NAME, DATE HERE
GĨkŨyŨ and KĨkamba 39
CHAPTER THREE
Harmonizing THE Orthography of GĨkŨyŨ and KĨkamba
Angelina Nduku Kioko, Martin C. Njoroge and Peter Mburu Kuria
Introduction The term orthography is derived from the Greek word ‘orthos’ which means ‘correct’, and ‘graphein’, which stands for ‘to write’ (Sampson, 1985). The orthography of a language describes or defines the set of symbols (graphemes and diacritics) used to represent the phonemic inventory of that language in the writing and the rules on how to write these symbols. According to Massamba (1986), a language takes a limited number of sounds from the central pool of speech sounds to form its phonetic inventory. In this chapter, orthography is used to refer to the system of symbols used in the writing system of Gĩkũyũ and Kĩkamba. There are three types of orthographies (Read, 1983: 143-152). The first is the ‘phonemic orthography’. In a ‘phonemic’ orthography there is a one-to-one correspondence between phonemes and graphemes. This type of orthography has a dedicated sequence of symbol or symbols for each phoneme. Examples of languages that have phonemic orthographies are Korean and Kiswahili. The second type is the ‘morpho-phonemic orthography’ which considers both the phonemic features and the underlying structure of words. In this case, words may be written in the same way despite differences in pronunciation. For example, the pronunciation of the plural marker in English {s} is conditioned by the phonetic environment in which it occurs, yet it is written with the same grapheme . The plural forms ‘cats’ and ‘dogs’ are pronounced as [kӕts] and [dɒɡz] respectively although the two final sounds are written 40 the harmonization and standardization of kenyan languages with the grapheme . The third type is ‘the defective orthography’ in which there is no a one-to-one correspondence between graphemes and phonemes in the language. For example, English and Arabic have sounds with more than one spelling. The palato-alveolar affricate sound [dӡ] in English, for instance, is represented by
The phonology and orthography Gĩkũyũ is a Bantu language of the A category characterized by a concordial feature agreement (Guthrie 1967; Heine 1978). According to Kenya’s Population Census (1999), there were just over seven million Agĩkũyũ in the country at that time. The Agĩkũyũ mainly occupy the central part of Kenya but are also distributed across other regions such as the Rift Valley and Coast provinces. They are also found in the major towns of Kenya. The first Gĩkũyũ orthography was designed by Christian mis sionaries according to Wanjau (1991). These were non-native speakers of Gĩkũyũ who did not represent the words as they were pronounced by the native speakers. There was thus no one-to-one correspondence between the phonemes and their graphemes. The United Kikuyu Language Committee in the 1940s resolved some difficulties in representing vocalic phonemes graphemically but did not resolve consonantal difficulties. Kĩkamba was originally spoken natively in four districts in the Eastern Province of Kenya. According to the 1999 Kenya National Census there were over 3.3 million Kĩkamba speakers at that time (8% of the Kenyan population). Kĩkamba is one of the Central Kenya Bantu Languages categorized by Guthrie (1967) as E55, that is, it is language unit 5 of the Kikuyu-Kamba group 50, situated in BANTU zone E. In most written literature, the morpheme ‘-kamba’ has been used to refer to the people, to the language, or to the land. However, this morpheme has no other semantic value except that of being a root morpheme. It can only be considered as a word if it has a noun class prefix affixed to it. Thus the correct reference forms are a+kámbá (noun cl.2) the people, mũ+ kámbá (noun cl.1) a person from the group, kĩ+ kámbá (noun cl.7) the language, and ũ+ kámbá (noun cl.11) their land.
The Vowels a. Gĩkũyũ vowels The Gĩkũyũ vocalic phonemes are /a/ /E/ /i/ /e/ /O/ /u/ and /o/, as presented in Figure 1 adopted from Mwihaki (1998:37). These phonemes are represented graphemically as ,
Figure 1: Gĩkũyũ vowels
Front Back High
Low
These vowels can occur in any phonetic environment within a word. But owing to the fact that Gĩkũyũ is predictably an open syllable system, vowels function more regularly in the word medial and word final positions. The Gĩkũyũ vowel is optionally short or long, and a single, geminate or contour unit (Carr 1993). Single vowels are simple units whose duration corresponds to one mora as in the Gĩkũyũ word ‘nda’ [nda] (stomach), which contrasts with the geminate vowel in the word ‘ndaa’ [nda:] (louse). Thus, as this illustration shows, vowel length is distinctive. A contour vowel is a mono-moraic unit consisting of non- identical properties and has a two-to-one linkage to skeletal slots when analysed within Autosegmental Framework (see Mwihaki 1998:37 for more details). Each of the vowel sounds in Gĩkũyũ has a grapheme representing it orthographically. Table 1 shows the grapheme, the phoneme and and an example of a word in Gĩkũyũ for each of these vowel sounds. We now turn to Kĩkamba.
a GĨkŨyŨ and KĨkamba 43
Table 1: Gĩkũyũ vocalic phonemes and their corresponding graphemes
Sound Sound Word Gloss grapheme (phoneme) illustration i /i/ ita strangle High ĩ /e/ ĩta call e /E/ eka hiccup a /a/ aka build o /O/ oya lift/pick up ũ /o/ ũria ask u /u/ uga say
b. Kĩkamba vowels The vowel chart representing the Kĩkamba vowels is identical to the one for Gĩkũyũ in Figure 1.
Figure 2: Kĩkamba vowels
Front Back High
Low
Kioko (2005:14) argues that length is phonemic in Kĩkamba and presents a chart showing both short and long vowels, but this is a position that is rather difficult to defend because there are many intervening variables. Every occurrence of two or three similar vowels cannot be taken as a case
a a ɔ uo 44 the harmonization and standardization of kenyan languages of a long vowel. At times these vowels represent syllables bearing distinct tones. Kĩkamba can have a sequence of up to eight vowels. This feature has risen from a historical consonant loss. According to Maundu (1980), Kĩkamba has historically lost two consonants as is seen when we compare lexical items with vowel sequences in Kĩkamba with the same lexical items in the related languages, as shown in Table 2.
Table 2: Comparison of vowel combination in Kĩkamba and related languages
Gloss Kĩkamba Gĩkũyũ Kimeru
banana /eio/ /iriØo/ /iriØo/
cook /ua/ /ruØa/ /ruØa/
buy /oa/ /Øora/ /Øora/
In many cases, the lost consonants leave behind a consonant position that can relate to certain phonological processes (Ford 1976). Again, a series of similar vowels does not mean that the vowels are in the same syllable, because the language has many vowel-only syllables. One of the commonly quoted minimal pairs – [ka:na] ‘kana’ (child) and [kana] ‘kana’ (four times) – suffices to demonstrate this uncertainty. The underlying form of [ka:na] is [ka+ana] ‘child’. Therefore, the length here results from the deletion of the noun root vowel. On the other hand, [kana] has an underlying form of [ka+na] ‘four’. These two do not form a minimal pair at all. Many of the other quoted minimal pairs differ in tone. The orthography is not uniform in its representation of length. In some cases, the orthography is that of a single vowel as in the case of ‘kana’ above, while in other cases it is a series of vowels of the same quality but not necessarily of the same tone. It is therefore more plausible to present the seven Kĩkamba vowels as in Table 3 and discuss the interaction of this set of vowels with tone and phonological processes. GĨkŨyŨ and KĨkamba 45
Table 3: Kĩkamba vocalic phonemes and their corresponding graphemes
Sound Sound unit Word Gloss grapheme (phoneme) example i /i/ ita strangle ĩ /e/ ĩta call e /E/ eka stop a /a/ aka build o /O/ osa take ũ /o/ ũka come u /u/ umya remove
The consonants a. Gĩkũyũ consonants Armstrong (1967), using the Southern Dialect, identified the fol lowing eighteen consonantal phonemes: /mb/ /F/ /m/ /ɲ/ /t/ /T/ /nd/ /r/ /n/ /∫/ /nʤ/ /k/ /ŋɡ/ /ŋ/ /w/ /h/ /j/ and /ɡ/. She used the phonemic principle to develop the orthography of Gikŭyŭ. These consonants are presented in Table 4, which is adapted from Njoroge (2006).
Table 4: Gĩkũyũ consonant phonemes
Place Labial Dental Alveolar Alveo- Palatal Velar Glottal Manner palatal m n Plosives b t, d k, ɡ, ŋɡ Fricatives F T ∫ h
Affricates t, nʤ
Nasals m n ɲ ŋ
Approxi- r j mants 46 the harmonization and standardization of kenyan languages
These consonant phonemes and their corresponding graphemes are presented in Table 5.
Table 5: Gĩkũyũ consonantal phonemes and their corresponding graphemes
Phoneme Grapheme Gĩkũyũ example English gloss
/ф/ b baba father
/mb/ mb mbara war
/m/ m maitũ mother
/t/ t tata aunt
/θ/ th thama migrate
/nd/ nd nda stomach
/n/ n nina finish
/r/ r rĩma dig
/ʃ/ c cama taste
/nʤ/ nj njĩra path
/ɲ/ ny nyama meat
/j/ y yake his/hers
/k/ k kana or
/ɣ/ ɡ ɡĩa possess
/ŋɡ/ nɡ nɡaya inherit
/ŋ/ nɡ’ nɡ’araɡu famine
/w/ w wĩra work
/h/ h haha right here GĨkŨyŨ and KĨkamba 47
As Table 5 indicates, the prenasalized phonemes /mb/, /nd/, /nʤ/ and /ŋɡ/ are represented with diagraphs. One grapheme represents the nasal element, which Armstrong (1967) argues should always be regarded as present, though sometimes in a very weak form. Indeed, research shows that it may not always be present and some writers do not even represent it. Consequently, the phonemes are sometimes graphemically represented as , , graphemes that are not present in the orthography of Gĩkũyũ. b. Kĩkamba consonants Arriving at the inventory of the Kĩkamba consonant phonemes is not an easy task. While a lot that has been written on the structure of the language, there is yet no agreement on how to represent the phonological characteristics of the consonants in the language. In this chapter we will adopt the approach in Kioko (2005), but the inventory presented here incorporates new evidence and a reanalysis of certain aspects of the phonological system. Kioko (2005:11) categorizes the consonant sounds into simple consonants and compound consonants. On the one hand, what she refers to as ‘compound’ phonemes represent three sets of consonants with complex articulation features. The term ‘simple’ refers to consonant sounds whose production involves a standard manner and place of articulation similar to that of the same sounds in other languages. On the other hand, the term compound has been used in the literature to refer to nasalized consonants and could be extended to other consonants with similar complex articulation features. The ‘simple’ phonemes can be presented in a manner and place of articulation chart, but we have found it clearer to discuss the ‘compound’ phonemes separately. Table 6 therefore presents only the simple phonemes. 48 the harmonization and standardization of kenyan languages Table 6: Kĩkamba consonantal phonemes Place Bilabials Dental Alveolar Alveo- Palatal Velar Manner palatal Stops t k Nasals m n ɲ Ŋ Fricatives ф ө s Affricates ʃ tʃ Glides w j w Laterals l The use of these consonants in Kĩkamba words is exemplified in Table 7. Table 7: Kĩkamba consonantal phonemes and their corresponding graphemes Phoneme Grapheme Kikamba example English gloss /Ф/ v vaya there /m/ m matu clouds /t/ t tala count /θ/ th thama migrate /n/ n nini small /l/ l linda protect /s/ s sama taste /ʃ/ sy syana children /tʃ/ ky kyama miracle/party /ɲ/ ny nyama meat /j/ y yake his/hers /k/ k kĩndũ thing /ŋ/ nɡ’ nɡ’ea slaughter /w/ w wĩa work /w/ w’ w’ĩa fear GĨkŨyŨ and KĨkamba 49 The information in Table 7 suggests that the language has a minimal number of consonants (15), but this is not the case. There are compound phonemes which fall into three categories: prenasalized compounds, palatalized compounds and labialized compounds. The prenasalized compounds receive the most coverage in literature on African languages. Omondi (1982) calls them ‘nasal compounds’ while Welmers (1973) refers to them as ‘prenasalized consonants’. Although compound phonemes are represented in the orthography by more than one orthographic symbol, there is agreement in the literature available that the sounds represented by these letter combinations constitute single consonant phonemes. Kĩkamba compounds a. Nasal compounds The nasal compounds or prenasalized consonants in Kĩkamba result from a complex articulation in which the airstream is released through both the nasal and the oral cavities. The stronger part of the sound however is the oral consonant. There are seven prenasalized consonants in Kĩkamba as listed in Table 8. Table 8: Kĩkamba nasal compounds Pre-nasalized Representation in the Word consonant orthography example 1 /nd/ nd ndaa (lice) 2 /ŋɡ/ ng ngunga (cave) 3 /nz/ nz nzamba (cockrel) 4 /ɲʃ/ nzy nzyĩma (hunting) 5 /ɲʤ/ ngy ngya (poor) 6 /mb/ mb mbakĩ (snuff) 7 /nð/ nth nthũ (enemy) From the presentation above, it is clear that the voiced obstruents in Kĩkamba ([b], [z], [d], [ɡ]) result from phonological processes, and do 50 the harmonization and standardization of kenyan languages not occur as simple phonemes. This leads to the conclusion that these surface realizations have underlying forms with voiceless consonants. In fact many of the nasal compounds in word initial position derive from the combination of the noun class prefix {n-} and the initial consonant author to of the root morpheme, as shown in (1). confirm O is (1) OK (a) [n +sOu] [nzoOu] elephant (b) [n + ko] [nɡo] firewood (c) [n +Фua] [mbua] rain (d) [n +tao] [ndao] a kind of tree Since Kĩkamba does not attest nasal compounds involving a nasal and a voiceless consonant, it is plausible to argue that the nasal compounds are underlyingly of the form involving only the attested simple phoneme consonants given above, i. e., ŋg results from ŋk, nd from nt, mb from mФ etc. This brings to question the argument of presenting nasal compounds as separate phonemes. However, on the basis of the syllable structure of Kĩkamba (where such compounds are considered single consonants) and on the basis of the cross-linguistic treatment of this phenomenon, we will analyse these compounds as phonemes. b. Labialised compounds Apart from /w/ and /w/,1 all the other simple consonants in Kĩkamba have a labialized version. The complete list is presented in Table 9. 1 These two approximants generate two sets of labialized consonants in Kĩkamba, which are distinctive sounds as illustrated here: kũtwa ( to harvest honey), kũtũa (to settle), kũtua (to brew), kũtw’a (to determine). Thus Table 9: Kĩkamba labialized compounds Labialized Representation in Word example consonant the orthography 1 /tw/ tw mũtwe (head) 2 /kw/ kw mũkwa (rope) 3 /sw/ sw tũswii (chicks) 4 /өw/ thw mũthwa (ants) 5 /фw/ vw vwana (lord) 6 /mw/ mw mwako (building) 7 /nw/ nw inwa (be sang) 8 /ɲw/ nyw kanywa (mouth) 9 /ŋw/ nɡ’w kĩnɡ’waa (crow) 10 /lw/ lw ũlwa (forget) A number of the labialized compounds result from gliding as seen in (2). (2) mũ-aka [mwaka] (year) kũ-aka [kwaka] (to build) tũ-ana [twana] (children) However, like the other two approximants discussed above, there are many for which there is no transparent motivation for positing a gliding process such as those given in (3). (3) kwata (catch) Фwana (lord) 52 the harmonization and standardization of kenyan languages c. Palatalized compounds The palatalized consonants fall into a similar pattern as the labialised. Table 10: Kĩkamba palatalized compounds Palatalized Representation in Word example consonant the orthography 1 /tj/ ty etya (make hard) 2 /өj/ thy thyaka (quiver) 3 /Фj/ vy ũvyũ (knife) 4 /mj/ my umya (remove) 5 /nj/ ny’ ngelany’o (cross) 6 /ŋj/ nɡ’y ũseng’yo (surprise) 7 /lj/ ly kũleelya (to loosen) The palatalized compounds involve a complex combination of a simple phoneme and the glide /j/. Many of these result from the phonological process of gliding exemplified in (4). j (4) mĩ-aka [m aka] (years) However, many of the palatalized phonemes cannot be derived syn chronically from the combination of a simple phoneme and the palatal glide. The articulation of the resulting consonant can be best described by the term ‘coarticulation’. Table 11: Complex compound consonants 1 – Pre-nasalized and labialized Prenasalized and Representation in Word example labialized the orthography 1 /ndw/ ndw ndwĩku (torn) 2 /nɡw/ nɡw ngwae (a kind of bird) 3 /nzw/ nzw nzwĩĩ (hair) 4 /nðw/ nthw nthwau (wings) 5 /nФw/ mbw mbwii (bell) GĨkŨyŨ and KĨkamba 53 Table 12: Complex compound consonants 2: Pre-nasalized and palatalized Prenasalized and Representation in Word example palatalized the orthography 1 /ndj/ ndy vindya (keep quiet) 3 /nzj/ nzy nzyawa (offspring) 4 /nðj/ nthy nthyũũa (dizziness) 5 /nФj/ mby mbya (horns) Difficulties in orthographies a. Challenges of the Gĩkũyũ orthography As noted in the foregoing, some difficulties in representing vocalic phonemes graphemically were resolved in the 1940s. But some difficulties still persist to date. One difficulty is in the use of double letters to represent long vowels as advocated by the International Institute of African Languages (Armstrong, 1967). Some writers do not use double letters while others do, for example in the name /njEEri/, in the words mbeca (money), beca (money) and kamupeni (campaign). It is noted that the word ‘kamupeni’ is borrowed and that the phonemic inventory of Gĩkũyũ does not have the sound [p]. Phonemic /nd/ is represented in conventional orthography with , Basically a sound is represented by the same grapheme in all its occurrences, but there are a few cases of variation. The bilabial fricative /Ф/, for example, is pronounced variably as /v/ /f/ /p/. The orthography developers used the grapheme myanda (gulleys) /mĩanda (portions) kyama (party/miracle) /kĩama (a kind of a tree) kyendi (favouritism) /kĩendi (one that goes) twende (let’s love) /tũende (let’s go) kũtw’a (to decide/determine) /kũtũa (to settle) Finally, because of this complexity, it is difficult to transfer literacy skills from another language to Kĩkamba. 56 the harmonization and standardization of kenyan languages Harmonizing the orthographies of Gĩkũyũ and Kĩkamba Harmonizing orthography requires systemic planning. Chanda (2002:30) identifies four kinds of orthography planning: orthography development, orthography reform, orthography standardization and orthography harmonization. The first three are relevant to one language but the last one is relevant to more than one language. Orthography development involves provision of a writing system for an unwritten language. Orthographic development can however be applied to a language that already has a writing system, especially when the existing system needs to be replaced with a new one. In orthographic reform, the writing system is modified especially by modernizing pronunciation and other aspects of grammar and morphology. There may be a re- codification of a language to account for new specialized terminologies of science and technology. Orthographic standardization on its part aims to provide a standard orthography through identification of inconsistencies, which are then purged. Chanda (2002), for example, reports on the standardization of orthography of seven Zambian regional official languages in 1977. Harmonization of orthography is seen in two senses. The first has to do with choices of graphemes and spelling, which are made in such a way that in all the languages concerned the same grapheme and the same spelling rules apply to the same elements or similar grammatical structures unless there is a specific reason not to do so. The second sense relates to a situation in which the languages do not have a writing system and are provided with a body of graphemes and spelling rules. During the harmonization process, the languages concerned are prescribed the same graphemes and the same spelling rules for the same sound and same elements of similar grammatical structures. There is a lot to use from this systematic approach in unifying the orthographies of Gĩkũyũ and Kĩkamba. In planning the harmonization of the orthography of Gĩkũyũ and Kĩkamba, there has to be a detailed description of the phonologies and the existing orthographies in the two languages. The morphophonemic and grammatical systems will need to be compared and contrasted to identify the similar sound segments which can be represented by the same grapheme in the two languages. The phonological processes will GĨkŨyŨ and KĨkamba 57 also need to be described, especially because morphology and phonology interact at the deep structure level to produce specific surface structures. Such a detailed study of grammar and phonology in both languages will guide the planners of harmonization to deal with any inconsistencies as they work out a unified orthography of the two languages. As the description of the phonologies and orthographies of Gĩkũyũ and Kĩkamba presented in this chapter shows, the two languages are more similar than different. For example, the seven vocalic sounds [a], [E], [i], [e], [O], [u] and [o] occur in the two languages, and are articulated at roughly the same levels within the resonance chamber, as Figures 1 and 2 show. Similarly, there is minimal variation in terms of consonant inventories. The marked differences are the presence of [r], [h] and [ɣ] in Gĩkũyũ and their absence in the Kĩkamba orthography. It is important though to note that [r] and [ɣ] sounds are present in some dialects of Kĩkamba and thus the presence of a grapheme in the harmonized orthography will facilitate literacy in such dialects. Kĩkamba also has the graphemes Table 13: Representation of similar sounds in orthographies of Gĩkũyũ and Kĩkamba Speech sound Gĩkũyũ grapheme Kĩkamba grapheme 1. /Ф/ b v 2. /ʃ/ c sy 3. /ɲʤ/ nj ngy 4. /tʃ/ c/ch ky To accomplish a successful harmonization of orthography in Gĩkũyũ and Kĩkamba, adhering to the following principles, adopted from Banda (2003), will be important. First, there must be consistency, with one grapheme representing one phoneme. This is in line with Nyombe’s (2002) assertion that the same grapheme should be used to represent one and the same sound in the two languages as much as is practical. Combinations of letters should only be used where no single letter currently exists to represent that particular sound. Further, the use of different graphemes for the same sound should be clearly spelt out. Second, accuracy should be ensured taking into account all the significant sounds in each language. In this connection, economy and precision should be ensured in the use of a grapheme by getting rid of redundant elements. Third, the same characters in both languages should represent the same sound. The descriptive account of the phonologies and orthographies of Gĩkũyũ and Kĩkamba shows similar phonological and orthographical systems and this makes it possible to adopt similar graphemes and spelling rules across the two languages. Benefits of harmonizing THE orthographies of Gĩkũyũ and Kĩkamba Harmonization does not entail the devaluation of mutually intelligible languages. It is a process that targets the literacy unification of a cluster of related languages such as the two languages discussed above. As Bokang (2002) notes, it is possible to have a unified writing system achieved by way of prescribing common alphabets, common spelling, common GĨkŨyŨ and KĨkamba 59 punctuation and a common word division system. This is particularly important because the spirit of language harmonization and unification should always be that of cooperation and not of conquest. Prah (2003) argues that African languages that are mutually intelligible can be clustered together and their orthographies harmonised. As the presentation above has shown, Gĩkũyũ and Kĩkamba fall into the category of very closely related languages. Guthrie (1967) classified both in the Bantu Zone E language group 50 in recognition of this genetic relatedness. In addition to this, historical interaction among speakers of the two languages makes it easy for an individual to follow and understand speech in either of the languages. This high degree of inter- comprehension between Gĩkũyũ and Kĩkamba can be enhanced greatly by the harmonization of their orthography. Banda (2003: 46) argues that properly designed orthographies can play a monumental role in promoting the use of African languages in all spheres of life, and hence contribute to the socio-economic development of Africans. The same argument can hold for the harmonization of orthographies of languages that share linguistic features, as the two languages described in this chapter. The harmonization of the orthographies of related African languages will enhance inter-ethnic communication not only in spoken language as at present but also in written communication. A number of benefits would be reaped from this harmonization of orthography. First, a harmonized orthography would make the production of literacy materials more cost effective because the same materials would not need to be written in two different orthographies, and bilingual speakers of the languages will easily develop the texts since the graphemes will be common. One of the aims of harmonizing orthography is to promote literacy for socio-economic development (Banda, 2003). Kĩkamba and Gĩkũyũ are both used as media of instruction in lower primary classes. The production of such materials will thus be made easier than it is at present. Secondly, harmonization of the orthography will make the process of translation from one language to the other much easier. The two languages have very many common lexical items and the cultures are quite similar. The harmonization of the orthography will not only lead to ease in reading what is written in the other language, but it will facilitate 60 the harmonization and standardization of kenyan languages translation. This will in particular enrich the literature in Kĩkamba because there is much more written in Gĩkũyũ. The exchange of written materials across the two languages, and the sharing of experiences throughout the written or translated texts will end up developing and enriching the lexis of the two languages (Chanda 2002). Related to translation is the transfer of literacy skills. After harmonization, people who develop literacy skills in one of the two languages will not need to go to school to learn how to read and write the other language. As long as they speak and understand the spoken language they will be able to read and write the other language. Writers will thus be guaranteed a larger audience for their texts. The number of readers will grow and the literacy levels of the speakers of the two languages will be enhanced. Harmonization will further enable knowledge contained in non- African languages to be made available to a wider population by using African languages (Roy-Campbell 2006), because what is translated into Kĩkamba or into Gĩkũyũ will be available to a wider audience. Having common writing conventions will facilitate reading and writing across the two languages, which will in turn make it easier to teach cross-linguistically. Such a direction would be beneficial to speakers of Kĩkamba and Gĩkũyũ. African languages and cultures are often rich with knowledge and patterns of behaviour which can be promoted through enhanced communication. Finally, the harmonization of the orthography will strengthen further the good relations that exist between the speakers of the two languages. Ease of communication establishes bonds between the communicators. The understanding between the users of the two languages will be strengthened by the respect and feeling of oneness that comes with the realization that “I can understand what you say and I can read and understand what you write”. Challenges and way forward The actual work of drawing a common orthography is not difficult because, as we have shown above, the sound systems of the two languages are very close. However, ensuring that the changes achieve the set objectives is another thing. According to Smalley (1964:34) four principles are involved in making decisions about orthographic changes: GĨkŨyŨ and KĨkamba 61 maximum acceptance, maximum representation of speech, maximum ease of reproduction, and maximum transfer. The changes proposed in the chapter are not drastic and so the key principle as we see it will be the maximum motivation principle. This refers to the acceptance by the users and by significant groups such as policy makers and policy implementers. If the change does not connect issues that motivate and shape the direction of change in these circles, then it could become an uphill task. There is no doubt that the harmonization of the orthographies of Gĩkũyũ and Kĩkamba will make cross-ethnic literacy easier. However, the proposed change has a number of implications which may negatively influence their acceptance. First, it will involve re-educating people who are already literate in the current orthography. Though recognising the benefits, such people may reject the change for the simple reason that they are already comfortable in the current difficulties. The motivation of policy makers and implementers is likely to be affected by the financial implications involved. The revision and reprinting of the literature already available in the current orthography will require money, as will in-service training programmes for teachers, writers and editors. The challenge will be to get these groups motivated enough to support the move. Another factor that may affect the acceptance of a harmonized orthography is ethnic identity. The fear of losing ethnic identity or fear of ethnic dominance may blind people to the benefits of a harmonized orthography. The target population would need to be assured that their oral languages will always be distinct and that the focus of the change is only on standardization of the written forms; that each language will retain its uniqueness. Countering such challenges requires funding and working with language experts. Finally, political considerations may also interfere with a move to harmonize the orthography. In a country like Kenya where political alliances are built on ethnic platforms, a move that vouches for greater communication across ethnic boundaries may be viewed with suspicion and denied the political good will. Thus, it may take a long time before an acceptable compromise is reached but with the pace of technological advancement, harmonization of orthographies of the two languages may become a reality soon. 62 the harmonization and standardization of kenyan languages Conclusion Many benefits are likely to accrue if related languages are to standardize and harmonize their orthographies. In this connection, Bantu languages should aim at unifying their written forms so that they attain an orthography that allows transfer of literacy skills from one language to another. Such a reform will lead to a stronger identity of Bantu languages. If a harmonized orthography is well mastered and correctly applied, it will give great momentum to the development and revitalization of African languages. Needless to say, if this harmonization is spearheaded by Africans themselves, it will be an essential component of the effort to empower African languages. As Roy-Campbell (2006) rightly argues, in the history of African languages, Europeans were the ‘authorities’ developing orthographies, and it is high time that Africans addressed their own languages in their own right. Success stories of already harmonized orthographies can impact significantly on the quest for unifying orthographies of African languages. Chinese, for example, has over 400 dialects, some of which are not mutually intelligible (Roy-Campbell 2006); nevertheless, these spoken varieties of Chinese share a common written language, which is used in education, the media, computer technology, and so on. If the move to unify related African language clusters acquires momentum, perhaps the whole Bantu language family will have only one unified orthography in future. References Armstrong, L.E. 1967. The Phonetic and Tonal Structure of Kikuyu. London: Dawsons of Pall Mall. Banda, F. (ed.), 2002. Language across Borders: The Harmonisation and Standardisation of Orthographic Conventions of Bantu Languages Within and Across the Borders of Malawi and Zambia. Cape Town: CASAS. Bokang, I. 2002. Standard in Setswana in Botswana. Unpublished MA thesis. Pretoria: University of Pretoria. Carr, P. 1993. Phonology. London: Macmillan. Chanda, V. 2002. Orthography Planning across Languages and Countries: Some Thoughts and Proposals. In: F. Banda (ed.), Language Across Borders, pp. 27-50. Cape Town: CASAS Ford, K. 1976. Tone in Kikamba and the Central Kenya languages. African Linguistics, 7(3)261-293. Guthrie, 1967. Classification of Bantu Languages. London: Dawsons of Pall Mall. Heine, B. 1978. A Typological of African Languages: Berlin: Dietrich Reimer Verlag. GĨkŨyŨ and KĨkamba 63 Kioko, A.N. 2002. A Case for Improved Orthography. In: F.R. Owino (ed.), Speaking African: African Languages for Education and Development, pp. 231-241. Cape Town: CASAS. ––––– Theoretical Issues in the Grammar of Kikamba: A Bantu Language. Munich: Lincom Europa. Kuria, P.M. 2005. An Examination of the Discrepancies between the Graphemic and the Phonological Structure of Gĩkũyũ. Unpublished MA thesis. Kenyatta University. Massamba, D. 1986. Phonological Theory: History and Development. Dar-es-Salaam: Dar-es Salaam University Press. Maundu, P.M. 1980. Reconstruction of Kikamba Consonantal Sounds. Unpublished MA dissertation. Nairobi: University of Nairobi. Mwihaki A.N. 1998. Loanword Nativization: A Generative View of the Phonological Adaptation of Gĩkũyũ Loanwords. Unpublished PhD thesis. Nairobi: Kenyatta University. Njoroge, M. 2006. Linguistic Variation in Spoken English as used by Teachers in Kenyan Primary Schools. Unpublished PhD thesis. Nairobi: Kenyatta University. Nyombe, B. 2002. Unified Orthography for Nilotic Languages. In: K.K. Prah (ed.), Writing African, pp. 25-42. Cape Town: CASAS. Omondi, L.N. 1982. The Major Syntactic Structures in Dholuo. Berlin: Dietrich Reimer Verlag. Prah, K.K. 2003. Harmonising and Standardising African Languages for Scientific and Technological Development: The CASAS Experience. Paper presented to the Icelandic Language Council: Conference on Language Policy and Standardization, Reykjavik, Iceland, 4 October 2003. Read, C. 1983. Orthography. In: M. Martlew (ed.), The Psychology of Written Language: Developmental and Educational Perspectives, pp 143-162. Chichester: John Wiley & Sons. Republic of Kenya. 1999. Totally Integrated Quality Education and Training: Report of the Commission of Inquiry into the Education System of Kenya. Nairobi: Government Printers. Roy-Campbell, Z. 2006. The State of African Language and Global Language Politics: Empowering African Languages in the Era of Globalization. In: A. Olaoba and M. Pemberton (eds.), Proceedings of the 36th Annual Conference on African Linguistics, pp 1-13. Somerville: Cascadilla. Sampson, G. 1985. Writing Systems. London: Hutchinson. Shroeder, L. 2008. Bantu Orthography Manual. Dallas, Texas: SIL International (SIL e-books). Smalley, W.A. 1964. How shall I Write this Language? In: Orthography Studies: Articles on New Writing System. London: Oxford University Press. Wanjau, G. 1991. Mwandikire wa GiGĩkũyũ Karin’ga: Ibuku ria Keeri. Karatina: Gakaara Press Ltd. Wa Thion’go, N. 2004. Murogi wa Kagogo. Nairobi: East African Educational Publishers Ltd. Welmers, W.E. 1973. African Language Structures. Los Angeles: University of California Press. in the words ‘ndiricha’, ndirisha’ and ‘ndirisa’ (window). The phoneme has allophones [s], [tʃ] and [ʃ] in the dialects of Gĩkũyũ. Consequently, the writers represent what they have in their perception and not what is available in the orthography. The bilabial fricative /Ф/ has its conventional orthography as . The phoneme is closely allied to /p/ a bilabial plosive and /v/ and /f/ both labial dentals. Writers represent the phoneme with