Some segmental phonological processes involving vowels in Nambya: a preliminary descriptive account.
Introduction: The Origins, Sociolinguistic Status and Genetic Affiliation of Nambya
This study explores some segmental phonological processes involving vowels in Nambya. However, before a detailed qualitative analysis of the phonological processes a discussion of the origins, sociolinguistic status and genetic affiliation of this language is in order. Nambya is a community or minority Southern Bantu language that is predominantly spoken in the Matabeleland North province of Zimbabwe. This is the geographical area which was demarcated by the Government of Zimbabwe as the area where Nambya should be taught from Grade 1 to Grade 3. Furthermore, the Nambya people, with a total population of about 100 000 (Hachipola, 1998), are found in large numbers in the Hwange district of Zimbabwe under Chiefs Hwange, Shana, Mweemba and Nekatamba. Also, the Nambya people are the traditional inhabitants of the area around Hwange National Park. Hwange which is formerly Wankie was named after a revered local traditional Chief called Hwange and is 100 km from the resort town of Victoria Falls in the far North West of the Matabeleland North province.
In addition, some Nambya people migrated and settled in and around Victoria Falls in search of opportunities that this resort town presents. In this regard, the majority of Nambya people are self-employed and they provide services like curio carvings and traditional dances. Some native Nambya speakers are found in Binga, Nyamandlovu and Tsholotsho areas. However, the majority of people who are in these areas are largely Kalanga speaking. During interviews with Nambya speaking people I gathered that historically the Nambya people came from South Africa and some of them settled in Hwange on their way to Namibia. However, I also established that the majority of the Nambya youth do not know much about their history. Much of the information concerning the history and culture of Nambya people was from the elderly people, especially village headmen and traditional Chiefs.
After analyzing a sample of data that I collected as part of my pilot study for this study I discovered that Nambya shares many of its structural characteristics with typical Southern Bantu languages such as Shona, Kalanga and Ndebele. These features include the fact that Nambya is tonal and has a simple, symmetrical, unmarked phonemic five vowel system comprising /a/, /e/, /i/, /o/ and /u/ and a typical Bantu consonant-vowel (CV) syllable structure. Although Nambya has onsetless syllables, their occurrence is restricted to the word-initial position (see Kadenge, 2007). Therefore, Nambya is typically a Southern Bantu language. Also, like many other Bantu languages, Nambya has a highly agglutinative morphology.
Guthrie (1948) classifies Nambya as belonging to Zone S.10 under the Shona group of languages. In this regard, he treats it as a dialect of Kalanga, which he also classifies as a Shona dialect (also see Chabata, 2007). Other scholars of Southern Bantu languages who treat it as a dialect of Kalanga are Doke (1931), Kangira (2001), Wentzel (1983), Evans (1991) and Chebanne et al. (1995). In this context, Doke (1931) categorises Nambya as a dialect of Kalanga together with other varieties such as Nyai, Rozvi, Talahundura, Lilima and Peri. This is the same approach that Fortune (1967) adopts when he says that Kalanga comprises Nanzva (Nambya), Lilima, Twamamba, Rozvi and Lemba. Also, it seems Wentzel (1983) and Fortune (1967) treat Nambya and Kalanga as having the same status. If this is acceptable then it becomes logical not to take Nambya as a dialect of Kalanga, but as a separate variety that shares a lot with the latter (Chabata, 2007).
While Doke (1931), Guthrie (1948) and Kangira (2001) classify Nambya as a dialect of Kalanga, there are a number of historical and socio-cultural reasons that justify the treatment of these two as separate languages. For example, Chigwedere (1985), Hayes (1977) and Chabata (2001) note that the Nambya people and the Kalanga people have two different cultures although their languages are closely related. Furthermore, my observation and findings suggest that the Nambya people maintain that their cultures, that is, dances, beliefs, value systems and sensibilities are different from those of the Kalanga people. They actually argue that their variety is a separate language and not a dialect of any other language. Furthermore, according to the Zimbabwean constitution, especially the Education Act, as amended in 1990, Nambya and Kalanga are officially recognized as separate indigenous languages alongside Tonga, Venda, and Shangani/Changani. Therefore, Nambya is clearly a separate language with its own distinct linguistic features.
It is also noteworthy that, in Zimbabwe, Nambya is officially regarded as a minority language. It is considered a minority language because its speakers are fewer than the speakers of Shona and Ndebele, which are the main official indigenous languages in Zimbabwe. Its minority status can be explained by the fact that Doke (1931) recommended that Shona shall be the official language for the Mashonaland parts of Zimbabwe while Ndebele shall be the official language for Matabeleland. Again, in Doke's (1931) recommendations, minority languages such as Nambya and Kalanga are marginalized and documenting them was discouraged. In this regard, Doke (1931:100) recommends that "no school books or other books be published in the Lilima or Nambzya dialects."
Although the Zimbabwe Education 1987 Act (as amended in 1990) in Section 55 of Part X1 has provisions for the teaching of minority languages in areas where they exist, very little is actually being done. Furthermore, in Zimbabwe, there is very limited electronic broadcasting and print media which is being done in Nambya. However, this study argues that it is high time Zimbabwean minority languages such as Nambya are studied, developed and documented. This will allow teaching on and in these languages since the teaching of minority languages is being hampered by lack of textbooks and other teaching resource materials.
This recommendation is in line with Hyman's (2003:1) observation that "all of linguistics seems now to accept, if not enthusiastically encourage, the study of 'endangered languages' as well as 'minority languages', or what are generally referred to as 'unempowered' languages." The promotion of minority indigenous languages such as Nambya is in line with the requirements of the SADC Protocol on Culture, Information and Sport, which in Article 12 under language policy formulation says:
1) Member States shall formulate and implement language policies that will aim at promoting indigenous languages for socio-economic development, where such languages exist.
2) Member States shall institute and put into practical effect policy measures that will aim at encouraging the learning and wider use of the official languages of the Member States and the working languages of SADC.
3) Member States shall encourage and promote the use of indigenous languages as medium of instruction where such languages exist.
However, the language under discussion has not received enough promotion and in schools it is only taught up to Grade 2. The Nambya people are simply anxiously waiting for their language to be used in everyday programmes such as agriculture, education, media and in the general national economic and political spheres. Having outlined the socio-linguistic status of Nambya and its genetic affiliation, the following section briefly looks at the nature of studies that have so far been done on Nambya.
Previous Studies on Nambya Grammar
As mentioned earlier in this investigation, like most indigenous languages in Zimbabwe, Nambya is a poorly studied language that requires urgent attention from language practitioners. In addition, there are no well researched teaching materials that can be used to teach Nambya in schools. During my field trip, I noticed that at Chamabondo Primary School in Victoria Falls where it is taught from Grade 1 to Grade 2 the only teaching guide that they use as a reference book is a small manual titled Lyeja Lyotanga ChiNambya prepared by the now defunct Curriculum Development Unit (CDU) of the Ministry of Education, Sport and Culture. Furthermore, Moreno (1988) is the only published piece of work on Nambya, which is a small Nambya-English dictionary that has a very brief general grammatical description of Nambya. The general grammatical description briefly looks at the morphology of Nambya and a few passing statements on the phonology of the language. So far, major researches that have been done on Nambya are two doctoral studies (see Kadenge, 2007; Chabata, 2007). In this regard, Kadenge (2007) examines the segmental phonology of the language while the Chabata (2007) investigates its verbal structure with special emphasis on the causative construction. These studies are too technical to be used as teaching material at elementary levels. Therefore, there is need for the development of well researched teaching guides, grammar books and textbooks on this language.
The development of Nambya and other local minority languages is important given the fact that "language development is part of the infrastructural development of any country" (Chabata, 2001:5). Therefore, I hope that this investigation is going unravel important linguistic facts about Nambya, which will then act as a spring board for more in-depth research on other aspects of its grammar. Having discussed the nature of studies that have so far been done on the Nambya language, the following section discusses the data collection methods that were employed in this study.
Data Collection Procedures
In order to gather natural data for this research, I embarked on an intensive fieldwork in Hwange and Victoria Falls. As mentioned earlier in this investigation, these are areas where Nambya is predominantly spoken. In Hwange, I stayed at Lukosi and Mweemba areas which are about 10 km and 20 km away from Hwange town respectively. In Victoria Falls, I stayed at Chonotimba and regularly visited Chamabondo Primary School where Nambya is taught from Grade 1 to Grade 2. Since I am not a native speaker of Nambya I had to select a representative sample of informants for this study. These are the people who provided all the utterances that were analyzed in this article. Initially, in selecting my sample of subjects I targeted the older generation because they are the ones who seem to have a fluent command of the language, which unfortunately is in the process of being discarded by the younger generation. The younger generation prefers to speak Ndebele because it is more prestigious than Nambya and it is the one that they learn from Primary school to University level.
However, following Maddieson's (2001) advice that for phonetic research there are some particular problems with collecting data from the oldest speakers such as lack of teeth or use of artificial teeth or deterioration in motor control of the articulators I included some middle aged and younger voices in order to come up with a representative corpus of Nambya data. Although I considered and documented variables such as speaker age and level of education, I did not limit myself to a particular demographic variable. I collected the data from 80 informants and the use of multiple subjects was done so as to provide a control against individual idiosyncrasies, whether due to differences in the shape of speech organs, different personal histories or other factors.
I began by eliciting a long list of basic words for items such as body parts, household items, local food stuffs, geographical and astronomical objects, tools, clothing, kinship and colour terms. The items were asked one at a time. At times the informants were asked to repeat their answers three or four times to enable me to capture as much phonetic details as possible. The terms were collected, transcribed and glossed at the same time. Also, the informants were tape-recoded as they were responding to questions that I prepared in an interview sheet (available upon request). The interviewees were tape-recorded as they were responding to some questions relating to their history, development programmes, cultural values, hunting, farming practices and other areas of general interest. Usually, the tape-recording was done as the informants were carrying out their daily chores and during meal times because that is when there was a lot of talking going on.
At Chamabondo Primary School in Victoria Falls, with permission from the school authorities, I tape-recorded three sessions of Grade 2 Nambya lessons that were delivered in Nambya by a first language Nambya speaking teacher. Each session was approximately 30 minutes long. All the tape-recordings were transcribed onto paper and then glossed with the help of a research assistant who is a native speaker of Nambya. After the data was transcribed, the tapes were dated, numbered and kept safely together with their transcriptions. Furthermore, the audio-recording was done where there was no background noise and where there was little or no reverberation. While in the field, all the data were checked with multiple native speakers. This was done in order to avoid individual peculiarities. The fieldwork progressed very well because Nambya people are enthusiastic for their language to be studied, hoping that studies such as this one may help in the elevation of the status of their language. Furthermore, the informants for this study were very useful in that they also provided me with relevant information concerning their understanding of the origins of their language, its status and how it has been influenced by neighbouring languages such as Ndebele, English, Kalanga, Shona and Tonga.
Phonological Processes Involving Nambya vowels
As mentioned earlier in this investigation, this study presents a descriptive analysis of some phonological processes involving Nambya vowels. In this regard, this study adopts a descriptive framework of data analysis because "as history has shown, it lends itself best to the data and makes possible relatively concrete classification as an aspect of linguistic description and analysis" (Mkanganwi, 1995:5). In this context, this study examines Nambya vowel harmony and hiatus resolution strategies such as glide formation, vowel elision and vowel coalescence. Furthermore, in Nambya, tone does not play a crucial role in vowel harmony and hiatus resolution. As a result, this study will not make any reference to tone and will not mark words for tone.
Progressive and Regressive Vowel Harmony in Nambya
This research examines the vowel harmony system of Nambya. According to Bakovic (2002:1), vowel harmony is "a process by which vowels in adjacent syllables in some domain (typically a word) agree with each other in terms of some distinctive feature(s)." In this regard, this study observed that in Nambya after extending verbs in order to create complex words the realization of the vowel of the verbal extension is determined by the immediately preceding vowel in the verb root. In order to clearly demonstrate the progressive vowel harmony that is exhibited in Nambya this study concentrates on the realization of the language's applicative extension. Let us see the examples given below:
 root + terminal root + applicative extension + terminal vowel /lil + a/ 'cry' [lil + il + a]' cry for' /bal + a/ 'write [bal + il + a]' write for /tum + a/ 'send' [tum + il + a]' send for' /tol + a/ 'take' [tol + el + a]' take for' /teng + a/ 'buy' [teng + el + a]' buy for'
Example  above shows that in Nambya the applicative morpheme is /-il-/ which is realized as [-il-] when immediately preceded by [i, a, u] and is realized as [-el-] when preceded by [e, o] in the verb root. In this regard, the nature of the vowel of the applicative suffix is determined by the immediately preceding vowel in the verb root. It should be noted that combining vowels from these two mutually exclusive sets is systematically prohibited. This explains why constructions such as * [bal-el-a] and * [tol-il-a] are semantically ill-formed in Nambya. This observation is accounted for by the fact that in Nambya the applicative verbal extension cannot surface with either [e] when the immediately preceding vowel is [i, a, u] or with [i] when the immediately preceding vowel in the verb root is [e, o]. This means that the acceptable forms of these words are [balila] 'write for' and [tolela] 'take for'. These observations also apply to the realization of the Nambya neuter and causative extensions (also see Kadenge, 2007).
This type of progressive vowel harmony is similar to the pattern that is found in Shona. In Shona, verbal extensions such as the causative have two allomorphs, namely, [-is-] and [-es-]. For example, in the Korekore dialect of Shona, the causative extension /-is-/ "is realized as [is-] when the last vowel in the in the radical is /a/, /i/ or /u/ and the consonant radicals; the allomorph /-es-/ occurs when the last vowel of the radical is /e/ or /o/" (Dembetembe, 1987:62). Similar vowel harmony patterns obtain in Kalanga (Mathangwane, 1999) and Luganda (Hyman, 2004).
This study also observed that the harmonization of vowels in Nambya is not limited to the realization of verbal extensions. In this regard, in Nambya, the realization of prefixal morphemes depends on the nature of the vowel of the immediately following vowel that belongs to the class affix of the noun. This observation is shown in  below:
[2a] Underlying Represent Surface Representation /a + mi-namato/ [e-mi-namato] 'of the prayers' /lwa + zwi-po/ [lwe-zwi-po] 'of the gifts' /na + tsi-nkwa/ [ne-t[??]i-nkwa] 'with bread' [2b] /zwa + ku-lja/ [zwo-ku-lja] 'food stuff /na + nu-lume/ [no-nu-lume] 'with men' /a + tu-sindi/ [o-tu-sindi] 'of squirres' [2c] /ja + Ba-kwajij [j a-Ba-kwajijana] 'of the foreigners' /ja + Ba-na/ [ja-Ba-na] 'of the children'
The general pattern that is emerging from examples [2a] through [2c] is that the vowel realization of immediately preceding morpheme depends on the nature of the place of articulation of the immediately following vowel of the class affix. This means that the class affix vowel regressively spreads its place of articulation and lip rounding features to the vowel of the preceding morpheme. In this regard, /a/ is realized as [e] when immediately followed by [i] (see [2a]) and is realized as [o] when immediately followed by [u] (see [2b]) and is realized as [a] when immediately followed by [a] (see [2c]) in the class affix. In this regard, /a/ is realized as [e] after regressively acquiring the features unrounded and front from [i]. Also, it is realized as [o] after assimilating the features rounded and back from the immediately following [u] and it remain [a] when followed by [a]. It is also clear from examples [2a] through [2c] that in Nambya the vowels of the class affixes are /i/, /a/ and /u/. On the basis of this observation, I can safely argue that Nambya was historically a three vowel language. This theory is based on the realization that diachronically the two secondary vowels [e, o] are phonologically derived from their basic counterparts /i, a, u/. Generally, this study has shown that vowel harmony in Nambya is progressive when it involves suffixal alternation and it is regressive when it involves prefixal alternation. Therefore, the Nambya language is characterized by bi-directional vowel harmony because it exhibits both progressive and regressive vowel harmonies. The section below looks at the various ways through which vowel sequences (hiatus) are removed in Nambya.
Nambya Hiatus Contexts and Hiatus Resolution Strategies
It is a well-known fact that many languages do not tolerate sequences of vowels (hiatus) in their phonologies (Casali, 1996). However, hiatus arises due to morphosyntactic concatenation. Furthermore, languages rely on morphophonemic processes such as glide formation, vowel deletion, vowel coalescence and consonant (usually glide) epenthesis to resolve hiatus (see Casali, 1996, 1997). This study observed that Nambya utilizes glide formation, vowel deletion and vowel coalescence in order to resolve hiatus. The following section explores the glide formation process.
In descriptivist terms, glide formation is the process whereby a high vowel of a class affix or gender concord changes to become a glide when followed by a vowel commencing stem. In Nambya, the glide formation process occurs with prefixes that have either /u/ or /i/. This means that class prefixes such as [mi-] and [mu-] surface in that form before consonant commencing stems. The data given below illustrate the glide formation processes that are operative in Nambya.
[3a] prefix + stem Surface Representation /mu + ana/ [mwana] 'child' /mu + anakwajijana/ [mwanakwajijana] 'cousin' /lu + endo [lwendo] 'journey' [3b] /mi + ojo/ [mjojo] 'hearts' /mi + ono/ [mjono] 'fish traps' /i + ose/ [jose] 'all of it'
Examples [3a] and [3b] show that in Nambya there two types of glide formation, namely, [w] and [j] formation. In this regard, [w] is formed when the prefix has /u/ and [j] is formed when it has /i/. The glide formation processes deal with the hiatus that is a result of prefixation. It is worth noting that the glide formation processes only turn the vowel of the prefix to a glide and not the vowel belonging to the stem. Furthermore, these processes deal with word-medial onsetless syllables since after the glide formation process the stem-initial vowel becomes the nucleus of the word-initial syllable. In other words, the stem-initial vowel takes over the position of the glided vowel. It is noteworthy that the glide formation processes help in maintain the canonical Nambya CV syllable structure. Furthermore, these processes remove the hiatus that is created by prefixation processes. Vowel elision is one of the phonological processes that are used to deal with vowel sequences in Nambya. The exploration of this process is the subject matter of the following section.
Vowel elision is one of the phonological processes that are used to eliminate hiatus in Nambya. This process results in the disappearance of a speech segment. This process occurs in Nambya when a consonant + vowel sequence is realized as a consonant only before vowel commencing morphemes generally and non-deverbative vowel commencing stems. The vowel elision process is demonstrated in the examples given below:
 Underlying Represent Surface Representation /mu + oto/ [moto] 'fire' /mu + ojo/ [mojo] 'heart' /tsi + oto/ [t[??]oto] 'fire place' /tsi + ulu/ [t[??]ulu] 'anthill' /zwi + ar)gu/ [zwangu] 'mine' /tsi + ar)gu/ [t[??]angu] 'mine' /wa + ar)gu/ [wangu] 'mine' /Ba + a)gu/ [Bangu] 'mine'
Example  above shows that in Nambya a sequence of two vowels created by prefixation results in the deletion of the first vowel. As the data above shows, the first vowel or the vowel of the prefix is deleted when a consonant + vowel sequence is followed by a vowel commencing morpheme. Therefore, Nambya is a first vowel deleting grammar. Furthermore, this process occurs within a word, that is, between a vowel final prefix (usually of the CV shape) and a vowel initial stem. It is noteworthy that the vowel elision process facilitates the creation of the canonical Nambya CV syllable structure. Also, words are resyllabified after the deletion of a vowel. In this regard, a word such as /mu + oto/ which has three syllables in the underlying representation is realized as a disyllabic word [moto] 'fire' after the deletion of a vowel.
This pattern of vowel elision is consistent Casali's (1996:11) observation, after a cross-linguistic survey of 92 languages, that "languages with the first vowel elision outnumber languages with second vowel elision by more than two to one and every language with second vowel elision also have first vowel elision in at least some contexts." Thus, there is overwhelming evidence that the first vowel deletion process is more common than the second vowel deletion process. As mentioned earlier in this investigation, the vowel elision process is important in Nambya phonology because it removes hiatus. The vowel elision pattern that obtains in Nambya is similar to the elision processes that have been reported to be common in some Bantu languages such as Bemba, Chichewa, IsiZulu, Luganda, SiSwati, Tsonga, Cinsenga and Xhosa. The following section analyses vowel coalescence as one of the phonological processes that are used to remove vowel sequences in Nambya.
Coalescence is the merging of two speech segments to form one. This investigation observed that in Nambya vowel coalescence involves the merging of two basic vowels /i, a, u/ to form a single secondary vowel, that is, either [e] or [o]. In this regard, this research observed that in Nambya the sequences of /a + i/ and /a + u/ are realized as [e] and [o] respectively in order to resolve hiatus. The pattern that this study established concerning vowel coalescence is that when the basic front high unrounded vowel /i/ is preceded by the basic low central vowel /a/ [e] is formed and when the basic back high rounded vowel /u/ is preceded by the basic low central vowel /a/ [o] is formed. The driving force behind the occurrence of vowel coalescence is that Nambya does not allow hiatus and word-medial onsetless vowels. Examples [5a] and [5b] below illustrate the phonological environments in which vowel coalescence occurs in Nambya. Example [5a] illustrates the morphosyntactic environments in which [o] is formed in Nambya:
[5a] Underlying Represent Surface Representation /hia + uno/ [hono] 'here he is' /hia + uju/ [hoju] 'this one' /hia + ujo/ [hojo] 'that one' /ja + uju/ [joju] 'this one' /ja + utu/ [jotu] 'these one' /ja + umu/ [jomo] 'there inside' /na + unu/ [nonu] 'with a person'
Example [5a] above shows that in Nambya [o] results from the coalescence of /a/ and /u/. As mentioned earlier in this study, the [o] formation process is triggered by the need to eliminate hiatus and word-medial onsetless syllables. As shown in [5a] above, during the formation of [o] it is /u/ which is onsetless. Due to [o] formation two syllables are merged into one syllable. This means that a word which is underlyingly trisyllabic (three syllables) is realized as a disyllabic word on the surface representation. For example, the word [ho.ju.] 'this one' is bisyllabic but is underlyingly trisyllabic [ha.u.ju.]. Therefore, the vowel coalescence process resyllabifies the word from three syllables to two syllables. The examples given in  below show the morphosyntactic domains in which [e] is formed in Nambya.
 Underlying Represent Surface Representation /ja + ijo/ [jejo] 'that one' /ja + it[??]o/ [jet[??]o] 'that one over there' /ja + it[??]i/ [jet[??]i] 'this one' /ma + i[??]o/ [me[??]o] 'eyes' /ma + ino/ [meno] 'teeth'
Example  above shows that in Nambya [e] is formed from the merging or coalescence of /a/ and /i/. It is also noteworthy that the underlying /i/ is an onsetless syllable. Therefore, it is coalesced with the immediately preceding syllable nucleus because Nambya does not allow word-medial onsetless syllables. Furthermore, the vowel coalescence process is triggered by the fact that Nambya prohibits hiatus. It is noteworthy that the pattern of the vowel coalescence processes that are operative in Nambya belongs to the pattern which occurs frequently in Bantu languages. The Nambya pattern of vowel coalescence processes is found in other Bantu languages such as Shona (Doke, 1931; Fivaz, 1966, 1970; Harford, 1997), Ndau (Mkanganwi, 1973), Cinsenga (Miti, 2001, 2006), Ndebele (Ndlovu, 1997; Khumalo, 2003; Sibanda, 2009) and IsiZulu (Doke, 1927).
Conclusion and Recommendations
In conclusion, I wish to emphasize that this research set out to simply make an empirical and/or "materials contribution" (Mkanganwi, 1995:83) on the structure of Nambya and I hope that the mere existence of this study will feed into most, if not all, of the existing theories in the area of Bantu linguistics. Furthermore, as is always the case in linguistic theory, description must always precede theory. This article has also made a number of observations concerning the sociolinguistic status of Nambya in Zimbabwe and its genetic affiliation. In this regard, this research has shown that Nambya is a poorly studied Southern Bantu minority language. In this regard, at the policy level, the Zimbabwean government is encouraged to effectively implement the Article 12 of the SADC protocol on Culture, Information and Sport which says that SADC Member States must formulate and implement policies that are aimed at promoting indigenous languages for socio-economic development, where such languages exist.
This explains why Pan African scholars such as Ngugi (1986) say that language is community owned and it is the right of the speakers to see to it that their language is eventually used as a tool to solve problems in that community, as well as in moments of everyday celebration of various events and occasions in that society. This is a crucial observation because language is an important tool of community development. In other words, it postures the whole being of a person, including the way people think and act. Therefore, the Nambya language must not be allowed to die because it is a crucial cultural resource and a symbol of identity and heritage for its speakers. Its development is some kind of cultural liberation from the Western world and it is the only means through which its speakers can effectively engage themselves in the process of globalization with the outside world.
Concerning the linguistic structure of Nambya, this research has made a number of observations concerning the language's hiatus resolution and vowel harmony processes. On Nambya vowel harmony, this study has shown that it is predictable because it is rightwards (progressive) when it involves suffixal alternation and it is leftwards (regressive) when it involves prefixal alternation. Furthermore, the findings of this research have shown that in Nambya hiatus is not permitted and it is consistently resolved through glide formation, vowel deletion and vowel coalescence. These processes help in removing word-medial onsetless syllables and help in maintaining the language's canonical or preferred CV syllable structure. Since this study has a restricted scope it cannot be expected to interrogate and resolve all issues concerning Nambya phonology in particular and its grammar in general. In this context, I hope that this research will be a means for further exploration of Nambya phonology and all kinds of other linguistic aspects in general.
Bakovic, E. 2000. Harmony, Dominance and Control. Unpublished doctoral dissertation. USA: Rutgers University.
Casali, R. F. 1996. Resolving Hiatus. Unpublished PhD Thesis. Los Angeles: University of California.
Casali, R. F. 1997. Vowel Elision in Hiatus Contexts: Which Vowel Goes? Language, 3: 493-533.
Chabata, E. 2007. The Nambya Verb with Special Emphasis on the Causative. Oslo: University ofOslo.
Chebanne, A. M. et al. Ngatikwaleni Ikalanga: A Manual for Writing Ikalanga as Spoken in Botswana. Botswana: The Botswana Society.
Chigwedere, A. 1985. The Karanga Empire. Harare: Books for Africa.
Dembetembe, N. C. 1987. A Linguistic Study of the Verb in Korekore. Harare: University of Zimbabwe Publications.
Doke, C. M. 1927. Textbook of Zulu Grammar. Witwatersrand: University of the Witwatersrand Press.
Doke, C. M. 1931. A Comparative Study of Shona Phonetics. Johannesburg: University of Witwatersrand Press.
Evans, J. 1991. Summary of the Kalanga Orthography Debate Proceedings. In van Waarden (ed.) Kalanga: Retrospect and Prospect. Gaborone: Gaborone Society.
Fivaz, D. 1966. Some Aspects ofShona Structure (Zezuru Dialect). Salisbury: Department of African Languages, University College of Rhodesia.
Fivaz, D. 1970. Shona Morphophonemics and Morphosyntax. Unpublished PhD Thesis. Johannesburg, South Africa: University of Witwatersrand.
Fortune, G. 1967. Elements of Shona. Harare: Longmans.
Guthrie, M. 1948. The Classification of Bantu Languages. London: Oxford University Press.
Hachipola, S. J. 1998. A Survey of Minority Languages of Zimbabwe. Harare: University of Zimbabwe Publications.
Harford, C. 1997. When Two Vowels Go Walking: Vowel Coalescence in Shona. Zambezia, xxiv (1): 69-85.
Hayes, M. E. 1977. The Nambiya People of Wange. NADA 11: 385-393.
Hyman, L. 2003. Why Describe African Languages? World Congress of African Linguistics, Keynote Address, June 18, 2003.
Hyman, L. 2004. African Languages. Berkeley: Universi8ty of California.
Kadenge, M. 2007. The Phonology of Nambya. Unpublished D.Phil Thesis. Harare: University of Zimbabwe.
Kangira, J. 2001. Transitivity in Kalanga. Unpublished M.Phil Thesis. Harare: University of Zimbabwe.
Khumalo, L. 2003. A General Introduction to Ndebele Grammar. Cape Town: The Center for Advanced Studies of African Societies (CASAS).
Maddieson, I. 2001. Phonetic Fieldwork. In Newman, P. and Ratliff, M. (eds.) Linguistics Fieldwork. Cambridge: Cambridge University Press.
Mathangwane, J. T. 1999. Ikalanga Phonetics and Phonology: A Synchronic and Diachronic Study. USA: CSLA Publications.
Miti, L. 2001. A Linguistic Analysis of Cinsenga: A Bantu Language Spoken in Zambia and Malawi. Cape Town: The Center for Advanced Studies of African Societies (CASAS).
Miti, L. 2006. Comparative Bantu Phonology and Morphology. Cape Town: The Centre for Advanced Studies of African Societies (CASAS).
Mkanganwi, K. G. 1973. An Outline of the Morphology of Substantives in Ndau, with a Preliminary Note on Ndau Phonology." Unpublished M.Phil Thesis. London: University of London.
Mkanganwi, K. G. 1995. Shona: A Grammatical Sketch. Unpublished Manual. Harare: University of Zimbabwe.
Moreno, A. 1988. Nambya Dictionary. Gweru: Mambo Press.
Ndlovu, H. 1997. The Historical Development of the Ndebele Language. Unpublished MA dissertation. Harare: University of Zimbabwe.
Ngugi, W. 1986. Decolonising the Mind: The Politics of Language in African Literature. London: James Currey.
Sibanda, G. 2009. Vowel Sequences in Nguni: Resolving the Problem of Unacceptable VV Sequences, in Matondo, M., McLaughlin, F. and Potsdam, E. (eds.) Selected Proceedings of the 38th Annual Conference on African Linguistics Cascadilla Proceedings Project, Somerville, MA, USA: 38-55.
Wentzel, J. J. 1983. The Relationship between Venda and Western Shona. Unpublished PhD dissertation. South Africa: University of South Africa.
Maxwell Kadenge, D.Phil.
Department of Linguistics, University of the Witwatersrand
Maxwell Kadenge is a Postdoctoral Research Fellow in the School of Literature and Language Studies at the University of the Witwatersrand. His research interests are mainly located within the domains of Southern Bantu phonology, morphosyntax and Zimbabwean English.