Kusunda language

For decades the Kusunda language was thought to be on the verge of extinction, with little hope of ever knowing it well. The little material that could be gleaned from the memories of former speakers suggested that the language was an isolate, but, without much evidence, it was often classified along with its neighbors as Tibeto-Burman. However, in 2004 three Kusundas, Gyani Maya Sen, Prem Bahadur Shahi and Kamala Singh,^[5] were brought to Kathmandu for help with citizenship papers. There, members of Tribhuvan University discovered that one of them, a native of Sakhi VDC in southern Rolpa District, was a fluent speaker of the language. Several of her relatives were also discovered to be fluent. In 2005 there were known to be seven or eight fluent speakers of the language, the youngest in her thirties.^[1] However the language is moribund, with no children learning it, since all Kusunda speakers have married outside their ethnicity.^[1]

It was presumed that the language became extinct with the death of Rajamama Kusunda on 19 April 2018.^[6] However, Gyani Maiya Sen and her sister Kamala Sen-Khatri contributed in further data collection, language training and revival of the language.^[7] The sisters, together with author and researcher Uday Raj Aaley, have been teaching the language to interested children and adults.^[8]

Aaley, the facilitator and Kusunda-language teacher, has written the book Kusunda Tribe and Dictionary.^[9] The book has a compilation of more than 1000 words from the Kusunda language. Uday Raj Aaley and Timotheus A. Bodt published “New Kusunda data: A list of 250 concepts,” in 2020.^[10] Nepal Archive has offered a free online basic to advanced level language lessons in their website.^[11]

David E. Watters published a mid-sized grammatical description of the language, plus vocabulary (Watters 2005), although further works have been published since.^[12] He argued that Kusunda is indeed a language isolate, not just genealogically but also lexically, grammatically and phonologically distinct from its neighbors. This would imply that Kusunda is a remnant of the languages spoken in northern India before the influx of Tibeto-Burman- and Indo-Iranian-speaking peoples; however it is not classified as a Munda nor a Dravidian language. It thus joins Burushaski, Nihali and (potentially) the substrate of the Vedda language in the list of South Asian languages that do not fall into the main categories of Indo-European, Dravidian, Sino-Tibetan, and Austroasiatic.

Before the recent discovery of active Kusunda speakers there had been several attempts to link the language to an established language family. B.K. Rana (2002) maintained that Kusunda was a Tibeto-Burman language as traditionally classified. Merritt Ruhlen argued for a relationship with Juwoi and other Andamanese languages; and for a larger Indo-Pacific language family, with them and other languages, including Nihali.^[13]

Others have linked Kusunda to Munda (see Watters 2005); Yeniseian (Gurov 1989); Burushaski and Caucasian (Reinhard and Toba 1970; this would be a variant of Gurov's proposal if Sino-Caucasian were accepted); and the Nihali isolate in central India (Fleming 1996, Whitehouse 1997). More recently a relationship between Kusunda, Yeniseian and Burushaski has been proposed.^[14]

Kusunda has several cases, marked on nouns and pronouns, three of which are the nominative (Kusunda, unlike its neighbors, has no ergativity), genitive, and accusative.^[1]

Other case suffixes include -ma "together with", -lage "for", -əna "from", -ga, -gə "at, in".

There are also demonstrative pronouns na and ta. Although it is not clear what the difference between them is, it may be animacy.

Subjects may be marked on the verb, though when they are they may either be prefixed or suffixed. An example with am "eat", which is more regular than many verbs, in the present tense (-ən) is,

Other verbs may have a prefix ts- in the first person, or zero in the third.

Donohue, Gautam & Pokharel (2014) argue that Kusunda lacks what typologists call "standard negation" – an unambiguous means of negating declarative verbal main clauses – contrary to the analysis in Watters et al. (2006), which had described several negative verbal suffixes.^[15]

Imperative clauses do distinguish polarity unambiguously, with the suffix -go for commands and -nin for prohibitives:^[15]

libuŋ ə-go	play do-IMP	'Play!'
libuŋ ə-nin	play do-PROH	'Don't play!'

Nonverbal predicates can also be clearly negated, using the negator otoq ('is not') for identity or the negative existential qaʕ-u ('does not exist') for existence:^[15]

tsi-je gimi	1SG-GEN money	'(That's) my money.'
tsi-je gimi otoq	1SG-GEN money NEG	'(That's) not my money.'

For non-imperative verbal clauses, however, no dedicated negative morpheme exists. What Watters et al. (2006) had analyzed as negative suffixes such as -aʕu and -daʕu are reanalyzed by Donohue et al. as combinations of the nominalizing suffix -da (expressing a "manifest characteristic") with the negative existential qaʕ-u. The resulting construction literally asserts the nonexistence of a nominalized property. For example:^[15]

gina g-əm-da qaʕ-u

3SG 3-eat-CHAR not.exist-IRR

'He/she did not eat.'

Two pieces of evidence support the reanalysis. First, the clitic -ba ('also, even') can be inserted between the nominalized verb and the negative existential (gina g-əm-da-ba qaʕ-u 'He/she did not even eat'), indicating a word boundary rather than a single suffix. Second, the construction is incompatible with future time reference (*goraq gina g-əm-da qaʕ-u 'He/she will not eat tomorrow' is ungrammatical), which follows from the semantics of -da, since it refers to an already-realized characteristic.^[15]

To express negation in future contexts, the only available strategy is the irrealis suffix -u, but this is inherently ambiguous between positive and negative readings:^[15]

goraq gina g-əm-u-a

tomorrow 3SG 3-eat-IRR-OPT

'He/she will not eat tomorrow.' / 'He/she will eat tomorrow.'

This situation is typologically unusual, since most known languages have some unambiguous way of negating verbal predicates across all tense contexts.^[15]

• Kusunda Swadesh list (basic vocabulary)
• Kusunda word list (1127 words)
• Gyani Maiya Sen-Kusunda
• Kamala Sen-Khatri
• Uday Raj Aaley
• Burushaski
• Nihali language
• Substratum in Munda languages