RESEARCH REVIEW (NS) VOL. 12, NOS1 ft 2 (1996) KASEM NOMINAL GENDERS AND NAMES AiLAwedoba JatndmsOom This paper aims at studying the collocational patterns involving Kasem pasonal names and gender detennineis. Such collocations are not common in languages although they may not necessarily be ungrammatical. In fact, they constitute special usages which facilitate and en- hance coninBmicatkmm some languages. Kasem exen^Mes this. The use of the redupUcatcd form of the Akan determiner a© in name utterances such as Akwesl aono (My dear Akwcsi) may be cited in mis regard. While scholarship on Indo-European languages has concerned itself with usages soch as these it does not appear mat Africanists have shared these interests; in any case not to the same degree. InthispaperlwillattemptanoutlinedescriptionofKasena names in the context of Kasem gender syntax. These are issues mat have received scanty attention, as far as I can recollect1 Kasem, like neighbouring Gur languages such as Buli (see Kroger 1992 and Naden 1988) and Gnnmni (see Rapp 1965), is a language with nominal classes. In Kasem these are marked syntactically, prinripauyby the sfanple determiners of CVM shapes (C stands for consonant, V for polar vowel2 and M for bilabial nasal). There is a set of nine determiners from which most nominal items, particularly noons, select appropriate determiners to match flieir grammatical number (singular andpraral choices) and the class to which the noniinalbcloiigs. Grammatical concord is found between nominals and determiners and among noniinalssachaspronominals, deictics, myf mmuyak exhihhing a syntagmatic relationship. However, it is not usual in Kasem for Barnes, which can also be considered as nominals, to have determiners appended to mem. When a name collocates with a detennmer mat name takes on board some of the characteris- tics of the gender to which the determiner usually belongs. This men becomes a subtle means fat the expression of attitudes, a feature mis paper seeks to explore. Kaaem Genders and Classes The term 'gender' has several meanings; here it is employed together with 'class'as a linguis- tic category. Hockett (1958:231) long ago described genders as ".. classes of nouns reflected in the behaviour of associated words". Coibett (1991:1) remarks mat "the dassfficatkm [of geo- ders] frequently corresponds to a real world distinction of sex, at least in part, but often too it does not." In Kasem it does not refer to biological gender or sex distinctions such as mascu- line and feminine. For example, the individuals designated by the nouns kaaae (woman) and nouns in question belong in the same class and gender and mat gender is neither feminine nor masculine. Similarly, the persons designated by the nouns bakeira(boy) and busankana (girl) differ in biological gender but in Kasem grammar they belong in a common class and gender which happens to be different from the gender to which baaro and kaane belong. Gender and class are co-referential terms in this usage. Although some writers usually prefer the term 'class' in discussions of similar subjects in languages I find it necessary here to retain both terms. Nouns or nominate belong in classes and these in turn belong in genders. See Callow (1966) for the morphological analysis of Kasem noun classes. A class (in Awcdoba 1979) is a grouping of nouns or nominals on syntactic grounds which exhibit common morphological and syntactic patterns such as suffix and prefix selections as well as choice of determiners. Some pronominal anaphorics will also be chosen on the basis of class. Most nominals, including determiners and pronominal anaphorics, make number dis- tinctions often by means of affixes. A class may be singular or plural and their syntactic, morphological, and semantic features distinguish classes. Kasem has nine nominal classes four of which are singular and the rest plural. Certain singu- lar classes form regular pairings with certain plural classes and it is these pairings that have been designated "genders" in this study. See also Sterk (1978) for a definition of Hohumono noun classes and genders along similar lines. In most cases a nominal will have its singular form in one class and its plural class in another class and these two classes would usually be the pairing that constitutes the gender to which that nominal usually belongs. There are five regular pairings or "genders". Each class has its determiner and each gender is illustrated, as in the table below, by citing its pair of determiners. A noun does not have to collocate with a determiner, but when it does take a determiner it becomes definite. The litmus test for the gender or class membership of a noun lies in the determiners that the item in question regularly collocates with. Thus we know that a name is being assigned to a certain gender when in usage a particular determiner is attached to it. This is inspite of the fact that Ms collocation is not a common occurrence. TABLE OF NOMINAL GENDERS AND CLASSES GENDER SINGULAR PLURAL 1 2 3 4 5 Nominal Class Determiner Nominal Class Determiner I m V VH vn worn [worn] dem [dim] kam [kam] kom|kum] kom [kum] n IV VI vin DC bam [bam] yam [yam] sem[sun] tern [tun] dem [dim] Genders and Dimensions Kasem genders are not a haphazard collection of nouns; yet the semantic bases of the genders are not always easily demonstrated and this is due to the miscellany of items found in any gender, see Prost (1970: 984-995) for a listing of the nouns that regularly belong in the gen- ders. The rules that govern or influence the allocation of nouns to genders have not been established completely for Kasem, to my knowledge. To appreciate the significance of the collocations that personal names enter into with gender determiners there is the need to under- stand the kinds of semantic associations that are attached to the genders. The most semantically determinate of the genders is Gender 1 which we may describe as the personal or human gender; human nouns predominate in it. Over 90% of human nouns belong in this gender. Even then, non-human items such as dwooru 'axe', gweeru 'leopard', duuru 'vulture', and sana 'millet beer' occur in the gender. Other genders are even more heteroge- neous in their membership. Gender 2 contains nouns for fruits, body parts, abstract and deverbative nouns and many more that cannot easily be classified on the above dimensions. The same can be said for Gender 3 which, like Gender 2, has diminutive associations. It in- cludes in its membership groups of items such as young plants, small mammals and insects, some body parts as well as nouns for human beings associated with diminutive stature or some negative or pejorative attribute or feature. In comparison with Gender 3, Gender 4 has aug- mentative associations. It includes many trees, animals, illnesses, as well as nouns that refer to people with some potentially negative attribute. Gender 5 has a smaller membership which includes domestic livestock. Its membership is less than 20 items. In fact the Kayoro dialect of Kasem has re-assigned the membership of this class to Gender4 If we may talk of attributes, the most salient semantic attributes of Kasem genders would include humanness, diminutiveness and augmentativeness. There appears to be an emic aware- ness of this in individual idiolects and it explains idiosyncratic derivations of diminutive or augmentative forms. This accounts for the neologism swora 'particularly tasty and mucilagenous stew' (Gender 3 noun) which was derived from sworo, (Gender 4), the more generic term for mucilagenous or okro-like stew. Adjectives which on the whole have the potential to occur in more than one gender illustrate fairly well the semantic character of the genders. When an adjective is assigned to a gender it usually acquires a new connotation that is associated with that gender. Certain noun stems that can enter more than one gender tend to change connotations with gender change. These changes are obviously not random. The point is illustrated by the nominal stem: kasen-; it occurs in Gender 1 to mean 'Kasena person'. When it occurs in Gender 2 it means Kasem language or culture, and as a Gender 3 noun it refers to a diminutive or debased Kasena person or attribute. Finally as a Gender 4 noun it refers to 'Kasenaland'. While the semantic bases of the genders cannot be denied, phonology does also play a role in the allocation of nouns to classes and genders. This is so in the case of borrowed nouns (Awedoba 1980). The final syllable of a noun, particularly the weight of the final vowel, plays a critical 10 role in the allocation of nouns to their genders3- Even complete homophones like buga [buge], Gender 3 'river' and buga [bugs], Gender 2 'tiredness' are assigned to different classes and genders on account of their suffixes which happen to bear a phonetic resemblance. In their underlying representations these vowel suffixes differ in weight hence the allocation of these two words to different noun classes and genders. On the Personal Name A name, according to Fortes(1955:349), is 'a document epitomizing personal experiences, historical happenings, attitudes to life, and cultural ideas and values'. Fortes was defining primarily Tallensi names (a nearby Voltaic people similar in many ways to the Kasena), nev- ertheless his definition aptly summarizes the nature and character of personal names among the Kasena and perhaps among other Upper-Easterners as well. Kasena personal names (yere singular and yera in plural) exhibit a varied structure and are amenable to several kinds of classifications which cannot be discussed exhaustively within the scope of this paper. There are formal or official names (what have sometimes been referred to as birth names), nicknames, as well as elliptical and reduced names. We may describe as official names those that have been allocated formally to the individual at birth or soon after (yeretete in Kasem). It is such names that refer to and are used to address a person in any formal context. Every Kasena person will have one such name and some individuals have additional names. Reduced names in Kasem (no lexeme exists in Kasem for this category of name) are derived from the formal names and are comparatively shorter, reduced names result from incorrect or affected pronunciation of the names of senior kin and those of the parental status by junior kin, especially infants. The mispronunciation may be deliberate but it often originates with toddlers in their attempts to call relatives. Such lisped or mispronounced names are not usually corrected; on the contrary children are encouraged to continue in their usage. Adults accent by repeating the mispronounced names when they talk to children about the individuals whose names the children tend to mispronounce. It is noteworthy that some of the initial education that Kasena children receive is about politeness in addressing their elders, especially persons of the parental generation, whose names should not be mentioned fully without justification. This is reminiscent of the Second Commandment in the Bible. Thus a name mat initially was a mispronunciation becomes concretised in a reduced name through continued usage. However where a reduced name already exists it is the older name that is likely to be retained by children and younger kin as the family encourage children to switch over to using such a name. A person, it would seem, becomes eligible for a reduced name when he or she becomes a parent; in mis respect tt becomes willy-nilly a status symbol. Firstborns are often me origina- tors of reduced names although I have known younger siblings to coin and perpetuate such names. The application of a reduced name symbolises respect and deference. Its coinage and continued use is reinforced by norms such as the prohibition enjoined on the junior generation to mention the names of senior generation persons without reasonable cause. The breach of U this norm could be interpreted as a sign of disrespect and likely to invite some form of penal sanction. The use of the reduced name is also a sign of close kinship. The alternative to this usage, for non-kin of lower status, is the application of addressive relationship terms. One consequence of the ^plication of these norms is the ignorance of the real names of parents on the part of young children. It also explains the reluctance of some children to disdose the names of their parents to outsiders4. See Kropp Dakota (unpublished typescript) for similar attitudes and conventions among the Dagomba. If formal names originate with the senior generations and reduced names with the junior gen- eration, nicknames5 (bateim or yerebaseiai) differ again in that, they originate with the peer group and tend to circulate among a group of individuals who joke with one another. People tend to acquire their nicknames later in life. Friends and colleagues may coin mem but indi- viduals have been known to select a name for themselves in response to peer group influence6 Whatever the type, most Kasena nicknames resemble the Penang 'friendship names' discussed in Needham (1971). Penang of Borneo, selected "friendship names" and used them to affirm and signify friendly relations in a way reminiscent of Kasena practices. Some Kasena nick- names have been known to eventually replace birth names; they certainly have the potential to become more popular than some birth names. The Morphology of Some Personal Names Kasena themselves generally hold that names are meaningful (yere jege kuri) and a name would not be a name if it did[not have a meaning that can be interpreted. For some names the meaning is transparent, yet for others they are opaque with a-double entendre. Name exegesis is a pertinent exercise among Kasena for whom this involves a paraphrase of the name utter- ance. When a name is apt and very suitable, it is said to be "sweet". Their attitude to names is comparable to their love for proverbs. Kasena names, most of all reduced names, differ from other nominals of the language. Struc- turally reduced names in Kasem can best be considered as unanalysable sequences since they cannot be reduced to conventional grammatical or lexical units. Other name types can be so analysed and therefore subjecttofolkitterpretaiion. Names like AgM>,Anao,Kftd0a,KacluHu etc are typical Kasena names which we can assign to the word unit7 Morphemically they contain a prefix and a noun root We can find analogous structures in the language. For ex- ample, the nouns kaduga (farm), kamogo (day artifact), kazsga (head pad), kazogo (mortar) are comparable since they too comprise a prefix and a root or stem. The two sets of words differ however. The prefix ka= attached to the noons is a formative affix which has function in the derivation of the noun from a verb root hi tins case the verbs are as follows: do/dugm(to sow crops), mo/moga (to draw or fashion out), zeae (to cany) and m (to pound). Ka= plays a similar role hi other nominalisation processes in Kasem, for example, ka ta kackaare [ka say 12 nonsense] (to say nonsense/ In names oa me other hand, the prefixes A- and Ka- are sex indicators which are attached to nouns, not verbs, in name derivation. In names like Agao, Anao.Kadoa and Kachana a noun is the base of the derivation. Secondly, it will be realised that nouns are morphologically more complex than names even when the personal name hap- pens to be derived from the noun. Kasena nouns have suffixes not to mention prefixes but personal names do not For example, kaduga [kcdug=A] and kazogo [kazog=U] have suffixes which mark their genders but the same cannot be said for the names above. If KKaduga* were a personal name comparable to the English "FiekT, it could only be said to have a pseodo- suffix (when we consider mat it is derived from the noun kaduga which has a legitimate suffix). What were prior to the derivation gender suffixes lose their status in the course of name derivation. Kasem nouns are analysable into stem and a suffix phis an optional prefix. Note that a suffix is more or less the same for each group of nouns belonging to the same gender. Names have no gender affixes. Names are different from not only noons but also other nominate in this respect Kasem nommals do exhibit gender affixes. While noons may collo- cate with determiners names do not usually do so. Names moreover do not usually make number distinctions in the way mat most nouns in Kasem do. Where names are pluralized the process is irregular and carries special significance, as is illustrated below at the conclusion of the paper. Names however resemble nouns in one respect: pronouns may be substituted for mem. Pro- nominal choices are restricted to the singular personal pronoun "o" as far as die name is concerned. In conclusion, we can say that Kasena personal names are genderless utterances, unlike most other nominals that can be assigned on morphological and syntactic grounds to one or more genders. Another feature of Kasena names is that most of mem, except for the reduced names, exhibit forms mat are derived from potential sentences. They may be regular statements, questions, commands or exclamations. Statements include the following: 1. Kodaadcyerane: la> dac a de yerane (It docs not apply to me alone) 2. Azeffiba: A zefliba(I have chased mem off). Examples of Questions: 3. Woarieflia: wo m t d ba (who owns them) 4. Atadewozaato w© tade wo (who should I complain to) Examples of Commands: 5. Yeke^pa: ye ke-n pa (don't give to others) 13 6. Yepafeonong: ye pa ko nwogi (don't let it out) Examples of Exclamations: Haya: (O dear) If grammatical analysis were applied to each of these utterances, constituent elements of clause structure such as subject, predicator, complement and so on'could be obtained. This is illus- trated below. 7) ko/dae/adeyerane it is not I conj. alone [Subj / Pred / Compl] 8) a/zeili/ba I chased them [Subj /Pred/Compl] If the above are simple one-clause sentences, note also that more complex name structures are found in the language as exhibited by Nakwebabaennabaniite, ('An elder should not bear tales and if you did, you would die'). This name is abbreviated sometimes as Nakwebabae ('An elder does not bear tales') sometimes as Nnabaente, ('if you bear tales you die') each of which is a potential sentence. This name constitutes a complex serial construction with a dependent serial clause and independent clause, as is illustrated below. 9) aakwe/ba/bae / / n / n a b ae Units elders don't bear(tales) you if bear tales you will die Since all names may be regarded as at least uni-lexemic, grounds would appear to exist for considering the more structurally complex names as rankshifted sentences or clauses, as the case may be. The concept of rankshifiing in systemic analysis accounts for the embedding of higher units, such as sentences clauses and phrases, in the structure of lower units or units of the same level. See Halliday (1961) for a discussion of the grammatical rankscale and the concept of rankshifting. Names aid Determiners The collocation of noun and determiner indicates not only the class and gender of a noun but also assigns definiteness to that noun. From another point of view, a noun differs from a name in that it refers to a class of individual items whereas a name refers to an individual and is ipso facto definite. Names do not therefore require that a determiner be appended to them to mark definiteness. When a determiner collocates with a name, rather than definiteness it is emotive meanings that are conveyed. These include personal attitudes and dispositions of the utterer towards the referred person. The emotive meanings involved which can be assigned to two categories are labelled below as Categories A and B. 14 Category A attitudes suggest either equality characterised by open and free relationship that is unrestrained by formality or inferiority of the person referred to. Category B attitudes signify mutual respect or even reverence for the other party named in the discourse. Whereas for Category A joviality is not out of place, for Category B restraint cannot be ruled out. These differences in attitude and formality find an echo in the application of pronouns of power differential and solidarity in some European languages, cf. the use of tu/vous hi French, (Brown and Gilman,1970 and (Robinson, 1972). In Kasem two determiners out of the lot may be used with personal names: the NCLI worn and NCL V kam determiners. Both of these are singular syntactic markers and denote genders 1 and 3 respectively, (see table above). These are the genders which, as noted above, have per- son/human and diminutive/negative associations respectively. The application of worn and kam enables the expression of Category B and A attitudes respectively. These are usages that are commonly found in Kasena greeting rituals (particularly the elabo- rate greetings), though the usage is not restricted to this genre. Kasem may be compared to some other languages in this respect We find similar communicative functions executed through greetings in other cultures and languages though the devices may differ. As Goody (1972;50) has pointed out, greetings in some West African societies involve communication and ex- changes of important information about the state of relationships and attitudes concerning not only kin, affines or friends but also the political structure. Courtesy rituals in Gonja society incorporate title references in their greetings and this is in consonance with their hierarchical social structure. Anwng the matrilinealAshand where there is a concern for ntoro or patrilineal group identification clan appellations feature in greetings (cf. Busia, 1954:199 and Fortes, 1950:267). Kasena greetings do not utilize titles as this society lacks them nor are clan references necessary since clans are rarely important reference points for social action; lin- eages which serve this function are smaller and localized. Nevertheless there are certain inter- personal relations that require to be signalled for what they are worth and the use of determiner collocations hi Kasem enables this to be done. Person to person attitudes are complex and may be expressed in a variety of ways, not all of which have to be by verbal devices. Name references are however important indicators of these altitudes. The use of reduced names as we have seen suggest not only mat the individual being addressed or mentioned is older chronologically but also mat he or she is kin or qualifies for the quasi-kin category and is of an older generation than the. speaker. Although a person of the senior generation may address a person of the junior generation by name, mis is not recip- rocal. Formality and restraint characterize inter-generation relations among Kasena. The same is incidentally true for most other Upper-Eastern Ghanaian people. Where the child-parent tie is concerned mere is even a degree of avoidance which is epitomized by the case of the first born son and the father, as Fortes(1959) has demonstrated cogently. Name avoidance can be 15 seen as an aspect of this. It must be pointed out that Upper-Easterners do not quite approximate Southern African peoples like the Sotho on this question. For the latter it is not enough that the forbidden name is not mentioned, its homophones mid semantically related words should be avoided or replaced by "hlonepha" and teknonyms,( cf. Laydevant,1948, Kunene,1958 and Kahler-Meyer,1971). Here too, as in the case of the Kasena, we see social norms conditioning linguistic deviation. Individuals of the same generation may refer to each other by birth name or nickname if they are of the same peer group. Wives may, because they enjoy a joking relation with the husband's junior agnates and his peer group, use the nicknames circulating among members of these groups. Between me unrelated wives of agnates mis kind of freedom does not exist and it is common to hear teknonyms used or the application of formulae such as 'mother of so and so'. The use of determiners with personal names thus becomes a resource for (he conveyance of additional attitudes. The following utterances illustrate the effects of determiner collocations: 10a) Adoa kam jege yezura aa ? (How is Adoa?) Adoa-detm-has- health -quest 10b) Adoa worn jege yezura aa ? (How is Adoa?) Adoa - detl - has -health -quest 10c) Adoa jege yezura aa ? (How is Adoa?) Adoa- has- health - quest To most native speakers of Kasem these three sentences are unlikely to be synonymous. Ex- ample 10c is unmarked; it poses a question without necessarily dropping a clue about the speaker's feelings towards the individual referred to. hi 10b), the speaker, by the inclusion of the Gender 1 determiner, signals his concern for the third person for whom the speaker may have profound respect, although he or she need not be of the senior generation. Example 10a also makes manifest the attitudes towards the referent In the latter case, these however are attitudes of ease and they suggest a degree of freedom that would allow for joking and even horseplay. 10a could also be an attempt to display sympathy for the person mentioned: Adoa. The sentence can in feet be paraphrased as "How is Adoa's health, poor flung". IHscuMioa asd Coadaakn We may wonder why out of nine determiners only wo« and km should be selected for this usage. The explanation lies in me semantie values of me genders to which these two determin- ers are syntactic m a r k ^m the singular. The attache means of incorporating ft into the gender to which the detennmernormaUybdoogs. This has the effect of loading the name with some of me peculiar attributes of the gender mquestioatt 16 thus becomes a device for displaying the speaker's assessment of the named and the inter- personal relationship between them. The value of that relationship need not be a reciprocal or a symmetrical one. Genders 1 and 3 are being put to mis use because these, together with Gender 4, are the most explicit semantically and their associations happen to be appropriate. If genders were to be paired contrastively, then Genders 1 and 3 could more naturally constitute one such pair by reason of the feet flat they contain the majority of human nouns between than. A pairing of Genders 3 and 4 enables me augnientatrve/dimmutrve contrast to be highlighted. When a noun belonging in one of these genders is introduced into the other, the remttag noon becomes the diminutive or augmentative of the original noun as the casemayWHK language exploits this with, some consistency. For example, cfcfcMra (dwarf-like bush spirit), a Gender 3 noun, and ctfcMni(ogrcorgroteftiMespiiit),aGfc»dcr4ao^lweaooaanonstM the Oaves Vaad VH mffixes respectively the different gorieraouas have been derived with Gender 1 as the specifically human gender is most suitable fix-polite attitudes, being the most positively valued gender in the language. Gender 3 too is suitable for the signalling of jocular relations because it suggests the attribute "lack of significance or weight" i.e. diminutiveness. Kasem is perhaps not unusual in this respect As Wolfowitz (1991:64) has pointed out, the use of diminutives in interlocution may represent a general case of what she calls 'downward- directed deference or respect'. In conclusion, the application of the Gender 1 determiner to a personal name indicates honor- able assignment of that name and its referent to the most prestigious gender, this suggests a perception of definitely positive human, as opposed to non-human, qualities hi the individual. It symbolises that me person referred to deserves respect Idiomatic expressions lend weight to this assertion. In Kasem, to say of a person mat o ye aoono (Xls a person ), would amount to saying that s/he is a good person. To describe a person as o dae noono (X is not a person) is to indicate that the individual concerned is a bad lot or is deficient in desirable human qualities. All this implies that humanness is a quality that can be allocated socially or withheld and therefore becomes an alienable quality to the human being. Just as the incorporation of a name into the human gender affirms personness, assigning an individual to Gender 3 detracts fiom the named individual's personness. This need not strike us as paradoxical; Fortes(1987) indicates that person status is not ascribed but must be earned or achieved. He goes on to specify Talleasi criteria for the postmortem achievement of person status. A Gender 3 classification amounts to a mild insult that does not necessarily give of- fence, after all this is a gender that also has some positive associations, especially when it is contrasted with Gender 4. In any case not all insults are resented, as we all know. Our interpre- tation of the offence hi an insult depends on where it is coming from, and me presence of malicious intent 17 Assigning a named person to Geadw 3 is of tire order of s c s ^^ This does not provoke hostility because ft assimilates to a category of behavkmrs resembling mock verbal duelling but whose aim is the assertion and maintenance of c«rtam Mendly relations between individuals and groups. The inappropriate application of this syntactic device can well raise eyebrows. • • • * • . • • •• •-•' • •; -:- ;" •' • • > Inedtortionalizied joking relatkms may be recalled to mtad, more especially as they are a coft* iaoo]^bedby aoeh ties bat tfusy slab aM between tiwK&top* gwop and ofeer <*ase*efe«ete«s derive from the pre-colonial predatory relations that obtained between the Zamberma slavers aBdtr«^j*Ba(l^stjb»Fn^Kj«!eaa; CHhef siavededert (flristoemdes Gonja) and thebvictiaB soctt'W th«K«0MK, Sisalaetc. SoeHokfc^l»5)f6r»4isca6sk)Dof theZan*enrainteive^^ tkmm what is today the Upper Regtom of (Siana". aa4 Analogy can be drawn between fee use of nick names, teknoayme, redsced astnes, idting relationships, appellations, titles and the application of determiners to pofisoBal names. All of these are amenable to a dual classification along the WOM/KAM dimension as set out In the table below. Jokes are often a mechanism that people who have suddenly been thrown together adopt m the wmt^ems^^iSadrscKMixoec&^kiBsl refetiotth^s m we find ameag the groups of raoM-eflWc ^RW^H" University studsats. A»oag rnese stodetits nidenanies are crarent,imdwhetlier Aey are the nan^tfe^tadwMaaklarve selected for th«nseh^ or those 4 t^ b h k d h d d t l i s l 5 at mm ti*e name conferred on at leetmor, all are of atype. WOM IMPLICATIONS KAM IMPLICATIONS 1,Nameswjihwflaiappeaded . ., 1.Na^eswaikamappended [+Addrcssive^rfercncc^mTaTns] h JokingBehaviowl ^ ,, ,. [-A4drcs«>-o^cfcraKC Kin Terms] .i[+j0kiB«8eha¥io«rlt [+ Avoidance] [-Avoidance] 18 Among those whose names may be mentioned with a Gender 3 determiner are some peer group members, kwo bia or members of paired joking groups, opposite sex affines of die spouse's generatkm, and penons of me grandchild gcaeratkm bat iwery names of the grand- parents12. Gender 1 determiner may go with the names of kin of the senior genention and affines of the tmbaaro and trakaane grade Le.'wife-givers', and those others who are en- titled to respect and deference fixwn the speaker. We may comr^this usage to the hxrapora- rfril1hrifiteP has shown, asymmetric usage of tide and last name suggests tmfiuuUiartty and inequality while mutual use of first name indicates the opposite; see also Brown and Gilman(1970). Cases can be found of personal names collocating with other determiners, particularly with ban, the Gender 1 plural determiner, or other plural dctenniners. These psage* have different Implications as hi the following utterances; , lib) Adwe sen daga Adoa det many (There are many people called Adoa) 1 la) Adoa ban daga. Adoa det many (Any Tom, Dick, or Harry may be called Adoa) In lla, the personal name Adoa has not been pluralized although the accompanying deter- miner is plural. Number concord has been violated yet the sentence is grammatical; and if a singular determiner had been involved the sequence would have been nngiauuuatkal. The application of a determiner in this case does not suggest any particular attitudes or meanings. This is an unmarked expression. Example lib exhibits bom number and gender concord al- though a name is involved here too. In the last example the name Ado* is being treated like an ordinary Kasena noun and assigned to the gender hi which the word doa (rain), from which the name is derived, belongs. This amounts to an exhibition of a characteristically depersonalized attitude to a name as a linqidstic hem as opposed to the name bearer. Some Kasena names like Adoa (Mr. Ram), as has been suggested above, have been derived from noons but example lib now illustrates how names may themselves be made into nouns and subjected to the ordinary rules of noun syntax. Kasem is perhaps not unusual in this respect aod similar examples can easily be cited from English usage as hi exemplified by utterances such as "She did a Ben JobnswT where "Ben Johnson" becomes an ordinary noun. 19 1. 2. 3. 4. 5. NOTES Callow (1966) was perhaps the first attempts* a systematic and d^ailcd study of the BKjiptotogy of Kascmnom classes. This writer owes a (fcfat of gratitude to Callow for his initial instruction on the Kasem language. Callow (op dt) generated much interest and accounted ftr a flurry of publications on the phonology of me Kasem nouns as illustrative of fte character of phonological roles. See Chomsky and HaDe (1968:358-364), Halle (1978), m^S(l9t^ aatf Awedoba (1984). The feature'polar3 enables high vowels: [u,u, i,l] and me low vowels: [a] and [a] to be grasped and distingnished from other -vowels in Kasem. This is useful in view of me distributional characteristics of these two sets of voweb-fa Kasem words. See Awedoba (1993) for the feature description of the vowels and die justification for the introduction of the polar feature. The final vowel which in this case may be light or heavy plays a role in the allocation ofnouns to genders.The final vowel in [bugs], Gender 2, is light but the one in [bugs], Gender 3 is a heavy vowel which has undergone lightening. See Awedoba (1993) for a further explanation of the weight feature in Kasem. We find this ignorance to be fairly prevalent in Ghanaian societies. It occasionally comes to public attention when a lost child is unable to name its parents and it there fore becomes a matter for the poflce to deal wim. We see advertisements concerning sach unfortunate eases in the Ghanaian Dailies quite often. Nktoianws, as the etytnotogy of the term basena of Terebaseim suggest^ pected to 'qualify* tent, or reflect certain attributes found in the name bearer or those that he or she lays claim to or wishes to emulate. In this respect nicknames may be positive or negative. The former could be names derived ftom some famous persons ia the society. The latter coaldbe based ottaa undesirable physical attribute or some peculiarity. Names like AymrKMr. Head), Mom** (Mr. Nose). Aid (Mr. Mouth), Ayi (Mr. Bye) etc. which refer to a part of fee hmnan anatomy are an example. Each of tliese is alluding to somediiHgmus«Btf about that nickname utterance. Negative nicknames may also be drawn from the names of indi viduals known to have some negative feature or characteristic and conf erred on those who are deemed to share or exhibit me attribute in question. These names are there- fore Insuhs wMdi are graceMly accepted whm used by tiwse enMed to use them. 6. This practice is illustrated in one folktale where Rabbit suggests to Hyena the impor- tance of finding memserves nicknames so as to avoid needless (»mpeiition between them at their in-law's compound. 20 7. 8. 9. 10. 11. 12. In Kasem systemic grammar, the word is ranked on the hierarchy of grammatical units between the phrase and die morpheme. Its structure is analysable in terms of morphemes but itself functions in the structure of the phrase. See Callow (1969) for a description of derivational processes in Kasem. The symbol / marks phrase boundary and // marks clause boundary, "pred" means predicator, "subj" means subject and "compl" stands complement; these are some of the dements of clause structure. The customary joking relation is referred to in the metaphor of siblingship and this is instructive. It confirms the point that those who enter into jocular types of relations see themselves as peers, or are trying to redefine a relationship and assert camarade rie. The connection between 'joking relations' and previous inter-ethnic hostilities has been suggested for other African peoples such as the Gogo of Tanzania, (cf. Rigby 1968). Writing about the West African dry Banton (1965:139) remarks that "Thus neighbouring peoples who are rivals and enemies in the rural districts become friends hi the city for they compare well with the incomprehensible people from other re- gions'. Though alternate generations maintain a jnking relatirrodiip and grandparent and grand child often joke with each other a grandchild does not refer to the grandparent by his or her full name. BIBLIOGRAPHY AwedobaAK 1979 Nominal Classes and Nominal Concord in Kasem M.A Thesis submitted to the Inst. of African Studies of the University of Ghana, Legon 1980 1984 1992 'Borrowed Nouns in Kasem l*>miaal Classes' Aathropologjcal Linguistics vol.22 no.6 p.247-53 'An Examination of the Phclps Analysis ofKasem Gender III'in Papers in Ghanaian linguistics no.4 eds: MEDakubu and Mensah 'Light and Heavy Vowels in Kasem' Afriluraistiscbe Arbetepapfcre 30, p. 135-153: Instirat fur Afiikanistik, Universttat zu Koln, Germany 21 1993 The PlKmetks and Phonology »fKajeia as Research Review Supplement 7:109 pages: Institute of African Studies, Univ. of Ghana, Lcgoa Banton, M 1965 'Social Alignment and Identity in a West African City' Urbuiza ikm and Migration in We«t Africa, California ed. Hilda Knper Brown R and GUman A 1970 "The Pronouns of Power and Solidarity' Readings in Ifae Sociology of Language ed. J A Fishman, Mouton BosiaKA 1954 'The Asharti of the GoM Coast'AfrfcanWorkU D.Fordeedpl90- 209IAI CallowJC 1965 'Kasen Nominals: A study in Analysis' JWAL 2,(1) P. 29-36 Callow JC 1965b Chomsky N and HaileM. 'Derivation in Kasem' Annales de Fuaiverritt d'Abidjan ser. H - Fascicule Hors Serie 9 Actes dn Hnitieme Congres International de Linguistique A&icaine) 1968 The Sound pattern of English, Harper and Row, New York. CoibettGG 1991 Gender, Cambridge, CUP Fortes M 1950 'Kinsh^andMarriageanrongtheAshanti'African Syrte«« of Kin skip and Marriage, eds. A.R.Radcliffe-Brown and D. Forde 1955 'Names among the Tallensi of the Gold Coast' Afrikanistkfce Studies, Deutsche Akad. d. Wissenschaften m Berlin, Institut £ Orieotforschung 26 p337-49 1959 Oedipus and Job in West African ReHgtoa. Cambridge, C.U.P 19S7 RefigfcwMoraBty and tte Person ed Jack Goody 1987 'The Concept of Person'in M. Fortes, Re&gkm Morality and the PersM p218-28 Goody, EN 1972 "Greeting", *1)eggmgf',ffl»duie presentation of respect' in He Interpretation of Ritual ed IS. LaFontaine, Tavistock Pdbl. Halle,E 1978 'Farther Notes on Kasem Nominate' in linguistic Analysis vol 4, no.2 22 Halliday, M A K 1961 'Categories of the Theory of Grammar1 WORD 17 p.241-292 Hockett, C F 1958 A Course in Modern Linguistics MacMillan, New York Holden, J J 1965 'The Zabarima Conquest of N-W Ghana.' Part I Transactions of the Historical Society of Ghana vol. VIII, p.60-86 Kahler-MeyerE 1971 'Niger-Congo, Eastern Bantu' Current Trends in Linguistics ed. Sebeok vol.7 Mouton Kleinewillinghofer, U 1995 Kropp-Dakubu, M E 'Don't use the name of my dead father, A reason for lexical change hi some Northwestern Adamawa languages (Northeastern Nigeria)' Afrika und Ubenee, Band 78, pl25-136 1983 'Personal Names of the Dagomba' Unpubl typescript in IAS Li brary, Legon Kruger, Franz 1992 Buli-English Dictionary Research on African Languages and Cul tures vol 1, Lit verlag, Munster Kunene, DP 1958 'Notes on Hlonepha Among the Southern Sotho' African Studies 17no.3p.l58-83 Laydevan,tF 1948 'La Coutume du Monepho Africa vol. 16 p.83-91 Naden,T 1988 "The Gur Languages' The Languages of Ghana ed. ME Kropp Dakubu, Kegan Paul International IAI p. 12-49 Needham,R 1971 'Penang Friendship Names' in The Translation of Culture ed.T.O. Beidelman Phelps, E 1975 'Simplicity Criteria in Generative Phonology: Kasem Nominate' in Linguistic analysis vol 1, no. 4 ProsLRPA 1970 "Deuxlanguesgourounsi: lekasemet lenuni" Buide 1'I.F.A.N. T. XXXII, ser. B, no 4 p975-1087 Rapp, E L 1%5 Die Gurenne-Sprache in NordGhana Leipzig, Veb. Verlag Enzylclopadie 23 Rigby, P 1968 'JokingRdadootirfpt,Kia Calories and Clandiq) Among the Gogo' in Aftic* 38 N6.2 p.133-155 Robinson, WP 1972 Lwguage aad Sodai Bdhariour Penguin Educational Books Stok, JP 1978 'Tbclfctomooo NdmdmSyitem' (kfimco) 13th. Wett African Languages Congrats, Freetowa, SenaLeooe Wolfowitz,C 1991 Laaguafe,Styka^SMMS^Mce:Styfctk CteiceiaSariMM Javanese, Chicago 24