The Structure of Kiranti Languages: Comparative Grammar and Texts.

Rapacha, Lal

The Structure of Kiranti Languages: Comparative Grammar and Texts Karen H. Ebert. Zurich: ASAS, Universiat Zurich, 1994. Pages: 283. ISBN 3-952010-5-2

Karen H. Ebert's grammar on Tibeto-Burman (T-B) Chianti languages (acronym SKI: CGT), to my knowledge is the first typological (although the term 'comparative grammar' has not strictly been used in Crystal's (1980:66 and 362-363) sense grammar of such type on geographically less-accessible and 'nearly extinct' (Crystal 2000: 20) Kiranti languages (see Appendix A for their regional distribution) spoken in Eastern hills of Nepal from the Likhu river in the west plus across the Nepal border to north-east India viz., Sikkim and Darjeeling. Phylogenetically (1), the generic phyla 'Kiranti' under T-B subfamily, includes not less than thirty-two (cf. Rai 1985, Hansson 1991a, Nishi 1992 and Pokharel 1994) scantily described languages. Or most of them are yet awaiting linguistic description and further documentation in any form of grammar or dictionary. To some extent, some of these languages have been investigated only recently after Allen's A Sketch of Thulung Grammar (1975) in the Kiranti linguistic literature.

Till the year 1994, all other grammatical descriptions and investigations were based on separate individual Kiranti languages, e.g. Toba 1984, Rai 1985, van Driem 1987 and 1993b, Michailovsky 1988, Ebert 1997a and Ebert 1997b. Besides these detailed works, there is a great deal of papers on several grammatical aspects of the Kiranti grammars published since the late 1960s and onwards. Therefore, I prefer to suggest Ebeq's grammar as the first typological account on the Kiranti languages because she has included six Kiranti languages viz., Khaling, Thulung, Caroling [Tsamling], Athpare, Bantawa and Limbu (Phedappe dialect) for the purpose of describing the grammatical structures of these languages. A reader trained in linguistics will be amazed by very much similar and dissimilar grammatical features ofthese mutually unintelligible languages termed as 'Kiranti.'

The SKL: CGT has been organized in six chapters and two appendixes. In the first appendix, Appendix A (pp.140-150) verbal paradigms on person and number affixes, basic tense and negative paradigms only of Athpare and Bantawa have been outlined. The second appendix, Appendix B (pp. 154-280) has included the texts from these languages and the sources of the texts are available in all languages except for Limbu. This grammar is the culmination of her fieldwork study in the eastern hills of Nepal mainly on Camling and Athpare languages and the rest of the data extracted are mainly from Allen (1975) for Thulung, Toba (1984) Khaling, Rai (1985) Bantawa and van Driem (1987) for Limbu. The narrative representation of fifteen myths and folktales explicate the Kiranti peoples' cultural, social and political worldviews.

The first chapter, 'Introductory Remarks' (pp.8-18), begins with a brief general observation and phonemic inventory of the six represented Kiranti languages. Some of her general conclusions are:

a. many of these languages are not even known by name to the linguistic world,

b. their phylogeneticl grouping and sub-groupings proposed till today (cf. Grierson 1909, Benedict 1972, Shafer 1953, Matisoff 1978, Hanpon 1991a and van Driem 1992) are all rather tentative due to the poor documentation of most members of the group,

c. most of them are threatened by extinction,

d. are SOV languages with a rather strict order of modifiers before heads,

e. South Eastern (SE) languages are mainly agglutinative and words can easily be split up into morphs, whereas the northern languages have more stem variation and portmanteau forms,

f. verb is characterized by a complex system of person and number markers,

g. agreement system is sensitive to the pragmatic constellation and agent-patient are usually marked on the verb,

h. are morphologically ergative, where Camling and Thulung exhibit a split between speech act participants, whereas in Limbu pronouns and nouns,

i. gender distinction is marginal,

j. the coding of space is a fascinating part,

k. make little use of converbs and participles in subordination- the northern and western languages somewhat more than the southern languages and Limbu. Hayu, the western-most language, has no converbs, but uses participles more frequently than Athpare and Camling [Tsamling]. Most subordinate clauses have fully inflected verbs followed by a case marker (often without and intervening nominalizer) or some other subordinator; and

l. make extensive use of compound verbs (pp.8-10).

Apart from these general observations, Ebert has made some generalizations on the phonemic inventory part as well. These languages lack fricatives except for Isl and/hi. Voiced stops are rare phonemes in the SE languages. Initial /g, gh, j, jh/ are restricted mainly to loan words. Her observation is only partially true, which will be explained later. In Limbu voiced stops are ailophones of unvoiced consonants after nasal, glottal stop and in intervocalic position; the only voiced phoneme in final position is /bl (la:b * la:ba (Limbu-Nepali- English Dictionary (hereafter LNED 2002:413 and 613 >> 'moon' vs. la:p 'wing'). In the northern languages Thulung and Khaling voiced and voiceless initials are approximately equal in frequency. Aspiration is phonemic in all the languages. Although these generalizations seem to be simplistic at first sight, her phonemic inventory is in a loss at least in the case of Camling, Bantawa and Limbu as van Driem (1997: 474) points out, "Tibeto-Burman comparativists are at a loss to distinguish with confidence between loan words, the result of sound laws, and the effect of analogy". One obvious reason is mainly because there are no exhaustive research on the phonetics and phonology of these languages.

Till this date no linguist is confident on the possible total number of phonemes in any Kiranti language given to him/her whatever literature available. "The-elephant-and-the-blind man" principle is at work. Another reason of this inconsistency or loss is that it is difficult to find out sufficient or any written records of these languages. As a result, Ebert is also inconsistent while making the phoneme inventory of Camling [Tsamling]. In this grammar (pp.14), she has listed twenty-nine consonant phonemes, /p, ph, b, bh, t, th, d, dh, c, ch, j, jh, k, kh, g, s, h, m, mh, n, h, n, nh, w, 1, lh, r, rh, y/ and six vowels /a, e, i, o, u / having (A) as optional. This inventory is self-contradicted in her latter grammar of Camling (1997b: 8-10) in which she has mentioned thirty consonant phonemes out of which four phonemes, e.g. /j/ [dz], /jh/ [dzh], (p.9) and /gh/ as optional. The number of optional vowels is increased up to three, e.g. /[??], A and b/ (p.10). This inventory has been challenged with twenty-seven consonant and eight vowels by B.S. Yalungcha's T samling-Nepali Dictionary.

Ebert (p. 14) is nearly accurate on listing the phonemes of Bantawa (p. 16) but she has missed out one glottal stop (?) phoneme, which has been listed in Oik Bantawa's (1998) grammar. Bantawa (ibid 2ft) has listed only six vowels, whereas Ebert has listed seven vowels with /A/ as optional (p.16). Similarly the LNED (2002) has over-shadowed all other previous available literature on the Limbu language and linguistics. Ebert's grammar is no such exception. The LNED (p.19) has listed twenty-five consonants out of which nine are allophones and only sixteen consonants have phonemic status, whereas Ebert (p.140) has listed eighteen consonants as phonemes. The number of vowels is also inconsistent between Ebert (p.16) and the LNED (p.19). This could have been examined closely and explained or justified if she had provided the distribution of phonemes accounted with considerable data. To add one more phonological feature of the Kiranti languages is that all of them are not tone-prone.

At the end of this chapter, the author has provided some additional notes on the phonemic transcription of Limbu, Bantawa, Thulung, Khaling and Nepali. She has critically pointed out that although N.K. Rai's (1985) dissertation differentiates between apico-alveolar (OFT) and dental stops (d/t), there seem to be no nominal pair (p.17) of O/T in Bantawa. Ebert's another clarification is on the orthography of Roman-Gurkhali 'ch' unaspirated and 'chh' aspirated to the transcribed as /cl and /ch/ respectively in accordance with linguistic convention. But ironically her own transcription of [tsamlin] has been misunderstood as (kremlin] by her book reviewers (Bhattarai and Vihwol 1999: 135-137). These transcription notes have justified her consistent transcription in the grammar in relation to its phonetic literature available.

In chapter 2, 'The verb' (pp.19-74) focuses on stems and their variation, transitiveintransitive and causative stems, non/finite forms, person and number affixes, basic tense, mood, converbs, participles, compound verbs and so on. Like Nettle's (1999:5) concept of a human linguistic pool, Kiranti languages are amazingly linked to each other in their linguistic pool. Here is an example of infinitive marker from the morphology of these languages: Limbu [Yakthumba] '-ma', Athapare '-ma', Bantawa '-ma', Rodung [Caroling] '-ma', Thulung '-mu', Khaling '-ha' (p.55), Wambule [RaDhu] '-cam', Jerung 'kha' (Rai 2002), and Koits [Sunuwar] '-ca' and Bayung [Babing] 'co' (my data). But on the contrary, Rai's (2002) data on Khaling '-ne' contradicts with that of Ebert. Although Ebert's data is based on secondary source, it seems to a certain extent that the western linguists also tend to nativise the Kiranti phonology, as they prefer to call them aliens or exotic or heathen at their own ease. Another point where she has missed to tap the Caroling nativizing morphemes are: '-muna' and '-ba/lama' misinterpreted as auxiliary. There is another dialectal variation '-bala' of these indivisible nativizing morphemes.

Similarly, chapter 3 explores pronouns, number, gender, numerals, classifiers, case/direction markers and nominalizing morphemes. All these languages represented here have first person dual exclusive -inclusive, and first person plural exclusive -inclusive distinction. All of them have three (singular, dual, plural) numbers. Possessive prefix for first person singular' -a' is common for all these languages except for Bantawa '-irj'. Classifiers and numbers are 'seldom used' (p.79) in those languages. Contrary to Ebert's claim, some Wentern Kiranti languages like Wambule [RaDhu] and K6its [Sunuwar] use 'nimpha' as numeral classifier and most of these languages have numbers for counting in order to meet their communication needs. The only reason these languages might have used Indo-Aryan Nepali numbers while communicating because of the country's assimilative socio-politicohistorical and monolithic language policy adopted for 230 years of history.

Chapter 4, 'Deixis and location' (pp.90-99) describes the fascinating part of space coding in Kiranti. They distinguish high, low and level locative in vertical case. Adverbs are always preceded by deictic plus verticality. At least three types of vertical verbs are common in Kiranti. But Ebert's data on Caroling (Chamling) and Bantawa for 'up above' contradict to each other with that of Rai (2002). This contradiction has created confusion for a comparativistl typologist reader and researcher as well. Her data 'pyupa-mo' cow-GEN (p.99) and 'gai-wa' cow-ERG (p.100) has lost its consistency in vocabulary use and semantic aspects.

Chapter 5, 'Simple sentences' (pp.100-111) and chapter 6, 'Complex sentences' (p.112-137) are interrelated to each other exploring all possible syntactic structures in Kiranti. The basic word order is s(ubject), o(bject)and v(erb). This order of Kiranti satisfies Greenberg's non-absolute universal 5 and 21 (Song 2001:6-7 and Comrie 1981:19) having postposition and the 'np' order string as-

'{DEM+GEN/poss (pron)} +NUM +ADJ/ATTR +poss (prefix) +N' (p.100).

The Kiranti copula and comparative structures, as case markers in Athpare, Caroling and Thulung'-lai' (Nep p.81) in chapter 3, have freely borrowed comparative markers in their structure, e.g. 'bhanda' (pp.106-7 Nep.) in Athpare and Bantawa. So does Camling in its reportive particle 'raicha' (pp. 191-240 text and wherever text cited). The ergative constructions exhibit a pattern of split ergativity based on a person hierarchy. As a result, ergativity is rather superficial trait of Kiranti morphology, where all third person or demonstratives are marked but not all first persons.

The complex sentences in Kiranti are basically of two types on the basis of degree of reduction. In the non-finite verb, which does not carry finite tense or person markers; subjects are always deleted (p.112). These sorts of clauses are maximally reduced, whereas the minimally reduced clauses are finite. This trait occurs only in Athpare (ibid). Ebert's generalization of western and northern languages generally and frequently having non-finite clauses, e.g. Hayu (which is not represented in the grammar) is mentioned only by way of reference without data for evidence. Her hasty claim that Kiranti has no coordination of sentence (p.12) is partially justifiable. The origin of subordinators in Kiranti, are commonly case markers and inflectional morphemes. In most Kiranti languages, the variational functions thus differ semantically (Matisoff 1978) of these verb inflections, e.g.-'sa' in Kiranti changes its meaning as a chameleon in pragmatically-oriented use.

As a whole, the SKL: CGr is magnum opus. The students of T-B linguistics in general and researches on T-B Kiranti languages in particular can immensely take advantage of it. Naturally, this is purely a linguistic description by a professional linguist and has less applicability in pedagogy. Thus, it is almost not accessible for the general readers. But one can easily extract material or structures out of this grammar for the purpose of writing a pedagogical grammar.

The linguistic map (p.9) of the T-B Kiranti languages has provided a fair idea of geographical location of these languages where spoken. However, the book has no single photographs of informants and has no index in it. If the author has included index, it would have facilitated reading. So far as the glossing is concerned, she used the interlinear translation (IT) programme developed by SIL (1998 version) except for Athpare. This has been done nicely. While reading and observing the narrative texts, one can actually trace the influence of the behaviourist method which had entered linguistics via Bloomfield's writings itself manifested slogan as, "accept everything a native speaker says in his language and nothing he says about it" (quoted in Sampson 1980:64). This influence has possessed confusion and ambiguity in Ebert's grammar, e.g., phonemic inventory (pp.13-18), 'bra vs. kwam' (p.85), 'ghum-balam' (p.72) in Canaling blocks the phonemic inventory. The contradiction with Dik Bantawa's (1998:34) 2sG imsa vs. imse 'sleep' (nos. IViii) means the data have to be rectified and accept what the native speakers say about their language. Another problem in her grammar is the free intervention of the Nepali loans which these languages have their own native vocabulary, e.g. 'sya?l' (Nep. p. 81) and 'ghDr' (p.124). The LNED (2002:57 and 672) has listed the native vocabulary 'Kidhi:ppa or Kidhiruppa' for 'jackle'. It has posed a serious problem of linguistic identity in Caroling as well as in Athpare also. Ebert's glossing -MAN (p.70) has not been mentioned in abbreviations. There is one typing error apetivizer* (p.7). Some of her sentences, e.g. mi-kota-hine 'you have seen it' (p.46) is not free from grammatical lapses to which Abbi (1994:77) phrases as 'ungrammatical b.ag'.

Despite such negligible errors of a linguist as non-native speaker, it suffices to say that Ebert's effort is Herculean. Moreover, it would have been more Herculian-like if she had added only one or two languages from Wallo Kirant 'Hither Kirant' (Grierson 1909:274 and 316) like that of Michailovky's 'Phonological typology of Nepal languages' (1998) so that a considerable number of Kiranti languages would see the daylight of the western as well as eastern world of linguistics and linguists. The representation of only six out of not less than thirty-two T-B Kiranti languages is meager if not mean. The larger representation of the 6 languages would mean the more accurate comparison and generalizations. However, Ebert's grammar has its own place in the literature of Kiranti Linguistics.

Appendix A:
Regional distribution of Kiranti languages in East Nepal

Wallo Kirant Majh Kirant Pallo Kirant
(Western Kiranti) (Central Kiranti) (Eastern Kiranti)

1. Khaling 1. Sangpang 1. Lohorung (North)
2. Dumi 2. Kulung 2. Lohorung (South)
3. Koi (Koyu) 3. Nachering 3. Yamphu/Yamphe/
 Newahang
4. Bayung (Bahing) 4. Mewahang 4. Limbu
5. Thulung 5. Saam/Pongyong 5. Chathare
6. Koits 6. Cukwa/Pohing 6. Athpare
7. Lingkhim 7. Dungmali/ 7. Chhilling/
 Chhulung/Chhitang
8. Hayu 8. Waling 8. Mugali
9. RaDhu (Wambule 9. Khandung 9. Phangduwali
10. Jerung 10. Bantawa 10. Lumba-Yakkha
11. Tilung (Tilling) 11. Puma 11. Yakkha
12. Coskule (no data) 12. Chamling 12. Belhare
13. Dorungkeca (not
data)

Cited also in Gurung 2004: 61 from Gerd Hansson 1991a)

References

Abbi, Anvita. 1994. Semantic Universals in Indian Languages. Simla: Indian Institute of Advanced Study.

Allen, Nicholas J. 1975. A Sketch of Thulung Grammar. Ithaka: Cornall University.

Bantawa, Dik. 1998 [2055 V.S]. Bantawa (Rai) Vyakaran [in Bantawa-Nepali]. Kathmandu: Laxmi Prakashan.

Benedict, Paul K. 1972. Sino-Tibetan: A Conspectus (contributing editor: James A. Matisoff) Cambridge: Cambride University Press.

Bhattarai, K. and D.Viwol. 1999 [2056 V.S]. 'Pra Karen H. Ebertko Kaemling' [Review article in Nepali]. Sayapatri. Vol.2:3. 135-137.

Comrie, Bernard. 1981. Language Universals and Linguistic Typology (2ndedition 1989). Oxford: Basil Blackwell Ltd.

Crystal, David. 1989. A Dictionary of Linguistics and Phonetics (2nd edition 1991). Oxford: Blackwell Publishers.

2000. Language Death. United Kingdom: Cambridge University Press.

Ebert, Karen H. 1997b. Camling [Chamling]. Munchen: Lincoln Europa.

--1997. A Grammar of Athpare. Munchen: Lincom Europa.

Grierson, George A. (ed.). 1909. Linguistic Survey of India (reprinted in 1990 and 1994). Vol.3, Part 1. New Delhi: Low Price Publications.

Gurung, Harka. 2004. Janajati Serophero [in Neapli]. Kathmandu: Nepal Adivasi Janajati Mahasangh.

Hansson, Gerd. 1991a. The Rais of Eastern Nepal: Ethenic and Linguistic Grouping (edited & provided with an introduction by Wemer Winter) Kathmandu: LSN & CNAS, Tribhuvan University.

Limbu-Nepali-English Dictionary (2nd edition). 2002 [2059 V.S]. Edited by late Iman Sing Chemjong et al. Kathmandu: Royal Nepal academy.

Matisoff, James A. 1978. Variational Semantics in Tibeto-Burman: The 'Organic' Approach to Linguistic Comparison. Philadelphia: Institute for the Study of Human Issues.

Michaolovsky, Boyd. 1988. La Langue Hayu (Nepal). Paris: CNRS.

--1988. 'Phonological typology of Nepal languages.' Linguistics of Tibeto-Burman Area. Vol. 11: 2. 25-501

Nettle, Daniel. 1999. Linguistic Diversity. Great Britain: Oxford University Press.

Nishi, Yoshi. 1992. 'A survey of present state of our knowledge about the Himalayan Languages.' 25th International Conference on Sino-Tibetan Languages and Linguistics, Berkeley.

Pokharel, Madhap P. 1994 [2051V.S]. 'Raiharuko bhasik samasya '[in Nepali]. Namsam. Kirat Rai Yayokkha.

Rai, Novel K. 1985. A Descriptive Study of Bantawa. Unpublished PhD Diss., Deccan College, University of Pune.

--2002. 'Spatial deixies in some Kiranti languages'. Paper presented at the 23rdannual coference of the Linguistic Society of Nepal, Tribhuvan University, Kirtipur, 26-27 November.

Sampson, Geoffery. 1980. Schools of Linguistics: Competition and Evolution. London: Hutchinson and Co.

Shafer, Robert. 1953. 'East Himalayish'. Bulletin of the School oj Oriental and African Studies. Vol.15: 2.

Song, Jae J.2001. Linguistic Typology: Morphology and Syntax. Singapore: Pearson Education Pvt. Ltd.

Toba, Sueyoshi. 1984. Khaling. Tokyo.

van Driem, George. 1987. A Grammar of Limbu. Berlin: Muton de Gruyter.

--1992. 'In quest of Mahakiranti'. Contributions to Nepalese Studies. Vol.19:2. 241- 247.

--1993b. A Grammar of Dumi. Berlin: Muton de Gruyter.

--1997. 'Sino-Bodic'. Bulletin of the School of Oriental and African Studies. 60: 3.455-488.

Yalungcha, B.S. 2002 [2059 V.S.]. Chamling-Neapli Shabdakosh [Tsamling-Nepali Dictionary]. Kathmandu: Ms Hem Kumari Tsamling for Kirat Rai Yayokkha.

Note

Some examples of Kiranti languages from Wallo Kirant (Western Kiranti) include: Jerung, Koits (Sunuwar), Bayung and Wambule (RaDhu) by way of crossreference in our discussion

Notes

(1.) The term 'phylogenetic' has preferably been used by Nettle (1999:115) after Nicholas' (1990) term 'genetic' in order to avoid confusion with genetics in the sense of DNA. Therefore, I have also here replaced Ebert's 'genetic' with phylogenetie.

(2.) The first draft of this review was submitted to Prof. Dr. Anvita Abbi, Centre of Linguistics and English, Jawaharlai Nehru University for MPhil/PhD coursework (Course LE630E Structures of Lesser Known Languages) evaluation in March 2003. It was slightly revised in June 2004.