P. 1
Kusunda Grammar

Kusunda Grammar

|Views: 16|Likes:
Published by raubn111

More info:

Published by: raubn111 on May 23, 2012
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less






  • Acknowledgments
  • List of abbreviations
  • 1.Introduction
  • 1.1.The current status of Kusunda
  • 1.2.Later research on Kusunda
  • 1.3.Recent contact with Kusunda speakers
  • 1.4.Context leading to the current study
  • 2.A socio-linguistic and anthropological profile
  • 2.1.Vocabulary
  • 2.2.Tibeto-Burman and other borrowings
  • 2.3 Marriage and kinship vocabulary
  • 2.4.Origin myths
  • 2.5.Religion
  • 2.5.1.Birth and death
  • 2.5.2.Illness
  • 3.Linguistic type
  • 4.Phonology
  • 4.2.Contraction
  • 4.3.1.Vowels and uvular consonants
  • 4.3.2.Pharyngealization of vowels and lowering of pitch
  • 4.3.3.Vocalic sequences
  • 4.3.4.Nasalization
  • 4.4.Consonants
  • 4.4.1.The labial series
  • 4.4.2.The apical series
  • 4.4.3.The laminal series
  • 4.4.4.The velar and uvular series
  • 5.Word classes
  • 5.1.Pronouns
  • 5.1.1.Possessive pronouns and the genitive
  • 5.1.2.Oblique forms and the accusative
  • 5.1.3.Interrogative pronouns
  • 5.1.4.Demonstrative pronouns
  • 5.2.Nouns and noun morphology
  • 5.2.1.Number marking
  • 5.2.2.Case marking
  • 5.3.Locative postpositions
  • 5.4.Adjectives
  • 5.5.Verbs and verb morphology
  • 5.5.1.Person marking in verbs
  • 5.5.2.Number marking
  • 5.5.3.Tense-aspect-modality (TAM)
  • 5.5.4.Negation
  • 5.5.6.Prohibitives
  • 5.5.7.Hortatives and optatives
  • 5.6.1.Adverbs of manner
  • 5.6.2.Adverbs of time
  • 5.6.3.Adverbs of location
  • 5.6.4.Adverbs of quantity
  • 6.Basic clause types
  • 6.1.Copular constructions
  • 6.1.1.Equative clauses
  • 6.1.2Existential copula
  • 6.1.3Locational copula
  • 6.1.4The inchoative and the causative
  • 6.2 Intransitive–transitive clauses
  • 6.2.1Intransitive suffixing pattern 1
  • 6.2.2Intransitive suffixing pattern 2
  • 6.2.3Suffixing pattern 3
  • 6.2.4Bitransitive verbs
  • 6.3 Clausal syntax
  • 7.Transitivity alternations – causatives, anti- causatives, and reflexives
  • 7.1 Causative
  • 7.1.1Morphological causative -a
  • 7.1.2Morphological causative -da
  • 7.2Anti-causative
  • 7.2.1Anti-causative -q versus lexical -q
  • 7.2.2Anti-causative -t
  • 7.2.3Passive
  • 7.3 Reflexive
  • 8.Morphology of subordinate structures
  • 8.1 Neutral forms
  • 8.1.1In Class II intransitive verbs
  • 8.1.2In Class II transitive verbs
  • 8.1.3In Class I verbs
  • 8.1.4Bare roots (concatenative structures)
  • 8.2 Embedded structures marked by -da
  • 8.2.1Non-finite forms marked by -da
  • 8.2.2Non-finite Class II verbs marked by -da
  • 8.2.3Inflected forms marked by -da
  • 8.3 Finite embedded structures
  • 8.4 Embedded structures marked by mutation
  • 9.Syntax of subordinate strucutures
  • 9.1 Adnominals and relative clauses
  • 9.1.1Finite adnominals formed from verbs
  • 9.1.2Non-finite adnominals formed from verbs
  • 9.1.3Case recoverablility strategies for relative clauses
  • 9.1.4Adjectivals
  • 9.2 Periphrastic causatives
  • 9.2.1Implicative causative – “make”
  • 9.2.2Causative “send”
  • 9.3 Complements
  • 9.3.1Purposive
  • 9.3.2Desiderative
  • 9.3.3Abilitive
  • 9.3.4Permissive
  • 9.3.5Preventative
  • 9.3.6Completive
  • 9.3.7See and hear
  • 9.3.9Know how
  • 9.3.10Believe
  • 9.3.11Show how, teach
  • 10.Clause chaining
  • 10.1Sequential events
  • 10.1.1Negation in sequential chains
  • 10.1.2A second negative
  • 10.2Overlapping events
  • 10.2.1Cause and effect
  • 10.2.2Temporal overlap
  • 10.2.3The benefactive construction
  • 10.2.4The “teach” converbal
  • 10.2.5Causative converbal – ‘send’
  • 11.Combinations of finite clauses
  • 11.1Coordination
  • 11.1.1Discontinuity of person reference
  • 11.1.2Discontinuity of time
  • 11.1.3Discontinuity of TAM marking
  • 11.2Conditionals and concessives
  • 11.2.1Continuity of person in irrealis mode
  • 11.2.2Discontinuity of person in irrealis mode
  • 11.2.3A concessive
  • 11.3Bi-clausal residue
  • References

Notes on Kusunda Grammar

A Language Isolate of Nepal

David E. Watters
Tribhuvan University, SIL International

With the participation of:

Yogendra P. Yadava, Madhav P. Pokharel, Balaram Prasain

Himalayan Linguistics
Archive 3

* First edition published 2005 by the National Foundation for the Development of Indigenous Nationalities, Kathmandu, Nepal; ISBN 99946-35-35-2 paperback. (For limited distribution within Nepal only) Himalayan Linguistics Archive 3. (2006) 1-182. © 2006 All rights reserved.

Table Of Contents
Acknowledgments ...........................................................................................................5 List of abbreviations.......................................................................................................7 1. Introduction.........................................................................................................9 1.1. The current status of Kusunda.........................................................................9 1.2. Later research on Kusunda ............................................................................10 1.3. Recent contact with Kusunda speakers ..........................................................11 1.4. Context leading to the current study...............................................................12 2. A 2.1. 2.2. 2.3 2.4. 2.5. 2.5.1. 2.5.2. 3. socio-linguistic and anthropological profile.............................................13 Vocabulary.....................................................................................................13 Tibeto-Burman and other borrowings............................................................15 Marriage and kinship vocabulary...................................................................16 Origin myths..................................................................................................17 Religion .........................................................................................................18 Birth and death.........................................................................................18 Illness ......................................................................................................19

Linguistic type ..................................................................................................20

4. Phonology ..........................................................................................................22 4.1. Harmony........................................................................................................22 4.2. Contraction ....................................................................................................23 4.3. Vowels...........................................................................................................24 4.3.1. Vowels and uvular consonants.................................................................25 4.3.2. Pharyngealization of vowels and lowering of pitch ..................................26 4.3.3. Vocalic sequences....................................................................................27 4.3.4. Nasalization .............................................................................................30 4.4. Consonants....................................................................................................31 4.4.1. The labial series .......................................................................................32 4.4.2. The apical series.......................................................................................34 4.4.3. The laminal series ....................................................................................37 4.4.4. The velar and uvular series.......................................................................37 4.4.5. Liquids, flaps, glides, etc. .........................................................................43 5. Word classes......................................................................................................44 5.1. Pronouns .......................................................................................................44 5.1.1. Possessive pronouns and the genitive ......................................................45 5.1.2. Oblique forms and the accusative.............................................................47 5.1.3. Interrogative pronouns.............................................................................47 5.1.4. Demonstrative pronouns..........................................................................49 5.2. Nouns and noun morphology........................................................................49 5.2.1. Number marking......................................................................................49 5.2.2. Case marking...........................................................................................50 5.3. Locative postpositions....................................................................................56 5.4. Adjectives.......................................................................................................58 5.5. Verbs and verb morphology...........................................................................59 2

Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 5.5.1. 5.5.2. 5.5.3. 5.5.4. 5.5.5. 5.5.6. 5.5.7. 5.6. 5.6.1. 5.6.2. 5.6.3. 5.6.4. Person marking in verbs ..........................................................................59 Number marking......................................................................................63 Tense-aspect-modality (TAM).................................................................66 Negation ..................................................................................................71 Imperatives...............................................................................................75 Prohibitives..............................................................................................78 Hortatives and optatives ...........................................................................79 Adverbs..........................................................................................................82 Adverbs of manner...................................................................................83 Adverbs of time........................................................................................83 Adverbs of location..................................................................................84 Adverbs of quantity..................................................................................84

6. Basic clause types..............................................................................................85 6.1. Copular constructions....................................................................................85 6.1.1. Equative clauses.......................................................................................85 6.1.2 Existential copula.....................................................................................87 6.1.3 Locational copula.....................................................................................89 6.1.4 The inchoative and the causative...............................................................90 6.2 Intransitive–transitive clauses.........................................................................93 6.2.1 Intransitive suffixing pattern 1 .................................................................94 6.2.2 Intransitive suffixing pattern 2 .................................................................94 6.2.3 Suffixing pattern 3...................................................................................95 6.2.4 Bitransitive verbs......................................................................................95 6.3 Clausal syntax................................................................................................95 7. Transitivity alternations – causatives, anti-causatives, and reflexives......97 7.1 Causative........................................................................................................97 7.1.1 Morphological causative -a......................................................................97 7.1.2 Morphological causative -da....................................................................98 7.2 Anti-causative.................................................................................................99 7.2.1 Anti-causative -q versus lexical -q..........................................................100 7.2.2 Anti-causative -t .....................................................................................101 7.2.3 Passive...................................................................................................103 7.3 Reflexive......................................................................................................104 8. Morphology of subordinate structures.........................................................106 8.1 Neutral forms...............................................................................................106 8.1.1 In Class II intransitive verbs...................................................................106 8.1.2 In Class II transitive verbs......................................................................107 8.1.3 In Class I verbs......................................................................................107 8.1.4 Bare roots (concatenative structures)......................................................107 8.2 Embedded structures marked by -da............................................................108 8.2.1 Non-finite forms marked by -da............................................................108 8.2.2 Non-finite Class II verbs marked by -da................................................109 8.2.3 Inflected forms marked by -da...............................................................109 8.3 Finite embedded structures ..........................................................................111 8.4 Embedded structures marked by mutation ...................................................112 3

........................1 Negation in sequential chains.1.....................................................121 9......2 Desiderative ...........................2 Discontinuity of time ..................................137 11..............1....5 Causative converbal – ‘send’............2 Non-finite adnominals formed from verbs..................................125 10...........9 Know how .......................1............132 11...........2 Periphrastic causatives ..180 4 ........................1........................................................................10 Believe .......................................2................................................................................................................................................................................................2.................................. Syntax of subordinate strucutures....................................3 Abilitive......8 Think ............................2 Causative “send”.........................................................................................122 9...........................................................................................122 9.......4 Adjectivals...........135 11..........4 The “teach” converbal .......................1 Discontinuity of person reference..... teach ..........................................................................................118 9.....136 11..........................122 9....126 10.........3 Case recoverablility strategies for relative clauses ...........113 9.......4 Permissive................................................2 Temporal overlap ..125 10.128 10...................................120 9..2.........116 9...........................................................................118 9................153 References ....113 9................132 10..................128 10..124 10.................................1 Coordination.............................................3..........3.....3................................................3 Bi-clausal residue.............120 9............................................................................................................................................................2 Overlapping events.7 See and hear...............................................................................................................139 Appendix B — Verb Paradigms ................135 11...........2.........................................................................134 11.....................3.......123 9..............................................................................................................................................113 9..............................................................1 Continuity of person in irrealis mode...............................................2...121 9................129 10...1.....................3..................................................................1.....................2 Conditionals and concessives..................2.........................................................3............6 Completive...........................136 11........................................2 Discontinuity of person in irrealis mode...................3 Complements......2........................ Combinations of finite clauses............120 9..... Clause chaining .............................................................1.....................2........3 The benefactive construction..........................Watters: Kusunda Grammar...1..............................................................................................3.......................3..............................134 11................................3 Discontinuity of TAM marking ..........................................1 Purposive..............................................................................................................1 Adnominals and relative clauses.....................................3.............3................................136 11.123 9..... Contents ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 9...................................11 Show how.................1 Cause and effect......................5 Preventative.............................1 Sequential events...........1.................................1 Finite adnominals formed from verbs ............................................2..................138 Appendix A — Kusunda Vocabulary...............................2..135 11....................................1 Implicative causative – “make”..........................117 9..124 9........................................114 9...............119 9..3 A concessive ...............3...................127 10................................................................2 A second negative ......................................129 10...................................

her uncle. Jagat Gurung. Their response was magnanimous. They were at the mercy of strangers. our greatest debt of gratitude goes to the Kusunda speakers who accepted the invitation to come to Kathmandu – Ms. our friends at the National Foundation for the Development of Indigenous Nationalities (NFDIN). Mr. It is impossible to overemphasize our indebtedness to them. Ukyab to help us in bringing Gyani Maiya and possibly other Kusunda speakers into Kathmandu for a three month period of intensive linguistic research. Once a week. Mr. All electronic media are now housed at the CDL library. anthropologists. and linguists.Acknowledgments No one could have guessed. early in 2004. We are grateful for his willing contribution. Tamla Ukyab. We knew almost nothing about the grammar – doubly lamentable in light of the fact that Kusunda was recognized as a language isolate. plus a short text that was largely uninterpretable. and Mr. first step in helping them and making a simple thing like land ownership a possibility for them. Linguistics Chair Prof. They also arranged a press conference in which they introduced the Kusundas to the public and to the academic community – sociologists. Only Gyani Maiya had been to Kathmandu before. Ms. Tribhuvan University. cheerfully sharing their language and their lives. Prem Bahadur Shahi Thakuri. keeping them busy the whole day long. Back in Dang. When they arrived. all we had were a couple of word lists. at the Central Department of Linguistics (CDL). never gave up. Without question. his niece. Gurung and Mr. for making digital recordings of many Kusunda sounds. Gyani Maiya Sen. and most of us had resigned ourselves to the notion that we would never know more about Kusunda than what we already knew. Ms. Four of us at CDL. We are also grateful to Mr. and myself. and to Dinesh Khadga of Kantipur Photo Studio for making video recordings of several elicitation sessions. that before the year was out we would get lucky with Kusunda. Gyani Maiya Sen. they quickly settled into their new situation. she could actually speak it! We. But for all of them. not knowing fully what to expect. Yogendra Prasad Yadava. with help from the National Indigenous Nationalities Women’s Forum. In some quarters. and make a valuable addition to their language archives. Sant Bahadur Gurung and Member Secretary Mr. It was there that we discovered that one of them. At the time. immediately made a proposal to Prof. we debriefed 5 . of SIL and CDL. also of NFDIN. to Dang to bring the speakers to Kathmandu. especially Vice-Chairman Prof. Mr. Prof. who lived in Rolpa. had more than a passive knowledge of Kusunda. daily sessions with our new-found friends. Dinesh Gurung. and Prem Bahadur contacted Kamala. one from the 1850s and another from the 1970s. In April 2004. Gyani Maiya contacted Prem Bahadur. Not only did NFDIN agree to fund the research and documentation project. dealt with many practical arrangements. Madhav Prasad Pokharel. In the midst of this unpromising circumstance. a three month stay in Kathmandu was a leap of faith. Kamala Singh Khatri. Balaram Prasain. the last speaker was reported to have died in 1985. Once in Kathmandu. Stephen Watters. related to no other language on earth. like finding lodging for the Kusundas and taking them on weekend tours to the zoo and to numerous religious sites. they brought three Kusundas to Kathmandu from Dang Valley to help them with citizenship papers – a practical. at the recent invitation of NFDIN. the situation was not hopeful. had separate. they also sent one of their own people. and though some Nepalese linguists and anthropologists were aware of one or two Kusundas with a vague memory of their ancestral tongue.

Watters: Kusunda Grammar; Acknowledgments ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– one another in public, “chalk-board discussions,” well attended by faculty and students of the linguistics department. I should make clear that the three participants do not necessarily agree with all my interpretations presented in this grammar. Any and all errors are my own. At the end of the research period, NFDIN hosted another press conference, at which the Kusundas were fêted, and the principal investigators were given opportunity to present their findings. An invited speaker, State Minister for Local Development Mr. Krishna Gopal Shrestha, promised his support for projects aimed at uplifting indigenous nationalities. Several other dignitaries, including Prof. Sant Bahadur Gurung, Prof. Yogendra Prasad Yadava, Prof. Madhav Prasad Pokharel, Prof. Novel Kishore Rai, and visiting Prof. Sueyoshi Toba, gave unanimous voice to the overriding value of language documentation and preservation efforts. The conference was reported in all the major Nepalese dailies for Asoj 24, 2061, BS (October 10, 2004). A special note of thanks goes to Professors R. M. W. Dixon and Alexandra Aikhenvald for inviting me as an Honorary Visiting Fellow to their Research Centre for Linguistic Typology (RCLT) at La Trobe University, Melbourne, Australia. I was able to spend three months, uninterrupted, analyzing my notes and writing the present grammar. I was also given opportunity to present my findings and analysis at an RCLT seminar, receiving valuable suggestions and comments from Robert Dixon, Alexandra Aikhenvald, Randy LaPolla, John Saeed, Gerrold Sadock, Ghil‘ad Zuckermann, and John Hajek. I would also like to thank Alexandra Aikhenvald, Scott DeLancey, David Bradley, Sueyoshi Toba, and Stephen Watters for reading an earlier draft of this grammar and making valuable comments. I have not incorporated all of them, and any shortcomings are my own. Last, but not least, I wish to thank the Himalayan Linguistics anonymous reviewer who took a close, critical view of the first edition of this grammar (the NFDIN version) and made a number of very helpful suggestions. Many of those recommendations I have followed and incorporated into this current edition. David E. Watters Kathmandu, Nepal 19 April 2006


List of abbreviations
1 1P 1S 2 2P 2S 3 3P 3S A ABL AC ACC ADJ adj. adv. aff. ALLT CAUS CDL CNV DAT DETRANS EMP ERG FOR GEN HORT IA IMM IMP IN INCHO INCOMP IND IRR LOC (N) n. NEG First person First person plural First person singular Second person Second person plural Second person singular Third person Third person plural Third person singular Subject of transitive Ablative Anti-causative Accusative Adjectivizer Adjective Adverb Affix Allative Causative Central Department of Linguistics Converbal Dative Detransitive Emphatic Ergative Benefactive Genitive Hortative Indo-Aryan Imminent Imperative In, Specific location Inchoative Incompletive Indicative Irrealis Locative Nepali Noun Negative 7

Watters: Kusunda Grammar; Abbreviations ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Nep. NEUT NF NFDIN NML NOM NP O OPT PAST PFV PL POSS pp. PRES PROH PTB PURP REAL S SG SUB SUBORD TAM TB TR V VDC vi. vt. WITH Nepali Neutral inflection Non-final verb National Foundation for the Developement of Indigenous Nationalities Nominalizer Nominative Noun phrase Object of transitive Optative Past Perfective Plural Possessive Postposition Present Prohibitive Proto Tibeto-Burman Purposive Realis Subject of intransitive Singular Subjunctive Subordinate Tense–aspect–modality Tibeto-Burman Transitive Verb Village Development Committee Intransitive verb Transitive verb Comitative


The current status of Kusunda Kusunda survives today. 46). in his Comparative Vocabulary of the Languages of the Broken Tribes of Nepal (1857).1. On the Chépáng and Kúsúnda tribes of Nepál. and were quite happy to refer to themselves as Kusundas. Kamala Khatri. but almost no other implement of civilization.1. and it is in the very skilful snaring of the beasts of the field and the fowls of the air that all their little intelligence is manifested. especially in western Nepal. to any other language or language family of South Asia.’ (See §2.” and may still be used in the sense of “savage. Hodgson. Nowadays. 5. Apparently the data was collected by linguistic assistants. introduced them. (See Appendix A for a full.” Van Driem (2001) reports that according to Ralph Turner (1931). Kusundas first came to the attention of the Western world in 1848 when Brian Hodgson.” are an ethnic group of Nepal who. 1. with the following words: Amid the dense forests of the central region of Népál. the term still has some currency. paragraph 2 and fn. due to the loss of vast tracts of forest lands their hunting bands have splintered and they have been compelled. some 223 in all. the British Resident to the Court of Nepal. I never could gain the least access to the Kusúndas.) It seems that Robert Shafer (1953) was the first to notice its unique status.” In our own experience. Kusunda was formerly used as “a term of abuse for the so-called Rajputs of Nepal. to the westward of the great valley. dwell. “During a long residence in Népál.) Nine years after his first mention of Kusunda.” Reinhard. of which the iron arrow-heads are procured from their neighbors. mihaq ‘the people. The most important consequence of Hodgson’s list was that it (should have) demonstrated unequivocally that Kusunda was unrelated to Chepang. their numbers have dwindled drastically and their language has all but ceased to exist. modern vocabulary of about 850 words. until recent historical times. in an article in the Journal of the Asiatic Society of Bengal. seemed either unaware or uncaring of its pejorative overtones. “They have bows and arrows. p. with one notable exception. however. too (1976:2). in varying degrees of fluency. because of a lack of marriageable Kusunda partners. and seeming like fragments of an earlier population. notes that in some areas of Nepal the word Kusunda makes reference to those who “do not listen to advice and behave rudely. for in his own words he confessed that. lived as semi-nomadic hunter-gatherers in central and midwestern Nepal. in scanty numbers and nearly in a state of nature. 1874:45) His description is even less flattering a few sentences later when he goes on to say that. together with the Chepangs. though aided by all the authority of the Durbar” (part II. As a result. although they continue to refer to themselves by their autonym. Our informants. a thirty-two year old 9 . (1848:650. Introduction The Kusundas. in only a handful of speakers — no more than seven or eight in toto. or for that matter. also known as Ban Rajas “Kings of the Forest. almost one hundred years later. to intermarry with other ethnic groups.1. published a list of Kusunda words. None of the descendants of the current speakers use the language. two broken tribes having no apparent affinity with the civilized races of that country.

and the Hill tribes of Arracan. K. a language isolate of west-central India (Fleming 1996. Kusunda. It is probable that other aboriginal languages existed alongside Kusunda in that prehistoric period. J. inscribed on the walls of caves and rock overhangs. published in 1909. Whitehouse 1997).” The time depth is enourmous and seems hardly plausible.” Surprisingly. who is credited with compiling Part I of Volume III. the very title of which implies a connection between Chepang. The latest proposal (Whitehouse et al. Kusunda is excluded entirely. 2 1 10 . attesting to the presence of possible multiple aboriginal populations. in which the only mention of Kusunda is an attempt to relate Kusunda ta≥ ‘water’ to Mikir la≥ ‘water’ and Newar la ‘water. continues to insist that Kusunda is Tibeto-Burman on the basis of a few putative cognates and some very unconvincing typological correspondences (2002). Introduction ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Kusunda woman of Sakhi VDC in southern Rolpa.” None of his actual comparisons. too. More than 150 years ago Brian Hodgson had already predicted that “The lapse of a few generations will probably see the total extinction of the Chépángs and Kusúndas. Vayus. Petroglyphs. one-hundred and twenty years after Hodgson’s original report. like Nihali. In the article itself. apparently on the basis of its implied association with Chepang in Hodgson’s first report.1 Since 1909. Many linguists agree that Kusunda is very likely the sole survivor of an ancient aboriginal population once inhabiting the sub-Himalayan regions before the arrival of Tibeto-Burman and Indo-Aryan speaking peoples.2 In 1968.’ The Ethnologue. and other TB languages. Moskva: Izdatel’stvo ‘Nauka’). and Bhramu. no further linguistic samples were made available on Kusunda until Grierson’s Survey of India. A few speculative proposals continue to make the rounds on the possible relationship of Kusunda to Munda or even to languages further afield. Vayu. That Kusunda has persisted to the present day is something of a linguistic miracle. the Chepangs. however. B. published a small sample of Kusunda including a few conjugated verbs. Rana. “possibly” Burushaski and languages of the Caucasus (Reinhard and Toba 1970). cited in van Driem 2001 as: “Kusunda – sinokavkazkie leksiceskie paralleli”. Konow also cites a full book by Forbes (1881). C. but most have been unimaginative rehashings of earlier notions about its relationship to Chepang.Watters: Kusunda Grammar. but only with Chepang. Kusunda was still classified as Tibeto-Burman. Forbes further states that “… an even closer connexion appears to exist between these tribes. continues to classify Kusunda as Tibeto-Burman. with “the possibility that Kusunda is a remnant of the migration that led to the initial peopling of New Guinea and Australia. can still be found in many parts of Nepal. 2003) advances the premise that Kusunda belongs to ‘Indo-Pacific’ (a highly speculative mega-family). Kamala Khatri is fully fluent and served as our primary language informant. other occasional articles and notices have appeared on Kusunda. Later research on Kusunda After Hodgson. are with Kusunda.2. Affinities of the Dialects of the Chepang and Kusundah Tribes of Nipál with those of the Hill Tribes of Arracan. or the Yenisseian languages of Siberia (Gurov 1989. Lingvisticeskaja rekonstrukcija i drevnejsaja istorija vostoka. and Kusunda. all of which he credits to “materials published by Hodgson. Languages of Further India. has used Kusunda on a daily basis since childhood and continues to use it with her aging mother. for example. Konow cites an article. Sten Konow. and therefore I apprehend that the traces now saved from oblivion of these singularly circumstanced and characterized tribes… will be deemed very precious by all real students of ethnology. F. Johan In his Kusunda section.” (1874:48) 1. Puni Thakuri. but they have long ceased to exist. by Capt. 1877. Forbes.

word began to spread that the few remaining speakers known to linguists and anthropologists were all deceased. an ethno-botanist living near Swayambhu in Kathmandu. saying that “three speakers were reported in 1970” and that “the last speaker was reported to have died in 1985. made contact with a lone speaker.3. In the text. found members of the Kusunda tribe in Gorkha district. Unfortunately. Caughley managed to elicit a few grammatical constructions and some additional words. Damauli (Tanahu). In the 1980s.). but neither were able to speak Kusunda. The Ethnologue began to report in the 1990s that Kusunda had gone extinct.” The claim is probably related to the death of Chudamani Ban Raja. and then in Surkhet district a year later. in Rampur village of Palpa district in January 1980 (p. Reinhard published some additional Kusunda data in 1976. In 1987. and his transcriptions are based on what he could hear from Reinhard’s tape recordings. but learned a couple of years later that she was no longer alive. Prof. a Japanese linguist studying in Nepal. a handful of Nepali scholars have been aware that one or two isolated speakers of Kusunda could still be found. the two co-authored an article – A Preliminary Linguistic Analysis and Vocabulary of the Kusunda Language – published by the Summer Institute of Linguistics and Tribhuvan University.” he recalls. Though the word list differed in some respects from Hodgson’s list. this time in 11 . Churamani Bandhu also traveled to Sahare village in Surkhet. Toba had to work without the benefit of an informant. 1. “a fairly fluent speaker. Ross Caughley. and that their language had died with them. but in Damauli he was able to meet two speakers – a man named Raja Mama and Raja Mama’s mother. Recent contact with Kusunda speakers Since the mid-1980s.Himalayan Linguistics: Archive No. the language was essentially the same. In 1989. Caughley traveled to some of the early Reinhard sites in Gorkha and met two Kusundas in and around Cheptar Bazaar. many morphemes and their glosses are necessarily speculative with very little to go on. with the help of Sueyoshi Toba.c. He was disappointed in Gorkha. in February 1983 and January 1984. but apparently his findings were not substantive enough to warrant publication (although he did include a single page of materials in his 1982 study of the Chepang verb). of the Summer Institute of Linguistics. Five years later. Perhaps a few things of significance could still be gleaned. an American anthropologist studying at the University of Vienna. albeit speakers with imperfect retention of their language – some with only vague memories of words used by their elders. Also. He collected a few words from the mother. and in 1970. The speaker was unaware of other Kusundas and reckoned that he was the last surviving speaker. but learned that he had already died. in 1985. it was better than anything we had to date. Still. wrote an article in the Nepali daily Gorkhapatra reporting that in his fieldwork he had encountered several Kusunda families living in Gorkha. but for the most part. and Dang. a linguist at Tribhuvan University and the Royal Nepal Academy. Both were related to Reinhard’s informant. Hemanta Raj Bhandari. Churamani Bandhu. followed up on Bhandari’s leads and traveled to Gorkha and Damauli the same year. but the prospects for fruitful research did not seem good. too. He collected a number of Kusunda words together with a short text narrated by Tek Bahadur. They belonged to a younger generation. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Reinhard. Chudamani Ban Raja. our knowledge of the Kusundas and their language remained almost static from 1970 to the present day. Caughley attempted to contact Chudamani a second time.

together with the author. We are exceedingly grateful for their patient help. A year later. in 1991. This time they made contact with Prem Bahadur and managed to collect a few new items of language. Madhav Pokharel. to Dang to fetch the speakers. Their response was magnanimous. and helped us extend our knowledge of Kusunda grammar well beyond its earlier bounds. Though not entirely extinct.” Rana also mentions Puni Thakuri. they also sent one of their own people. Sangini Rana Magar. then away from his village on seasonal work. Context leading to the current study In recent years. 1. Ballabha Mani Dahal. and it was on this occasion that the two met for the first time. and B. agreed to come to Kathmandu for a three month period. Yogendra Yadava. including Tamla Ukyab. Kamala Khatri. also known as Janajati Pratisthan. The results were beyond our expectations. Whole systems were moribund. Prem Bahadur further agreed to make the journey to Tunibot in Rolpa to fetch another bhanji ‘niece. K. the mother of one of our informants. even he was able to recall only 67 items on the Swadesh 100-word list. The first occurred in May of 2000. K. along with two younger relatives who did not speak the language. and Prof. in Rolpa district. They turned out to be considerably more fluent than their maternal uncle Prem Bahadur. and another in April. Gyani Maiya and Kamala. and was unable to form sentences of any complexity.Watters: Kusunda Grammar. along with its earlier incarnations. The next year in Dang valley.3 In May 2004. well beyond the reach of the memories of its last faltering speakers.’ daughters of two brothers. reports that though Prem Bahadur was the more proficient of the two speakers. Prof. Bandhu returned to Dang with Prof. Puni Thakuri and her daughter come from Tunibot village. of the Central Department of Linguistics at Tribhuvan University. not a speaker of Kusunda. are ‘cousin-sisters. Kusunda was lacking its former vitality. 2004 introducing 67 year old Gyani Maiya Sen of Dang. Not only did they fund the research and documentation project. He met the informant’s wife. Rana. Furthermore. Bandhu learned of another speaker by the name of Prem Bahadur Shahi Thakuri.’ Kamala Khatri (already mentioned as the daughter of Puni Thakuri). Dinesh Gurung.4.’ Prem Bahadur Shahi Thakuri (Churamani Bandhu’s 1990 contact). Rana. 3 12 . advanced a proposal to NFDIN for a cooperative effort in bringing Gyani Maiya and possibly other Kusunda speakers into Kathmandu for a three month period of intensive linguistic research. introducing Raja Mama of Tanahu (Churamani Bandhu’s early contact). a few weeks after the April 2004 symposium. and learned that the man he sought was no longer alive. Gyani Maiya and her mama ‘maternal uncle. Prof. or conducted basic-level research on Kusundas in the field. Several of its members. Introduction ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– search of one of Reinhard’s informants of twenty years before. have written about. Madhav Pokharel. In a footnote to his 2002 paper “New Materials on the Kusunda Language. he could only partially recall the most rudimentary verbal paradigms. Gyani Maiya was apparently an earlier contact of B. formed by an act of Parliament) has taken an active interest in endangered groups like the Kusundas. The National Foundation for the Development of Indigenous Nationalities (NFDIN. with at least two recent symposiums organized in Kathmandu in which Kusundas were introduced to the press and to the academic world. The two women. Sakhi VDC. who had also met Raja Mama of Tanahu. one of the contributors to this grammar.

” capable of generating no more than the most rudimentary sentences. our most fluent informant. Prem Bahadur was born in the jungle near Hapur. They are currently settled in villages where the only means of communication is through Nepali.4 Gyani Maiya’s sister. and Raja Mama is likely the last surviving semi-speaker in the “Gorkha–Tanahu group. Sant Bahadur Gurung. Communication between spouses must be conducted in a common language.” Reinhard’s informant being the last. reports that her mother is fully fluent.” All seven speakers in the Dang–Rolpa group are only recently removed from a hunter-gatherer existence in the jungles. reports that although they have conducted extensive searches for Kusundas in the past few years.’ and ‘chicken’. they have managed to locate no more than twenty-eight. there were no eligible Kusundas around. Chhetri. all indications by the speakers themselves would put the estimate much lower – perhaps as few as seven or eight speakers. Vocabulary There are several aspects of the Kusunda language that seem unusual for huntergatherer people. the destruction of vast tracts of forest land. 87 of whom are reported to speak the Kusunda language. I personally visited Bhalkot on at least two occasions while studying Kham. 4 13 .’ and ‘gold. usually Nepali. Kamala’s cousin-sister.2. and Gyani Maiya’s three younger brothers married Kham-speaking Magar women from the village of Bhalkot. and their mama. Kamala Khatri.” It is probable that there are no surviving speakers in the “Surkhet group. self-propagating hunting bands being some of the major ones. Cross-tribal marriage is one of the major contributors to the death of the Kusunda language. Deeper causes.1. and Prem Bahadur know. too. reportedly also a Kusunda speaker. but she has had no contact with any of them for more than thirty years.’ They also have native words for fifteen castes (3) – Brahmin. They still recall subsisting in the jungle on wild roots and hunted animals. contribute to the necessity of inter-tribal marriage – overpopulation among the general populace.’ ‘goat.’ ‘money.’ ‘sheep. Prem Bahadur and Gyani Maiya married Nepali-speaking Magars. Though we have no independent figures on which to dispute this number. Prem Bahadur. and children grow up (at best) with only a passive understanding of a few words in Kusunda. is also a speaker. as was Gyani Maiya and her three brothers. Vice-Chairman of NFDIN. barely speaking Nepali. forcing them to marry outside their caste. but was unaware of any Kusunda speakers living there. She is now too old to travel. A socio-linguistic and anthropological profile According to the 2001 Census of Nepal. can be rated only as a “semi-speaker. Pyuthan. When they reached marriageable age. and the resultant splintering of earlier self-sufficient.’ ‘police. of course. Gyani Maiya reports that three of her younger brothers speak fluent Kusunda. Prof. Gyani Maiya. 164 people in Nepal call themselves Kusunda. lives in the same village. but even during their childhood their band was small and splintered. 2. in the western extreme of Baglung district. these seven kins-people are the only speakers in the “Dang–Rolpa group.’ ‘cow. only four or five of whom are speakers (person communication). but speaking only Nepali or Kham. Kami. They have native words for several domestic animals (1): ‘horse. As far as Kamala. as well as for (2): ‘king. Gyani Maiya.

appears to contain the TB root tup ‘beat. too. l. Jugi. Gaini. Magar. b. like the Chepang. and it is probably this term that is the source for the word beg˙i. a. in the Kham dialects of the region. b. A socio-linguistic and anthropological profile ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Damai. c. h. j.’ which is common. being used as an even stronger pejorative term than Kusunda. are renowned for making minute anatomical distinctions. e. Kumal. agricultural life. b. the former word. horse cow sheep goat chicken king police money gold greetings Kusunda Ban Raja Brahmin Chhetri Kami Damai Sarki Gaini Badi Jugi Kumal Tharu Newar Magar Sesi (Kham-Magar) pæyaks˙m numba goloq aidzi tap mo≥ gimtsamba gimi k˙pdza≥ sodzaq beg˙i < ? Nep. d. It is possible that many such distinctions have been forgotten by the Kusundas since settling into sedentary. A significant distinction is made between beg˙i ‘Kusunda’ (their gloss) and mihaq ‘Ban Raja’ (their gloss). Sarki. a musician and tailor caste. leather. Badi. f. e. and Sesi Kham. a. k. Equally unusual for a hunter-gatherer people is the lack of distinction between ‘heart’ and ‘lung’ – both gobloq – or between ‘snake’ and ‘bug’ – both tu.) 5 14 . not beg˙i Kusunda. begari (pejorative) mihaq ~ mehaq kæ˙rk˙la r˙ktsa koˆodi p˙itoba i≥gid˙t dz˙≥t˙l d~odzil˙q k˙l˙≥gu ya≥b˙ru înor˙t ~ ih˙≥ yeg˙mbu kiptsak (2) (3) The word for Damai (3f). apparently.Watters: Kusunda Grammar. m. g. Tharu. Newar. The words mentioned above are given in (1–3): (1) a. they were careful to make clear that they were mihaq Kusunda.’ The word for ‘police’ gimtsamba (2b) contains three syllables and may be a neologism. d. d. c. c. n. e.5 Though our informants seemed quite happy to accept the word “Kusunda” as their Nepali exonymic caste name (even referring to themselves as such). Surrounding tribes with hunting-gathering histories. o. The word for Sarki (3g) contains the Kusunda root gid˙t ‘skin. (Intervocalic ‘r’ is frequently dropped from the middle of Kusunda words borrowed from Nepali. Though Reinhard The Nepali term begari (‘forced laborer’) is sometimes added to Kusunda (in Kusunda-begari) as an abusive insult for a worthless person. though we are unaware of the individual morphemes. i.

that the vast majority of lexical items in Kusunda show no resemblance whatever to other languages or language families of the region. however.’ as in ‘horn’ ipi gidza≥. Furthermore.’ and ‘inside’ gem˙t is also the word for ‘stomach. literally ‘head body. d. and makes reference to ‘page’ in Nepali. e. b.’ It is possible that the Magar forms were borrowed from Kusunda. j. Tibeto-Burman and other borrowings We encountered several words in Kusunda that appear to have Tibeto-Burman origins. the Magar words are decidedly unusual for Tibeto-Burman. *ip is the Proto-Tibeto-Burman etymon for ‘sleep. or Chepang. Chepang and Magar use different etyma.’ The word for ‘yellow’ derives from the word for tumeric. 23). gwa ‘egg’ and tu ‘bug’ bear resemblance to Magar words with similar meanings.Himalayan Linguistics: Archive No. In Kusunda. f. Hodgson does not list heart or liver. insect’ Gamale Kham Gamale Kham Chepang Takale Kham Magar.’ and bu for ‘bug. h. More common TB forms are ba or bwa for ‘chicken. and Takale Kham has s-ep. a much more recent phenomenon than borrowing from Tibeto-Burman. the word for ‘leaf’ has been extended to also mean ‘book. Karen ‘eat’ Yamphu Though the first two words. Given. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– and Toba show a distinction between ‘heart’ g˙mu and ‘liver’ kammu. Most locative expressions have developed as metaphorical extensions of body-part names. Gamale Kham has im. gwa tu h~aÚ p˙itoba p˙yet ran sat ip/im am hampe ‘egg’ ‘bug’ ‘face’ ‘Damai’ ‘leech’ ‘millet’ ‘comb’ ‘sleep’ ‘eat’ ‘where’ < < < < < < < < < < gwa du h~aÚ tæubla pyat r~aÚd˙i sat ip/im am hampe Magar ‘chicken’ Magar ‘bug. we are probably looking at borrowings from Kham. Proto Kham Proto Kham Chepang ‘rice’.’ A protrusion or body-part extension is referred to as a ‘body. are very likely different renditions of the same word. Magar.’ It seems unusual that such a word would be borrowed.’ and the word in its proto form cannot be found in the surrounding languages. g. may be purely accidental. It occurs in none of the surrounding languages. except as ‘rice’ in Chepang. Though such metaphor is common in the languages of the world. and may be the source for Kusunda ip/im.(also occurring in certain paradigmatic forms in Kusunda).’ The word for ‘leaf’ is commonly found as a classifier root for ‘book’ or ‘page’ in South and Southeast Asia. ‘Behind’ s˙mba is also the word for ‘buttocks. Perhaps more difficult to explain is (4h). common also in TB languages of the region (see fn.as a causative ‘put to sleep. these words.’ One or both forms were certainly in use in Proto Kham. the word for ‘sleep. the specific associations found in Kusunda are not common in the languages of the immediate region.2. 2. Our informants were reluctant to give us Nepali 15 . Many more words have been borrowed from Nepali. i. Following are the words: (4) a. c. too. The verb am ‘to eat’ in (4i) which corresponds to Karenic.

along with others. gei buda buja. 2. Second ascending generation father’s father mother’s father father’s mother mother’s mother 2. We did. which in some cases are different from those collected by Reinhard: 2. The groom. is to some extent reflected in kinship terminology. gi'ogi Prasain’s list nep noqæok nya nya Reinhard’s list ni'ap.3 Marriage and kinship vocabulary According to Johan Reinhard (1976:6) The Ban Raja. like several other groups in Nepal (including the Thakuri). ni'ep ni'ap.e.2. however. bro.1. We elicited no additional marriage customs from our informants. but that it is also common for a couple to elope on their own. i. where marriage to the daughter of one’s mother’s brother is preferred. however. This practice. Unfortunately. marriage with one’s father’s sister’s daughter. A socio-linguistic and anthropological profile ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– loans. Reinhard also reported that the Kusunda disclaim any tradition of patrilateral crosscousin marriage.” As a result.’s wife father’s second brother father’s third brother wife’s father wife’s mother Prasain’s list y˙i m˙i nyam nyabi --gimdæ˙r --y~atsa maqæadzæi --budun gyaudzi 16 Reinhard’s list yei mai ni'am ni'ambe numu pæusai (N) -yei mijar mai mijarni yei mijut mai mijutni imala îsala ~ bud˙n. goes to the home of the bride where he gives her gifts of clothing and jewelry. First ascending generation father mother mother’s brother mother’s brother’s wife father’s sister father’s sister’s husband father’s brother father’s older brother father’s older brother’s wife father’s younger brother father’s y. collect the following kinship terms (the work of Balaram Prasain). To the parents he gives a small gift of money.3. Child marriage is apparently unknown. saying rather that “we don’t have a word for that. this means that we have no way to gauge the level of Nepali incursion. The wedding ceremony itself is simple. along with a small group of friends and relatives.e. i. most lexical items represented by Nepali loans simply do not occur on our word list. practice matrilateral cross-cousin marriage. It was also reported by Reinhard in 1976 that marriage partners are normally arranged by the parents.3. ni'ep ni'a' ni'a' . but that it was tolerated in recent times due to a scarcity of marriageable women.Watters: Kusunda Grammar.

The Origin of the Kusunda. m˙m˙nji b˙yei' bai' bin˙i' (N) m˙do. left his inheritance to the 17 . and could produce nothing analogous on their own. Prasain’s list duktsi pinda duktsi ni≥gitse getse bai qatraq duktsi ---Reinhard’s list dukcæi. As a translation. get them to orally translate the story back into Kusunda based on Reinhard and Toba’s Nepali translation.4.3. m˙nau' j˙wa~î (N) ---jetæaju (N) dewar (N) amaju (N) n˙nd˙ (N) jetæu (N) sala (N) mijarni sali (N) Second descending generation Prasain’s list grandchild gebon Origin myths 2. it lacks spontaneity. however. the father of three sons. First descending generation son eldest son daughter son’s wife daughter’s husband brother’s son brother’s daughter sister’s son sister’s daughter 2. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 2.Himalayan Linguistics: Archive No. from a Kusunda man named Tek Bahadur in central Nepal. Our informants were unfamiliar with the story. Yogendra Yadava did. Own generation husband wife older brother younger brother older brother’s wife younger brother’s wife older sister younger sister older sister’s husband younger sister’s husband eldest sister second sister third sister husband’s older brother husband’s younger brother husband’s older sister husband’s younger sister wife’s older brother wife’s younger brother wife’s older sister wife’s younger sister 2. dugutsi -nicæe. a king.3. and Reinhard and Toba provided a transcription and translation of the text in their 1970 article. ninyitsi bai' j˙wa~î. According to the story. nicæi bæanja (N) bæ˙tiji (N) bæ˙tiji (N) bæanja' (N) bæanji (N) Reinhard’s list nati (N) Prasain’s list dui ¯~aˆdi mom bæ˙ya --bai bin˙i m˙ndu -unibai m˙lli bai sali bai -----gesala -gesali Reinhard’s list duw˙i ni'and˙i. dutci. ni≥d˙i mam bæai˙' m˙nji.5.4. Reinhard recorded a short Kusunda text.3.3. nisi.

Birth and death Reinhard (1976) reports that at childbirth both mother and baby and are required to observe an eleven day period of pollution. does not support such a claim. A socio-linguistic and anthropological profile ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– eldest son. Both myths make a deliberate connection between the Kusunda and the high caste Thakuri. Sita quietly picked up the child and left. The latter two names are associated with hunting and hunting implements. They also worship ancestor spirits.Watters: Kusunda Grammar. and when milking his buffalo. The word qaoli seems to refer to gods in general. Religion The most obvious religious scruple practiced by the Kusundas is their refusal to drink cow’s milk or use any other cow product. begging and hunting. ruled over the forests and became the father of the Kusundas. The son was unsuccessful in all that he did – his crops failed and weeds grew in their place. The linguistic evidence. Arimal. one of them involving Balmiki. and to claim a relationship with Thakuri (and hence to the king and the ruling class) is fairly common. the baby was missing. The first son ruled the cultivated lands and became the father of the Thakuri. of course. too. being made by several other aspiring tribes in Nepal as well. The Thakuris are his descendants.5. According to the story Sita asked Balmiki to watch her baby while she went to the river to wash. The Ban Raja are his descendants. One day the second son wanted to do a special sacrifice that required the blood of a pig. He agreed. The one made from kus grass. 2. not a specific god. a famous sage. our younger informant. The second son became king in his brother’s stead and was successful in all that he did. The mourning period is for thirteen days. is reported by Reinhard (1976). and the Magars (who keep pigs) are his descendants. He rubbed oil on its legs and a black man sprang out. Interestingly. the Kusundas being descended from the “unsuccessful” son and the “imitation” son. It is also reported by Reinhard that Kusundas bury their dead. 18 .1. The title “Ban Raja” concedes the same – “the king is King of the Durbar.5. When Balmiki awoke. Reinhard (1976) reports that shamans in seance call on the help of specific deities Bhuiyar. Kamala. she raised both children as her own. Ban Jhankri. we are Kings of the Forest. telling him that because of his lack of success he could no longer be king. and he fashioned an identical baby for her out of kus grass.” 2. including cow dung for cooking or plastering. followed by a purification ritual for both. both myths concede a lower status for Kusunda. So as not to disturb him. The second son confronted his older brother. There are other origin myths as well. it produced blood in place of milk. In some cases personal utensils and uncooked rice are left behind for the deceased’s afterlife. who became king in his place. The youngest son volunteered. but he could find no one to kill it. and Gwang. but when Sita returned Balmiki was deep in meditation. Little is known about the gods and deities of the Kusundas. This one. depositing the body in a newly dug grave close to a river. and the goddess Sita. was careful to observe this prohibition and refused to use milk in her tea. named Kusha. Later. and had to live in the forest.

” with magical drawings and mantras drawn on a piece of paper. sprinkled with water to purify it. or by counting and matching grains of rice in a divination ceremony. and conducts a specialized ceremony. They enter through his head. Eventually he begins to shake violently with the presence of his spirits. eating corn prematurely from the field before the firstfruits had been offered to the gods. When asked how she could be so sure. Serious illness requires that the shaman conduct a seance. usually in the night. that her malady was the result of her three year old son. For the seance. Earlier in his life. During our short time of association with Gyani Maiya and Kamala. and spread with a mat of sal leaves. Such determination is done by the shaman and based on divination – by feeling for diagnostic signs in the patient’s pulse. after first being involuntarily possessed. tightly folded and tied around the neck of the patient. back in the village. She had been convinced. Illness Ban Rajas employ shamans in curing illness. but we witnessed none of the actual transactions. He gave her some homeopathic pills. Only the shaman has the ability to be possessed by a tutelary spirit. There is no formal initiation ceremony. however. She was much improved the next day and pursued it no further. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 2.2. The ability is not inherited. she responded that it was obvious – her neck was stiff! Reinhard (1976) reports that the treatment of illness begins with a determination of the cause and category of the illness. Gyani Maiya was having some health problems and employed the services of a Tamang shaman from Kirtipur. he learns the necessary formulas and spells to become possessed at will. makes libations of liquor to the tutelary spirits. 19 . a special sitting place is prepared for the shaman. Kamala. recites charms.Himalayan Linguistics: Archive No. The shaman washes. painful neck one morning and was eager to seek the help of a shaman. too. Where none are available in their own clans they are quite willing to obtain the services of shamans from other castes. An area is cleared on the ground. Lesser illnesses require a jantra or “curing charm.5. lights incense. The cure includes the requirement of a sacrifice. suffered from a stiff.

There is considerable free variation between the two sets.3. it is the active articulator that counts as contrastive. apical. One of the most striking and unusual aspects of Kusunda grammar is the means by which “marked” structures are distinguished from “unmarked” ones. dental. the marked polarity is negative. and vowels from the upper set shift to the lower set. and uvular/pharyngeal. to no other language on earth – it is a true linguistic “isolate”. a few lexical borrowings from surrounding languages. there are sounds found throughout the region. the marked transitivity is 20 . Vowels in Kusunda occur partially in harmonic sets. and glottal. an upper set comprising three vowels and a lower set comprising three vowels. and occurs today only in a very small subset of high frequency verbs. The “mark” is not an affix. Conversely. But all such borrowings are relatively recent and have nothing to do with its genetic lineage. the point of articulation in Kusunda is more-or-less immaterial. Phonologically. o. the marked modality is irrealis (while the unmarked category is realis). pharyngeal. The uvular consonants are accompanied by the concomitant features of pharyngeal stricture and lower fundamental frequencies on preceding vowels. and only in a few words can opposing vowels in upper and lower sets be shown to be contrastive. Linguistic type Kusunda is related to no other language or language family of South Asia. having a lesser functional load than in surrounding languages. as far as we can tell. a. indeed. we can be reasonably safe in assuming that throughout most of its history Kusunda developed in isolation. Vestiges of this striking morphophonological process are still apparent in several categories and may represent the morphologization of a single process in different contexts – all ‘marked’ categories. is to describe consonant sounds in terms of their active articulator – labial. retroflex. a harmonic autosegmental process (which we will refer to as “mutation”) that spreads “retraction” across the entire word. What is more material for Kusunda. The original language provides only a substrate. to be sure. that do not occur in Kusunda. laminal. lexically. Such features. and grammatically distinct. palatal. alveolar. and only in recent times has it had contact with other linguistic types. This suggests that Kusunda may have at one time had a simple. The voicing contrast in Kusunda is neutralized in many consonants. Thus. Furthermore. The status of some linguistic isolates can be extremely difficult to determine. are non-contrastive. Kusunda possesses sounds that do not exist in other languages of the region – namely. Kusunda has not escaped at least some such influence.e. uvular. it is phonologically. velar. alveolar. are often the primary auditory cues to their presence. The process appears to be very old. or retroflex. and there can sometimes be considerable variation even in single words. but rather. but. it remains a typological isolate – i. both from Indo-Aryan and from Tibeto-Burman. Consonant sounds in Kusunda occur at many points of articulation – bilabial. Most of the points. Thus. velar consonants become uvular. by and large. however. then. There are. as typical of the region or indeed of the languages of the world. like retroflex consonants. Apical consonants become laminal. An apical consonant. velar. three vowel system – i. such languages may have been sufficiently influenced through long-term contact with surrounding languages that they begin to resemble them both grammatically and lexically. for example. can be dental. uvular and pharyngealized consonants. in fact.

and applicatives. is that such embedded structures appear not to be nominalized (although our so-called ‘neutral’ verbs are possible candidates for nominalization. verbal complements.Himalayan Linguistics: Archive No. Unusual. and inherently transitive verbs made intransitive – the latter a kind of “middle” derivation. is. in fact. There are derivational operations in Kusunda whereby inherently intransitive verbs can be made transitive.” i. the person and number of the subject actant being recoverable from obligatory person–number agreement markers in the verb. different morphemes that just happen to be homophonous must be assumed – -da is a plural number agreement in verbs. and a morphological causative.1 and §9. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– causative. is marked by the auxiliary ‘give’ in place of the auxiliary ‘make. Both operations are morphological. with alternative orders used to mark specialized pragmatic notions. to do. The basic TAM system in Kusunda makes a binary distinction between realis and irrealis. (In mutating verbs.’ In most cases. is that converbs in Kusunda do not mark sequential events. in which case the caused event is embedded to the causativizing matrix verb ‘to make. however. The majority of verbs belong to Class II. Class II verbs have also developed a past tense distinction. unusual for the typology of the region. A structure which on the surface looks deceptively like the benefactive in surrounding languages. In terms of grammatical case marking. at least for some non-inflecting (Class II) verbs. but an affix -da occurs as a distinguishing morpheme in numerous systems.2). the marker of an animate object in transitive clauses. all inflection falls on an auxiliary based on the verb ‘to make. Subject NPs are often deleted in running discourse. it requires no derivational machinery and is formed simply by the addition of a dative argument. Sequential chains are marked by a series of fully finite verbs. Periphrastic causatives are also possible. subordination is also marked by mutation. and causativization always yields a Class II verb. There are very few case marking affixes in Kusunda. A striking feature in Kusunda. This is common in the linguistic area. Such verbs can be thought of as “non-inflecting. see §8. AOV constituent order. Verbs in Kusunda divide broadly into Class I and Class II. In other cases. whereas in Class II verbs both person agreement and number is marked by suffixes. only an overlapping converbal structure. The benefactive. In some cases a single etymological source can be plausibly inferred – -da is a local case marker. Certain periphrastic causatives. with a new.) Some. to do. In Class I verbs. Syntax in Kusunda follows a basic SV. All of these systems utilize mutation as a mark (at least to a partial extent). do. Most subordinate structures are fully finite and their embedded status is signalled entirely by their syntax. Kusunda follows a nominative–accusative case marking alignment – something unusual for the languages of the region. innovative person marking distinction developing in some contexts. however. but “overlapping” ones.e. and occurs in some complement clauses. a distinction that is generally missing from Class I verbs. periphrastic causatives.’ Verb subordination (or embedding) occurs in numerous Kusunda structures – relative clauses. person agreement is marked by prefixes and number by suffixes. as well as some complements (like ‘teaching how to do something’) are marked by overlapping converbs. Class I verbs are old with a lot of irregularity in the paradigms and an abundance of suppletive forms. but not all. 21 .’ Class II intransitive verbs are basically bereft of person marking. a marker of incompletive aspect. clause chains in Kusunda are marked by converbs.1. and the marked dependency is dependent.

but it does play a contrastive role in some grammatical systems. Fairly predictably. Some of these issues will require careful spectographic analysis coupled with sound theoretical insights. Roediger 1874. it becomes apparent that the two sets are. u) and a lower set (e. [u] would occur in conjunction with [˙]. and it is particularly in these ‘back’ environments that [ßk] is hardly distinguishable from [q]. with no apparent difference in meaning.2. Our analysis shows a contrast between the cardinal values for velar and uvular consonants.) Consonants are diverse with bilabial. but because of a certain amount of variation at the extremes we often had difficulty in distinguishing certain varieties of velar consonants from certain varieties of uvular consonants. for example. contrastive and that the variation is related to an intrinsic tendency of consonant–vowel “harmony” in certain parts the language.5 Laminals and uvulars tend to co-occur in harmonic sets. occur in two harmonic sets — an upper set (i. and a post-velar [ßk] occurring before back vowels. [Ê] [c] – and at times the choice of one over the other seems unprincipled. Harmony Vowels in Kusunda.’ the lower set predominates. retroflex. The already cited contrast between [n˙g˙n] ‘You went’ and [¯a an] Alexandra Aikhenvald has pointed out to me that a distinction between the phonology of roots and the phonology of inflectional morphology can also be found in languages like Hebrew. ˙.4. in fact. pharyngeal. has numerous allophones – [@t]. o). uvular. however. for example. and a consonant system more varied (and also different) than Tibeto-Burman or Indo-Aryan. as in [n˙g˙n] ‘You went’ versus [¯a an] ‘You will go.4. 5 22 . a. appears to have two major variants – a pre-velar [k] occurring before front vowels. Our attempts at nailing down vowels were constantly frustrated.1 on Mutation.5. [t]. dental. But the fact remains that some grammatical contrasts depend upon the velar/uvular contrast. Upon closer investigation. for example. One day the form for ‘I cooked’ would be [hul˙dßn] and the next day it would be [holadßn]. Phonology A definitive overview of Kusunda phonology is far from complete and what we present here is both preliminary and tentative.4 on Apcial nasals and §5.1. but in other environments there is considerable free variation between the two sets – [i] varies with [e]. It bears little resemblance to Nepali or Indo-Aryan and even less resemblance to the surrounding Tibeto-Burman languages. Kusunda has a vowel system in some ways simpler (but also of a different kind) than either Tibeto-Burman or Indo-Aryan. and glottal points of articulation. but not all are contrastive. though. at least to a certain extent. 4. [˙] varies with [a]. See. like the environment of uvular ‘q’ or ‘ . An apical sound represented by /t/. palatal. and [o] would occur with [a]. There are several complex issues which will require more study and research before they can be termed complete. A striking aspect of Kusunda phonology is that it defies many South Asian typological norms. In some cases the harmonics mark a grammatical category. It is hoped that what we present here will point the next researcher in the right direction. with "servile" consonants being a subset of "radical" consonants. alveolar. In some environments.’ (We will treat this in more detail in §4. and [u] varies with [o]. The laminal/palatal series appears to have only weak contrastive lexical status. The phoneme [k]. as in [holadßn].3. velar. as in [hul˙dßn].

We expect that. velar ‘g’ and uvular ‘ ’ are routinely lost. the most fluent speaker. our wordlist will undergo considerable revision and refinement.Himalayan Linguistics: Archive No. Phonological precision grew as the morphology became more transparent.’ Thus. compounds and polymorphemic strings. the speaker with the least native-speaker ability uses contraction least. and Kamala. but so do consonants. or even three. highly assimilated. In Kamala’s speech. uses contraction most. “hidden” in the compound dui˙Ôe are two. and it is very likely that some are contractions of what may turn out to be. cycles of phonological contraction. includes a contracted form for [getse] ‘child. It is also doubtful now that [z] or [Ô] are phonemes at all in Kusunda. upon further investigation. occurring only in intervocalic environments of compressed speech as allophones of [ts] or [dz]. offspring. but in any register. We found ourselves having to play one speaker off the next in order to effectively isolate the sounds and morphemes in a given string. for example. [duktsi getse] > b. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– ‘You will go’ is one such example. In our early transcriptions.’ 6 23 . (supported by tape recordings) our transcription from Kamala’s speech for ‘man’ (a male human being) was: (5) (6) (7) [dui˙Ôe] [dui getse] *duktsi getse (lit. In still other cases the harmony principle works as a marking mechanism for a grammatical category. We emphasize that all the variables at work in Kusunda phonology are not fully understood or even identified. not just in rapid speech. For some words we have identified a kind of harmonic principle – not only do vowels tend to occur in harmonic sets. the loss of the latter sound being compensated for by a kind of pharyngealization on the surrounding vowels and a lowering of pitch (more on that later). ‘boy child’) which has now been modified to: which in turn (possibly) comes from an earlier: The contraction process is as follows: (8) a. very likely. 4. The glossary that we include in this description has about 850 items. strings. which also. She compresses a lot of speech into fairly short. [du i ˙Ôe]6 The form duktsi itself is very like a contraction of an earlier *du getse [male child] ‘son.2. pending further research on Kusunda. It is interesting to note that in Reinhard and Toba’s (1970) description. Prem Bahadur. Contraction Another factor contributing to much difficulty in isolating Kusunda sounds has been the Kusunda penchant for extreme contraction of sound strings. the word for ‘son’ is listed as duguci [dugutsi].

pæoÚdzi a. minimal pairs for vowels. but one word utilizes a vowel from the upper set and the other word utilizes the corresponding vowel from the lower set – i. as in the following: (14) (15) a. Kusunda has six vowels. gidz˙≥ ~ k˙pa≥ b. mu≥ ~ mo≥ c. gulu≥ ~ golo≥ b.3. as in the following: (9) Kusunda vowels: i ˙ e a In terms of absolute contrasts. Following are examples: (10) Variation between [i] and [e]: ~ getse a. gitsi b.e. there is a certain amount of free variation between the two sets (as borne out in tape recordings from one day to the next). Vowels As already mentioned. tukan ~ holadßn e. three forming an upper set and three forming a lower set. gip˙n b. gip˙n ~ gidze c.’ ‘language’ ‘flower’ 24 . it is difficult to find contrastive pairs of words in which the consonants remain constant. gipan ‘He jumped.’ ‘body’ ‘tumeric’ ‘I cooked it. swollen’ ‘know’ ‘I cooked it. the variation reported in (10a) constitutes a minimal pair in careful speech: (11) a. gidzi ~ mehaq d. hul˙dßn ‘round’ ‘king’ ‘fat. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 4. yi≥gu ‘child’ ‘language’ ‘name’ ‘Kusunda’ ‘stone’ o u Interestingly.’ (13) As with the contrast between gitsi ‘thorn’ and getse ‘child’ illustrated in (11). contrastive pairs can be found. pæuÚdzi b. In many words. gitsi ~ gep˙n b. getse ‘thorn’ ‘child’ Likewise. with these vowels too.Watters: Kusunda Grammar. k˙p˙≥ ~ holadßn c. mihaq ~ ye≥gu e. hyudzi ~ hyodzi ~ tokan d. similar variation occurs between [u] and [o] as well as between [˙] and [a]. especially within polysyllabic morphemes. as illustrated in the following examples: (12) Variation between [u] and [o]: a.’ ‘He shot. hul˙dßn Variation between [˙] and [a]: ~ gidza≥ a.

e. e. imperative’ ‘day before yesterday’ ‘trans. c. occur only with vowels [o] and [a]. then. displaying a kind of harmony within a syllable. goloq ‘itch’ ‘sharp’ ‘Come!’ ‘bad’ ‘owl’ ‘sheep’ (20) This occurs also in Eskimo-Aleut. c. somewhat ambivalent in their preference of upper or lower vowels. qolom b.) 4. kan c. o.3. -di i-dzi tap-i wi imbatu tugun -˙gu guben nu ˙di imb˙-di gip˙n dz˙g˙l l˙pa ‘first person past’ ‘he said’ ‘chicken’s’ ‘house’ ‘think’ ‘bug’ ‘I came’ ‘negative’ ‘log’ ‘you’ ‘I became’ ‘I returned’ ‘language’ ‘old woman’ ‘blood’ [e] -de e-dzi tape we emblaq [o] -to to on -˙go gobba nok [a] adi imba-di gipan dza lab˙‘converbal marker’ ‘he gave’ ‘to here’ ‘good’ ‘witch’ ‘intrans.7 It is not altogether surprising. Vowels and uvular consonants Only vowels [o] and [a] can occur following uvular consonants. however. the velar sounds. ˙] and lower vowels [e. [g].1. b. as in the following: (19) a. d.‘become’ and causative a. 8 7 25 . It is significant. kolumu c. that uvular sounds. and [a] occur before labial or velar consonants. but in contact with uvular consonants they become [o]. d. and [ˆ]. ˙ga a. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– It turns out that much of the free variation we encountered occurs in polysyllabic words where [h] or one of the velar sounds [k]. [˙] a. is unidirectional – it is a constraint on the consonant. [u] a. qan b.8 The constraint. though they too seem to have found some phonemic evidence for a [k] vs. b. in this respect. a]: (16) [i] a. to find their neighbors. [i]. too. and [a]. c. u.4 and example (243).Himalayan Linguistics: Archive No. Other consonants. Though minimal contrasts are rare in our data set. In Reinhard and Toba’s transcription all [q]s are followed by an [a] or [o]. imperative’ ‘tall’ ‘you plural’ ‘I made’ ‘I thought’ ‘flower’ ‘fire’ ‘to plant’ (17) (18) (See also the contrast between inchoative ˙. can occur with [o] or [a]. [q] contrast in central vowels. [e].‘make’ in §6.1. not on vowels. Bergsland (1997:21) reports that the vowels [u]. the following words and morphemes further illustrate the contrast between upper vowels [i. or [≥] occurs. b. all retracted. d. e. especially the voiced ones [ ].

Pharyngealization in Kusunda produces sounds very different from the sounds of breathiness in Tibeto-Burman. The uvular consonant [q] is less restricted in terms of the vowels that can precede it. Pharyngealization is manifested by a “ratchety” quality on the vowel. “low register” is linked historically to the voicing of syllable initials. not so much following ones. *s-ma≥ > Magar mha≥. is historically and qualitatively different from what has been described for Tibeto-Burman or Indo-Aryan. ukæi-q-ßn9 b. though. as in the left side of Figure 1: The -q here fits the morphological system as -q (anti-causative). Kham m~ïÚh Where breathiness occurs on syllables beginning in voiced stops. have “voice-register” systems. but includes also a “retracting” shift in the word’s consonants (see §5.c. Gerrold Sadock (p. in which “lax” register vowels (opposed to “modal” or “normal” register vowels). as in: (22) a. represent them as voiced aspirates.1). are manifested by breathy phonation and a concomitant lowering of absolute pitch levels on the syllable. its presence being perceived mostly by its modifying effect on preceding vowels. namely Magar and Kham. What we refer to here as “pharyngealization” in Kusunda.3. though also yielding a concomitant lowering of pitch on the affected syllable. 10 9 26 . ˙r˙q ‘ruined’ ‘garlic’ Pharyngealization of vowels and lowering of pitch Tibeto-Burman languages spoken in the regions surrounding Kusunda. they arise from different phenomena. In both languages.Watters: Kusunda Grammar. and practical orthographies in Devanagri often represent the difference between low and high registers as a difference in voicing. Ghil‘ad Zuckermann (p.c. in fact. front vowel. but may be phonetically altered to -k by the preceding high. almost as dramatic as the cawing of a crow.) reports that voiced uvular [ ] in Greenlandic Eskimo is barely audible. Kham ruhs b. it affects preceding vowels.5. Pharyngealization in Kusunda is linked to the following voiced uvular consonants [ ] or [ˆ]. and on a spectogram displays a wave form with a series of rapidly occurring strirations.on the syllable. i. *s-rus > Magar rhus. Ladefoged and Maddieson (1996:168) claim that the so-called pharyngeal fricatives of Semitic languages are “often neither pharyngeals nor fricatives” but better termed as approximants. as in: (21) 4. breathiness can sometimes be traced to the loss of a prefixed *s. written in Devanagri. We have examples of [q] at the end of a syllable preceded by [i] or [˙]. and John Saeed (p. the auditory impact is somewhat similar to the “voiced aspirates” of Indo-Aryan languages. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– This observation will be important to our argument later on.c) reports a similar pharyngealization in Arabic. that the “mark” in certain grammatical systems is more than a shift from high vowel to low vowel.e.1 0 The effect is regressive. In other Tibeto-Burman languages of Nepal.3. a. especially in the Tibetic and Tamangic languages. The practical orthographies of both languages.) the same for Cushitic languages. Uvular and pharyngeal consonants have been noted elsewhere too for their pharyngealizing effect on preceding vowels.2.

Instead.4.Himalayan Linguistics: Archive No. Each syllable is written as a combination of glide and vowel.3. The presence of [ˆ] is signalled additionally by nasalization on the affected vowel. Diphthongs Most vowel sequences are diphthongs. above’ In Kamala’s speech. [ ] and [ˆ] have virtually been lost. monomorphemic diphthongs. these are the primary auditory cues to their presence. In fact. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Figure 1. each having a prominent peak. the historical underpinnings for these consonants are a little easier to perceive. In Gyani Maiya’s speech.1.3. and those that move in and out of the center towards the periphery. 4. Vocalic sequences There are numerous vocalic sequences in Kusunda. the vowel denoting the peak. those that move in vowel space from one periphery to another. Following are examples of monosyllabic.2. but none are interpreted as triphthongs or quadriphthongs.3. see §4. beginning with central vowels moving outwards: (23) CENTRAL FRONT: a. sometimes almost surfacing in careful speech.3. and most occur within the domain of a single syllable and a single morpheme. Spectogram of [a k˙] (< a k˙) ‘up. being replaced by compensatory pharyngealization on the preceding vowel and a lowering of pitch levels. (For more discussion on the actual consonantal value of these sounds.4. [˙i] ˙igiag˙i m˙i p˙idzi s˙i ‘to live’ ‘dog’ ‘mother’ ‘clothing’ ‘bear’ 27 . long sequences are divided into multiple syllables. There can be up to three or four vowels in a single sequence.) 4.

as in the following: (26) a. HIGH-FRONT LOW-CENTRAL: [ia] yaq ‘snow’ byaq-ßn ‘overturned’ gya ‘male of a species’ gya-q-ßn ‘collapsed’ dæya ‘old man’ myaq ‘leopard’ 28 . [˙u] ˙ul˙dz˙u p˙u-tsi-n ao aota qaoli qao-dzi b. Thus. BACK FRONT: [ui] wi dwi gwidzi pwi twi [ue] ~ [oe] wen / we-y˙n [oi] b. as in the following examples: (25) a. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.Watters: Kusunda Grammar. [ao] Some monosyllabic diphthongs go from one periphery of the vowel space to the other. FRONT BACK: [eo] ~ [io] hyoq-ßn hyo-w˙-n yeodze [iu] ~ [eu] yu myu dzyu-t-ßn p˙≥gyu ‘house’ ‘husband’ ‘tortoise’ ‘iguana’ ‘bee’ ‘good’ (no examples to date) ‘hide’ ‘swollen’ ‘now’ ‘ear’ ‘arrow’ ‘leaked’ ‘lizard’ Other vowel sequences that move from one periphery to another are those from high-front to low-central and from high-back to low-central. and another from front to back. a common sequence is from back to front. [ai] ai-t-ßn qæai-d-ßn qaida la~î BACK: ‘I beg’ ‘I was frightened’ ‘other’ ‘cucumber’ ‘to sell’ ‘pain’ ‘I am tired’ ‘dirt’ ‘door’ ‘god’ ‘rotten’ (24) CENTRAL a.

a morphemic rendition is given on the left (written in a working orthography) and a syllabic rendition is given on the right: (28) FOUR VOWELS. HIGH-BACK LOW-CENTRAL: [ua] gwa ‘egg’ wa≥ ‘firefly’ so-wa-d-ßn ‘I washed it. TWO SYLLABLES: a.3. The first example is monomorphemic and the second bimorphemic: (29) FOUR VOWELS. c.’ There are also a few homorganic diphthongs with little or no movement from one vowel to another. [iu˙a] b. In the examples following. are monomorphemic: (30) THREE VOWELS. d. TWO SYLLABLES: a. [uiii] morphemic qaw˙i twi-yi syllabic [qa-w˙i ] [twi-yi] ‘jackal’ ‘bee’s’ In like manner. BACK BACK: (no examples to date) 4. as in the following: (27) a.2. but not all.’ gæu-wa-t-ßn ‘I moved it.Himalayan Linguistics: Archive No. THREE SYLLABLES: a. Double diphthongs ‘firewood’ ‘here’ ‘tree’ ‘eye’ ‘mine’ ‘stone’ ‘foot’ There are a few examples in Kusunda of up to four vowels in a single sequence. All such sequences are polymorphemic and polysyllabic. e. Some spread across as many as three syllables – two diphthongs and one monophthong.’ dæ~u-wa-d-i ‘I stood it up. b. [aio] [aia] [aoi] [˙ui] [˙io] morphemic qaiyo qæaiya awi p˙wi p˙iyo syllabic qai-yo qæai-ya a-wi p˙-wi p˙i-yo 29 ‘skirt’ ‘danger’ ‘hand’ ‘bamboo’ ‘thirst’ . there are also sequences of three vowels that divide into two syllables.3. [iu˙˙] morphemic kiw˙-a-t-ßn piw˙-˙go syllabic [ki-w˙-at-ßn] [pi-w˙-˙-go] ‘I pinched it. FRONT FRONT: [ei] deidzi teisa [ii] yi yi≥ tsi-yi [ee] ye≥gu yen b. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.’ ‘Peel it!’ Following are sequences of four vowels that divide into two syllables. [au˙i] b. Most of the following examples.

3. o p.3. 30 . q.’ ‘father’ ‘pus’ ‘mine’ ‘skinny. no diphthong Where a sequence [ia] or [i˙] is broken by a morpheme boundary. 4. Nasalization Nasalization appears to be a new phenomenon in Kusunda. g.’ ‘Dance!’ ‘Keep it!’ Apart from the sequences shown in (26). FRONT [e˙] CENTRAL: leÚ-˙-g-˙n p˙tse-˙go seÚ-˙-go CENTRAL: ‘He swept. In such cases the central vowel has its own peak. i. These are not diphthongs: (31) [ia] [i˙] abi-a-t-ßn abi-˙-g-˙n limi-˙go gwi-˙go ‘I carry. [˙u˙] [˙ie] [ui˙] [iu˙] [i˙i] [iao] [iii] [eia] [ei˙] [eu˙] [eo˙] [ou˙] tæ˙w˙ p˙yet gwi-˙go ts-i-w˙-n y˙i gyao tsi-yi keya≥ ey˙≥ bew˙hyo-w˙-n mo-w˙-d-i tæ˙-w˙ p˙-yet gwi-˙go tsi-w˙n y˙i gyao tsi-yi ke-ya≥ e-y˙≥ be-w˙ hyo-w˙n mo-w˙-di ‘place’ ‘leech’ ‘Keep it!’ ‘I brought it.3. j. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– f.3. Following are examples: (32) a. the syllable also breaks.’ Two vowels. BACK [u˙] sw˙tt˙i ruÚ-˙go c. each with its own peak. l.’ ‘Bury it!’ ‘I spoke.4. most two vowel sequences whose final vowel is central occur across morpheme boundaries. m.’ b. the same speaker may use both variants – a nasal consonant or a nasal vowel. h. k.’ ‘Give it back!’ ‘Catch!’ ‘all’ ‘Pull!’ ‘Buy it!’ ‘I called. In the latter case. CENTRAL CENTRAL: [a˙] dza-˙go [˙a] umb˙-a-t-ßn [˙˙] l˙b˙-˙go ts˙-˙-d-i 4. occurring either in Nepali loan words or in native words as compensation for a following lost velar or uvular nasal consonant. It is assumed that the nasal consonant is the underlying form.’ ‘He carries. n. thin’ ‘upper back’ ‘be born’ ‘swollen’ ‘I boiled it.Watters: Kusunda Grammar.

can be dental. What is more material for Kusunda. d. uvular. dental. velar. c. is still marginal in many contexts. voicing originally being the unmarked norm. e. it seems. or retroflex. An apical consonant. gæ~as b~ara p~asula a~ola bæ˙~îra tsæ~a-d-i dæ~u-tsi-n ~ dæu≥h~ukui ~ hu≥k˙i r~ako ~ ra≥kwa t˙~îna ~ t˙ndin m˙l~a-a-t-ßn n~oÚdze y~a-˙-g-˙n gæ~aÚ-d-i ‘grass’ ‘pot. are non-contrastive. f. b. velar. (34) a. and that in relatively recent times. c. apical. g. laminal. c. Fricativization. however. h. [Ô]. and uvular. is to describe consonant sounds in terms of their active articulator – labial. pharyngeal. Many of the points.4. b. b. apart from phonemic /s/ and allophonic variants of most stop consonants ([∏]. e. c. Following are examples: (36) NO KNOWN UNDERLYING CONSONANT: a. and [≈]) is non-existent. d. and there can be considerable variation even in single words. The voicing opposition on consonants. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (33) NEPALI LOANS: a. b. alveolar. it is the articulator that counts as contrastive. above’ ‘defecate’ ‘I ground it’ In some cases it is unknown if the nasalized vowel can be attributed to a lost nasal consonant. and many of the modern lexemes that utilize it can be traced to TB or IA. cooking vessel’ ‘shin bone’ ‘finger’ ‘sparrow’ ‘I tied it’ < chan ‘hobble’ ‘I stand’ ‘small’ ‘millet’ ‘today’ < < < < m˙la≥ nu≥ ya≥ gæ~aˆ˙n ye≥gu ‘slowly’ ‘rise’ ‘feces’ ‘grindstone’ WITH UNDERLYING NASAL CONSONANT: WITH POSSIBLE UNDERLYING NASAL CONSONANT: ‘I was late’ ‘up. and glottal. b~aqa la~î lib~u-dzi ¯~aÚ-d-ßn p~ago p~ayi h~utu h~aÚ ‘sickle’ ‘cucumber’ ‘to play’ ‘I removed it’ ‘five’ ‘cloud’ ‘far’ ‘face’ Consonants Consonant sounds in Kusunda occur at many points of articulation – bilabial. [©]. palatal. (35) a. it has been gaining in non-native features.Himalayan Linguistics: Archive No. f. 4. due to the influence of TB and IA. The actual point of articulation can be extremely variable. 31 . d. It appears that the original sound system of Kusunda may have been quite simple. for example. retroflex. for example. alveolar. We have recorded only the nasal vowel. d. Aspiration appears to be new.

even if it is a voiced consonant.vs. many of the allophonic problems that occur elsewhere in the phonology also occur in the labial series. [amba] b.4.Watters: Kusunda Grammar. b-): a. There are five sounds to which we have accorded phonological status – [p]. and [m] (plus the semivowel [w]). [pitadi] b. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– labial apical laminal velar uvular glottal stop – voice p t k q ÷ + aspir pæ tæ kæ qæ + voice b d g ( ) + aspir bæ dæ gæ fricative s (≈) affricate – voice ts (t ) + aspir tsæ + voice dz + aspir dzæ nasal m n (¯) ≥ ˆ lateral l rhotic r semi-vowel w y (palatal) –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Table 1. another with aspiration. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– In Table 1 is an inventory of contrastive consonants in Kusunda. as in the following examples: (37) CONTRAST (p. [b]. (¯). [bi itßn] (38) NEUTRALIZATION: ‘I untied it. However. (t ). [tab-i gwa] Neutralization does not occur if the preceding sound is a consonant. and ( ) is the most common phonetic realization of ( ). One problem has to do with voicing. [tap] b. as in: (39) a. and (≈) are somewhat questionable consonants. the contrast is sometimes neutralized intervocalically (especially in rapid speech and if the syllable in question is unstressed). [bæ].1. CONSONANT INVENTORY IN KUSUNDA 4. and still another with fricativization. and we will do well to deal with them here.’ ‘chicken’ ‘chicken egg’ a.’ ‘It dawned. The labial series The labial series in Kusunda is fairly simple and straightforward. [pæ]. [hampe] ‘meat’ ‘where’ 32 . Though there is a clear voicing contrast word initially.

in fact. [d˙∏˙ i] b.’ 4. but some appear to be partial borrowings. found in Kham.4. as in: (41) a. underlying /b/). in fact. we can show that underlying /p/ surfaces as [b] intervocalically and underlying /b/ surfaces as [∏]. We suspect. Following are examples: (40) FRICATIVIZATION: a. If. resembles Nepali [pæeri]. The fricative variants do not have phonemic status. bæ˙≥ero We have two instances of [bæ]. c. but we do not yet have sufficient evidence to support our hypothesis.1. d. b. The word [k˙pæera] ‘again.1. pæopæepßn pæirutßn pæotoq pæurlu≥ pæuÚdzi pæyapæyaks˙m ‘shoot a gun’ ‘drop’ ‘full’ ‘short’ ‘red’ ‘jump’ ‘wash clothes’ ‘horse’ ‘break (rope)’ ‘sparrow’ < Nep.2.’ It is unnecessary to trace the two TB loans back to PTB. turn out to be [∏]. due to areal influence from Indo-Aryan and Tibeto-Burman. Magar.1. (i. one of which appears to have a Nepali source: (43) a. Words with aspirate [pæ] that appear to have a native origin are the following: (42) a. e.4. bæ˙~îra 33 . [ i∏a] ‘there’ ‘quickly’ Without the information in (40) or in (41b). and may very well be a novel innovation. that this is what is happening.’ for example. may. There is a possibility too that some intervocalic sequences that we have represented as /w/. [pæ˙laq] ‘bed’ resembles Nepali [pæ˙le] ‘plank’. we would be supported in representing (41a) ‘there’ as /sape/ and (41b) ‘quickly’ as /siba/. Magar. bæ˙raqßn b. [sabe] b. Bilabial fricatives Voiced labial consonants often undergo fricativization when occurring between vowels. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 4. h.e.Himalayan Linguistics: Archive No. g. conventional wisdom would recommend that we write [sabe] as /sabe/. and some Kham dialects). there are cognate forms closer at hand. however. and [pæela≥] ‘flat’ resembles TB *ple≥ ‘flat. f. We have 13 instances of [pæ] in our wordlist. Bilabial aspirates Aspiration has marginal status in Kusunda. as in: [d˙∏˙ i] ~ [d˙o i] ‘later. and Chepang. [yi∏iu] ‘later today’ ‘scratch’ A question arises with respect to polysyllabic morphemes in which a voiced bilabial stop occurs intervocalically in some words and a voiced fricative in others. [pæurpæur] ‘run’ resembles Tibeto-Burman (TB) *pur ‘fly’ (which is aspirated in Chepang.

[coq] ~ [tok] h. tends to occur preceding high. for example. sounds in the apical series occur at several non-contrastive points of articulation – dental. [tu] d. [t˙m˙n] RETROFLEX: e. and the palatal variant in harmony with uvular consonants. The apical series The apical consonants are those articulated with the apex (or tip) of the tongue. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 4.2. retroflex.2. In terms of raw phonetic values.4. one of which is Nepali. The stop variants are not absolute. the retroflex variant preceding [o] or [a]. [Êoba] PALATAL: g. [t]. it is the active articulator that is contrastive. The dental variant. The four voiceless sounds are variants of an apical /t/ and the four voiced sounds are variants of an apical /d/. b. e.Watters: Kusunda Grammar. alveolar. the alveolar variant in neutral contexts preceding [e]. [Êa≥] f. and another which has a possible Kham source: (46) a. We have already seen that aspiration has a marginal status in Kusunda. c. Apical stops The following raw stop consonants occur in the apical series in Kusunda – [@t]. [Î]. dæab˙-di . possibly from Kham. and [c] in the voiceless series. [u]. ‘rhododendron’ (lit. [Ê]. and some palatal variants. and [@d]. as in the following examples: (44) DENTAL: a. d. We have five examples of [tæ]. front vowels. and [Ç] in the voiced series. [d].1. but phonological tendencies. dæundi b. or [˙]. ‘place’ < Nep. [qocu] • Aspiration. tæal tæ˙w˙ tæag˙n ye≥gu tæum˙n ye≥gu witæu ‘plate’ < Nep. [teisa] c.4. three of which are from Nepali: (45) a. Here again. and the series is rife with phonological variants. 4. [@tigi] ALVEOLAR: ‘our’ ‘here’ ‘snake’ ‘I ate’ ‘water’ ‘sew’ ‘we’ ‘bird’ b. This is true also for apical consonants. tæa~u ‘flint’ (lit. related to the Kham custom of young men [dhapa] wearing rhododendrons in their hair) ‘shut’ 34 c. dæapa gipan ‘fog’ < Nep. thoknu ‘strike’) ‘mortar’ ‘slippery’ We have six examples of [dæ]. ‘striking stone’ < Nep. dhapa flower.

e. (i. it is an underlying [t]. [ i en] b. [o enti] c. too. dæ~uf.2. as in the following: (48) [ ] variant: a. occurs as a variant of [dz] or [ts] in intervocalic environments in which the following vowel is [i] or [e]. The phoneme /s/ has a palatalized variant [ ] which occurs preceding front vowels. under closer scrutiny and more careful speech. [dui˙Ôe] Apical affricates < < [deidzi] [dui getse] ‘firewood’ ‘male human’ The apical affricates [ts] and [dz] occur commonly in our data set. It occurs only in “compressed” or “contracted” speech. but there is also a possible contrastive [t ]. Thus. [sodzaq] ‘rice (in field)’ ‘girl’ ‘eagle’ ‘winnowing tray’ (Nep. (See §4.2.2 immediately below.) The two affricates commonly occur with other vowels. and -dzi being a frequently occurring aspectual marker.1. been shown to have an underlying [t]. described in §4. we have not recorded [Ï] anywhere. [deiÔi] b. As with fricatives. (We are viewing these as allophonic variants in such environments.) ‘bear (the animal)’ ‘greetings’ (49) The voiced counterpart of [ ]. [supo] b.4. In fact.3 and in the section on “mutation” – §5.) 4.Himalayan Linguistics: Archive No.4. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– d. [gid˙t] ~ ~ [n˙ti] [git˙t] ‘who’ ‘skin’ As in the discussion following example (41). [Ô]). first person tsi occurs as [t i]. as in the following: (47) a. we suspect that in many places where [d] occurs intervocalically. and the aspectual marker -dzi occurs as [dÔi]. Our case would be stronger if we also found [Ï] in environments analogous to the environment for [∏]. [s˙i] c. a.3. although [z] also occurs as an allophonic variant of voiced and voiceless affricates [dz] and [ts] respectively.3. as in the following examples: 35 . dæya ‘put on fire’ ‘stand’ ‘old man’ • Intervocalic voicing.4. Apical fricatives Only the apical fricative [s] occurs in our data set with contrastive status.2. as in: (50) 4.2. [ i≥ki] [s] variant: a. Some words originally transcribed with intervocalic [d] have.5. dæaÚe. [n˙di] b. Stops in the apical series display some of the same characteristics presented for stops in the labial series – primarily a tendency for intervocalic voicing. the affricates are palatalized before front vowels. tsi being the first person singular pronoun.4.

/s/.1. but there is also a possible contrastive [¯]. [~oÚ¯i] • Syllabic ‘n’. and palatal points of articulation. Following are examples: (54) a. Syllabic ‘n’ is usually preceded by a stop consonant.’ 36 . abi-a-t-ßn carry-do-1-REAL b.’ 4.’ < Nep.Watters: Kusunda Grammar.’ ‘sickness’ ‘old woman’ ‘soak’ d. One of them appears to be derived from Nepali. [dz˙g˙l] f. retroflex. [dzugui] e. described in §4. but there are also cases where it is preceded by [n]. the resonant consonant ‘n’ occurs very frequently with its own syllabic status. [m˙¯¯i] b. chan ‘hobble’ ‘He weeps.3 and in the section on “mutation” – §5. In our discussion of apical consonants so far. /d/.2. borlo-q-ßn spill-AC-REAL d. alveolar. This is generally true for the apical nasal consonant /n/ as well. front vowels (analagous to the palatal variants for the other apical consonants /t/. as in: (52) a. In our data set. bal-ßn descend-REAL c. [n˙ti] b. tsæ~aÚ-d-i b. [tso on] AFFRICATE DZ: ‘to be (existential)’ ‘I went.’ ‘I put.4.’ ‘It spilled.’ ‘I/you descended.’ ‘You hit it. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (51) AFFRICATE TS: a. palatal consonants have been treated as an allophonic variant of a generic apical series which includes dental. [dzoqto] We have a only two cases of aspirated apical affricates. pumba-n-ßn hit-2-REAL ‘I carried it.) (53) NON-PALATALIZED: a. and /ts/). in which case it is written as [ßn]. [numba] PALATALIZED: ‘who’ ‘cow’ ‘many’ ‘big’ a. with a palatal variant [¯] preceding high. (We are viewing [¯] as an allophonic variant of [n] in such environments. [tsu] b.5.4.3. [ts˙g˙n] c.4. dzæ˙m-dzi Apical nasals ‘I tied it.

the only laminal consonants are /¯/ and /t /.2). as in: (55) a. and for that reason they will be discussed together.4. Thus. Though there is good evidence for a contrast between velar and uvular consonants. both somewhat difficult to justify. 4. but we have so far been unable to distinguish it clearly from a fricative series: 37 .3. [¯~aÚ ˙di] ‘this’ ‘I removed it’ This particular contrast is somewhat suspect and may. as in: (56) a.’ ‘You will go. [na] b. One such environment is where it “harmonizes” with a following uvular cosonant. As we have just seen in example (53).’ ‘I will go. in the following example. the phoneme /k/ can have pre-velar and post-velar variants depending on whether the associated vowel is high-front or low-back. Thus. but there are some minimal environments that dispose us to treat it as a separate phoneme. Apart from the palatal glide or semi-vowel /y/ (see §4. [ts˙g˙n] d. As with consonants in the apical series. absolute points of articulation may vary slightly from one word to the next according to the vowels which follow. [n˙g˙n] b. as are also [ts˙g˙n] and [t a an]. [t a an] ‘You went.3. for example. Fortunately. as in: (57) a. the fact is that in certain grammatical environments the palatalization of both ‘n’ and ‘ts’ is part of a larger autosegmental “mark” that spreads vocalic and consonantal harmony across the entire word.’ ‘I went. in fact. There may be an aspirate series for uvular consonants.4. we have often found it difficult to distinguish between a post-velar [ßk] and a uvular [q].’ Though it might be possible to write the laminal as a simple /n/ and rely on the following uvular consonant to trigger the appropriate variant [¯]. [¯aqtsin] ‘wife’ ‘I wait. of the same verb.4. especially because they commonly occur in identical contexts. [n˙g˙n] and [¯a an] are realis and irrealis forms. in contrast to the tip of the tongue in apical consonants. [¯~aˆdi] b.4. The velar and uvular series There are difficulties inherent in distinguishing the velar and uvular series of consonants.’ The phoneme [¯] also occurs in at least one environment where it is unpredictable and contrasts with [n]. [¯a an] c.Himalayan Linguistics: Archive No. [¯] can occur as a variant of /n/. respectively. The laminal series Laminal consonants are formed by the blade of the tongue against the palate. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 4. turn out to be [¯~aˆ ˙di].5). the task of distinguishing velar and uvular points of articulation is simplified in voiced consonants – primarily because of the concomitant features of pharyngealization and lowering of fundamental frequencies in the uvular series (see §4.

floor’ ‘moon’ In Kamala’s speech. In this respect. Following are examples of [k]. b. c. or to Ø. The velar consonant [k] occurs in all contexts – syllable initial. Kusunda differs noticeably from Tibeto-Burman languages in which nasal [≥] is very common syllable initially. b. a non-fricative [g] occurring between vowels can be an indicator of underlying [k]. as in the following contrastive sets: 38 . b. Intervocalically. it is common for ‘g’ to occur as a fricative [©]. [≥] a. b.4. [a©˙i] [d˙©˙i] [du©˙] [nigu] < *niku ‘dog’ ‘he went’ ‘ground. No word in Kusunda that we have found begins with nasal [≥]. qæai-dzi ~ ≈ai-dzi b. the fricative is reduced even further. and [≥] in various contexts: (59) [k] a. syllable final. [ßkan] [ßkampe] [ßkolde] [kiptsaßk] [ki] [kiladi] [garo] [gebusa] [gimi] [guhu] [goloq] [gwa] [gida≥] [ha≥nu≥] [dæu≥to] [mo≥] [dzi≥] [k˙p˙≥] [ye≥gu] ‘sharp’ ‘where’ ‘knife’ ‘Sesi Kham’ ‘louse’ ‘thief’ ‘wall’ ‘curd’ ‘money’ ‘bone’ ‘sheep’ ‘egg’ ‘fluid’ ‘shadow’ ‘Stand!’ ‘king’ ‘oil’ ‘tumeric’ ‘stone’ (60) (61) • Fricative variants for velar consonants. [g]. e. d. e. e. Its voiced counterpart [g] occurs in the same contexts except syllable final. [g] a.Watters: Kusunda Grammar. d. By contrast.1. f. f. either to a very light fricative. and with any vowel. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (58) a. As in the labial and lingual series. noqæok ~ no≈ok c. m~aqæ ~ m~a≈ Velar consonants ‘afraid’ ‘maternal grandfather’ ‘dream’ 4. d. c. the velar nasal consonant [≥] occurs only syllable finally. c. f. g. c.4. d. as in (62d): (62) a.

to be a fricative [x] also occurs with equal frequency as an aspirate [kæ].1. kæola lekæadßn kæurpa kæ˙rgun kæaidzi kæamdzi kæaÚ u kæakæ˙≥gu kæ˙rwi nikæedi ukæ˙‘stream’ < Nep. f. khurpa. h˙ra-q-ßn open-AC-REAL b.is likely another instance of third person inherent possession. because of its instability in a predictable direction. at least three of which have a Nepali source. gya-q-ßn collapse-AC-REAL > h˙ra-≥ > open-AC:REAL > gya-≥ > collapse-AC:REAL ‘open’ (participle) ‘collapsed’ (participle) • Aspiration.2. We have twelve examples of [kæ]. c. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (63) Gyani Maiya: a. g. 39 .1. d. b. b. the word for Ban Raja. when spoken rapidly comes out as [mæya≈]. In rapid speech. as in: (66) a. and one of which has a Kham or Magar source: (67) a. [kæaidzi] [kæaÚ u] [kæ˙≥gu] [nikæe-d-i] ~ ~ ~ ~ [xaidzi] [xaÚ u] / [≈aÚ u] [x˙≥gu] [nixe-d-i] ‘food’ ‘is not’ ‘cold (of food)’ ‘I laughed’ • Consonant [≥] as a “coalesced” consonant.’ which marks realis. the sound [≥] sometimes arises as a coalescence of anti-causative ‘q’ plus syllabic ‘n. [oraÚi] ‘He killed it’ (realis) ‘He killed it’ (past) ‘He killed it’ (realis) ‘He killed it’ (past) (64) It is difficult to determine if there is a velar fricative [x] in the data set. j. mihaq. Magar khur ‘knife’ ‘red’ < Kham kæ˙ir. The word in (68d) shows up in the Reinhard and Toba data as gihan. Following are examples: (65) a. lekhnu ‘knife’ < Nep. h. [oragi] Kamala: a.1 1 and suggests another possible source for voiced aspiration – CVhV. one of which is Nepali. and another which is likely from Nepali. See §5. What sounds. we are assuming that /kæ/. k.‘brownish-red’ ‘food’ ‘bite’ ‘is not’ ‘parch grain’ ‘cold (of food)’ ‘wheat’ ‘laugh’ ‘ruin’ We have five examples of [gæ]. c. l. d. 11 The gi. Kham.Himalayan Linguistics: Archive No. ‘write’ < Nep. Indeed. [oraÚn] b. e. [orag˙n] b. in many contexts. Given that velar sounds are often fricativized anyway. has a variant [x]. i.

[yaqa] b.Watters: Kusunda Grammar. This fact by itself would lead us to believe that [q] and [ ] are in complementary distribution. In the following examples. [aqa] b. and not voiceless ones ([q]). ‘hot’ < ? Nep gæ˙r˙m ‘move’ ‘genitals (female)’ ‘grind’ 4. we are interpreting the sound. b. [ˆ]). never surfaces as an actual stop. The solution is not so simple here. [ ]. Uvular consonants The uvular consonants [q]. and we experimented with [h] preceded by pharyngealization on the vowel. as in [Ù]. (indicating pharyngealization. c. pharyngealization will be a useful diagnostic for distinguishing [q] from [ ] in certain contexts.4. and [ˆ] contrast with the velar consonants [k]. however. [ya o] ‘porcupine’ ‘cold (weather)’ The problem we run into here is similar to the problem we have seen with other consonants – voiceless consonants. and we will write the sound in our text examples as / /. In our early transcriptions. though numerous words begin with [q]. as in the following contrastive pair: (69) a. but more often as a pharyngealized approximant [ ].4. in our data. many of our [ ]s are transcribed as [÷].1 2 later we began to realize that the sound had an anticipatory “glottal” or “pharyngeal” quality. and [≥].4. As we collected more data we began to detect distinct pharyngeal stricture during the production of the consonant. If we are correct in asserting that anticipatory pharyngealization on vowels is a feature that accompanies voiced uvulars ([ ]. intervocalically. [okti] b. [a k˙] ‘medicine’ ‘chest’ ‘below’ ‘above’ 12 In Reinhard and Toba’s transcription (1970). There are. realized especially on the preceding vowel). pharyngealization makes the difference between two otherwise similar sequences: (70) (71) a. no word begins with [ ]. it is sometimes articulated with a certain amount of fricativization. and though the sound is phonetically closer to [ ] than anything else. • A test for distinguishing [q] from [ ] in some contexts.1). [G] occurs as an interovcalic phenomenon. gæ~as gæ˙run gæutßn gæya≥ gæ~aÚ‘grass’ < Nep. 40 . As far as we can tell. as [ ]. and intervocalic [∏] is an underlying [b]. cases of [q] also occurring intervocalically. The caveat is that cardinal [ ] does not occur. With labial consonants we were able to say that intervocalic [b] is an underlying [p]. Rather. e. we wrote [ ] as [h] or [÷]. within the larger system. d. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (68) a. [o tsi] a. We have already commented that velar and uvular points of articulation are more easily distinguished in the voiced series for the simple reason that voiced uvular consonants are accompanied by other phonetic indicators – pharyngealization and a lowering of fundamental frequencies on preceding vowels.2. frequently surface as voiced sounds (see §4. [g]. [G].

b. like the contrast between velar [g] and uvular [ ] plays a role in TAM marking (see §5. [gi¯aq] [¯aqtsin] [¯a an] [¯aˆ an] [¯~aˆ di]1 3 [¯~aˆ gitse] [¯~aˆ dzi] ‘trousers’ ‘I wait. [t aÚ u] b. as in the following: (72) a.3.1). and claims that initial ≥. K. this would hint at a certain amount of variation between [q] and [ ] and weaken our test for distinguishing one from the other.Himalayan Linguistics: Archive No. If so. f. In the first pair below we give the affirmative forms for ‘I went. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– We also have a contrastive pair – gyaqai ‘breath’ vs. It is likely that the consonant in the causativized version is also an underlying [q]. In a process of anticipatory assimilation. [n˙g˙n] a. as evidenced by pharyngealization on the vowel preceding [ ]. We have found no occurrences of initial ≥. The negative construction is one such context. There are a number of intransitive verb roots which end in [q]. Uvular [ ] is compressed to such an extent in these kinds of high frequency environments that only its secondary (concomitant) phonetic features are perceptible. the alveolar nasal consonant in the sequence [na] is realized as a palatal consonant [¯a] wherever the following consonant is uvular.’ [t a u] [¯a u] ‘I did not go. gya ai ‘strength’ – which may turn out to be the same word. now voiced in the new environment. c. and until we are sure about voicing in causativization.5.’ In (74a) and (74b).is evidence of TB origins for Kusunda. Following are examples: B. There are several constructions in Kusunda in which an underlying [ ] can be posited based on the two principles discussed immediately above – pharyngealization (plus lowered pitch) and consonantal harmony. we are not yet sure if voicing plays a role in causativization.’ ‘You sat. the sequence [aÚ u] comes from an underlying [a u]. [¯aÚ u] < < ‘I went. 13 41 . Rana (2002) transcribes this as ngyangdi [≥ya≥di].’ ‘You will go.’ ‘wife’ ‘daughter’ ‘You look. or in any other grammatical marking. • Nasal consonant harmony with uvular consonants. g. e. [ts˙g˙n] b.’ and ‘You went.in our investigation. we will err on the side of caution and continue to write the intervocalic uvular as / / in causative contexts.’ • Establishing underlying [ ] in contexts where it is “compressed”. d. • The voicing contrast between [q] and [ ].’ ‘You did not go. Evidence for underlying [ ] is further supported by the change from [ts] and [n] affirmative to [t ] and [¯] negative in (74a–b). However. which when causativized end up as [ ] sandwiched between vowels. Unusual contrasts.’ In the second pair are the negative forms: (73) (74) a.’ ‘You went.

syllable finally. (81) [gi¯aq] [goloq] [ab˙q] [ukæiqßn] [¯aqtsin] INTERVOCALICALLY: a. not [q]. [byaq-ßn] spill-REAL CAUSATIVE: ‘It spilled. [hyøÙ ødi] ~ [hyø wødi] hyo -o-di hide-CAUS-1:PAST (77) [myáÙ atßn] ~ [myá atßn] mya a-t-ßn dig-1-REAL ‘I hid it.’ WITH NO INTRANSITIVE COUNTERPART: ‘I dug it. indicating that we are dealing with [ ]. b.’ a. it is difficult to distinguish between [q] and post-velar [ßk] which occurs before [o] or [a].Watters: Kusunda Grammar. As already mentioned. c. [qasti] b. and /o/ is lowered to [ø]. Following are examples of [q] in various environments: (79) SYLLABLE INITIAL: a. ha o b. [yaqa] .’ b. intervocalically. Phonology ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (75) INTRANSITIVE: a. Where [ ] occurs intervocalically between [a] and [o]. [byáÙ atßn] ~ [byá atßn] bya -a-t-ßn spill-CAUS-1-REAL (76) INTRANSITIVE: ‘I spilled it. [aqa] b. [hyoq-ßn] hide-REAL CAUSATIVE: ‘He hid.’ b.’ It should also be noted that in the three examples of / / above. or between two [o]s. The voiceless uvular consonant [q] occurs in a variety of contexts – syllable initially. as in (see also (69b)): (78) a. an approximate like sound [w] surfaces between the vowels. [qotu] (80) SYLLABLE FINAL: ‘one’ ‘bird’ ‘trousers’ ‘sheep’ ‘greens’ ‘ruined’ ‘I wait’ ‘below’ ‘porcupine’ 42 a. d. • [G] and semivowels. to on [ha wø] [to wøn] ‘rhesus monkey’ ‘day before yesterday’ • Uvular [q]. e. all are accompanied by pharyngealization on the preceding vowel. and in a few cases.

Kusunda has the liquid [l]. Most of these features have already been illustrated. c. lowering of fundamental frequencies. as in mehaq [meha≈] ‘Ban Raja’ or haq [ha≈]‘leaf. d. Following are examples: (83) LIQUID: a. glides. 4. c. [¯a an] b. and harmony with palatal nasals. d. e. however. it is this feature alone that distinguishes [ˆ] words from [ ] words. In this sense it is parallel to [≥]. No word begins with nasal [ˆ].4. b. [¯~aˆ an] ‘You will go. burlap’ ‘descend’ ‘old woman’ ‘Chettri’ ‘red’ ‘garlic’ ‘light weight’ FLAP: Glides (semi-vowels) occur as non-peak elements in vocalic sequences belonging to the same syllable.Himalayan Linguistics: Archive No. insist that they are able to distinguish the two. b.’ Even with a good digital recording it is extremely difficult for us to hear the difference between (82a) and (82b). f. (84) a. ye≥gu yaq p˙yet wa≥ li≥wa witæu p˙wi ‘stone’ ‘snow. c. Our informants. etc. and glides [y] and [w]. g. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Where [q] occurs word finally with a preceding [h] it surfaces as [≈]. In addition. as in (72d–g). the nasal consonant [ˆ] imposes nasalization on the preceding vowel.5.’ • Uvular [ˆ]. la an ligin balßn dz˙g˙l r˙ktsa pæurlu≥ ˙r˙q k~aw˙r ‘village’ ‘nettle. hail’ ‘leech’ ‘firefly’ ‘heavy’ ‘slippery’ ‘bamboo’ 43 . the flap [r]. The uvular nasal consonant [ˆ] has many of the characteristics and concomitant features found in [ ] – pharyngealization of preceding vowels. flaps. as in: (82) a. Liquids. as in: (85) GLIDES: a. d. b. In some words.’ ‘You sat.

5. nok THIRD PERSON: e. and gina can be modified by special pronominal numbers 1 5 which are different from cardinal numbers1 6 – thereby creating duals.’ ‘stomp away. (for the equivalent of ‘they. Van Driem (2001).’ pinjang ‘four. Word classes The major word classes. and other In Reinhard and Toba’s report (1970). The difference between their data and ours may be attributable to dialectal differences or to further erosion of grammatical categories in the language. verb. and question words. The pronouns tok. based on Gorkha and Surkhet dialects. These numbers resemble the numbers in Hodgson’s (1857) list: ghinga ‘two.2.’ ‘go in long strides. This is a category of the type in which basic verbs like ‘go’ are modified by ideophonic adverbs to mean things like ‘saunter. our current knowledge of [-k ~ -÷] as a pronominal plural marker. both in Indo-Aryan and Tibeto-Burman languages. she. nok. in refuting some of the fanciful notions that have been suggested for genetic relationships with Kusunda. this is now unlikely. numerals. We were unable to determine the extent to which Kusunda makes use of “ideophones” or “rhyming couplets” for modifying verbs. gina ‘he. Class II person and number marking).’ daha ‘three. We will also look at some minor word classes. Syllable final [q] also occurs in verb paradigms as a plural marker (see §5.2. The phonomenon is well represented in the region. and the [-i] is of uncertain origin. nu d. and adjective are well represented in Kusunda.’ see (89)) The first and second plural in Reinhard and Toba’s description is [to÷i] and [no÷i] respectively. however.1 4 The five pronouns are: (86) FIRST PERSON: a.’ and pangang ‘five. like adverbs. The segment [-k ~ -÷] is certainly a plural marker. 5. Given. tok SECOND PERSON: ‘I’ ‘we’ ‘you’ ‘you (plural)’ c. We will deal with them one at a time in this section with a separate sub-section for pronouns.5. noun. 44 . tsi b.1. indeed. The category needs study in Kusunda. third person singular is git and third person plural gid˙i.’ 16 15 14 The cardinal numbers resemble the numbers in Reinhard and Toba’s (1970) list. that’ f.’ etc. plurals. He cites the “final element” in to÷i and no÷i as a possible reflection of the PTB first and second person plural morpheme *-i. Plural number does not occur with third person pronouns. argues reasonably that it would make just as much sense to posit TB origins for certain morphemes. Pronouns There are five basic pronouns in Kusunda — three persons with singular number and only first and second persons with plural number. In Kamala’s speech the two plurals are [to÷] and [no÷]. it has not yet been determined if it exists at all.

c.1.1. In addition. a. Where they do.1.‘third person’ paradigm.‘second person. b. e.’ n. The “alienable–inalienable” distinction found in some languages appears to be irrelevant for Kusunda. the same pronominal roots (tsi. e. d.Himalayan Linguistics: Archive No. they are possessive prefixes. c. a. c. Possessive prefixes In first and second persons singular. d.‘first person. and tok becomes tig-.1. d. e. together with first and second person stem alternations with -i in possession. and differ from the possessive pronouns shown in (90) in much the same way that ‘mine’ differs from ‘my’ in English. 5. ni-gimi 2SPOSS-money ‘my money’ ‘your money’ 45 .1. tok dzi≥a tok da tok pya≥dza≥ tok pa≥dza≥ tok m˙nni nok dzi≥a nok da nok pya≥dza≥ nok pa≥ja≥ nok m˙nni gina dzi≥a gina da gina pya≥dza≥ gina pa≥dza≥ gina m˙nni ‘we two’ ‘we three’ ‘we four’ ‘we five’ ‘we many’ ‘you two’ ‘you three’ ‘you four’ ‘you five’ ‘you many’ ‘they two’ ‘they three’ ‘they four’ ‘they five’ ‘they many’ (88) (89) 5. e. tsi-gimi 1SPOSS-money b. for example. d. b. Following are examples: (91) a. but with vocalic modification from back to front in all roots with back vowels (the two plural roots and second person singular).’ g. Thus. and the possessive suffix -yi as evidence that Kusunda belongs to an “IndoPacific” mega-language family. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– number combinations up to five. tsi-yi tig-i ni-yi ?nig-i gina-yi ‘mine’ ‘ours’ ‘yours’ ‘yours (pl) ‘his’ < *tsi < *tok < *nu < *nok < *gina Whitehouse. without the genitive. b. nu becomes ni-. as in the following: (90) a. Following are examples: (87) a.and ni-) can occur as pronominal prefixes on nouns. b. the usual genitive suffix -yi ~ -ye (allomorph -i following consonants) attaches to the pronominal root. Possessive pronouns and the genitive Possessive pronouns are clearly related to free pronouns. c. et al (2003) take the t.

gim˙t b. ‘your stomach. such words come from the domain of body parts (and may be restricted to a general “parts” domain). The following examples show them in free variation: (94) a. qai b. for example. gina-yi gimi he-GEN money ‘his money’ First or second person singular possession can be marked by either form – the prefixal form or the genitive marked form – though there are likely pragmatic differences between the two. tig-i (93) gimi 1PPOSS-GEN money ‘our money’ a. the default word for ‘stomach’ is gim˙t and the second person possessed form. [gya-qai] ‘wind’ ‘breath’ (lit. while the word for ‘breath’ is either ata qai ‘mouth wind. ta≥ b. *gina-gimi b. ‘its water’) ‘saliva’ (lit. in Reinhard’s 1976 list. gida≥ c. without a prefix. In contrast. ge-qai. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– In first and second persons plural.2. ata gida≥ ‘water’ ‘sap’ (lit.’ which.’ (97) a. This is true also for third person gina. indeed is possible. but the preferred form is ts-im˙t. ni-gim˙t c. ts-im˙t ‘stomach’ ‘your stomach’ ‘my stomach’ Interestingly. (96) a.’ is ni-gim˙t. ni-gimi ~ ni-yi gimi n˙ti-ye who-GEN gimi money ‘my money’ ‘your money’ The genitive occurs also on question words. while we have elicited upto.is very likely the vestige of an old third person possessive prefix.Watters: Kusunda Grammar. tsi-gimi ~ tsi-yi gimi b. The word for ‘wind’ is qai. as in: (92) a.1. the possessive root cannot occur as a possessive prefix attached directly to nouns. Thus. From this we might expect tsi-gim˙t for ‘my stomach. *tig-gimi b.’ or ge-qai ‘its wind. ‘knee’ in Reinhard’s list is t-uputu with a prefix (likely meaning ‘my knee’). ‘its wind’) A word that commonly occurs in combination with other words to mean something like ‘fluid’ or ‘juice. as in: (95) 5.’ is very likely the etymon for ‘water’ with a lexicalized third person prefix.1. ‘mouth water’) 46 . ‘whose money?’ Lexicalized third person possession There are several lexical items in the language in which an initial g. ‘mouth wind’) ‘breath’ (lit. but must be accompanied by the genitive. ‘stomach’ is listed as imat. ata qai c. Typically. as in the following: (98) a.

GENITIVE ACCUSATIVE tsi tsi-yi t˙n-da tok tig-i (to÷-da) nu ni-yi n˙n-da nok nig-i (no÷-da) gina (gina-yi) gin-da –––––––––––––––––––––––––––––––––––––––– Table 2. b. n˙n. g.might be. Rana (2002) claims similarity between this word and the Kham word ge-pa≥ ‘our language. b.2. d. OBLIQUE PRONOMINAL FORMS IN KUSUNDA 5. the initial element in Kham is a first person plural prefix. n˙ti is The basic form for this word is mihaq ~ mehaq. B. language’ ‘fat’ ‘thorn’ Other candidates for a lexicalized third person prefix are the following: 5. e. i≥ gida≥ e. Interrogative pronouns The interrogative pronouns of Kusunda correspond to the so-called WH-words of English and include the following seven: (100) a. g. taken together with the genitive form. can be expressed by n˙ti. what’ ‘what. twi-yi gida≥ (99) a. ‘eye water’) ‘honey’ (lit. c. Thus.1. and third person has a slightly reduced form. 18 17 47 . why’ ‘what. gemehaq1 7 gidza≥ gidzi git˙t gip˙n1 8 gisi gitsi ‘tears’ (lit. ‘bee water’) ‘Ban Raja’ ‘body’ ‘name’ < ? *g-i-dzi ‘he says’ ‘skin’ ‘word. d.1. or n˙tßn. one for the genitive and another for the accusative. e. but it is not clear what their semantic differences are. n˙ti n˙n n˙tßn hampe ˙sa ˙si ˙saÚ i ‘who. Kusunda has two “oblique” modifications of the pronoun. c. K. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– d.1 9 Used with an animate demonstrative.3. p˙n and pa≥ bear some similarity. Oblique forms and the accusative First and second person pronouns have an oblique form that occurs with the accusative. ‘What. f.’ for example. as summarized in Table 2: –––––––––––––––––––––––––––––––––––––––– FREE PRON.Himalayan Linguistics: Archive No. all obviously related forms.’ Granted. why’ ‘where’ ‘how’ ‘how much’ ‘when’ The semantic domain of some of these interrogatives is not yet clear. while in Kusunda it is a third person singular. and it is not clear to us what the function of the prefix ge. f.

n˙tßn qæaiya kæaÚ u what danger not ‘There is no danger.1.’ c. which usually means ‘what’ can also mean ‘why. cold ‘Why won’t you drink it hot? (Then) drink it cold!’ The possessive ‘whose’ is expressed with the interrogative pronoun n˙ti. but a semantic relation between ‘how’ and ‘how much’ seems odd. all beginning with ˙s.’ or TB Kham su-z˙ male [who. n˙ti ta what this (inanimate) ‘Who is this?’ ‘What is this?’ Likewise. by ‘what.’ and with an inanimate demonstrative. n˙tßn. nowhere” series. hampe kæaÚ u where not ‘nowhere’ The finals -ti and -ßn may be old classifier morphemes and occur in other sequences like qas-ti ‘one’ and qas-ßn ‘one’. as in IA Nepali ke-hi chhaina [what-EMP is. nothing. The semantic relation between ‘how much’ and ‘when’ seems plausible. n˙ti kæaÚ i who not ‘no one’ d. although it occurs also in English.EMP is.3): (101) a. n˙tßn kæaÚ u tsim˙n-gya what not touch-OPT ‘Nothing will happen. for example. n˙ti na who this (animate) b. Negative indefinite pronouns are derived from the more basic interrogative pronouns seen in the preceding section.not] ‘no one.Watters: Kusunda Grammar. The pattern is pervasive in South Asia and often appears with an emphatic marker on the question word as well. 19 48 . its basic form is the same.’ as in the following (see §5.’ Though it occurs without an emphatic in Kusunda. as in: (103) gina k˙l˙m n˙ti-ye that pen who-GEN ‘That pen is whose?’ The forms in (100e–g) are obviously related. • Negative indefinite pronouns. uses the interrogative pronoun plus an existential negative. kæ˙≥gu qwon drink:IMP hot why 2-eat-NEG:IRR. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– translated by ‘who. plus the genitive suffix. as the following examples show: (104) a.not] ‘nothing.’ as in: (102) gæ˙run n˙tßn n-˙m-˙o. let him touch it!’ b. The “nobody.

(See also §5. Plurality of referents can be marked using numerals or quantifiers. 5. as in the usual “proximate” and “distal” distinctions in both IA and TB languages of Nepal.Himalayan Linguistics: Archive No. make’ to carry the appropriate inflection. using demonstrative pronouns.’ More than half of the verbs in Kusunda are formed from such elements. This sets them apart from pronouns. have the same morphosyntactic properties.” They require the presence of an auxiliary verb ‘to do. however.1.1. as in the following (see also example (89)): 49 .4. to distinguish between ‘this’ and ‘that. Another class of words that might be called “nominal” (though not so in this grammar) are “action nouns. na being an animate demonstrative. We are not aware of a way.1. still weak. It is possible that na is the functionally unmarked member of the two. which we saw in §5. There is no nominal plural marker. Demonstrative pronouns As far as we can tell. as in na pæurlu≥ tsu ‘This is red’ to refer to an inanimate object.2.1. the distinction appears to be based on animacy. and ta an inanimate demonstrative.2. Though they can be counted. Rather. as far as we currently know. n˙ti ta what this (inanimate) ‘Who is this?’ ‘What is this?’ Unfortunately.) 5.2. More investigation needs to be done on their properties. Here we repeat the examples in (101): (105) a. this generalization breaks down in many contexts. concrete kinds of things that can be manipulated and counted.3 we give some evidence.) In §5.2. n˙ti na who this (animate) b. for a possible “locational” subclass of nouns.4. The class is composed of prototypical nouns. (See §6.’ However. there are only two demonstrative pronouns in Kusunda – na and ta. Number marking There is no number marking for nouns in Kusunda. Rather than referring to them as “nouns” in light verb constructions. none of them can be pluralized. Nouns and noun morphology All nouns in Kusunda. and the two terms seem to be interchangeable. we will refer to them as “non-inflecting verbs.1. can be marked in the NP by numerals or quantifiers. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 5. like plurality in nouns. The distinction does not appear to be deictic. Plurality in pronouns. the third person pronoun sometimes occurs as a demonstrative. or by plural agreement affixes in the verb.’ There are no plural demonstrative pronouns. and in that function appears to express a distal demonstrative ‘that’: (106) gina k˙l˙m gina-yi that pen he-GEN ‘That pen is his.2. or by plural agreement affixes in the verb.” This is a highly restricted class and occurs only in constructions as a complement of the verb ‘to do’ – notions like ‘to do thinking’ or ‘to do hitting.

twi-ye bee-GEN ag˙i dog gida≥ water ‘Ram’s dog’ ‘bee’s honey’ There are numerous other cases in which the possessive relationship is not marked at all.’ We have already seen some case marking in pronouns – the oblique forms used for the genitive and the accusative. ata mouth b. nor are there oblique forms for pronouns other than in the genitive and in the accusative.1. First or second person possession is marked primarily by possessive prefixes. the underlying shape of the pronoun was also modified (see Table 2).’ ‘They ate. gina da g-˙m-da-n he three 3-eat-PL-REAL d. The genitive The genitive suffix -yi/-ye (allomorph -i/-e following consonants) occurs primarily on possessive pronouns (see §5. Following are examples of third person possession: (108) a. the first is marked by the genitive. the semantically similar “body part relationships” are not generally marked. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (107) a. the second two are not. As far as we know. gina g-˙m-˙n 3-eat-REAL he b.2.2. m˙¯¯i many 5. especially in “part–whole” constructions. Case marking g-˙m-da-n 3-eat-PL-REAL ‘He ate.2. gina g-˙m-da-n 3-eat-PL-REAL he c.2. This modification presumably reflects an older case marking system.1) or on third person possessors in a possessor–possessee relationship. 5. Thus. though ‘honey’ is marked for possession in (108b). numba cow amba meat amba meat 50 ‘goat meat’ ‘cow meat’ . as in: (109) a. We do not know if the choice is lexically determined or determined by pragmatic notions like “degree or level of semantic union.’ ‘Many ate. aidzi-gi 2 0 goat-GEN b. i≥ eye gida≥ water ‘saliva’ ‘tears’ gida≥ water What might be termed as “type relationships” are sometimes marked by the genitive.’ ‘They three ate. In both cases. in addition to being marked by suffixes.” In the following examples. sometimes not. (110) a.Watters: Kusunda Grammar. ram-e Ram-GEN b. there is no oblique modification for noun stems in Kusunda.1.

and the accusative is marked by the nominal suffix -da. The -gi here seems to point to the same etymon. Following are examples of nominative and accusative marking in intransitive and transitive clauses: (113) a. Nor is there a marking distinction between third and non-third persons in transitive clauses (as in “person-based” split ergatives).Himalayan Linguistics: Archive No. and under what conditions as -gi. Without further study we do not know under what conditions the genitive surfaces and -yi. as in (112b): (112) a.2. pyana tsi-Ø n˙n-da imba-d-i yesterday I-NOM you-ACC think-1-PAST ‘Yesterday I thought about you. it appears that the genitive is not used to mark other semantic relationships between nouns. duga ab˙q ground vegetable ‘stone house’ ‘mushroom’ The genitive is also used in first or second person possession where a modifier occurs between the possessor and the possessed item.’ b.’ c. ye≥gu 1S-GEN stone wi house ‘my house’ ‘my stone house’ The nominative and the accusative Situated in a sea of ergative marking languages. All of these ergative or ergative-variant systems occur in the languages of the region. Kusunda has no ergative of any kind. which has been related by some to Tibetan -ki (DeLancey 1984). The nominative argument in Kusunda is zero marked (unmarked). tsi-yi 5.2. gina-Ø u-g-˙n he-NOM come-3-REAL ‘He came. In Kusunda. Kusunda. ye≥gu wi stone house b. like “composed of” or “location of.2. as far as we can tell. there is no distinction between transitive and intransitive subjects – both are zero marked.” as shown in the following examples: (111) a. tsi-wi 1S-house b. with its nominative-accusative case marking system.’ The Kusunda genitive -yi/-y e looks a lot like the Kham genitive -yi/-y e. but. qotu bird getse offspring ‘chick’ From the data we have so far. or “aspectually-split” systems). tsi-Ø limi-t-ßn I-NOM dance-1-REAL ‘I danced. stands out as a typological island. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– c. 20 51 . or “stative-active” systems. or between “more transitive” and “less transitive” clauses (as in “volitionality” systems.

Ø tsi-h~ukui-da ip-da-d-i 1S-small-ACC sleep-CAUS-1-PAST ‘(I) put my little one to sleep. like many languages of the region. marks the “terminus” of a causal chain.” a well established pattern both in Indo-Aryan and Tibeto-Burman languages of the region (with the exception of Kham. The grammatical category. and -da marks the animate “primary object. the same morpheme that marks accusative also marks “dative recipient” in bitransitive clauses and the subject in “dative subject” clauses. but it is clear that animacy plays a big role.’ NOM In fact.” Following are examples of both: (114) DATIVE RECIPIENT: a.” a notional indirect object (Dryer 1986).’ Note that the verb form in (114a) is identical to the form in (113d) even though one functions in a transitive clause and the other in a bitransitive clause. ‘hunger is not to me’) c. as in: (115) a. tsi qotu-da ho-wa-d-i I bird-ACC fly-CAUS-1-PAST ‘I made the bird fly. What may be regarded as the same morpheme -da also marks embedded clauses of purpose (glossed in the examples 52 . -da marks the “dative subject. (114a) adds a “secondary object” letter.’ • The object marker -da as a ‘purposive’ marker. tsi wi a-t-ßn I house make-1-REAL ‘I built a house’ / ‘I make houses. then. t˙n-da ida≥ kæaÚ u I-DAT hunger is:not ‘I am not hungry.’ DATIVE SUBJECT: b. Reinhard and Toba show this structure without a dative subject.’ e. We have numerous cases of inanimate objects being unmarked. gina ta≥ kola-g-i he water draw-3-PAST ‘He drew water. In (114b–c).Watters: Kusunda Grammar.’ (lit. t˙n-da tsya ts-oˆ -˙n ola≥ ota≥ I-DAT tea 1-drink-REAL tasty seems ‘To me drinking tea is tasty.’ f. gina-Ø t˙n-da haq t˙mb˙-g-˙n he-NOM I-DAT letter send-3-REAL ‘He sent me a letter. It is not known what semantic or pragmatic parameters delimit the use of -da in transitive clauses.’ b. gina-Ø t˙n-da t˙mb˙-g-˙n he-NOM I-ACC send-3-REAL ‘He sent me. one of Kusunda’s immediate neighbors). Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– d. and might be better termed an “endpoint.

3. goraq tsi da≥ t -a -an tomorrow I Dang 1-go-IRR ‘Tomorrow I will go to Dang.Himalayan Linguistics: Archive No. It is likely that the marker is more attuned to the pragmatic notion of “endpoint” or “goal” than it is to the syntactic notion of transitive “object.2. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– as PURP). Very likely it is a kind of noun. dzæola g-o -on bag 3-put-REAL ‘He put it in the bag.2).’ • The locative -da. The so-called “preposition” inside. we present the following examples: (116) a.2. and will be treated more fully in §8. The marker occurs rarely. for example. marks at least one other kind of relationship. only three locative case markers have emerged. 5. This may constitute some evidence for a separate “locational” subclass of nouns (see §5. though some of them are suspect. In the meantime. depending on its usage. the “endpoint” of the eventuality named by the matrix verb is the semantic notion being marked. DAT.2. Local case markers In our data set. Here too. but seems to occur almost obligatorily on the word for ‘road’2 2 and also commonly on the word for ‘forest. tsi wi ˙n-da ol~aˆ -t-ßn I house make-PURP want-1-REAL ‘I want to build a house. of course.’ From the example in (116a) it is clear that the matrix verb need not be transitive to warrant the felicitous use of -da.’ b. Reinhard and Toba list more. given as gem˙t.3. especially if the semantic connection is sufficiently understood by the collocation of the verb and the locational noun. discussed in §5.’ 22 21 53 .2. The embedded verb retains many of its finite features.” Throughout this grammar we will continue to gloss -da as ACC.3. (117) a. is very likely the word for ‘stomach. a locative one.2. The marker -da. or PURP.’ b. serve as a ‘relator noun’ for inside in the same way that the word for buttocks in our data serves as a relator noun for behind.2 as an accusative. t-˙m-da t-ug-un 1-eat-PURP 1-come-REAL ‘I came to eat. See §5. un-da myaq p˙rm˙-d-i qæai-tsi-n road-LOC leopard meet-1-PAST afraid-1-REAL ‘I met a leopard on the road and was frightened.’ The word for stomach can.’2 1 None of the locative case markers appear to be obligatory.’ (118) a. It appears possible that the frequent combination un ‘road’ and -da ‘locative’ may be the source for the verb ‘to show’ unda a-d-i [show do-1-PAST] ‘I showed it.

’ c. specific locations.’ c. tsya-˙na ¯~a-˙go tea-from extract-IMP ‘Remove it from the tea. wi-g˙ dzu≥-dzi house-IN hang-3:PAST ‘It’s hanging from the house. Recall that -da is obligatory on most occurrences of ‘road. diffuse locations. This is probably the concatenation of several morphemes. gag˙ri-ga ta≥ o -to water put-IMP jug-IN ‘Put water in the jug!’ d. marked by the suffix -˙na. ha o yi-g˙ tsi monkey tree-IN is ‘The monkey is in the tree.’ Notice that in (118b).Watters: Kusunda Grammar. while -ga marks nearby. -˙na. indicates direction ‘away from’ or ‘out of. the final one of which. The locative marker -ga appears to mark a nearby. -da occurs twice. wi-˙na ˙g-a house-ABL come-IMP ‘Come out of the house!’ b. • The locative -ga/-g˙.’ as in the following: (120) a. as in the following: (119) a. The ablative. ao-g˙ l˙b˙-˙-go dirt-IN bury-do-IMP ‘Bury it in the dirt!’ • The ablative.’ e.’ d.’ b. un-da ip-dzi road-LOC sleep-3:PAST ‘He slept on the road (on the way). wi ˙-g-i tæ˙w˙-g˙ ip˙n u-g-i house make-3-PAST place-IN corn come-3-PAST ‘In the place where he built a house corn came up. In its second occurrence it marks what appears to be an inanimate object – ‘road.’ Comparing locative -da with locative -ga (following). specific location. and will be glossed IN.’ In fact. tsi-gimi gila≥-da me-˙-g-˙n 1S-money forest-LOC lose-do-3-REAL ‘He lost my money in the forest. the distinction may be that -da marks distal. cor54 . gin-da t˙mb˙-˙-go un-da miÚ-d-wa he-ACC send-do-IMP road-LOC lose-REAL-NEG ‘Send him.’ Reinhard and Toba (1970) list the word for ‘from’ as aa÷piaadana [a÷piad˙n˙]. we are interpreting it as a locative marker. he won’t lose the way. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.

it collocates with the locational copula tsi. In “reciprocal” kinds of verbs.’ (124) tsi-ma qasti tsi/tsu I-WITH one is ‘I have one.’ (lit. as a destination or location. The allative appears to be no longer productive. morphologically. As a possessive. (121) a. As a location. The form given in (121) for house.” and the “instrumental. or doing an action with another actant.’ • The allative.’ As in most other languages of the region. as in: (125) tsi gina-ma limu-tsi-n fight-1-REAL I he-WITH ‘I fought with him. wa. wa u-g-i house:in come-3-PAST ‘He came home. the comitative also marks possession.’ There is at least one bitransitive verb in Kusunda in which -ma marks the “source. wa ts-o ˙-d-ßn do-1-REAL house:in 1-put ‘I put it in the house. tsi wa sip-tsi-n I house:in enter-1-REAL ‘I entered into the house. it collocates with the existential copula tsu.2. as in ‘With me is one rupee’ meaning ‘I have one rupee. This contrasts with the nominative form wi.” the “benefactive. but occurs also in the following: (122) a~o-a sip-tsi-n inside-ALLT enter-1-REAL ‘I entered (to) inside. ‘with me is one’) The language is ambivalent in how it interprets -ma.4. Reinhard and Toba (1970) list ‘to the house’ as w˙haa. They are: the “comitative. behave like intransitives.” 55 .” the “comparative.’ c.2. may include the vestige of an old allative marker -a. has its own special suppletive form which incorporates location – wa. as in the following: (123) gina-ma tsi ts-˙g-˙n 1-go-REAL he-WITH I ‘I went with him.Himalayan Linguistics: Archive No. Here we will treat a few nominal suffixes that do not neatly fall into the grammatical or locative class of markers. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– responds with our data. Such verbs.’ b.” • The comitative. The comitative marks a notion of association. -ma marks the reciprocal actant. House.’ Other case markers 5.

Following is an example of the lack of an instrumental: (129) na kæurpa-Ø kis˙-g-˙n this knife cut-3-REAL ‘This is cut by a knife. Reinhard and Toba list a morpheme -ya as the instrumental.’ s˙mba derived from s˙mba ‘buttocks. ‘with me’) • The “benefactive”. Watters 2002). All attempts have yielded nothing. for example. their nominal source is transparent. see §10. the word ceases to function as a full noun.’ etc. Here. Locative postpositions Most locative postpositions are specialized nouns that specify some kind of location – notions like ‘on top of.3.’ • Comparative. Thus. The comparative in Kusunda is based on the Nepali bæ˙nda. For details on the latter kind of benefactive construction. meaning ‘after’ or ‘later. Some have referred to such nouns as “relator nouns” (Starosta 1985.’ (See §8.2. too.’ as in: (126) tsi-ma gimi ai-dzi d-˙g-˙i I-WITH money beg-3:PAST 3-go-PAST ‘He begged money from me and left. It seems strange that we have been unable to elicit an instrumental marker. literal sense.) The word has a temporal function. What we refer to here as the “benefactive” is a case marker that occurs on NPs without concomitant valence adjusting modifications on the verb.’ / ‘(He) cut this with a knife. In other cases.’ (See fn. as in the following: (128) na bæ˙nda ~oˆ ˙n o -da-n this than big come-PL-REAL ‘Bigger than this come.’ (lit. a~o-˙na b~adza ˙g-a inside-ABL outside come-IMP ‘Come out from the inside!’ 56 .2. we can get: (130) a. Thus.) 5.2.’ ‘beside.Watters: Kusunda Grammar.3.’ (in reference to mosquitos) • Instrumental. as in the following: (127) tsi-lage limu-dzi I-FOR fight-3:PAST ‘He fought for me. A transparent case of a noun serving as a postposition is the word for ‘behind.1 for the “neutral” conjugation of this verb – something that yields an intransitive reading. a~o ‘inside’ and b~adza ‘outside’ can take the ablative in the same way that ‘house’ can take the ablative. 21. the benefactive case marker is based on the Nepali ko lagi. and serves only in a locative function.’ That many such words are basically nouns can be shown by the fact that they are capable of supporting the genitive or the locative case marking suffixes shown in §5. Where these nouns still function elsewhere in the grammar as full nouns with their original. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– The verb is ‘beg.

There may be a three-way deictic distinction – proximate. distal. under’ ‘up.’ Again.’ and sape ‘to here’ may have formed a paradigm at one time in the language. t˙isa b. paÚtse ‘down. below. under’ ‘down. The forms ape ‘there. c. akp˙i b. The words are: (132) (133) a.2) The last word. above’ Another three locational nouns are the following: (134) ‘across’ ‘uphill’ ‘downhill’ • Deictic pronouns/locatives.’ and another two for ‘up. ape c. d. Likewise. above’ ‘up. one for static location and another for “direction to. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. but we have been unable to verify this. (131d).’ We have two lexemes for ‘down. asne c. below. tape b. but in the modern language the forms are semantically disparate. are the following: (131) a. wi-yi b~adza house-GEN outside ‘outside the house’ / ‘the house’s outside’ Other words that are probably nouns. though we do not have enough data to prove it. aˆ dze ~ a. sape ‘to here’ ‘to here’ ‘to there’ 57 .” The three forms in each set do not form regular paradigms. n~oˆ dze b. above’ (see the spectogram under §4. ama a a. remote – in the Kusunda deictic locatives.’ as in pinda duksi ‘the first/eldest son. the individual members being suppletive. below. t˙wa c. aqa b. gidzaq kampya pinda a k˙ ‘between’ ‘to one side’ ‘in front of’ ‘on top of. we are dealing with a noun that has become grammaticalized as a locational postposition. There are two sets. also occurs in our elicitation lists as the noun for ‘sky.’ We have nothing to indicate what the difference might be. or above. The modern paradigms are as follows (it is not known how (135b) and (135c) differ): (135) STATIC: a. pinda in (131c) can be used to mean ‘first. or under.3.Himalayan Linguistics: Archive No. b.’ tape ‘to there. nupa ‘here’ ‘there’ ‘there’ The locatives in (135) occur with verbs like ‘sit’ or ‘sleep’ in constructions like ‘Sit here!’ or ‘Sleep there!’ (136) DYNAMIC: a.

4 for derived Adjectivals. black.” etc. and green.’ for example. The Kusunda word for tumeric is k˙p˙≥. but unlike Tibeto-Burman. h~ukun suta bæ˙ra-q-ßn small thread break-AC-REAL ‘The small thread broke. precede the noun they modify. That is. that the class of adjectives in Kusunda matches the class of adjectives in any other language. This does not mean. but only as secondary innovations. isna ‘to there’ The locatives in (136) occur with verbs that include motion – like ‘Come here!’ or ‘Go there!’ or even ‘Give it here!’ We have examples of both tape and t˙wa occurring with the command ‘Come here!’ 5. nu ta kam ˙n wey˙n kæaÚ u you this work do good not ‘It’s not good for you to do this work. red. Following are examples: (137) MODIFICATION: a. gina-ma limu-tsi-n I strong am.1. white. its negative counterpart).’ We spent very little time studying adjectives. as in some of the TB languages of the region.2 3 is based on the color of tumeric powder. One thing that did catch our attention was the paucity of basic color terms – black. (See also §9.” “shape. and red are the colors predicted by Berlin and Kay’s (1969) study for a three-term system. na amba ola≥ tsu this meat tasty is ‘This meat is tasty.) Adjectives in Kusunda.’ b. of course. when occurring in a modifying function. and in a predicative function they occur with the existential copula tsu (or kæaÚ u.4. is a verb in Kusunda. that occur as verbs in Kusunda.Watters: Kusunda Grammar. Yellow. white.” “color. Even the simplest TB systems have a four-term system – black. tsi gya ai tsu. the property notions of “size. form an inherent class of words not derived from other word classes (as with the class of stative verbs in TB). The 58 .’ (138) PREDICATION: a. ‘Fat. he-WITH fight-1-REAL ‘I am strong. Still. and red. huigin ta≥ pokæla-yin dirty water bathe-PROH ‘Don’t bathe in dirty water!’ b. I fought with him.. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– d. for example. white. Yellow and green are also possible in Kusunda. but an adjective in English. and may encounter surprises in a later study. There are several words that occur as adjectives in English. Kusunda has a distinct “adjective” class. Adjectives Like Indo-Aryan. and ‘yellow’ is k˙p˙≥ b˙yo.’ c.’ c. tsi ˙gi tu ts-~aˆ -dzi I live snake 1-see-PAST ‘I saw a live snake.

Verbs and verb morphology Verbs in Kusunda are morphologically complex with a fair bit of morphological irregularity reflecting varying layers of grammaticalization.’ It is impossible to speculate on the source of such verb roots and whether they were inflecting at an earlier stage or not. Prefixing patterns (Class I) Verbs are marked for the person (first. Most verbs. too. 5. may mean something like ‘har-like’ (from Nepali ‘green’) and ‘kæ˙r-like’ (from Kham ‘reddish-brown’).5.Himalayan Linguistics: Archive No. they produced the word h˙rgun. In both IA and TB.1 on Adnominals and relative clauses. and a newer system which adds a past distinction to realis events.’ The words h˙r-gun and kæ˙r-gun. Person and number are separate indices. or having developed on an analogy with. the word for tumeric is borrowed from Nepali h˙lid˙. a prefixing pattern and a suffixing pattern. “light verb” constructions – constructions like ‘to do hunting. Person marking in verbs Person agreement patterns in the Kusunda verb are of two basic types. Class I verbs are defined as those in which person is marked by prefixes and number is marked by suffixes.’ ‘eat. For a treatment of these.5. The suffixing pattern is modeled closely on the conjugational pattern of the verb ‘to do.’ and has every appearance of being. claimed that Kusunda had no word for ‘green.’ ‘to do breaking. Person agreement marking can be prefixing or suffixing. and where the patterns are suffixing they interact with the tense-aspect-modality (TAM) system in complex and unpredictable ways. 5. The Kusunda word for ‘bad’ stands out as unique in the languages of the region.’ 23 59 . which appears to be related to Nepali h˙riyo. ‘bad’ qolom is a separate lexical item from ‘good’ wey˙n.1. see §9. In Kusunda. second. The words for ‘black’ and ‘white’ respectively are soksodi ~ soksog˙r˙m and kasige.1. The underlying syllable shape of the verb root appears to have at least some bearing In Kham.’ and ‘sleep’ – and appears to be the older of the two patterns.’ or ‘to do thinking.5. for several weeks. then. depending on the verb.’ Eventually. TAM systems. are of at least two types – an older system which marks a basic distinction between realis and irrealis events. make. There is no object agreement. too. and ‘yellow’ is built off that word – h˙lidyaso ‘tumeric-like. can be used in a modifying function. The likelihood of this being true is more convincing in the light of a second word for ‘red’ – kæ˙rgun – based on a Kham corruption of the Nepali word for ‘rust. for example. ‘bad’ is commonly expressed only as NEG-good. Our informants. 5.’ In Kham the word for ‘reddish-brown’ is kæ˙iryaso ‘rust-like.” all inflection being carried by the auxiliary ‘to do.1. The prefixing pattern occurs with a limited set of “high frequency” verbs – verbs like ‘come.’ but we have found it in no other contexts.’ More than half the verbs in Kusunda take this form and we will present them as “non-inflecting verbs.’ ‘go. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– word b˙yo is similar to the Nepali bæ˙yo ‘became. or third) and number (singular or plural) of the subject argument.

’ and ‘go’ in realis mode: (141) EAT SLEEP COME GO -------------------------------------------------------------------------------------1s t-˙m-˙n ts-ip-ßn t-ug-un ts-˙g-˙n 1p t-˙m-da-n ts-ip-da-n t-ug-da-n ts-i÷-da-n -------------------------------------------------------------------------------------2s n-˙m-˙n n-ip-ßn n-ug-un n-˙g-˙n 2p n-˙m-da-n n-ip-da-n n-ug-da-n n-i÷-da-n -------------------------------------------------------------------------------------3s g-˙m-˙n g-ip-ßn ug-˙n / ugi d-˙g-˙n / d-˙g-˙i 3p g-˙m-da-n g-ip-da-n u-d˙i d-˙g-˙i -------------------------------------------------------------------------------------The verbs ‘eat’ and ‘sleep’ are the most regular of the four verbs shown in (141) and will serve as our starting point. ‘go. with many irregular and suppletive forms.’ etc. The prefixing pattern occurs primarily. The forms are clearly based on the shape of the free pronouns tsi ‘I. Following are examples of all persons and all numbers for ‘eat. or otherwise. as in first person ts. Other Class I verbs with first person ts. grammatical. and third person by g-. In the latter case. the determination of third person g. n-.’ The determination of first person ts.’ etc. respectively. Other Class I verbs with first person t.’ nu ‘you. n-.vs. With the third verb.possibly fills a person slot in the synchronic grammar. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– on affixation type. Other Class I verbs with third person Ø.’ and gina ‘he. Ø is lexical and cannot be determined by rule.are the following: ts-o ˆ-˙n ‘I drank. or gaccording to whether the actant is first.’ t-uk-an ‘I knew. or zero. it.vs.is purely lexical and cannot be determined by rule – phonological. n-.or t. The first element. Many others take g-.’ the forms are ts-. first person agreement is signaled by ts. ‘come.’ ts-o -on ‘I put. second person by n. can be discounted on historical grounds. Again. is t-/ts-.’ uk-an-dzi ‘he knows.’ the pronominal agreement forms are t-.in the same slot. on verb roots that begin in a vowel.are the following: o -˙n ‘he died.as a third person marker. historically it is the first element of the root and still surfaces in certain forms like the imperative – da ‘Go!’ D. Class I verbs are generally messy. second. Plural number is indicated by a suffix -da immediately following the root. and with the final verb.Watters: Kusunda Grammar. make’ In general.are the following: t-o -˙n ‘I died. then.in the first prefix slot. and (seemingly) d-. though d. but not exclusively.’ ‘come. the (a) forms in (139) and (140) illustrate the rule. or third person. t. she. t-. a prefix. This is very likely a consequence of their considerable age and maintained by high frequency 60 . or Ø.’ ‘sleep. d-.’ etc. Thus. and the suffixing pattern occurs primarily on verb roots that begin in a consonant. and the (b) forms illustrate the exception: (139) PREFIXING PATTERN: BASE FORM FIRST PERSON SECOND PERSON GLOSS a) am b) da (140) t-˙m˙n ts-˙g˙n n-˙m˙n n-˙g˙n SECOND PERSON ‘eat’ ‘go’ GLOSS SUFFIXING PATTERN: BASE FORM FIRST PERSON a) pumb˙ b) a pumb˙ a-d-i a-d-i pumb˙ a-n-i a-n-i ‘beat’ ‘do.vs. however.

and third person h-.Himalayan Linguistics: Archive No. A case in point are the plural forms for ‘go’ – ts-i÷-da-n ‘I went’ and n-i÷-da-n ‘You went. second person ¯-.’ ‘We sit. meaning ‘fetch’ is often interchangeable with ‘go’ and is likely the source for this form.e. and third person -g. 24 A root i-. as in: h~a ˆ-no ‘Sit!’ 5. Class II verbs are characterized by person marking suffixes. the difference being that in Class II they are suffixed while in Class I they are prefixed. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– usage.’ ‘You [pl] made.’ ‘You made.’ ‘You [pl] sit. (143) 1s 1p 2s 2p 3s a-t-ßn a-d-˙-n a-n-ßn a-n-˙-n ˙-g-˙n [make-1-REAL] [make-1-PL-REAL] [make-2-REAL] [make-2-PL-REAL] [make-3-REAL] ‘I made.’ As in Class I.2.2 4 Another irregular verb is the verb ‘sit.’ The vocalism [i] in the plural root [i÷] cannot be explained in the synchronic grammar and nothing of the original root da remains. Suffixing patterns (Class II) In contrast to Class I verbs. For Class II transitive verbs.’ It is possible that the underlying root in this case has an /h/ onset.’ ‘They sit. The /h/ emerges also in the imperative.1. the pronominal elements here are first person -t/-d. as in the following: (142) 1s 1p 2s 2p 3s 3p s-~aˆ an s-~aˆ an ¯-~aˆ an ¯-~aˆ an h-~aÚdzi h-~aÚdei ‘I sit.2. as in the following: (144) 1s 1p 2s 2p 3s 3p BUY CARRY CUT dza-a-t-ßn dza-a-d-˙n dza-a-n-ßn dza-a-n-˙n dza-˙-g-˙n dza-˙-g-˙n abi-a-t-ßn abi-a-d-˙n abi-a-n-ßn abi-a-n-˙n abi-˙-g-˙n abi-˙-g-˙n kis-a-t-ßn kis-a-d-˙n kis-a-n-ßn kis-a-n-˙n kis-˙-g-˙n kis-˙-g-˙n In the pronunciation of these verbs. 61 .4.’ ‘We made. We end up with a suppletive form. to make.4). • Transitive patterns. there are distinct syllabic peaks on each of the vowels – one on the final vowel of the root (as in dza ‘buy’) and the second on the vowel which belongs to the onset of the verb ‘to make’ (as in a-t-ßn).’ which is given in (143). the agreement patterns appear to be modeled on the verb ‘to do.’ ‘He/she made.5. second person -n. Numerous transitive verbs are patterned on this paradigm. it forms its own syllable without the aid of vowels (see the discussion on Syllabic ‘n’ under §4.’ with first person s-.’ ‘You sit. The -ßn symbol is a syllabic ‘n’ – i.’ ‘He/she sits. The majority of verbs in Kusunda belong to this class.

’ ‘I dressed (someone).’ • Transitive–intransitive distinction.’ Such conjugations are developing yet another set of pronominal markers. very nearly a complete copy of the free pronoun. they are ambitransitive. as in: tsi sip-ßn-tsi ‘I entered. as illustrated in the following: (146) a. but here we will treat the transitive–intransitive distinction which is marked by affixation type. such verbs occur in “neutral” conjugations. the same verb root is used in both a transitive and an intransitive sense.’ and nu sip-ßn-nu ‘You entered. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– • Intransitive patterns.’ 62 . intranstive reading of some verbs depends solely on whether the affixation pattern used is a transitive pattern or an intransitive pattern. INTRANSITIVE bem-tsi-n fall-1-REAL b. Most inherent intransitive verbs fall into this class. the transitive pattern is fashioned after the verb ‘to do. Following are examples: (145) 1s 2s ENTER DANCE DESCEND FALL sip-tsi-n sip-n-in limi-tsi-n limi-n-in b˙l-tsi-n b˙l-n-in pæep-tsi-n pæep-n-in Very commonly. to make.1). As mentioned earlier.’ and may.’ Thus.’ ‘I entered.’ ‘I entered. There are several transitivizing devices in Kusunda (see §7. In a sense. TRANSITIVE bem-˙-t-ßn fall-do-1-REAL ‘I dressed. The correct reading (transitive or intransitive) depends on the type of affixation. TRANSITIVE tul-a-t-ßn dress-do-1-REAL (148) a. tsi I b. tsi b˙l-ßn ‘I descended.’ ‘I fell. tacking them on at the end. tsi I sip-ßn enter-REAL sip-tsi-n enter-1-REAL sip-ßn-tsi enter-REAL-1 ‘I entered. That is. bereft of person markers. as in: tsi sip-ßn ‘I entered.Watters: Kusunda Grammar. we get considerable variation for intransitive person markers. A slightly different suffixing pattern emerges for Class II intransitive verbs. INTRANSITIVE tul-tsi-n dress-1-REAL b.’ nu b˙l-ßn ‘You descended. in fact. tsi I c. The pronominal marker is more complete.’ nu sip-ßn ‘You entered’. be a verb concatenation (as we treat it in the following glosses): (147) a. the transitive vs.’ ‘I made (someone) fall. then. In the following examples.

and past tense. Though first person -t. t-˙m-˙n 1-eat-SG:REAL b. t-˙m-du 1-eat-SG:IRR ‘I will eat. as shown in the following contrastive pairs: (150) FIRST PERSON a. person marking is prefixed and number marking is suffixed. and the plural realis morpheme -da-n is replaced by -da-k. the combinatory patterns that emerge are slightly different for realis. 5.’ ‘They ate. the singular realis morpheme -˙n is replaced by -du. Class I number marking In Class I verbs. both systems are suffixed. 5. we have illustrated person marking patterns only in realis mode.3 on TAM marking. g-˙m-˙n 3-eat-SG:REAL b. and third person -g are constant across all TAM categories.’ 63 . irrealis. causing the two systems to intersect. keeping the two systems separate. In Class II verbs.5. n-˙m-˙n 2-eat-SG:REAL b.5.’ a. t-˙m-da-n 1-eat-PL-REAL (151) SECOND PERSON ‘I ate.Himalayan Linguistics: Archive No.1. Number marking Number marking in Class I and Class II verbs is basically the same.2. INTRANSITIVE hyoq-tsi-n hide-1-REAL b. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (149) a.’ a. second person -n. n-˙m-da-n 2-eat-PL-REAL (152) THIRD PERSON ‘You ate.’ For irrealis mode. TRANSITIVE hyo -a-t-ßn hide-do-1-REAL ‘I hid.2. as in the following: (153) FIRST PERSON a. with slight and definable variation.’ ‘You [pl] ate. The variation is due to the fact that in Class I verbs. There is also an irrealis mode.5.’ Until now.’ ‘I hid (something). realis plural (S/A only) is marked by -da while singular goes unmarked. g-˙m-da-n 3-eat-PL-REAL ‘He/she ate. These differences will be discussed in §5. and for all transitive verbs a past tense.’ ‘We ate.

pumba-n-u beat-2-SG:IRR b.’ Here.’ ‘You [pl] will beat (someone).’ a.2. Class II person and number marking In Class II irrealis mode. the Class I and Class II forms are composed of different morpheme sequences (clearly as a consequence of reanalysis due to person suffixation in Class II – see also summary in (166–167)): (156) Class I Class II SINGULAR -du (SG:IRR) -d-u (1-SG:IRR) PLURAL -da-k (PL-IRR) -d-˙k (1-PL:IRR) The Class II -d-u / -d-˙k forms are in paradigmatic opposition to second person -n-u / -n-˙k and third person -g-u / -g-˙k.’ ‘They will eat. as in: (157) FIRST PERSON a. pumba-d-˙k beat-1-PL:IRR (158) SECOND PERSON ‘I will beat (someone).2. g-˙m-du 3-eat-SG:IRR b. g-˙m-da-k 3-eat-PL-IRR ‘He/she will eat. pumba-d-u beat-1-SG:IRR b.’ 64 .’ in which plural -da remains invariant and realis -n changes to irrealis -k. however. realis t-˙m-da-n ‘We ate’ contrasts with irrealis t-˙m-da-k ‘We will eat. In fact.’ ‘We will beat (someone). singular is marked by zero (or unmarked) and plural is marked by -da. t-˙m-da-k 1-eat-PL-IRR (154) SECOND PERSON ‘We will eat. the sequences -d-u (157a) and -d-˙k (157b) may appear superficially to be equivalent to the -du (153a–155a) and -da-k (153b–155b) of Class I verbs.’ a.’ ‘You [pl] will eat. at least in the synchronic system. Nevertheless. pumba-g-u beat-3-SG:IRR ‘He/she will beat (someone). in the synchronic system. There is some evidence to suggest that the plural function was originally marked by -k.Watters: Kusunda Grammar.’ a. too. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. 5. n-˙m-da-k 2-eat-PL-IRR (155) THIRD PERSON ‘You will eat. pumba-n-˙k beat-2-PL:IRR (159) THIRD PERSON ‘You will beat (someone).5. much as it is in free pronouns. n-˙m-du 2-eat-SG:IRR b.’ a.

Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. pumba-g-˙k beat-3-PL:IRR ‘They will beat (someone).’

Likewise, the realis forms also distinguish between -d ‘first person,’ -n ‘second person,’ and -g ‘third person,’ as in: (160)

a. pumba-d-ßn beat-1-REAL b. pumba-d-˙n beat-1-PL:REAL] (161)

‘I beat (someone).’ ‘We beat (someone).’

a. pumba-n-ßn beat-2-REAL b. pumba-n-˙n beat-2-PL:REAL (162)

‘You beat (someone).’ ‘You [pl] beat (someone).’

pumb˙-g-˙n beat-3-REAL

‘He/she/they beat (someone).’

Class II verbs also have a past tense. As is true for irrealis and realis modes, so too in past tense the suffixes -d, -n, and -g mark first, second, and third person, respectively. Following is the paradigm: (163)

a. pumba-d-i beat-1-SG:PAST b. pumba-d-ei beat-1-PL:PAST (164)

‘I beat (someone).’ ‘We beat (someone).’

a. pumba-n-i beat-2-SG:PAST b. pumba-n-ei beat-2-PL:PAST (165)

‘You beat (someone).’ ‘You [pl] beat (someone).’

a. pumba-(g)-i beat-3-SG:PAST b. pumba-(g)-ei beat-3-PL:PAST

‘He/she beat (someone).’ ‘They beat (someone).’

Thus, in Class I verbs, where person markers are prefixed, the sequences -da and -du are indivisible, single morphemes – -da functions as a plural marker and -du as a singular irrealis marker. In Class II verbs, -da and -du are forced to coalesce with suffixed person markers -d, -n, and -g, resulting in a reinterpretation of the data. The -du of Class I becomes the -d-u, -n-u, and -g-u of Class II; the -da-n of Class I becomes the -d-˙n, -n-˙n, and -g-˙n of Class II; and, finally, the -da-k of Class I becomes the -d-˙k, -n-˙k, and -g-˙k of Class II. 65

Watters: Kusunda Grammar; Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– In summary, we can identify the following person, number, and mode affix patterns for the two classes of verb: (166)

singular plural b. SECOND PERSON: singular plural c. THIRD PERSON: singular plural (167)

Class I t- ∑ -˙n t- ∑ -da-n n- ∑ -˙n n- ∑ -da-n g- ∑ -˙n g- ∑ -da-n Class I t- ∑ -du t- ∑ -da-k n- ∑ -du n- ∑ -da-k g- ∑ -du g- ∑ -da-k

Class II ∑ -d -ßn ∑ -d -˙n ∑ -n -ßn ∑ -n -˙n ∑ -g -˙n ∑ -g -˙n Class II ∑ -d -u ∑ -d -˙k ∑ -n -u ∑ -n -˙k ∑ -g -u ∑ -g -˙k

singular plural b. SECOND PERSON: singular plural c. THIRD PERSON: singular plural 5.5.3. Tense-aspect-modality (TAM)

For most verbs in Kusunda there is, what appears to be, a three-way tense distinction – past, present, and future. By contrast, and central to our understanding of the three-way distinction, however, Class I verbs have only a two-way distinction. Past and present time are generally grouped together into a “realis” mode, and its opposite, “irrealis,” marks future time and possibility. Realis, in some cases, however, can also occur with future events, especially where certainty is high; such events are “perceptually located in the real world” (Elliott 2000). Elliott claims that this is the case for a number of Australian languages; “tense does not necessarily reflect reality status” (Elliott 2000:68). Evidence suggests that the realis–irrealis distinction is older and more primary than the so-called a “past–present–future” distinction. In fact, past tense (see Table 3) appears to be only an added category; it occupies some of the same semantic space as realis (see §5.5.4 on Negation) and, more significantly, its absence does not constitute a “present” tense (indicated by the “strike through” in Table 3). Rather, the absence of “past” is only realis. The major difference between past and realis is that past has an unequivocal past–completive reading, while realis is used more frequently in utterances with a kind of “neutral” or “timeless” sense, as in the verb ‘build’ in the following sentences: (168) a. pyana tsi wi a-t-ßn, t˙~îna yesterday I house build-1-REAL, today ‘Yesterday I built a house, today I will sleep.’ 66 ts-ip-du 1-sleep-IRR

Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. pyana tsi wi a-d-i, t˙~îna ts-ip-du yesterday I house build-1-PAST, today 1-sleep-IRR ‘Yesterday I built (finished) a house, today I will sleep.’ Past–present–future in Kusunda, then, is only the fiction of a three-way division. The real distinction is past–realis–irrealis. Thus, in Table 3, Class I verbs are those with a two-way distinction, and Class II verbs add a further distinction for a total of three. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– -n -du / -dak Class I | < ––––––––––– realis ––––––––––> | < –––– irrealis ––––> | -n -u / -k | < ––––––––––– realis ––––––––––> | < –––– irrealis ––––> | -i | < –––– past ––––> | present future ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Table 3. TWO-WAY CLASS I DISTINCTION, AND THREE-WAY CLASS II DISTINCTION From Table 3 we see that both verb classes mark future and unrealized (irrealis) events with -u / -k, and realis events with -n. Class II verbs, in addition, mark past completive events with -i (singular) or -ei (plural). The realis form is partially compatible with future time adverbials, although not in independent utterances as in (169a). It can be used, however, with future time adverbials in what appears to be a kind of neutral, timeless sense, as in (169b): (169) a. *goraq tsi ts-˙g-˙n 1-go-REAL *tomorrow I *‘I went tomorrow.’ b. t˙~în wi a-t-ßn, goraq ts-ip-ßn today house 1-make-REAL, tomorrow 1-sleep-REAL ‘Today I build a house, tomorrow I sleep.’ Irrealis, on the other hand, is fully compatible with future time. Thus, it can occur with future time adverbials, or with present time adverbials extendable to a future reading, as in the following: (170) a. goraq tsi t -a -an 1-go-IRR tomorrow I ‘Tomorrow I will go.’ b. d˙baq tsi t -a -an 1-go-IRR now I ‘I will go now.’ (‘I haven’t started yet, but I’m ready to go.’) Irrealis is not compatible with past time, (unless it occurs with incompletive -da, in which case it carries a sense of ‘would have’ or ‘was about to’ – see § Its incompatibility with past time adverbials is illustrated in the following example: (171) *pyana tsi t -a -an *yesterday I 1-go-IRR *‘Yesterday I will go.’ 67

25 68 . similar phenomena have been reported in isolated cases.’ nu n-˙g-˙n you 2-go-REAL b. 173b.5. a Totonacan language of Mexico.2. and .’ t-˙m-du 1-eat-SG:IRR ‘I will eat. respectively. t’. 25). ˙ becomes a. the process is old. second person subject agreement in verbs is marked by glottalizing all glottalizable consonants – p. the marked parameter is marked by mutation. ts.’ ‘You are going. then. REALIS ‘I went. velar becomes uvular.1. Though such marking systems are rare in the languages of the world. Beginning with the realis conjugation as the default. ch. for example.5.3. and u becomes o (see fn.Watters: Kusunda Grammar. as in: dzu≥-dzi ‘It hangs’ vs. Mutation In a small subset of Class I verbs (high frequency verbs like ‘go’ and ‘come’) a process of “mutation.2. k.4. and q become p’. k’. t-˙m-˙n 1-eat-SG:REAL ‘I ate. ts’. it is being replaced by a newer and more regular affixation pattern. §9.2 5 Apparently.3.’ Whatever the original function of ‘mutation’ may have been. not restricted to the realis–irrealis distinction alone. Jim Watters (1987) reports that in Tlachichilco Tepehua.’ ‘I am going.3. mutated dzoˆ-a-dzi ‘He hung it. the realis–irrealis distinction in the majority of Class I verbs is marked by affixation patterns. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 5. Following are examples: (172) FIRST PERSON a. t. Following are examples of irrealis forms marked by the affixes -du (for singular) and -dak (for plural): (174) REALIS IRREALIS a. there are two equivalent irrealis forms – those marked by mutation (172b. all consonants and vowels used for the irrealis form are “shifted” back one point of articulation – apical becomes laminal.’ See also sections §8. In such cases the irrealis morpheme is a kind of autosegmental harmony process which spreads across the whole of the root affecting all consonants and vowels. IRREALIS nu ¯-a -an you 2-go-IRR 5. IRREALIS tsi t -a -an I 1-go-IRR (173) SECOND PERSON a. In some transitivity diathesis alternations.3.” is used in marking the realis–irrealis distinction.1. §9. and §9. REALIS tsi ts-˙g-˙n I 1-go-REAL b. For “mutating” verbs.4.8 for the ‘mutating’ pattern in subordination. Non-mutating Class I verbs ‘You went. and 175) and those marked by -k (176). ch’.’ In contrast to the mutating Class I verbs. it shows evidence of having later been morphologized to operate in several grammatical systems.

5.Himalayan Linguistics: Archive No. -di marks past tense for singular participants. t-˙m-da-n 1-eat-PL-REAL ‘We ate.’ n-˙m-du 2-eat-SG:IRR ‘You will eat. This is seen as a trend away from the older mutating pattern to a more modern and widespread affixation pattern. Following are examples of verbs participating in both patterns (with no apparent difference in meaning): (175) MUTATING PATTERN a. n-˙m-da-n 2-eat-PL-REAL ‘You [pl] ate. nu ¯-a -ak you 2-go-IRR ‘You will go. and -dei past tense for plural participants. the actual TAM marking is only -u and -˙k for irrealis (singular and plural respectively). the tense 69 . can participate optionally in this conjugational pattern. too. TAM in subordinate clauses The distinction between realis and the more specific (more “marked”) past-completive is relevant to subordinate clauses as well.2. we have already seen (as in 153–155 and in 174) that -du marks irrealis for all singular participants and -dak for all plural participants (thereby distinguishing number in the suffixes. and ˙ > a.3. but not person).2.5.’ e.3.’ Comparing the forms in (175) and (176) with the base forms in (172a) and (173a).’ 2-eat-PL-IRR n-˙m-da-k ‘You [pl] will eat. Likewise. n > ¯. In Class II. As we saw in §5. and -i and -ei for past (singular and plural respectively).’ b.’ d.5. nu ¯-a -an you 2-go-IRR ‘You will go.’ (176) AFFIXATION PATTERN a. we see that even the irrealis conjugation utilizing the basic affixation pattern (174) also retains mutating characteristics – ts > t . tsi t -a -ak I 1-go-IRR ‘I will go. g > . however. those in which the matrix verb implies that the subordinated event actually happened.4.’ It turns out that mutating verbs. With the so-called “implicative” verbs. 5. n-˙m-˙n 2-eat-SG:REAL ‘You ate.’ t-˙m-da-k 1-eat-PL-IRR ‘We will eat.’ b.’ f. TAM in Class II verbs For Class I verbs.’ c. g-˙m-˙n 3-eat-SG:REAL ‘He ate.’ 3-eat-PL-IRR g-˙m-da-k ‘He will eat.3. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. in Class II verbs the -d is separable as a distinct person marking morpheme – a first person morpheme (see 157–165). 5. then. tsi t -a -an I 1-go-IRR ‘I will go.’ g-˙m-du 3-eat-SG:IRR ‘He will eat. g-˙m-da-n 3-eat-PL-REAL ‘He ate.

collocates with both realis and irrealis modalities. see §11. as in the following: 70 . Incompletive Another aspect. marked by -da. In realis mode. where the eventuality named by the matrix verb and the eventuality named by the subordinate verb are different events.’ b.5. and collocates only with realis or past tense.’ c.3.” or “past continuous” sense with habitual overtones. as in the following: (178) a. in some sense. but tentatively we have glossed it as a kind of incompletive (or imperfective).3. meaning ‘about to.5.2 and §11. 5.’ is marked by ben.3. as in the following: (179) a.3.’ 5. gina amba hab˙-g-i tsi he meat roast-3-PAST I ‘I saw him roasting meat. tsi ts-˙g-˙n ben I 1-go-REAL IMM ‘I am about to go. -da yields a kind of “past imperfective.’ c.6. wi a-t-ßn tumb˙-d-ßn house make-1-REAL finish-1-REAL ‘I finished making the house. is.5. Imminent aspect “Imminent” aspect.’ b. This can be illustrated by the following: (177) a. (See also §11. already asserted to be fact. tsi gin-da pumba-d-i ben I he-ACC beat-1-PAST IMM ‘I am about to beat him.Watters: Kusunda Grammar. The event. wi a-d-i tumb˙-d-i house make-1-PAST finish-1-PAST ‘I finished making the house. tsi wi a-t-ßn l˙mba-d-u I house make-1-REAL 1-learn-IRR ‘I will learn how to build houses. bem-dzi ben fall-3:REAL IMM ‘He’s about to fall.) We have not yet determined all the variables. the TAM marking need not match.’ ts-~aˆ -dzi 1-see-TAM For further discussion on subordination. though not yet realized.’ b. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– apparatus of both verbs must be matching. *wi a-t-ßn *house make-1-REAL (ungrammatical) tumb˙-d-i finish-1-PAST In other constructions.

tok ts-ip-d˙k-da we 1-sleep-PL:IRR-INCOMP ‘We would have slept.5. pinda t-˙m-da e-g-aÚ u-da. past does not have a negation pattern distinct from realis. and he died.’ With a non-volitional verb like ‘fall.’ as in the following: (183) gina nu bem-dzi-da that person fall-3:IRR-INCOMP ‘That person was about to fall.Himalayan Linguistics: Archive No. i. tsi wi a-d-i-da I house make-1-PAST-INCOMP ‘I was making a house.’ or ‘was about to. while past is a later development. the construction can also mean ‘while X was happening. earlier 1-eat-PURP give-3-NEG:REAL-INCOMP yeodzi t-˙m-da e-g-i 1-eat-PURP give-3-PAST now ‘He didn’t used to let me eat.’ b.’ In the right context.e. someone knocked on the door. used with irrealis.’ both senses compatible with non-fact modality. tsi-ag˙i pumb˙ ˙-g-i-da o -dzi 1S-dog beat be-3-PAST-INCOMP die-TAM ‘He used to beat my dog.’ the interpretation is more often the latter sense – ‘was about to.’ The same -da. Following are examples: (182) a.’ d.5. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (180) a.’ / ‘I used to go.’ It is not clear how this form contrasts with the “imminent” aspect illustrated in §5.3.’ and nu≥-dzi-da ‘was about to get up. yields a reading of ‘would have. 71 . nu e-n-u-da you give-2-SG:IRR-INCOMP ‘You would have given it. This lends support to our earlier hypothesis that realis and irrealis are the original and basic distinctions in Kusunda. now he lets me.’ b.’ as in the following: (181) tsi ts-ip-ßn-da n˙ti aoda d˙i-˙n I 1-sleep-REAL-INCOMP who door knock-NEUT ‘While I was sleeping.’ Other verbs that behave the same way are ip-dzi-da ‘was about to sleep. Negation There are two negation patterns – one for realis and another for irrealis. 5. In Class II verbs the distinction between past and realis is neutralized in the negative.5. tsi ts-˙g-˙n-da 1-go-REAL-INCOMP I ‘I was going.’ c.4.

-d (along with -n and -g) belongs to the person marking system and not to the negative suffix.” Thus realis affirmative statements assert something to be factual.” but is more closely related to “actual” and “possible.2 6 (See also §6. too. while irrealis asserts that even possible truth is not likely to become fact. Recall that the mutated form is related to several marked structures – something also true of negation.’ ‘He/she did not give.pl give-2-NEG e. nu e-n-a u you give-2-NEG d.3. The suffix -du. second. marks singular actant for first. In Class II verbs -d is separated off as a first person marker. the -d has become part of the negative suffix.2 Negative existential. and third persons. Following is a Class I negative–realis paradigm: This suggests that the original sequence is something like -a u. “realis” asserts as true that something did not happen (or is not happening). Negative realis in Class I verbs ‘I did not give. the imperative marker. Following is a Class II negative–realis paradigm: (184) CLASS II — NEGATIVE–REALIS: a. 5.5. Thus.1.) As in the affirmative. tok e-d-a u we give-1-NEG c.5. and third person is -g-aÚ u.4.4. and irrealis affirmative statements assert something to be possibly true.1. in fact. prefixing or suffixing) affects other affixation patterns as well.) 26 72 . In Class I verbs. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– It also becomes more clear under negation that the distinction between realis and irrealis is. nok e-n-a u you.’ ‘We did not give. tsi e-d-a u I give-1-NEG b. which interestingly.5.2.e. not related to “real” and “not real. In Class II negative verbs.’ As already alluded to.Watters: Kusunda Grammar. In Class II verbs. recall from (153–155) that with fully contrastive prefixed person markers in the affirmative (Class I verbs) the only distinction marked by suffixes is the number distinction. Negative realis in Class II verbs Negative realis-assertion is marked by a suffix -aÚ u. then. In the negative. with pharyngealization on the first vowel and falling–rising pitch on the full sequence. too. Class I verbs mark negative realis in all persons with the suffix -daÚ u. gina e-g-a u he/she give-3-NEG 5.’ ‘You did not give. that in the negative there is no marking in the verb to distinguish between singular and plural actants. while -dak marks plural actants. Such information is signalled by the free pronoun.’ ‘You [pl] did not give. second person is -n-aÚ u. (See §5.1 on Mutation. the affixation type of person markers (i.2. we have already seen repeatedly that -d is a person marker – first person negative is -d-aÚ u. of course. Note. is precisely the mutated form of -˙go.

and -g-aÚ u for third person. i.) Significantly. nok you. (See also Negative indefinite pronouns under §5. tok t-˙m-daÚ u we 1-eat-NEG:REAL c.’ ‘We did not eat. The most basic (and probably original) negative marker is a simple -u suffix added to the marked form of a mutating verb – yielding -a -u (as in 187a.) Next. gina m˙nni he many g-˙m-daÚ u 3-eat-NEG:REAL See §10. tsi t-˙m-daÚ u I 1-eat-NEG:REAL b. in Class II verbs. c) and phonetically yielding [-aÚ -u].1.’ Here we encounter a few clues about the composition of the more elaborate negative suffixes.e. REALIS n-˙g-˙n 2-go:REAL-IND ‘You went. 73 .Himalayan Linguistics: Archive No. Thus. It is derived from the negative existential or locative copula kæaÚ u or kæaÚ i and may be a past negative as opposed to a realis negative.3. 5.’ ‘They did not eat.1. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (185) CLASS I — NEGATIVE–REALIS: a.4.pl n-˙m-daÚ u 2-eat-NEG:REAL ‘I did not eat.’ ‘He/she did not eat. the modern realis negative marker includes -d as part of the negative sequence – -daÚ u for all persons (as in 185a–f). gina he/she 3-eat-NEG:REAL f. Finally. the slightly elongated form -aÚ u follows person markers yielding sequences like -d-aÚ u for first person.’ ‘You did not eat.5. both irrealis and negative are marked/mutated patterns. (186) a.3.2 7 Following is a negative–realis paradigm for mutating verbs (illustrating the simplest of the negative suffixes): 27 See footnote 26. Negative realis in mutating verbs All mutating verbs belong to Class I. and the negative suffix -u is added to the mutated root. (But not all Class I verbs are mutating. NEGATIVE ¯-a -u > [¯-aÚ -u] 2-go:IRR-NEG ‘You did not go.’ b. Mutation is part of the history of negation and occurs in both the simple and in the more elaborated forms of the morpheme. nu n-˙m-daÚ u you 2-eat-NEG:REAL d.’ ‘You [pl] did not eat.’ c. -n-aÚ u for second person. in non-mutating Class I verbs.2 for a second negation pattern found frequently on negated medial verbs of a clause chain.’ g-˙m-daÚ u e. IRREALIS ¯-a -an 2-go:IRR-IND ‘You are going. negative realis in mutating verbs employs mutation as part of its marking strategy. person marking indices are prefixed to the verb root.

nok n-id-aÚ u you. Note. but it seems to be more prevalent in third person contexts. as in (188) and (189) below.4. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (187) a.’ b. One possibility that suggests itself is that -i and -u are somehow related to the existential and locative copulas (see §6. This problem will require more study. We also saw from (150–152) that plurality is regularly marked by -da. as in the following examples: (188) CLASS I: a.Watters: Kusunda Grammar. We do not know if there is an implied semantic distinction between -i and -u.pl 2-go:PL-IRR:NEG ‘You [pl] did not go.’ < d-a -i Unlike the non-mutating verbs.’ e.’ 74 . however.4. This -wa is very likely related to the negative -u discussed immediately above. tsi t -aÚ -u 1-go:IRR-NEG I ‘I did not go. Parsing the forms. tsi t-˙m-wa I 1-eat-NEG:IRR ‘I won’t eat it.’ (189) CLASS II: a. gina d-aÚ -i he/she go-IRR-NEG ‘He/she did not go. 5.’ c.5. tsi ts-ip-wa I 1-sleep-NEG:IRR ‘I won’t sleep. that the negative in third person uses an -i suffix in place of the -u suffix. too. tsi a-d-u-wa I make-1-IRR-NEG ‘I won’t make it. The negative -i occurs also in other persons. tok ts-id-aÚ u we 1-go:PL-IRR:NEG ‘We did not go. here we see a distinction between the singular and plural forms of the negative – distinct plural forms occur in (187b) and (187d). From (141) we saw that the plural forms for ‘go’ involve a vowel change in the verb root from ˙ to i. There may be evidential distinctions here.’ < ¯-a -u < t -a -u d. Negative irrealis Negative irrealis-assertion is marked by a suffix -wa. nu ¯-aÚ -u you 2-go:IRR-NEG ‘You did not go.1).’ b. in the case of first person. from refusal.is glossed as ‘go:PL’. Negative irrealis-assertions can arise either from low probability or. is far from transparent. In (187b) and (187d) both processes occur – thus -id.

varying 75 . as in: (192) a. translated as ‘not yet.’ b. sometimes giving rise to potential ambiguities. as yet.g. Negatives with a strong component of “refusal” sometimes mark the pronoun twice on the verb – the old prefixed form and a newer suffixed form.’ Imperatives 5. undiscovered morpheme. as in: (191) tsi t-˙m-wa-tsi I 1-eat-NEG:IRR-1 ‘I won’t eat it. as in: (190) a.’ The same construction can also convey a sense of negative habitual: (193) numba amba t-˙m-daÚ u-tok cow meat 1-eat-REAL:NEG-1P ‘We don’t eat cow meat. suppletive forms for marking the imperative.’ c. most verbs follow regular patterns and we will deal with those first. tsi t-ug-u I 1-come-IRR ‘I will go.1. tok t-˙m-daÚ u-tok we 1-eat-REAL:NEG-1P ‘We haven’t eaten yet. Recall that historically *-u has been posited as the negative morpheme in (187) (see the discussion surrounding those examples). with -u the negative. 5.5.Himalayan Linguistics: Archive No.4. tsi t-ug-u < t-ug-wa I 1-come-NEG:IRR ‘I won’t go. • Refusal. Intransitive imperative The regular marker for the imperative in intransitive verbs is the suffix -to. high frequency verbs (e. and -a some.’ b.’ It is formed by suffixing a pronominal copy at the end of the realis negative verb.5. Another aspectual category is the negative perfect. tsi t-˙m-daÚ u-tsi I 1-eat-REAL:NEG-1S ‘I haven’t eaten yet.5.5. especially in old.’ Negative perfect 5. and each has to be learned separately. nu a-n-u-wa you make-2-IRR-NEG ‘You won’t make it.5.’ There is also a possibility that -wa is a bimorphemic sequence (< *-u-a). 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. There are a large number of irregular.5. see (197)).’ It should be mentioned that the irrealis negative morpheme -wa is often contracted to -u. Still. nu n-˙m-daÚ u-nu you 2-eat-REAL:NEG-2S ‘You haven’t eaten yet.

e. ip-to o -do bem-to hyoq-to b˙l-to limi-to dzoq-to tul-to sip-to pæu-to yu≥-to lib~u-to nu≥-to dæu≥-to byo≥-to dzæ˙m-to pæep-to hu. Following are examples of -to (-do) on intransitive verbs: (194) a. however. m. In fact. though we lack the data to make that claim. r. the form of the imperative is -no. l. n. the thing climbed is marked for locative case. d. b. o. 76 . verbs without a fully affected object are both semantically and morphologically “less than transitive” (Hopper & Thompson 1980).’ Another two verbs that are transitive in force but intransitive in morphology are the verbs ‘to beg.u-do ‘Sleep!’ ‘Die!’ ‘Fall!’ ‘Hide!’ ‘Descend / Come down!’ ‘Dance!’ ‘Swim / Get wet!’ ‘Dress (yourself)!’ ‘Enter!’ ‘Jump!’ ‘Lie down!’ ‘Play!’ ‘Get up!’ ‘Stand!’ ‘Squat!’ ‘Weep!’ ‘Drop!’ ‘Fly!’ • Nasal allomorph -no.’ In both cases. the suffix occurs also with verbs that one would expect to be transitive. a bare root without affixation. c. f.Watters: Kusunda Grammar. are formed by -to: am-to2 8 ‘Eat!’ and ¯aq-to ‘Wait!’ It may be that these two verbs are ambitransitive. however. the notional object is marked with the comitative morpheme: ‘to beg with someone’ and ‘to fight with someone. h. More difficult verbs to “explain” are kham-to ‘Bite!’ and o -to ‘Set it aside!’ 28 The imperative of ‘eat’ is also am. Other verbs which might be expected to take transitive morphology are the verbs ‘to eat’ and ‘to wait. p. i. j. g~aˆ -no ‘Sit!’ ‘Look!’ Some verbs which seem to be transitive in force also take the intransitive imperative marker (as in (195b) ‘look’). as in the following: (195) a.’ as in soq-to ‘Climb!’ In the case frame of this verb. Another such verb is ‘to climb. h~aˆ -no b. Where the verb root ends in a uvular nasal consonant. g. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– with -do.’ Their imperatives. not as an object. q. mount. k.’ Apparently. as in: (196) tsi-h~ukui-da yi-g˙ soq-indzi 1S-child-ACC tree-ON climb-let ‘Let my child climb in the tree.’ the imperative of which is – ai-to ‘Beg!’ – and ‘to fight.

n-ug-un ‘You came’. In such cases the imperative verb is conjugated for second person.or a-. e. in (197e) the imperative takes almost the same form as the bare verb root. b. n-i-di ‘You brought it’.’ Apart from the shared ‘d’ in third person forms. c. ts-o ˆ-˙n ts-i-di t-ug-un ts-˙g-˙n t-˙m-˙n qon ˙iga ˙ga da am ‘Drink!’ ‘Bring!’ / ‘Fetch!’ ‘Come!’ ‘Go!’ ‘Eat!’ In (197a). as in the following sample: (198) a. d. d-˙g-˙n ‘He/she went. bem ˙-go pumba ˙-go d~aba ˙-go kisa ˙-go unda ˙-go hab˙ ˙-go t˙mb˙ ˙-go borl˙ ˙-go dzaÚ ˙-go abi ˙-go toba ˙-go ‘Trip him!’ ‘Beat him!’ ‘Find it!’ ‘Cut it!’ ‘Show it!’ ‘Roast it!’ ‘Send it!’ ‘Boil it!’ ‘Buy it!’ ‘Carry it!’ ‘Sew it!’ Because most transitive verbs take the syntactic form of a “light verb” (i. go ˆ˙n ‘He/she drank. i. b. as in: ts-˙g-˙n ‘I went’.e. g-i-di ‘He/she brought it.’ For (197b). c. many such verbs are intransitive in force. e. Among them are the following: (197) 1ST PERSON FORM IMPERATIVE FORM VERB GLOSS a. the regular conjugated forms have i. as in: t-ug-un ‘I came’. This phenomenon is 77 .’ In (197d). compared to the rest of the paradigm. the regular conjugated forms have ug.as a root. some nominal notion plus the semantically empty verb ‘to do’).2.’ For (197c). d. n-˙g-˙n ‘You went’.as a root. e. the imperative occurs in an unrecognizable form. d. There are several imperatives formed by irregular or suppletive conjugation. the ‘d’ element in the imperative form da surfaces only in some third person forms. b.5.Himalayan Linguistics: Archive No. imba ˙-go nikæe ˙-go lol˙ ˙-go ya≥ ˙-go an ˙-go ‘Think!’ ‘Laugh!’ ‘Cry out!’ ‘Defecate!’ ‘Urinate!’ An exception to the generalization that transitive imperatives are fashioned after the verb ‘to do’ is in the case of Class I verbs – those with prefixed person marking.5. is unrecognizable: tso ˆ˙n ‘I drank’. Transitive imperatives Transitive imperatives are much more regular than their intransitive counterparts. f. Finally. as in: ts-i-di ‘I brought it’.’ whose root form is ˙. h. 5. k. ug-˙n ‘He/she came. Most transitive imperatives are fashioned after the imperative of the verb ‘to make. to do. g. c. j. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– • Irregular and suppletive forms. the form qon. as in the following: (199) a. The morpheme ˙-go transfers across to (almost) all transitive imperatives. no ˆ˙n ‘You drank’.

though only a single example has been found – ‘Know it!’ (see (200). -yin) following the verb root. n-˙m-in 2-eat-PROH c.1. c.6. n-o -˙n 2-die-PROH (203) a. both transitive and intransitive.) (202) CLASS I INTRANSITIVES (includes pronominal prefix): a.5. As with imperatives. n-ip-in 2-sleep-PROH b. bem-in fall-PROH b. First person realis is t-ukan. e.1. bem ˙-yin pumba ˙-yin d~aba ˙-yin kisa ˙-yin unda ˙-yin hab˙ ˙-yin ‘Don’t trip him!’ ‘Don’t beat him!’ ‘Don’t allow it!’ ‘Don’t cut it!’ ‘Don’t show it!’ ‘Don’t roast it!’ 5.Watters: Kusunda Grammar. are marked by the suffix -in (allomorphs -en. n-im-in 2-forget-PROH b. This form is identical with second person realis n-uk-an (2-know-REAL). Intransitive and transitive Class I prohibitives Intransitive prohibitives and Class I prohibitives are simpler than those illustrated in (201) in that they occur without the “helping” verb ‘to do. n-ukan2 9 Prohibitives ‘Know it!’ / ‘Figure it out!’ Prohibitives. as can be seen in §5. on the surface. -˙n. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– much more common in intransitive verbs. b. (Note that the pronominal prefix also occurs potentially on transitive imperatives.6.5. phu-yin jump-PROH ‘Don’t sleep!’ ‘Don’t die!’ CLASS I TRANSITIVES (includes pronominal prefix): ‘Don’t forget it!’ ‘Don’t eat it!’ ‘Don’t look at it!’ CLASS II INTRANSITIVES (without pronominal prefix): ‘Don’t fall!’ ‘Don’t jump!’ A strange alternative set is available for the forms in (202) which. make. f. make’ – ˙-yin ‘Don’t do it!’ Following are examples: (201) a.5. ¯-aˆ -yin 2-look-PROH (204) a. We have a single example in transitive verbs: (200) 5. the root is preceded by the second person pronominal prefix n-. and third person ukan-dzi.6. most (but not all) transitive prohibitives are built on the prohibitive form of the verb ‘to do. 29 78 .’ In addition. with Class I verbs (see (202–203)). d.

EAT REALIS: Class II .5. ‘Don’t sleep!’ ‘Don’t die!’ Verbs that participate in both -in and ˙-yin prohibitive Some verbs are capable of taking either the -in prohibitive or the prohibitive ˙-yin based on the verb ‘to do. 5.e. d. -tsi marks some unknown category: (205) CLASS I INTRANSITIVES (alternative pattern from (202)): a. n-o -tsi-n 2-die-1-PROH 5. there is generally a defective or poorly developed past perfective marking (sometimes missing altogether). c.’ / ‘I beat. paralleling the forms for ‘beat’: (207) Class I .’ / ‘You beat.7.’ realis and irrealis forms occur predictably as follows. t-˙m-˙n b. i. n-˙m-˙n . with a Class I verb like ‘eat. Hortatives and optatives The “hortative” and “optative” in Kusunda specify similar semantic notions. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– appear to include the first person affix -tsi.1.’ In the majority of cases.’ / ‘We beat. e.Himalayan Linguistics: Archive No. those with prefixed person marking.’ ‘You eat. b.’ -n-i ‘second person.’ and -g-i ‘third person. t-˙m-da-n c.’ In verbs without the past distinction (some of the Class I verbs) there is only a realis–irrealis distinction.) The verbs that concern us here. differing primarily in the person involved in the predication. The hortative is a kind of first person imperative expressing notions like ‘Let’s go!’ whereas the optative is a kind of third person imperative expressing a notion of ‘Let him go!’ or ‘May he go!’ 5.5.2 on Periphrastic causatives. In Class I verbs. Recall from examples like (163–165) that past marking is characterized by -d-i ‘first person. Could this possible mean something like ‘I say to you.’ ‘We eat. this makes for an intransitive/transitive difference. not first person past. bem ˙-yin ‘Don’t make him fall!’ (See §9.6. are those which take either ending without a perceptible difference in meaning. don’t sleep!’? If not.7.2. Hortative Our understanding of the hortative in Kusunda is still tentative. It is in precisely these cases that -di marks hortative.BEAT pumba-d-ßn pumba-d˙-n pumba-n-ßn 79 ‘I eat.5. Following are examples: (206) a. as in: bem-in ‘Don’t fall!’ vs. Thus.’ a. n-ip-tsi-n 2-sleep-1-PROH b. however. n-o -˙n / n-o ˙-yin n-o ˆ-˙n / n~o ˆ ˙-yin pumba-n-in / pumba ˙-yin hab-en / hab˙ ˙-yin tsim-in / tsim˙ ˙-yin ‘Don’t die!’ ‘Don’t drink!’ ‘Don’t beat him!’ ‘Don’t roast it!’ ‘Don’t touch it!’ Further research may reveal varying levels of transitivity or even levels of intentionality in such verb pairs. Part of the problem is due to a close similarity in form between hortatives and first person past forms.

’ / ‘He/she beats. c.’ / ‘Let me buy it!’ ‘I touched it. d. A question arises as to how the hortative is marked in Class II verbs. b. f.’ ‘They beat.’ ‘Let me (go/get) beat it!’ The grammaticalization of a hortative from verbs like ‘come. fetch’: (211) a.’ / ‘You [pl] beat.’ ‘Let’s eat!’ / ‘We beat.’ / ‘I will beat.’ / ‘They beat. they are unsure about their grammaticality and where to place them.’ ‘go. b. For some verbs we have homophonous forms for the hortative and first person past: (210) a.’ ‘I will eat. Kamala says “no”. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– d. n-˙m-da-n e. go get.’ / ‘They will beat. though there appears to be some confusion on that count.’ a.’ ‘You beat. e.’ ‘He/she beats. t-˙m-du t-˙m-da-k n-˙m-du n-˙m-da-k g-˙m-du g-˙m-da-k In the past paradigm.’ ‘He/she eats. d.’ / ‘Let me give it!’ ‘I bought it. c.’ and ‘get’ is not unusual and has been reported in numerous languages (see Aikhenvald forthcoming: on 80 . t-˙m-d˙i c.’ / ‘We will beat. Unfortunately.’ ‘You [pl] will eat / beat.’ ‘We will eat. pumba-di / -d-i beat-HORT / -1-PAST b.’ / ‘You will beat. e. d. however.’ ‘They eat. g-˙m-˙n f. Though the forms n-˙m-di and g-˙m-di have a familiar ring to them. f. not past tense: (209) Class I . Gyani Maiya and Kamala are in disagreement on whether a form n-˙m-di is possible – Gyani Maiya says “yes” (as a past tense).’ ‘He/she will eat. -di and -dei mark singular and plural hortatives. Class I forms are defective – i. our notes are sketchy and inconsistent on this matter and we cannot give a definitive answer.BEAT pumba-d-i pumba-d-˙i pumba-n-i pumba-n-˙i pumba-g-i pumba-g-˙i ‘Let me eat!’ / ‘I beat. g-˙m-da-n (208) IRREALIS: pumba-n˙-n pumba-g-˙n pumba-g-˙n pumba-d-u pumba-d˙-k pumba-n-u pumba-n˙-k pumba-g-u pumba-g˙-k ‘You [pl] eat. but past tense in Class II. t-˙m-di b. a-d-i e-d-i dzaÚ-d-i tsim˙-d-i ‘I made it. Furthermore.’ ‘You [pl] beat.e.’ / ‘Let me touch it!’ For other verbs we have the homophonous form plus a disambiguated form that makes use of the Class I hortative form of the verb ‘get.’ ‘You will eat.EAT PAST: Class II .’ a. The analogy between Class I hortative and Class II past tense apparently confuses them.’ / ‘Let me make it!’ ‘I gave it.Watters: Kusunda Grammar.’ ‘They will eat. the forms expected in (209c–f) are generally missing from Class I. Forms -di and -d˙i (209a–b) mark the hortative in Class I.’ / ‘He/she will beat. pumba ts-i-di 1-get-HORT beat ‘Let me beat it!’ / ‘I beat it.

whereas the optative (either the negative. has a paradigmatic relationship to the imperative. The negative imperative (the prohibitive). c. the hortative has formal ties to future tense. Wherever g. ‘do.7. In the languages surrounding Kusunda. and following universal expectations. 5. ‘die’ g-o -ya o -wa --------------------------------------------------------------d. mostly transitive. but generally they are so confused and incomplete that we will not attempt to present them here. Optative • Optative in Class I verbs. the paradigm is very irregular. • Optative in Class II verbs. c.e. ‘come’ g-u-ya g-u-yin e. ‘go’ ‘come’ ‘die’ ‘sleep’ ‘drink’ ‘put’ ‘see’ PROHIBITIVE OPTATIVE n-iw-˙n n-ug-in n-o -˙n n-ip-in n-~oˆ -˙ni ?? ˘-~aÚ-yin g-ya g-u-ya g-o -ya g-ip-da g-oˆ -na g-o -ta g-~aˆ -na Notice here. i. make’ ‘give’ ‘beat’ ‘buy’ ‘carry’ IMPERATIVE OPTATIVE ˙-go e-go pumba ˙go dzaÚ ˙go abi ˙go 81 ˙-ge e-ge pumba ˙ge dzaÚ ˙ge abi ˙ge . that Class I verbs. The optative. Apart from this simple generalization. are the ones that diverge.5. but it is interesting to note that negative -wa occurs only in the absence of third person g-. There are also negative optatives in our data set (as in ‘Let him not go’). Class II verbs. b.2. ‘eat’ g-˙m-˙ g-˙m-in Our data is insufficient to make any predictions. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Origins of imperatives). especially in Class I verbs. for example.’ Recall that such verbs were also very regular in the imperative. whether transitive or intransitive. and others the irrealis negative suffix -wa: (213) POSITIVE NEGATIVE a. empoying the suffix ˙-go ‘Do it!’ / ‘Make it!’ (see examples in (198)). positive. the negative suffix used is the prohibitive -in. d. or both) employs the third person prefix g-.Himalayan Linguistics: Archive No. The most regular optative forms are those for Class II transitive verbs. except to say that some of them appear to take the prohibitive suffix -in. too. employs the second person prefix n-. ‘go’ g-ya d˙-w˙ b. e. g. ‘sleep’ g-ip-da ip-d-wa c. Here the suffix empoyed is -˙ge.occurs. e. those patterned after the verb ‘to do. f. follow similar conjugational patterns. Following are examples: (212) a. fashioned after the optative form of ‘do’: ˙-ge ‘May he do it!’ Following are examples: (214) a. b. d. make. What is unusual is a formal similarity between past tense and hortative.

is based on the optative form of ‘go. which might suggest that this is an intransitive form. 5. wherever it occurs. j. and dziu-gya ‘Let it leak!’ Significantly. occurs in the optative form of all Class I verbs. as we saw earlier. This is analogous to the intransitive imperative formed by the suffix -to (allomorphs -do and -no in the same environments as optative -da and -na) and fits our analysis well. is almost always transcribed without the intervening schwa: -gya. e.’ ‘place. as in kis˙-gya ‘Let him cut it!’.’ ‘time.’ Are we dealing here with a transcription error. 30 82 .6. Among them are lim-gya ‘Let him dance!’. to become. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– f.’ In our current data set we do not have hui-da ‘Let it fly!’ or m˙b˙ ˙ge ‘Let him hear!’ If we could confirm these kinds of forms our analysis would be on safer ground. f. wear’ ‘fall’ ‘hide’ ‘see’ ‘sleep’ ‘squat. b.’ and ‘quantity’ respectively. Given further that the optative form of the verb ‘to go’ is g-ya. too.’ and ‘how many’ as adverbs of ‘manner. The more traditional ones are relatable to question words like ‘how. and tsim˙n-gya ‘Let him touch it!. m˙b˙-gya ‘Let him go hear!’3 0 (We also saw in (211b) that ‘go/get’ is used as a basis for some hortative constructions. b˙l-gya ‘Let him descend!’. dzæ˙m-ya There are several heterogeneous classes of words traditionally called adverbs that have in common that they modify events or states (i.Watters: Kusunda Grammar. verbs or adjectives). i. however. g. sip-gya ‘Let him enter!’. d. or -ya. h. Following are examples: (215) a.’ ‘when. the form occurs just as frequently on transitive verbs. c. but not ˙-gya. sit’ ‘weep’ IMPERATIVE OPTATIVE kæai-to o -do tul-to bem-to hyoq-to g~aˆ -no ip-to byo≥-to dzæ˙m-to kæai-da go -ya tul-ya bem-ta. that the source for the transitive Class II opatative -˙ge. the form. we are left with a residue of intransitive verbs that form the optative with the suffix -ta (allomorphs -da and -na). and that -gya is a ‘go’ optative. ‘cook’ ‘find’ ‘hear’ ‘kill’ ‘roast’ hol˙ ˙go d~ab˙ ˙go m˙b˙ ˙go o da ˙go hab˙ ˙go hol˙ ˙ge d~ab˙ ˙ge m˙b˙ ˙ge o da ˙ge hab˙ ˙ge There is some confusion about another optative form -gya that occurs on the verb ‘to be.’ ‘where.) Assuming that our analysis is correct. it is probable that we are looking at a grammaticalized verb ‘go’ tacked on as a suffix as in: hui-gya ‘Let it go fly!’. with ˙gya being the same form as ˙ge? Probably not.being separable as a third person prefix (which. i. It is possible. and will be organized as such here. with g. Adverbs ‘be afraid’ ‘die’ ‘dress.e. bem-ya hyoq-ta g~aˆ -na gip-ta byo≥-da dzæ˙m-ta. g. see (212)). Interestingly. h. even likely.’ as in: ˙-gya ‘Let it be!’ The form occurs erratically on other intransitive verbs as well.

qa tse ‘slowly’ ‘tardily’ ‘fast. gin ˙di ‘doing like this’ ‘doing like that’ Other adverbs of manner have to do with the relative speed in which an event occurs.6. quickly’ ‘alone’ ‘together’ Others adverbs refer to the associative aspect of the referents performing the action: (218) Still others adverbs of manner answer questions about an event’s level of certainty or necessity. early’ ‘evening’ ‘afternoon’ ‘this year’ ‘next year’ ‘last year’ deosi goradze isi.’ and ‘in that manner’: (216) a. pene to on (221) TIME WITHIN A DAY: ‘today’ ‘tomorrow’ ‘day after tomorrow’ ‘yesterday’ ‘day before yesterday’ ‘later today’ ‘morning. as in the following: (217) a. m˙la≥ c. though it must be pointed out that the different domains have no significance or relevance in grammatical terms: (220) DAYS: t˙ina.Himalayan Linguistics: Archive No. as in: Yesterday. Adverbs of time ‘again’ ‘almost’ Adverbs of time are broader in scope that manner adverbs. Adverbs of manner Two adverbs of manner that are generic are based on ‘this’ and ‘that’ and combine with the ‘make’ verb root to form pro-forms – ‘in this manner. about the margin of possibility by which it failed to occur. REALIS: k˙pæera b. In the following examples. t˙ndin goraq qoturaq pyana. m˙st˙m˙ b. myaqa b. and if the event did not occur. and characterize entire events.1. we list them in broad semantic domains. Prem Bahadur went to the forest. iskin oˆ ts˙ (222) YEARS: t˙inan b˙s˙ oran b˙s˙ pyai d˙g˙n b˙s˙ 83 . as in: (219) a. IRREALIS: hunko 5. siba a.6.2. ai ˙di b. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 5.

2. b. as in: Kamala sat across from me.4. Adverbs of location ‘later’ ‘earlier’ ‘before.3 as a class of nouns – “relator nouns” – and will not be repeated here. h~utu b. Those listed here are locative postpositions. c. e. Locative suffixes also specify location or direction. earlier’ ‘long ago’ ‘long ago’ ‘in a moment’ ‘now’ ‘always’ Adverbs of location specify the physical location of an event or of its participants. We will add here two other adverbs – adverbs of relative distance – that are somewhat different from physical location: (224) 5.3.2. (225) a. ist˙ Adverbs of quantity a.6. d. as in ‘in the tree’ or ‘to the house’ and were treated in §5.Watters: Kusunda Grammar. godza≥ m˙nni h~uku h~uku sw˙tt˙i m˙ta. Word classes ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (223) GENERAL: s˙mbaq pya pinda pyai pyai pinda pinda yeodze d˙baq ˙sae 5.3. Locative postpositions were treated more generally in §5.6. m˙ ‘few’ ‘many’ ‘a little bit’ ‘all’ ‘only’ ‘far’ ‘near’ Following are examples of adverbs of quantity: 84 .

na tsi kila-d-ßn gimi this I steal-1-REAL money ‘This is money that I stole.’ 85 . 6.not ‘This money is not mine. as in the following: (226) a. and 3) intransitive.’ in which ‘this’ is the predicator. nu m˙b˙-n-i gip˙n n˙tßn what you hear-2-PAST talk ‘What is the matter/talk you heard?’ A reordering of elements in any of these equative clauses makes for a slight difference in meaning. but in this section we will attempt to summarize and bring together all the salient features that serve to define them. Copular constructions There are three basic copular constructions in Kusunda: a verbless equative. The construction is formed by the juxtaposition of two noun phrases (NPs) without grammatical case marking. for example. an existential copula. which can occur virtually anywhere the zero copula occurs.1.1. and bitransitive constructions. and the complement functions as the predicator. an infrequently used equative copula does exist). A reording of (226d).5 we have already noted several distinctions between transitive and intransitive verbs. na gimi tsi-yi odoq this money I-GEN is.’ e.1. 2) existential and locative copular constructions. The negative of (226b) and (226d). gina k˙l˙m n˙ti-ye that pen who-GEN ‘That pen is whose?’ d.1. Basic clause types Basic clause types in Kusunda are 1) verbless equative constructions. In §5. na gimi tsi-yi this money I-GEN ‘This money is mine. Equative clauses The equative clause is most commonly expressed without a copula (though. 6. and a locative copula. The order of the elements has grammatical implications – the subject is the presupposed. is as follows: (227) a. The first NP in the sequence is the copula subject and the final NP is the copula complement. 6.1 Negative equative The negative equative copula is odoq.6.1. as we shall see.’ c. for example. transitive. n˙ti na what this ‘What is this?’ b. topical element.1-REAL money = this] ‘The money I stole is this. yields tsi kila-d-ßn gimi = na [I steal.

the equative copula b˙t can also occur with adjectives.’ b. the implication is that one particular item is chosen from among many as having the relevant feature – ‘This one is long’ or ‘This is the long one. However. however. because the positive construction is verbless.’ not just ‘I am a king. the predicate adjective copula does occur in this environment (as in na pæurlu≥ tsu ‘This is red’.’ There is still a possibility that we have interpreted (228a) incorrectly.’ In effect.’ c. an environment in which one might expect the “predicate adjective” copula.Watters: Kusunda Grammar. however. in which case (228a) should be interpreted to mean ‘This is heavy. Following are examples: (229) a. we are assuming that the adjective in the verbless construction. This is similar to what occurs in Nepali and other languages of the region. Of course. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– odoq b. too.’ Likewise. ta hol-di amba this cook-ADJ meat ‘This is cooked meat.’ b. and there may be pragmatic constraints on its use.2 An equative copula There is an equative copula b˙t.not ‘This one is not long’ / ‘This is not the long one.’ We have recorded a few cases in which the equative copula (Ø or odoq) occurs with adjectives.1. like Kham. 86 . tsi b˙t I equative ‘It’s me. Perhaps adjectives are “verbal” enough in their semantics that they can be used as simple predicators. 6. In Kusunda. the copula b˙t occurs only rarely. in which case the adjective functions as a nominal: (230) b˙t tsi h~ukui I small equative ‘I am a small/unimportant (person).’ b˙t equative Note that in (229b) the implication is similar to what we described for (228a) – the referent is singled out from among many – ‘I am the one who is king. is functioning as a nominal. tsi mo≥ b˙t I king equative ‘I am the king.’ / Not *I am small. that the negative equative is possible here. but where the equative occurs. the adjective is rendered as a nominal.1. na li≥wa Ø this heavy (copula) ‘This one is heavy’ / ‘This is the heavy one. na tsi kila-d-ßn gimi this I steal-1-REAL money is.not ‘This is not money that I stole. see example (237a) below).’ Given. it is more difficult to determine the correct interpretation in that context: (228) a. na l˙≥ka odoq this long is. which is in some ways the positive counterpart of the negative equative described above.

The same copulas (positive and negative) used for existential senses are also used in predicate adjective constructions.1) that the actual negative sequence is -aÚ u. tu dukæu tsu snake two exist ‘There are two snakes.” The former expresses that something exists in a given location.1.1 Existence It is important to distinguish in Kusunda the difference between “existence in a location” and “location.’ < kæ-a u 87 . number. or for tense (that we are aware of). 6.1. although there are some limited aspectual notions supported by it. ta≥ kæ-aÚ u water neg.2.2 Negative existential The negative existential is kæaÚ u.’ whereas the latter focuses on the actual location of a given referent.’ The existential copula is used primarily to express the former – existence in a location. We noted earlier (see the discussion following example (74) and §5.2. with kh-a u as the negative existential. This leaves kh.’ (tsi would mean ‘Two snakes are there’ / ‘There are two snakes there’) 6.2 Existential copula The existential copula is tsu in the positive and kæaÚ u in the negative. is rare in Kusunda: (231) a. a suppletive form with no apparent relationship to tsu.4. Following are examples: (232) a.1. tsi-˙i tsu my-father exist ‘My father is’ / ‘My father is living’ / ‘I have a father.’ b. It does not conjugate for person.as the negative existential verb root.’ a. ta≥ tsu water exist ‘There is water. yi-g˙ ha o tsu tree-IN monkey exist ‘In the tree is a monkey’ / ‘There is a monkey in the tree. as in any language.Himalayan Linguistics: Archive No.’ d. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– We are unaware of any tense distinctions for this copula or for the Ø equative copula.exist-not ‘There is no water. with pharyngealization on the first vowel and falling-rising pitch on the full sequence. as in ‘The monkey is in the tree. Pure existence. like ‘In the tree is a monkey. wu tsi-˙i tsu home my-father exist ‘My father is (at) home.’ c.5. This suggests that the original negative sequence is something like -a u. 6.

3 Possession As reflected in the glosses in (231a) pure existence is rare. ‘with you how many exist?’) c.’ (lit.’ b.not ‘I don’t have even one child.’ (lit. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 6. and used in conjunction with possessive clauses. nu-ma ˙si tsu? you-WITH how.4 In “dative subject constructions” feelings/emotions/pain are construed as existing ‘to’ a referent. ‘with me one exists’) Negative possession uses the same negative existential verb that we saw earlier – kæaÚ u.1.5 The “predicate adjective” construction – ‘He is tall. the construction is similar to the possessive clause construction.many exist ‘How many do you have?’ (lit.’ ‘I am big. ‘hunger is to me’) t˙n-da ida≥ kæaÚ u me-DAT hunger is.2.not ‘I am not hungry. as in: (237) a. tsi-h~ukui li≥wa tsu my-small heavy is ‘My child is heavy.1. and ‘to me’ in dative clauses. the interpretation is one of possession – ‘I have a father’ (though the literal translation is ‘My father is’).2. – predicates adjectival notions through the help of the existential copula tsu. ‘hunger is not to me’) Predicate adjective The negative uses the same existential negative kæaÚ u that we saw earlier: (236) 6.2.Watters: Kusunda Grammar. Following is an example: (234) tsi-h~ukui qasti b˙ kæaÚ u my-little one even is. Following are examples of typical possessive clauses: (233) a.’ Dative subject constructions 6.’ (lit. the major difference being in the marking of the topical argument – ‘with me’ in possessive clauses.’ 88 . na pæurlu≥ tsu this red is ‘This is red.’ In terms of syntax. tsi-h~ukui dahat tsu my-small three exist ‘I have three children. tsi-ma qasti tsu I-WITH one exist ‘I have one. (235) t˙n-da ida≥ tsu me-DAT hunger exist ‘I am hungry.’ etc.’ (lit. as in ‘Hunger is to me. ‘my little ones three exist’) b.1.

’ ‘Are you (there)?’ ‘Are you there?’ c. Following are examples: (239) POSITIVE: a.neg ‘The monkey is not (located) in the tree. suta bæ˙ra-q-ßn tsu string break-AC-REAL is ‘The string is broken.1. whether explicitly in a place or not. ben amba ola≥ kæaÚ u raw meat tasty is. tsi tsi I loc. Contrasting kæaÚ u with kæaÚ i we have further evidence that the sequences were originally tri-morphemic – kæ-a -u versus kæ-a -i.Himalayan Linguistics: Archive No.’ ‘I am here. Recall that the negative existential was kæaÚ u. ha o yi-g˙ kæaÚ i monkey tree-LOC loc.) 6.copula a.’ b. nu tsi? you loc.’ 89 .’ The predicate adjective construction is also used to predicate attributes designated by “participles” (here partially-finite adnominals). in that the latter expresses that the referent exists. k˙re k˙la-q-ßn lan tsu pot break-AC-REAL like is ‘The pot is about to break.’ (See also (230).copula ‘I am (here). the adjective can occur also without the copula. tsi t˙isa I here tsi loc.copula b.not ‘Raw meat is not tasty. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– The negative uses the same existential negative kæaÚ u we saw earlier: c. It differs from the existential copula. nu isna you there (240) NEGATIVE: tsi? loc. in which case it is interpreted as a nominal in an equative clause – ‘I am a small person.’ d. The locational copula is tsi in the positive and kæaÚ i in the negative.copula d. tsi-h~ukui li≥wa kæaÚ u my-small heavy is. as in the following: (238) a.3 Locational copula The locational copula is used to predicate the location of a referent – it is in such and such a place.’ (lit. ‘is like it will break’) As shown in (228a).not ‘My child is not heavy.

LOCATIONAL: ha o yi-g˙ tsi monkey tree-LOC loc. ‘money exists with me.’ / ‘There is a monkey in the tree.2). where the real predication is done by a nominal and the verb is semantically empty). ‘did an egg’) 90 . we saw that in possessive clauses marked by the comitative -ma ‘with.Watters: Kusunda Grammar.’ both existential and locational locatives can be used felicitously.’ Notice that sentences (239a) and (239b) are very nearly synonymous.’ (lit.copula ‘I have money. ‘money is located with me. 2) they have an intransitive–transitive paradigmatic relationship to one another. as are (239c) and (239d). tap gimi gwa ˙-g-˙n chicken female egg do-3-REAL ‘The hen layed an egg. gina kæaÚ i he/she loc. too.’ (lit. LOCATIONAL POSSESSIVE: tsi-ma gimi tsi I-WITH money loc.’ (lit.1.neg ‘He is not (there). Following are contrastive examples: (241) a.’ In §5. EXISTENTIAL POSSESSIVE: tsi-ma gimi tsu I-WITH money exist ‘I have some money on me.5.copula ‘The monkey is in the tree. Notice.2.2. all transitive. p˙idzi bun˙ ˙-g-˙n cloth weave do-3-REAL ‘She is weaving cloth. example (124). EXISTENTIAL: yi-g˙ ha o tsu tree-LOC monkey exist ‘In the tree is a monkey.4.’) b.e. and 3) they form the basis for verbal conjugation in “non-inflecting verbs” (see §5. how this copula contrasts with the existential copula. Where the existential is used the assertion is concerned with the existence of the item.’ b. Following are some light verb constructions. What is implied in sentence (a) and (c) is made explicit by a locative adverb in (b) and (d).4 The inchoative and the causative The inchoative and the causative are treated here as subtypes of copular constructions because 1) they serve frequently as verbs in “light verb” constructions (i. where the locational is used the assertion is concerned with the location of the item: (242) a.’ as in the following: (243) a.’) 6.’ (lit. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. ‘does weaving’) b. and all of which utilize the verb ‘to do.1.

REALIS: 1s ˙-ni-n. Nevertheless. There may be a single ambi-transitive verb with the interpretation based on pragmatics. IRREALIS: 1s ˙-d-u 1p ˙-d˙-i 2s ˙-n-u 2p ˙-n˙-k 3s ˙-g-u 3p ˙-g˙-k We have a few sentences in our corpus in which there is a clear distinction between “inchoative” and “causative” roots: (245) tsi mo≥ ˙-d-u s˙h˙r a-d-u make-1-IRR:SG I king become-1-IRR:SG city ‘I will become king and I will make a city. and we can say nothing substantive about it. It is difficult to say at this stage. As we have shown earlier (see §5. i. tsi du≥dzi a-d-i I drum do-1-PAST ‘I beat the drum.2).’ The status of the inchoative realis is confusing.1.’ There is some question about whether there is a phonological distinction between the inchoative ‘become’ and the causative ‘make. with what appears to be a lot of variation. make. Class II transitive conjugations are patterned after the “causative” paradigm. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– c.’ Under careful articulation we recorded distinct paradigms. but on the following day our informants would often give either interpretation for either paradigm. PAST: CAUSATIVE (‘make’) 1s 1p 2s 2p 3s 3p ˙-d-i ˙-d˙-i ˙-n-i ˙-n˙-i ˙-g-i ˙-g˙-i 1s 1p 2s 2p 3s 3p 1s 1p 2s 2p 3s 3p 1s 1p 2s 2p 3s 3p a-d-i a-d˙-i a-n-i a-n˙-i a-g-i a-g˙-i a-t-ßn a-d˙-n a-n-ßn a-n˙-n ˙-g˙-n ˙-g˙-n a-d-u a-d˙-k a-n-u a-n˙-k a-g-u a-g˙-k b. based on the auxiliary verb ‘to do. ˙-yi-n 3p c.Himalayan Linguistics: Archive No. we give here two separate paradigms: (244) INCHOATIVE (‘become’) a.e. ˙-n˙-n.5.’ as in the following contrastive examples (‘do’ marks the Class II auxiliary verb root): 91 . ˙-ni 1p 2s 2p 3s ˙-gi-n.

’ (malefactive) 92 . example (356): (248) a.3. BASE FORM: nu tsi-wi ukæ˙ a-n-i you 1S-house ruin do-2-PAST ‘You ruined my house. The verb ‘give’ is the prototypical bitranstive verb and forms the basis for some benefactive constructions in Class II verbs. Most Class II verb. or in a vowel radically different from the first vowel [a] of the causative root ‘make. In much the same way that ‘do’ functions as a transitive marker in Class II verbs. with the exception that the root is e-. tsya d˙baq ˙-g-u tea later be-3-IRR:SG ‘Tea will be made later.’ Where the verb root ends in a consonant. We have a few examples in which the inchoative version of the verb root [˙-] can be used in a kind of “passive” sense.2. so too ‘give’ functions as a bitransitive marker for some of the same verbs. Following are examples: (247) a. Where the verb root ends in [˙] or [a]. and we do not yet know what the constraints are. (246b). Following are examples (see also §10. The paradigm is identical to the “causative” paradigm given in (244).Watters: Kusunda Grammar. cannot participate in this construction. tsya d˙baq hul˙ ˙-g-u tea later cook be-3-IRR:SG ‘Tea will be cooked later. CLASS II TRANSITIVE tsi abi-a-t-ßn I carry-do-1-REAL ‘I carried it.’ c. the causative root surfaces unscathed – a-t-ßn [make-1-REAL]. gina tsya d˙baq hul˙ a-g-u he tea later cook make-3-IRR:SG ‘He/she will cook tea later.’ b.’ b.’ The difference between the verb in (247b) hul˙ ˙-g-u and its active counterpart in (247c) hul˙ a-g-u is so minimal that the contrast may depend more on contrastive syntax than anything else – the “passive” version is without an agent.’ / ‘will cook later’ c. CLASS II TRANSITIVE tsi hul˙-d-ßn < hul˙ a-t-ßn I cook-1-REAL ‘I cooked it.’ d. but this is almost expecting too much of the distinction between [˙] and [a] where the distinction is almost non-existent elsewhere.’ (246c). the two vowels tend to coalesce. as in (246d).’ b. The construction is rare in our corpus and needs more research. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (246) a.2. • Benefactive light verb. not a-. MAKE (CAUSATIVE) tsi a-t-ßn I make-1-REAL ‘I made it. however. BENEFACTIVE FORM: nu tsi-wi ukæ˙ e-n-i you 1S-house ruin give-2-PAST ‘You ruined my house for me. CLASS II TRANSITIVE tsi dzo -a-t-ßn I wet-do-1-REAL ‘I soaked it.

as shown in Figure 2: 93 . BENEFACTIVE FORM: ni-ip˙n abi e-d-u 2S-corn carry give-1-IRR:SG ‘I’ll carry your corn for you.2.2 Intransitive–transitive clauses Morphologically. tok nikæ e-d-˙k we laugh give-1-PL:IRR ‘We will laugh (at him). The major morphological differences for verbs. There is no object marking. there are few differences between transitive and intransitive verb types.’ as in the following: (250) a. though there are no morphological differences between transitive and intransitive Class I verbs. lie in the distinction between Class I and Class II – those with prefixed or suffixed person marking.’ b. then. We have already seen some of those differences in §5. there are minor differences between transitive and intransitive Class II verbs. Thus. Both are indexed for a single referent – the subject referent.’ b.5. in terms of morphological differences.5.1. BASE FORM: ni-ip˙n abi a-d-u 2S-corn carry do-1-IRR:SG ‘I’ll carry your corn.1 and §5. However.’ 6. the Class I–Class II distinction is primary.2).5.Himalayan Linguistics: Archive No. For Class I there are both transitive and intransitive verbs.’ There are at least two verbs in the language whose transitive morphology is based on the verb ‘to give’ (in much the same way that Class II transitive verb conjugations are based on the verb ‘do’). and for Class II there are both transitive and intransitive verbs. The differences between Class II transitive and Class II intransitive verbs lie in person marking patterns. This may be because both verbs imply a dative goal – ‘write to someone’ and ‘laugh at someone. tsi haq lekæ e-d-i I letter write give-1-PAST ‘I wrote a letter. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (249) a.1. The differences have implications for number marking as well (see §5. and the transitive–intransitive distinction is secondary. respectively.

Watters: Kusunda Grammar. person marking distinctions developed in these ambiguous cases. Person agreement inflections in the Kusunda verb 6. SECOND PERSON: nu sip-ni-n you enter-2-REAL ‘You entered.’ 94 . immediately before the realis marker.2.2.’ b. As we have already seen in example (146).) third person suffix (+ realis) first person suffix (+ realis) second peron suffix (+ real. secondary development.’ Intransitive suffixing pattern 2 b.) third person suffix (+ realis) –– intrans Class II — –– trans Figure 2. very nearly a copy of the free pronoun. The distinction between persons lies only in the free pronoun: (251) a. FIRST PERSON: tsi b˙l-ßn I descend-REAL ‘I descended. there is no person marker. Class II intransitive verbs are bereft of person marking – the suffix -ßn shown in Figure 1 is a realis marker. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– –– intrans Class I — — trans tsngtsng-(ßn) / -ts-(in) -(ßn) / -n-(in) -dzi / -(ßn)-dzi -t-(ßn) -n-(ßn) -g-(˙n) first person prefix second person prefix third person prefix first person prefix second person prefix third person prefix first person suffix (+ realis) second person suffix (+ real. SECOND PERSON: nu b˙l-ßn you descend-REAL ‘You descended.’ b. FIRST PERSON: tsi sip-tsi-n I enter-1-REAL ‘I entered.’ a. is suffixed to the root.2 In what was probably a later. as in: (253) a. analagous to the distinctions found elsewhere in the grammar. FIRST PERSON: tsi sip-ßn I enter-REAL ‘I entered. SECOND PERSON: nu sip-ßn you enter-REAL ‘You entered.’ (252) 6.1 Intransitive suffixing pattern 1 In the most basic pattern. a pronominal marker.

We have not recorded “suffixing pattern 3” (§6. as in: (254) a.2.3 1 6. is usually elided. This needs further verification. Except in very carefully articulated speech.’ etc. the verb ‘give’ is identical to Class II transitive verbs (see Figure 1).4 Bitransitive verbs The only convincing bitransitive verb in Kusunda is the verb ‘to give. hul˙-d-ßn and hul˙-n-ßn (the basic forms) are pronounced identically – as hul˙-d-ßn – giving rise to the need for disambiguation. d~aba-n-ßn ‘You found it’. the only exception being the verb root itself.3 Clausal syntax Syntax in Kusunda follows a basic SV. unda-n-ßn ‘You showed it. because the person and number of the subject (S or A) referent is marked in the verb.2. the subject is usually excluded from the scope of assertion (Givón 1984. Except in careful speech.’ b.’ The triple -n-ßn-n in (255b).’ with all other bitransitive notions being benefactive constructions that utilize the verb ‘to give.Himalayan Linguistics: Archive No. Where the subject NP does occur. though they may exist.’ b. 6. The suffixing pattern described in §6. the verb coming last. This is true of all second person forms in Class II transitive verbs: pumba-n-ßn ‘You hit him’.3 is also available to Class II transitive verbs.’ Morphologically. This is in keeping with universal expectations that in unmarked clauses.2. 1991). the subject NP. TRANSITIVE FIRST PERSON: tsi hul˙-d-ßn-tsi I cook-1-REAL-1 ‘I cooked it. all are pronounced with a -d.’ • Pattern 3 in transitive verbs. a full copy of the free pronoun is tacked on to the end of the verb complex.3 Suffixing pattern 3 In yet another development. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 6. m˙b˙-n-ßn ‘You heard it’. TRANSITIVE SECOND PERSON: nu hul˙-n-ßn-nu you cook-2-REAL-2 ‘You cooked it.3) for all these verbs. is a bit of phonological fiction. though morphologically accurate. AOV constituent order. SECOND PERSON: nu sip-ßn-nu you enter-REAL-2 ‘You entered.2. In Kusunda. FIRST PERSON: tsi sip-ßn-tsi I enter-REAL-1 ‘I entered. as in the following: (255) a. This order is typologically pervasive throughout South Asia. even as a pronominal form. 31 95 . especially in connected speech. following the realis marker.

unmarked order. ‘Climb. occur closer to the verb than the pronominal referent.) 96 .’ for example.’ ‘beg. however. (The order in (256). the notional object is marked by a locative ‘on’ for ‘climb’ and a comitative ‘with’ for ‘beg’ and ‘fight’. Thus.2.2. then. INTRANSITIVE: goraq tsi da≥ t -a -an tomorrow I Dang 1-go-IRR ‘Tomorrow I will go to Dang. for an earlier question: (256) tsi t -a -an I 1-go-IRR ‘I’ll go.’ (That’s what I’ll do) In general. All happen to be non-accusative in case marking. For more details and examples. There are a few transitive verbs in Kusunda which appear to be “less-than-transitive” – they have some of the morphosyntactic characteristics of intransitive verbs. say. that is. Case marking in Kusunda follows a basic nominative–accusative alignment – S and A arguments are unmarked (nominative) and the O argument is marked by the accusative suffix -da (see example (258b)). all take intransitive imperative markers.’ b. see §5. or it serves as an anchor.) This is relevant to clauses that have peripheral elements like time or location. Basic clause types ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– it is either contrastively marked.’ • Case marking.’ c.’ (I’m volunteering) ‘I’ll go. (See the discussion following example (196).’ Where both peripheral elements occur – time and location – the element with closer semantic ties to the verb occurs next to the verb.Watters: Kusunda Grammar. time. TRANSITIVE: pyana tsi agai-da pumba-d-i yesterday I dog-ACC beat-1-PAST ‘Yesterday I beat the dog. and temporal elements which occur outside the case frame also occur syntactically outside: (258) a. This order mirrors the basic order for transitive predications too – the object occurs next to the verb. in (257a–c) the peripheral elements. Peripheral elements tend to attract the focus of assertion. place. tsi da≥ t -a -an I Dang 1-go-IRR ‘I will go to Dang. and the other element occurs as an adverbial specification to the whole predication.’ b.’ and ‘fight. and comitative. This is the least marked order: (257) a. gina tok-ma u-g-i he/she us-WITH come-3-PAST ‘He/she came with us. more topical elements occur early and less topical ones later.2. tsi pyana ts-˙g-˙n I yesterday 1-go-REAL ‘I went yesterday. is also a basic.

and the intransitive member is derived by a process of anti-causativization. With intransitive affixation they are intransitive verbs. TRANSITIVE tul-a-t-ßn dress-TR-1-REAL ‘I dressed.1 Morphological causative -a In §5. the language which surrounds Kusunda on all sides is the opposite – most events. How such things are lexicalized is a matter of cultural specific construal and depends in some way on whether the eventuality named by the verb can be conceptualized as occurring without an agent.’ Recall that most transitive verbs.1.2. In some cases the basic. 97 . Such verbs have been classified primarily as Class II intransitives (see Figure 1) for the simple reason that the transitive forms closely resemble the verb ‘to do.5. make’ still occurs and will be treated in §9. INTRANSITIVE tul-tsi-n dress-1-REAL b.1.’ tacked onto the intransitive root. take the form of the structure in (259b). It is not immediately obvious from the general semantics of a verb whether its inherent status will be transitive or intransitive. with morphological rules for deriving one from the other. including many which are not causativized. and reflexives There are a number of derivational operations by which intransitive verbs can be made transitive. one member of which is transitive and the other intransitive. 7. and transitive verbs made intransitive.7.2. leaving periphrastic causatives and other syntactic devices to §9. This suggests that they originally derive from a periphrastic causative construction. In Kham. in which case the transitive member is derived by a process of causativization. even the ‘break’ type ones.) The example we gave in (147) is representative and is repeated here as (259): (259) a. for example. unmarked form is intransitive. Transitivity alternations – causatives.’ ‘I dressed (someone). make. and with transitive affixation they are transitive verbs. are lexicalized as inherently spontaneous intransitive events and are made transitive only by causativization. The former process increases the valence of the verb and the latter decreases its valence. anticausatives.2 (in the subsection on Transitive–intransitive distinction) we saw that some verbs appear to be ambi-transitive and can take either transitive or intransitive affixation patterns. In other cases the base form is transitive.1 Causative There are a number of verbs in Kusunda that occur in alternating pairs. In Kusunda. the ‘break’ type verbs are inherently agentive and can be made intransitive only by derivation. We will deal here only with morphological operations. 7. (A periphrastic causative using the verb ‘to do.

TRANSITIVE o -da-d-i die-CAUS-1-PAST ‘I slept. Prefixed person marking is lost: (260) a. after derivation they become Class II verbs – i. the verb ‘kill’ illustrated in (261b) is very often pronounced ora. Such verbs can be made transitive either by the morphological causative -a or by the morphological causative -da. There is a great deal of variation between the two forms.Watters: Kusunda Grammar.’ ‘I killed him. but both form the causative by -da. INTRANSITIVE ts-ip-ßn 1-sleep-REAL b.2 Morphological causative -da Most intransitive verbs can be made transitive by adding the morphological causative suffix -da. But so also do ‘fall’ and ‘wear’ end in consonants.‘kill.1. in the following examples. having suffixed person marking.’ ‘I put him to sleep. Where such verbs belong inherently to Class I. The verbs ‘sleep’ and ‘die’ both end in consonants and the causative is formed by -da. INTRANSITIVE hyoq-tsi-n hide-1-REAL b. there may be a difference in “force” for “directness” of causation). though. ‘Scorch’ and ‘swell’ end in vowels. but the causative is formed by -a (though -da is also possible with no apparent difference in meaning. TRANSITIVE I hyo -a-t-ßn hide-CAUS-1-REAL c. INTRANSITIVE t-o -˙n 1-die-REAL b.e. The choice of one over the other does not seem to be based on phonological factors or the syllable type of the root. TRANSITIVE ip-da-d-i sleep-CAUS-1-PAST (261) a. Transitivity alternations ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 7. Thus.’ ‘I died.’ i.’ Note that the causative of ‘die. of course. having prefixed person marking.’ ‘I hid (something).e. the (a) sentences are inherently Class I verbs and their transitive (b) counterparts are Class II. Following are verbs which can take either form: (262) a.e.’ ‘I hid (something).’ For Class II intransitive verbs there is no shift from prefixing to suffixing under causativization – they begin with suffixed person marking.’ 98 .and is an example of a verb developing what may someday become a suppletive pair whose developmental pathway is opaque – -o ‘die’ versus ora. TRANSITIVE II hyo -da-t-ßn hide-CAUS-1-REAL ‘I hid. i.

Such causatives are typologically common in the area. in addition to a causative morpheme.’ ‘I rescued (him).) (265) a. TRANSITIVE I sip-a-t-ßn enter-CAUS-1-REAL c.’ Some intransitives when causativized yield unexpected meanings. In such verbs the “middle” or “anti-causative” <AC> is marked by the suffix -q.’ The implication is that the subject is ultimately responsible for the eventuality. The pair ‘to survive’ versus ‘to rescue’ is one such pair (from ˙igi. suggesting the possibility of contact induced meanings. The ‘break’ verbs are among them. even if only by neglect or inadvertence.’ but rather ‘laugh at’ and ‘say to. in both TB and IA languages: (266) tsi yi-da qau-da-d-i I wood-ACC rot-CAUS-1-PAST ‘I rotted the wood.’ Some verbs.’ 7. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (263) a. as in the following sets of contrastive pairs: 99 .’ Causativized by the morphological causative they do not mean ‘make laugh’ or ‘make say.to ˙gi-): (264) a. ˙igi-d-ßn live-1-REAL b. gin-da nikæ-e-d-i he-ACC laugh-CAUS-1-PAST ‘I laughed at him. Two examples are the verbs ‘to laugh’ and ‘to say. he cries.Himalayan Linguistics: Archive No.’ b. INTRANSITIVE sip-ts-in enter-1-REAL b. manifest some phonological change in the root morpheme itself.’ ‘I inserted (something). n˙tßn gip˙n ts-o -da-d-ßn dzæ˙m-in-dzi what word 1-say-CAUS-1-REAL weep-<??>-HAB ‘No matter what I say to him.’ It should also be noted that morphological causativization in Kusunda works even on intransitive “process” verbs which would normally require a periphrastic causative in English – verbs like ‘rot.’ ‘I inserted (something).’ (Similar semantic shifts occur in Nepali. TRANSITIVE II sib-da-d-ßn insert-CAUS-1-REAL ‘I entered. ˙gi-da-d-ßn live-CAUS-1-REAL ‘I survived.2 Anti-causative The anti-causative is a detransitivizing derivation on inherently transitive verbs yielding intransitive and middle senses.

followed by the morphological causative -a. TRANSITIVE: suta bæ˙ra-g-˙n string break-3-REAL ‘(He/she) broke the string. occurring on verbs whose inherent status is middle. We will not separate the -q in these cases.’ (269) a. the marker is semantically motivated. but. -q is not a derivational marker per se. Transitivity alternations ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (267) a.’ b. the middle marking remains (often becoming . TRANSITIVE: tsi ip˙n sola-d-ßn I corn break-1-REAL ‘I broke (harvested) the corn. but treat it as part of the underlying root. as in the following: (270) a. we will mark them as “middle marked” <MM>.’ b.’ b.) Under causativization. INTRANSITIVE (MIDDLE): ip˙n gimdzi sola-q-ßn corn self break-AC-REAL ‘The corn broke by itself.’ 100 .Watters: Kusunda Grammar. ta≥ byaq-ßn water spill-REAL ‘The water spilled.2. TRANSITIVE: tsi k˙reÚ k˙la-d-ßn I jug break-1-REAL ‘I broke (shattered) the jug. Rather. there are no corresponding transitives without the -q.’ 7. INTRANSITIVE (MIDDLE): k˙reÚ k˙la-q-ßn jug break-AC-REAL ‘The jug broke. INTRANSITIVE (MIDDLE): na suta bæ˙ra-q-ßn lan this string break-AC-REAL like ‘This string is like it will break. In such cases.’ b.1 Anti-causative -q versus lexical -q tsu is Somewhat problematic are intransitive verb roots in which a final -q is part of the underived lexical representation. recognizing that we are not dealing with a derivation. in fact. (If we do find occasion to dissect such forms.).’ (268) a. such verb roots look like derived middles. On the surface. tsi wi gya -a-d-ßn I house collapse-CAUS-1-REAL ‘I collapsed/razed the house.’ (271) a. wi gyaq-ßn house collapse-REAL ‘The house collapsed.

gag˙ri-ga ta≥ pæiru-t-ßn water fill-<T>-REAL jug-LOC ‘Water filled in the jug. tsi soq-ßn I climb-REAL ‘I climbed. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. ta≥ byaq-ßn water spill-REAL ‘The water spilled. in (272) -q is lexical and the transitive form is derived.Himalayan Linguistics: Archive No.’ 7.‘climb. There are numerous intransitive verbs.2.’ b. dimi n~odze gwi-t-ßn smoke above gather-<T>-REAL ‘Smoke (soot) gathered above. tsi ta≥ bya -a-d-i I water spill-CAUS-1-PAST ‘I spilled the water.’ b.’ <— detransitivized by -q Another verb with lexical -q in the intransitive member is soq. however.’ b. -t followed by syllabic -ßn signifies first person followed by realis.’ as in: (274) a. “Is -t part of the lexical root or is it a grammatical morpheme?” We know already that -t here is not 101 .2 Anti-causative -t More difficult to interpret are intransitive verb roots which end in -t.’ b.’ (273) OPEN SYLLABLE: <— transitivized by -a a. In the former case. tsi k˙reÚ k˙la-d-ßn I jug break-1-REAL ‘I broke (shattered) the jug. Thus.’ It is important to notice the contrast between verbs whose underlying root ends in -q and verbs whose underlying root is an open syllable. k˙reÚ k˙la-q-ßn jug break-AC-REAL ‘The jug broke.’ We have glossed -t in (275a–b) with a tentative <T>. and in (273) -q is derivational and the intransitive form is derived: (272) UNDERLYING -Q: a. tsi ta≥ bya -a-d-i I water spill-CAUS-1-PAST ‘I spilled the water. in which a stem final -t followed by syllabic -ßn does not indicate first person: (275) a. Recall that in many transitive verb conjugations (like a-t-ßn ‘I made’). whereas in the latter case -q occurs only in the intransitive. -q (or . yi-g˙ soq-a-d-ßn tree-IN climb-CAUS-1-REAL ‘I treed him. The question is.) occurs in both transitive and intransitive forms.

’ and gyaq-tsi-n ‘I collapsed.’ Significantly. of “floating” -t in our data base. we need to compare these forms with their causativized counterparts: (276) a.’ We have numerous other examples. This strengthens our hypothesis that -t is not part of the intransitive root: (278) a. thus demonstrating that it too is an element divisible from the verb root. separating the -t from the verb root. hampe t -a -an kæai-tsi-t-ßn 1-go-IRR fear-1-AC-REAL where ‘Wherever I go I get scared. of course. gag˙ri-ga ta≥ pæir-a-t-ßn water fill-CAUS-1-REAL jug-LOC ‘I filled water in the jug. b˙l-n-in descend-2-REAL c.’ 102 . tsi d˙idzi gwi-˙-d-i I firewood gather-CAUS-1-PAST ‘I gathered firewood.’ c. be satisfying to our claims if anti-causative -q also floated to the right of person markers. It would.’ The stem final -t does not occur in the causativized version (*gwi-t-a-t-ßn) and we are left to conclude that -t is an anti-causative morpheme analogous to the -q we saw earlier. the -t occurs after person marking. Transitivity alternations ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– first person.’ ‘You descended. too. where the same optional person marking is employed in intransitive verbs with anti-causative -t.’ ‘I fell. In §5. Following the same line of reasoning we used above.’ ‘You hid.2 (subsection intransitive patterns) we saw that Class II intransitive verbs have the option of including a person marker immediately after the root-final consonant: (277) a.Watters: Kusunda Grammar. nok gwi-ni-t-n-a-n you[pl] gather-2-AC-2-PL-REAL ‘You all gathered. Where -t occurs it is always anti-causative <AC>. In fact we have no data in our corpus to prove this.’ b.1. that we have no verbs in our corpus with root final -t. It turns out.’ b. un-da ts-˙g-˙n miÚ-tsi-t-ßn road-LOC 1-go-REAL lost-1-AC-REAL ‘Going along the road I got lost. and as our claims would predict. in fact. but we will not list them all here.5. pæep-ts-in fall-1-REAL d. (Nor do we have anything to disprove it. sip-ts-in enter-1-REAL b. hyoq-n-in hide-2-REAL ‘I entered. -q remains attached to the root. as in: borloq-tsi-n ‘I got boiled.) We have examples of person marking only on verbs in which -q is part of the root.

but the resemblance may be accidental.’ Note that (279a) is made intranstive by virtue of the <AC> marker -q.2. and appears to be detransitivized by the inchoative -in. ACCOMPANIED BY AN ANTI-CAUSATIVE: dza h~oÚ-˙-go borlo-q-in-dzi fire burn-CAUS-IMP boil-AC-INCHO-TAM (?) ‘Stoke the fire.1 “Passive” in ambi-transitive verbs It is difficult to say unequivocally whether some verbs are inherently transitive or intransitive. it might indicate that the second clause event is a consequence of the first clause event. The suggestion warrants looking into. rises’.2. anticausative derivation discussed in §7. It might be equally plausible to think of -in as the intransitive (inchoative) version of ‘to do. roast’ is one such verb. ON AN INTRANSITIVE: n˙n gip˙n ts-o -da-d-ßn dzæ˙m-in-dzi what word 1-say-CAUS-1-REAL weep-INCHO-TAM (?) ‘Whatever I say to him. it (the water) will boil. appears to be available only to select transitive verbs. On Class II transitive verbs.” Following are examples of its usage (but with no attempted explanation on how the construction works)3 2: (279) a. (279b) begins as a transitive verb. The closest analogue is the detransitivizing. squats. We will label it as <INCHO> for “inchoative. meÚ-in-dzi ‘He gets lost’.” The derivation. Another structure which is not well understood applies to transitive and intransitive verbs alike.’ b. We have several examples of intransitive verbs marked by this construction. -in. he cries. the “auxiliary” ending ‘to do.3 Passive There is no canonical passive in Kusunda (of the English sort) in which the former object is promoted to subject status. see example (265b)). dæ~u-in-dzi ‘He stands.2 which produces a kind of “middle. WITHOUT ANTI-CAUSATIVE: qai u-g-i ip˙n sola-in-dzi wind come-3-PAST corn break-INCHO-TAM (?) ‘Wind comes and the corn breaks. make.’ is replaced by a sequence -in-dzi. The first affix.marker in all the examples listed occurs on the second clause of a complex sentence. but are uncertain about how the meaning differs from ordinary intransitive conjugation without -in: kæai-in-dzi ‘He is afraid’. -Dzi occurs elsewhere as a “present tense” marker (also not well understood. by~o-in-dzi ‘He sits.’ 7. and the intransitive version is passive-like. as in the following contrastive pair: My anonymous reviewer has suggested that because the -in.Himalayan Linguistics: Archive No. 32 103 . however. resembles the Nepali passive marker -in. The verb ‘to burn.’ c. and (279c) is intransitive already.3. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 7. make’ (see (244b)). They can be conjugated as transitive or intransitive without mediating derivational morphology.

’ Many third person realis forms. today 1-eat-REAL ‘Yesterday it was cooked. the most neutral inflection is a generic third person prefix (see §8. however.2. as in the following verb used in a modifying function: (282) g-˙m-˙n kæaidzi 3-eat-REAL food ‘eaten food’ / ‘food eaten by (someone)’ In like manner.1. and §9. but relies on the reflexive pronoun gimdzi ‘self. (On this topic see also sections §4. independent verbs in coordinate structures (see §11.’ The reciprocal uses the same mechanism. as in (284).Watters: Kusunda Grammar. t˙~îna t-˙m-˙n yesterday cook-3-REAL. as in the following: (281) ts-i-da-n tok hyoq-ßn we hide-NEUT 1-go-PL-REAL ‘We went to hide.’ In Class I verbs. TRANSITIVE: tsi amba dza-g˙ hab˙-d-ßn I meat fire-IN burn-1-REAL ‘I roasted meat in the fire.2 A “functional” passive In §8.) (284) pyana hul˙-n. illustrated in (280b).1.2. as in: (283) pyana hul˙-g-˙n.’ 7. today I’ll eat it. today 1-eat-REAL ‘Yesterday it was cooked.3 Reflexive The reflexive construction in Kusunda does not require detransitivizing morphology in the verb. is passive in force and must be translated in English as ‘got burned.’ The intransitive version. In such cases.1 we will discuss verbs that occur without person marking – our so-called “neutral” inflection. neutral inflection on the subordinated verb indicates an identity of person (same subject) with the matrix verb. today I’ll eat it.3).4. In subordinate structures like the “purposive” clause. INTRANSITIVE: dza-g˙ ts-˙g-˙n hab˙-tsi-n fire-IN 1-go-REAL burn-1-REAL ‘I went into the fire and got burned.1.3. examples (63–64). as in: 104 .’ 7. pronounced as -˙g˙n by Gyani Maiya.1. it is not normally possible to be rid of all person marking – person marking occurs as a prefix.’ b.4. Transitivity alternations ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (280) a. but with two occurrences of the reflexive pronoun. are pronounced simply as -˙n by Kamala.1) can be interpreted as having passive force where the medial “passive” verb is inflected as a generic third person and the final verb is specific for person. t˙~îna t-˙m-˙n yesterday cook-REAL.

’ This is not to say that detransitivized constructions are not compatible with the reflexive pronoun. gagri pæiru-t-ßn jug fill-AC-REAL ‘The jug filled. the construction requires a fully transitive verb. as in the examples of (285). as in: (287) a. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (285) a. If. tsi a-d-aÚ u.’ The same reflexive pronoun is used to indicate exclusive ownership. gina gimdzi gimdzi wi-nu t-ei-dzi he/she self self house-TO go-PL-TAM (?) ‘They each go to their own houses. they are.Himalayan Linguistics: Archive No. nu ni-gimdzi wa ¯-a -an home 2-go-IRR you 2S-self ‘You will go to your own home.’ ANTICAUSATIVE WITH REFLEXIVE PRONOUN: b. ip˙n gimdzi suml-˙-q-ßn break-TR-AC-REAL I do-1-NEG. RECIPROCAL: gimdzi gimdzi o -da-˙-g-˙i self self die-CAUS-TR-3-PAST ‘They killed each other. however. the speaker wishes to express that the agent acted on him/herself. gagri gimdzi pæiru-t-ßn jug self fill-AC-REAL ‘The jug filled on its own.’ 105 .’ c. corn self ‘I didn’t do it. the reflexive pronoun only signifies that the eventuality named by the verb happened without outside agency (not that it acted upon itself): (286) ANTICAUSATIVE WITHOUT REFLEXIVE PRONOUN: a. Used with intransitive constructions. REFLEXIVE: gimdzi pæo-˙-g-˙n self beat-TR-3-REAL ‘He hit himself.’ b. inherent or derived. the corn broke by itself.’ b. in which case the reflexive pronoun precedes the possessed item.

1 In Class II intransitive verbs One verb type in which “no marking” is possible is in Class II intransitive verbs.1).1 Neutral forms We will define a neutral form in Kusunda as having no person inflection. 8. tul-ßn wear-NEUT cloth ‘worn clothes’ This may not be the same -da.2. Here we will explore the basic morphological structure of subordinate structures and talk later (in §9) about how they function within hosting structures. the subordinated verb root is marked by a suffix -da. In one.1. a state of affairs that is modified in finite contexts (see §6. Languages of the region. purposive clauses and the like – are cast in one of three basic construction types. sentential complements. Neutral forms occur in subordinated clause environments (or as modifiers) where person–number and tense–aspect specifications are not relevant – such specifications are marked in the matrix verb. a causative -da. and here a subordinate -da.2. Such verbs are marked only with a suffix -n (variant -˙n). the neutral form itself is a nominalized form). although it might be argued that the presence of -da is sufficient evidence to assume some kind of nominalization – -da is also an NP object marker (see §5. wi-g˙ house-LOC enter-NEUT go-OPT ‘Let him go enter the house!’ ts-i-da-n b. in which case -da marking is still an available option (see §8. there will necessarily be some overlap. as described in §6. Though presented from different points of view.3 3 Lack of nominalized structures makes Kusunda typologically exceptional. 8. Morphology of subordinate structures Clauses which are subordinated or embedded in other structures – adnominals in relative clauses. of course. both Nepali and TB languages like Kham. an aspectual -da. tok hyoq-ßn we hide-NEUT 1-go-PL-REAL ‘We went to hide.2. nominalize all sorts of embedded clauses. and in a third type the subordinate verb is fully finite. None of the structures appear to have any kind of nominalizing morphology. Such forms are not “citation” forms. the verb is cast in a “neutral” non-finite form without person marking. but which is well suited for non-finite contexts. which may or may not be the realis marker -n. Here we will gloss it <NEUT> for “neutral. In Class I verbs neutral marking is not always possible. 33 106 . periphrastic causatives.2. there is an accusative -da. a plural -da.1.8. nor do they include any kind of nominalizing morphology (unless. in another.2.” as in the following: (288) sip-ßn g-ya a.’ p˙idzi c.2). and in some TB languages there are multiple types of nominalization. -Da occurs in many contexts in Kusunda.2 for the modification). Recall that such verbs are bereft of person marking.

1.2 In Class II transitive verbs Class II transitive verbs. it is difficult to know if the so-called neutral forms are only contracted third person realis forms – ˙-g-˙n (do-3-REAL) > -˙n. In such cases. a third person form is used (somewhat analogous to the form in (290) in which third person seems to be understood as generic. (For less finite forms for some of these verbs see §8. we will reserve the label “neutral” to those forms that are without third person marking (as in (289)). can occur in a neutral form.’ Forms like the one in (290) are problematic in that they use a third person form in the complement (‘to beg’). as in: (291) g-˙m-˙n kæaidzi 3-eat-REAL food ‘eaten food’ / ‘food eaten by (someone)’ Bare roots (concatenative structures) 8.Himalayan Linguistics: Archive No. Many third person realis forms.’ 107 .3 In Class I verbs Class I verbs are the least amenable to non-finite conjugation for the simple reason that such verbs normally include a prefixed person marker. Thus. as in: (289) a.’ ts-i-d˙-i c. a form technically incompatible with first person in the matrix verb (‘I came’). pronounced simply as -˙n by Kamala. The third person -g tends to surface more readily in open syllable roots: (290) t-ug-un kæaidzi ai-g-˙n food beg-3-REAL 1-come-REAL ‘I came to beg for food. are pronounced -˙g˙n by Gyani Maiya.’ t-ug-un b.) 8. This possibility is apparently lexically determined and is not available for all verbs: (292) kola t-ug-un a. This is especially true if the verb root ends in a consonant. although forms like ai-g-˙n.2.1. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 8. in a concatenative structure. tsi amba hab-˙n I meat roast-NEUT 1-come-REAL ‘I came to roast meat.2. tok kæaidzi dza -˙n ts-i-da-n we food buy-NEUT 1-go-PL-REAL ‘We went to buy food. without the final -n. too. ta≥ water draw 1-come-REAL ‘I came to draw water. amba pad-˙n meat hunt-NEUT 1-go-PL-HORT ‘Let’s go hunting (for animals). especially where the form is incompatible with third person. for the most part. too.4 There are a few verbs for which only the bare root occurs in neutral constructions. are neutral or generic in function.1.’ In these Class II transitive verbs.

occurs in its inflected forms with prefixed person marking – ts-ip-ßn ‘I sleep’.’ An alternative is to identify -da with the object marker -da. the verb ‘to sleep. g-ip-ßn ‘He sleeps.’ n-ug-un b.2. “non-inflecting verbs”) – here followed by a conjugated form of the verb ‘to make.2.’ in which case it would be analogous to purposive constructions in some TB languages.’ not unusual in the world’s languages. n-ip-ßn ‘You sleep’.) Interestingly. -da does not occur on relatives. all Class II transitives modelled after the verb ‘to do’ are a special case of bare-root neutral verbs (i.’ u-g-i c.’ Likewise with the verb ‘to eat’: 108 . the desiderative. gina ip-da he sleep-PURP come-3-PAST ‘He came to sleep.2). as in the following Kham example: (294) ≥aÚ zya-na ≥a-hu-ke I eat-GO 1-come-PFV ‘I came to eat. however.2 Embedded structures marked by -da There are several subordinate clause types – like the purposive. and non-implicative causatives – that are marked by the suffix -da on the embedded verb. occurs in this “pure” form (we will see modified forms later) on Class I verbs (i. Morphology of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– t-˙m-du b. to do.’ In the non-finite -da structure. tsi ip-da I sleep-PURP 1-come-REAL ‘I came to sleep. (Ultimately. of course. We will refer to this morpheme as the “purposive” <PURP> morpheme. the two may be etymologically related.1 Non-finite forms marked by -da Some types of subordinate clauses are marked by the purposive morpheme -da suffixed to the non-finite.Watters: Kusunda Grammar.’ for example.e. Thus.’ 8. however.’ In a sense.’ Following is an example of a typical Class II verb: (293) qaÚi u-g-i ip˙n suml˙ wind come-3-PAST corn break ‘Wind comes and breaks the corn. This -da may be related to the verb ‘go. the object marker itself having derived from ‘go. hol cook 1-eat-IRR:SG ‘I’ll cook and eat. bare root of the embedded verb. but is exclusive to complements.occurs with -da suffixed: (295) t-ug-un a. nu ip-da you sleep-PURP 2-come-REAL ‘You came to sleep. only the bare root ip.’ ˙-g˙-n-dzi do-3-REAL-TAM (?) 8. and only after the person marking prefix has been stripped away (an exception is given in §8. those with potential prefixed person marking).e. and it is in these kinds of contexts that -da ‘go’ begins to specialize as -da ‘object. The -da suffix.

’ Inflected forms marked by -da 8. tsi am-da I eat-PURP 1-come-REAL ‘I came to eat. the structures described in this section are available to both Class I and Class II verbs.2 that the -˙n “neutral” forms for Class II transitive verbs may. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (296) t-ug-un a.5.’ u-g-i c.1.Himalayan Linguistics: Archive No. It is as follows: (299) ai-da t-ug-un beg-PURP 1-come-REAL ‘I came to beg.as follows: (298) ˙n-da ~oˆ la-t-ßn a. 109 .1. ‘To drink’ is another such verb: (297) tsya qon-da t -a -an tea drink-PURP 1-go-IRR:SG ‘I will go to drink tea. and that -da occurs invariant. there is also a less finite structure for the verb ‘to beg’ than the one illustrated in example (290) of §8. Where the inflected forms of the verb are a-d-i ‘I made’. in fact.2.2. nu am-da you eat-PURP 2-come-REAL ‘You came to eat. a-n-i ‘You made’. Furthermore.5.2.2. For at least some verbs of that class there is also a suppletive form of the verb which combines with -da to form a non-finite structure.’ Non-finite Class II verbs marked by -da 8.’ Likewise. gina am-da he eat-PURP come-3-PAST ‘He came to eat. ˙-g-˙n ‘He made’.’ t˙mb˙-d-i b. make’ is one such verb.’ am-da t-ug-da-k d. Both of these points will contrast with “inflected -da structures” in §8. The transitive verb par excellence ‘to do.3. tsi wi I house make-PURP want-1-REAL ‘I want to build a house. the non-finite form is ˙n.2 Recall from §8. no-k kæaidzi hul-ak to-k cook-IRR we-PL eat-PURP 1-come-PL-IRR you-PL food ‘If you cook the food. The bare root forms for these Class I verbs are the same suppletive forms we saw earlier in §5. we’ll come to eat it.3 The forms in this section will contrast with those described in §8. be generic third person forms whose full form is -˙g˙n. as -da throughout.2.1 (Irregular and suppletive forms). kam ˙n-da work do-PURP send-1-PAST ‘I sent him to do work.’ The important thing to notice in examples (295) and (296) is that the embedded verb occurs without person agreement prefixes.’ n-ug-un b.1.

Thus. as we had in §8.1 can also appear in their person-inflected forms.’ n-ug-un c. n-i-di ‘You fetched it’. gina g-˙m-da he 3-eat-PURP come-3-PAST ‘He came to eat. the forms are inflected for first and second persons – -d-a and -n-a.’ c. g-i-di ‘He fetched it. this form contrasts with the uninflected -da forms in the following way: (300) UNINFLECTED -DA: a. tsi am-da t-ug-un I eat-PURP 1-come-REAL ‘I came to eat. nu n-˙m-n-a you 2-eat-2-PURP 2-come-REAL ‘You came to eat. Also. second. respectively. With the verb ‘to fetch.’ Some Class I verbs have suppletive roots that are used with the -da construction. the third person form of ‘eat’ includes a third person prefix g-. and (302b) shows the suppletive form: 110 .2. all forms take the second person -na form of the purposive and a third person g. g-i-na u-g-i 3-fetch-PURP 2-come-REAL ‘He came to get (it). such that -d-a marks first person environments and -n-a marks second person environments.Watters: Kusunda Grammar.2. tsi t-˙m-d-a t-ug-un 1-eat-1-PURP 1-come-REAL I ‘I came to eat. The embedded verbs that we saw in §8. Morphology of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– • On Class I verbs. We get ip-da but not *g-ip-da.is a person prefix can be seen in the following inflections: ts-i-di ‘I fetched it’. Thus. Third person reverts back to -da (there is no -g-a). such a form is disallowed.’ In (300b) and (300c) the purposive marker is no longer simply -da. The paradigms for these verbs are irregular. The verb ‘to go’ is one of them.’ INFLECTED -DA: b. g-i-na t-ug-un 3-fetch-PURP 1-come-REAL ‘I came to get (it). third person inflection for the purposive morpheme -da.’ That g.’ however. the form of the embedded verb includes prefixed person markers – t-. get’ there is no first.’ b. as is generally true for all Class I verbs. the purposive morpheme -da participates in a person agreement paradigm. and g-.’ u-g-i d. (302a) shows the base form as a benchmark.1 and also in (300a). n-. Rather. In such cases. g-i-na n-ug-un 3-fetch-PURP 2-come-REAL ‘You came to get (it). There is no apparent difference in meaning (though there certainly may be subtle differences). With the verb ‘to sleep. Rather.prefix on the verb stem irregardless of person: (301) a.

Thus. (An exception can be found in §8. we have seen the following distinctions: (304) CLASS I: CLASS II: -di -da -du -da 8. the following (a) and (b) sentences contrast in form. but we are unaware of any semantic differences: (305) a.2. tsi amba hul˙-d-a I meat cook-1-PURP 1-come-REAL ‘I came to cook meat. contrasting with second person -n and third person -g. and third persons. In Class I verbs.equative ‘I wouldn’t go there. the -da structure in Class II verbs is always an “inflected -da. or marked by -da.’ b. second.’ n-ug-un b. can also be marked as finite.2. sape ts-˙g-˙n there 1-go-REAL ‘I went there.” i. non-finite form. but part of the following morpheme. the -d that occurs in an identical phonological sequence is not the same -d.3 ‘past’ ‘plural’ ‘irrealis’ ‘purposive’ -d-i -n-i -d-a -n-a -d-u -n-u -d-a -n-a ‘first person–past’ ‘second person–past’ ‘first person–plural’ ‘second person ‘plural’ ‘first person–irrealis’ ‘second person–irrealis’ ‘first person–purposive’ ‘second person–purposive’ Finite embedded structures Many constructions whose embedded verb can be marked by a neutral.’ u-g-i c. sape ts-i-da odoq there 1-go-PURP neg. gina amba hul˙-g-a he/she meat cook-3-PURP come-3-PAST ‘He/she came to cook meat. tsi wi ˙n-da ~oˆ la-d-i I house make-PURP want-1-PAST ‘I wanted to build a house.’ • On Class II verbs. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (302) a. the morpheme has different representations for first.e. tsi wi a-t-ßn ~oˆ la-d-i I house make-1-REAL want-1-PAST ‘I wanted to build a house.’ 111 .Himalayan Linguistics: Archive No. nu amba hul˙-n-a you meat cook-2-PURP 2-come-REAL ‘You came to cook meat.) Here.’ b.’ Recall that in Class II verbs the suffix -d is almost always a marker of first person. So far. a third person -g-a is also possible: (303) t-ug-un a. To our knowledge.

4 Embedded structures marked by mutation We have noted already that mutation is a grammatical “mark” in Kusunda. 25. could also be interpreted to mean ‘I cooked food and came. especially Class II transitives. the greater the likelihood that two different events are being coded.Watters: Kusunda Grammar.’ c. In the following examples. wi a-d-i tumb˙-d-i house make-1-PAST finish-1-PAST ‘I finished building the house. t˙n-da wi-g˙ t -a -an ˙-g-i me-ACC house-LOC 1-go-SUBORD make-3-PAST ‘He made me go home.’ 112 . as in the following: (306) a. n˙n-da ¯-a -an e-g-i you-ACC 2-go-SUBORD give-3-PAST ‘He let you go.’ There are limits on the finiteness of the subordinated verb.’ Based on the same principle.’ b.’ / ‘I cooked food and came.’ b. and in §10 on co-subordinate structures. for example. *wi a-t-ßn tumb˙-d-i *house make-1-REAL finish-1-PAST We will deal more with the specifics of TAM interplay when we deal with specific constructions in §9.1. but also the marked member of the independent–subordinate dichotomy. Both verbs convey the same event: (307) a. kæaidzi hab˙-d-a t-ug-un food cook-1-PURP 1-come-REAL ‘I came to cook food. kæaidzi hab˙-d-ßn t-ug-un food cook-1-REAL 1-come-REAL ‘I came to cook food. §5. 8. The sentence in (306c). kæaidzi hab˙-˙n t-ug-un food cook-NEUT 1-come-REAL ‘I came to cook food. the more finite the embedded verb.’ c.5. but subordination: (308) a. the embedded verb in some constructions must be marked with the same TAM as the matrix verb. wi a-t-ßn tumb˙-d-ßn house make-1-REAL finish-1-REAL ‘I finished building the house.3. the mutated variants of n˙g˙n and ts˙g˙n do not indicate irrealis.5.’ b. Morphology of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– For some verbs. and fn. (See §5.4. there are three contrasing forms.) Here we see that the mark applies not only to the marked member of the realis–irrealis dichotomy.3. In general.

’ Realis marked adnominals are similar in structure: (310) a. and can be TAM-marked either as realis or past tense. apparently.’ In other languages of the region. 9. n˙n-da e-d-u-wa this I find-1-PAST money.1 Finite adnominals formed from verbs The adnominal modifier in Kusunda utilizes at least two of the structures we saw in §8 – they can be cast as “neutral. etc. both IA and TB. relative clauses. (Santali also. Following are examples of finite verbs in past tense functioning as modifiers within relative clauses: (309) a. Though such adnominals are formed from verbs and are modifying in function. a single category can be encoded in more than one way. in fact. 113 . In some cases. has fully finite verbs used in modification (see Masica 1976:121). So-called “relative clauses” are composed of a finite verb in adnominal function which modifies the following head noun. Kusunda is exceptional in this respect and does not fit the areal typology.1.” or they can be fully finite. i. those denoting established fact.e. hab˙-g-i amba roast-3-PAST meat ‘meat that he roasted’ b.1 Adnominals and relative clauses Kusunda has no relative pronouns. and it would not be surprising to find a preponderance of “factitive” verb forms coding them. turn out to be a nominalization). are the locus of presupposed propositions. Most have a one-to-many relationship. na tsi d~aba-d-i gimi. of course. – and explore the different ways in which they can be encoded. tsi e-d-ßn gimi hampe hur˙-n-i I give-1-REAL money where throw-2-PAST ‘Where did you throw the money I gave you?’ We have no examples of irrealis marked adnominals (though they may exist – further research needs to be done here).e. as in: ‘the I-kicked-it ball’ in place of ‘the ball that I kicked. you-ACC give-1-IRR-NEG ‘This is money that I found.) 9. I will not give it to you. different encodings imply different nuances of meaning. adnominals formed from verbs are typically nominalized structures. various complements.9. i. Neither appears to be a nominalization (although the “neutral” form may. tsi t-˙m-˙n kæaidzi I 1-eat-REAL food ‘food that I ate’ b. Relative clauses. Syntax of subordinate strucutures In this section we will look at individual grammatical categories – periphrastic causatives. they have no morphological form that can be said to be different from an ordinary finite verb (and hence not strictly speaking ‘participles’) – they are fully inflected for person. and in other cases we are unaware of any semantic differences.

but does have what appears to be a realis marker -˙n. In the section on phonology. REALIS: hab˙-n amba roast-REAL meat ‘roasted meat’ < hab˙-g-˙n roast-3-REAL 114 . As in §8. and it is precisely in these verbs that the deletion of third person affixes from “neutral” forms is significant (see §8.1). Its lack is a kind of deletion.1. a verb type that inherently includes person marking. 9. tul-ßn p˙idzi wear-NEUT cloth ‘worn clothes’ b.4.Watters: Kusunda Grammar.1 ‘Passive participles’ Earlier (§7. Third person -g is missing.1.2). we noted the tendency in Kamala of using third person verb forms -˙n (REAL) or -˙i (PAST) where Gyani Maiya uses -˙g-˙n ( 3-REAL) or -˙g-i ( 3-PAST).1.1. as in: (313) a.1.1. the embedded verb has no person inflection. as in (311b).1 and §9. dza -˙n tsya buy-NEUT tea ‘purchased tea’ The verb form in (311a) occurs on a Class II intransitive.2). however. appears to be unavailable for many transitive verbs. that the lack (or deletion) of subject person marking in transitive verbs. For such verbs.” A past neutral form also occurs.2. The following two forms we have seen before in non-embedded structures: (311) a. however. too. k˙la-q-ßn k˙re break-AC-NEUT jug ‘broken jug’ b. occurs on a Class II transitive verb.1 and §7.1. The possibility exists that Kamala is reacting to the lack of a semantic equivalent for a passive participle in some transitive forms. Person marking is not deleted to create a neutral form.2 Non-finite adnominals formed from verbs In non-finite (or “neutral”) adnominals. marked by a suffix -q or -t. as in: (312) a. then. yields a result semantically equivalent to a passive participle. especially §4.4. Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 9. a verb type that is inherently bereft of person marking (see §8. The form in (311b).1. Notice. however. a construction semantically equivalent to a “passive participle” is available (in which detransitivization is also marked). we will gloss the marker as <NEUT> for “neutral” (recognizing too that it may be a nominalization or a true morphological participle).2. gwi-t-ßn dimi gather-AC-NEUT smoke ‘accumulated soot’ This derivation. and is in the process of creating a new one by the deletion of third person – what we refer to in this grammar as “neutral.2. we saw an anti-causative derivation for some transitive verbs. respectively.

hul˙-g-˙n amba cook-3-REAL meat ‘cooked meat’ / ‘meat cooked by someone’ / ‘meat cooked by him’ 115 . (b). The most transitive are those with fully finite transitive verbs (315e).2. yo-˙n amba cook-REAL meat ‘cooked meat’ / Nep. The difference between a generic and specific sense seems to lie with the presence of a free pronoun: (314) a.1.o)’ Nep. hul-˙n amba cook-REAL meat ‘cooked meat’ ‘meat that has been cooked (by s.2 Levels of transitivity in adnominal modifiers There can be various levels of transitivity in adnominals derived from verbs. In Kamala’s speech. and the least transitive are those with Class II intransitive verbs where S equals O (315a). ‘pakeko masu’ (intransitive) b. and (c) by using Nepali equivalents in addition to English. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.3 we noted that Class I verbs are not as vulnerable to the erosion of third person markers as Class II verbs for the simple reason that they occur as prefixes. (Not all transitive verbs have a Class II intransitive counterpart as the following verb does. 9. ‘p˙kaune masu’ c. ‘pakeko masu’ (intransitive) ‘cooking meat’ / Nep. There. GENERIC: g-˙m-˙n kæaidzi n-˙m-du? 3-eat-REAL food 2-eat-IRR:SG ‘Will you eat eaten (defiled) food?’ In Gyani Maiya’s speech. ‘p˙kaeko masu’ (transitive) d. PAST: hab˙-i amba roast-PAST meat ‘roasted meat’ < hab˙-g-i roast-3-PAST In §8.Himalayan Linguistics: Archive No. the same principle seems to hold for Class II verbs. hol-di amba cook-ADJ meat ‘cooked meat’ / Nep. third person forms are commonly used in a generic sense.) It will be easier to describe transitivity levels (a).1. there may be a detransitivization of generic transitive forms. We begin with the least transitive: (315) a. or there may be an erosion of her phonology. SPECIFIC: gina g-˙m-˙n kæaidzi n-˙m-du? 2-eat-IRR:SG he/she 3-eat-REAL food ‘Will you eat food eaten (defiled) by him?’ b.

used primarily by Kamala and derived from an earlier third person generic form.) With ‘hide. as in: (316) a. we have at least three relative clause types in our data set – a “subject relative clause.” and a “locative relative clause. the inherent form of the verb is intransitive. It is instructive to contrast the ‘cook’ verb above with another verb whose intransitive semantics implies that S equals A.” 116 .3 Case recoverablility strategies for relative clauses All relative clauses. The third person generic form is given in (315d). The form in (315e) is a first person form which is never used in a generic sense. Rather. It can also be a specific third person form. The form in (315c) is a neutral form. like the forms in (312). *hyoq-ßn gimi *hide-REAL money Our informants’ explanation for the ungrammaticality of (b) is “because money cannot hide itself. and the only “passive” form available to such verbs is Kamala’s “neutral” form or Gyani Maiya’s “generic third person” form: (317) a.1. Gyani Maiya: hyo -a-g-˙n gimi money hide-CAUS-3-REAL ‘hidden money’ (by someone) This contrasts sharply with those forms that are inherently transitive and are detransitivized by anti-causative derivation. hul˙-d-ßn amba cook-1-REAL meat ‘meat that I cooked’ The verb in (315a) is a suppletive intransitive form of the verb ‘to cook’ and always implies intransitivity with S equalling O. where there is no inherent O intransitive verb. The Kusunda verb for ‘hide’ is such a verb. Based on this criterion.’ because S equals A. It is transitive. The -di form in (315b) is an adjectival form that is available to only a few transitive verbs.Watters: Kusunda Grammar. but with no agent in view. 9. whether they are “subject relative clauses” or “object relative clauses.and a transitive hol-. Furthermore. the intransitive adnominal must modify an animate noun. (For ‘cook’ there are two inherent forms – an intransitive yo. and implies full transitivity.” Thus. the passive adnominal must be formed off the causativized version of the verb.” an “object relative clause. hyoq-ßn amba hide-REAL animal ‘hidden / hiding animal’ b. Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– e. There is no morphological marker in the embedded verb to distinguish different types.” are coded in the same ways – as neutral or as fully finite. Kamala: hyo -a-˙n hide-CAUS-REAL ‘hidden money’ gimi money b. the distinction is based entirely on the identity of the role of the argument that occurs as the head of the NP.

3 Locative relative clause In a locative relative clause.3. hab-di roast-ADJ ‘roasted’ 117 .Himalayan Linguistics: Archive No.1. a peripheral argument is made head of the noun phrase – in this case a location. a fully finite verb without derivational morphology). Some intransitive and detransitivized verbs can form adjectives by the suffix -di. wi ˙-g-i nu house make-3-PAST person ‘the person who built the house’ 9. along with its predicate (again. along with its predicate (a fully finite verb without derivational morphology). d˙-g-˙n nu hampe go-3-REAL person where ‘Where is the person who went?’ b. and it is not yet known what the parameters are. h˙ra-q-ßn aota open-AC-REAL door ‘open door’ d. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 9. as in the following: (319) a.3. modifies the new head noun.1.1.1 Subject relative clause In a subject relative clause the erstwhile subject is extracted and made head of a new NP. as in: (320) hi o -da-g-i tæ˙w˙ pig die-CAUS-3-PAST place ‘the place where he killed a pig’ Adjectivals 9. however.1. The rest of the embedded clause.2 Object relative clause In an object relative clause the erstwhile object is extracted and made head of a new NP. The derivation.’ b. seems not to be available to all intransitive verbs. nu pumb˙-n-i ip˙n corn you beat-2-PAST ‘corn that you beat (thrashed)’ 9. gimi t˙mb˙-g-i n˙ti money send-3-PAST who ‘Who sent the money?’ / ‘Who is the money sender?’ c. tsi k˙la-d-ßn d˙idzi kila-˙-g-˙n I chop-1-REAL firewood steal-TR-3-REAL ‘He stole the firewood I chopped. as in the following: (318) a. We give the following examples: (321) a. modifies the new head noun.4 We include adjectivals under Adnominals and relative clauses because of their close formal similarity. The rest of the embedded clause.3.

’ The embedded verb is coded in much the same way as the clauses we saw in §8.2. hyo-dzi swell-ADJ c. kila-di steal-NML b. hyoq-di hide-ADJ d. hol-di cook-ADJ c. o -dzi die-ADJ b. One of the Kusunda periphrastic causatives implies that the event coded in the embedded clause already occurred. whereas we are unsure of the level of implicature in the second causative type.2 Periphrastic causatives ‘dead’ ‘swollen’ ‘dry’ ‘spoiled. Following are examples of its occurence in adjectival contexts: (323) a. borlo-q-di boil-AC-ADJ (322) a. This is one of the most puzzling suffixes we have encountered because of its occurrence in numerous and varied contexts. h˙ra-q-di open-AC-ADJ f. Following are neutral forms: 118 . As with universal expectations.1 and §8.Watters: Kusunda Grammar. rotten’ In addition to the morphological causative discussed in §7. however.2.1 and §7. Kusunda also has at least two periphrastic causatives. qai-dzi dry-ADJ d.2 – either as “neutral” or finite (or as a mutating structure as we saw in §8. Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b. the periphrastic causative implies less direct causation than the morphological causative. 9. none of the -da encoding we saw on “purposive” clauses.1. There is.1 Implicative causative – “make” The implicative causative is a complement verb structure whose matrix verb is the verb ‘to make. qau-dzi spoil-ADJ 9. bem-di fall-ADJ e.4).1. sokso-di black-NML ‘cooked’ ‘hidden’ ‘fallen’ ‘open’ ‘boiled’ We have two examples of -di acting as a formative for nominalization: ‘thief’ ‘black’ Some intransitive verbs make use of a suffix -dzi for adjectivals.

’ b.2. is something we have not yet encountered. realis mode. (325) a. like the verb in (325a). The marked modality is irrealis.1) is the irrealis form.’ 119 . t˙n-da amba hab-˙n me-ACC meat roast-REAL ‘He made me roast meat. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (324) a.’ b. and will not properly discuss until §10. 25. something that needs to be accounted for (we broached this topic first in §8. t˙n-da wi a-t-ßn ˙-g-i me-ACC house make-1-REAL make-3-PAST ‘He made me build a house.4) we will see the pattern again. Earlier (see fn. t˙n-da wi-g˙ t -a -an ˙-g-i me-ACC house-LOC 1-go-IRR:SG make-3-PAST ‘He made me go home. We introduce it here simply as an “overlapping converb.3.3. We do not seem to have any examples of the embedded form in past tense. and here. the marked polarity is negative.” Following are two examples: (327) a. tsi-gimtsi-da limu-n 1S-friend-ACC fight-REAL ‘I got my friend to fight. gina t˙n-da wi a-de t˙mb˙-g-u he me-ACC house make-CNV send-3-IRR:SG ‘He will get me to build a house. 26). the form of the embedded verb is realis (which is what one would expect for an implicative verb). gin-da limi-dzi a-d-i he-ACC dance-3:REAL make-1-PAST ‘I made him dance.1. the embedded form of “mutating” verbs (see §5. the marked dependency is dependent.5.4). a transitive verb which means ‘send’ in its literal sense. as its matrix verb. Irrealis would seem to imply a non-implicative status for the embedded verb.5. d-a -an-dzi a-d-u 3-go-IRR-TAM make-1-IRR:SG ‘I will make him go.’ b. however. (326) a.Himalayan Linguistics: Archive No.’ c.’ In an unexpected twist. In Class II verbs. Whether it is possible or not needs to be verified. The embedding structure.2 Causative “send” The “send causative” uses. and specifically in §10.’ ˙-g-i make-3-PAST a-d-i make-1-PAST The following embedded forms are finite. d-a -an-dzi a-d-aÚ u 3-go-IRR-TAM make-1-NEG:REAL ‘I didn’t make him go.’ 9.2. In the “permissive” construction (see §9. we noted that the mutating pattern very likely originated in an old “marked–unmarked” distinction and was later morphologized in various systems.

’ 9.1 Purposive The “purposive” clause is a complement to matrix verbs ‘go/come. With literal ‘send.’ ‘allow.’ and ‘know how to.’ the complement is cast as a “purposive” clause.1).’ ‘finish. as in: (328) tsi gin-da am-da I him-ACC eat-PURP ‘I sent him to eat.3 Complements There are numerous verbs which are capable of supporting sentential complements as an object (or destination) argument – ‘go/come for the purpose of.’ ‘want.’ We have already seen numerous examples of this clause type when we illustrated the morphology of -da in §8.’ 9. WITH COME: nu n-˙m-n-a you 2-eat-2-PURP ‘You came to eat. WITH GO: tsya qon-da t -a -an tea drink-PURP 1-go-IRR:SG ‘I will go to drink tea. complements of the desiderative can be marked “neutral” or with the suffix -da: (330) a.3. when used with non-sentential arguments.’ or ‘send/bring.’ ‘see/hear that.’ ‘think that.2 Desiderative The desiderative embeds complements to the matrix verb ‘to want.’ As with the purposive.’ n-ug-un 2-come-REAL c.’ The causative use of ‘send’ contrasts with the literal sense of ‘send’ through the complementation type used.Watters: Kusunda Grammar. it does not imply that the event named by the embedded verb occurred.’ b. gina wi ˙g-˙n ~oˆ la-g-˙n he house make-REAL want-3-REAL ‘He wants to build a house. Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.’ Literally.3.2. the Kusunda verb means ‘to search for. 9.’ ‘be able to. WITH SEND: gin-da kam ˙n-da t˙mb˙-d-i he-ACC work do-PURP send-1-PAST ‘I sent him to do work. Here we reiterate some of the examples used elsewhere: (329) a. nu gin-da wi a-de t˙mb˙-n-i you him-ACC house make-CNV send-2-PAST ‘You got him to build a house.’ 120 . The purposive can also be marked by “neutral” morphology (see §8.’ t˙mb˙-d-i send-1-PAST This latter sense of ‘send’ is non-implicative.

Himalayan Linguistics: Archive No. the embedded verb is marked by -da or as neutral: (333) a.3. A second “abilitive” construction contrasts with the one above in §9.’ As in many other languages of the region.4 Permissive The permissive construction employs the matrix verb ‘to give.’ c. tsi gin-da p˙rm˙-da ~oˆ la-d-i I him-ACC meet-PURP want-1-PAST ‘I want/wanted to meet him. tsi amba pad˙-n I meat hunt-REAL ‘I can hunt animals. tsi wi ˙n-da ~oˆ la-d-ßn I house make-PURP want-1-REAL ‘I want to build a house.’ 121 . whereas this one implies “finding opportunity. by strange circumstances. Thus. t-˙m-˙n d~aba-d-aÚ i 1-eat-REAL find-1-NEG:REAL ‘I wasn’t able to eat.’ Notice that in (332a) that the mutating verb ‘go’ is marked by the mutated form of ‘go.’ As in the other complements we have seen.’ hyoq-tsi-n able-1-REAL hyoq-tsi-n able-1-REAL • Opportunitive. with ‘want’ the subject of matrix and embedded clauses must be the same.” It uses the matrix verb ‘to find’: (332) a.3. though. The former implies physical ability. Syntactically. it is not possible with this structure to say ‘I want him to go’ – only ‘I want to go’ or ‘He wants to go.4. t -a -an d~aba-d-aÚ i 1-go-SUBORD find-1-NEG:REAL ‘I wasn’t able to go.’ indicating that it is subordinate.’ 9.) 9.’ b. indeed. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.3 Abilitive The verb ‘to be able’ is homophonous with the intransitive verb ‘to hide’. t-˙m-˙n hyoq-wa 1-eat-REAL able-IRR:NEG ‘I can’t eat.3. gina t˙n-da wi ˙n-da eg-i he me-ACC house make-PURP give-3-PAST ‘He allowed me to build a house.’ c. tsi wi ˙n-da I house make-PURP ‘I can build houses.3. (See §8. the verb allows a sentential complement marked either with -da or as neutral. as in the following: (331) a.’ b. it may be the same verb. The abilitive conjugates as an intransitive verb.

Not surprisingly. We repeat here example (177): (336) a.3. wi a-t-ßn tumb˙-d-ßn house make-1-REAL finish-1-REAL ‘I finished building the house.5 Preventative The preventative construction employs the causative of the verb ‘to stand. stand-CAUS-PROH ‘Let him come. and both are fully finite. Following are examples: 122 . Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– b.’ 9.’ b. g-u-ya.2. only this complement structure requires both verbs (the matrix verb and the embedded verb) to be coded in the same TAM.’ 9. pinda t-˙m-da e-g-auÚ-da.’ Two separate events are coded. now he lets me eat. *wi a-t-ßn tumb˙-d-i *house make-1-REAL finish-1-PAST 9. dæo ˆ-a-yin 3-come-OPT.’ b. (334) a. t-˙m-da e-g-i before 1-eat-PURP give-3-NEG-INCMP.Watters: Kusunda Grammar.4.1. Following are examples: (335) a.3. don’t stop him. this construction casts embedded mutating verbs in the “marked” form (elsewhere. wi a-d-i tumb˙-d-i house make-1-PAST finish-1-PAST ‘I finished building the house. the “irrealis”). t˙n-da t -a -an e-g-aÚ u me-ACC 1-go-SUBORD give-3-PAST:NEG ‘He didn’t allow me to go. and also with the implicative causative in §9. t˙n-da t -a -an e-g-i me-ACC 1-go-SUBORD give-3-PAST ‘He allowed me to go.6 Completive The completive matrix verb is the transitive verb ‘to finish.’ c. each of them expressing person–number and TAM specifications for the immediate clause only.3. 1-eat-PURP give-3-PAST ‘He didn’t used to let me eat. both verbs are marked as fully finite. autonomous event from the matrix verb event.’ b.’ As we saw earlier in §8.7 See and hear The verbs ‘see’ and ‘hear’ regularly support complements in which the embedded clause is a separate.’ and means ‘to stop someone doing something.’ As far as we know. g-˙m-˙n dæoˆ -a-d-i 3-eat-REAL stand-CAUS-1-PAST ‘I stopped him eating.

3.’ as in tsi n˙n-da imba-d-i (I you-ACC think-1-PAST) ‘I remembered you.’ (See §9.’ 9. tsi-gimtsi t˙n-da wi a-t-ßn g-~aÚ-dzi 1S-friend me-ACC house make-1-REAL 3-see-TAM (?) ‘My friend saw me building a house.’ 123 .’ while the embedded event is ‘I build a house.10.’ b.) In our corpus of data we have only cases in which both the matrix verb ‘know’ and the embedded complement verb are coded as realis: (340) a.’ In (337a) the main clause event is ‘I saw.’ Notice that here.’ The events are entirely autonomous.out-3-PAST hear-1-PAST ‘I heard him cry out.8 Think The verb ‘think’ in Kusunda also means ‘to remember. tsi-gimtsi wi ˙-g-˙n tsi 1S-friend house make-3-REAL I ‘I saw my friend building a house.3.’ which is used as a matrix verb for ‘believe.’ b. Knowing a person employs a different verb tsirma. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (337) a. loÚl˙-g-i m˙b˙-d-i cry. tsi wi a-t-ßn t-uk-an I house make-1-REAL 1-know-REAL ‘I know how to build a house.’ while the embedded event is ‘He builds a house. 9.’ ts-~aÚ-dzi 1-see-TAM (?) b. The same holds for the verb ‘hear.9 Know how The verb ‘to know’ in Kusunda expresses the notion of ‘how to do something’ and can include abstractions like gina gip˙n uk-an-dzi (he language know-REAL-<??>) ‘He knows Kusunda (= He knows how to speak Kusunda)’.’ Likewise. in (337b) the main clause event is ‘He saw.Himalayan Linguistics: Archive No.‘recognize. tsi wa t-ug-un m˙b˙-g-i I home 1-come-REAL hear-3-PAST ‘He heard me coming home. nu wi a-n-ßn n-uk-an you house make-2-REAL 2-know-REAL ‘You know how to build a house. like we saw in (329) and (334).3. the embedded “mutating” verb ‘go’ takes the marked case.’ We have only one example of the verb with a sentential complement: (339) tsi da≥ t -a -an imba-d-i I Dang 1-go-SUBORD think-1-PAST I think I will go to Dang. too.’ as in the following: (338) a.

2).3.’ 9.3.’ We have only one example: (341) ta≥ g-i-w˙n3 4 tsirma-t-ßn water 3-go-REAL recognize-1-REAL ‘I believe it will rain. teach 9.11 ‘Showing’ or ‘teaching how’ to do something employs a complement construction we have not yet seen (except briefly in §9.’ Show how.10 Believe The complement construction ‘to believe something’ employs the matrix verb whose literal meaning is ‘recognize.Watters: Kusunda Grammar. Syntax of subordinate structures ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– c.2.2 on Overlapping events. 34 The verb for ‘rain’ is surprising in Kusunda – ‘water goes’ instead of ‘water comes’. 124 . nu ip˙n sola-n-ßn n-uk-wa you corn break-2-REAL 2-know-NEG:IRR ‘You won’t know how to break (harvest) corn. and we will hold off discussing it till we get to §10.

2.’ (I hid and slept. wa u-g-i ip-dzi home come-3-PAST sleep-PAST ‘He came home he and slept.” In the surrounding languages.’ c. hyoq-ts-in ts-ip-ßn hide-1-REAL 1-sleep-REAL ‘I hid and I slept. aota h˙ra-g-˙n t˙n-da daha-g-i door open-3-REAL me-ACC seat-3-PAST ‘He opened the door and he seated me.’ 125 hul-ts-in cook-1-REAL .10. Clause chaining “Clause chaining” is a clause combining strategy very common in the languages of South Asia and syntactically distinct from English-type “co-ranking” structures like He went to the forest and set traps. these are “co-ranking” structures (Longacre 1985) and differ from the “co-subordinate” structures we will look at in §10. Following are examples: (342) a. eating occurs first and leaving second. Kusunda is typologically distinct from the languages immediately surrounding it. First of all it distinguishes on a temporal scale between “sequential” events and “overlapping” events. medial verbs have no special marker to distinguish them from final verbs. on the other hand. it marks sequential events as “co-ranking” and overlapping events as “co-subordinate. are less finite.’ e. it is termed “co-subordinate.’ g. ~oˆ la-d-i d~aba-d-aÚ i search-1-PAST find-1-PAST:NEG ‘I searched and/but I didn’t find it. With respect to clause chaining.” It is subordinate in that it depends on the final verb for certain finite specifications. sometimes pared down to a simple verb root plus some kind of non-final marker. Kusunda has such sequences. Verbs belonging to chain-medial clauses.’ f. gina po≈la-g-i g-˙m-˙n 3-eat-REAL he bathe-3-PAST ‘He bathed and he ate. 10. like Nepali and Kham. but coordinate in that it lacks many features of true subordinates (Haspelmath 1995). and secondly.1 Sequential events Sequential events occur in temporal chronology – Having eaten he left is an example. there is no formal distinction between sequential and overlapping events.) b. being marked for things like person/number agreement and tense/aspect distinctions. Furthermore. Typically. but all verbs in the chain – medial and final – are fully finite. and the only strategy employed is the co-subordinate one. As such. A basic feature common to classical clause chaining languages is the distinction between chain-medial and chain-final verbs. the verb belonging to the chain-final clause is the most finite. Thus. borlo-q-di ta≥-da sip-ts-in boil-AC-ADJ water-LOC enter-1-REAL ‘I entered boiling water and I got cooked.

Watters: Kusunda Grammar; Clause chaining ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– All of the sequences above retain the same subject for both verbs and would occur in other languages of the region as classic clause chains – the first verb would occur as a non-finite medial verb (plus linking morphology), and only the final verb would occur as fully finite with person inflection. English, too, would treat all the sentences in (342) as “same-subject” chains by deleting the second pronoun — ‘You hid and you slept.’ The kinds of sequences illustrated in (342) are not formally different from the coordinated structures we will look at in §11.1. The only difference, if any, might be based on semantic inferences. The structures here were elicited as sequential chains (using Nepali -er˙), while the structures in §11.1 were elicited as non-chains like After I eat I will go. 10.1.1 Negation in sequential chains

One consequence of having finite marking on both verbs is their logical independence of one another. Under negation, it appears (tentatively, at least) that the scope of the negative applies only to the negated verb, something not necessarily true for classical clause chaining languages. In classical clause chaining languages, negation on the final verb can extend backwards to the medial verb as well, as in the following example from Kham (Watters 2002:327): (343) m˙d˙ cæokoraÚ o-d˙ ta-s˙-m~ï:h-si-c-yo wine beer drink-NF PROH-CAUS-drunk-DETRANS-2P-IMP ‘Don’t get yourselves drunk drinking beer and wine!’

The inference is that you should not drink and get drunk – drinking is also prohibited. However, cause and effect must be semantically well integrated before the scope of negation on the final verb extends back to include the medial verb. Unfortunately, our corpus of data does not include examples of negative clause chains where the two events can be thought of as semantically well integrated, and under such circumstances it is impossible to know if the scope of the negative in semantically integrated events is more inclusive than it is in semantically discontinuous events. Negation of the medial verb Where the medial verb in a clause chain is negated, the scope of negation is local – it applies to that verb alone. Following are examples: (344) a. tsi t-˙m-daÚ u t-ug-un 1-come-REAL I 1-eat-REAL:NEG ‘I came without eating.’ (lit. ‘I didn’t eat, I came.’) b. po≈la-g-aÚ i g-˙m-˙n bathe-3-REAL:NEG 3-eat-REAL ‘He ate without bathing.’ (lit. ‘He didn’t bathe, he ate.’) c. teina tumba-d-aÚ u, goraq tumb˙-d-u today finish-1-REAL:NEG, tomorrow finish-1-IRR ‘Not finishing it today, I’ll finish it tomorrow.’ Negation of the final verb We have very few examples of this structure, but the ones that we do have often imply “condition,” which will be treated more fully in §11.2. Following are two 126

Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– examples: (345) a. okti t-˙m-da-k t-o -d˙-o medicine 1-eat-PL-IRR 1-die-PL-NEG:IRR ‘(If) we eat the medicine we won’t die.’ (lit. ‘We will eat the medicine and we won’t die.’) b. nu t˙n-da pumba-n-u tsi qæai-d-wa I fear-1-NEG:IRR you me-ACC beat-2-IRR, ‘(Even if) you beat me I won’t be afraid.’ (lit. ‘You will beat me, I won’t be afraid.’) A third example implies opposition: (346) gina po≈la-g-i g-˙m-daÚ u he bathe-3-PAST 3-eat-REAL:NEG ‘He bathed (but) he didn’t eat.’ A second negative


Both medial and final verbs sometimes employ a second negative. Though we can surmise on the origin of its structure, we do not know how this negative differs semantically from the one illustrated in §10.1.1 above. We give two examples of a medial verb negated by this second negative (a–b), and one example of a final verb (c): (347) a. gina po≈la kæaÚ i g-˙m-˙n he bathe be:REAL:NEG 3-eat-REAL ‘He didn’t bathe and he ate.’ (He ate without bathing.) b. wa uga kæaÚ i un-da ip-dzi home come be:REAL:NEG trail-LOC sleep-TAM ‘He didn’t come home and he slept on the trail.’ OR ‘Not coming home, he slept on the trail.’ c. gimi meÚ-d-i d~aba-n˙ kæaÚ i be:REAL:NEG money lose-1-PAST, find-2 ‘I lost the money and you didn’t find it.’ Beginning with (347a), the expected negative form for a finite, realis Class II verb would be po≈la-g-aÚ i (bathe-3-REAL:NEG), the -g being a third person agreement marker. The negative kæaÚ i, on the other hand, is the negative existential copula, now used as negation marker on non-copular verbs. (See Negative existential under § The use of copular verbs in modality systems, of course, is widespread in the languages of the world, South Asia included. This is the first instance, however, that we have seen it in Kusunda. In §8.1.4 we saw examples of bare verb roots used in concatenative structures like hol t-˙m-du (cook 1-eat-IRR:SG) ‘I’ll cook and eat.’ The first root, hol, is followed by a second finite root, t-˙m-du, without intervening morphology of any kind. The negative construction that we are dealing with here is a parallel structure – the negative copula follows the bare root. It is reasonable that such “bare root” verbs should be negated by the existential, without all the finite machinery of conjugated verbs. The form of the final verb in (347c) is also unexpected. We would expect d~aba-n-aÚ i (find-2-REAL:NEG). Final verbs are always finite, being marked for TAM and person/num127

Watters: Kusunda Grammar; Clause chaining ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– ber information, and (347c) manages to tack a second person marker -n˙ onto the end of the first root. It is very likely, in fact, that the grammaticalization pathway for d~aba-n-aÚ i is < *d~aba-n˙ kæaÚ i. It may turn out, then, after further study, that the “concatenative” structure is the “real” clause chaining device and the structures illustrated in (342) are coordinated “co-ranking” structures on the English model. 10.2 Overlapping events

Kusunda has a second kind of clause chaining device whose structure (but not function) is more in keeping with areal expectations. In this second device, medial verbs are non-finite and marked by a clause-chaining suffix -de3 5, while final verbs are fully finite for person marking and TAM. The surprising thing about Kusunda, however, is that this structure does not mark sequential events. (This was far from obvious in our early work because of the perceived similarity with Nepali sequential events.) The non-sequential nature of the structure was especially difficult to detect in contexts like the following where it is difficult to imagine anything but a sequential reading: (348) a. am-de u-g-i eat-CNV come-3-PAST ‘He came while eating.’ (i.e. before finishing his meal)3 6 b. ip-de t-ug-un sleep-CNV 1-come-REAL ‘I came while sleeping.’ (i.e. before the night was over) c. hyoq-de ts-ip-ßn hide-CNV 1-sleep-REAL ‘I slept (while) hiding.’ In the linguistic literature, the opposite of a sequential event has often been referred to as a “simultaneous” event, as in While she was cooking she tended the baby. As we will show, however, the term “overlapping” event represents the data in Kusunda better than the term “simultaneous,” and is the one we will use throughout. Thus (346a) does not necessarily mean (though it can mean) that he was eating while coming down the road, which is what a simultaneous reading would give. 10.2.1 Cause and effect

For expressions of “cause and effect” it might seem that a sequential event structure would be more appropriate than an overlapping event structure. But presumably because of close topic continuity between the events in a cause–effect vector, they are frequently coded as overlapping, as in:

The suffix -de is also the converbal marker for some dialects of Kham (with variants -di and -d˙ in other dialects). Classical Tibetan has (s)-te. When asked for the meaning of sentences like these, our informants’ response in Nepali would typically be khaer˙ ayo 'Having eaten he came.’ But when asked for the Kusunda equivalent of khaer˙ ayo they were prone to respond with the sequential construction g-˙m-˙n u-g-i (3-eat-REAL come-3-PAST) ‘He ate and he came.’



2. tsilgari soq-de katæm˙ndu t-ug-un airplane climb-CNV Kathmandu 1-come-REAL ‘I came to Kathmandu by climbing (riding) a plane. (350e).3 and §10. Except in a restricted case (which we will see in a subsection below). h~aÚ-de ts-ip-ßn sit-CNV 1-sleep-REAL ‘I slept sitting up.’ 10. dz~o-wa-de o -da-d-i hang-CAUS-CNV die-CAUS-1-PAST ‘I killed it by hanging it. gilas-ga sib-a-de g-o -on glass-IN enter-CAUS-CNV 3-put-REAL ‘He put it in the glass by stuffing it.3 The benefactive construction The benefactive construction is marginal in Kusunda.’ c.’ c.’ b.’ b. and compared with the surrounding languages. 129 .2.’ d.’ The last sentence. am-de gip˙n a-t-ßn make-1-REAL eat-CNV talk ‘I talked while eating.4 respectively) give further evidence of the “overlap” reading of the Kusunda converb. Such a reading comes as a result of overlap or simultaneity.Himalayan Linguistics: Archive No. 10. tsi amba hul˙-de t-ug-u I meat cook-CNV 1-come-NEG:IRR ‘I won’t come while cooking meat.2. but ‘stuffing’ describes the manner in which it was put. pumba pumba-de gina h~aÚ hyo-da-d-i he face swell-CAUS-1-PAST beat beat-CNV ‘I made his face swell by beating and beating him.’ d. is typologically exceptional. na am-de t-o -da-k this eat-CNV 1-die-PL-IRR ‘Eating this we will die.’ e. hol-de t-˙m-du cook-CNV 1-eat-IRR:SG ‘I will eat while cooking.2 Temporal overlap Here we give examples of sequences in which the overlap is more temporal than topical: (350) a.2. The ‘putting’ and the ‘stuffing’ occur at the same time. the benefactive requires no derivational machinery. is a good example of the “adverbial” reading of some converbs. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (349) a. The benefactive construction and the ‘teach’ converbal (which we will treat in §10.

TRANSITIVE USE: apa-ce the-ta buh-ri pit-ci father-ERG he-DAT field-to send-TAM ‘Father sent him to the field. however. has transitive and bitransitive senses for verbs like ‘send’ and ‘show. the transitivity of verbs is fixed. Clause chaining ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Usually. like Tamang. TRANSITIVE USE: no-e ≥a-lai he-ERG me-OBJ ‘He sent me.’ Such verbs have transitive and bitransitive uses. and even verbs like ‘send’ and ‘show’ are inherently transitive.’ Such a construction is also available to some TB languages. but only on verbs that are marginally bitransitive – verbs like ‘send’ or ‘show. BITRANSITIVE DERIVATION: no-e ≥a-lai cæiti p˙r~î-d-y-~aÚ-ke-o he-ERG me-OBJ letter send-NF-GIVE-1S-PFV-3S ‘He sent me a letter.2. like Kham. requires a benefactive derivation which employs a non-literal usage of the verb ‘to give. For such verbs to be used in a bitransitive (benefactive) sense requires an applicative derivation in the verb. BENEFACTIVE: tsi gin-da wi a-t-ßn I him-DAT house make-1-REAL ‘I built a house for him.’ b. Nepali. BASIC TRANSITIVE: tsi wi a-t-ßn I house make-1-REAL ‘I built a house.4.’ For verbs with a fixed transitive reading. See the ‘benefactive’ under §5. as in: tsi-lage limu-dzi (me-FOR fight-TAM) ‘He fought for me’ (on my behalf).’ p˙r~î-na-ke-o send-1S-PFV-3S b. BITRANSITIVE USE: the-ce apa-ta chiTi pit-ci he-ERG father-DAT letter send-TAM ‘He sent a letter to Father.’ Nepali. 37 130 .’ b. all that is required is the addition of a dative argument3 7 to a transitive clause. as in the following: (351) a.Watters: Kusunda Grammar. as in the following example from Tamang: (352) TAMANG (from Taylor 1973:106-107) a. like Kham (and Tamang). The derivation employs the verb ‘to give’ (with a non-literal reading) in a clause chaining construction: (353) KHAM (from Watters 2002:239) a.’ It is also possible to use a modification of the Nepali -ko lagi.’ In other TB languages.2.

gina t˙n-da wi a-de e-g-˙n he me-ACC house make-CNV give-3-REAL ‘Building a house he gave it to me. tsi-kæaidzi hul˙-de e-go 1S-food cook-CNV give-IMP ‘Cooking my food give it to me!’ (not ‘Cook my food for me!’) Recall from §9. The sentence in (c) would be a “malefactive.’ (not ‘He built a house for me.4 that ‘give’ is used in Kusunda in a permissive sense with the embedded verb marked by -da. reveals that ‘give’ is not benefactive.2. they can produce such sentences based on a Nepali template. only Telegu “appears to lack the benefactive use of give.’ b. followed by the verb ‘to give’ as a final verb. but retains its literal sense. Structurally.3. but after discussion they reject them on the basis that they are incompatible events.Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 10. the benefactive is required on all sentences in (354) for the simple reason that doing something to one’s possessed item is equivalent to doing it to the person – it forces an extra argument.’) b. Careful probing. saying that the money could not be given after it had already been sent. 39 38 131 . the knife would have to be uncovered. it uses the converb -de on a medial verb. Interestingly.’ c. Interestingly. ?tsi-kolde du-ga l˙b˙-de e-g-˙n 1S-knife dirt-IN bury-CNV give-3-REAL ?‘Burying my knife in the dirt he gave it to me. Masica (1976:148) notes that throughout the Indian area. on the surface. ‘give’ has a permissive meaning only. Sentence (c) they rejected outright. however.3.3 9 For a language like Kham.” None of these observations apply to Kusunda. appears to be almost identical to the benefactive construction used in Nepali and Kham (as illustrated by Kham in (353b)).1 A deceptive look-alike Kusunda has a construction which. Our informants could force a reading of sentences (a) and (b).3 8 We first began to realize this when sentences like the following created confusion and heated debate: (354) a.’ The verb ‘to give’ in all three sentences in (354) is incompatible with a literal meaning for ‘give’ – hence their ungrammaticality or questionable grammaticality. We were able to verify our hypothesis on the following sentences which are not incompatible with a literal reading of ‘give’: (355) a. but insisted that they represented two very incompatible events – for (a). *tsi-gimi sabe t˙mb˙-de e-g-˙n 1S-money there send-CNV give-3-REAL *‘Sending my money there he gave it to me. ?tsi-gimi kila-de e-g-˙n 1S-money steal-CNV give-3-REAL ?‘Stealing my money he gave it to me.” He further notes that as an auxiliary.

’) 10.’) b. Clause chaining ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– c.4 The “teach” converbal The “teach” converbal employs the overlapping converbal marker -de. followed by a finite form of the verb ‘to show.Watters: Kusunda Grammar. ni-bæari abi e-d-u 2S-load carry give-1-IRR:SG ‘I’ll carry your load for you. gina t˙n-da amba hul˙-de unda-i he me-ACC meat cook-CNV show-3:PAST ‘He showed me how to cook meat.’ 10.’ (not ‘I’ll bring it for you.’ The following examples show that the verb ‘show’ is not in a sequential relationship with the preceding verb. tsi n˙n-da ts-i-de e-d-u I you-ACC 1-bring-CNV give-1-IRR:SG ‘I’ll bring and give it to you. pumb˙ e-go beat give-IMP ‘Beat it (for me).’ c. t˙n-da hul˙ e-g-˙n me-DAT cook give-3-REAL ‘He cooked (it) for me.’) b. make. d.’ b. It implies teaching by example. tsi-wi nu ukæ˙ e-n-i 1S-house you ruin give-2-PAST ‘You ruined my house.3. tsi n˙n-da wi a-de unda-d-u I you-ACC house make-CNV show-1-IRR:SG ‘I’ll show you how to build a house. as in: (356) a. but we are now in a position to discuss it more fully.’) 10. nu t˙n-da ip˙n sula-de unda-n-u you me-ACC corn break-CNV show-2-IRR:SG ‘You’ll show me how to harvest corn.’ Some of the same verbs (but apparently not all of them) can substitute the verb ‘to give’ in place of ‘to do. ‘showing how’: (357) a.1.2. we get a benefactive reading.2.2. The ‘send’ causative contrasts with the ‘make’ 132 .5 Causative converbal – ‘send’ We saw the periphrastic causative ‘send’ earlier in §9.2.2.’ (not: ‘He cooked meat and showed me.’ Where this occurs.4 that many transitive verbs are modelled after the verb ‘to do.2 A lexical benefactive Recall from §6. but an overlapping relationship.’ (not: ‘You’ll break corn and show me.’ (not: ‘I’ll build a house and show you.’ as in: abi a-t-ßn (carry do-1-REAL) ‘I carry it.

The accusative marked object is an object of the higher verb and the embedded object is unmarked.’ (not: ‘I’ll make a house and send you.) The overlapping converb carries with it the implication that the two verbs in the clause chain are at least partially continuous. First of all. it implies that the embedded event also occurred. ‘She made him leave. in which the initial verb-event occurs first and the final verb-event later. This contrasts with a sequential interpretation. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– causative in at least two ways. such an implication seems to be weaker (though we are not yet sure).’ 133 .e. (The ‘make’ causative did not. With the ‘send’ causative. tsi n˙n-da wi a-de t˙mb˙-d-u I you-ACC house make-CNV send-1-IRR:SG ‘I’ll make/have you build a house. the ‘send’ causative employs the “overlapping converb” as part of its structure. nu gin-da wi a-de t˙mb˙-n-i you him-ACC house make-CNV send-2-PAST ‘You made/had him build a house. Secondly. the ‘make’ causative is an implicative causative. Following are examples: (358) a. which means something like. i.Himalayan Linguistics: Archive No. These clauses are bitransitive with a hierarchical structure. both temporally and thematically.’) Part of the force of the ‘send’ causative is captured in the English idiomatic expression (from Shakespeare) She sent him packing.) b.’ (not: ‘You built a house and sent him.

down-1-REAL 1-sleep-REAL:NEG ‘I laid down. but there we were contrasting it with co-subordinate strucures. as in: He came home and slept.’ Thus. the same subject in both verbs and their English translations. though in Kusunda. But because there are no explicit clause connectors. especially where there is also continuity of time reference between the two verbs. It has a basic coordinate structure. The same Kusunda sentences could also be translated as. (but) I didn’t sleep. We looked briefly at coordination in §10. and the latter are “co-subordinate. Longacre 1985). he slept’ or ‘He came home and he slept. of condition. The adversative is based entirely on semantic inference. Continuity between clauses implies conjunction.’ in Kusunda.1. but one in which the second event of the bi-clausal structure can be thought of as an unexpected outcome of the first event: (360) tsi yu≥-tsi-n ts-ip-daÚ u I lie. hyoq-ts-in ts-ip-ßn hide-1-REAL 1-sleep-REAL ‘I hid and slept. and discontinuity implies disjunction. Two very broad factors that are relevant to our study are 1) the level of continuity between clauses. (The sentences were originally elicited as sequential chains using Nepali -er˙. In this chapter we will look further at co-ranking structures – those clause combinations in which both verbs are finite. all verbs being finite (and hence equal in rank. the elided translations given in (359). here we contrast coordination with other co-ranking structures. and 2) those that code events in which there is some kind of temporal or semantic overlap. Following are examples of coordination. they are. their English translations elide the second pronominal reference. but does not require. This is not to say that the inferences are not dependent on certain grammatical and pragmatic factors. for example. 11. and 2) the particular TAM marking employed. like ‘and’ or ‘but. wa u-g-i ip-dzi home come-3-PAST sleep-3:PAST ‘He came home and slept. hence. Combinations of finite clauses In §10 we looked at clause chains of two types – 1) those that code events which occur in temporal sequence. of adversity and concession.’ Subject continuity allows.” making a distinction between medial verbs (non-finite) and final verbs (finite).’ b. There is a formal structural difference – the former are “co-ranking” in structure. the person and number of the subject is marked on both verbs. These are clauses of coordination. translated by English ‘and’: (359) a.1 Coordination Where the person reference in both verbs of a co-ranking structure is identical we get something equivalent to coordination.11. ‘He came home. It also turns out that irrealis marking on the final clause of a two-clause construction almost always implies a conditional construction.’ 134 . the semantic relationship between any two co-ranking clauses is a matter of inference.) • The adversative.

Again.’ b. today meet-1-PL-REAL ‘Yesterday I thought of you. d˙baq a-t-ßn wi 1-eat-REAL.’ (364) DIFFERENT PRONOMINAL REFERENCE: nu n-ug-u s˙mba tsi t-˙m-du I 1-eat-IRR:SG you 2-come-IRR later ‘After you come I will eat.1. Here. tsi un meÚ-d-u me-ACC send-PROH. there are no explicit coordinators marking such predications.2 Where one or both clauses are marked by adverbs of time. pyana n˙n-da imba-d-i. In the following example both clauses are in future time. I road lose-1-IRR:SG ‘Don’t send me.’ In examples (359–361) all verbs were cast in past time – either as realis or past tense.2). I’ll sleep later. today we met. later ‘I’m building a house. a disjunction of events is implied by the two verbs in the co-ranking structure. but with the same discontinuity of person reference. ‘You’ll come. tsi kæaidzi t-˙m-˙n. a disjunction of events is implied.1. In the following examples. now I’ll build the house.1. and the second verb for irrealis: 135 . t˙~îna p˙rm˙-d-a-n yesterday you-ACC think-1-PAST. tsi wi a-t-ßn. as reflected in the following translations (here the second pronoun cannot be elided): (361) a. The semantic inference to be drawn in such cases is not different from the inferences we saw in (363) and (364).’ (lit. I’ll eat later’) 11. the second clause implies “reason”: (362) t˙n-da t˙mba-yin.1 Discontinuity of person reference Where two verbs in a sequence are marked for different subjects. there can also be an accompanying discontinuity in TAM marking between the two verbs of the construction.1. s˙mba ts-ip-ßn 1-sleep-REAL I house make-1-REAL. even where the pronominal reference for both events is identical.’ Discontinuity of time 11. the first verb is marked for realis or past. t˙n-da un unda-g-i t-ug-un me-ACC road show-3-PAST 1-come-REAL ‘He showed me the road and I came. nu n-ug-un tsi hyoq-ts-in hide-1-REAL you 2-come-REAL I ‘You came and I hid.’ b.’ c. I’ll lose the road.Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 11. the relationship between the two events must be inferred: (363) SAME PRONOMINAL REFERENCE: a. now make-1-REAL house I food ‘I ate food.3 Discontinuity of TAM marking Where there is discontinuity of time (as in §11.

The following example. contrasts with the examples shown in (365): (366) na t-˙m-da-n t-o -da-k this 1-eat-PL-REAL 1-die-PL-IRR ‘If we eat this we will die. kæaidzi t-˙m-˙n. d˙baq t -a -ak 1-go-IRR before bathe-1-PAST. now I’ll go. a conditional.2 Disjunction of person in irrealis contexts for bi-clausal constructions can occur with or without continuity in the TAM marking of the two verbs. We saw such an example in (361). I’ll build the house later. An exception to this is in a bi-clausal structure having the same properties but with an explicit temporal specification on one clause. the protasis and the apodosis. are marked for irrealis. (The surprising thing is that irrealis marking is not always required on the protasis. it is often implied that the first clause is a kind of conditional with the main clause its apodosis. Combinations of finite clauses ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (365) a.2. We have already seen in §11. the default relationship is a conditional one.1 Continuity of person in irrealis mode Irrealis marking on the final verb of a bi-clausal construction can occur with or without person continuity between the two clauses. then. I’ll sleep later.1 With continuity of TAM We have numerous examples in our corpus in which both sides of a conditional clause.2. the protasis side of the conditional should also be marked irrealis. later make-1-IRR:SG food ‘I’m eating food.2 that the final clause of a conditional construction must be marked irrealis. 11.2. 11. 11. pinda po≈la-d-i.2 Conditionals and concessives wi house In a language like English. a matter that we will discuss in 136 . later ‘I haven’t slept (yet). now ‘Earlier I bathed. and it is not at all surprising that in a language for which irrealis marking is available. tsi ts-ip-d-aÚ u.’ b.e.2. there is no formal dependency marking – both verbs are fully finite – but there is an implied semantic dependency. where the semantic relationship is ‘after this… then that. Where the final (i. The result in both cases is the same – it implies condition. In Kusunda.’ Without the explicit adverbial specification. conditional and concessive clauses are subordinate. In the following sections we present both types.Watters: Kusunda Grammar. s˙mba ts-ip-du 1-sleep-IRR:SG I 1-sleep-REAL-NEG. the protasis and apodosis elements depending upon one another.’ There are also cases of discontinuity in TAM marking which are not accompanied by time adverbials – such often imply “condition” and will be dealt with next.’ Discontinuity of person in irrealis mode 11. s˙mba a-d-u 1-eat-REAL.’ c. however. main) clause of a bi-clausal structure is marked for irrealis.

’ (lit. goraq ta≥ g-iw-˙n tsi ts-˙g-wa 1-go-NEG:IRR tomorrow water 3-go-REAL I ‘If it rains tomorrow.’ 11. nok n-ug-da-k qa tse b˙jar ts-i-da-k together bazaar 1-go-PL-IRR you. it might be that with realis marking in the protasis the speaker is stating something about general knowledge – “If you don’t drink the tea. goraq ta≥ g-i-wu tsi ts-˙g-wa 1-go-NEG:IRR tomorrow water 3-go-IRR:SG I ‘If it rains tomorrow I won’t go.’ It is not known how the semantic interpretation of the sentences in (368) should differ from those in (367). For example. Here we get a disjunction of TAM marking.2 With discontinuity of TAM In conditional constructions it is also possible that the protasis side of the construction be marked for realis. I won’t be afraid’) 137 . and the speaker makes no commitment about the outcome. though conditional. it will get cold” – or even about his level of commitment to the outcome – “If it rains tomorrow (and I expect it to do so). Some kind of evidential might be implied here.’ c. but when!” With irrealis in the conditional.” In any event. nu t˙n-da pumb-˙n tsi you me-ACC beat-REAL I ‘If you beat me I will hide.’ b. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– the next subsection. I will not go. I won’t go. will actually occur – something similar to the differences reflected in the English saying “Not if. ‘You will beat me.pl 2-come-PL-IRR ‘If you come we will go together to the bazaar.2. it will get cold.Himalayan Linguistics: Archive No. as in the following: (368) a. it is likely that the speaker is more strongly committed to the possibility that the event in the first clause. nok kæaidzi hul-a-k tok am-da t-ug-da-k you. 11.2.2. nothing is certain. n-˙m-daÚ u kæ˙≥-g-u 2-eat-NEG:REAL cold-3-IRR:SG ‘If you don’t drink it.’ b.’ hyoq-ts-u hide-1-IRR c. replacing the expected irrealis.) Following are examples: (367) a.3 A concessive The concessive in Kusunda is no more than a conditional with implied contraexpectation between the protasis and the apodosis.pl food cook-PL-IRR we eat-PURP 1-come-PL-IRR ‘If you cook food we will come to eat. as in the following: (369) nu t˙n-da pumba-n-u tsi qæai-d-wa you me-ACC beat-2-IRR:SG I fear-1-NEG:IRR ‘Even if you beat me I won’t be afraid.

6. We saw earlier. those reported in §10 and §11 are all that our current data base will support. and the translation given to us was a time adverbial. This is a topic that clearly needs more study.3. There must certainly be many more co-subordinate and co-ranking constructions in Kusunda. make explicit some of the dependency relationships we have already discussed.2. under section §11. but for the time being. a future subjunctive in (a) versus a past subjunctive in (b). a notion superimposed upon the “lower level” TAM marking which is relevant to individual clauses. Thus: (371) a.2. On the other hand. -da may be an irrealis subordination marker.’ In other cases. The difficulty is resolved if the subjunctive is marked as an add-on category.’ OR ‘If he had cooked we would have eaten. We have one example in which only the medial verb is marked by -da. 138 . Here. both verbs are marked by -da.5.3 Bi-clausal residue There are other bi-clausal constructions in Kusunda in which one or both clauses are bounded by a suffix -da. This has direct bearing on the overall interpretation. -Da appears to be a subordinator and may. it may be that the -da here is no more than the same “incompletive” aspectual marker -da we saw in §5.’ It appears that the difference between the (a) and (b) sentences in (371) is the TAM of the main verb – in (a) it is irrealis and in (b) it is realis. gina hul-˙n-da tok t-˙m-da-k-da he cook-NEUT-SUB we 1-eat-PL-IRR-SUB ‘If he cooks we will eat. in fact.’ (??) b. ta≥ g-iw-˙n-da tsi sowa-t-ßn-da I bathe-1-REAL-SUB water 3-go-REAL-SUB ‘If it had rained I would have bathed.2 Discontinuity of TAM that a bi-clausal construction with a subjunctive for the first clause can mark that clause as realis – a seeming contradiction of terms (see 367). as in: (370) tsi ts-ip-ßn-da n˙ti aoda d˙i-˙n 1-sleep-REAL-ADV who door knock-REAL I ‘While I was sleeping someone knocked at the door.Watters: Kusunda Grammar. Combinations of finite clauses ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 11.

an˙-go (imp). followed by the imperative form (usually -to or -˙-go).Appendix A Kusunda Vocabulary Following is a basic vocabulary of about 850 items in Kusunda. up. speak' (see also ts-˙ ˙-d-ßn and n-˙ ˙-n-i) -˙g vi. how many ˙uba≥ n. flesh. an˙-g-˙n (3rd). to hunt (lit. abi a-d-i. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– -˙ vt. from ˙nin. meat amba pad˙-g-˙n vt. to hunt meat) ambu / ambo n. dog ag˙i getse n. how much. ground a~o / waha pp. make hole ˙ ab˙q / ˙bo≈ n. ayin. ai-dzi. ˙n˙-yin (proh) vi. across alu (Nep) n. udder. when ˙si interrog. breast. and we have tried to list first person realis forms (usually -d-ßn). alms aidzi n. n-˙g-˙n. to burn food ao-dzi vi. greens. dirt. ˙raq / ˙ra≈ are variant pronunciations of the word for ‘garlic. inside ao bæa≥ n. shirt (Nep bhoÊo) ao n. live. ˙ini. to survive ˙ina (Nep) n. to beg aidzi n. pigeon ˙mbyaq n. to embrace a≥gi n. ˙niyu (neg) vi. vegetable abi a-t-ßn. dog pup ai ˙di adv. ˙igi-d-i. to urinate a≥ga ˙-go (imp) vt. to pierce. followed by first person past (if one exists) (usually -d-i). to squeeze ˙rtsa n. to be. armful ˙ol˙-d-i. to rescue ˙iga vt. root of the verb 'go' (see also ts-˙g-˙n. to squeeze udder) amdzi n. ao-da ˙-yin (proh) vt. other aidzi / ˙idzi n. ao-da ˙-go (imp). potato amaÚ a. living ˙gi-da-d-ßn.’ In some cases the variants represent different interpretations by different members of the research team. below amba / ˙mba n. garlic ˙rdzo ˙-g-˙n vt. to sell ˙raq / ˙ra≈ n. Entries with multiple words separated by a forward slash (/) are variants of the same word. to milk (lit. nostril ˙uba≥ ˙-go vt. Forms ending in ˙-g-˙n are third person realis forms. above ~ ag˙i n. to carry adze pp. ˙ol˙ ˙-go (imp) vt. goat akp˙i / akæpe pp. milk ambu tsilip ˙-g-˙n vt. to live. to burn (of food) a~ola (Nep) n. There is no citation form for verbs. ai-yin (proh) vt. mirror ˙itsi n. urine an˙-d-i. carrying net an n. ai-to (imp). ˙gi-da-d-i vt. root of the verb 'talk. bring' (see also ts-iw-˙n and -i) ˙igi-d-ßn. mango -˙na suff. to make live. dove. from when. ai-d-i. For example. imperative form of 'fetch. always ˙sa i. hole in ground ao-da-d-i. ˙saÚ i interrog. this manner ai-t-ßn. d-˙g-˙i) ˙gi adj. how ˙sae adv. become (Nep b˙n-nu) ˙≥gago n. finger 139 A . needle ˙sa interrog. amaqa pp. ˙igi-no (imp) vi. abi ˙-go (imp) vt.

climb down b~aqa n. to winnow bem-ßn. ginger beg˙i n. spill byo≥-dzi. noose. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– aoqa pp. under. ahun n. bæorl˙-d-i. b˙l˙-go (imp). uncooked bew˙-d-ßn. on top of aosi / aose / ˙usi n. uphill at n. younger sister biy~a≈ n. sparrow bæ˙psar n. trip bem˙-g-˙n vt. younger brother bæab ˙-go (imp) vt. to emerge bultsum n. sickle (kæurpa) beg˙i n. sky. spill byaq-ßn vi. pull in two bæ˙ra-q-ßn vi. beard ata gida≥ / atta gita≥ n. to boil water bæyon-to vi. footprint. halt BH C (See TS) . to dawn bibi-dzi. b~akæ˙ n. to be born bi it-ßn vi. b˙l-to (imp) vi. spicy. outside bai n. headstrap. byo≥-to (imp) vi. snare ata gi n. track a-t-ßn. bæorl˙ ˙-go (imp). a-d-i. to walk bin˙i n. b˙l˙-yin (proh) vt. hot byagorok n. door ape loc. b˙l-tsi-n. radish byaq-a-t-ßn. ˙-go (imp). wrist. to break rope. arm aw˙n / ˙hun. to mend. mustache. bæorl˙ ˙-yin (proh) vt. Kusunda beg˙n n. hand. type of bird bul-dzi (3rd) vi. bew˙-d-i vt. mouth ata≥ n. to carry wound in a coil bæorlo-q-ßn vi.Watters: Kusunda Grammar. to bear child 140 B bæ˙~îra (Nep) n. beneath asne pp. to stop. thumb budun n. patch bæi≥ga ˙-g-˙n vt. saliva (lit. ˙-yin (proh) vt. type of bird (Nep l~acæe) bal-ßn. bw~akæ. tobacco leaf budi a~ola (Nep) n. bæ˙ra-d-i vt. wide b~adza pp. to squat b˙ adv. byaqa ˙-go (imp) vt. byo≥-tsi-n. to fall. to weave (lit. elder sister b~akæa. there aqa / akæa pp. fox bæ˙ra ˙-go (imp) vt. also b˙k˙k ˙-go vt. imperative form of 'say' (see also ts-oˆ -on) awi / aw˙i n. byo≥-ni-n (2nd). to uproot bw~ak. down. burm˙-go (imp) vt. year b˙t vi. spicy. 'do loom') burm˙-d-i. flea bun˙ ˙-g-˙n (3rd) vt. feather bew˙-tsi-n vi. to boil bæorl˙-d-ßn. to descend. atta n. to fall ben adj. to remove from fire b˙m n. to overturn. hot beho vt. to fold b˙l˙-d-i. raw. frog b˙s˙ (Nep) n. byaq-a-d-i. pepper beg˙n adj. cloud b˙mlu / bamlo n. to be (predicational) b˙tsæ˙i adj. mouth water) ata. bw~a≈ n. to make at-to vt. to overturn. bem-to vi. goiter bya adj. bem-tsi-n. bibi-to vi. chilli. father-in-law bukæra n. to break (of rope) bæ˙ya n. fingernail aota / awata n. to snatch bæ˙ra-d-ßn.

to shut. to hurt. to make stand DH dz˙g˙l n. DZ . to seat s. be sick dz˙≥g˙l (Nep) n. strike d˙s˙-d-i. basket doma n. cliff dibæa≥ n. ashes dzidzyon vi. splinter dzo -a-d-ßn. fog dæya n. ground. brother’s son 141 dæ˙b˙-d-ßn. coal dz˙pak n. dæ˙b˙-d-i. wild man dzaÚ-d-ßn. to trap dæapa gipan n. dæaÚ ˙-go (imp) vt. dus˙ ˙-go (imp) vt. d~ab˙ ˙-g-˙n (3rd) vt. d~aba-d-i. dz˙u-tsi-n.o. sari d˙≥ga n. set load down dahat num. ground mushroom dui n. kaiu n. charcoal. to cover. sand. dz˙u ˙-g-˙n (3rd) vi. a kind of tree dowa-d-i. dæ˙b-in (proh) vt. dzapin ˙-go (imp) vt. three da≥bwa adj. get d~ab˙i / da≥b˙i n. dzogoro≈ n. fire dza h~oÚ-d-i vt. dzapin ˙-g-i (3rd). find. d˙bakæ. to soak. causative suffix. three da vi. Badi (beggar caste) doko (Nep) n. to put on fire dæundi n. make wet dzo≥dzo ˙-g-˙n vi. weeds dz˙≥t˙l / dza≥tal n. dark deidzi n. earth dus˙ ˙-d-i. purposive marker d~aba-d-ßn. close dæap ˙-g-˙n vt. to burn fire dzapin ˙-g-˙n. sour d~odzil˙q n. sickness dz˙u-dzi. smoke dimtu / d˙mt˙u adj. walking stick. d˙s˙-go vt. to do drum) duga. to hit. imperative form of 'go' -da aff. dug˙ n. dæab˙ (imp). to meet. mountain. man dukæu num. old woman dz˙i migo n. drum (Nep madal) d~ow˙dzi ˙-g-˙n vt.. floor dug˙ ab˙q n. soil. traditional singer dz˙≥˙n / dza≥ adj. oil dzi≥a num. dza ˙-go (imp) vt. dui getse n. dowa-go vt. pain. d˙wak adv. shaman daha-d-ßn. to stop d~ow˙dzi / du≥dzi n. whip da num. rhododendron dæaÚ a-t-ßn. to quit doing. dæaÚ ˙-d-i. old man dæ~u-tsi-n. plural suffix. d˙ba≈. Gaini. husband duidze. to leak dzogoroq. to leave. firewood deosi. abandon dzandza n. feel pain. to beat drum (lit. to push (See TSH) d˙baq. later today diba≥ n. two duktsi / duktse n. to bloom. dæu≥-to (imp) vi. third person form of 'go' (see also ts-˙g-˙n and da) d˙i ˙-g-˙n (3rd) vt. now d-˙g-˙i vi. son. object marker. d˙b˙si adv. human male. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– CH D dum n. to stand dæ~u-wa-d-ßn. daha-d-i vt. to shoot a bow d˙rpu ˙-go (imp) vt. blossom dzi≥ n. two dziu a-t-ßn vt. crowd dimi n. dzaÚ-d-i. dæ~u-wa-d-i vt.Himalayan Linguistics: Archive No. new dza n. to buy dzei / dz˙i n. a kind of yam dz˙u n. dzo -a-d-i vt. release d˙~ure. to bear fruit.

tsi-m˙t. to blow on fire dzugui / dzug˙i n. git˙t n. he. child getseb˙i n. sap gidzaq loc. odor. imperative form of 'see'. belly. e-go (imp). fat gis˙k˙la n. wife’s sister. g˙r˙m-dzi vt. 'smell comes') g˙r˙m-d-i. tobacco g˙~utsi n. intestine gesala n. child getse. name gid˙t. between gidza≥ n. curd gem˙t dz˙u n. aroma g˙ndzibur n. to force g˙ki n. high 142 dzæ˙m ˙-g-˙n (3rd). breast. bond friend gida: (Nep) n. fur giban n. wife’s younger brother gesali n. oats gita≥ n. dz~ow˙-d-i vt. offspring. money (< gimi haq ‘female leaf’) gimi e-d-i vt. look' (see also ts-aˆ -dzi) G . stomach. self. dz~oˆ -a-d-i. e-yin (proh) vt. forest mushroom gila≥ tap n. to get wet dz~ow˙-d-ßn. dz~oˆ -˙-g-˙n (3rd). to earn gimtsi n. upper back g˙i (Nep) n. wound g˙ini adj. friend gimihaq-ni / gimya≈ni n. eblaq / emblak n. language gisi n. dzoq-to (imp) vi. kidney g˙ndzi n. that gina pron. grandchildren gebusa n. maternity garo n. whetstone g~aˆ -no (imp) vt. wall gebæusa n. female Ban Raja gin a-di adv. it gina-yi pron. hair. to give eirak adj. to weigh dzyah~ade n. daughter-in-law gi n. word. gæokæba adj. body gidzi n. she. illness dzulo n. bad smell g˙ndzi ˙-go (imp) vt. bark. uncle gimdzamba / gimtsamba n. a kind of tree ga≥no (imp) vt. baby. ni-gim˙t n. sickness. father’s brother. cry dzæ~o n. stench. to knock down g˙sa n. to smell. smell. gæi≥gæyak n. to hang dzw~a ˙-go vt. female of a species gimi n. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– flourish dz~oˆ -a-d-ßn. dz~oˆ -˙-go (imp) vt. witch ey˙≥ n. his own gimi n. h~ukui n. diarrhea gem˙t. trousers gipan n. seed gida≥ n. to weep. to pay (lit. to encircle. louse egg DZH E e-d-ßn. to hang dzoq-ßn. to smell something g˙ndzi u-g-i (3rd) vi. sweets gitsi n. sister-in-law getse n. thick g˙kæya ˙-go vt. poor emblaq. fence in g~aˆ ˙n ye≥gu n. e-d-i. tinder dzu≥-dzi (3rd) vi. tall. dzæ˙m-in (proh) vi.Watters: Kusunda Grammar. forest gila≥ ab˙q n. his gi¯aq / gi≥yak. give off odor (lit. that manner gina dem. flower gip˙n / gep˙n n. dzæ˙m-to (imp). skin. milk gebon n. leather gila≥ / gela≥ n. Kalij Pheasant gimdæ˙r n. police gimdzi pron. 'to give money') gimi ˙-go (imp) vt. thorn gobba / gæo≈ba≈.

to move gæue ˙-go (imp) vt. gæ˙r˙-go (imp) vt. little bit gudz~a≥ vt. grass gæu a-t-ßn. sheep gol˙≥d˙i n. guhu n. gya ˙-go (imp). gwi ˙-go (imp) vt. h˙mdzi n. gya. third person form of 'put. gæo ˙-go vt. gæ˙r˙-d-i. to chase gune≥gi / guna≥ge n.a. egg gwa ˙-g-˙n vt. round gum˙-d-ßn. rhesus monkey hai-t-ßn. heat gæ˙sa n. gyaÚi / gyakæa n. got˙≥ n. goro≈ n. bone marrow guidzi / goidzi n. hot gæ˙r˙-d-ßn. to sting gæya≥ n. male of a species gya numba n. hold gya. gæ˙r˙o adj. raksi gæ˙run. green h˙r˙-q-ßn vi. mother-in-law gæ˙i n. gorokæ. to gather. pus gyakæ˙r ˙-g-˙n vi. bat h˙r-a-d-ßn. to grind gæak ˙go vt. to snore ha≥nu≥ n. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– gobloq n. brain gyaq-ßn. gæ~aÚ-d-i. face h~aÚ / tsi-ha≥ n.Himalayan Linguistics: Archive No. humming bird h˙ila≥ ˙-go vt. tobacco gæa-o (imp) vt. allowance gulu≥ adj. to gather gya n. forehead hab˙-d-ßn. tomorrow goto≥. gyaq-to (imp) vi. gum˙ ˙-go (imp) vt. gæ~a-yin (proh) vt. hai-dzi (3rd) vi. to thrash. beat gæ~aÚ-d-ßn. to take guhu masi n. ata qaÚi n. lung godoq / gorok. keep. gyaq ˙-go (imp). gyaÚwo n. where hampe kæaÚu pron. root godza≥ adv. to make hot. soup gu. to copulate GH h˙b˙-o (imp) vt. set aside' (see also ts-o -on) goloq / gol˙≈ n. imperative form of 'sit' 143 H . log gudza≥ adv. hara≥ ˙-go (imp) vt. save gwi-t-ßn. shadow haˆ -no vi. cause to open h˙rdze a-t-ßn vt. gya-d-i. morning. to win h˙rgun adj. heart. strength gya-qai. to plough gæurma ˙-go vt. gya ˙-yin (proh) vt. gyaq ˙-yin (proh) vt. to rest gyao n. gwi ˙-d-i. 'do egg') gwame n. hab˙-d-i. ox gya-d-ßn. to hold in hand h˙g˙dzun n. to open on its own h~aÚ n. boil gæ˙run ta≥ n. to pierce gæan ye≥gu n. gum˙-d-i. grindstone gæ~as (Nep) n. female genitals gæy~a-d-ßn vt. to shake h˙m˙tsi / h˙mitsi. to burn. bone guben n. to press. g-o -an (irr) vt. nowhere hanti n. roast ha o. breath gya o. adams apple ha≥ki ˙-g-˙n vi. ha wo n. few g-o -on vt. langur monkey gwa n. soya bean goradze adv. centipede gwi ˙-d-ßn. hab˙-go (imp) vt. to reduce g-ug-un (3rd). neck ha≥ki gun n. to move gæu-t-ßn vi. tortoise gulundi n. cheek h~aÚ / ha≥ n. to lay egg (lit. gyaq-a-d-i. turtle. early today goraq / gor˙≈ adv.u. gwi-tsi-t-ßn vi. gæ~aÚ ˙-go (imp).ai. to open. to yawn hampe interrog. cummerbund ha≥ki n. to collapse gyaq-a-d-ßn. h˙r-a-d-i. to make collapse gyaudzi n.

return in-da-d-ßn. to cook hul˙n n. hunger t˙n-da ida≥ 'I am hungry' (lit. leaf. to swell hyo-w˙n. to doze i≥ i≥-dzi adj. night i≥dz~u. Sarki. pig. paper note (lit. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– (see also s-aˆ -an and ¯-aˆ -an) haq / ha≈ n. root form of the verb 'sleep' (see also ts-ip-ßn and n-ip-ßn) ip˙n n. eyeball in˙≥ gi n. wild boar hi gimi n. maize ip-da-d-ßn. imb˙ ˙-go (imp) vt. throw away huÚ-dzi. salty h~uku h~uku adv. imperative form of 'sleep' ira≥ ˙-go vi. hul˙-go (imp) vt. eyelash i≥ n. swollen hyo -a-d-ßn. salt huki am adj. small h~ukun kolde n. Tharu ~ inu / in˙u n. imba-d-i. shoemaker caste (lit. remember imb˙-d-ßn. to be unable to hyoq-ßn. sun i≥ / in˙≥ n. to exchange. iskin adv. to chase hunko adv. far hu≥k˙i bai n. hunger is to me) idu / id˙u n. in-da ˙-go (imp). to hide hyoq-a vt. knife h~ukun ˙ota n. root form of the verb 'put in' (see also ts-ig-˙n) 144 I J (See DZ) . a kind of tree hoig˙n.vt. almost hur˙-d-i / hurra ˙-g-˙n (3rd) vt. spade. to hide. near iyu n. little bit h~ukui / hu≥k˙i adj. to there -it. to throw. tongue i≥gidat / i≥gid˙t n. ip-da ˙-go (imp) vt. wheat flour ip-to vi. hul˙-d-i. head ipi gidza≥ n. huÚ-do (imp). eye i≥ gida≥ n.Watters: Kusunda Grammar. huÚ-in (proh) vi. root form of the verb 'say' (see also ts-it-ßn and n-it-ßn) ist˙ adv. deaf -i vt. book haq a-d-ßn vt. corn. be able hyu-dzi adj. hyo -a-d-i vt. liver -ig. paper. litter hugya≥ n. to read haq gimi n. to set (of sun) i≥ao / i≥au n. filthy. bring' (see also ts-iw-˙n and ˙iga) ida≥ n. horn ipi guhu n. rubbish. in-da-yin (proh) vt. evening isna pp. idzi≥ / idz˙≥ n. dirty. to feed. hyo-da-d-i vt. Newar ima n. imb˙ ˙-go (imp) vt.vi.vi. fat ih˙≥ / eh˙≥ n. ear iyu ˙ndoq adj. nose in˙≥ getse n. to put to sleep ipi n. hyo-dzi adj. window hul˙-d-ßn. sow pig hila≥ n. shovel huki / hukki n. farmland imala n. to think. ip-da-d-i. huigin adj. female leaf) het ˙-go (imp) vi. to fly h~utu adv. in-da-d-i. hyoq-tsi-n vi. father’s third brother ~ isi. foot leather) -ip. skull ippa kadzi n. to call out hi / he n. tears i≥ ipdzi vi. imb˙-d-i. the youngest elder sister hyo-da-d-ßn. sleepy i≥ miÚ-dzi vi. nurse înor˙t / ino˙r˙t n. root form of the verb 'fetch. cook humba ˙-g-˙n vt. field. father’s second younger brother imba-d-ßn. to stretch îsala n.

flasehood kum kum ˙-g-˙n vi. k˙la ˙-go (imp). kola-d-i. louse kila-d-ßn. kami koyu n. crow koidzu n. cooked rice kæaiya. qaida n. Sesi Kham kirna (Nep) n. kola ˙-go (imp) vt. crab katse gumbo n. scorpion kat˙≥ n. again k˙re / k˙ri n.Himalayan Linguistics: Archive No. a kind of tree (Nep bæ˙layo) ki n. a kind of tree (Nep sajæ˙) kasi. to make tall koˆ odi / k˙~ud˙i n. to pile up kaitæwa ˙go (imp) vt. to cut. to tease kon-da-d-ßn. to parch grain kæaidzi n. to one side kan adj. other k˙la-d-ßn. to offer k˙ita. kæa ˙-go (imp) vt. to bite. bad kolumu / kolomu n. lie. to separate k˙la-q-ßn vi. kumba-d-i. to sharpen kapu n. Brahmin kæ˙rwi / kæ˙rug˙i n. to lie (lit. kila-d-i. kampe pp. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– JH K (See DZH) k˙ba n. ≈˙≥gu adj. to quarrel kya. to give drink konda-gi vt. to break. whistle kulum n. Blacksmith. kila ˙-yin (proh) vt. khukri. to hold under arm k~aw˙r adj. everywhere kampya. white. to raise. to promise god kaitæo ˙-go vt. kæam-tsi-du. owl komba komba ˙-g-˙n (3rd) vt. k˙la ˙-yin (proh) vt. thief kiptsak n. sharp kan˙ ˙-go vt. vase. to separate k˙l~a ˙-g-˙n vt. rear kola ˙-g-˙n (3rd) vt. gold k˙bule vt. kiw˙ ˙-go (imp). tumeric. light weight kaw˙t n. k˙la-d-i. to distribute. a kind of fruit tree (Nep kapæ˙l) kasa n. water container k˙reÚ n. knife kolom. k˙la-g-˙n (3rd). thin. itch 145 kawa ˙-go (imp) vt. skinny kæ˙rgun adj. to dump k˙mboi n. to cause to drink. yellow k˙pæera adv. to break. to say a lie / to do a lie) k˙bdza≥ / k˙pdza≥ n. kæa ˙-g-i (3rd). k˙la-q-dzi (3rd) vi. bangle kayak n. kasige / kasigi adj. jug k˙r˙≥ (Nep) n. falsehood k˙ba aÚdzi / kaba-dze ˙-g-˙n vt.a≥ adj. lie. divide k˙l˙≥gu n. to cut meat. kæam-dzi. danger kæam-tsi-n. k˙la-d-i vt. kon-da-d-i vt. to stir kola-d-ßn. leprosy kampa kampe adv. wheat kæ˙≥gu. clear brush kiw˙ a-d-ßn. rib k˙u ˙-g-˙n vt. k˙mba ˙-go (imp) vt. tick kisa-d-ßn vt. smash (as a jug). hornet k˙pa≥ n. chew KH . cold (of food) kæa ˙-d-i. ≈aiya n. to pinch kog n. light color katse n. kæam-in (proh) vt. to steal kiladi n. kæam-to (imp). split wood k˙la-d-ßn. smash k˙la-q-tsi-n. Jugi k˙mba-d-ßn. red kæ˙rk˙la n. to draw water kolde n. kiw˙-n (proh) vt. food. qolom adj. to join kamk˙m n. kila ˙-go (imp). pot. kila ˙-g-˙n (3rd). red deer kok ˙-go vt. to water animals kondok ˙-g-˙n vt. besar k˙pa≥ b˙yo n.

to be stamped. imperative form of 'forget' (see also ts˙-mi-n) miÚ-t-ßn vi. limi ˙-go (imp) / l˙ima ˙-go vt. mami n. to lose -mi vt. lib~u-nin (proh) vi. m˙d˙u n. meÚ-d-i vt. to plant la an / la≥ n. quarrel li≥wa adj. elder brother mo≥ / mu≥ n. m˙Úb˙-yin (proh) vt. to taste la~î / l~a¯e. anxiety m˙ina (Nep) n. to write leÚ n. heavy lol˙ ˙-d-ßn. leÚ ˙-go (imp) vt. laÚ ˙-go (imp) vt. listen m˙≥ / ma≥. to swallow motsa n. mother m˙imi n. teach l˙p-dzi vi. buffalo m˙hi-dzi adj. topi mame. blood vessel mehaq. not (existential) kæola (Nep) n. to adhere. to stamp. village laÚ ˙-d-i. elder sister’s husband m˙nni / m˙ni adv. to make dance. to be shy l˙ a-d-ßn. to drown miu / myiu n. long lab ˙-g-˙n vt. make hop around limit-ßn. to do a lie) leÚ ˙-g-˙n (3rd). limi-to (imp) vi. m˙la ˙-yin (proh) vi. be lost mihaqtse / mæaq-tse / maktse n. song m˙≥ ˙-g-˙n vt. to go mad kæurpa n. month m˙iti ˙g˙i adj. l˙ppa n. with malu n. bread maqæadzi / ma≈kadzæi n. mixed m˙i n. ring M . lol˙ ˙-go (imp). to bury l˙mba-d-ßn. to play libu≥dzi n. wild cherry lekæa-d-ßn / lekæe-d-ßn vt. print l˙b˙-d-ßn. gemehaq / gimya≈ n. knife kæya ˙-g-˙n vi. only m˙Úb˙-d-ßn. lol˙ ˙-d-i. limu-tsi-n vi. to apply. cucumber l~asi n. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– kæaÚu. naked m˙la≥ adv. to cry out 146 L m˙hi a-t-ßn. lie. stream kæomba adj. arrow mom n. to make mistake mitæu ˙-g-˙n vi. second elder sister m˙la a-t-ßn. l˙mba ˙-go (imp) vt. to sing (lit. blood l˙≥g˙i / l˙nk˙i n. father’s younger brother’s wife m~a≈ / m~akæ n. stick l˙p-ßn vi. l˙≥kan n. imprinted l˙pa. to lose. to be late m˙ndu. root form of the verb 'forget' (see also ts˙-mi-n) miÚ-to vt. Nepali cap. l˙b˙-d-i. to dance limu-dzi. to hear. 'do song') -ma aff. lol˙ ˙-yin (proh) vi. m˙la a-d-i. lib~u-tsi-n. slowly m˙ta. Ban Raja meÚ-t-ßn.Watters: Kusunda Grammar. l˙ ˙-g-˙n / l~ah~ago vt. rub in l˙bu u-g-˙n / l˙bbo vi. m˙ih n. ma≥la≥ n. dream m~a≈ ts-~aˆ -dzi vi. king motæa ˙-go (imp) vt. to move lib~u-dzi. to lie (lit. m˙ adv. l˙b˙ ˙-go (imp) vt. mad kæomba d˙mdzi vi. nettle. m˙Úb˙-go (imp). to mix m˙hi. burlap lime-d-ßn. limi-tsi-n. m˙hi ˙-go vt. slowly m˙lli bai n. cigarette l˙≥ka. player lig˙n / ligin n. see a dream) matsa n. to learn. lib~u-to (imp). to be angry (lit. l˙mba-d-i. falsehood leÚ ˙-g-˙n vt. many. much m˙st˙m˙ adv. to dream (lit. anger to come) l˙b˙-d-i vt. to fight. khukri. boy mitæi≥ vt. to sweep libæ˙ ˙-g-˙n vi. laÚ ˙-n-i (2nd). m˙Úb˙-d-i. banana. ≈aÚu v. l˙≥k˙i adj. plantain mundo (Nep) n.

cat (see also ts-it-ßn and -it-) n-iw-˙n vi. mwø ˙-d-i. nigu n. mwø ˙-go (imp). moon nimbu n. prohibitive form of 'forget' (see also ts˙-mi-n) ni≥gitse / ni≥tsi n. alone myaq˙-t-ßn. calf nupa pp. ¯aq-ni-n (2nd). mother’s brother’s wife nyam n. second person form of 'speak' n˙bæ˙i n. mwø ˙-yin (proh) vt. ¯aq-to (imp). this (animate thing) nep n. to boil food myaq / mya≈ n. to kill (see also or˙-d-ßn) o tsi. myaka adv. second person form of 'go' n˙n-da pron. thou numba n. nikæe-d-i. to hold in two hands nyabi. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– murut˙≥ ˙-g-˙n vt. human female. fish n˙sa (Nep) n. ¯~aÚ ˙-yin (proh) vt. grandfather n-ig-˙n vt. whose na dem. which. chest . person. nyabe n. o -da-d-i vt. above n-o -on vt. n~owdze pp. nothing n˙n noh˙n conj. second person form of 'put. grandmother nya ˙-go (imp) vt. mother’s brother nyase n. what. wife ¯~aˆ didze. man. lemon ni-mi-n vt. you plural nokok. to rise. accusative case of second person pronoun n˙n kæaÚu pron. no one n˙ti-ye interrog. mankind nu. up. myaq˙-go (imp) vt. ¯~aÚ-d-i. myaq˙-g-˙n (3rd). n˙b˙ ˙-go (imp) vt. second person form of 'put in' (see also ts-ig-˙n and -ig-) ni-gimdzi pron. nikæe ˙-g-˙n (3rd) vi. get up nwatsiu adj. n˙tßn interrog. occasion n-˙ ˙-n-i vi. roll over mu≥ni n. nu≥-to (imp) vi. to release muru≥-to. myaq˙-d-i. yours n~odze. there nu≥-tsi-n. remove. ¯~aÚ ˙-go (imp). maternal grandfather nu. your own nig-i pron. woman ¯-aˆ -dzi vt. not (equative) o -da-d-ßn. ¯~aˆ di getse n. cow numba getse n. ¯aq-nin (proh) vi. no≈ok n. extract. deep nya n. queen mwø ˙-d-ßn. pick up ¯~aˆ di / ¯~a≥di n. to marry n-˙g-˙n vi. niu n. you. set aside' (see also ts-o -on) nok pron.Himalayan Linguistics: Archive No. because n˙n. daughter n-ip-ßn vi. who n˙ti kæaÚi pron. leopard myaqa / myakæa. to turn over. to lift. why n˙sa n. to dig myaqo getse n. second person form of 'sleep' (see also ts-ip-ßn and -ip-) n-it-ßn vi. oÚ tsi n. n˙n-da pron. to wait ¯ O odoq v. expel. yours plural nikæe a-t-ßn. upper body. to laugh niku. second person form of 'see' (see also ts-aˆ -dzi and gaˆ -no) ¯aq-tsi-n. feast n˙b˙-d-i. human. second person form of 'say' 147 N ¯~aÚ-d-ßn. prohibitive form of 'go' and 'bring' (see also -˙g and ˙iga) ni-yi pron. m˙r˙≥ ˙go vi. sinew n˙ti adv.

pita-n-i (2nd). long ago pit ˙-go (imp) vi. to help. bile piÚ˙gu num. 'do return') p˙tse e-d-i vt. give aid pæeb-a-d-ßn. give back (lit. leafy vegetable. before. big oˆ la-d-ßn. sweet. hay p˙≥ga n. spider p˙≥gyu n. medicine ola-d-ßn. pæeb-a-d-i vt. four piw˙ ˙-go (imp). po≈la-d-i. pita ˙-go (imp) vt. four p˙idzi n. pyene / peni adv. to give back (lit. pæ~˙ ˙-g-˙n vt. a kind of tree (Nep malu) p˙o-tsi-n vi. knock over . last year pyai pyai adv. iguana pumba-d-ßn. to peel p~otæa ˙-go (imp) vt. to feel drowsy pinda adv. old pwahan n. to step on ola≥ adj. to kill (see also o -da-d-ßn) oran b˙s˙ adv. po≈la ˙-yin (proh) vt. pumba ˙-yin (proh) vt. next year osenti n. bamboo p˙yet n. to return. oˆ la-d-i. om˙q / omok / omokæ n. long ago pyana. p˙tse ˙-go (imp) vt. to want oˆ ta≥ n. po≈la ˙-go (imp). onion pyai d˙g˙n b˙s˙ adv. wing. to finish (be consumed) pæ˙≥. ora-d-i. tailor p˙nahak n. to untie pitta n. spirit otok vt. to beat pumba-d-ßn. to leave. to scrub outside pimba ˙-go (imp) vt. pluck paÚtse / patsæe pp. pumba-d-i. to be tired p˙rm˙-d-ßn. downhill pa≥dza≥ num. pilgrim omoq. or˙-g-˙n (3rd) vt. cloth. woman’s garment ~oÚ¯i adj. to hit. shin p~aÚyi / p~ai n. oˆ la ˙-go (imp) vt. ola-d-i. pumba-d-i. 'remove clothing') p˙idz˙bo n. lentil (Nep mas) p˙itoba / p˙itokba n. girl otisge n. Damai. in front of pinda duktsi n. to bathe pui n. tasty olum n. depart pita-d-i. to search for. piw˙ ˙-yin (proh) vt. earlier pyadz (Nep) n. to count pimpi≥dzi vi. 'give return') p˙wi / p˙ui / p˙u˙i n. afternoon ~ oˆ ¯i n. leech p˙y~akæ n. ghost. to undress (lit. five p~asula (Nep) n. umbrella oqtsi. strike purano (Nep) adj. cloud pewa ˙g˙n vt. to jump over po≈la-d-ßn. ola-g-i (3rd) vt. owts˙ / ˙~utsa adv. paÚ ˙-go (imp) vt. clothing p˙idzi ¯~aÚ ˙-go (imp) vt. bed pæ˙u-n vi. yesterday pya≥dza≥ num. chest or˙-d-ßn. sleeve om˙q / oma≈ n. pumba ˙-go (imp). a tree mushroom paÚ-d-i. blouse. to reject papdore ab˙q n. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– okti (Nep) n. to pick. o tsi. five 148 P PH pæ˙laq n. smoking pipe oˆ ts˙.Watters: Kusunda Grammar. upside down p˙tse a-d-i. arm. eldest son pinda pinda adv. to return. hard p˙tsæe haq n. lizard p~ago num. cemetary pya adv. to digest pæ~a ˙-g-˙n vt. p˙rm˙-d-i vt. to make fall. graveyard. pumba ˙-go (imp). green leaf p˙ts˙idzi adj. pumba ˙-yin (proh) vt. to meet p˙s˙o adj. branch. oÚ tsi / okætsi n.

that direction sa. there. qaidzi adj. skirt qo e-d-ßn. koÊou n. pæira ˙-go vt. seÚ ˙-go (imp) vt. steam qaÚi hur-˙n vt. pine tree sabe pp. to bow. qau-da-d-i. nest qoturaq adv. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– pæelad˙≥ n. to burst pæwotoq adj. qo e-d-u vt. pæu-yin (proh) vi. pumpkin r~ako. qauda ˙-go (imp) vt. pæeb-in (proh) vi. buttocks s˙mtoq n. bird’s beak qotu wi n. dry. drag s˙ ˙-go (imp) vi. bitter qatraq n. s~˙li n. to tear. short pæurlu≥ adj. to sit sat n. to run pæuÚ-dzi. to drop. qæai-nin (proh) vi. ruÚ˙-go (imp) vt. qai un. Chettri r˙nta n. ra≥kwa n. full pæo a-d-ßn. pæurpæu-to (imp) vi. pæep-to (imp). fall over pæira-d-ßn. qæai-dzi-n. to wash clothes pæyaks˙m n. to salute s˙i n. day after tomorrow QH R qæai-d-ßn. be torn qas˙-d-i vt. to fear. behind s˙mba n. imperative form of 'drink' (see also ts-oˆ -on) qotu / kotou. flat pæel~ade n. pillow qaoli n. strike. to tear. to drop pæep-ßn. itch qaÚsßn num. qæai-to (imp). be frightened q˙mba ˙-d-i / q˙mba ˙-go vt. pæep-a-d-i vt. pæuÚ-d-i. third elder sister s-aˆ -an. to 149 Q r˙ktsa / r˙ktsæya n. to pull. rotten qawai n. millet rambenda (Nep) n. rain shield sen n. to make decay qaudzi adj. qo e-d-i. rip off qasti. nasal mucus ruÚ˙-d-i. air. low pæya ˙-go (imp) vt. son-in-law qau-˙n vi. squirrel s˙~uli. together qaien. lentil (Nep g˙g˙t) pæela≥ adj. paddy seÚ-d-ßn. to hit. to decay qau-da-d-ßn. throw breath) qamtsi n. adj. to fill pæirun. one qas˙-d-i vt. pæo ˙-g-˙n (3rd) vt. red pæurpæu-d-i. to tear qato / kat˙u adj. to move from one place to another qomba-a-t-ßn. swing r˙≥gunda / r˙mkuna n. jackal qayo n. qaÚsßn num. pæokto adj.Himalayan Linguistics: Archive No. wither qaÚi / kai n. qomba-a-d-i. one qaÚs-ßn vi. to jump pæwa ˙-g-˙n vi. to move.am / sam n. wind. haˆ -no (imp) vi. to S . bear s˙mba adv. star sali bai n. mushy argue qom˙-d-i vi. after. horse pæyal˙m adj. seÚ-d-i. bird qotu uhu n. beaten rice pæep-a-d-ßn. to breathe (lit. comb sat ipi b˙n˙-d-i vt. qæai-ni-n (2nd). soft. to pour qa tse adv. pour from one container to another qon vt. god qaÚn n. pæu-to (imp). qomba ˙-go (imp) vt. millet ruga (Nep) n. later. tomato ran n. pæirut-ßn v. to shoot a gun pæotoq / pæoqto≈. to comb hair segu (Nep) n.

to wash pots supo n. ours toba ˙-d-i. water ta≥ dziu-n. to dress tul-a-d-ßn. day before yesterday tu n. sip-in (proh) vi. tasi n. chicken tap getse n. n-˙ ˙-n-i (2nd) vi. sola ˙-go (imp). ˙ga (imp). d-˙g-˙i (3rd). n-iw-˙n (proh) vi. sutta n. chick tap gimi n. ni-mi-n (proh) vt. tul˙-yin (proh) vi. to know tul-tsi-n. ts-˙ ˙-d-i. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– catch siba. sola-d-i. rain. ni-mi-di (2nd). tamb˙ ˙-go (imp) vt. here tigi-i pron. siwa adv. n-˙m-in (proh) vt. accusative form of first person pronoun t˙wa n. sip-tsi-n. to die to on adv. this (inanimate things) tahu-dzi vi. speak ts-˙g-˙n. am-to (imp). to go ts˙-mi-n. soq-tsi-n. to break (as a pencil) solo-q-ßn vi. to cause to climb soq-ßn. sib-a-d-i. fast sib-a-d-ßn. greetings sok ˙-go vi. tul-a-d-i / tul-a ˙-go (imp) vt. am. hen 150 T tæ˙w˙ / tæ˙wa n. miÚ-to (imp). to pierce (by a thorn) tæum˙n ye≥gu n. winnowing tray suta. ladder sowa-d-ßn. to leak ta≥ p˙iyo n. boys and girls tasin. 'thirst water') tei sa≥-dzi vi. bow twi n. black sola-d-ßn. sib-a ˙-yin (proh) vt. to enter sisen / sisin n. soq-a-d-i. sowa-go vt. plate tæamdzi pp. string. tul-to (imp). to eat ta dem. snake. n˙-du (imp). to forget TS . dziu-t-ßn vi. n-ug-in (proh) vi. beer. to reach tap n. soksodi adj. brew simi (Nep) n. to finish tut n. quickly. to hold tightly tap gya n. thread sw˙ttei / sw˙tt˙i adv. to talk.Watters: Kusunda Grammar. to extinguish t-ug-un. to remain teisa pp. to climb soqdzi yi n. between tæamdzi vt. o -do (imp). sib-a ˙-go (imp). soq-to (imp) vi. n-ukan (imp) vt. cock tape. rooster. sowa-d-i. sola ˙-yin (proh) vt. tul-ni-n (2nd). rope. sip-to (imp). insect tuba-o (imp) vt. to send t˙n-da pron. hello. place tæal (Nep) n. to graze soksog˙r˙m. pass through sidza≥ n. n-˙g-˙n (2nd). soq-a-go (imp) vt. to dress someone tumb˙-d-ßn. beans sip-ßn. we toldok n. to sew tok pron. all sya ˙-go (imp) vt. to enter. t˙ndin / t˙ndi≥ adv. here. to this direction taqsi n. today t-˙m-˙n. bug. paddy. honey t˙inan b˙s˙ adv. tapye pp. rice milk t-o -an. n-˙m-˙n (2nd). rice field si≥ki n. to break in two soq-a-d-ßn. n-o ˙-yin (proh) vi. bush ta≥ n. toba ˙-go (imp) vt. to come t-uk-an. t˙mb˙-d-i. ts˙-mi-di. soq-ni-n (2nd). this side t˙wa pp. tumb˙-d-i vt. this year t˙mb˙-d-ßn. bee twi-yi gida≥ n. thirst (lit. to here t˙~îna. da (imp). mortar TH ts-˙ ˙-d-ßn. eagle sodzaq greet.

¯-aˆ -dzi (2nd). to roam umb˙ a-t-ßn. snow yaqa. u≥ n. cold (weather) yaq n. road. father’s younger brother yebu n. mine ts-o -on. make a bed ya. yakæa. look tsi vi. to say ts-oˆ -˙n. steel striking stone) ts˙m˙k n. u. civet cat yeg˙mbu n. shit. tsæ~aÚ ˙-g-˙n vt. wey˙n adj. flint (lit. credit ts-iw-˙n. steel for striking fire ts˙maq tæag˙n ye≥gu n. to call un n. leg yen amba n. to see. un-da-d-i. to fetch. to show unibai u≥nibai n. to bring tsi-yi pron. n-ig-˙n (2nd) vt. set aside ts-oˆ -on. to milk tsim tsim ˙-g-˙n v. y~a ˙-go (imp). n-iw-˙n (proh) vt.Himalayan Linguistics: Archive No. dung ya≥ ˙-g-˙n / y~a ˙-g-˙n (3rd). path. g-ip-ßn (3rd). I tsidan n. to be (locational) tsi. ts-i-d-i.a / yakæ. my own tsilip ˙-g-˙n (3rd). umb˙ ˙-go (imp) vt. tea ukæi-q-ßn vi. yak˙l˙k adj. qon (imp). ˙iga (imp). porcupine ya≥ n. road un. tinder. 'leg meat') . to blow uhu. bedbug usni n. tsim˙ ˙-go vt. ip-to (imp) vi. to put. hail. to feel tsim˙-d-i. to sleep tsirma-t-ßn. to ruin ukæ˙ ˙-g-˙n vt. yaÚ u / ya≈˙u adj. thigh (lit. to be (existential) tsula (Nep) n. house witæu / oitæ˙u adj. tsæ~aÚ-d-i. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– ts˙maq n. sugar tsinu n. n-o -on (2nd). to defecate ya≥b˙ru / ya≥buru n. y~a ˙-yin (proh) vt. firefly wen. lower back. t˙n-da pron. knee. to tie. g-aˆ -dzi (3rd). father ya ˙-go vt. foot. to ruin ulum-to (imp) vi. tooth 151 y˙i / ei n. straw mat (Nep ghundro) ya si. tsilip ˙-go (imp) vt. to say (quotative) ts-iw-˙n vt. yaÚqsi / yase n. to bring ts-ig-˙n. trail. to squeeze. firepit tsya n. bind tsæut ˙-g-˙n vt. mud Y TSH U tsæ~aÚn. kneecap uris (Nep) n. Magar yekæolak. waist ya u. yam yeg˙mba n. trail unda-d-ßn. vomit wa≥ n. ya≈ n. n-it-ßn (2nd). to put in (Nep hal-nu) tsi-gimdzi pron. gaˆ -no (imp) vt. dumb yen n. n-ip-ßn (2nd). ya≈a n. good wi / uhi. n-oˆ ˙-yin (proh) vt. to spread. ui n.u n. eldest sister upto / opÊo n. a kind of tree ts-ip-ßn. Kumal y~atsa n. g-o -on (3rd) vt. wring. unda ˙-go (imp) vt. to touch tsini (Nep) n. girl W wak (Nep) n. at-to (imp) vt. to recognize ts-it-ßn. i-dzi (3rd) vi. slippery witæu dum n. umb˙ a-d-i. to cross river ub˙-go (imp) vt. tsirma-d-i. fire fibers ts-aˆ -dzi. to drink tsu / tsiu vi. tsirma ˙-go (imp) vt.

-gi aff. shoe (lit. to make tight ye≥ d˙i ˙-g-˙n vt. stone door) ye≥gut / yi≥gut n. soft yow˙n vi. -hi. 'foot leather') yeodze adv.Watters: Kusunda Grammar. to scratch yiÚ / ihi n. genitive yibiu ˙-g-˙n vt. weasle. rat. mouse -yi. one moment yer˙≥ ˙-go vt. yu≥-tsi-du. cave (lit. stone. tail 152 . pestle ye≥gu ˙ota n. tree yolaq adj. now. to kick (lit 'hit with leg') ye≥gu n. to ripen yu≥-tsi-n. stone) ye≥gu / yi≥gu n. yu≥-to (imp) vi. coin (lit. Appendix A — Vocabulary ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– yen gid˙t / yen git˙t n. to lie down yu≥s~u / gusu≥ n.

Past forms are generally -d-i. There are three persons (first. as well as third person realis. ‘I/you descend. and the former form is avoided altogether. as in bal-ßn. -d-˙n. only Class II verbs regularly allow it. We collected about 200 verbs. ts-ip-da-n. Because of our limited time.Appendix B Verb Paradigms Verbs in Kusunda are inflected for tense–aspect and for the person and number of the subject argument. -n-˙n. n-ip-da-k. g-ip-da-k ‘sleep. g-ip-du. For some Class I verbs. in which case the plural will be -dei. but have complete paradigms for only some of them. g-ipda-n. -d-˙k. The past tense distinction is missing. past plural forms in -dei have contaminated some Class I realis paradigms. Irrealis forms are generally -d-u. n-˙m-da-k. for many verbs we elicited only first person past.’ and ‘take. -n-i. -n-ßn. n-˙m-˙n. Class II intransitive verbs are partially defective. and third) in two numbers (singular and plural). Attempts to use it are interpreted as causative. unless the bal-tsi-n forms are past tense – but we have been unable to confirm such. Not all verbs allow a past tense. second. -n-i. as in ‘see/look. There is no distinction between third person singular and plural unless the singular form is -dzi. t-˙m-da-k. Realis forms are generally -d-ßn. and -g-˙k.’ Mutation in such verbs also marks dependency (see §7. 153 . -n-˙k. t-˙m-da-n. at least in our data set. n-ip-du. except in a few verbs like ‘forget’ and ‘fetch. Basic TAM categories are past. -n-u. realis. n-˙m-da-n. Other forms. g-ip-ßn. n-ip-ßn. past tense forms have infiltrated some of the third person forms. -g-i. Class II transitive verbs are the most regular and were the easiest to elicit. and -g-˙n. -g-u. g-˙m-˙n. n-˙m-du. n-ip-da-n. ts-ip-du. -d-ei. ts-ip-da-k. There is still much to sort out in this respect. In realis mode such verbs are generally bereft of person marking. g-˙m-da-k ‘eat’. t-˙m-du. whether transitive or intransitive.’ and where person marking does occur it is inserted between the verb root and the realis marker. are generally limited to a realis–irrealis distinction. irrealis.’ The marking of irrealis by mutation occurs only with the Class I verbs ‘go. the difference between a would-be hyoq-di and the causative hyo -di is minimal.’ ‘come. indeed. The regular past tense markers (-d-i. Class I verbs. and -g-ei.’ The most regular Class I verbs are ‘eat’ and ‘sleep’ – t-˙m-˙n. can generally be predicted from these four. g-˙m-da-n. ts-ip-ßn. g-˙m-du. Especially in verbs with a final uvular consonant (as in hyoq-ßn ‘I/you hide’). and irrealis. especially in Class II verbs. -g-i) may be avoided in some Class II intransitive verbs for reasons of ambiguity.’ Also. realis. -n-ei. as in bal-tsi-n ‘I descend.’ Past tense is generally missing from such verbs.4).

.................................. Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Mutating Verbs Past GO 1s 1p 2s 2p 3s 3p ....... ...... ...... However. ............. ......... d-˙g-˙i te-i-dzi COME ..... . u-g-i u-dei TAKE ........... Class I verbs do not have a Past tense.Watters: Kusunda Grammar...... ........... n-i÷-na-n d-a -an-dzi n-o -da-n o -da-n-dzi o -da-n-dei g-o -an g-o -da-n t-a -~a-dei 154 .... . COME t-ug-un t-ug-da-n n-ug-un n-ug-da-n ug-˙n ug-da-n TAKE t-ug-un t-ug-da-n n-ug-un n-ug-da-n g-ug-un g-ug-da-n Irrealis GO 1s 1p 2s 2p 3s 3p ts-a -an ts-i÷-da-n ¯-a -an COME t-o t-o n-o -an -da-n -an TAKE t-o -an t-o n-o n-o -da-n -an -da-n Only in third persons is there a difference between the verbs ‘come’ and ‘take’ In general..... ... the third person forms for ‘go’ and ‘come’ have what appears to be both past and realis forms............. ..... Realis GO 1s 1p 2s 2p 3s 3p ts-˙g-˙n ts-i-da-n n-˙g-˙n n-i-da-n d-˙g-˙n ..... ..... .....

................... ...... n-i-dak g-i -ak g-i-dak 155 ......Himalayan Linguistics: Archive No........... the third person forms for ‘go’ and ‘come’ have what appears to be both past and realis forms.... u-g-i u-dei Realis GO 1s 1p 2s 2p 3s 3p ts-˙g-˙n ts-i-da-n n-˙g-˙n n-i-da-n d-˙g-˙n ....... ... .. ............ However.... Class I verbs do not have a Past tense.. d-˙g-˙i te-i-dzi COME .. ..... 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Mutating Verbs with newer morphology Past GO 1s 1p 2s 2p 3s 3p . COME t-ug-un t-ug-da-n n-ug-un n-ug-da-n ug-˙n u-d˙i Irrealis GO 1s 1p 2s 2p 3s 3p ts-a -ak ts-i-dak ¯-a -ak COME t-ug-u t-u-da-k n-ug-u n-u-da-k g-ug-u g-u-da-k In general....... .....

’ for example regularly uses -dzi in place of -di. the plural forms are usually -dei.. and -gi. The verb ‘see. The -dzi forms in third singular ‘die’ and ‘sleep’ – o -dzi.... -ni. Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Intransitive Verbs Realis SIT 1s 1p 2s 2p 3s 3p s-aˆ -an s-aˆ -an ¯-aˆ -an ¯-aˆ -an h-aˆ -dzi h-aˆ -dei DIE t-o t-o n-o n-o -˙n -da-n -˙n -da-n SLEEP ts-ip-ßn ts-ip-da-n n-ip-ßn n-ip-da-n g-ip-ßn / (g)-ip-dzi o -˙n / o -dzi o -dei / o -˙n-dei (g)ip-da-n / (g)-ip-dei Irrealis SIT 1s 1p 2s 2p 3s 3p s-aˆ -du s-aˆ -d˙-k ¯-aˆ -du ¯-aˆ -d˙-k h-aˆ -du h-aˆ -d˙-k DIE t-o -du SLEEP ts-ip-du ts-ip-da-k n-ip-du n-ip-da-k g-ip-du g-ip-da-k t-o -d˙-k n-o -du n-o -d˙-k g-o -du .. (g)-ip-dei. Where such occur.. 156 ..Watters: Kusunda Grammar.. (g)-ip-dzi – are alternative realis forms occurring frequently with certain verbs... based on analogy with the plural forms in past paradigms – o -dei..

..’ ‘bring........... BRING ts-iw-˙n ts-i-da-n n-iw-˙n n-i-da-n g-iw-˙n g-iw-˙n PUT IN ts-ig-˙n ts-i-da-n n-ig-˙n n-i-da-n g-i-n .............. g-imi-du ...........Himalayan Linguistics: Archive No.......... . Differences between ‘fetch....... g-imi-di ....................... n-imi-du . .................... .......... ............ ......................... n-imi-di . BRING .... 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Transitive Verbs Past FORGET 1s 1p 2s 2p 3s 3p ts-˙mi-di .. ..................... . n-i-du ............ ....... 157 . FETCH ts-i-di ts-i-dei n-i-di n-i-dei g-i-di g-i-dei Realis FORGET 1s 1p 2s 2p 3s 3p ts-imi-n ts-imi-n n-imi-n n-imi-n g-imi-n g-imi-n FETCH ts-i-dßn ts-i-dan n-i-dßn n-i-dan g-i-dzi g-i-dan Irrealis FORGET 1s 1p 2s 2p 3s 3p ts-˙mi-du . g-i-du ....... FETCH ts-i-du ..... ..... ........’ and ‘put in’ are very slight... PUT IN ........ BRING ts-i-du ts-i-d˙k n-i-du n-i-d˙k g-i-du g-i-d˙k PUT IN ts-i-d-u ts-ig-˙k n-i-d-u n-ig-˙k g-i-yu ..

. singular is marked by -du (a singular–non-past portmanteau) and plural by -da.. -ni..Watters: Kusunda Grammar... In Class I irrealis... SEE/LOOK ts-aˆ -du ts-aˆ -d˙-k ¯-aˆ -du ¯-aˆ -d˙-k g-aˆ -du g-aˆ -d˙-k The -di..or t-. 158 . Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Transitive Verbs Realis EAT 1s 1p 2s 2p 3s 3p t-˙m-˙n t-˙m-da-n n-˙m-˙n n-˙m-da-n g-˙m-˙n g-˙m-da-n KNOW t-uk-an t-uk-da-n n-uk-an n-uk-da-n (uk-an-dzi) (uk-da-n-dei) SEE/LOOK (ts-aˆ -dzi) ts-aˆ -dei (¯-aˆ -dzi) ¯-aˆ -dei (g-aˆ -dzi) g-aˆ -dei Irrealis EAT 1s 1p 2s 2p 3s 3p t-˙m-du t-˙m-da-k n-˙m-du n-˙m-da-k g-˙m-du g-˙m-da-k KNOW t-uk-an-(du) t-uk-da-k n-uk-an-(du) n-uk-da-k .... second person n-. Class I verbs mark the person of the subject with a verbal prefix – first person ts. . and singular goes unmarked. d-... So too in the third person forms for ‘know.. or Ø. Realis plural in such verbs is marked by a suffix -da immediately following the root.. -gi forms are replaced in ‘see’ by -dzi forms. and third person g-.’ The -dei forms are on an analogy with past plural forms... Realis is marked by -n following number markers.. Irrealis (by a purely synchronic analysis) is marked by -k following the plural marker and unmarked following singular.

....... g-o -a-d-ßn n-o -a-d-ßn PUT 2 ts-o -a-d-ßn n-o -d˙-k g-o -du g-o -d˙-k 159 ................ .. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Transitive Verbs Realis DRINK 1s 1p 2s 2p 3s 3p ts-oˆ -˙n ts-oˆ -na-n n-oˆ -˙n n-oˆ -na-n g-oˆ -˙n g-oˆ -na-n PUT ts-o -on ts-o -da-n n-o n-o g-o g-o -on -da-n -on -da-n Irrealis DRINK 1s 1p 2s 2p 3s 3p ts-oˆ -du ts-oˆ -d˙-k n-oˆ -du n-oˆ -d˙-k g-oˆ -du g-oˆ -d˙-k PUT ts-o -du ts-o -d˙-k n-o -du PUT 2 ......... .............Himalayan Linguistics: Archive No. ..... ..... ..........

. 160 ..... Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs Realis FALL 1s 1p 2s 2p 3s 3p bem-ßn bem-da-n bem-ßn bem-da-n bem-dzi bem-dzi bem-ni-n bem-tsi-n HIDE hyoq-ßn hyoq-ts-˙n hyoq-ßn hyoq-¯a-n hyoq-dzi hyoq-dei Irrealis FALL 1s 1p 2s 2p 3s 3p bem-du bem-d˙-k bem-du bem-d˙-k bem-du bem-d˙-k HIDE hyoq-ts-u hyoq-ts˙-k hyoq-n-u hyoq-¯˙-k . Further study may reveal that one is Past and the other Realis. ....Watters: Kusunda Grammar... hyoq-ni-n hyoq-tsi-n Class II intransitive verbs have two conjugations in realis mode — one bereft of person markers (as in bem-ßn) and the other with person marking added (as in bem-tsi-n).............

..... 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs Realis DESCEND 1s 1p 2s 2p 3s 3p bal-ßn / bal-tsi-n bal-da-n bal-ßn / bal-ni-n bal-da-n bal-dzi bal-dei GET WET dzoq-ßn dzoq-da-n dzoq-ßn dzoq-na-n dzoq-dzi dzoq-dei LAUGH nix-ßn nix-ßn nix-ßn nix-ßn nix-ßn nix-ßn Irrealis DESCEND 1s 1p 2s 2p 3s 3p bal-du bal-d˙k bal-du bal-d˙k bal-du / -dzi bal-d˙k GET WET dzoq-d-u dzoq-d-˙k dzoq-n-u dzoq-n-˙k dzoq-dei .Himalayan Linguistics: Archive No. LAUGH nixe-d-u nixe-d-˙k nixe-n-u nixe-n-˙k nix-ßn nix-ßn DANCE limi-t-u / limi-tsi-du DANCE limi-t-ßn / limi-tsi-n limi-da-n limi-n-ßn limi-da-n limi-dzi 161 .......

.. ..... WEEP ............... ............................ ........................ ..................... ... ............ ............... Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs Past HANG 1s 1p 2s 2p 3s 3p ............ WAIT ....... ............ ..... ............ . .Watters: Kusunda Grammar... .... 162 .. .. .............. dz˙m-dzi dz˙m-dei Realis HANG 1s 1p 2s 2p 3s 3p dzu≥-tsi-n ?? dzu≥-ni-n ?? dzu≥-dzi ?? WEEP dz˙m-tsi-n dz˙m-tsi-da-n dz˙m-ni-n dz˙m-ni-da-n dz˙m-gi dz˙m-gi-da-n Irrealis HANG 1s 1p 2s 2p 3s 3p ?? ?? ?? ?? ?? ?? WEEP dz˙m-tsi-du ?? dz˙m-ni-du ?? dz˙m-gi-du ?? FEAR kæai-d-u kæai-d-˙k kæai-n-u kæai-n-˙k kæai-dzi kæai-dzi WAIT ¯aq-tsi-du ¯aq-tsi-d˙k ¯aq-ni-du ?? ?? ?? FEAR kæai-tsi-n kæai-tsi-da-n *kæai-ni-n (neg) kæai-na-n kæai-in-dzi ?? WAIT ¯aq-tsi-n ?? ¯aq-ni-n ?? ?? ?? FEAR ..................... ......... ............ ..

.. ... ENTER sip-d-i ?? sip-n-i ?? sip-dzi ?? 163 ..... Realis BE TIRED 1s 1p 2s 2p 3s 3p Irrealis BE TIRED 1s 1p 2s 2p 3s 3p nu≥-dzi dæu≥-du nu≥-n-u dæ~uÚ-ni-nu / dæu≥-n-u RISE nu≥-dz-u STAND dæu≥-tsi-du ENTER p˙o-dzi nu≥-dzi dæu≥-dzi sip-ßn-dzi p˙o-ni-n p˙o-tsi-n RISE nu≥ / nu≥-tsi-n nu≥-dza-n nu-n / nu≥-ni-n dæ~uÚ-n-ßn sip-ni-n STAND dæu≥-tsi-n ENTER sip-ßn / sip-tsi-n STAND .. ............ ..Himalayan Linguistics: Archive No.................. ....... ............. ......... ............. ................... ............. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs Past BE TIRED 1s 1p 2s 2p 3s 3p p˙o-d-i ?? p˙o-n-i ?? p˙o-g-i ?? RISE ...... ..........

......... Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs Past RUN 1s 1p 2s 2p 3s 3p pæurpæu-d-i pæurpæu-d-ei pæurpæu-n-i pæurpæu-n-ei pæurpæu-dzi pæurpæu-dei Realis RUN 1s 1p 2s 2p 3s 3p pæurpæu-d-ßn pæurpæu-d-˙n pæurpæu-n-ßn pæurpæu-n-˙n pæurpæu-g-˙n . Irrealis RUN 1s 1p 2s 2p 3s 3p pæurpæu-d-u pæurpæu-d-˙k pæurpæu-n-u pæurpæu-n-˙k pæurpæu-dzi pæurpæu-dei 164 .Watters: Kusunda Grammar..

. ................ ....................... ....................... ... ................... .................. BREAK k˙la-q-tsi-n ?? ?? ?? ?? ...... BE LOST miÚ-tsi-t-ßn miÚ-tsi-t-na-n ?? ?? ?? . .. ................Himalayan Linguistics: Archive No.. ..... ................... ........ MOVE . gwi-dzi gwi-dei BE LOST .. .... .. ............. .......................... 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Derived Intransitive Verbs Past GATHER 1s 1p 2s 2p 3s 3p . Irrealis GATHER 1s 1p 2s 2p 3s 3p gwi-tsi-du gwi-tsi-d˙k gwi-ni-t-nu gwi-ni-t-n˙k ?? gwi-t-d˙k ?? BE LOST miÚ-tsi-du miÚ-tsi-d˙k mi-t-ni-du mi-t-ni-d˙k miÚ-du ?? miÚ-d˙k ?? MOVE ?? ?? ?? gæwi-ni-t-d˙-k ?? ?? BREAK ?? ?? ?? ?? ?? ?? MOVE gæwi-tsi-t-ßn gæwi-tsi-t-da-n gæwi-n-ßn ?? gæwi-ni-t-na-n ?? .. Realis GATHER 1s 1p 2s 2p 3s 3p gwi-tsi-t-ßn gwi-tsi-da-n gwi-t-na-n gwi-ni-t-na-n gwi-t-ßn .... .. ...................... BREAK .... 165 ...... ...........

Watters: Kusunda Grammar; Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past MAKE 1s 1p 2s 2p 3s 3p a-d-i a-d-ei a-n-i a-n-ei ˙-g-i ˙-g-ei Realis MAKE 1s 1p 2s 2p 3s 3p a-t-ßn a-d-˙n a-n a-n-˙n ˙-g-˙n ˙-g-˙n GIVE e-d-ßn e-d-˙n e-n e-d-˙n e-g-˙n e-g-˙n Irrealis MAKE 1s 1p 2s 2p 3s 3p a-d-u a-d-˙k a-n-u a-n-˙k a-g-u a-g-˙k GIVE e-d-u e-d-˙k e-n-u e-n-˙k e-g-u e-g-˙k BEG ai-t-u ai-d-˙k ai-n-u ai-n-˙k ai-n-dzi ai-n-dzi / ai-g-˙k CUT kisa-d-u kisa-d-˙k kisa-n-u kisa-n-˙k kisa-g-u kisa-g-˙k BEG ai-t-ßn ai-t-˙n ai-n-ßn ai-n-˙n ai-dzi ai-dei kisa-g-˙n kisa-n-ßn CUT kisa-d-ßn e-g-i ai-g-i e-n-i ai-n-i GIVE e-d-i BEG ai-d-i CUT kisa-d-i kisa-d-ei kisa-n-i kisa-n-ei kisa-g-i kisa-(g)-ei


Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past OPEN FOR 1s 1p 2s 2p 3s 3p h˙ra-d-i h˙ra-d-ei h˙ra-n-i h˙ra-n-ei h˙ra-g-i ........... t˙mb˙-d-i t˙mb˙-d-ei t˙mb˙-n-i t˙mb˙-n-ei t˙mb˙-g-i t˙mb˙-(g)-ei Realis OPEN 1s 1p 2s 2p 3s 3p h˙ra-d-ßn h˙ra-d-˙n h˙ra-n-ßn h˙ra-n-˙n h˙r˙-g-˙n ........... SEND t˙mb˙-d-ßn t˙mb˙-d-˙n t˙mb˙-n-ßn t˙mb˙-n-˙n t˙mb˙-g-˙n ........... Irrealis OPEN 1s 1p 2s 2p 3s 3p h˙ra-d-u h˙ra-d-˙k h˙ra-n-u h˙ra-n-˙k h˙ra-≥ ........... SEND t˙mb˙-d-u t˙mb˙-d-˙k t˙mb˙-n-u t˙mb˙-n-˙k t˙mb˙-g-u t˙mb˙-g-˙k DRAW WATER kola-d-u kola-d-˙k kola-n-u kola-n-˙k kola-(g)-u ........... SEARCH oˆl~a-d-u oˆl~a-d-˙k oˆl~a-n-u oˆl~a-n-˙k oˆl~a-g-u oˆl~a-g-˙k kola-g-˙n ........... kola-n-ßn DRAW WATER kola-d-ßn SEARCH oˆl~a-d-ßn oˆl~a-d-˙n oˆl~a-n-ßn oˆl~a-n-˙n oˆl~a-g-˙n ........... kola-d-i ........... kola-n-i ........... kola-g-i ........... oˆl~a-d-i oˆl~a-d-ei oˆl~a-n-i oˆl~a-n-ei oˆla-(g)-~î oˆla-(g)-ei SEND DRAW WATER SEARCH


Watters: Kusunda Grammar; Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past FEED 1s 1p 2s 2p 3s 3p in-da-d-i in-da-d-ei in-da-n-i in-da-n-ei in-da-(g)-i REMEMBER imba-d-i imba-d-ei imba-n-i imba-n-ei imba-g-i imba-(g)-ei Realis FEED 1s 1p 2s 2p 3s 3p Irrealis FEED 1s 1p 2s 2p 3s 3p in-da-d-u in-da-d-˙k in-da-n-u in-da-n-˙k in-da-g-u in-da-g-˙k REMEMBER imba-d-u imba-d-˙k imba-n-u imba-n-˙k imba-g-u imba-g-˙k LOSE meÚ-d-u meÚ-d-˙k meÚ-n-u meÚ-n-˙k meÚ-g-u meÚ-g-˙k BUY dzaÚ-d-u dzaÚ-d-˙k dzaÚ-n-u dzaÚ-n-˙k dzaÚ-g-u dzaÚ-g-˙k in-da-g-˙n imba-g-˙n in-da-n-ßn imba-n-ßn in-da-d-ßn REMEMBER imba-d-ßn LOSE meÚ-d-ßn meÚ-da-n meÚ-n-ßn meÚ-na-n meÚ-g-˙n BUY dzaÚ-d-ßn dzaÚ-da-n dzaÚ-n-ßn dzaÚ-na-n dzaÚ-g-˙n dzaÚ-ga-n LOSE meÚ-d-i meÚ-d-ei meÚ-n-i meÚ-n-ei meÚ-(g)-i dzaÚ-g-i dzaÚ-n-i BUY dzaÚ-d-i


. Irrealis SHOW 1s 1p 2s 2p 3s 3p unda-d-u unda-d-˙k unda-n-u unda-n-˙k unda-g-u unda-g-˙k COOK hul˙-d-u hul˙-d-˙k hul˙-n-u hul˙-n-˙k hul˙-g-u hul˙-g-˙k KILL ora-d-u ora-d-˙k ora-n-u ora-n-˙k ora-g-u ora-g-˙k BEAT pumba-d-u pumba-d-˙k pumba-n-u pumba-n-˙k pumba-g-u pumba-g-˙k KILL ora-d-ßn ora-d-˙n ora-n-ßn ora-n-˙n ora-g-˙n ..Himalayan Linguistics: Archive No... KILL ora-d-i ora-d-ei ora-n-i ora-n-ei ora-(g)-i ora-(g)-ei BEAT pumba-d-i pumba-d-ei pumba-n-i pumba-n-ei pumba-(g)-i pumba-(g)-ei 169 ....... BEAT pumba-d-ßn pumba-d-˙n pumba-n-ßn pumba-n-˙n pumba-g-˙n ............... COOK hul˙-d-ßn hul˙-da-n hul˙-n-ßn hul˙-na-n hul˙-g-˙n . 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past SHOW 1s 1p 2s 2p 3s 3p unda-d-i unda-d-ei unda-n-i unda-n-ei unda-(g)-i unda-(g)-ei COOK hul˙-d-i hul˙-d-ei hul˙-n-i hul˙-n-ei hul˙-g-i hul˙-(g)-ei Realis SHOW 1s 1p 2s 2p 3s 3p unda-d-ßn unda-d˙-n unda-n-ßn unda-n˙-n unda-g-˙n .................

................. WASH sowa-d-ßn sowa-d˙-n sowa-n-ßn sowa-n˙-n sowa-g-˙n sowa-ei Irrealis BURN 1s 1p 2s 2p 3s 3p hab˙-d-u hab˙-d-˙k hab˙-n-u hab˙-n-˙k hab˙-g-u hab˙-g-˙k WASH sowa-d-u sowa-d-˙k sowa-n-u sowa-n-˙k sowa-Gu sowa-Gu BATHE po≈la-d-u po≈la-d-˙k po≈la-n-u po≈la-n-˙k po≈la-g-u po≈la-g-˙k tumb˙-g-u tumb˙-n-u FINISH tumb˙-d-u BATHE po≈la-d-ßn po≈la-d-˙n po≈la-n-ßn po≈la-n-˙n ?? ......Watters: Kusunda Grammar. Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past BURN 1s 1p 2s 2p 3s 3p hab˙-d-i hab˙-d-ei hab˙-n-i hab˙-n-ei hab˙-g-i hab˙-(g)-ei WASH sowa-d-i sowa-d-ei sowa-n-i sowa-n-ei sowa-d-ei sowa-d-ei Realis BURN 1s 1p 2s 2p 3s 3p hab˙-d-ßn hab˙-d-˙n hab˙-n-ßn hab˙-n-˙n hab˙-g-˙n . tumb˙-g-˙n tumb˙-n-ßn FINISH tumb˙-d-ßn BATHE po≈la-d-i po≈la-d-ei po≈la-n-i po≈la-n-ei po≈la-(g)-i po≈la-(g)-ei tumb˙-g-i tumb˙-n-i FINISH tumb˙-d-i 170 .

bæ˙ra-n-ßn bæorl˙-n-ßn BREAK II bæ˙ra-d-ßn BOIL WATER bæorl˙-d-ßn bæ˙ra-g-i bæorl˙-g-i bæ˙ra-n-i bæorl˙-n-i BREAK II bæ˙ra-d-i BOIL WATER bæorl˙-d-i 171 ...Himalayan Linguistics: Archive No............. bæ˙ra-g-˙n . 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs Past BREAK 1s 1p 2s 2p 3s 3p k˙la-d-i k˙la-d-ei k˙la-n-i k˙la-n-ei k˙la-g-i k˙la-g-ei Realis BREAK 1s 1p 2s 2p 3s 3p k˙la-d-ßn k˙la-d˙-n k˙la-n-ßn k˙la-n˙-n k˙la-g-˙n ........... Irrealis BREAK 1s 1p 2s 2p 3s 3p k˙la-d-u k˙la-d-˙k k˙la-n-u k˙la-n-˙k k˙la-g-u k˙la-g-˙k BREAK II bæ˙ra-d-u bæ˙ra-d-˙k bæ˙ra-n-u bæ˙ra-n-˙k bæ˙ra-g-u bæ˙ra-g-˙k BOIL WATER bæorl˙-d-u bæorl˙-g-˙n .......

Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Causative Verbs Past MAKE FALL 1s 1p 2s 2p 3s 3p bem-˙-d-i bem-˙-d-ei bem-˙-n-i bem-˙-n-ei bem-˙-(g)-i bem-˙-(g)-ei MAKE WET dzo -a-d-i dzo -a-d-ei HIDE hyo -a-d-i hyo -a-d-ei hyo -a-n-i hyo -a-n-ei hyo -a-(g)-i hyo -a-(g)-ei dzo -a-n-i dzo -a-n-ei dzo -a-(g)-i dzo -a-(g)-ei Realis MAKE FALL 1s 1p 2s 2p 3s 3p bem-˙-g-˙n bem-˙-n-ßn bem-˙-d-ßn MAKE WET dzo -a-d-ßn HIDE hyo -a-d-ßn hyo -a-n-ßn dzo -a-g-˙n hyo (hyo -a-g-˙n -a-ei) Irrealis MAKE FALL 1s 1p 2s 2p 3s 3p bem-˙-d-u bem-˙-d-˙k bem-˙-n-u bem-˙-n-˙k bem-˙-g-u bem-˙-g-˙k MAKE WET dzo -a-d-u HIDE hyo -a-d-u hyo -a-d-˙k hyo -a-n-u hyo -a-n-˙k hyo -a-dzi dzo -a-d-˙k dzo -a-n-u dzo -a-n-˙k dzo -a-(g)-u dzo -a-g-˙k 172 .Watters: Kusunda Grammar.

3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Causative Verbs Past DROP 1s 1p 2s 2p 3s 3p Realis DROP 1s 1p 2s 2p 3s 3p Irrealis DROP 1s 1p 2s 2p 3s 3p byaq-a-g-u ˙igi-g-u pæeb-˙-n-u byaq-a-n-u ˙igi-n-u pæeb-˙-d-u OVERTURN byaq-a-d-u SURVIVE ˙igi-d-u pæeb-˙-g-˙n byaq-a-g-˙n ˙igi-g-˙n byaq-a-n-ßn ˙igi-n-ßn pæeb-˙-d-ßn OVERTURN byaq-a-d-ßn SURVIVE ˙igi-d-ßn pæeb-˙-g-i byaq-a-g-i ˙igi / ˙igi-g-i pæeb-˙-n-i byaq-a-n-i ˙igi-n / ˙igi-n-i pæeb-˙-d-i OVERTURN byaq-a-d-i SURVIVE ˙igi / ˙igi-d-i 173 .Himalayan Linguistics: Archive No.

Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Mutating Verbs — Negative Realis / Past GO 1s 1p 2s 2p 3s 3p ts-a u ts-a u ¯-a u ¯-a u d-a i d-a i COME t-ug-a u t-ug-a u n-ug-a u n-ug-a u ug-a u ug-a u Irrealis GO 1s 1p 2s 2p 3s 3p ts-˙g-wa tsi-d˙g-wa n-˙g-wa ni-d˙g-wa d-˙g-wa d-˙g-wa COME t-ug-u t-u-dak n-ug-u n-u-dak g-ug-u g-u-dak 174 .Watters: Kusunda Grammar.

3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Intransitive Verbs — Negative Realis SLEEP 1s 1p 2s 2p 3s 3p ts-ip-da u ts-ip-da u n-ip-da u n-ip-da u g-ip-da i g-ip-da i DIE t-o -d-a u t-o -d-a u n-o n-o -n-a u -n-a u SIT s-aˆ -da u s-aˆ -da u ¯-aˆ -da u ¯-aˆ -da u h-aˆ -da i h-aˆ -da i Irrealis SLEEP 1s 1p 2s 2p 3s 3p ts-ip-wa ts-ip-d-wa n-ip-wa n-ip-d-wa g-ip-wa g-ip-d-wa DIE t-o -d-wa t-o -d-˙o n-o -n-wa SIT s-aˆ -n-wa s-aˆ -n-wa ¯-aˆ -n-wa ¯-aˆ -n-wa h-aˆ -n-wa h-aˆ -n-wa o -da i o -da i n-o -n-˙o o -d-wa o -d-˙o 175 .Himalayan Linguistics: Archive No.

Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class I Transitive Verbs — Negative Realis EAT 1s 1p 2s 2p 3s 3p t-˙m-da u t-˙m-da u n-˙m-da u n-˙m-da u g-˙m-da u g-˙m-da u KNOW t-uk-da i t-uk-da i n-uk-da i n-uk-da i ok-da i ok-da i Irrealis EAT 1s 1p 2s 2p 3s 3p t-˙m-wa t-˙m-wa n-˙m-wa n-˙m-wa g-˙m-wa g-˙m-wa KNOW t-ug-wa t-ug-wa n-ug-wa n-ug-wa ug-wa ug-wa SEE/LOOK ts-~aÚ-d-wa ts-~aÚ-d-˙o ¯-~aÚ-n-wa ¯~-aÚ-n-˙o g-~aÚ-n-wa g-~aÚ-d-˙o SEE/LOOK ts-~aÚ-da i ts-~aÚ-da i ¯-~aÚ-da i ¯-~aÚ-da i g-~aÚ-da i g-~aÚ-da i 176 .Watters: Kusunda Grammar.

Himalayan Linguistics: Archive No. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Intransitive Verbs — Negative Realis FALL 1s 1p 2s 2p 3s 3p bem-d-a u bem-d-a u bem-n-a u bem-n-a u bem-d-a i bem-d-a i Irrealis FALL 1s 1p 2s 2p 3s 3p bem-d-wa bem-d-˙o bem-n-wa bem-n-˙o bem-d-wa bem-d-wa HIDE HIDE 177 .

Watters: Kusunda Grammar. Appendix B – Verb paradigms ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs — Negative Realis MAKE 1s 1p 2s 2p 3s 3p a-d-a u a-d-a u a-n-a u a-n-a u a-g-a u a-g-a u GIVE e-d-a u e-d-a u e-n-a u e-n-a u e-g-a u e-g-a u Irrealis MAKE 1s 1p 2s 2p 3s 3p a-du-wa a-du-wa a-nu-wa a-nu-wa a-g-wa a-g-wa GIVE e-du-wa e-du-wa e-nu-wa e-nu-wa e-g-wa e-g-wa 178 .

u unda.i pumba.i KILL ora-d-a i ora-d-a i ora-n-a i ora-n-a i ora.i ora.i Irrealis BEAT 1s 1p 2s 2p 3s 3p pumba-d-wa pumba-d-˙o pumba-n-wa pumba-n-˙o pumba-g-wa pumba-g-˙o KILL ora-d-wa ora-d-˙o ora-n-wa ora-n-˙o ora-g-wa ora-g-˙o SHOW unda-d-wa unda-d-˙o unda-n-wa unda-n-˙o unda-g-wa unda-g-˙o SHOW unda-d-a u unda-d-a u unda-n-a u unda-n-a u unda. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Class II Transitive Verbs — Negative Realis / Past BEAT 1s 1p 2s 2p 3s 3p pumba-d-a i pumba-d-a i pumba-n-a i pumba-n-a i pumba.u 179 .Himalayan Linguistics: Archive No.

180 . 1967. Tibetan dialects. Ross C. Martin. Givón. Grierson. 2002. Journal of the Asiatic Society of Bengal XVII (2): 650–658. Vol. Haspelmath. Fleming. and discourse patterns in selected languages of Nepal. J. Forbes. Canberra: The Australia National University [Pacific Linguistics. F. Elliott. Bergsland. Population census 2001. Martin and Ekkehard König (eds. 1986. Dor Bahadur. Matthew. Harold C. the Himalayan dialects and the North Assam groups. Knut.References Aikhenvald. Fairbanks: University of Alaska [Alaska Native Language Center Research Papers 10. And other essays. “Etymological notes on Tibeto-Burman case particles”. C. Vol. Berkeley: University of California Press. F. Brian H. 1848. Caughley. Linguistics of the Tibeto-Burman Area 8.4:808–845. Language 62. Hale.] Berlin. 1996. Hodgson. and antidative”. Hemanta Raj. T. 1984. Kathmandu: Sanivasiya Parisißstank. Part I]. Central Bureau of Statistics. “On the Chépáng and Kúsúnda tribes of Nepál”. T. 1877. 1995. Forbes. Alexandra. Phirante jivan choßddai aeka Kusußnßda jati: lop hune sthitima (The Kusunda people giving up their nomadic lifestyle: in an endangered state). ß Bista. Austin and David Watters (eds). Forthcoming. Converbs in cross-linguistic perspective 1-55. Amsterdam and Philadelphia: John Benjamins. Norman. Syntax: A functional–typological introduction. 1973. In Haspelmath. National Report. Dryer. His Majesty’s Government of Nepal. Delhi: Motilal Banasidass [Linguistic survey of India. George A. I. “Affinities of the dialects of the Chepang and Kusundah tribes of Nipál with those of the Hill Tribes of Arracan”. Vol III. DeLancey.1:59-77. Basic color terms: Their universality and evolution. “Primary objects. Givón. “Realis and irrealis: forms and concepts of the grammaticalisation of reality”. 1969. OK: Summer Institute of Linguistics and University of Oklahoma. II. Kathmandu: Ratna Pustak Bhandar. Reprinted 1967. Gorkhapatra (daily newspaper). 1990. 1987. C. Series B. Aleut grammar. Bhandari. Journal of the Royal Asiatic Society IX: 421–424. “Origins of imperatives”. Comparative grammar of the languages of further India: A fragment. “The converb as a cross-linguistically valid category”. Clause. Jennifer R. London: Allen and Co. Tibeto-Burman family. 84]. The syntax and morphology of the verb in Chepang.). Amsterdam and Philadelphia: John Benjamins. “Looking to the West and North: Nihali and Kusunda find links.” Mother Tongue 2:67–74. 1997. 1881. People of Nepal. Vol. 2000. Syntax: A functional–typological introduction. Berlin and New York: Mouton de Gruyter. sentence. Linguistic Typology 4: 55-90. Capt. Brent and Paul Kay. II. Scott. 1909. secondary objects. 1984. J. National Planning Commission Secretariat. 1982. Capt.

251-299. Journal of the Institute of Nepal and Asian Studies. 1857.. 1931. I–III. “East Himalayish”. Veneeta Z. 1973. Fairbanks 111-133. Watters. 2005. Starosta. E. Chicago and London: University of Chicago Press.). Ralph. Shopen. “Comparative vocabulary of the languages of the broken tribes of Nepal”. Need. A comparative and etymological dictionary of the Nepali language. Balaram Prasain. Chicago: Chicago Linguistic Society. In A. Clause. Yadava. with the participation of Yogendra P. “Transitivity in grammar and discourse”. Turner. 2002. London: Asher and Co. Madhav P. 2001. Brian H. II:81-174. Paper presented at the Fourth Harvard Roundtable on the ethnogenesis of South and Central Asia. literature. “Clause patterns in Tamang”. (ed. Language 56:2. “The Bana Rajas – A vanishing Himalayan tribe”. For Gordon H. and religion of Nepál and Tibet. Taylor. Watters. Vols. and discourse patterns in selected languages of Nepal. Bulletin of the School of Oriental and African Studies XV (2): 356-374. David E. Roediger. (eds. Paul and Sandra Thompson.1-21]. Lincom Europa [Languages of the World/Materials 162]. A grammar of Kalaallisut (West Greenlandic Inuttut). Kathmandu: National Foundation for the Development of Indigenous Nationalities. E. USA. Doreen. 1976. Timothy. Masica.). 1985. Language typology and syntactic description Vol. 1987. George. Pokharel. Language typology and syntactic description. “New materials on the Kusunda language”. II. Sadock. Essays on the languages. 1970. 181 . David E. Cambridge MA. A preliminary linguistic analysis and vocabulary of the Kusunda language.Himalayan Linguistics: Archive No. 235-286. 1985. Robert E. Brian H. part 2: Parassession on autosegmental and metrical phonology. 1874. 1976. Bosch. Schiller (eds. 1874.). James K. 2002. (ed). Reinhard. Van Driem. Longacre. Journal of the Asiatic Society of Bengal XXVI: 317–371. Harvard University. sentence. Kathmandu: SIL and Tribhuvan University. Rana. Kathmandu [Contribution to Nepalese Studies. A grammar of Kham. 2002. 1985.. Papers from the 23rd annual regional meeting of the Chicago Linguistic Society. Johan and Sueyoshi Toba. Shafer. 1953. B. Leed. In Hale and Watters. multiple tiers. London: Trübner and Company. B. 388-402. “Sentences as combinations of clauses”. Johan. and Richard L. “Relator nouns as a source of case inflection”. Cambridge: Cambridge University Press [Cambridge Grammatical Descriptions]. Tribhuvan University. “Underspecification. Robert. Hopper. Stanley. Reinhard. Notes on Kusunda grammar. 2003. Honolulu: University of Hawaii [Oceanic Linguistics Special Publication 20]. Gesenius’s Hebrew grammar. Defining a linguistic area: South Asia. Languages of the Himalayas: an ethnolinguistic handbook of the Greater Himalayan region. Hodgson. In Timothy Shopen. 1980. May 11–13. Leiden: Brill. Watters. Cambridge: Cambridge University Press. London: Trübner and Co. Vol. In Acson. K. Colin P. eds. Jerrold M. 3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Hodgson. and Tepehua phonology”.

References ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Whitehouse. Proceedings of the National Academy of Sciences 101. “The external relationships of the Nihali and Kusunda languages. Paul and Timothy Usher. 182 . Whitehouse. Merritt Ruhlen. William S.15: 5692-5695.” Mother Tongue 3:4-49. 1997. “Kusunda: An Indo-Pacific language in Nepal”.Watters: Kusunda Grammar.-Y. 2003. Paul. Wang.

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->