The Origins of Postverbal Negation in Kuki-Chin

Note: It will be easiest to format yourpaper correctly if you simply type into this

document. Save the file as:AuthorSurname_AbbreviatedTitle_NEILS#_ForRev

iew.doc or .pdf (or better, both .docand .pdf). Example:

Morey_AhomTones_NEILS2_ForReview.doc.

The Origins of Postverbal Negation inKuki-Chin

Scott DeLanceyUniversity of Oregon

1. Introduction

In the vast majority of the Eastern andWestern Tibeto-Burman languages the finiteverb is negated by a preverbal particle*ma-, which is reconstructible for Proto-Tibeto-Burman (Matisoff 2003: 488). But in asubstantial number of languages in andaround North East India, we find variouspatterns in which negation is markedpostverbally:

The overall pattern of the position ofnegative morphemes in Tibeto-Burman canbe summarized as follows. VNeg order isdominant in an area corresponding roughly

1





to the section of India east andnortheast of Bangladesh, including mostBodo-Garo, Tani, and Kuki-Chin languages,while NegV order is dominant in twoareas, one to the west, in Bodic, and oneto the east, including Nungish, Jinghpo,Northeast Tibeto-Burman, and Burmese-Lolo-languages. (Dryer 2008: 70)

This paper is a preliminary exploration ofthe reasons for this variation. I willmostly restrict my attention to Kuki-Chinlanguages, following on observations byKonow (1904) and C. Y. Singh (1992), andleave consideration of analogous phenomenain “Naga”, Karbi, Boro-Garo, and otherlanguages for another time.

2. The negative prefix *ma-

The *ma- negative occurs across the family,and is uncontroversially reconstructed tothe proto-language. Outside of our area italways preverbal, and almost always aprefix, though it is also attested as aphonologically independent particle, as inLahu. Thus Matisoff reconstructs a “negativeadverb” rather than a prefix. But even inLolo-Burmese it always precedes the finiteverb (Bradley 1979:372) and in everylanguage where the phonological structure ofthe language allows, it is prosodicallydependent on the following verb.

2





2.1. *ma- outside of North East India

Tibetic languages provide useful exampleswhich illustrate the basic TB negativeconstruction and typical secondarydevelopments. In Classical Tibetan thenegative particle ma1 always directlyprecedes the finite verb:

(1) ngas ma bor roI.ERG NEG throw FINAL‘I didn’t throw [it]!’

The Tibetan writing system has no means ofindicating phonological dependence of onemorpheme on another, but in all contemporaryTibetic languages the negative form isattached as a prefix to the verb, andpresumably this was also the case in olderforms of the language.

Two other facts about Tibetan negationprovide a model which can explain some ofthe developments in negative marking inKuki-Chin and elsewhere. First, note that,as in other verb-final languages, negationtypically attaches to the final verb in asequence, so that in constructions with1 There are two forms of the negative particle inClassical Tibetan: ma with perfective and imperativestems, and mi with present and future stems. Only thefirst is relevant to our present concerns.

3





auxiliary verbs, the negative prefixattaches to the auxiliary rather than thelexical verb. Over time, as auxiliaryconstructions become more grammaticalized,this can result in constructions in whichthe negative marker occurs between the verbstem and a TAM suffix, as in Lhasa Tibetan:

(2) phyin-songwent-PERF ‘[S/he] went.’

(3) phyin ma-songwent NEG-PERF ‘[S/he] didn’t go.’

In Dryer’s (2013) scheme this constitutesa shift from pre- to postverbal negation,but for our present purposes it does notcount as such, because the negative morphemecannot occur as the only verbal operatorfollowing the verb, but must be followed bya TAM element on which it was originally aprefix.

The second phenomenon of interest inTibetan is the special negated forms of thecopulas. The equational and existentialcopulas yin and yod have irregular negativeforms min and med. In the modern languagesmany tense/aspect forms are based on thesecopulas. Since a copula auxiliary is the

4





final verbal element in a clause, it carriesnegation, resulting in negativeconstructions which could be synchronicallyanalyzed as having tensed postverbalnegative forms, as in Lhasa Tibetan:

(4) nga zos=gi yodI eat-IMPF PERSONAL‘I am eating’

(5) nga zos=gi medI eat-IMPF PERSONAL.NEGATIVE‘I am not eating’

We will see forms in NEI languages whichappear to have a similar origin.

2.2. *main North East India

The original negative construction occurs insome languages of NEI. Mongsen Ao showsconsistent prefixing of the negative mə-,even in the presence of TAM suffixes (Coupe2008: 292):

(6) mə-tʃhuwa-əɹNEG-emerge-PRES‘doesn’t return’

(7) mə-phuʔ-i-uʔNEG-steal-IRR-DEC

5





‘won’t steal’

Mongsen Ao also has a negative suffix,which co-occurs with the prefix, only inpast tense:

(8) mə-tsəpha-laNEG-fear-NEG.PST‘did not fear’

Ao is conservative among its neighbors;many other “Naga” languages have innovativenegative constructions. But in some of theselanguages we find frozen forms which attestto the earlier use of the olderconstruction. For example, Sumi (Amos Teo,personal communication) regularly negatesthe finite verb, either stem or auxiliary,with postverbal -mo (see Section 3.2):

(9) pa=je à-lè phè-mo3SG=TOP NRL-song sing-NEG‘He will not sing.’

But the preverbal negative is preserved infossil form in mtha ‘not know’ (ithi ‘know’)and composite postverbal operators -mphi ‘notyet’ (aphi PROGRESSIVE) and -mla ‘unable to do’(-la ‘able to do’).

In many Northwest Kuki-Chin languages wefind a reflex of *ma- occurring as part of

6





the string of verbal suffixes, alwayspreceding some other TAM morpheme. Anexample is Anal (Sengoi Singh 1992). Likeother NW KC languages, Anal uses theprefixal indexation paradigm in affirmativeclauses, and the postverbal paradigm innegative clauses (DeLancey 2013a, b).Clearly the negative construction originatedas *ma- prefixed to a postverbal auxiliaryni:

(10) ni kə-ca-waI 1SG-eat-TNS‘I eat.’

(11) ni ca-mə-ni-ŋI eat-NEG-TNS-1SG‘I do not eat.’

The only reported example that I know ofof an apparent reflex of *ma occurringpreverbally is Daai Chin am (So-Hartmann2009: 252-254):

(12) ah khhyu:=noh ta am3SG.POSS wife=ERG FOC

NEG

dang-yah mjohsuspect=NON.FUT EVID

7





‘His wife did not suspect, it is told.’

While the resemblance to the pan-TBnegative prefix is obvious, both the form(am rather than ma) and the phonologicalindependence of this form from the verb stemremain to be accounted for.

3. Postverbal #mak2

The most widespread postverbal negativeconstruction is clearly related to PTB *ma-.This is a postverbal, often phonologicallyindependent syllable mak or maʔ, whichappears to have the same kind of origin asthe Tibetan min and med forms described inSection 2.1. Some languages are reported ashaving postverbal ma, with no finalconsonant. In some cases this may simply bea case of failing to transcribe a finalglottal stop, or it could represent furtherphonological erosion mak > maʔ > ma. Theseforms occur in a number of languages inNorthern Naga3 and the languages formerlylumped together as “Naga”, includingLiangmai, Maram, Maring, and Zeme (Marrison2 I adopt Bauman’s (1975) practice of using # to indicate a comparative set which has not yet been systematically reconstructed.3 In the Northern Naga languages the final /k/ is a 1SG agreement marker, but in the other “Naga” and KC languages it must have a different origin.

8





1967: 127). Within Kuki-Chin they arereported only in the Northwestern subbranch;in the next section we will see someexamples.

3.1. #mak and its origin

The Linguistic Survey of India reports someform of mak as a postverbal negator in allof the “Old Kuki”, i.e. Northwestern KC,languages. In modern descriptions it appearsthat these forms are found only in non-future tenses. For example, Koireng (C. Y.Singh 2010:114-5), and Moyon (Kongkham 2010)have distinct negative morphemes in therealized/non-future and unrealized/futuretense. The realized negative is -mək-; theunrealized negative represents another formwhich I will discuss in Section 4.3:

Realized negative

Unrealized negative

1SG

-mək-iŋ -no-ni-ŋ

1PL

-mək-uŋ -no-mə-ni

2S -mək-ci -no-ti-ni4

4 The Koireng Grammar has a misprint in example 24, p. 114: -niti should be -tini. The correct form is given in

9





G2PL

-mək-ci-u -no-ti-ni-u

3SG

-mək-e -no-ni

3PL

-mək-u -no-ni-u

Table 1: Koireng negative paradigms

The fact that the negative forms -mək- and-no- are followed, and in the 2nd personunrealized forms, preceded, by agreementmorphemes argues that they originated asauxiliary verbs. This is probably also thehistory of no, which we will return to below(4.3). But mək is more complex; it appearsto have the same kind of origin as Tibetanmin and med, that is, the coalescence ofwhat was originally a copula with a negativeprefix.

This hypothesis concerning the origin ofmak is not original; already in the LinguisticSurvey of India Konow had discerned the relationbetween our postverbal mak forms and thegeneral Tibeto-Burman preverbal *ma-:

It is … probable that māk is acompound, consisting of thenegative prefix ma and a verb

the text above on p. 114.

1





substantive … On the whole it maysafely be assumed that thenegative suffixes in the Kuki-Chinlanguages contain a negative prefixwhich is not, however, prefixed tothe principal verb but to the oldcopula which is added as anassertive suffix. The negativeverb would, accordingly, be acompound. The negative particle isusually inserted between the rootand the tense suffixes, a factwhich well agrees with thesupposition of its being a verbforming a compound. (Grierson1904: 19, emphasis original)

There are independent arguments for acopula #yak or #yik at the root of theNorthern and Northwestern KC “agreementwords” (DeLancey to appear), so we mightpropose that mək < *ma-yak. But we also haveto reconstruct a negative copula #kay for PKCand even earlier (Section 4.2), so anoriginal source *ma-kay is also plausible.

3.2. Open syllable /m-/ forms

In a few Northern and Northwestern KClanguages there is a postverbal #ma formwithout the final -k:

1





Lamkhang (Thounaojam and Chelliah 2007: 63)

(13) ə-čak-mə2-eat-NEG‘Don’t eat!’

Thadou (Haokip 2012: 12)

(14) kà dám mɔɔ eeI well NEG DECL‘I am not well.’

These could be explained away asreflecting phonological reduction of oldermak, but it is not obvious that this is theonly possible explanation for these forms.This question cannot be resolved withoutmore detailed information on otherNorthwestern KC languages.

4. Other postverbal negative forms

There are three well-attested postverbalnegative forms besides mak, probablyreconstructable as *law, #kay, and *no. Forthe first two there is evidence outside oftheir current status to show that they wereoriginally negative copulas.

1





4.1. *law

VanBik (2009: 253, #1035) reconstructs anegative morpheme *law for PKC, based on HakaLai lăw, Fallam Lai làw, Mizo lò, Thadou lòw(Haokip 2012); see also Bawm Chin lo (Reichle1981), Sukte -ləw (C. Y. Singh, personalcommunication). Outside of KC proper, noteMeithei te (non-future) / loy (future)(Singh 2000: 144-5). More recently VanBik(2013) suggests that this originates in*law1, *lawʔ2 ‘disappear/lose’ (2009: 249,#1011).5 Examples of this form in Kuki-Chinare:

Mizo (Chhangte 1993:92)

(15) kán-ṭhû-toʔ-low1SPL-sit-PPF-NEG‘We are not sitting anymore.’

Hakha Lai (Peterson 1998)

(16) law ʔa-ka-thloʔ-piak-lawfield 3SG.SU-1SG.OB-hoe2-BEN-NEG‘He didn’t hoe the field for me.’

5 Verbs in Kuki-Chin languages typically have two distinct stem forms, generally referred to as Stem 1 and 2; this distinction is indicated here and in the glosses to the examples by subscript numerals.

1





Falam (Yu 2007)

(17) a-thàj dîŋ a-sī-làw3SG-know IRR 3SG-be-NEG‘S/he shouldn’t find out.’

Thadou (Haokip 2012)

(18) ipîi nâ bɔɔl lòw hâmwhat you do2 NEG INTERR‘What is it that you do not do?’

4.2. #kay

Another negative form in Kuki-Chin is #kay;an example is the postverbal negator -kəy inSukte (C. Y. Singh, personal communication):

(19) ken ən ne-kəy-iŋI rice eat-NEG-1SG‘I do not eat rice.’

We will see additional examples in Section4.4. The origin of this form is not clear,but the forms bear a striking resemblance tothe Bodo-Garo negative existential copulagwi. Both may be related to Jinghpaw kòi‘avoid, shun’ (Hanson 1906:242) or ké‘scarce’ (Hanson 1906:232).

1





Mindat (Southern Chin) has a preverbalnegator which looks as though it could berelated to this form (Jordan 1969: 54-55):

(20) käh kah law khaiNEG 1SG come FUTURE‘I will not come.’

(21) käh ei pha ciNEG eat PERF FIN‘[He] has not eaten.’

4.3. *no

One more negative formative which occurs inseveral languages is no. We have alreadyseen this in the unrealized paradigm inKoireng (Section 3.1. Another example, alsofrom NW KC, is Chhothe (Jayalata Devi 1992),where negative -no is suffixed to the verb,preceding other suffixes:

(22) nəŋ dəŋ-ŋeyou know-TNS‘You know it.’

(23) nəŋ dəŋ-no-eyou know-NEG-TNS‘You don’t know it.’

1





(24) kəy bu bək-iŋI rice eat-1SG‘I eat rice.’

(25) kəy bu bək-no-ŋ-eI rice eat-NEG-1SG-INTERR‘I do not eat rice.’

In example (25) we see that negative -no-takes the 1st person agreement suffix,providing evidence for its verbal origin. InHmar it occurs as an uninflected particlewith a verb, but conjugated with thecharacteristic KC prefixes as a negativecopula (Baruah and Bapui 1996: 133-135):

(26) ká-fè: á nih1SG-go ASP FUTURE‘I am going.’

(27) ká-fè: nɔ: ní-ŋ1SG-go NEG FUTURE-1SG‘I do not eat rice.’

(28) zirtí:rtu ká-nihteacher 1-COP‘I am a teacher.’

(29) zirtí:rtu kán-nɔhteacher 1-NEG‘I am not a teacher.’

1





4.4. Languages with two postverbal negatives

Many Kuki-Chin languages use more than oneof these negative constructions. We havealready seen the occurrence of both *mak and*no in Koireng and Moyon (Section 3.1). InPaite (Northern KC) we see both *law and #kay(C. Singh 1992, N. Singh 2006):

The declarative negation is alsoformed by adding the suffix ‘key’to (1) a verb, (2) a be-verb, and(3) a copula verb. But in the caseof sentences containing the verb‘hi’ [the copula], the declarativenegation may also be formed byadding ‘ləw’ to the main verbalso. (N. Singh 2006: 152)

That is, a finite verb is negated bysentence-final key (N. Singh 2006: 152)

(30) əmáʔ ə-kəphe 3-weep‘He weeps.’

(31) əmáʔ ə-kəp-keyhe 3-weep-NEG‘He doesn’t weep.’

1





The copula hí used as an auxiliary may benegated the same way:

(32) əmáʔ kəp ə-híhe weep 3-COP‘He weeps.’

(33) əmáʔ kəp ə-hí-keyhe weep 3-COP-NEG‘He doesn’t weep.’

But alternatively the lexical verb may benegated instead, in which case the negativeform is ləw:

(34) əmáʔ kəp-ləw ə-híhe weep-NEG 3-COP‘He doesn’t weep.’

5. Conclusion

Postverbal negative constructions in Kuki-Chin have arisen through two broad pathways:grammaticalization of an auxiliary negatedwith the PTB *ma- prefix, andgrammaticalization of some other serializedverb, either an inherently negative copulaor another verb with inherently negativemeaning. In a brief survey we have seen each

1





of these pathways in different languagesinvolving different specific sourceconstructions. First, we can conclude fromthe multiplicity of negative constructionswith a single branch that this represents afairly recent process; if the development ofa new postverbal negative construction hadbeen completed by Proto-Kuki-Chin we wouldexpect the daughter languages to share thesame construction. Second, the fact that allthe KC languages have innovated postverbalnegation, and through several differentpaths, suggests a consistent tendency towardpostverbal negation in these languages. Wecould imagine a typological tendency, aspost-head operators are considered to becharacteristic of SOV languages, or an arealphenomenon, given that there is lessevidence for such a tendency in other TBlanguages.

References

Baruah, P. N. Dutta, and V. L. T. Bapui. 1996. Hmar Grammar. Mysore: Central Institute of Indian Languages.

Bauman, James. 1975. Pronouns and pronominal morphology in Tibeto-Burman. Ph.D. dissertation, University of California,Berkeley.

1





Bradley, David. 1979. Proto-Loloish. London: Curzon.

Coupe, Alexander. 2008. A Grammar of Mongsen Ao. Berlin: Mouton de Gruyter.

Dahl, Östen. 1979. “Typology of sentence negation.” Linguistics 17: 79-106.

DeLancey, Scott. to appear. “Morphological evidence for a Central branch of Trans-Himalayan (Sino-Tibetan).” Cahiers de Linguistique – Asie Orientale 44.2.

DeLancey, Scott. 2013a. “Verb agreement suffixes in Mizo-Kuki-Chin.” In G. Hyslop, S. Morey and M. Post, eds., Northeast Indian Linguistics V, 138-150. Delhi:Cambridge University Press.

DeLancey, Scott. 2013b. “The history of postverbal agreement in Kuki-Chin.” Journal of the Southeast Asian Linguistics Society 6:1-17.

Dryer, Matthew. 2008. “Word order in Tibeto-Burman languages.” Linguistics of the Tibeto-Burman Area 31.1: 1-83.

Dryer, Matthew. 2013. “Order of Negative Morpheme and Verb.” In: Dryer, Matthew & Haspelmath, Martin (eds.) The World Atlasof Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology. (Available online at http://wals.info/chapter/143, Accessed on 2014-01-14.)

2





Grierson, George (ed.). 1904. Linguistic Survey of India, Vol. 3, part 3: Specimens of the Kuki-Chin and Burma Groups. Calcutta: Office of the Superintendent of Government Printing. Repr. 1967: Delhi: Motilal Banarsidass.

Haokip, Pautang. 2012. “Negation in Thadou”.Himalayan Linguistics 11.2: 1-20.

Jayalata Devi, Naorem. 1992. A Brief Study of Negation in Chothe. MPhil dissertation, Department of Linguistics, Manipur University.

Jordan, M. 1969. Grammar of the Chin language. Appendix to Chin Dictionary and Grammar. Paris.

Peterson, David. 2003b. Hakha Lai. In: Thurgood and LaPolla (eds.), The Sino-Tibetan languages, 409-426. London and New York: Routledge.

Sengoi Singh, Sanasam. 1992. Negation in Anal. MPhil dissertation, Department ofLinguistics, Manipur University.

Singh, C. Yashawanta. 1992. “Negation in Kuki-Naga languages”. In Anvita Abbi (ed.), Languages of Tribal and Indigenous Peoples of India, 387-400. Delhi: Motilal Banarsidass.

Singh, C. Yashawanta. 2002. Tarao Grammar. New Delhi: Akansha.

Singh, C. Yashawanta. 2010. Koireng Grammar. New Delhi: Akansha.

2





Singh, Naorem Saratchandra. 2006. A Grammar ofPaite. New Delhi: Mittal.

VanBik, Kenneth. 2009. Proto-Kuki-Chin: A Reconstructed Ancestor of the Kuki-Chin Languages. (STEDT Monograph 89). Berkeley: Sino-Tibetan Etymological Dictionary and Thesaurus Project, University of California.

VanBik, Kenneth. 2013. “Grammaticalization in Kuki-Chin”. presented at the 46th International Conference on Sino-Tibetan Languages and Linguistics, Dartmouth College.

2

The Origins of Postverbal Negation in Kuki-Chin

Documents