The languages and cultures of the Mishmi

peoples

Roger BlenchKay Williamson Educational FoundationMcDonald Institute for Archaeological Research, CambridgeVisiting Professor, University of New England, ArmidaleAcademic Visitor, Museu Emilio Goeldi, Belem, Brazil

MPI, Nijmegen, 8th May, 2015

Linguistic diversity in Arunachal Pradesh

Arunachal Pradesh (and NE India in general) is a remarkable area for ethnolinguistic diversity (probably due to the mountainous terrain)Due to its special status in colonial times, and more recently proximity to China, its languages have been very little studiedThe region is an interface of three global religions, Buddhism, Hinduism and Christianity and their interaction with local religious traditionsThe societies are also structurally very diverseAnd there is now a strong urge among local peoples to document their own language, culture and customsThis presentation results from a field trip in February/March 2015 based on a collaboration with language committees for the Idu, Tawra and Kman languages

Ethnolinguistic

map of NE India

Mishmi: three peoples, one name

The Mishmi peoples of Arunachal Pradesh consist of three different groups, speaking three distinct languagesThese are the Idu [also Kera], the Tawra [=Taraon] and the Kman [=Miju]Two of these languages are presumably related, Idu and Tawra, but the third, Kman, is quite differentThese languages are usually classified as Sino-Tibetan, i.e. together with Tani and Chinese, but the evidence for this is very weakThey may well be language isolatesA fourth language, Meyor, spoken in Walong and Kibitho, has many common features with Kman, but is probably not related genetically

Mishmi: three peoples, one name

The Mishmi peoples also share many common cultural features which make them distinct from other peoples of Arunachal PradeshThe most important of these is polygamy, which is reflected in the very distinctive house type, the long houseIn this, each wife had her own space, with associated pig-rearing and the rooms were joined by a long internal corridorHeadhunting was also (formerly) practised as the custom of displaying the skulls of sacrificed animalsMultiple language registers (including poetic, hunting and shamanisticAll of this suggests centuries of common interaction, although we know almost nothing about the long history of the Mishmi

The Mishmi peoples

SociolinguisticsEach language has a few thousand speakers; estimates of levels of competence vary widelyThe main competing languages were formerly Tibetan and AssameseHowever, today Hindi is the most important language; schooling is in Hindi as is much of the mediaHence there is a widespread local perception that these languages are very threatened and that young people are losing their languageMy own observation is that the situation is not as bad as is locally feared, but there is a significant loss of specialised vocabulary, for example plant and animal namesFortunately, world religions have not made much impact

Linguistics: pre-existing work

None of these languages have been the subject of intensive modern workThey have been covered in Indian publications coming from the Arunachal Pradesh research directorateThese publications are only marginally worth having, as the phonology is extremely simplified and the semantics viewed through the lens of Hindi culture

Linguistics: pre-existing work

There have been some initiatives, apparently semi-clandestine, to analyse the languages from outside and produce primersThese always have the feature that they try and treat Tawra and Kman in the same document, which is culturally approved, but problematic linguistically

Linguistics: pre-existing work

Linguistics: pre-existing work

There has also been work on the Chinese side, on Idu, Daruang [Tawra] and Geman [Kman]There are only a few peripheral villages on the Chinese side, so it is difficult to tell if differences with mainstream speech are dialectal or issues of transcription and elicitationBroadly the phonology corresponds but the tones seem wildly over-differentiated

Linguistics: general profiles

The Mishmi languages are broadly very similar typologically, indeed they resemble most of the other languages of Arunachal PradeshConsonant inventories are quite small, but Idu in particular has both a great variety of contrastive prosodies and vowel articulations, reduced in Tawra and much reduced in KmanTone systems are basically two tones with rising and fallingNumber marking is very reduced, and there is a (small) system of numeral classifiersWord order is basically SOV, with VAs following the verb but there is significant variationIt is an open question as to whether there is any bound morphology

Phonology: IduIdu consonants

Bilabial

Labio-dental

Alveolar

Retro- flex

Palato-alveolar

Palatal

Velar Glottal

Plosive p, pʰ b, bʰ

t, tʰ d, dʰ

ʨ k, kʰ g, gʰ

ʔ

Fricative [s] ʃ hAffricate ʧ ,

ʧʰ ʤ

Nasal m n ɲ ŋFlapped rLateral l ɭApproximants

w ʋ ɽ y

Phonology: IduIdu vowels

Front Central BackClose i u [ũ]

Close-Mid e [ẽ] o [õ]

Open-Mid ɛ [ɛ] ə, ə ɔ [ɔ]Open a [a]

One of the most interesting vowels is the retracted schwa [ə]. Idu probably has contrastive vowel length for all vowels, and contrastive nasalisation, making a very large inventory

Phonology: IduTones

Like most regional languages, Idu has a relatively simple tone-system. There are two level tones, High and Low and a rising and [rare] falling tone.

fǔ cursefù suspendfû to sprinkle powder

Phonology: KmanKman consonants

Bilabial

Labio-dental

Alveolar

Retro- flex

Palato-alveolar

Palatal

Velar

Glottal

Plosive p, pʰ b

t, tʰ d

k, kʰ g

ʔ

Fricative [f]

s z

ʃ h

Affricate ts dz

ʧ , ʧʰ ʤ

Nasal m n ɲ ŋFlapped rLateral l ɭApproximants

w ʋ ɽ y

Phonology: KmanKman probably has nine vowels;

Kman vowelsVowels Front Centr

alBack

Close i iU u Close-Mid

o

Open-Mid

ɛ ə ʌ ɔ

Open a

It is uncertain that the two central vowels /iU / and /ə/ are distinct.Nasalised vowels and long vowels are marked in the two existing sources Das Gupta (1977) and Boro (1978). However, in striking contrast to Tawra and Idu, these are in fact absent. Phonetic nasalisation of final vowels is often heard due to a prior nasal, and sometimes final -ŋ is in free variation with a nasalised vowel.

Morphology: KmanA key feature of Kman are the many consecutive sequences of consonants without vowels. Some important examples are;

Kman CC and CCC sequencesGloss Kmanankle tlnbamboo sp. máy mbrtcough, productive

tt

cup wadngrass tfngreen krmpartridge mnmsnake ɽlsnake nest entrance

rbn

storm t.mk òwáŋwild buffalo tl.wì

This is very counter-intuitive to anyone raised on the traditions of writing European languages; we expect vowels to follow consonants. So attempts to write the language usually insert extra vowels. For example, the name of the language is often locally spelt ‘Kaman’ and in the Chinese source ‘Geman’.

ClassifiersNumber is marked with a uniform suffix in all languages and is not marked if plurality can be identified by other indicators such as a numeralAll three languages have classifiers, between nine and twelve, as well as collectivesThis is significantly less than the neighbouring Tani languagesMany nouns don’t take classifiers. Is the system building or collapsing?Bamboo appears to play a dominant role in the conceptual world they illuminateOnly Kman applies a classifier to humans

Classifiers: Idu Idu numeral classifiersForm Applied toala clothba mithunsbo treesbra spherical things (potatoes, beads, grains,

oranges)bru bangles, maize, bananasdo bamboo clusters, houses, dead deer, dead

pig, snake [some use only]go fish, ratspra leaves, paper, flat things, some frogspo packets, bundles, necklacespu elephants, chickens, cucumbers, papaya,

pumpkinto bamboo single plant

Classifiers: KmanKman numeral classifiersKman Semantic clusterbòŋ bamboo, genericbràt spherical things, oranges, grains, pebblesbrɨl small trees, twigs, bamboo splintersbyɔŋ long pieces of wood, bamboo, cane dùm short pieces of wood, bamboo, caneglà human beingsklòŋ flat things, leaveskòw human beingskròŋ longhousesnáw individualised living things, not plantsŋgòw half tubular shapes, bamboo split lengthways[1]

phál flat things, paper, plankphòŋ bamboo in clusterstàŋ cloth, mats, bagsthùm medium pieces of wood, bamboo, canetùl trees, bushes

Classifiers: TawraTawra numeral classifiersTawra Semantic clusterbra small round objects, but also cylindrical

objects like thick treesbra circular, small spherical objectsbriU l posts, cylindrical, longbru long thin objects, tree, banana, pen,

speardõ buildings, large man-made structureshadum large trees, bushes, clusters of bamboona thin flat objects, cloth, paper, leafpla circular flat objects, long sharp objects,

banknote, plankpum large round/square ? hollow objects, box,

elephant, birdti openingstyaŋ animals except elephant not fish, insects

or birds

Examples of Kman word order

katuwat bwí hula pay kiklashaman house came then shout the shaman came to the house and then began to chantbajar tay yey sawkə siU n syup yeymarket

go let us

so meat buy let us

let us go to the market so that we can buy meat

ki sut.ke wa ʤaŋ.minI dao.INST cut bamboo

Kman word order seems remarkably variable;

Anthropology: documenting indigenous communities

Mishmi peoples are equally concerned about cultural preservation and documentationAn interesting aspect of the region is the relatively successful negotiation of modern technology; mobile phones are common, but these have not lead to language loss and youth disaffection common elsewhere Indian museums as a whole are fairly poor on minorities. In AP, the state museum shows each minority in a diorama, static figures in so-called typical costume and activitiesNeedless to say village life isn’t like that and the possibility of using even vaguely modern technology has evidently passed them by.Museums do play an important role in other parts of the world in interacting with local communities and responding to their wishes, but not yet in India.

Anthropology of the Idu

Arunachal Pradesh State Museum

Kman

Tawra

Idu

What binds Mishmi together? The three Mishmi peoples, despite their distinct languages have much in common culturallyPart of that shared heritage underlies the attempts to draw together Mishmi as a single groupThis is not necessarily helpful in language workBut other features, such as architecture and shamanism are clearly held in common

ArchitectureThe three Mishmi peoples all share the tradition of the longhouse, which is connected with polygamyEach wife has a separate living space and these are accessed by a long internal corridor Idu longhouse

Kman longhouse

ShamansShamans are known as igu in Idu, gwak in Tawra and katowat in KmanThey are called upon to recite chants and conduct sacrifices for sick persons, sometimes with drumming and gongsPigs, chickens and mithuns are sacrificedThe chants are in a special language which is only known to the shamans

Idu ShamanThe language of shamans is not a ‘secret’ language, and people are quite willing to teach it to you if you have a command of the base language

Shared cultureAnother very striking common feature is the ‘trophy wall’ where the skulls of larger animals, either shot by a hunter or sacrificed, are displayed along the inner corridor of the house

Hunters’ trophies: Idu

Hunters’ trophies: Kman

Multiple language systems Idu, Tawra and Kman all share a system of multiple language registers These are;

a) ordinary speech b) speech of hunters c) speech of priests/shamans d) poetic/lyrical register

Multiple language systems The speech of hunters involves lexical substitution, the replacement of animal names and others by special forms, sometimes short poemsPriests’ speech is more complex, involving much language which is difficult to understand as well as lengthy descriptions of sacrificial animalsThe poetic/lyrical register is mainly about the substitution of poetic lexicon for ordinary words. The surprising thing is that we know nothing of the origin of these words.The Mishmi peoples seem to be unique in the development of so many speech registers. Something to be celebrated.

Developing writing systems [aka scripts]

The government publication systems were intended to provide a writing system for the languages, but since they were based on an entirely erroneous conception of the sound system, they were unlikely to succeed

There have been other attempts by (probably) missionaries to develop orthographies, but with a lack of continued input, these have had no impact

The CIIL has also sent emissaries to propose orthographies, but some of their suggestions, such as @ for schwa and non-diacritic nasalisation were never going to fly

Other proposals involve developing Devanagri, which will also fail for cultural reasons

Developing writing systems II The communities are now very anxious to develop orthographies, so work is under way to make this happen

An interesting development has been the local evolution of orthographies for texting and for Facebook.

At least one book has been published in Kman. However, these scripts are very tied to a small universe of expressive discourse, i.e. they can be understood because you already half-know what someone is likely to say

Tonal differences, for example, are represented by spelling workarounds

A children's story in Kman ‘Facebook script’

An earlier publication on Tawra and Kman writing

Developing writing systems III

So, although the development of a local orthography which works in context is in some ways good, it would be hard to teach in any consistent way

So it was decided to start from scratch in analysing the phonology and morphology to make consistent proposals for a writing system

Kman and Tawra culture

The orthography is quite

irregular but this type of publication

contains much of cultural interest

Achievements so far Initial documentation of all three languages has been carried out and stored in digital formThis has been written up as a series of proposals for writing systems which have been distributed to interested personsThese are highly preliminary, that they should be read and considered by the communityIn the meantime, the digital data is being transcribed fully and sent back to the community for further revision and additionsIn particular we hope to develop an agreed writing system in the next couple of years, along with first versions of dictionaries.Preparation of grammar sketches is under way

Idu Language Development Committee

THANKSTo the Idu, Kman and Tawra communities for interest and support

Special thanks to Dr. Miti Lingi, Sokhep Kri and Jogin Tamai for patient language work