Form and Function of Case Marking in Kambaata
urn:nbn:de:0009-10-3790
Zusammenfassung
Die kuschitische Sprache Kambaata zeichnet sich durch eine reiche nominale Morphologie aus. So unterscheidet sie acht verschiedene Kasus an Nomina und Pronomina: Akkusativ, Nominativ, Genitiv, Dativ, Ablativ, Instrumental-Komitativ-Perlativ, Lokativ I und II. Kasus wird nicht nur durch (segmentale) Suffixe, sondern auch durch eine bestimmte Position des Wortakzents (suprasegmental) gekennzeichnet. Der erste Teil des Artikels widmet sich diesen formalen Regeln der Kasusbildung und diskutiert im Detail die Deklinationen von Nomina und Eigennamen, die Kasusmarkierung an nominalen Modifikatoren und Personal-, Demonstrativ- und Interrogativpronomen. Adjektive, Zahlwörter und Demonstrative kongruieren mit ihren Kopfnomina in Genus und Kasus, ihr Kasussystem ist jedoch auf die Unterscheidung von Nominativ, Akkusativ und Oblique beschränkt. Der zweite Teil des Artikels thematisiert die grammatischen Funktionen und die Semantik jedes einzelnen Kasus anhand elizitierter Sätze, Daten aus Texten und spontanen Äußerungen. Kasus ist das wichtigste Instrument, um die syntaktischen Abhängigkeiten im Satz zu markieren, da Kambaata keine Adpositionen hat.
<1>
Kambaata is a Highland East Cushitic (HEC) language spoken by more than 600,000 speakers ( Gordon 2005 ) about 300 km southwest of Addis Abeba in the Kambaata-Xambaaro-Zone of the Southern Nations, Nationalities and Peoples Regional State, Ethiopia. It is mutually intelligible with its immediate relatives Alaaba and Qabeena. While the few works published on Kambaata so far more or less concentrate on the verbal morphology and on morpho-phonological processes (Abebe et al. 1985, M.G. Sim 1988, R.J. Sim 1988), little is known about the noun. Only sketchy information is found in Hudson (1976), Korhonen et al. (1986:108), and Sheleme (1989).
<2>
According to Blake (1994:1), “[c]ase is a system of marking dependent nouns for the type of relationship they bear to their heads.” In Kambaata the head is either a verb, noun or adjective. There are no adpositions. Kambaata is a strict verb-final language; modifiers precede the head. Example (1) presents a first impression of the structure of a Kambaata sentence. [1] The nominative encodes the subject, the accusative the object (semantic role: patient), the ablative an oblique object (semantic role: source). The genitive encodes a noun modifying another noun.
1) |
Án |
qoxíichch |
boqqollí |
zerettá |
dirr-iishsh-óomm . |
1 sg.nom |
storage:place. m.abl |
maize. m.gen |
seed. m . acc |
lower- caus1(+pal+gem)-1sg.pvo |
|
I took down the maize seeds from the storage place (above the cattle). |
<3>
The present article aims, firstly, at analysing the formal means of case marking in Kambaata (see section 2), presenting the different declensions and the formation rules for different case forms, and discussing syncretism. The importance of accent for case marking is furthermore demonstrated. A major distinction is made between case marking on nominal heads and nominal modifiers. Secondly, the grammatical and semantic functions of each exponent of case are investigated (see section 3).
<4>
The Kambaata noun (fig. 1) consists of a root, optional derivational morphemes and a case marker. Root and derivational morphemes make up the nominal stem. The root is never used in isolation; it requires a case marker. Several optional morphemes can be attached in addition: a pragmatically determined n-morpheme of which the function still needs to be examined, a gender marker (diachronically derived from a demonstrative; see Sasse 1984), possessive suffixes, markers of coordination.
e.g. |
oll - |
eechch - |
oon - |
ta - |
nnée |
|
root- |
sg- |
f.icp- |
ta- |
1pl.poss.crd |
||
'and with our neighbour (f.)' |
Case is marked segmentally by suffixes, auto-segmentally by a specific position of the accent. Suppletion only occurs in the paradigm of personal pronouns (see section 2.8). Kambaata differentiates at least eight cases: nominative, accusative, genitive, dative, ablative, instrumental-comitative-perlative (icp), short (I) and long (II) locative. The status of the vocative is not yet clear.
<5>
Not all nominals make a formal distinction between all cases, e.g. nominal modifiers generally have a reduced number of distinctions. Owing to syncretism and the reduction of case distinctions on the one hand, and the polysemy of some cases (see section 3) on the other hand, the question arises as to how many cases one should assume. I here follow the distributional approach of Blake (1994:20f) and the guidelines of Comrie (1991): If two cases are distinguished formally by at least one noun phrase, then they are considered distinct in general. The same degree of distinction is assumed to apply to all declensions, although the opposition between one case and the other can be neutralised in some of them. This approach allows one to formulate more general rules about the syntactic-semantic distribution.
<6>
It also has to be determined how many declensions are to be differentiated. Here I follow an approach that might seem over-specific to the reader, since it results in a high number of declensions and sub-declensions (Table 1). This high number gives, at first sight, the impression that the case system is extremely complex, although this is, in fact, not the case. As the analysis of the hitherto little described Kambaata language is still in its early stages, this approach is nevertheless favoured in order not to mask distinctions that might turn out to be relevant at a later stage of analysis.
Table 1: Case paradigms of common and proper nouns [2]
decl |
acc |
nom |
gen |
dat |
abl |
icp |
locII |
locI / voc |
|
F1a |
-á-ta |
´-a-t |
-á |
-áa(ha) |
-áachch |
-áan |
-áan |
´-a |
|
F1b |
-á |
´-a |
-á |
-áa(ha) |
-áachch |
-áan |
-áan |
´-a |
|
F2a |
-í-ta |
´-i-t |
-é |
-ée(ha) |
-éechch |
-éen |
-éen |
´-e |
|
F2b |
-é |
´-[i] |
-é |
-ée(ha) |
-éechch |
-éen |
-éen |
´-e |
|
F3a |
-ú-ta |
´-u-t |
-ó |
-óo(ha) |
-óochch |
-óon |
-óon |
´-o |
|
F3b |
-ó |
´-u |
-ó |
-óo(ha) |
-óochch |
-óon |
-óon |
´-o |
|
F4 |
-áa-ta |
-áa-t |
-áa |
-áa(ha) |
-áachch |
-áan |
-áan |
-áa |
|
F5 |
-ée-ta |
-ée-t |
-ée |
-ée(ha) |
-éechch |
-éen |
-éen |
-ée |
|
F6 |
-óo-ta |
-óo-t |
-óo |
-óo(ha) |
-óochch |
-óon |
-óon |
-óo |
|
M1a |
-á |
´-u |
-í |
-íi(ha) |
-íichch |
-íin |
-áan |
´-a |
|
M1b |
-á |
´-[i] |
-í |
-íi(ha) |
-íichch |
-íin |
-áan |
´-a |
|
M2 |
-í |
´-u |
-í |
-íi(ha) |
-íichch |
-íin |
-éen |
´-e / ( ) |
|
M3a |
-ú |
´-u |
-í |
-íi(ha) |
-íichch |
-íin |
-óon |
´-o |
|
M3b |
-ó |
´-[i] |
-í |
-íi(ha) |
-íichch |
-íin |
-óon |
´-o |
|
M4 |
-é |
´-[i] |
-é |
-ée(ha) |
-éechch |
-éen |
-éen |
´-e |
|
M5 |
-ó |
´-u |
-ó |
-óo(ha) |
-óochch |
-óon |
-óon |
´-o |
|
M6 |
-áa |
-óo |
-ée |
-ée(ha) |
-éechch |
-éen |
-áan |
-áa |
|
M7 |
-ée |
-óo |
-ée |
-ée(ha) |
-éechch |
-éen |
-éen |
-ée / ( ) |
|
M8 |
-óo |
-óo |
-ée |
-ée(ha) |
-éechch |
-éen |
-óon |
-óo |
|
M9 |
-úu |
-úu |
-íi |
-íi(ha) |
-íichch |
-íin (?) |
-úun |
-úu / ( ) |
Notes and Examples:
-
F1a: largest feminine declension, e.g.: maccáta ‘ear’, most plurative nouns, e.g. boorráta ‘oxen’
-
F1b: feminine proper nouns, e.g. Besá
-
F2a: e.g. gatíta ‘backyard’
-
F2b: most feminine proper nouns, e.g. Aacaamé; some common nouns, e.g. shumagé ‘hare’
-
F3a: e.g. xinkúta ‘riddle’, feminine singulative nouns in –(ich)chúta and feminine agent nouns in –aanchúta
-
F3b: feminine proper nouns, e.g. Ayyaantó, some common nouns, e.g. xorbó ‘ball’
-
F4: e.g. mashsháata ‘knife (for enset food)’, associative nouns in –’áata
-
F5: e.g. qunculéeta ‘scraper’, associative nouns in –’éeta
-
F6: e.g. hizóota ‘sister’, associative nouns in –’óota, e.g. Xummiso’óota ‘Xummiso and his associates’
-
M1a: largest masculine declension, ishimá ‘brother of mother’, most loanwords, e.g. muuzá < Amh. muz 'banana'
-
M1b: masculine proper nouns, e.g. Baafá
-
M2: e.g. foolí ‘soul’
-
M3a: e.g. utubú ‘center pole (house)’, masculine singulative nouns in –(ich)chú and masculine agent nouns in –aanchú
-
M3b: masculine proper nouns in -aamó, -amó, -eebó, -aabó see Makkeebó
-
M4: masculine proper nouns (see F2b), e.g. Boqé
-
M5: most masculine proper nouns (see F3b) e.g. Salfisó, some common nouns, e.g. hagasó ‘bird:sp’
-
M6: e.g. zaanzáa ‘centre of enset corm’
-
M7: (so far only) qoqée ‘throat’
-
M8: e.g. elóo ‘pit’
-
M9: (so far only) hagúu ‘dry season’
<7>
The order of the case forms in the paradigm (Table 1) is in accordance with the following considerations (Plank 1991c): Vertically, less complex case forms are prior to more complex ones. Forms which are considered the base of the following are more to the left than the forms that are derived from them. Cases that are at times not formally distinguished are adjacent. The accusative as the unmarked case (see section 3) and the case with the highest number of allomorphs precedes all others. It is followed by the subject case (nominative) and the case of nominal modifiers (genitive). All cases to the right can more or less be derived from the genitive (dative, ablative, icp) or the accusative (locative I and II), i.e. the diversity of forms within the declensions is limited. Accusative and genitive are “diagnostic” or “reference forms” (Plank 1991b:32). Locative II and icp, genitive and dative are neighbours because frequently they are not differentiated and subject to syncretism. The position of the locative I/vocative is problematic. It is not a complex form, instead probably derivable from the accusative and forming the base of the locative II. For the time being and in spite of the fact that this leads to the unfortunate situation that locative I follows locative II, it is put in the last column, because the rules that govern its formation are not yet fully understood due to its restricted occurrence in the database.
<8>
Horizontally, Table 1 is divided into two major parts, feminine nouns and masculine nouns. The majority of feminine nouns is characterised by an additional t-suffix in the accusative and nominative (see section 2.5). The genitive of feminine nouns is formed by changing a high short accusative vowel into its corresponding non-high vowel (i.e. i>e, u>o), non-high and long accusative vowels remain unchanged. Masculine nouns ending in a short vowel have –u or –i in the nominative and –i in the genitive (except M4 and M5). Masculine nouns ending in a long vowel have –óo in the nominative and –ée in the genitive (except M9). For the derivation of case forms to the right of the genitive general rules can be posited, irrespective of the gender. Dative, ablative and icp build on the genitive. They are derived either by vowel lengthening only (dative) or by vowel lengthening plus attaching a consonantal suffix (ablative, icp). Locative I/vocative and locative II are derived from the accusative: A high short accusative vowel is changed into its corresponding non-high vowel (i.e. i>e, u>o, non-high and long vowels remain unchanged) to receive the locative I / vocative, from which the locative II can be derived by vowel lengthening and attaching an –n.
<9>
Although common nouns and proper nouns can be differentiated on semantic grounds and although they concentrate in specific declensions, an impermeable dividing line cannot be drawn. Proper nouns are names of persons, domestic animals and places. The first two types of proper nouns differ from common nouns in not having the ‑tV-morpheme in the feminine gender. The third group, place names, occupies an intermediate position. Some of the feminine place names (e.g. Duuraamí-ta = capital of Kambaata) contain the –tV-morpheme, others do not, or the speakers differ in their judgments (e.g. Qajeelé, Jooré ~ Joorí-ta = places near Duuraame).
<10>
Some common nouns formally have to be grouped with proper nouns, because they have non-high accusative vowels, and if feminine, no –tV- in the nominative and accusative (F2b, F3b, M5). This small group contains some terms for animal species, e.g. shumagé (f) ‘hare’, xaafuluuqé (f) ‘bird:sp’, terms from the women’s respect vocabulary (Treis 2005), e.g. abbó (m) ‘rain’, idé (f) ‘sun’, from the infant’s vocabulary, e.g. kookkó (m) ‘hyena, dog, cat’ and from other semantic fields, e.g. xorbó (f) ‘ball’.
<11>
Proper nouns dominate the declensions F1b, F2b, F3b, M1b, M3b, M4, and M5. Note that only indigenous Kambaata names were considered. Most masculine proper nouns mark the nominative by [–i ] (which does not occur orthographically), a pattern rarely attested for common nouns.
<12>
When analysing case in Kambaata, it is especially important to take accent into consideration. Kambaata has no lexical accent, but grammatical minimal pairs do exist. Each case form has a characteristic accentual pattern [3] , some forms are even differentiated by accent only. The accusative is characterised by an accent on its case marker, see foolí ‘soul’ (acc), gatí-ta ‘backyard’ (acc) and meselée-ta ‘girl’. In the nominative, the accent is shifted to the front, if the accusative vowel is short, see fóolu (nom) and gáti-t (nom), but it remains if the accusative vowel is long, see meselée-t (nom). Dative, ablative, icp and locative II nouns are accented on the case-marker. Usually, the accent is not affected if a noun is modified by a possessive suffix, except in the genitive case. The genitive is characterised by an accent on the last syllable, see foolí (gen), gaté (gen), meselée (gen). The accent even moves further to the right, if the noun is modified, see fooli-sí ‘of his soul’, gate-sé ‘of her backyard’, meselee-nné ‘of our girl’. The locative I/vocative has an accent on the penultimate and often differs from the genitive only in the accent position, see gáte ‘in the backyard’ vs. gaté ‘of the backyard’ and the feminine proper name Aasháame (voc) vs. Aashaamé (acc).
<13>
Homonymous word-forms are a recurrent phenomenon in the case paradigms of Kambaata. If this homonymy is systematic, it is called syncretism (Haspelmath 2002:275). Identical locative II and icp forms are common, because feminine nouns generally do not distinguish these cases, as some masculine proper nouns (M4, M5) do. Despite this syncretism, ambiguities seldom arise, the semantics of the nouns and the verbs help to disambiguate. Inanimate referents are usually instruments, human referents usually companions. Ambiguities could arise for feminine nouns that denote locations, but if the location is to be interpreted as the place along which one passes, the super-ordinate verb is usually accompanied by the converb hig- ‘pass’. In contrast to Kambaata, Sim (1989:113) reports for Hadiyya about an accentual difference between feminine locative (II) and icp nouns.
<14>
Genitive-dative syncretism is a general feature of nouns with a long accusative vowel (F4-6, M6-9), because the dative's optional –hV-suffix is usually deleted in allegro speech. Genitive and dative do not only overlap formally in some declensions, but rather do so functionally throughout the language. Attributive possessors are genitive (‘X’s Y’), whereas possessors in nominal predications are dative (‘X has a Y’ expressed as ‘a Y is for X’) (see section 3). In identificational expressions (‘Y is X’s thing’, or ‘Y is a thing for X’), the speaker even has the choice to encode the possessor in the genitive or dative.
<15>
Sasse’s observation (1984:111) that the distinction between subject and object is often abandoned in Cushitic due to the interference with other grammatical categories (gender and focus) does not hold for Kambaata. The formal identity of nominative and accusative is rare, not systematic, and the distinction is actually neutralised only in a small fraction of all nouns (M8, M9, as well as nouns in F4-6 if modified by a possessive suffix). Nominative and accusative in the majority of declensions, irrespective of gender, are differentiated by accent. The recent grammaticalisation of case- and gender-sensitive demonstratives into gender markers (-hV/-tV) resulted only in an additional, not a new coding of the distinction.
<16>
Accusative–genitive syncretism is found among proper nouns (see F1b, F2b, F3b, M4, M5) and among some common nouns (M2). Other pairs or sets of identical forms are restricted to one or the other declension of nouns with a long final vowel.
<17>
The case systems of Kambaata and Alaaba (Schneider-Blum forthcoming) are by and large similar, but differ in details. The distribution of the –tV/-hV-morphemes (see Fig. 1) after the case marker of the noun is one of these details. As Sasse (1984) pointed out, it is likely that these morphemes originated in the gender- and case-sensitive demonstratives *ka > -ha (m.acc), *ku > -hu (m.nom), *ta > -ta (f.acc), *ti > -t[i] (f.nom) (see Table 3) and developed into gender suffixes and secondary case-suffixes. Before, gender had only been marked by agreement; case had been marked by simple case-suffixes. After the grammaticalisation of the demonstratives, an additional device of marking gender and case found its way into the paradigms of Kambaata, Alaaba and Qabeena (see also Sasse 1984 on Burji and Oromo). Compare feminine nouns in Sidaama and Kambaata: giira ‘fire’, maxine ‘salt’ (nom=acc) (Anbessa 2000:48) vs. giiráta (acc) – gíirat (nom) ‘fire’ and maxiníta (acc) – maxínit (nom) ‘salt’. However, the distribution of –tV/-hV in Kambaata is idiosyncratic and not yet explicable. Whether Kambaata represents an incomplete instance of grammaticalisation, or an overt gender-marking system in recession is yet unclear and needs to be investigated in a comparative approach. The distribution of -tV/-hV seems to depend on
-
whether the noun is a common noun or proper noun,
-
the case of the noun,
-
whether the noun has additional suffixes (see fig. 1: possessive suffixes, –n-morpheme, coordination morpheme) or only consist of stem and case marker.
<18>
All these aspects (i)-(iii) interact, and result in a very complicated picture. To compound matters, ‑tV/-hV are sometimes optional, sometimes obligatory and not always gender-sensitive. Due to restrictions of space, the exact distribution cannot be elaborated upon and the reader has to be referred to forthcoming publications.
<19>
Kambaata is a "word-marking language" (Blake 1994:100): not only the head in an NP is case-marked, but also the modifiers, i.e. adjectives, numerals and demonstratives. Adjectives are a word class separate from nouns. The modifier position is decisive for differentiating the two word classes: While nouns always appear in the genitive when modifying another noun, adjectives show case and gender agreement. Apart from this, adjectives can be modified by converbs. Numerals may be assumed to form a sub-class of adjectives characterised by a vowel change between masculine and feminine forms.
<20>
When used as heads, adjectives, numerals, and demonstratives display the full case paradigm. According to their final accusative vowel, adjectives and numerals [4] can be assigned to the declensions for nouns in Table 1, e.g. matú ‘one’ belongs to declension M3a. As modifiers, however, their case distinctions are reduced. Three cases, nominative, accusative and rest (= oblique), at the most, are distinguished. Genitive, dative, ablative, icp, locative II and locative I/vocative forms are systematically syncretic, or phrased differently, modifiers are under-specified for case.
<21>
Contrary to the observations of Schneider-Blum (forthc.) for Alaaba, the modifiers in Kambaata carry accents. There is no indication that the accents on modifiers get lost. On the contrary, in Kambaata the accent on modifiers is decisive for certain case distinctions. See declension A1, abbá m.acc, ábba m.obl, or as another minimal pair danáame manchó oddishsháta ‘the clothes of the beautiful woman’ vs. danaamé oddishsháta (a) ‘the clothes of the beautiful one’, (b) ‘the clothes of Danaame (= woman’s name)’.
<22>
By analogy with nouns, adjectives and cardinal numerals can be grouped into different declensions (Table 2).
Table 2: Adjectival declensions [5]
decl |
fem |
|||
acc |
nom |
obl |
||
A1 |
f |
-á-ta |
´-a-t |
´-a |
m |
-á |
´-u |
´-a |
|
A2 |
f |
-ú-ta |
´-u-t |
´-o |
m |
-ú |
´-u |
´-o |
|
A3 |
f |
-í-ta |
´-i-t |
´-e |
m |
-ú |
´-u |
´-o |
|
A4 |
f |
-í-ta |
´-i-t |
´-e |
m |
-á |
´-u |
´-a |
|
A5 |
f |
-óo-ta |
-óo-t |
-óo |
m |
-óo |
-óo |
-óo |
Notes and examples:
A1: |
largest declension class of adjectives, e.g. zatimá(-ta) ‘meagre’, plurative-marked adjectives |
A2: |
e.g. qeraa’rrrú(-ta ) ‘long, high, tall’, derived adjectives, e.g. sal-aanchú-ta ‘pregnant’ (< saláa ‘embryo, foetus’) |
A3: |
most numerals, e.g. matú/matí-ta ‘one’, derived adjectives wo’-aamú/wo’-aamí-ta ‘juicy’ ( < wo’á ‘water’) |
A4: |
numerals, e.g. lamalá/lamalí-ta ‘seven’, and hoolamá/hoolamíta ‘much, many’ |
A5: |
(so far only) haaróo/haaróo-ta ‘new’ |
Ordinal numerals are invariant modifiers with the accentual pattern of genitive nouns. As heads of an NP, they require additional nominalising morphology.
<23>
Kambaata distinguishes four demonstratives, (i) ka/ta, (ii) híkka/hítta, (iii) káaph/táaph and (iv) hikkáaph/hittaaph. From the speaker’s perspective, (i), (ii) and (iv) refer to near, far and very far objects. The function of (iii), the hearer’s perspective and the discourse functions still need to be investigated, the reader is referred to forthcoming publications. Formally the demonstratives constitute two classes, (i)/(ii) and (iii)/(iv). Demonstratives differ from adjectives in two respects: As heads they consistently distinguish masculine and feminine plural forms. The opposition between locative II and icp in the feminine plural paradigm is noteworthy, because this is not attested in any other feminine paradigm in the language. Additionally, demonstratives have a unique accent pattern not comparable to the one of nouns or adjectives.
Table 3: ka/ta- and híkka/hítta-demonstratives [6]
m |
f |
m.pl |
f.pl |
||
acc |
dep |
(hík)ka |
(hít)ta |
- |
- |
idp |
(hik)káan |
(hit)táan |
(hik)kará |
(hit)tará |
|
nom |
dep |
ku / híkku |
ti / hítt |
- |
- |
idp |
(hik)kúun |
(hit)tíin |
(hik)kurú |
(hit)tirú |
|
gen |
dep |
(hík)ka ~ (hik)kánn |
(hít)ta ~ (hit)tánn |
- |
- |
idp |
(hik)kanní |
(hit)tanné |
(hik)karrí |
(hit)tarrí |
|
dat |
dep |
(hík)ka ~ (hik)kánn |
(hít)ta ~ (hit)tánn |
- |
- |
idp |
(hik)kanníi(ha) |
(hit)tannée(ha) |
(hik)karríi(ha) |
(hit)tarríi(ha) |
|
abl |
dep |
(hík)ka ~ (hik)kánn |
(hít)ta ~ (hit)tánn |
- |
- |
idp |
(hik)kanníichch |
(hit)tannéechch |
(hik)karríichch |
(hit)tarríichch |
|
icp |
dep |
(hík)ka ~ (hik)kánn |
(hít)ta ~ (hit)tánn |
- |
- |
idp |
(hik)kanníin |
(hit)tannéen |
(hik)karríin |
(hit)tarríin |
|
locII |
dep |
(hík)ka ~ (hik)kánn |
(hít)ta ~ (hit)tánn |
- |
- |
idp |
(hik)kannéen |
(hit)tannéen |
(hik)karráan |
(hit)tarráan |
|
locI |
dep |
- |
- |
- |
- |
idp |
(hik)kánne |
- |
- |
- |
Table 4: káaph/táaph- and hikkáaph/hittáaph-demonstratives
m |
f |
m.pl |
f.pl |
||
acc |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)káaph |
(hit)táaph |
(hik)kaaphíra |
(hit)taaphíra |
|
nom |
dep |
(hik)kúuph |
(hit)tíiph |
- |
- |
idp |
(hik)kúuph |
(hit)tíiph |
(hik)kuuphíru |
(hit)tiiphíru |
|
gen |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)ka’í |
(hit)ta’é |
(hik)kaaphirí |
(hit)taaphirí |
|
dat |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)ka’íi(ha) |
(hit)ta’ée(ha) |
(hik)kaaphiríi(ha) |
(hit)taaphiríi(ha) |
|
abl |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)ka’íichch |
(hit)ta’éechch |
(hik)kaaphiríichch |
(hit)taaphiríichch |
|
icp |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)ka’íin |
(hit)ta’éen |
(hik)kaaphiriin |
(hit)taaphiríin |
|
locII |
dep |
(hik)káaph |
(hit)táaph |
- |
- |
idp |
(hik)ka’éen |
(hit)ta’éen |
(hik)kaaphiráan |
(hit)taaphiráan |
|
locI |
dep |
- |
- |
- |
- |
idp |
(hík)ka’e |
- |
- |
- |
<24>
The (pro)nominal interrogative ayé ‘who?’ displays the case distinctions and forms of declension M4 (see Table 1). má ‘what?’ has an irregular, but noun-like paradigm (Table 5).
Table 5: The Interrogative má ‘what?’
‘what?’ |
|
acc |
má |
nom |
mí |
gen |
míi |
dat |
míi(ha) |
abl |
míichch |
icp |
míin |
locII |
máan |
The interrogatives hákka/hátta ‘which?’ and me’ú/me’íta ‘how much?’ can either be used as modifiers (with a reduced case paradigm) or as heads, i.e. independently. hákka/hátta ‘which?’ has the same paradigm as the (hík)ka/(hít)ta-demonstratives (Table 3), differing in all forms only in the stem-vowel a vs. i. The interrogative me’ú/me’íta ‘how much?’ behaves like a numeral of declension A3 (see Table 2), having the same degree of formal distinction and the characteristic vowel change u ~ i. As a modifier its paradigm is reduced (acc – nom – obl). When used independently, it fits into the M3a declension of masculine nouns or F2a declension of feminine nouns, respectively.
<25>
The interrogatives do not make up a morphologically homogeneous group. They either have a (pro)noun-like, demonstrative-like, or numeral paradigm. There is a straightforward formal relation between question word and questioned word. [7] The questioned entity is typically in the same category like the interrogative. The (pro)nominal interrogatives ayé and má question a (pro)noun, an answer to a question containing hákka/hátta typically involves a demonstrative (Q: which one? A: this one/that one), and me’ú/me'íta enquires about an amount which would be expressed by a numeral or quantifier.
<26>
In the pronominal paradigm (Table 6) nine personal pronouns differentiate at least six case forms each. For 3m.sg locative II and icp are distinguished in addition. As a result of innovations in the personal pronoun paradigm, different stems can be found for different case forms of one and the same person. Case forms can no more be derived regularly from accusative and genitive. Apart from the forms shown in the table, Kambaata has vocative pronouns: kóo (2m.sg.voc) and tée (2f.sg.voc) are used independently, the short forms kó and té are used if a head noun follows, e.g. kó adabáa ‘hey, boy!’, té meselée ‘hey, girl!’. In the vocative case of the second person pronouns gender (masculine vs. feminine) is distinguished. This is insofar surprising as all other second person pronouns refer to feminine as well as masculine referents. The vocative pronouns are only used for persons younger than the speaker. Plural forms do not exist. If there are several addressees, ki’nne’eeta (2pl.acc) can follow the singular pronouns.
Table 6: The personal pronouns
nom |
acc |
gen |
dat |
abl |
icp |
locII |
dep |
|
1sg |
án |
ées |
íi |
esáa(ha) |
esáachch |
esáan |
esáan |
-’e / -’[i] |
2sg |
át |
kées |
kíi |
kesáa(ha) |
kesáachch |
kesáan |
kesáan |
-(k)k |
2hon |
á’nnu |
ki’nnéta |
ki’nné |
ki’nnée(ha) |
ki’nnéechch |
ki’nnéen |
ki’nnéen |
-’nne ~ ‑(k)ki’nne |
3m |
ís |
isú |
isí |
isíi(ha) |
isíichch |
isíin |
isóon |
-s |
3f |
íse |
iséta |
isé |
isée(ha) |
iséechch |
iséen |
iséen |
-se |
3hon |
íssa |
issáta |
issá |
issáa(ha) |
issáachch |
issáan |
issáan |
-(s)sa |
1pl |
na’óot |
nées |
níi |
nesáa(ha) |
nesáachch |
nesáan |
nesáan |
-nne |
2pl |
a’nno’óot |
ki’nnéeta ~ ki’nne’eeta |
ki’nnée ~ ki’nne’ée |
ki’nnée(ha) ~ ki’nne’ée(ha) |
ki’nne’éechch |
ki’nne’éen |
ki’nne’éen |
-’nne ~ ‑(k)ki’nne |
3pl |
isso’óot |
isso’óota |
isso’óo |
isso’óo |
isso’óochch |
isso’óon |
isso’óon |
-(s)sa |
Apart from free personal pronouns, Kambaata has a set of dependent pronouns (see last column of Table 6) that refer to the possessor if they are suffixed to a noun and to an object if suffixed to a verb. [8]
<27>
The previous section concentrated on the formal aspects, the following discusses which cases encode which syntactic/grammatical functions and semantic roles. Besides the common and presumably well-known, some more remarkable functions of the different cases are pointed out. Case is of considerable importance in Kambaata, because there are no adpositions and the order of NPs in a sentence is relatively free. Like its closest relatives Alaaba (Schneider-Blum forthc.) and Qabeena (Crass 2005) and other Cushitic languages, Kambaata has a marked nominative system. The nominative does not have more phonological weight than the accusative (see section 2), but the accusative is more frequent in absolute terms: it is used in more contexts than the nominative. 'Marked nominative system” here refers neither to the morphological nor to the phonological markedness of the nominative, but instead to a notional markedness.
The nominative encodes the subject of intransitive and transitive verbs (see (1) above), as well as of nominal predications.
<28>
The accusative has a wide range of functions. First of all, it marks the direct object of a verb whose semantic role is prototypically that of a patient (see (1) above). Motion verbs encode their goals in the accusative, see hadáta ful- ‘go out’ (lit.: 'go to outside”), lagá orooqq- ‘leave for the river, leave (to go) to the river’. Some verbs which one would expect to be intransitive, can take accusative 'cognate objects” (Crass 2005:277-79), e.g. gisanáta gis- ‘sleep (a sleep)’ and shumáta shuma’- ‘urinate (urine)’. Apart from marking objects of different semantic roles, the accusative is also an 'adverbial case' (Sasse 1984). The plain accusative form of a temporal noun expresses adverbials of time, either points in time (2a) or durations (2b). In minor instances it encodes manner adverbials (2c).
2a) |
Ga’áata / sanná |
mar-áamm. |
tomorrow. f.acc / Monday. m.acc |
go- 1sg.ipv |
|
I will go tomorrow / on Monday. |
2b) |
Matú |
sanná |
horánka |
xíjj-ee-se. |
one. m.acc |
week. m.acc |
whole. m.acc |
be:sick (+pal+gem)-3m.pve-3f.obj |
|
She was sick for a whole week. |
2c) |
Xuummá |
hóshsh-eemm. |
peace. m.acc |
pass:the:day (+pal+gem)-1sg.pve |
|
I passed the day peacefully / in peace. |
<29>
Some adjectives can govern accusative complements, like (hasú) qoorá ‘clever (e.g. in searching)’, (ontíta lokkáta) xillaallá ‘(e.g. five feet) deep’. See also (3).
3) |
Buxichchí |
sá’u |
buudá |
qabára-a. |
poor:man. sg.m.gen |
cow. m.nom |
horn. f.acc |
flat- m.cop |
|
A poor man’s cow has flat horns / is flat regarding the horns. |
<30>
The accusative often serves as citation form. In addition, one may consider it the case of nominal predicates (4). [9]
4) |
Kúun |
íi-haa |
ishíma. |
idp:dem1.m.sg |
1sg.gen-m.cop |
mother’s:brother. m.acc |
|
This is my mother’s brother. |
<31>
The term I adopt for the direct object case is not uncontroversial. Instead of 'accusative' a widely adopted term among Cushitists (and beyond) is 'absolutive'. Sasse (1984) favours the latter term on the basis of the wide range of additional functions apart from encoding the direct object. I have nevertheless opted for 'accusative’ in order to avoid the frequent association of 'absolutive' with 'ergative' which is definitely not found in Cushitic.
<32>
The genitive is the case of nominal modifiers. There is an array of different semantic relations between modifying and modified noun, the most prototypical being the possessor-possessed relationship or the whole-part-relationship. As Kambaata has no adpositions, spatial expressions consist of a genitive (or ablative) noun referring to a whole and the locative II or icp-form of a relational noun referring to a part (e.g. ‘top’, ‘inside’, ‘bottom’). See also (5).
5) |
Mánchu |
haqqí-na |
miní |
mereeróonee-t. |
man. sg.m.nom |
tree. m.gen-crd |
house. m.gen |
middle. m.locII.vv-cop |
|
The man is between the tree and the house. |
<33>
The dative is the marker of indirect objects and some adverbials. The semantic roles recipient and beneficiary are encoded by the dative (6).
6) |
Íi |
hizóoha-’nn |
gaxá |
abbáashsh-eemm. |
1sg.gen |
sister. f.dat-n.(met+)1sg.poss |
floor. m.acc |
sweep (+pal+gem)-1sg.pve |
|
I swept the floor for my sister. |
The purpose is expressed by the dative which itself is frequently found on deverbal nouns (7b).
7a) |
Maxée |
afúushsh! |
rainy:season. m.dat |
keep (.caus1).2sg.imp |
|
Keep it for the summer / rainy season! |
7b) |
Ajjoorí |
faamá |
xuud-íi |
márr-eemm. |
Ajjoora. m.gen |
waterfall. m.acc |
see- m.dat |
go (+gem)-1sg.pve |
|
I went to see the Ajjoora falls. |
<34>
The dative marks adverbials of time, but in contrast to the accusative, it refers to an approximate period of time. Compare (8) below with (2b) above.
8) |
Máto |
hezzeetíi |
xíjj-ee-se. |
one. m.obl |
week. m.dat |
be.sick (+pal+gem)-3m.pve-3f.obj |
|
She was sick for about one week. |
<35>
Whereas the attributive possessor is encoded by the genitive, the dative marks the possessor in nominal predications (9).
9) |
Masaaláan |
meentíi |
batináashshat |
hújit |
yóo-ssa |
masaala. m.locII |
women. m.dat |
a:lot. f.nom |
work. f.nom |
be.3 -3pl.obj |
|
On Masaala [holiday in September] the women have a lot of work. |
<36>
The ablative marks oblique objects whose semantic role encompasses source, origin and starting point (see (1) and (10a)). In addition, it encodes what is to be avoided (Blake 1994: 'aversive/evitative’) (10b), a related concept ‘apart from’ (10c), and the maleficiary of an action (10d).
10a) |
sanníichch |
harbá |
iill-án |
qaxée |
Monday. m.abl |
Friday. m.acc |
arrive- 3m.ico |
until |
|
from Monday to Friday |
10b) |
Mashsháachch |
qoraphph-í-t |
waasá |
mur! |
knife:sp. f . abl |
stay. refl-ep-2sg.pco |
enset. m.acc |
cut.2 sg.imp |
|
Beware of the knife when cutting waasa [= enset food]! |
10c) |
kambaatíichch |
wólit |
[…] |
minaagad-áakkat |
Kambaata. m . abl |
other. f.nom |
[…] |
ethnic.group- pl.f.nom |
|
other ethnic groups apart from the Kambaata (Kambaatissata III:10) |
(10d) |
Cíilu |
reh-ée-se. = |
Cíilu |
iséechch |
reh-ée'u. |
baby. m.nom |
die-3 m.pve-3f.obj |
baby. m.nom |
3f.abl |
die- 3m.pve |
|
The baby died on her. |
<37>
However, not only verbal heads, also relational nouns (11a) or adjectives (11b), govern ablative nouns. The ablative marks the base of comparison (11c).
11a) |
Ginjiríichch |
biríta |
orooqq-ée'u. |
breakfast. m . abl |
before. f.acc |
go:out-3 pl.pve |
|
He went out before breakfast. |
11b) |
Hotéelu |
miníichchi-s |
abbíshsh |
qée'rra-a |
hotel. m.nom |
house. m . abl - 3m.poss |
exceed. 3m.pco |
far -m.cop |
|
The hotel is very far from his house. |
11c) |
Át |
esáachch |
ábba-a |
2 sg.nom |
1sg . abl |
big- m.cop |
|
You (m) are bigger than me. |
Some relational nouns (e.g. zakkíin ‘after’, aazíin ‘underneath’, alíin ‘above’, biríta ‘before’) require an ablative (11a) instead of a genitive modifier (5).
<38>
Case is syntactically/grammatically determined when marking a noun for a specific grammatical relation to the head, but the choice of one or the other case form can also be subject to semantic/pragmatic considerations. For adverbials of time, the choice of one or the other case form results in slight meaning differences. A point in time can either be encoded by the accusative, locative II or ablative. The ablative nouns have the special connotation ‘about’, ‘around’, i.e. oroobíichch ‘about Wednesday (but possibly also one day before or after)’, and is used when the speaker is in doubt about the exact day. The meaning of a locative time adverbial, oroobáan, is similar (‘probably on Wednesday’), whereas the accusative, oroobá, refers to an exact point in time (‘on Wednesday’).
<39>
Besides marking temporal and local adverbials, the locative II encodes oblique objects. A locative noun can alternate with a relational noun NP. The locative noun in the first variant of (12) denotes the place, but gives no exact information as to whether the spoon is put into the mug or laid across the mug, or alternatively beside it (the listener must deduce this from the context). The relational NP is more explicit, specifying the search domain and clearly stating that the spoon is put into the mug.
12) |
Án |
mooqúta |
xenqóon |
~ |
xenqó |
aazéen |
afuushsh-éemm. |
1 sg.nom |
spoon. f.acc |
mug. f.locII |
~ |
mug. f.gen |
interior. m.locII |
put (+pal+gem)-1sg.pve |
|
I put the spoon into the mug. |
The instrumental-comitative-perlative form of inanimate nouns expresses means and instruments.
13a) |
Xaasá |
baacaséen |
fánn-eemm. |
tin. m.acc |
sickle. f.icp |
open (+gem)-1sg.pve |
|
I opened the tin with a sickle. |
13b) |
Faashshíin |
wáall-eemm. |
horse. m.icp |
come (+gem)-1sg.pve |
|
I came on horseback. |
The icp encodes companions or additions (‘plus’).
14a) |
Án |
kesáan |
mar-áamm. |
1 sg.nom |
2 sg.icp |
go- 1sg.ipv |
|
I will go with you. |
14b) |
birrá |
shilingíinee-t. |
birr. m.acc |
shilling. m.icp.vv-cop |
|
It is [one] birr and [one] shilling. (= Price: 1.50 Birr) |
Motion verbs require the icp to mark the place through, across or along which one moves.
15) |
koofíin |
hig- |
area:between:legs. m.icp |
pass- |
|
to pass through the legs (e.g. a ball) |
<40>
Some relational nouns may be locative II-marked or icp-marked to express differing, but related spatial relations. The noun aazí ‘interior’ differentiates the icp aazíin ‘under(neath), below’ and the locative II aazéen ‘inside, under’; alí ‘top’ has the icp alíin ‘above’ and the locative II aléen ‘on’.
<41>
Blake (1994) only reports about languages which have a separate marker for the perlative, but maybe the polysemy of the icp-marker is typologically not as surprising as it might seem at first sight, because Luraghi (1991:67) reports about a polysemous case reconstructed for Proto-European and Old Indic which also encodes the instrument and the place passed.
<42>
The status, distribution, and semantics of vocative and locative I are not entirely clear. If Comrie’s (1991) recommendations for delimiting cases are applied strictly, it is not even sure yet whether we are dealing with two cases or with just one. So far no declension was found that differentiates vocative and locative I. But note that for some declensions in Table 1 the vocative and locative I cell remained empty due to lack of data. If no formal difference between the two is found for any declension, one would have to assume a single polysemous case marker.
<43>
The status of the vocative as a case is questioned in the literature. If case is defined according to Blake (1994:1) as marking dependent nouns for the type of relation they bear to their heads, then, strictly speaking, the vocative cannot be considered a case, since its most frequent function is to address. Apart from this it serves as citation form, not only for proper nouns, but for common nouns as well (like the accusative). As an example of its use in texts consider (16a).
16a) |
Ciilíi |
su’mmá |
Ayyáano |
y-í-n |
fushsh-i-néemm. |
infant. m.dat |
name. m.acc |
proper:noun. m.voc |
say- ep-1pl.pco |
take:out- ep-1pl.pve |
|
We called the child 'Ayyaano”. |
16b) |
Ti |
ázut |
zomború |
y-am-án-t |
ga’’-an-táa’u. |
dep:dem1.f.nom |
milk. f.nom |
first:milk. m.acc |
say- pass-pass(+ass)-3f.pco |
call- pass(+ass)-3f.ipv |
|
This milk is called zomboru [first milk of a cow]. |
Instead of a vocative, the verb y- ‘say’ can also have an accusative complement (16b). Accordingly, the functions of vocative and accusative overlap.
<44>
The locative I encodes oblique objects and local adverbials. It is rarely used in elicited sentences, but in natural discourse and texts it appears quite frequently, even though not as often as the locative II. A certain set of nouns occurs in the locative I in preference to the locative II. The nouns of this set seem to be inherently definite, e.g. míne ‘in the house’, gáte ‘in the backyard’, anshára ‘under the bed (= the only one that one usually owns)’ (17a). Frequently attested is the locative I form of place names (17b), whereas the locative II of these nouns is rather uncommon. Temporal adverbials can occur in the locative I (17c).
17a) |
Lokkáachchi-s |
fushsh-í |
anshára |
afuushsh-ée-’e. |
feet. f.abl-3m.poss |
take:out- 3m.pco |
area:under:the:bed. m.locI |
put- 3m.pve-1sg.obj |
|
He took me from his feet and put me under the bed. (Kambaatissata III:48) |
17b) |
Biishsháta |
meseléeta |
Beqéra |
moog-éemma. |
red. f.acc |
girl. f.acc |
proper:noun. m.locI |
bury- 3hon.ipv |
|
A red girl is buried in Beqera. (= Riddle) |
17c) |
Fúlo |
miní |
xuud-áno |
áago |
hadáta |
xuud-áno. |
go:out. m.locI |
house. m.acc |
see- 3m.ipv |
Enter. m.locI |
outside. f.acc |
see-3 m.ipv |
|
When going, out it sees the house, when entering, it sees the outside. (= Riddle) |
<45>
Sometimes the locative I is in free variation with the icp and encodes a companion or instrument (18a, b). [10]
18a) |
Bajigo-’óo |
ánnu |
fóole ( ~ foolíin) |
yóomma-ba’a. |
proper:noun -assoc.f.gen |
father. m.nom |
soul. m.locI ( ~ soul. m.icp ) |
be. 3hon-neg |
|
The father of Bajigo and his siblings is no longer alive. (Kambaatissata III:100) |
18b) |
ánga ( ~ angáan) |
xuunsh-amm-ée-na |
hand. f.locI ( ~ hand. f.icp ) |
squeeze- pass(+gem)-3m.pve.rel-crd |
lókka ( ~ lokkáan) |
xuunsh-amm-ée |
hoogó |
bu’llaháa |
|
foot. f.locI ( ~ foot.f.icp) |
squeeze- pass(+gem)-3m.pve.rel |
enset:juice. f.gen |
enset:starch. m.acc.crd |
|
[…] and starch from the enset juice which was squeezed [out of the enset pulp] by hand and squeezed [out of the enset pulp] by foot. |
<46>
In contrast to vocative nouns (19a), locative I nouns cannot be modified (19b), neither by an adjective or relative clause nor by a possessive pronoun. If one wants to modify a locative noun, it has to be locative II-marked.
19a) |
Adabéechcho-’e! |
boy. sg.m.voc-1sg.poss |
|
My little boy! |
(19b) |
*ábba |
míne |
/ |
*míne-’e |
big. m.obl |
house. m.locI |
/ |
house. m.locI-1sg.poss |
|
in the big house |
/ |
in my house |
<47>
Apart from this syntactical difference (+/- capable of governing a modifier), it is still an unsettled problem whether locative I and II or locative I and icp (18a, b) are semantically different. It is not known either whether there are verbs which only tolerate one or the other locative case form. There are, however, many cases like in (20) in which both can be used interchangeably.
20a) |
It-eennó-ra |
hi’rrú |
agúrr |
eat- 3hon.ipv.rel-pl.m.acc |
buy. m.acc |
leave. 3m.pco |
gizzá-s |
haraqéen |
~ |
haráqe |
xoof-áyyoo’u. |
|
money. m.acc-3m.poss |
alcohol:sp. f.locII |
~ |
alcohol:sp. f.locI |
finish- 3m.prog |
|
Instead of buying food, he spends his money on alcohol:sp. |
<48>
While the present article had to be confined to case marking on nominals, case in Kambaata is not only used to mark nominals for the type of relation they bear to their heads, but also to mark subordinate verbs, or rather clauses, for the type of relation they bear to their main clause. Kambaata does not only lack adpositions but also conjunctions. The markers of subordinate clauses are of nominal origin and are case-marked. When in the course of time these nominal subordinate markers became more and more attached to the preceding verb (a modifying relative verb), they developed into enclitics and suffixes, so that today subordinate verb plus head noun plus case marker create one complex whole. This phenomenon, however, will have to be discussed in a separate publication.
Acknowledgments
<49>
The data for this article was collected during two field trips to Duuraame (KX-Zone, SNNPRS, Ethiopia) in October and November 2002 and from April to October 2003. I especially thank my informants Aberra Gebrewold, Filiphos Paulos, Mathewos Shagana, Tessema Handiso, Tseganesh Leggese, and Markos Wedajo with whom I worked on the topic of this article, the Deutsche Forschungsgemeinschaft for generously funding the project Grammatische und lexikalische Dokumentation des Hochlandostkuschitischen (2002-2005) and Gerrit J. Dimmendaal for very helpful comments on an earlier draft of the article. Further thanks go to the organisers and participants of the 4th International Conference of Cushitic and Omotic Languages in Leiden, April 10-12, 2003.
Abbreviations |
|
ABL |
ablative |
ACC |
accusative |
Amh. |
Amharic |
ASS |
assimilation |
ASSOC |
associative |
CAUS |
causative |
CRD |
coordination marker |
COP |
copula |
DAT |
dative |
DEM |
demonstrative |
DEP |
dependent/determining |
EP |
epenthesis |
F |
feminine |
GEM |
gemination |
GEN |
genitive |
hon |
honorific/impersonal |
ICP |
instrumental-comitative-perlative |
ICO |
imperfective converb |
IDP |
independent/pronominal |
IMP |
imperative |
IPV |
imperfective |
LOCI |
short locative |
LOCII |
long locative |
M |
masculine |
MET |
metathesis |
NEG |
negation |
NOM |
nominative |
OBJ |
object |
OBL |
oblique |
PAL |
palatalization |
PCO |
perfective converb |
PVO |
-o -perfective |
PVE |
-e- perfective |
PL |
plural |
POSS |
possessive |
PROG |
progressive |
REFL |
reflexive |
REL |
relative clause marker |
SG |
singular |
sp. |
species |
References
Abebe Gebre-Tsadik, Margaret and Ronald J. Sim, Charlotte and Klaus Wedekind 1985
'The verb morphophonemics of five highland East Cushitic languages including Burji,' Afrikanistische Arbeitspapiere 2
Anbessa Teferra 2000
A Grammar of Sidaama. Unpublished PhD Thesis. Hebrew University of Jerusalem
Blake, Barry J. 1994
Case. (Cambridge Textbooks in Linguistics) Cambridge: Cambridge University Press
Comrie, Bernard 1991
'Form and Function in Identifying Cases,' In: Plank 1991a, pp.41-55
Crass, Joachim 2005
Das K’abeena. Deskriptive Grammatik einer hochlandostkuschitischen Sprache. Köln: Köppe
Gordon, Raymond G. Jr. (ed.) 2005.
Ethnologue: Languages of the World, Fifteenth edition. Dallas, Tex.: SIL International ( Kambaata )
Haspelmath, Martin 2002
Understanding Morphology. (Understanding Language Series) London: Arnold
Hudson, Grover 1976
'Highland East Cushitic,' in Bender, M. Lionel (ed.) The Non-Semitic Languages of Ethiopia, pp.232-277. East Lansing, MI: African Studies Center, Michigan State University
Kambaatissata. Rosaanchi Maxaafa. Sakki Kifila 1989
[Kambaata Language. School Book. Grade III]. E.C. Southern Nations Nationalities and Peoples’ Regional State: Education Bureau
Kambaatissata. Rosaanchi Maxaafa. Shoolki Kifila 1989
[Kambaata Language. School Book. Grade IV]. E.C. Southern Nations Nationalities and Peoples’ Regional State: Education Bureau
Korhonen, Elsa, Mirja Saksa and Ronald James Sim 1986
'A dialect study of Kambaata – Hadiyya,' Afrikanistische Arbeitspapiere 5:5-41, 6:71-122
Luraghi, Silvia 1991
'Paradigm size, possible syncretism, and the use of adpositions with cases in flective languages,' In: Plank 1991a, pp.57-74
Plank, Frans (ed.) 1991a
Paradigms: The Economy of Inflection. Berlin, New York: Mouton de Gruyter
Plank, Frans 1991b
'Of abundance and scantiness in inflection: a typological prelude,' In: Plank 1991a, pp.1-39
Plank, Frans 1991c
'Rasmus Rask’s Dilemma,' In: Plank 1991a, pp.161-196
Sasse, Hans-Jürgen 1984
'Case in Cushitic, Semitic and Berber,' In: Bynon, James (ed.). Current Progress in Afro-Asiatic Linguistics: Papers of the Third International Hamito-Semitic Congress, pp.111-126. Amsterdam, Philadelphia: Benjamins
Schneider-Blum, Gertrud forthc.
'Case System in Alaaba (HEC).' To appear In: Proceedings of the 15th 'International Conference of Ethiopian Studies, 20-25 July 2003, University of Hamburg
Sheleme Kitata 1989
Noun Morphology of Kambata. Unpublished B.A. thesis, Addis Ababa University
Sim, M.G. 1988
'Palatalisation and Gemination in the Kambaata Verb,' Journal of Afro-Asiatic Languages 1,1:58-65
Sim, R.J. 1988
'The Verb in Northern Highland East Cushitic.' In: Marianne Bechhaus-Gerst and Fritz Serzisko (eds.). Cushitic – Omotic. Papers from the International Symposium on Cushitic and Omotic Languages, Cologne, January 6-9, 1986, pp.433-452. Hamburg: Buske
Sim, R.J. 1989
Predicate Conjoining in Hadiyya: A Head-Driven PS Grammar. Unpublished PhD Thesis. University of Edinburgh: Department of Linguistics
Treis, Yvonne 2005
'Avoiding Their Names, Avoiding Their Eyes: How Kambaata Women Respect Their In-Laws,' Anthropological Linguistics 47,3:292-320
[1] All examples of this article are written in the official Kambaata orthography which follows the Oromo orthography. It deviates from IPA-standards in the following respects: (a) Ejectives: <ph> = [p’], <x> = [t’], <q> = [k’] and <c> = [tʔ’], other consonants: <ch> = [tʔ], <sh> = [ʔ], <’> = [ʔ]. (b) Both letters of a digraph are reduplicated when the consonant is geminate, e.g. [ʔ:] = <shsh>. (c) Long vowels are indicated by two vowels in a sequence, e.g. [a:] = <aa>. (d) Due to a questionable orthographic convention sonorants after glottal stop are always doubled (’nn, ’mm, ’ll, ’rr), although they are actually clusters of only two consonants (ʔ + n/m/r/l). (e) Every Kambaata word ending in a consonant orthographically, has a final unvoiced [i] phonetically. Being the default final vowel, it is not written, i.e. the nominative marker of feminine nouns, /-t/, for instance, is phonetically [ti].
In contrast to the official Kambaata orthography, which marks glottal stop only between identical vowels (e.g. [aʔu] = <au>), here all glottal stops apart from the predictable word-initial ones are written.
[2] General note on this and the following tables: Optional elements are found in round brackets. Empty round brackets (…) indicate that the forms are not (yet) attested in the database. Hyphens (-) indicate forms that were ruled out. In square brackets […] phonetic information is presented. Some temporal nouns (e.g. barí ‘day’, gassimá ‘morning’) could not be put in one of the declensions with any certainty. They show irregular vowels in some case forms and not all forms are attested in my database.
[3] Note the deviation from the nominal case patterns in the demonstrative paradigms (Table 3 and 4).
[4] Independent demonstratives (i.e. demonstratives as heads / pronouns), however, cannot be assigned to the nominal declensions in Table 1, because they differ considerably in their formal means of case marking (see section 2.6.2).
[5] The (short) oblique forms given in the table alternate with long forms that have – ta (f) or – (h)a (m), additionally. The accent position remains unaffected by the extension. Informants always stressed that both, the short and the long form, are equally possible. See maléesa ~ maléesa - ta meentichchóo ‘to the wise woman (dat)’, maléesa ~ maléesa -a nubabíi ‘to the wise old men (dat)’.
[6] The tilde (~) in Table 3 indicates that in the oblique two forms are in free variation.
[7] I am indebted to Gerrit J. Dimmendaal (p.c.) for drawing my attention to the relation between the word-class of a question word and the word-class of its typical answer.
[8] The allomorphs with the geminate initial consonant are found after vowels and non-sonorant consonants, the other ones after sonorants only.
[9] The predicative form of the noun only looks like an accusative segmentally; autosegmentally it exhibits a non-accusative accentual pattern (-´- instead of --´). The divergence from the characteristic accusative accentual pattern cannot yet be accounted for.
[10] As examples (18a) and (18b) show, the term 'locative' is not really adequate for the short case form. But because the term 'oblique' is reserved for the non-nom-acc case of nominal modifiers and should not be applied again and because the short case form formally resembles the locative II most, the term 'locative I' is favoured.
License
Recommended citation ¶
Treis Y (2006). Form and Function of Case Marking in Kambaata. Afrikanistik online, Vol. 2006. (urn:nbn:de:0009-10-3790)
Please provide the exact URL and date of your last visit when citing this article.
Fulltext ¶
- Volltext als PDF ( Taille 446.2 kB )
Comments ¶
There are no comments yet.
Do you have any additions or comments?
Submit comment.