DESCRIPTIVE AND TYPOLOGICAL STUDY OF

JARAWA

Thesis submitted to the Jawaharlal Nehru University in partial

fulfillment of the requirement for the
award of the degree of

DOCTOR OF PHILOSOPHY

Pramod Kumar

under the supervision of

Prof. Anvita Abbi

&
Prof. Bernard Comrie

CENTRE FOR LINGUISTICS

SCHOOL OF LANGUAGE, LITERATURE, AND
CULTURE STUDIES
JAWAHARLAL NEHRU UNIVERSITY
NEW DELHI – 110067
2012.
 Centre for Linguistics
School of Language, Literature & Culture Studies
lawaharlal Nehru University
New Delhi-ll0067, India

Dated: 2-0 . 0\ · '2.-011-

CERTIFICATE

This thesis titled "Descriptive & Typological Study of Jarawa" submitted by

Mr. Pramod Kumar, Centre for Linguistics, School of Language, Literature,
and Culture Studies, Jawaharlal Nehru University, New Delhi, for the award of
the degree of Doctor of Philosophy, is an original work and has not been
submitted so far in part or in full, for any other degree or diploma of any
University or Institution.

ntis may be placed before the examiners for evaluation for the award of the
degree of Doctor of Philosophy . .

~~
(PROF. ANVIT A 1\B]31) (PROF. BERNARD COMRIE)
SUPERVISOR JOINT-SUPERVISOR
.! ew I, ~

, • >

. I .. ' '" !_¢."7

PROF. FRANSON D. MANJALI
CHAIRPERSON

Tel : +91-11-26704199 Fax: +91- I 1-26741586

 DECLARATION BY THE CANDIDATE

This thesis titled "Descriptive & Typological Study of Jarawa" submitted by me

for the award of the degree of Doctor of Philosophy, is an original work and has
not been submitted so far in part or in full, for any other degree or diploma of
any University or Institute.

PRAMOD KUMAR
Ph. D. student
Centre for Linguistics
SLL&CS/JNU
i

Acknowledgements

In the completion of this thesis, I am indebted to a lot of people whom, I must

acknowledge. First of all, I would like to express my deepest gratitude to my

supervisors Professors Anvita Abbi and Bernard Comrie. Without their

support and encouragement this kind of work would not have been possible.

They stood by me, offering me support and guidance at every step. From both

my supervisors I learnt a lot about academics. From Professor Abbi in

particular, I learnt a lot about every walk of life. More than a supervisor to

me, she has been a mentor, and a constant source of inspiration and gentle

encouragement. Thanks, Ma’am, for always being there for me.

Second, I would like to express my gratitude to the Jarawa people for not only

trusting me in all possible ways, but also teaching me their language and

about themselves. They always supplied me new words for new objects, and

corrected me whenever I mispronounced something. I really miss the people I

met in the Andaman Islands. I am very sorry that I have not yet been able to

fulfil their wish that I return to Andaman to meet them again.

I am grateful to the Department of Linguistics, Max Planck Institute for

Evolutionary Anthropology (MPI-EVA), Leipzig, Germany, for a doctoral

fellowship and financial support for my Ph.D. I sincerely thank Claudia

Büchel, Julia Cissewski, Brigett Packendoff, Patty Epps, Don Stilo, Orin

Gensler, Juliette Blevins, Martin Haspelmath, Susane Michaelis, Eva Schultze-

Berndt, Tom Güldemann, Christfried Naumann, Jan Wohlgemuth, Sven

Grawunder, and Rocco Buchholz for their endless help and support during

my stay at Leipzig. I am also thankful to the visiting scientists at MPI-EVA for

the information I gathered from them, enriching my own academic calibre.

i
ii

I express my thanks to Professor G. K. Chaddha, the then Vice-Chancellor of

Jawaharlal Nehru University (JNU), for his support. I wish to thank all my

teachers at the Centre for Linguistics, Professors P. K. S. Pandey, Vaishna

Narang, Franson D. Manjali, and Ayesha Kidwai, for introducing me

linguistics.

I sincerely thank the Anthropological Survey of India staff stationed at Port

Blair, Mr. B. K. Ghosh (the then Head of Office), Mr. Samir Biswas, Mr. Bidyut

Kumar Das, Mr. Umesh Kumar, and Mr. S. A. Bariq. They provided much

needed support, both practical and moral, during my research in Andaman.

I acknowledge the help and support of the library staff at JNU, New Delhi,

and at MPI-EVA, Leipzig. My special thanks go to the library staff of MPI-

EVA who helped me procure all the articles and books I ever needed for my

research.

I also take this opportunity to thank the staff at the Centre for Linguistics,

JNU, Mr. S. S. Rawat, Naveen, and Bhagawati Madam for their support and

encouragement.

I express my sincere gratitude to Jyoti Iyer, a student of the Centre for

Linguistics, JNU, who copy-edited the thesis and gave it its present form.

I also thank Kamal Choudhary, and K. Krishnamurthy and his wife S. Devi,

for the strength and encouragement they managed to give me even during the

short span of our meeting in Leipzig. I express my gratitude to my friends

Dinesh, Pawan, Subodh, Shashi Bage, and Sulo Da from my hometown, and

Alok, Vinay, Sanjayji, Keshari, Bibek, Ramkrishna, Arwind and Debu from

Delhi, who helped me in various ways.

ii
iii

This kind of work could not have been done without the support and

encouragement of my parents and family. I am very grateful to them. My

special thanks go to my wife Pratibha Rani, and my son Pratyush Pranjal for

their love, understanding and untiring faith in me. They not only spared me

for the completion of my Ph.D., but were also deprived of time they needed

with me.

I sincerely acknowledge the help provided by Mr. Anoop Kumar Mondal

(Tribal Welfare Officer, Kadamtalla) and his wife Ms Anjana Mallik during

my stay in Kadamtalla. I also express my gratitude to other Aadi Andaman

Janjati Vikas Samiti (AAJVS) Field Staff, Debrat Kumar Mandal alias Dippu

(and his family members), Bibhuti Biswas, Bhajan Biswas alias Uncle, Swapan

Hawldar, Late Vinay Kumar Tete, Lokanath, Ramesh Madhu alias Rana,

Prashant Saha, and Balram Yadav. I thank them for never letting me feel that I

was away from home. I wish to thank the local people of Kadamtalla for their

support during my stay, especially Mr. Panchanand Prasad, Mr. Om Prakash

Prasad, Mrs. Nirmala Devi, and Ms. Jyoti Kumari.

The greatest and foremost problem in even beginning of my research was to

get permission to visit the Jarawa Reserve for the collection of linguistic data.

I am indebted to the Government of India, in particular the Ministry of Home

Affairs and Tribal Affairs, and the Andaman administration, for their constant

defence of my research permit. They deserve a special mention. I thank Mr. N.

Gopalaswamy (then Principal Secretary, Ministry of Home Affairs), Mr. Jalali

(then Joint Secretary, Ministry of Home Affairs), Mrs. Aindri Anurag, (then

Director, Andaman and Nicobar Islands, Ministry of Home Affairs), Mr. Rajiv

Kumar (then Joint Secretary, Ministry of Tribal Affairs), Mr. B. S.

Karmawphlang, (then Director, Andaman & Nicobar, Ministry of Tribal

Affairs), Mr. Gyanesh Bharati (then Deputy Commissioner, Andaman Islands,

iii
iv

at Port Blair), Mr. S. A. Awaradhi (then Director, Directorate of Tribal

Welfare, Andaman and Nicobar Islands, at Port Blair), and Mr. Acchelal Singh

(then Research Officer, Directorate of Tribal Welfare, Andaman and Nicobar

Islands, Port Blair).

It is not only in the reckoning of names, but in the success of my work that I

express my utmost thanks to all those I have acknowledged here.

Pramod Kumar

PLACE: New Delhi

DATE: 20th January, 2012

iv
v

Table of Contents

Acknowledgements i-iv

Abbreviations xiii-xiv

Tables and Figures xv

Chapter 1 Introduction
1.0 Introduction 1

1.1 Landscape 3

1.2 Demography 5

1.3 History of Islands 7

1.3.1 Pre-colonial Period (till 1775) 8

1.3.2 Colonial Period (1776 – 1947) 9

1.3.3 Independence and After (1947 – 1997) 13

1.3.4 Post-1997 Period (1998 – till date) 14

1.4 Inhabitants of the islands 15

1.4.1 Great Andamanese 16

1.4.2 Onge 18

1.4.3 Jarawa 18

1.4.4 Sentinalese 20

1.4.5 Jangil 21

1.5 The Jarawa people 21

1.5.1 Social structure and customs 21

1.5.2 Tools and Artefacts 24

1.6 Previous studies 26

1.6.1 Nair (1974) 26

1.6.2 Rajasingh & Ranganathan (2000) 27

1.6.3 Senkuttuvam (2000) 27

v
vi

1.6.4 Sreenathan (2001) 28

1.6.5 Rajasingh (2002) 28

1.6.6 Abbi (2006) 29

1.6.7 Blevins (2007) 29

1.7 Present Study and its Methodology 30

1.7.1 Present Study 30

1.7.2 Methodology 30

[Link] Selection of consultants 31

[Link] The selection of the area 31

[Link] Tools for the study 32

[Link].1 The questionnaire 32

[Link].2 Systematic observation 33

[Link] Field Procedure 33

1.7.3 Limitations of the study 34

1.8 Organization of the present work 35

Chapter 2 Phonetics & Phonology

2.0 Introduction 36

2.1 Sounds and Phonemes 37

2.2 Phonemic Contrasts 39

2.2.1 Vowels 40

[Link] Qualitative Contrasts 41

[Link] Quantitative Contrasts 44

2.2.2 Consonants 45

[Link] Plosives and Retroflex 46

[Link] Nasals 54

[Link] Fricatives 55

[Link] Trill, Flap and Laterals 58

[Link] Approximants 58

vi
vii

2.3 Phonemic distribution 59

2.3.1 At syllable level 59

[Link] Consonants 60

2.3.2 At word level 62

[Link] Vowels 62

[Link] Consonants 63

2.4 Syllabic Structure 65

2.5 Syllabification 66

2.6 Word Structure 66

2.7 Morphophonemic 69

2.7.1 mi vs ma 69

2.7.2 Reduced Forms 71

2.8 Conclusion 72

Chapter 3 Morphology

3.0 Introduction 73

3.1 Morpheme/ word 73

3.2 Word class 75

3.3 Noun Morphology: Nouns and Noun Phrases 76

3.3.1 Pronouns 76

[Link] Personal pronouns 76

[Link].1 Morphological nature of personal pronouns 80

[Link] Demonstrative pronoun 85

[Link] Interrogative pronoun 86

3.3.2 Nouns and noun classification 86

[Link] Gender 94

[Link] Number 95

[Link] Case marking 96

3.3.3 Possession 96

vii
viii

[Link] Inalienable possession 97

[Link] Alienable 99

3.3.4 Noun Phrase 100

[Link] Basic Noun Phrase 100

[Link] Demonstratives 101

[Link] Numerals and number marking 102

[Link] Adjectives and Modifiers 103

3.4 Verbal morphology 104

3.4.1 Verbs and verb roots 104

3.4.2 Verbal affixes 104

3.4.3 Agreement 106

3.4.4 TAM 107

3.4.5 Function of suffixes: evidentiality & modality 108

[Link] Evidential Marker 108

[Link] Non-Evidential Marker 113

[Link] Modality Marker 114

3.4.6 Function of prefixes 116

[Link] Pronominals 117

[Link] Referential/definiteness marker {ɖi} 117

[Link] Co-referencing element {hi-} 118

[Link] Conflict between {ɖi} and {hi-} 119

3.4.7 Causatives 120

[Link] Morphological causatives 120

[Link] Syntactic causatives 122

3.5 Adjectives and Adjective Phrases 124

3.5.1 Descriptive adjectives 125

3.5.2 Demonstrative adjectives 125

3.5.3 Degree of adjectives 126

viii
ix

3.5.4 Adjective Phrases 128

3.6 Adverbs 128

3.6.1 Intensifier 128

3.6.2 Temporal adverbs 129

3.6.3 Spatial adverbs 129

3.7 Postposition 129

3.7.1 Use of {ɖə} 130

[Link] Place of location 130

[Link] Source of location 131

[Link] Goal of location 131

[Link] Temporal adverbs 131

[Link] Spatial adverbs 132

3.7.2 Use of {ɲacʰe} 133

3.8 Copula {ʈʰi} 134

3.9 Word formation 138

3.9.1 Compounding 138

3.9.2 Denominalization 139

3.9.3 Reduplication 140

3.9.4 Empty morpheme 141

3.10 Conclusion 142

Chapter 4 Syntax

4.0 Introduction 143

4.1 Clause Structure 143

4.2 Verbless clauses 145

4.2.1 Noun as head of verbless clause 145

[Link] Pronominal as S argument of nominal predicate 145

[Link] Noun as S argument of nominal predicate 146

ix
x

4.2.2 Adjective as head of verbless clause 147

[Link] Pronominal as S argument of adjectival predicate 148

[Link] Noun as S argument of adjectival predicate 148

[Link] Adjectival predicates with suffix {-g} 149

4.3 Verbal Clauses I: Core arguments 152

4.3.1 Intransitive clauses 153

4.3.2 Denominalized verbs 156

4.3.3 Transitive clauses 158

4.3.4 Ditransitive clauses 160

4.4 Verbal Clauses II: Peripheral arguments 162

4.5 Negative clauses 163

4.5.1 Negation with ‘naɖem‘ 164

4.5.2 Negation with suffix {-ma} 166

4.5.3 Use of ‘ɲag ‘ 167

4.6 Imperative clauses 168

4.6.1 Simple Imperative clauses 168

4.6.2 Prohibitive clauses 169

4.7 Interrogative clauses 170

4.7.1 Yes/No question 170

4.7.2 Wh-question 172

4.8 Complex sentences 176

4.8.1 Co-ordination 177

4.8.2 Complementation 177

[Link] Complement internal clauses 177

[Link] Complement external clauses 181

4.9 Function of {ɖi} 184

4.9.1 Grammatical relations 188

4.9.2 Definiteness 191

x
xi

4.10 Cleft sentences / relativization 192

4.11 Conclusion 196

Chapter 5 Jarawa and Typology

5.0 Introduction 197

5.1 Language typology 197

5.2 Phonological typology of Jarawa 199

5.3 Morphological typology of Jarawa 200

5.3.1 Root and Affixes 200

5.3.2 Noun morphology 201

5.3.3 Pronominal system 202

5.3.4 Adjective class 202

5.3.5 Verbal morphology 203

5.4 Word order typology of Jarawa 203

5.4.1 Word order in clauses 204

5.4.2 Peripheral arguments 205

5.4.3 Complement clauses 206

5.4.4 Adpositions 207

5.4.5 Comparative adjective 208

5.4.6 Noun-adjective order 208

5.4.7 Demonstrative-numeral-noun 209

5.4.8 Genitive-noun (Possessive) 209

5.4.9 Adjective-Adverb (Intensifier) 210

5.4.10 Verb-adverb (manner of adverbs) 211

5.4.11 Copula construction 211

5.4.12 Causative construction 213

5.4.13 Negation 213

5.4.14 Question particle 215

5.4.15 Cleft sentences/relativization 218

xi
xii

5.5 Conclusion 218

Chapter 6 Conclusion 220-231

Bibliography 232-260

Appendix IA: Word List-Noun 261-266

Appendix IB: Word List-Verb 267-274

Appendix IC: Word List-Adjective and Adverb 275-276

Appendix II: Sentences 277-332

Appendix III: Distribution of verbal suffixes on verbs 333-340

Appendix IV: Map I: Andaman & Nicobar Islands 341

Appendix IV: Map II: Jarawa Reserve Area 342

xii
xiii

Abbreviations

PN - Personal Name

REF - Referentiality/Definiteness

POSS - Possessive

LOC - Locative

COP - Copula

DEM - Demonstrative

DEM-A - Demonstrative Adjective

PDEM - Proximate Demonstrative

DDEM - Distant Demonstrative

PM - Place Marker

VS - Denominalization / Verbalizer Suffix

INTM - Interrogative Marker

BENF - Benefactor

INTSF - Intensifier

VEVD - Verifiable Evidential Marker

NVEVD - Non-Verifiable Evidential Marker

EVD - Evidential Marker

NEVD - Non-Evidential Marker

ASS - Assertive Marker

HYP - Hypothetical Marker

NEG - Negative Marker

EM - Empty Morpheme

PL - Plural

COMPM - Comparative Marker

CAUS - Causative

xiii
xiv

1 - First Person

2 - Second Person

3 - Third Person

ACC - Accusative

xiv
xv

Map & Tables

Map 1 Location of Andaman and Nicobar Islands, Republic of India 4

Table 1.1 Population of non-autochthons in Andaman and Nicobar Islands 6

Table 1.2 Religion-wise population of non-autochthons in Andaman and Nicobar Islands 7

Map 2 Comparative picture of indigenous population in Andaman in 1800 and 2000 16

Map 3 Distribution of the communities of Andaman prior to 1860 17

Table 2.1 Jarawa vowel phonemes 40

Table 2.2 Jarawa consonant phonemes 46

Table 2.3 Distribution of Jarawa consonants in syllable 60-61

Table 2.4 Distribution of Jarawa vowels at word level 62-63

Table 2.5 Distribution of Jarawa consonants at word level 63-65

Table 2.6 Usage of mi form and ma form 70-71

Table 3.1 Jarawa personal pronouns 77

Table 3.2 ma and mi pronominal forms 78-79

Table 3.3 Independent pronominals and their prefixed forms 81

Table 3.4 Names of body parts in Jarawa 87-90

Table 3.5 Jarawa nouns used for kinship terms 90

Table 3.6 Names of fauna 91-92

Table 3.7 Names of flora 92

Table 3.8 Jarawa words related to nature and the cosmos 93

Table 3.9 Names of cultural artefacts 94

Table 3.10 Possessive forms 97

Table 3.11 Numerals 102

Table 3.12 Functions of {ɖə} as postposition with reference to semantic roles 132-133

Table 3.13 Examples of Jarawa compound nouns 139

Table 3.14 Examples of Jarawa denominalised verbs 140

xv
 1

Chapter 1: Introduction

1.0 Introduction

This dissertation aims to present a descriptive grammar of the Jarawa language.

Sound system, morphology and syntax of the language have been examined and

analyzed to arrive at conclusion. This study also makes an attempt to classify the

language typologically. According to Radcliff-Brown (1948), the word ‘Jarawa’

has been derived from the Aka-Bea language of the Great Andamanese group,

which meant ‘stranger’ ‘or someone who should be afraid of’. The people who

speak Jarawa belong to a hunter-gatherer community living along the western

coast of the Andaman & Nicobar Islands, a part of Union Territory of India. The

term ‘Jarawa’ is used for both the language and the people/community, although

Jarawas call themselves əŋ1.

The Andaman Islands have been the home for indigenous communities

comprising of Great Andamanese, Onge, Jarawa and Sentinelese (see §1.4 for

detailed discussion). All these communities represent ‘the first Palaeolithic

colonizers of South-East Asia’ (Thangaraj et al, 2003) and ‘remained in isolation

for a longer period than any known population of the world’ (Kashyap et al,

2004). The archaeological evidence for an early occupation of Andamans is scant

because of the nomadic lifestyle of the inhabitants and the limited number of

explorations that have taken place. At present, the oldest confirmed radio-carbon

1 Instead of using the term əŋ for this community, I will make use of the term ‘Jarawa’ throughout
the dissertation, because this community and their language are known worldwide by this term
for centuries. Changing the term may create confusion among readers.

1
2

date is just >2000 years old and there are no artefacts to suggest contact or trade

with the world outside the archipelago (Cooper 1993). Native islanders’

relationship with Palaeolithic colonizers and their isolation from outside world

for longer times have been the subject of interest and inquisition of scientists,

anthropologists, archaeologists, human evolutionists and linguists for centuries.

Though, the focus of the present study is the Jarawa language and people using

this language, it will be difficult to understand this community in isolation; the

reason being that the different communities of the islands have witnessed

conflict and contact with the outside world at different stages of history which

had affected all of them individually and collectively in varying ways. Thus, in

order to understand Jarawa, their habitat, and their plight in the course of

history, it is necessary to explore the geography of the islands and examine the

historical stages of the region with special reference to all the indigenous

communities of Andaman and Nicobar Islands.

The present chapter concentrates on a brief discussion of the islands, its history,

and native population. Section 1.1 presents a geographical description of the

islands. Section 1.2 discusses the demography of the Islands with reference to

Non-Autochthons (non-indigenous) population of the islands. The history of the

islands is discussed in Section 1.3. Section 1.4 presents a brief introduction of the

natives of the islands. Section 1.5 focuses on Jarawa, their physical characteristics,

social structure and customs, as well as tools and artefacts used by Jarawas.

Review of literature is provided in Section 1.6. Section 1.7 depicts the

methodology adopted for fieldwork in the Andaman Islands. The organization of

this dissertation is discussed in Section 1.8.

2
 3

1.1 Landscape

An archipelago of 572 islands, rocks and rocky outcrop covering an area of 8249

sq km in the Bay of Bengal constitutes Andaman and Nicobar Islands which is a

part of the Union Territory of India. It has two sets of islands, namely Andaman,

and Nicobar. The two sets of islands are separated by a 150 km wide channel

called the Ten Degree Channel. Andaman is spread over 6408 sq km, and is

made up of about 550 islands, rocks and rocky outcrop out of which only 26 are

permanently inhabited. Nicobar covers an area of 1841 sq km stretching over 24

islands out of which 11 are permanently inhabited.

The islands of the archipelago lie in a crescent that stretches from Cape Negrais

of Myanmar to the Banda Arc of Sumatra of Indonesia. The Andamans are

considered to be the extensions of the submerged Arakan Yoma mountain range

of Myanmar, while the Mentawei Islands to the south and southwest of Sumatra

are presumed to be a southern continuation of the Nicobars.

The Andaman Islands lie between latitudes 10°13ʹ and 10°30ʹN and longitudes

90°15ʹE and 93°10ʹE. The entire set of Andaman Islands is divided into two: Great

Andaman and Little Andaman. Little Andaman is 120 km south from the town of

Port Blair. Great Andaman has three main islands, namely North, Middle and

South Andaman, separated by narrow passages. The nearest landmass in the

north is Myanmar, roughly 280 km from Landfall Island, the northern most

islands in the group. The closest landmass to Great Nicobar in the south is

Sumatra, which is located at a distance of 145 km. The capital of Andaman and

Nicobar Islands, Port Blair, is a small town in South Andaman, at a distance of

1,225 km from Kolkata and 1,190 km from Chennai.

3
4

Map 1: Location of Andaman and Nicobar Islands, Republic of India

(Source: Thangaraj K, Singh L, Reddy AG, Rao VR, Sehgal SC, Underhill, PA,

Pierson M, Frame IG and E Hagelberg (2003) Genetic affinities of the Andaman

Islanders, a vanishing human population. Current Biology 13:86–93.)

To the east of Middle Andaman is a group of islands known as Ritchie’s

Archipelago, while the Labyrinth group of islands is situated southwest of South

Andaman. Ninety per cent of the total area of the Andaman group comprises

reserve forests and protected areas. Thirty six per cent of this land is demarcated

4
 5

as tribal reserves. Narcondam and Barren are two islands of volcanic origin. The

former is supposed to be extinct, while the latter is still active.

The Nicobar group is spread over an area of 1841 sq km of which 1542 sq km

covered by forests. This group consists of 24 islands in three distinct clusters. The

northern group consists of Car Nicobar and Batti Malv; the central or the

Nancowry group includes Tillangchong, Chowra, Teressa, Bompoka, Trinkat,

Kamorta, Katchal, and Nancowry. The southern group consists of the two large

islands of Little and Great Nicobar together with Pigeon, Megapode, Kondul,

Pilo Milo, Menchal, Teris, Trak and Meroe Islands. The entire Nicobars have

been declared a tribal reserve.

This group of islands is situated in the equatorial belt and has a tropical climate

which is warm, moist, and equitable. The temperature here ranges from 18°C to

35°C. They receive rains from both the North East and the South West monsoons,

with average annual rainfall in the range of 3000–3500 mm.

1.2 Demography

The present day Andaman and Nicobar Islands is home to people belonging to

different sects and religions. The arrival of non-autochthons (non-indigeneous)

started with the establishment of the British colony when the colonizers brought

in with them labourers, convicts and others. Many of them settled in the Islands.

Government policies in the Post-Independence era were also marked by

inhabiting the Islands with non-autochthons to colonise the Islands. In this era

the first settlers on the islands were groups mostly consisting of refugees from

Bangladesh (then East Pakistan) and Sri Lanka. The inflow of non-autochthons

continued and islands were populated by different types of communities. The

5
6

group of non-autochthons included people who came to the Islands to seek jobs

(some as contract labourers) and for trades. This resulted in a varied population

which settled in various pockets throughout the Islands. Some of these pockets

are located on the fringes of Jarawa Reserve Area. The population of Andaman

and Nicobar Islands is presented in Table 1.1, with a religion-wise break-up in

Table 1.2 based on the Census Report of India (1971-2001).

Table 1.1: Population of non-autochtons in Andaman and Nicobar Islands

1971 1981 1991 2001

TOTAL POPULATION 115133 188741 280661 356152

Male 70027 107261 154369 192972

Female 45106 81480 126292 163180

RURAL POPULATION 88915 139107 205706 239954

Male 53195 78401 111986 128961

Female 35720 60706 93720 110993

URBAN POPULATION 26218 49634 74955 116198

Male 16832 28860 42383 64011

Female 9386 20774 32572 52187

6
 7

Table 1.2: Religion-wise population of non-autochtons in Andaman and Nicobar

Islands

RELIGION 1971 1981 1991 2001

Hindu 70134 121793 189521 246589

Christian 30342 48274 67211 77178

Muslim 11655 16188 21354 29265

Sikh 865 991 1350 1587

Buddhist 103 127 322 421

Jain 14 11 17 23

Others 2020 1357 886 1089

1.3 History

Little is known about the history of Andaman and Nicobar Islands before the

establishment of the British colony, because of which it is very difficult to project

a complete and precise account of these islands. What we do know is either in

the form of scattered Government reports after the settlement of the British

colony, or a few books and travelogues which had been prepared based on

observation and speculation. To produce a detailed study of the islands with

focus on the Jarawa people, we need to view their history in terms of different

periods, based on either (1) direct contact with the people after colonisation of

7
8

the islands or (2) observation/speculation made by voyagers. The history of the

Andamans can be divided broadly into four periods: Pre-Penal Settlement Period

(till 1776), Colonial Period (1776–1947), Independence and after (1947–1996), and

Post-1997 Period (till date).

1.3.1 Pre-Colonial Period

The first description of the islands is found in Ptolemy’s map of the world,

dating back to the second century AD. This map showed some spotted tiny

islands, referred to as Buzacata, in the Indian Ocean. Historical sources show that

China and Japan knew about the islands in the first century AD. Another

description is found in the narrative of two Arab travelers who traversed the

greater part of India and China in the ninth century. These Arab travelers in 857

AD described the people of Andaman as follows:

The people on this coast eat human flesh quite raw; their complexion is black, their hair

frizzled, their countenance and eyes frightful, their feet are very large, and almost a cubit

in length, and they go quite naked. They have no sort of barks or other vessels; if they

had, they would seize and devour all the passengers they could lay hands on.

(Mukherjee 2003:i)

Ajaib-al-Hind, a text of the eleventh century noted:

No one to this day has landed on the Andaman where people are cannibals.

(Mukherjee 2003:ii)

Marco Polo after coasting by the Andaman and Nicobar Islands in 1290 on his

way to Italy mentioned Andaman as Angamanain and described the people in

the following manner:

8
 9

Angamanain is a very long island. The people are without a king, and are Idolaters, and

are no better than wild beasts. And I assure you all the men of this Island of Angamanain

have heads like dogs, and teeth and eyes likewise; in fact, in the face they are all just like

big mastiff dogs! They have a quantity of spices; but they are a most cruel generation,

and eat everybody that they can catch, if not of their own race.

(Mukherjee 2003:i)

These descriptions were not made by observing, watching and making notes

with close examination of these people, but from a brief sail past the islands,

and a short telescopic view of the inhabitants. A description of the land and

the people is to be found after the British established their colony, first in

1789 and then in 1858. After the independence of India in 1947, the Indian

government took over the territory and settled the Indian population from

the mainland.

1.3.2 Colonial Period

The 1771 survey of the Andaman Islands by John Ritchie, a hydrographer of the

East India Company, may be considered as the beginning of a new epoch in the

history of the islands, and the starting point of the Colonial Period. It seems that

to establish a base in the Indian Ocean to protect and escort crew members and

victims of shipwrecks to safety was an important requirement of the British

colonisers. It also became inevitable for its strategic importance, following the

establishment of a Danish mission in the Nicobar Islands and the steady

consolidation of Dutch control over the East Indies. In 1788–89, the Government

of Bengal sent Lieutenants R. H. Colebrooke and Archibald Blair of the Royal

Indian Marine (the British naval force in India) to carry out a survey of the region

and prepare a report. In 1789, Blair established a base on the south-east bay of

9
10

South Andaman. It was named Port Cornwallis after the then Governor-General,

and is known as Port Blair at present. In 1792 the base was shifted northwards to

a better strategic location, to present-day Diglipur. The second settlement

suffered extreme ill-health and a high mortality rate, leading to abandonment of

the settlement on these islands in 1796. However, the impact of contact between

the colonisers and their convicts and native islanders had been long and

insidious. Portman (1899) acknowledges:

Of what took place between 1794 and 1858 we have no knowledge, but when we re-

occupied the Andamans in the latter year, the Aka-Bea-da were by far the stronger and

more numerous tribe.

(Portman 1899: 703)

It appears to me quite possible that some disease was introduced among the Jarawas by

the people of Lieutenant Blair’s Settlement which reduced this tribe considerably in

numbers, and thus enabled the Aka-Bea-da to obtain the upper hand.

(Portman 1899: 703)

In the year following the first Independence Movement in 1857 the British again

established a settlement in Andaman. Nonetheless, there are conflicting views

regarding the setting up of a penal colony in the islands. Some Indian historians

are of the opinion that the main objective of the establishment of a penal colony

in the islands was to transport the rebels to a place where they posed no political

threat to the British. However, Portman (1899) recorded the contrary. He

explicitly noted:

I have often heard it asserted that we occupied the Islands after the Mutiny in order to

have some distant and safe place across the sea to send our rebels and criminals to, when

in the state of the country it would not have been safe to keep them in India; and that the

10
 11

Andamanese have been killed off as a natural result of the occupation, and the country

taken from them without their consent and for no fault of their own. But such was not

the case. Long before the Mutiny the conduct of the Andamanese had made it imperative

that the Islands should be occupied and friendly relations established with the

Aborigines, and this would have done sooner had the Mutiny not broken out.

(Portman 1899: 186)

After the First Independence Movemement, the East India Company was

abolished and its administration taken over by a Court of Directors in

London under the Queen. Taking a lesson from the extinction of Tasmanians,

the Court of Directors maintained that the islanders should be treated “with

every kindness” so long as these islanders recognized British superiority or

did not pose any obstacle to the British interest in the process of colonisation.

Portman (1899) noted:

The English have nothing to reproach themselves with regarding the Andamanese, whatever

may have been the case in Tasmania; and, having the unfortunate experiences of that penal

colony, and our treatment of the aborigines there, before them, the Government of India

adopted a policy towards the aborigines of the Andaman Islands which has made them above

all races of savages, the most carefully tended and petted.

(Portman 1899:209-210)

The first phase of colonisation (1789–1796) witnessed two events. The Jarawa

people behaved in a friendly manner with the colonisers whereas the Aka-Bea

people attempted to resist every effort of colonisation. The second phase of

settlement (1858) was marked with similar resistance from the Aka-Beas, while

the Jarawas were of a peaceful disposition. By 1863, some of the Great

Andamanese communities were befriended and many of them were put in

Homes made for them in and around Port Blair. The rationale behind keeping

11
12

them in Homes was to civilise them. It was soon observed that inmates of the

Homes started falling sick. A number of diseases proliferated first among the

inmates of the Homes and then spread over the territory deep inside the forest.

As a result the Homes were closed down and their inmates sent back to their

habitat. Portman (1899) clearly mentions that his predecessors could do very

little for the communities other than saving them from one disease or another.

Even after realising the consequences of contact, the colonisers were keen to

bring them within the fold of friendliness. The raison dʹêtre of this action was

threefold. First, they wished to ensure that victims of shipwrecks would be

treated well and brought back to Port Blair; second, to ensure that the plan of

colonisation would not be resisted; and finally, to ensure that natives would help

them recapture fugitive convicts (Portman 1899: 49). To achieve this colonial

interest, the authorities tried to befriend the Jarawas, ignoring the consequences.

Portman (1899) records:

In January 1876 two expeditions were sent to search for the Jarawas, and one party

visited four villages, in which they left quantities of unsuitable presents, and brought

back specimens of the Jarawa weapons, and utensils. It was unfortunate that, at the

outset, the Jarawa huts should have been looted thus, and the presents left, being such

things as matches, pipes, tobacco, and looking glasses, the uses of which were unknown

to the savages, were useless to them, and by no means compensated them for the

articles taken away.

(Portman 1899: 717-8)

These expeditions were not the last ones. For years, such expeditions were

organised. Members of such expeditions belonged to Great Andamanese,

particularly Aka-Bea communities and also included convicts. The Aka-Bea

community was traditionally an enemy of the Jarawas. They never missed an

12
 13

opportunity to kill the Jarawas or plunder their camps (Mukhopadhyay 2002:

33). This seems to be one of the reasons why the peaceful Jarawas were hostile to

outsiders.

Later, colonisers abandoned the idea of colonising the Jarawas, but they formed a

Bush Police and established its posts along the periphery of the Jarawa territory.

The objective of these Bush Police posts was to prevent Jarawas from entering the

settlement areas. However, the expeditions did not entirely stop till the colony

fell.

1.3.3 Independence and After (1947-1996)

After India’s Independence in 1947, Andaman and Nicobar Islands came under

the administration of the government of India which changed the policy towards

the Jarawas. According to Singh (2002), India adopted a friendly attitude towards

the Jarawas and instructed the Bush Police to do the same. However the

government’s policy to settle the refugee population in Andaman Islands paved

the way for immigration of more non-Jarawa population to the islands. The

justification for colonisation by the Indian government was to grow more food

and make the islands self-sufficient in the matter of food (Census 1961: Xliv).

This resulted in the clearing of a large number of forest areas on the Andaman

Islands. This reduced the territory of the Jarawas. Till 1950, the Jarawas were

spread over a large area, from Ferrarganj in the south to Mayabunder in the

north. By 1960, large numbers of villages were settled through the Jarawa

territory from Port Blair in south to Diglipur in the north. In the process of

rehabilation of the refugees from Bangladesh in the Andamans, the

administration felt the need to create a buffer zone; the Jarawa Reserve Area was

thus demarcated in 1957. The primary purpose of the buffer zone was to create

13
14

land for the rehabilitation programme and to establish Bush Police posts on the

periphery of the reserve area to protect the settlers and forest workers from the

Jarawas’ fury. Secondly, the land cleared for the settlement was often very close

to the Jarawa habitat. This close proximity of two distinct demographic groups

who had no knowledge of each other created a mutual distrust. Jarawas were

unfamiliar with such demarcation. For them, the settlement of villages was an

encroachment into their territory and habitat. The population of non-Jarawas

went on increasing. Moreover, the exploitation of natural resources (especially

timber, and occasionally other resources which may be of use to the Jarawas) by

the employed workers and settlers continued. Timber factories were also set up.

The government was keen to befriend the Jarawa people. The post-Independence

policy significantly meant no more sending of punitive expeditions to catch or

kill Jarawas. In 1968, three Jarawas were caught in Kadamtalla by villagers, who

were taken to the police station and were kept for over a month, but then sent

back to their habitat with a lot of gifts like bananas and coconuts (Singh 2002).

This helped the administration to extend a friendly hand to the Jarawas.

Meanwhile a contact party was also formed, and sent frequently to the same area

in the Kadamtala region. The objective of the contact party was to build trust

among the Jarawas by offering gifts. There are no reports of destroying their huts

or taking their belongings away, unlike with previous colonisers. By the end of

1997, the Jarawas made contact with friendly gestures.

1.3.4 Post-1997 Period

The sudden appearance of Jarawas in the open during the day in October 1997

was a surprise to both local settlers and the administration. This was the first

group of Jarawas of the Kadamtalla region in Middle Andaman who came out to

14
 15

break the mutual mistrust and hostility that had prevailed for a century. Their

intention was to extend a friendly gesture, rather than confrontation. The

administration responded with a similar gesture by offering them food and gifts,

like bananas and coconuts. This resulted in a relation of trust between the

Jarawas and the administration. Consequently, more Jarawas started coming out

of their earlier designated areas, meeting with local settlers, the police force, and

tourists. Finally in October 1998, Jarawas of South Andaman also came out into

the open like the Jarawas of Middle Andaman.

For the Jarawas, this friendly contact was initially to have an experience of other

world. When they realized that the other world is a store house of food,

implements, and easy transportation, they started exploiting these amenitites.

The administration soon realized that this contact may prove fatal for the Jarawa.

According to Mathew (2002), half of the population was infected with several

diseases — measles, pneumonia, acute upper and lower respiratory tract

infection and fungal infection of skin — between the third week of September

and the third week of October, 1999. Consequently, the administration decided

to stop this contact. Whenever Jarawas appeared on the road or settlement areas,

the administration tried to send them back to their habitat.

1.4 Inhabitants of the islands

Based on linguistic and cultural affinities, Radcliffe-Brown (1948) classified the

Andaman Islanders into two broad categories. One group comprises ten

communities of Great Andaman; the other is made up of Jarawa of Great

Andaman, Onge of Little Andaman, and the inhabitants of Sentinel Island. The

former peoples are called Great Andamanese and the latter Little Andamanese.

15
16

Map 2 presents a comparative picture of indigenous population in Andaman in

1800 and 2000.

Map 2: Inhabitants of the Andamans

(source: [Link])

1.4.1 Great Andamanese

Great Andamanese is a category which applies to ten communities inhabiting the

region from South Andaman to North Andaman. These communities include

Aka-Cari, Aka-Kora, Aka-Bo, Aka-Jeru, Aka-Kede, Aka-Kol, Oko-Juwoi, Aka-

Pucikwar, Akar-Bale and Aka-Bea. Map 3 presents the distribution of these

communities before 1860.

16
 17

Map 3: Distribution of the communities of Andaman prior to 1860

(source: [Link])

All the ten communities of the Great Andamanese group were spread over the

entire Great Andaman region before the establishment of the British colony in

1858. They had their own distinct territory. Jarawas and Jangil occupied a small

area of southern Andaman. Since Great Andamanese communities met with

heavy loses when they came in contact with the colonisers. According to the 1901

census, the population of Great Andamanese had shrunk to 625. The 1931 census

reported their number as 74 plus 16 (74 of pure breed and 16 of half breed) and

only 23 survived till 1974 (Chakraborty 1990), and only a small increase in

population has been observed, and their numbers have now crossed 40 (Abbi

2006).

Radcliffe-Brown (1948: 11–12) notes that all the languages of the Great

Andamanese group were closely related to one another. They had the same

grammatical structure, and a large number of roots were the same in all or in

several of them. Abbi (2006) found that the present day Great Andamanese is an

17
18

amalgam of all the surviving communities of Great Andamanese. She had

further observed that there were still a few speakers of four distinct Great

Andamanese languages, namely Bo, Jeru, Khora and Sare. However, after the

death of Jirake, Boro, Nao Junior, and Boa in the last seven years, the speakers of

only two languages survive now. It was further observed that the languages

were not passed on to the next generation (Abbi et al 2007: 325-343). Jangil

became extinct in the beginning of the twentieth century.

1.4.2 Onge

The Onge community once occupied the entire Little Andaman but they are now

settled in two settlement areas or camps: Dugong Creek and South Bay. Initially,

they came into contact with the British who befriended them only to have control

over the island. Later, after 1950, the non-Onge population started settling in

once the Indian government decided to move refugees to the Andaman Islands.

These non-Onge populations brought diseases and disturbed the ecological

balance of the islands; consequently, the Onge population started declining. At

present their population is around 104. Their language appears to be similar to

Jarawa (Abbi 2006; Blevins 2007).

1.4.3 Jarawa

The Jarawas inhabit the west coast of South and Middle Andaman Islands. In the

initial phase of colonisation and in the early years of its second phase, the

Jarawas occupied the southern part of South Andaman and Rutland Island

further south. In the second phase of settlement, most of the harbour and penal

settlements were established in these areas. By the end of the nineteenth century,

the entire area came under settlement and forest operations. This resulted in the

18
 19

forceful eviction of the Jarawas from their landAnd they were compelled to move

northward. By that time there were decline in power and population of the Great

Andamanese group of people. This paved a safe route for the Jarawas to move

north.

As Portman (1899) noted, the Jarawas were friendly to the colonisers in the first

phase of colonisation. However, the initiative of the British authorities to

befriend them and resultant expedition parties sent off to search and catch

Jarawas became a cause of distrust. This distrust was further strengthened in

post-independent India when the administration of the Andaman region settled

villages on the fringes of Jarawa territory during 1950s–60s and constructed the

Andaman Trunk Road during the 1970s, which cuts through the middle of the

Jarawa territory. Singh (2002) admits that it was a historic blunder on the part of

the government. Singh (2002) further claims that the first contact with the

Jarawas was established in Kadamtalla region in 1969 by the Bush Police. After

that the administration took the decision to continue to befriend the Jarawas and

build trust. The contact party comprised of administrative officials and a handful

of experts. Initially, they would leave gift items hanging from trees on the sea

shore. After some time it was observed that the Jarawas had begun to tie creepers

from trees so that the visitors could hang gifts. This was the beginning of

building trust; though the Jarawas avoided visitors till the beginning of 1974. In

April 1974 the contact party succeeded in meeting a group of Jarawas of the

Kadamtalla region. The friendly relation thus formed was restricted to the

contact party; the entry of other non-Jarawas into Jarawa territory was

persistently resisted. It was observed from the beginning that the Jarawas never

resorted to attack until non-Jarawas entered their territory and indulged in

undesirable activities (which included hunting pigs and deer, and the felling of

19
20

trees). Many unpleasant encounters were also reported when the Andaman

Trunk Road was constructed (Singh 2003).

In April 1996, a Jarawa boy named Enmei in the Kadamtalla region was found

with a broken leg, and was sent to Port Blair for medical treatment. He was

discharged in October 1996 and sent back to his habitat with a lot of gifts. This

became the turning point which marked the end of earlier hostilities and distrust.

After a year, in October 1997, a group of young Jarawas appeared in Uttara Jetty,

with Enmei among them. Again, they were sent back with a lot of gifts. After

that, their visits to Jetties became regular, and more and more Jarawas started

coming out in the open (Report of the Expert Committee on the Jarawas of Andaman

Islands 2003, henceforth Report 2003).

When I visited the area in Feb–March 2007, the Jarawa population was 273. Now

they have crossed 340 (gathered from personal communication with the

Secretary, Tribal Welfare, Andaman and Nicobar Administration in 2010).

Jarawas are monolinguals. However, they have now started picking up some

Hindi words from settlers and tourists. They still prefer their hunting-gathering

life style, though they are attracted to settlement areas and enjoy receiving gifts

from settlers and tourists.

1.4.4 Sentinalese

The Sentinalese are the inhabitants of Sentinel Island. They have resisted all

possible contact with the outsiders till date. Hence little is known about them

and their language. It is their resistance that has helped them to survive.

20
 21

1.4.5 Jangil

Little is known about this community. The Jangils became extinct at the

beginning of the twentieth century.

1.5 The Jarawa people

Jarawas belong to the Negrito stock. They have a short stature. The colour of

their skin is black and hair is frizzy. Children have dark reddish-brown hair

while adults have black curly hair. Signs of ageing are not easily observable;

adults thus often have child-like expressions. They are, in general well-built but

lean, and are healthy and strong. The basic unit in the Jarawa social structure is

the nuclear family, with further organisational levels which make up the

community. These facts have been discussed in the following sub-sections.

1.5.1 Social structure and customs

The social structure of the Jarawas consists of three units: the nuclear family, the

local group or band, and territorial groups. Of territorial groups, there are three.

Division into territorial groups is not based on any difference in social, cultural

or linguistic make-up. Rather, it denotes authority over a particular territory, and

game-hunting and foraging activities within the territory. Thus, the Jarawa

community has three socially discernible territorially divided groups. The

northern group occupies the western part of Middle Andaman Islands while the

central and southern groups occupy the western part of South Andaman. Among

Jarawas, the northern, southern and central groups are known as Tanmad, Boiab,

and Thidong respectively. Among non-Jarawas, these groups are known as

Kadamatalla Jarawa, Tirur Jarawa and R.K. Nallah Jarawa respectively. While no

one from another territory can forage or collect honey within a group’s territory

21
22

without prior permission from the group, movement of Jarawas within and

beyond the territory is not restricted. Generally, when members of one group

enter theterritorial area of another group, they are treated as guests with rights

over the territory; the host group also invites the guests for foraging activities.

When the guests leave the area, they are offered gifts like honey, boar’s meat,

and fish. (Report 2003)

The unit of social organisation, that is, the nuclear family, consists of a husband,

a wife and their children. Children can be from the present marriage or earlier

marriages. Apart from the nuclear family, there exist two other distinct groups:

the bachelors (unmarried young boys above 6-7 years of age) and the maidens

(unmarried young girls above 6-7 years of age). Widowers live with the bachelor

group while widows stay with the maiden group.

Each local group or band is composed of several families and members of

bachelor and maiden groups camp and move together from one place to another.

Most of the members of these groups are related to each other through

consanguinal or affinal relations. The affiliation of the members of these local

groups is not permanent in nature. Rather, processes of fusion and fission

operate. It is possible that members of a local group may split into smaller

groups and move in different directions. It is also possible that they coalesce

again later. Meanwhile, these small breakaway groups may unite with another

group for a short or long period.

A territorial group comprises of several local groups or bands of the same

territory. The territorial division enjoys autonomy. Social interactions between

territorial divisions occur regularly through marriage and social visits. Each

member of each territory is known to all by his or her name.

22
 23

The Jarawa community follows certain kinds of customs. The most important of

them is the ceremony performed on the attainment of adolescence. Known as

lepa, the ceremony is generally performed when a boy attains the age of thirteen

or fourteen years. During lepa, the boy hunts a wild boar on his own (he may be

assisted by his kin) and offers the game to his kin and others present. For girls,

the opamame ceremony celebrates the attainment of puberty. From the very first

day of menarche, her movement is restricted to a small square area surrounded

by four posts. A mixture of əlam (a reddish soil), pig fat, and gum extracted from

a creeper, is applied on her head, neck, and face. During her menstrual period

she does not talk to any person, does not bathe, and remains seated or lies down

on dewa leaves. Every morning these leaves are changed. Some food restrictions

are also observed: she is not supposed to consume pig meat and honey. She only

eats a kind of mollusc and drinks water. After three days, she takes a bath and

consumes pig meat, especially pig fat. She also takes small children into her lap

and prays for fertility. On that day all women and girls of that camp assemble,

dance and sing around the girl. She is decorated with floral ornaments. After lepa

and opamame, the boy and the girl are renamed. (Report 2003)

For marriage, generally the parents of the child or the elder members of the

family start discussing the match when the child is one or two years old. After

the age of seven or eight, the prospective bride or groom may stay in the local

band of the would-be in-laws, and move from one place to another with that

band, but they belong to the bachelor and maiden groups till the marriage is

solemnised. Sometimes such proposed marriage is not solemnised. Age is not

counted in calendar years, but it appears to be that the general age of marriage is

eighteen to twenty years for boys, and fourteen to fifteen years for girls. If the

proposed husband of a girl dies before the marriage, the girl may be compelled

23
24

to marry some widower. The widow and widower may also marry with mutual

consent. (Report 2003)

During the menstrual period of the wife, the husband does not sleep with her.

During the first pregnancy, both the partners remove all ornaments and do not

paint their bodies with white clay, breaking away from normal practice. During

parturition the woman is kept on planks of jungli supari (a kind of wild tree of

which the nut is widely used in India) in a corner, with a fire nearby. Elderly

women of the camp attend the birth of the child. After delivery the umbilical

cord is cut with the head of an arrow, and the mother feeds the baby her

colostrum, which is a secretion similar to milk produced just before birth. Both

the parents take care of the child from birth onwards. The child is breastfed till

the age of two or three, though after teething begins, breastfeeding is

supplemented with food. Usually the mother provides breast milk to the child,

but if necessary other women of the same group also do so. Generally this kind

of situation arises when the mother is not present. It is also observed that an

unmarried girl may also put her breast into the child’s mouth to stop it from

crying when the mother has gone to collect food. Thus, the responsibility to take

care of the young children is shared by the elders and teenage girls. During the

movement of the group from one place to another, the infants are carried along.

1.5.2 Tools and Artefacts

Jarawas mostly make use of forest products for their tools and artefacts, though

they also employ iron and fibres. They have been collecting iron and fibres for

many years, earlier from shipwrecks on the seashore and later from

settlements/villages. They use iron to make arrow tips, knife and knife-like

implements, while fibres are used to make fishing nets. The tools and artefacts

24
 25

are mostly gender-specific and function-oriented. Men make and use tools and

implements necessary for hunting while women make and use fishing nets and

cane baskets. Foraging activities are distributed among men and women.

Hunting wild boar is limited to men. Fishing in streams, nallahs (inlets from the

sea), and other small pits is done by women and children. Fishing in and around

the sea is done by both men and women but preferably by men. Collecting honey

and other forest products, like fruits, is carried out by both; timber collection is

mostly done by women.

As far as their habitation is concerned, the Jarawas construct two major kinds of

huts, (1) semi-permanent and (2) temporary. Known as cadda huʈʰu, semi-

permanent huts are large oval or round community huts constructed at

frequently visited places. Though meant for all seasons, they are especially

favoured as places to stay during the rains. They can accommodate twenty or

more families at a time. They also have accommodation for bachelors and

maidens. Generally bachelors (including widowers) and maidens (including

widows) live in dormitory style, situated at two ends of the semi-permanent

huts. The area for bachelors is known as thorkalang cadda, and that for maidens

thorkongo cadda. Besides these, each family has a demarcated place, called thulə, to

live, with fire nearby. Each family marks their place with four posts where the

husband, wife and their small children sleep. Each semi-permanent hut has a

central fireplace used for baking or boiling meat, besides the individual hearth

for each family. There is a hanging platform for storing pig-fat and meat, above

the central fireplace. One or two hanging platforms are also used for household

items. Pig skulls are also kept hanging from the roof tied with cane strips.

(Report 2003)

25
 26

Temporary huts are of two types. One is very temporary in nature and usually

used during hunting and gathering when there is a requirement for shelter for

the night or for a few days. The other kind is more durable and used for a longer

period. Both types of temporary huts are lean-to-type huts and have a definite

pattern of construction. Temporary huts built for short duration accommodate

the members of a hunting or gathering squad while temporary huts built for a

longer duration can house a family. Durable temporary huts are used mostly

during dry seasons. Lean-to-type huts in general are of two types, either they

face an open space or face away from an open space. Inside the forests these

kinds of huts generally face an open space, but along the sea shore, both kinds of

huts are found. (Sarkar 1990)

1.6 Previous studies

There has been plenty of work done on Andaman and Nicobar Islands, but my

concern in the present work is restricted to the studies conducted on the Jarawa

language. The first work on Jarawa language is by Nair (1974), followed by

Rajasingh and Ranganathan (2000), Senkuttuvam (2000), Sreenathan (2001),

Rajasingh (2002), Abbi (2006, 2009) and Blevins (2007).

1.6.1 Nair (1974)

Nair (1974) presents a preliminary report on Jarawa. His analysis is based on a

small amount of linguistic data. He described the sound system of Jarawa with

not much discussion of other parts of grammar.

26
 27

1.6.2 Rajasingh and Ranganathan (2000)

Rajasingh and Ranganathan (2000) had a specific objective in working on Jarawa.

They collected data with the intention to prepare a primer for the administrative

staff and police force working in and around the Jarawa Reserve Area. They

collected words and sentences used by the Jarawas. Their work includes words

and sentences and their pronunciation alongside translation into English and

Hindi. In the introduction, a list of alphabets (symbols) is provided which are

used in determining the pronunciation of words and sentences of Jarawa. At the

phonetic level there are a lot of critical gaps in this list of alphabets. For example,

the sound [v] is not represented, but there are a large number of words in the

primer which employ the consonant [v]. Moreover, the list provided by the

authors contains a voiceless bilabial aspirated plosive [p ], but their data, rather

shows a voiceless bilabial fricative [φ]. Hence, the data fails to represent the

former, i.e., [p ].

1.6.3 Senkuttuvam (2000)

His work is an attempt to present the phonological sketch of Jarawa language

from his first-hand data (as he was the part of Expert Committee constituted by

the Government of India). He claims that Jarawa is closer to the Dravidian

language family but does not provide historical evidences such as cognates. The

work reveals many phonological misrepresentations. For instance, he tentatively

posits a back velar stop /q/ and glottal stop /Ɂ/ in the Jarawa inventory. Similarly,

he puts forward labio-dental fricatives /f/ and /v/, but the occurrence of /f/ is rare

in his data while /v/ shows up in many words. The bilabial approximate /w/ is

shown substitutable by /v/ in his description, for example /ya:w/ vs. /ya:v/ ‘tail’.

27
28

He describes five nasals as phonemes but in his description the dental, alveolar,

retroflex, palatal, and velar are shown to be occurring as homorganic nasals. He

also considers length as phonemic but his data-set lacks any of this evidence,

since /va:ya:/ vs. /va:ya/ ‘grandmother’, and /va:na:/ vs. /va:na/ ‘weep’ in his data-

set, demonstrates that length is not phonemic in this language. Moreover, he

does not distinguish between the phonetic symbols for dental and retroflex

sounds in his work which creates further confusion.

1.6.4 Sreenathan (2001)

Sreenathan (2001) makes an attempt to present the psychological and cultural

state of the Jarawas through their language. He collected first- hand data to this

end (as he was the part of the Expert Committee constituted by the Government

of India), and presents, in this work, a phonological analysis of the language. His

attempt to relate language to the psychological state of the Jarawas is not my

concern here, but rather the fact that in the area of language analysis, he failed to

represent the sounds properly. There are many sounds represented by one and

the same symbolwhich creates confusion. For example, his phonetic chart

demonstrates that dental and retroflex stops are two different phonemes. But he

employs the symbols /t/ and /d/ for both types of stops. He also fails to note the

existence of voiceless aspirated plosive sounds, which are in abundance in

Jarawa.

1.6.5 Rajasingh (2002)

Rajasingh (2002) presents a brief description of the sound system of Jarawa based

on his earlier collaborative work (Rajasingh and Ranganath 2000). He provides

the placement of vowels and consonants on the phonetic chart, and a word as

28
 29

evidence for each phoneme. A drawback of his work is that he does not use

different symbols for different types of sounds. For instance, he posits dental

stops and retroflex sounds, but uses the same symbol for both types of sounds.

Similarly, dental, alveolar, palatal, and velar nasals are represented by the same

IPA symbol. In his data, there are labio-dental fricatives /f/ and /v/, but he does

not place these sounds in the phonetic chart. Furthermore, what is described at

one place as bilabial fricative is transcribed as labio-dental fricative /f/ in another

place. Moreover, he does not clarify why he assumes that these sounds are

phonemes in this language at all.

1.6.6 Abbi (2006)

Abbi (2006) described the languages of Andaman with focus on Great

Andamanese, Onge and Jarawa; this work being a comparative study of the three

languages. This study is based on fieldwork conducted in 2001–2002. She also

provides supplementary data on Great Andamanese from her 2004–2008 project.

Based on the linguistic evidence she collected, Abbi is the first linguist who

proposed that Great Andamanese and Jarawa-Onge are two different language

families. She is also the first linguist to identify retroflex and voiceless aspirated

plosives in these languages, giving detailed phonetic charts and description of

phonotactic occurrences.

1.6.7 Blevins (2007)

Blevins (2007) made an attempt to relate Jarawa-Onge to Austronesian language

families. By employing comparative methods of historical linguistics, she

demonstrated that Jarawa and Onge are the two lost sisters of the Austronesian

29
30

language family. Her analysis is based on earlier works on Onge and Jarawa,

including my field-data-set2 then available to her.

1.7 The Present Study and its Methodology

1.7.1 Present Study

As I have mentioned in the beginning of this chapter, the present study was

undertaken to arrive at a descriptive grammar of Jarawa,. For this purpose,

Jarawa data was collected during fieldwork conducted during November 2003 to

July 2006. Based on the data elicited, an attempt has been made to present this

descriptive grammar and its position in the typology of languages.

1.7.2 Methodology

At the outset, I would like to admit that the data elcitation was not an easy task.

In my effort to collect data among the monolingual Jarawa community, I came

across three significant challenges: firstly, to accustom my ear to sound system of

the target language; secondly, to initiate data collection from the monolingual

community and thirdly, to learn the language. Each challenge had its own sets of

problems and required careful steps to move forward. Another substantial task

was to avoid cultural barriersbetween the researcher and the consultants, as we

did not know each other’s cultural and social norms. I did know that any minor

violation of their cultural and social norms might land me in a situation of

mistrust which might further lead to unacceptability within the community.

Even my research permit given by the administration contained restrictions and

2 I had been doctoral fellow at the Max Planck Institute for Evolutionary Anthropology, Leipzig/
Germany and she had been scientist there when the paper was written.

30
 31

conditions about how to behave; how, when and where to contact these people,

with instructions about what to do and what not to do. Within these restrictions,

I followed some careful principles and techniques based on the situation. In

subsequent sub-sections I will present an account of the general method of field

linguistics and some of the ways I adopted to elicit data from the target

community.

[Link] Selection of consultants

Selection of consultants is a fundamental issue in conducting any linguistic

research, particularly for writing a grammar of an unknownor not very well

known language. The first requirement for the selection of consultants is that the

consultant must be a native speaker. This was not the problem in this

community, since everybody was a native speaker of Jarawa. We generally also

divide the consultants based on age, sex and region. Data collection was done

primarily in the Kadamtalla region (see § [Link]). Both the sexes were accessed

for this work. All age-groups were consulted but the corpus of this study

contains people above 15 years of age. The determination of age is ad-hoc and

based on the guesses of the researcher; for the community does not have any

system to count the age in terms of a calendar year. Nor is there any

demographic record of Jarawas available in Government documents.

[Link] The Selection of the area

Jarawa people are divided into three groups based on their territorial divisions

(see § 1.5.2). The data was mostly collected in Kadamtalla region. This is because

the research permit had clear instructions to visit Jarawa Reserve Area with the

staffs of Adim Adivasi Janjaati Vikash Samiti (AAJVS) stationed at Kadamtalla

31
32

and to collect data in the presence of staff. Nonetheless this region had many

advantages. It was easy to visit the Jarawas of RK Nallah. It was also easy for me

to visit the Jarawas through the sea-route. Thecorpus of this thesis hence contains

data provided by the Jarawas of other two regions also. This could be made

possible because of Jarawas of other regions visiting the Kadamtalla region and

also because of my own initiative and the availaibility of opportunities to make

short visits to the other two regions along with the AAJVS staff.

[Link] Tools for the study

I used two tools in the course of this research, namely, (1) systematic observation

and (2) questionnaire. In systematic observation, participant observation seemed

to be the best option since the community is monolingual. In such circumstances,

a structured questionnaire generally does not work. Hence an unstructured

questionnaire was adopted as a tool; however structured questionnaire was also

employed whenever it was possible.

[Link].1 The questionnaire

There are two kinds of questionnaire, structured and unstructured. I adopted

both the methods. The method of unstructured questionnaire was, however,

followed for the major part of the fieldwork. There were two reasons to adopt the

unstructured questionnaire: whenever I visited the community, I came across

different situations and different speakers. Mostly depending on the situation

and the consultant, I made an attempt to elicit data accordingly. Many times I

used these occasions as opportunities to cross-check previously elicited data.

Varied situations and varying speakers made my data rich.

32
 33

The structured method was adopted only in restricted cases. There are two

things to be noted in this regard: the structured method was adopted while

pictorial method was employed, i.e. collecting data with the help of pictures; and

there was an attempt to verify the data.

[Link].2. Systematic observation

Participant observation

In order to collect linguistic data in a monolingual community, participant

observation is one of the significant methods. This method not only allowed me

to observe the way they speak, but also provided me an opportunity to know

how they perform other activities. It also helped me note down the conversation,

and its context. Moreover, it helped consultants get rid of their hesitation in the

presence of an outsider, since Ibecame a part of their daily lives. This provided

me the opportunities to learn many things about their social and cultural norms

without disturbing them. Throughout my fieldwork in the Andamans, I adopted

participant observation method along with other methods which are described in

§ 1.7.4.

[Link] Field Procedures

I followed different techniques to collect data in the field. These techniques

included pictorial method, showing objects to them to know the name of those

objects; listening to their conversations, and recording those conversations. In the

initial phase, carefully listening to their conversations, recording pieces of

conversations, and listening to those conversations back home where I used to

stay were the major steps I took. This helped me accustom my ear to the sound

system of the language. Meanwhile, I started collecting data with help of pictures

33
34

related to body-parts, birds, animals, fish etc. It was followed by noting words by

pointing to objects that were lying around. I also attempted to list words related

to daily activities such as run, walk, sleep, eat, drink etc. by pointing to these

activities if somebody was enacting them. Moreover, I kept noting down simple

sentences which I could understand (after collecting some data). I also noted the

contexts of sentences/conversations. As time passed I picked up the language

and tried to speak small phrases. This gave me an opportunity to learn, converse,

and collect further data on the language.

1.7.3 Limitations of the study

Despite all the efforts made by me to elicit maximum amount of, there were

limitations to this study. Firstly, there was a very limited amount of time

available to be in contact with the community. As mentioned earlier, the

condition that AAJVS staff must be present at all times caused a severe

restriction of contact between the researcher and the consultants. Secondly,

whenever the group of Jarawas stayed far away from the road side along the sea

shore, contact with Jarawas depended on the short visits made available with

AAJVS staff. It is true that I was totally dependent on the mercy of the AAJVS

staff. Thirdly, it had, always, been a problem and challenge to obtain data from a

monolingual community to write a descriptive grammar. I am aware that there is

a lot of grammatical information which may have escaped my attention. Readers

will also agree with me that it is not possible to unearth each and every

grammatical information of the language in a limited time period. Hence, this

work is a modest attempt to present a descriptive grammar of the Jarawa

language from a typological point of view based on the data I could elicit from

my consultants during the period of my stay at Great Andaman.

34
 35

1.8 Organization of the present work

Chapter 2 describes the sound system of the language. This chapter begins with

the identification of sounds and their phonemic status in the language. To

establish the phonemic status of the sounds, minimal and sub-minimal pairs are

provided. Moreover, the distribution of sounds at syllable and word level is also

discussed. This chapter also investigates syllable structure and other

phonological processes found in this language.

Chapter 3 discusses the morphology of the language. This includes noun

morphology, verbal morphology, adjectives and adverbs. Copulas and

postpositions are also described. Noun morphology also deals with pronouns

and nouns, and their behaviour in the language. Verbal morphology consists of

suffixes and prefixes which are attached to the verbs, and their usage. There is

also a brief discussion on adjectives and adverbs in the same chapter.

Chapter 4 gives an account of the syntactic structure of the language. This

description includes verbless predicates and verbal predicates. This chapter also

focuses on negative constructions, interrogative clauses and imperative

structures. Complex constructions are also included in this chapter.

Chapter 5 concentrates on the typological features of the language. An attempt is

made to classify and categorize the language typologically.

Chapter 6 presents the conclusion of the thesis.

35
36

CHAPTER 2: PHONETICS AND PHONOLOGY

2.0 Introduction

The preceding chapter was a detailed description of Andaman, its history and

culture, the Jarawa community, their history, culture, and society. The present

chapter discusses the phonetics and phonology of the language with focus on

segmental phonology and describes the vowel and consonant inventory.

Furthermore, I will make an attempt to enquire into the phonemic status of the

sounds3 present in Jarawa. This will be followed by a discussion of syllable

structure and other phonological phenomena in the language.

The chapter is organized in the following manner. Section 2.1 focuses on the

problems of sounds and phonemes in this language and the criteria that have

been employed to claim the phonemic status of a sound. It also talks about

phonemic contrasts of sounds, where I provide minimal pairs. I go on to describe

the phonemic distribution of sounds in syllables and words. Then I provide the

occurrences of sounds in different environments. In section 2.2 the syllable

structure of Jarawa is discussed. Section 2.3 describes syllabification, and section

2.4 deals with word structure. Section 2.5 focuses on the morphophonemics of

Jarawa phonology. A brief conclusion of the chapter is presented in section 2.6.

3 The term sound, in this dissertation, is used for phone.

36
 37

2.1 Sounds and Phonemes

In this section I discuss different techniques that were employed in establishing

whether a particular sound is a phoneme in this language. I also present the

methodological principles adopted.

The sound inventory of Jarawa contains 41 sounds, of which 28 are consonants

and 13 are vowels. However, it is difficult to claim that all the sounds are

phonemes in this language, for it is difficult to provide minimal pairs for every

pair of sounds in order to establish the phonemic status of a particular sound.

Even so, I have made an attempt to provide minimal pairs wherever the corpus

contains such data. The main problem in researching the sound system of Jarawa

is that we are not able to claim with authority that a particular sound is a

phoneme in this language, because the language demonstrates considerable

variation. Since there is a lot of variation at the level of sounds, it is pertinent to

discuss this aspect first. There are instances where two pairs of sounds, for

example /p/ and /b/ can be used interchangeably in the words like bəᵼʈʰe ‘go’. The

word bəᵼʈʰe can be pronounced as pəᵼʈʰe. Since we have minimal pairs which

demonstrate that these two sounds are phonemes in this language, we may

deduce that in such words they are in free variation. But in the case of palatal

plosives /č/ and /ǰ/ it becomes hard to decide whether they are phonemes or they

are allophones of the phone /č/ since we lack minimal pairs.

We can deduce from the preceding observation that whether a sound is a

phoneme or allophone is not always distinct. This difficulty arose during

research mainly due to lack of minimal pairs for each pair of sounds. Further,

those pairs of sounds which have minimal pairs also demonstrate some variation

37
38

in the words. In order to assert a conclusive result in such circumstances we

needed to enquire the phonological knowledge of speakers. Phonological

knowledge of speakers contains the phonological contrast which amounts to

minimal pairs. However, when a community is monolingual, it becomes difficult

to decipher every aspect of phonological contrast for each phoneme. In this

context it is important to state the methodological principles adopted to identify

phonemes. Four methodological principles were employed to state the phonemic

status of a sound in this language.

1. Phonological contrast: This highlights whether the particular sound is

contrastive in nature by looking at minimal pairs. There are some pairs of

sounds which have minimal pairs.

2. Variability among sounds: If there is no minimal pair for a pair of sounds,

an attempt was made to find out if there is any variability between the

sounds; and, if it exists, what the contexts of occurrences are for those

sounds in the language. If there is any variability and if it is difficult to

claim the phonemic status of the sounds, I have adopted pattern congruity

as the method to support the claim; for example in the case of the sounds

/č/ and /ǰ/.

3. Distribution of sounds: I have also made an attempt to investigate the

distribution of sounds at syllable level and word level. Examining the

occurrences of sounds at different levels assists in determining whether

the variability of the sounds is due to some kind of conditioning

(phonological or lexical) or it is an idiolectal variation.

4. Statistical criterion: Since some pairs of sounds show alternation among

themselves, it becomes difficult to infer which sound has the actual

representation in the words. To determine the actual representation of

38
 39

sounds in words, I have adopted a statistical criterion where the

frequencies of occurrences of two sounds are [Link] is, if one

sound is used more frequently in a word than the other, the former is

considered to be the phoneme, and latter an idiolectal variation. This

criterion is employed if there is no viability of having definite sets of rules

which apply to a set of data to claim the allophonic status of certain

sounds.

2.2 Phonemic Contrasts

In this section, I will attempt to provide minimal pairs to establish the phonemic

status of vocalic and consonantal sounds wherever it is possible.

Vowels in this language can be classed in three groups: two front vowels, two

back vowels and three central vowels. The front vowels are [i] and [e], back

vowels [u] and [o], while the central vowels are [ᵼ], [ə] and [a]. There are also two

mid-low vowels, [ɛ], and [ɔ], but their phonemic status is not clear. Their

occurrences in the data of our corpus are limited, and I could not obtain any

minimal pairs. Length is also phonemic in this language; hence we have short

and long vowels. However, no length distinction was found in the case of the

high-mid central unrounded vowel. Thus we have 13 vowels. Some nasalized

vowels were also found, but their phonemic status can not be established in the

present work.

Among the 28 consonants, there are voiced and voiceless plosives along with

voiceless aspirated plosives. Sounds like nasals, trill and retroflex flap, lateral

and retroflex lateral also occur in this language. There are two approximants,

labial and palatal. There are also a few fricatives like pharyngeal fricative, and

39
40

bilabial fricative. Two labialised consonants exist, pharyngeal fricative, and

voiceless aspirated velar plosive.

2.2.1 Vowels

Jarawa exhibits the basic seven vowel system which is illustrated in table 2.1.

Table 2.1: Jarawa vowel phonemes

Front Central Back

High i u

High-mid ᵼ

Mid e ə o

Low a

This basic seven vowel system can further be categorized into two groups: one

which has qualitative contrasts in itself and another which contains quantitative

differences. In qualitative difference, Jarawa employs three way distinctions in

the height: high, mid, and low; three way distinction of tongue position: front,

central, and back; and two way distinction in the position of lips: rounded and

unrounded. There is one vowel which may be categorized as a high-mid central

unrounded vowel [ᵼ]4. The minimal pairs and sub-minimal pairs are given in the

next section to demonstrate their phonemic contrasts.

4 Though this symbol represents high central unrounded vowel in IPA, I have used this symbol
for high-mid central unrounded vowel in order to avoid confusion with schwa.

40
 41

[Link] Qualitative Contrasts

Qualitatively, vowels in Jarawa consist of high, mid, and low vowels in front,

central, and back positions with rounding and non-rounding features. This three-

way combination of articulation yields seven vowels. These vowels are /i, e, u, o,

a, ə, ᵼ/. The phonemic status of /ᵼ/ has been established in this language but it is

still difficult to provide acoustic evidence to claim that the phoneme /ᵼ/ is a high-

mid central unrounded vowel. At the auditory level, this phoneme seems to be a

little bit higher than the mid central unrounded vowel /ə/ but lower than the

high central unrounded vowel /ᵼ/. Moreover, two mid-low vowels have been

encountered, but their phonemic status is not clear. Their occurrences are also

limited in the data and I do not have any minimal pairs at present. I will present

their distribution in phonological structure in section 2.1.2. Below I present some

minimal pairs and near-minimal pairs to establish the phonemic status of

vowels.

/i/ vs. /e/

The close high front unrounded vowel /i/ stands in contrast with the half close

mid front unrounded vowel /e/.

(2.1) /li/ ‘Proximate Demonstrative’

(2.2) /-le/ ‘Plural Marker’

/i/ vs. /u/

The high close front unrounded vowel /i/ contrasts with the high close back

rounded vowel /u/.

41
42

(2.3) /hiɽu/ ‘black’ /innə-gə cew/ ‘smells good’

(2.4) /hulu/ ‘hot’ /unnə/ ‘return home’

/i/ vs. /ᵼ/

The high close front unrounded vowel /i/ is also in contrast with the central half

close mid unrounded vowel /ᵼ/

(2.5) /ukᵼ/ ‘bird’s wing’

(2.6) /ugi-k-əᵼʈʰe/ ‘fly[V] (compound word made of ugi-k- əᵼʈʰe, here əᵼʈʰe

means ‘go’ -k- is an empty morpheme, and /ugi/ may

correspond to a bird or its feather)

/e/ vs. /o/

The half close mid front unrounded vowel /e/ stands in contrast with the half

close mid back rounded vowel /o/

(2.7) /tape/ ‘moon’

(2.8) /tapo/ ‘good’

/u/ vs. /o/

The high close back rounded vowel /u/ contrasts with the half close mid back

rounded vowel /o/

(2.9) /oʈʰa/ ‘mind/head’

42
 43

(2.10) /uʈʰa/ name of a person

/o/ vs. /a/

The half close mid back rounded vowel /o/ differs from the open low central

unrounded vowel /a/

(2.11) /topo/ ‘snake’

(2.12) /tapo/ ‘good’

/a/ vs. /ə/

The open low central unrounded vowel /a/ contrast with the half close mid

central unrounded vowel /ə/

(2.13) /hʷawə/ ‘watercourse’

(2.14) /hʷəwə/ ‘wild boar’

/ᵼ/ vs. /ə/

The half close mid central unrounded vowel /ə/ is in contrast with the half close

high-mid central unrounded vowel /ᵼ/

(2.15) /ukᵼ/ ‘bird’s wing’

(2.16) /-ugə/ ‘flesh’ (part of a compound word /hu-ugə/ ‘pig’s

flesh’)

43
44

[Link] Quantitative Contrasts

Whether quantitative contrast appears across the board in this language is not

clear because not all the quantitative contrasts have minimal pairs. However,

there is evidence of some minimal pairs, which indicate there is quantitative

difference in this language. The data presented here covers both types of cases. I

have divided the examples into two groups; one which has minimal pairs and

another which does not.

Vowels Group A. This group consists of those vowels which have

minimal pairs and near-minimal pairs. It is evident from these examples that

length is phonemic in this language.

/a/ vs. /a:/

(2.17) /tapo/ ‘good’

(2.18) /ta:po/ ‘want/seek[v]’

/i/ vs. /i:/

(2.19) /ipo/ ‘bark’

(2.20) /i:po/ name of a person

/ə/ vs. /ə:/

(2.21) /čəw/ ‘wild potato’

(2.22) /lə:w/ ‘honey’

44
 45

Vowels Group B. In this group I have placed those vowels for which we

do not have minimal pairs but there are instances where lengthening of

vowels has been observed.

/e/ vs. /e:/

(2.23) /pela/ ‘soil’

(2.24) /pe:l/ ‘shell fish’

/u/ vs. /u:/

(2.25) /ulli/ ‘stone’

(2.26) /u:ɖɖu/ ‘cough’

/o/ vs. /o:/

(2.27) /olla/ ‘today’

(2.28) /o:lleg/ ‘Lakralunda’ (name of a place in Kadamtala region)

2.2.2 Consonants

Jarawa has 28 consonantal sounds which are given in table 2.2. Among these 28

consonants are voiced and voiceless plosives, and voiced nasals. It also has

voiceless aspirated plosives. It employs five places of articulation for voiced and

voiceless plosives and voiced nasals. The aspirated counterparts of the voiceless

plosives are also attested. The voiceless aspirated plosives are found at four

places, dental, retroflex, palatal and velar. Non-nasal continuants like trill,

retroflex flap, lateral, and retroflex lateral also occur in this language. Their

45
46

distribution in the phonological structure of this language is limited. It has two

approximants, labial and palatal; a trill and a retroflex flap, and two laterals,

aleveolar, and retroflex. There are a few fricatives like pharyngeal fricative, and

bilabial fricative, though the bilabial fricative was not found in abundance. There

is one labialised pharyngeal fricative /hʷ/ and one labialised voiceless velar

aspirated plosive / kʰʷ/.

Table 2.2: Jarawa consonant phonemes

Bilabial Dental Alveolar Retroflex Palatal Velar Pharyngeal

Unaspirated p b t d ʈ ɖ č ǰ k g
Plosives
Aspirated tʰ ʈʰ čʰ kʰ
Plosives kʰʷ
Nasals m n ɲ ŋ
Fricatives ɸ ʃ h
hʷ
Trill r
Tap/flap ɽ
Lateral l
Approximant w j

[Link] Plosives and Retroflex

Jarawa uses voiceless and voiced plosives. It also distinguishes between

aspirated and unaspirated plosives. Voiceless and voiced unaspirated plosives

occur at five places of articulation, bilabial, dental, retroflex, palatal and velar.

Thus the inventory contains voiceless plosives /p, t, ʈ, č, k/ and voiced plosives

are /b, d, ɖ, ǰ, g/. For aspirated voiceless plosives, only four places of articulation

46
 47

are used: dental, retroflex, palatal, and velar. Thus /tʰ, ʈʰ, čʰ, kʰ/ are the voiceless

aspirated plosives. I have categorized the plosive sounds into three groups

(Plosives Group A, B and C below): one which has minimal pairs, the second

which does not have minimal pairs, and the third which lacks minimal pairs but

presents a lot of variability. I also discuss the variation in the usage of these

sounds. I have adopted the methodological principle stated in section 2.1 to

decide whether there are phonemes or allophones.

Plosives Group A. In this group I describe those plosives and retroflex

which have minimal pairs. Though there exists variation in the usage of the

sounds, I will attempt to discuss the variation, and thereafter based on the

methodological principles, decide the phonemic status of the sounds.

/p/ vs. /b/

The voiceless unaspirated bilabial plosive /p/ differs from the voiced unaspirated

bilabial plosive.

(2.29) /pe:l/ ‘shell fish’

(2.30) /be:l/ ‘dewli fish’

(2.31) /poɖ/ ‘honey found inside the tree’

(2.32) /bod/ ‘make net basket’

These minimal pairs demonstrate that the sounds /p/ and /b/ are phonemes in

this language. As said earlier, there are instances where these sounds seem to be

in free variation. Thus, the word /bəᵼʈʰe/ ‘go’ is also pronounced pəᵼʈʰe. Similarly,

47
48

/piʈʰi/ ‘bad’ is, sometimes, articulated as biʈʰi. But instances of /b/ realised as /p/ in

words like bəᵼʈʰe are around 10–20%. Similarly, the instances of /p/ realised as /b/

in words like /piʈʰi/ are again around 10–20%. The frequencies of usage of /p/ or

/b/ in these words led me to consider /bəᵼʈʰe/ as the form for ‘go’. Similarly I have

taken /piʈʰi/, not /biʈʰi/ as a form for ‘bad’ used in this language. It is difficult to

ascertain whether it is a kind of a synchronic phenomenon or there is some

diachronic process that the language has undergone, since its closest sister, Onge,

still does not have the bilabial voiceless plosive (Abbi 2001, 2005; Sharma and

Dasgupta 1978).

/t/ vs. /d/

The voiceless unaspirated dental plosive /t/ contrasts with the voiced

unaspirated dental plosive /d/.

(2.33) /pato/ ‘ornament’ (ornamental string of leaves around waist)

(2.34) /pado/ ‘rope’

With this pair of sounds I have not found any variation.

/t/ vs. /tʰ/

The voiceless unaspirated dental plosive /t/ stands in contrast with the voiceless

aspirated dental plosive /tʰ/.

(2.35) /pato/ ‘ornament’ (ornamental string of leaves around waist)

(2.36) /patʰo/ ‘fishing net’

48
 49

I have not observed any variation in this pair of sounds.

/tʰ/ vs. /d/

The voiceless aspirated dental plosive /tʰ/ differs from the voiced unaspirated

dental plosive /d/.

(2.37) /patʰo/ ‘fishing net’

(2.38) /pado/ ‘rope’

Again, no variation was found in this pair of sounds.

/tʰ/ vs. /ʈʰ/

The voiceless aspirated dental plosive /tʰ/ stands in contrast with the voiced

aspirated retroflex /ʈʰ/.

(2.39) /patʰo/ ‘fishing net’

(2.40) /paʈʰo/ ‘arrow’

This pair of sounds again lacks variation.

/ʈʰ/ vs. /ɖ/

The voiceless aspirated retroflex /ʈʰ/ is contrastive with voiced retroflex /ɖ/.

(2.41) /ʈʰi/ copula

49
50

(2.42) /ɖi/ referential marker

No variation has been observed in this pair of sounds.

/č/ vs. /čʰ/

The phoneme /č/ is a voiceless unaspirated palatal plosive while /čʰ/ is a voiceless

aspirated palatal plosive.

(2.43) /četali/ ‘the day before yesterday’

(2.44) /čʰedali/ ‘now’, ‘at present’

Again there is no variation between the two sounds.

/k/ vs. /g/

The phoneme /k/ is a voiceless unaspirated velar plosive while /g/ is a voiced

unaspirated velar plosive.

(2.45) /ukᵼ/ ‘bird’s wing’

(2.46) /ugə/ name of a person

This pair of sounds has some variations. For example, the word /kᵼyə/ ‘forest’ can

be articulated as /gᵼyə/. But such variation is limited to this word. We have a

word /gəgapa/ ‘talk’ which does not demonstrate any variation. Similarly there is

a word /kag/ ‘breast’ which does not show any variation. Hence, /k/ and /g/ are

considered phonemes in this language.

50
 51

Plosives Group B. In this group I place those sounds for which no

minimal pairs exist in the corpus, but which seem to be phonemes. I will

attempt to discuss if there is any variation between the pairs of sounds.

/d/ vs. /ɖ/

The voiced unaspirated dental plosive /d/ differs from the voiced unaspirated

retroflex /ɖ/.

(2.47) /bod/ ‘make net basket’

(2.48) /poɖ/ ‘honey found inside the tree’

No variation observed in this pair of sounds.

/ʈ/ vs. /ɖ/

The voiceless unaspirated retroflex /ʈ/ contrasts with the voiced unaspirated

retroflex /ɖ/.

(2.49) /oʈə:re/ ‘back of the neck’

(2.50) /oɖə/ ‘hair’

This pair of sounds lacks variations.

/t/ vs. /ʈ/

The voiceless unaspirated dental plosive /t/ stands in contrast with the voiceless

unaspirated retroflex /ʈ/.

51
52

(2.51) /t-ita/ ‘eat[V]’

(2.52) /ʈiʈa innə/ ‘cockroach’

This pair of sounds does not show any variation.

/ʈ/ vs. /ʈʰ/

The voiceless unaspirated retroflex /ʈ/ differs from the voiceless aspirated

retroflex /ʈʰ/.

(2.53) /ʈiʈa innə/ ‘cockroach’

(2.54) /ʈʰilʈə inə/ ‘weevil’

No variation has been observed between this pair of sounds.

/k/ vs. /kʰ/

The phoneme /k/ is a voiceless unaspirated velar plosive while /kʰ/ is a voiceless

aspirated velar plosive.

(2.55) /kopel/ name of a person

(2.56) /kʰo/ ‘hold[V]’

This pair of sounds does not demonstrate any variation.

Plosives Group C. In this group there is only one pair of sounds. In this

pair we have palatal plosives. However, the corpus lacks minimal pairs to

52
 53

confirm that they are phonemes. A lot of interchangeability has been

observed in this pair, but this is not true of all the words where one or the

other occurs. If we take pattern congruity as the principle to establish the

phonemic status of this pair of the sounds, we may claim that these sounds

are phonemes in this language.

/č/ vs. /ǰ/

The phoneme /č/ is a voiceless unaspirated palatal plosive while /ǰ/ is a voiced,

unaspirated palatal plosive. Their existence as distinct phonemes has not been

established. The data so far demonstrates that they are in free variation in this

language, and no minimal pairs exist on record to claim that they are distinct

phonemes. We have seen above in the case of bilabial sounds /p/ and /b/ that

though they can sometimes be substituted for one another, they are still distinct

phonemes, since minimal pairs for them have been found. The same is the case

with the sounds /k/ and /g/. In the case of /č/ vs. /ǰ/, there are words like /čə:w/

‘wild potato’ which may be pronounced as /ǰə:w/. But also attested are words like

/čawa:ja/ ‘walk[v]’ which is never articulated as /ǰawa:ja/. There are, in fact, a lot of

words that begin with /č/ which cannot be replaced by /ǰ/. This creates the

suspicion that these sounds could be considered as distinct sounds in this

language. However, the question is still open as to whether /č/ and /ǰ/ are distinct

phonemes in this language or they are allophonic variants of the same phone.

(2.57) /čawa:ja/ ‘walk[V]’

(2.58) /ǰagijə/ ‘fall[V]’

53
54

[Link] Nasals

Jarawa uses voiced nasals. It employs four places of articulation for the nasals,

bilabial, dental, palatal, and velar. Thus, we have four nasals in this language,

/m, n, ɲ, ŋ/. They seem to be in contrast with each other. However, the minimal

pairs for these nasals are not straightforward. Rather, I will demonstrate here

that contrasts exist between the following pairs: /m/ and /n/, /m/ and /ɲ/, /n/ and

/ɲ/, and /n/ and /ŋ/. However, minimal pairs have not been found for the pair /ɲ/

and /ŋ/.

/m/ vs. /n/

The phoneme /m/ is a voiced bilabial nasal while /n/ is a voiced dental nasal.

They stand in contrast in Jarawa.

(2.59) /m-apo/ ‘my elder brother’

(2.60) /napo/ ‘fish’

/m/ vs. /ɲ/

The phoneme /m/ is a voiced bilabial nasal, while /ɲ/ is a voiced palatal nasal.

They are in contrast in this language.

(2.61) /ma:g/ ‘mine’

(2.62) /ɲa:g/ ‘no’

54
 55

/m/ vs. /ŋ/

The phoneme /m/ is a voiced bilabial nasal, while /ŋ/ is a voiced, velar nasal.

They stand in contrast in this language.

(2.63) /mi ~ ma/ 1st Person Pronoun

(2.64) /ŋi~ŋa/ 2nd Person Pronoun

/n/ vs. /ɲ/

The phoneme /n/ is a voiced dental nasal, whereas /ɲ/ is a voiced palatal nasal.

The examples (2.65-2.66) demonstrate their contrastive nature.

(2.65) /inijəla/ ‘know[V]’

(2.66) /iɲijačʰe/ ‘forget[V]’

/n/ vs. /ŋ/

The phoneme /n/ is a voiced dental nasal, while /ŋ/ is a voiced velar nasal. The

examples below show contrast.

(2.67) /ən-/ generic possessive marker for humans

(2.68) /əŋ/ ‘Jarawa’

[Link] Fricatives

Jarawa has a few fricatives. Among those fricatives is the voiceless bilabial

fricative /ɸ/, though at present, evidence of minimal pairs which demonstrate its

55
56

existence as a separate phoneme has not been found. Similarly, the voiceless

palatal fricative /ʃ/ is found in this language, but as the corpus suggests, it seems

to be an allophone of the voiceless aspirated plosive /čʰ/; they exist in free

variation. Other kinds of fricatives are the voiceless pharyngeal fricative /h/ and

the labialized voiceless pharyngeal fricative /hʷ/. According to the corpus, it is

difficult to establish whether they are distinct phonemes or they stand in

allophonic distribution. In the present study, I have considered /ɸ/ and /h/ as

phonemes, and /ʃ/ and /hʷ/ as sounds, but further inquiry is required to ascertain

whether they are phonemes or are in allophonic distribution in this language.

/ɸ/

(2.69) /aɸle/ ‘fasten[V]’

I was unable to find any alternate sounds for /ɸ/. Further, the Jarawa corpus does

not contain minimal pairs of /ɸ/ with other corresponding sounds to support a

claim that it is phonemic in the language. This sound can occur syllable-initially

and -finally, but its occurrences at word level are limited to initial and medial

positions. Thus, based on the corpus, it is difficult to claim whether it is a

phoneme or a sound in this language. Since there is no variability, the present

work assumes that it is a phoneme.

56
 57

/čʰʰ/ vs. /ʃʃ/

/ʃ/ is a voiceless palatal fricative while /čʰ/ is a voiceless aspirated palatal plosive.

They are in free variation in this language; /ʃ/ can always substitute the sound

/čʰ/. The examples in (2.70) and (2.71) demonstrate their variation.

(2.70) /ʃɛla/ name of a woman

(2.71) /čʰɛla/ name of a woman

Here the phoneme /čʰ/ and sound /ʃ/ seem to be in allophonic distribution as they

are in free variation. However, in one word /naʃa/ ‘penis’ in my corpus , it seems

that the phoneme /ʃ/ is not substituted by /čʰ/. /čʰ/ is, moreover, a phoneme, as is

evident from examples (2.43) and (2.44), but /ʃ/ does not seem to be. Hence, /ʃ/ is

considered to be in free variation with /čʰ/ in this language.

/h/ vs. /hʷ/

The phoneme /h/ is voiceless pharyngeal fricative and /hʷ/ is its labialised

counterpart.

(2.72) /huʈʰu/ ‘big/fat’

(2.73) /hʷəwə/ ‘wild boar’

Since they cannot be substituted for each other, I will consider them as

phonemes. /h/ appears syllable-initially and syllable-finally, while /hʷ/ can only

57
58

occur syllable-initially. There is no evidence to say that these two sounds can be

substituted. There is also no evidence as to whether these sounds can be replaced

by any other sounds in this language.

[Link] Trill, Flap and Laterals

At the sound level, Jarawa exhibits one trill, one tap, and two lateral

approximants — an alveolar lateral and a retroflex lateral. Thus, the inventory is

/r, ɽ, l, ɭ/. It is evident from examples in (2.74– 2.77) that the sounds /r, ɽ, l/ are

phonemes in this language. The occurrence of sound /ɭ/ is limited in my corpus

and I do not have evidence to claim that the sound /ɭ/ is a phoneme, nor to claim

that it is a variant of another sound, particularly /l/.

/r/ vs. /ɽ/

(2.74) /wara/ ‘crow’

(2.75) /waɽa/ ‘swim[V]’

/ɽɽ/ vs. /l/

(2.76) /hiɽu/ ‘black’

(2.77) /hulu/ ‘hot’

[Link] Approximants

Jarawa has two approximants, bilabial /w/ and palatal /j/. The examples

presented in (2.78) and (2.79) are minimal pairs which demonstrate that they are

58
 59

phonemes in this language. Moreover, we have not observed any variation

between this pair of sounds.

/w/ vs. /j/

(2.78) /wa:na/ ‘weep[V]’

(2.79) /ja:nə/ ‘egg’

2.3 Phonemic distribution

In this section, I will discuss the distribution of sounds, examining which sounds

can occur syllable-initially, -medially, and -finally. I will also make observations

on the occurrence of sounds at the word level, that is, which sounds which can

occur word-initially, -medially, and -finally. In Jarawa, there are different

structures for the distribution of vocalic and consonantal sounds. Vowels always

form the nucleus of a syllable. Therefore, it is unnecessary to examine the

occurrences of a vowel in syllable-initial syllable-medial or syllable-final

positions. Hence, I have dealt with vocalic sounds only vis-à-vis words where

they can occur word-initially, -medially or -finally. For consonants, I have made

an attempt to examine whether they can appear as onset or coda in a syllable.

Moreover, I have also tried to see whether there is a possibility to consider their

occurrences at the word level. No consonant clusters have been found at syllable

level.

2.3.1 At syllable level

Syllables in Jarawa consist of an optional onset and an optional coda. Data

suggests that Jarawa constitutes a syllable using one consonantal sound at onset

59
60

and one at coda position. However, there are restrictions on the usage of sounds

at different levels. Hence, a word consisting of CVCV contains two syllables, CV

and CV. In the case of a word of CVCCV structure, I have considered this word

as consisting of two syllables, CVC and CV (for a detailed discussion of

syllabification, see section 2.5.

[Link] Consonants

Most of the consonants are found syllable-initially and -finally. The sounds /ʈʰ, č,

čʰ, ǰ, ɲ, r, hʷ/ are not found syllable-finally. Table 2.4 presents the occurrences of

consonants at syllable level. I have considered polysyllabic words to demonstrate

that these consonants can occur syllable-finally because there is no evidence for

the occurrence of these sounds in monosyllabic words. Wherever possible, I have

made an attempt to provide monosyllabic words.

Table 2.3: Distribution of Jarawa consonants in syllable

Consonant Initially Finally

p paʈʰo ‘arrow’ a:p ‘jackfruit’

ɸ ɸi:na ‘a kind of mosquito’ aɸle ‘bind[v]’

b boi͂ja ‘small’ le:b ‘road/path’

t taŋ ‘tree’ ʈʰotkili ‘small vehicle’

d dojə ‘girl’ bod ‘make net basket[V]’

tʰ tʰulə ‘sleep [V]’ ən-ahotʰwag ‘tie [V]’

ʈ aʈiba ‘speak [V]’ peʈpel ‘butterfly’

ʈʰ ʈʰomaje ‘wave’

60
 61

ɖ ɖag ‘coconut’ ənahoɖ ‘tooth’

c čawa:ja ‘walk[V]’

cʰ čʰumijə ‘stick used to make

fishing net’

j ǰagijə ‘get down [V]’

k ka:ja ‘mother’ tʰitʰəkdəhə ‘candwa fish’

kʰ hʷəʈkʰele ‘bald’ ənikʰwa ‘ear’

g gəgapə ‘sing[V]’ oha:g ‘flower’

m mala ‘many’ na:m ‘firewood’

n na:po ‘fish’ ʈʰinon ‘hillock’

ɲ ɲoɲo ‘to feed milk’

ŋ ŋi ‘you’ čəla:ŋ ‘snake’

r čokogere ‘helicopter’

ɽ waɽa ‘swim[V]’ həɽkʰəjə ‘a kind of insect’

l lə:w ‘honey’ pe:l ‘shell fish’

w wa:na ‘weep[V]’ a:w ‘bow’

y jakeka ‘day after tomorrow’ oʈʰaj-le ‘boil[V] (potato)’

h huʈʰu ‘big’ hʷa:h ‘stream / creek’

hʷ hʷəwə ‘pig’

61
62

2.3.2 At word level

At the word level, most of the sounds occur word-initially, word-medially and

word-finally.

[Link] Vowels

At the word level there is ample evidence to claim that vowels can occur word-

initially, -medially or -finally. However, their distribution is not uniform across

these positions. Only short vowels can occur in all three positions, whereas long

vowels may occur only word-initially or-medially. I have presented the

occurrences of short and long vowel in table 2.4.

2.4: Distribution of Jarawa vowels at word level

Vowels Initially Medially Finally

i iŋkə ‘thirst’ hiɽu ‘black’ piʈʰi ‘bad’

e ekane ‘bring[V]’ peʈpel ‘butterfly’ ča:lahe ‘stay[V]’

u unnə ‘come back [V]’ hulug ‘hot’ hiɽu ‘black’

o ohag ‘flower’ halog ‘cane tree’ paʈʰo ‘arrow’

a aɸle ‘bind [V]’ taŋ ‘tree’ pela ‘land’

ə əŋka:b ‘old man’ itəŋ ‘a kind of tree’ iŋkə ‘thirst’

ᵼ wᵼwə ‘rain’ ukᵼ ‘wing’

i: i:nnen ‘non-Jarawa’ əni:pi:l ‘hand’

e: e:poči ‘name’ le:b ‘road/path’

u: u:ɖɖu ‘cough’ bu:čʰu ‘vessel/pot’

62
 63

o: o:tʰahe ‘sneeze [V]’ no:ha ‘bird’

a: a:p ‘jackfruit’ ta:po ‘want/seek [V]’

ə: əŋ ‘Jarawa’ lə:w ‘honey’

[Link] Consonants

The table below illustrates the occurrences of consonants at the word level in

Jarawa.

2.5: Distribution of Jarawa consonants at word level

Consonants Initially Medially Finally

p paʈʰo ‘arrow’ čapʈʰe ‘climb’ a:p ‘jackfruit’

ɸ ɸi:na a kind of aɸle ‘bind[V]’

mosquito

b boi͂ja ‘small’ ʈʰapabad a kind of le:b road/path

flower

t taŋ ‘tree’ ʈʰotkili ‘small

vehicle’

d dojə ‘girl’ pado ‘fishing net’ bod ‘make net

basket [V]’

tʰ tʰulə ‘sleep [V]’ ən-ahotʰwag ‘tie [V]’

ʈ aʈiba ‘speak [V]’ peʈpel ‘butterfly’

toʈkʰolo a kind of

63
64

flower

ʈʰ ʈʰomaje ‘wave’ paʈʰo ‘arrow’

ɖ ɖag ‘coconut’ ənahoɖ ‘tooth’

c čawa:jə ’walk[V]’

cʰ čʰumijə ‘stick used to ačʰu ‘bad smell’

make fishing net’

j ǰagijə ‘get down[V]’

k ka:ja ‘mother’ tʰitʰəkdəhə ‘candwa

fish’

kʰ kʰo ‘hold[V]’ ənikʰwa ‘ear’

hʷəʈkʰele ‘bald’

g gəgapə ‘sing[V]’ ənkegili ‘dumb’ oha:g ‘flower’

m mala ‘many’ əmal ‘hive’ na:m ‘firewood’

n na:po ‘fish’ ʈʰinon ‘hillock’ ʈʰinon ‘hillock’

ɲ ɲoɲo ‘feed milk[V]’ iɲijačʰe ‘forget[V]’

ŋ ŋi ‘you’ čəla:ŋ a kind of

snake

r čokogere ‘helicopter’

ɽ waɽa ‘swim[V]’

həɽkʰəjə ‘a kind of

insect’

64
 65

l lə:w ‘honey’ ullel ‘big water’ pe:l ‘shell fish’

ipilhunu ‘one-

handed’

w wa:na ‘weep[v]’ əŋhowal ‘pool of a:w ‘bow’

water’

y jakeka ‘day after oʈʰaj-le ‘boil[V] oɫɫa:j ‘name’

tomorrow’ (potato)’

h huʈʰu ‘big’ noha ’bird’ hwa:h

‘stream/creek’

wa:h a kind of

wild tuber

hʷ hʷəwə ‘pig’

2.4 Syllable Structure

The syllable structure of Jarawa consists of an optional onset and coda, and a

nucleus. Thus, it attests all the four variants of minimal possible syllable

structures — V, CV, VC, and CVC. Examples in (2.80) illustrate this.

(2.80)

(a) V e: ‘yes’

(b) CV mi 1st P Pronominal

(c) VC əŋ ‘Jarawa’

(d) CVC ka:g ‘breast’

65
66

2.5 Syllabification

Jarawa attests all the four variants of minimal possible syllable structures, but the

present corpus suggests that only one consonant is allowed in onset and coda

positions. If a word is composed of more than one syllable, the parsing of the

syllable would be as shown in (2.81).

(2.81)

(a) CVCV [Link]

(b) CVCCV [Link]

(c) VCVC [Link]

2.6 Word Structure

Words in Jarawa may be monosyllabic, disyllabic, or polysyllabic entities. These

combinations are explained with reference to the occurrences of vowels in the

beginning of words, or medially or finally. I consider here words which end with

a consonant (closed syllable) or without a consonant (open syllable). Let us

consider the following words:

Monosyllabic

Open

(2.82) V e: ‘yes’

(2.83) CV ka interrogative marker

66
 67

Closed

(2.84) VC a:w ‘bow’

(2.85) CVC waŋ ‘some’ ɲa:g ‘no’

Disyllabic

Open

(2.86) VCV enɛ ‘coconut milk’

(2.87) VCV ulli ‘stone / pebble’

(2.88) CVCV noha ‘bird’ wᵼwə ‘rain’

(2.89) CVCCV čapʈʰe ‘climb[V]’

Closed

(2.90) VCVC əmal ‘hive’ otaŋ ‘coconut fibre’

čonel ‘banana’

(2.91) VCCVC ullel ‘big water’

(2.92) CVCVC ʈʰomaj ‘salt water’ ʈʰinon ‘hillock’

(2.93) CVCCVC peʈpel ‘butterfly’

67
68

Polysyllabic words

Trisyllabic

Open

(2.93) VCVCV aʈiba ‘speak[V]’

(2.94) CVCVCV čiləhe ‘star’ hʷečʰebə ’earthquake’ janela ‘egg’

(2.95) VCCVCCV orgonʈʰo ‘short’

(2.96) CVCVCV ʈʰujuni ‘ female pig’

Closed

(2.97) VCVCVC ičʰitaŋ ‘root of a plant’

(2.98) VCCVCVC əŋhowal ‘pool of water’

(2.99) CVCVCVC ʈʰapabad ‘a kind of flower’

Tetrasyllabic

Open

(2.100) VCVCVCV ilijema ‘[Link][V]’

(2.101) CVCVCVCV ʈʰučaʈəgə ‘where’

(2.102) VCCVCVCV ənkegili ‘dumb’

68
 69

(2.103) VCVCCVCV ipilhunu ‘one-handed’

Closed

(2.104) VCVCVCVC ənohabag ‘a piece of glass used to look one’s face’

2.7 Morphophonemics

In this section I will discuss the phonological changes which occur in this

language. There are two kinds of phonological processes which will be discussed

here. Section 2.7.1 focuses on two forms of pronominals, which I have termed the

‘mi form’ and the ‘ma form’, conditioned according to how they are used. Section

2.7.2 presents a complex phenomenon which Jarawa exhibits, wherein two words

are taken together and the form of the second word is reduced. For example, mi

‘1GEN’ and paʈʰo ‘arrow’ are taken together and pronounced as /maʈʰo/ ‘my arrow’.

Here paʈʰo is only pronounced as /ʈʰo/. This phenomenon is very common in

Jarawa.

2.7.1 mi vs. ma

There are two forms of pronominals in each person in Jarawa. For the first person

there are /mi/ and /ma/, for second person /ŋi/ and /ŋa/ or /ni/ and /na/, and for

third person /hi/ and /hiwa/, though we have shorter forms for these pronominals

which are not taken into consideration in this section. These pronominals are

used in the positions of subject and object of a verb, and as possessive pronouns.

Throughout the dissertation I have taken /mi/, /ŋi/, and /hi/ form as ‘mi form’ and

/ma/, /ŋa/, and /hiwa/ or /wa/ as ‘ma form’. However, there is a great difference in

the usage of the mi form and the ma form in the syntax, especially in possessive

69
70

constructions (for detailed discussion, see § 3.3.1). In the present chapter, I

enquire into whether there is any phonological reason to use mi vs. ma forms,

particularly in possessive constructions. In these constructions, for first and

second person any either form may be used, but only /wa/ is permitted for third

person. The examples in table 2.6 illustrate some of the patterns of usage of mi

and ma forms.

Table 2.6: Usage of mi form and ma form

1st Person 2nd Person 3rd Person

mi form ma form mi form ma form

mi čaɖɖa ma-čaɖɖa ŋi čaɖɖa ŋa-čaɖɖa wa-čaɖɖa

‘my hut’ ‘your hut’ ‘his/her hut’

mi paʈʰo ma-paʈʰo ŋi paʈʰo ŋa-paʈʰo wa-paʈʰo

‘my arrow’ ‘your arrow’ ‘his/her arrow’

mi ipaɖa m-ipaɖa ŋi ipaɖa ŋ- ipaɖa h- ipaɖa

‘my injury’ ‘your injury’ ‘his/her injury’

mi a:w m-a:w ŋi a:w ŋ-a:w wa-a:w

‘my bow’ ‘your bow’ ‘his/her bow’

70
 71

mi əččəle5 ŋi əččəle wa əččəle

‘my child’ ‘your child’ ‘his/her child’

m-itəǰile ŋ-itəǰile wa-itəǰile

‘my friend’ ‘your friend’ ‘his/her friend’

In the above data, there are three things that are evident. First, there is no change

in the possessive prefix of the third person; second, the mi form can attach to any

word, across the board; third, the ma form occurs with those words which have

back vowels in the first syllable of the word. The data also confirms that the ma

form is only used when there is affixation; i.e. attaching of the possessive prefix

to the possessed noun. If there is no affixation and rather, both the possessive

pronominal and the possessed noun are used as two independent words, the mi

form is generally used. Note that the environments showing vowel harmony

suggested that the process at work is, in fact, affixation.

2.7.2 Reduced forms

There is a general tendency in Jarawa to reduce the form of a nominal or verb

when it is used in a possessive construction or in compound words. As

mentioned previously, /mi paʈʰo/ ‘my arrow’ can be articulated as /maʈʰo/. In

/maʈʰo/ there are two morphemes, mi ‘1GEN’ and paʈʰo ‘arrow’, where /paʈʰo/ is

reduced to /ʈʰo/ and combines with /ma/ to become /maʈʰo/. The general rule to

reduce the form is to take last syllable of the possessed noun and join it with the

5 It is observed that there is no affixation of possessive prefix with əččəle ‘child’.

71
72

possessive prefix. Another example is with the compound word ugi-k-əᵼʈʰe ‘fly’.

In this compound are two words, /ugi/ and /bəᵼʈʰe/ where the phoneme /b/ has

been deleted in /bəᵼʈʰe/. There is no evidence so far to show why /k/ is inserted in

this compound. Therefore, I will consider /k/ as an empty morpheme here.

2.8 Conclusion

This chapter has discussed the problems associated with the identification and

establishment of sounds in Jarawa, and the methodological principles adopted to

decide whether a sound is a phoneme or not in this language. I have also covered

vowels and consonants. In the discussion of the distribution of the sounds, I have

demonstrated the occurrences of consonants at both syllable and word level and

the occurrences of vowels at word level. The chapter also analysed the

phonological processes in which there was an issue related to the use of ‘mi’ and

‘ma’ forms in possessive constructions. I have also observed that there is a

tendency in this language to use reduced forms in the case of possessive

constructions and compounds of two words, where the second word is reduced.

72
 73

Chapter 3: Morphology

3.0 Introduction

The preceding chapter was a detailed examination of phonetics and phonology

of Jarawa. This chapter focuses on Jarawa morphology with specific emphasis on

different word classes. Jarawa is an agglutinating language and has simple

morphology in terms of affixation, i.e. it takes prefixes and suffixes. There are

two kinds of prefixes: one is pronominal which attaches to verbs, adjectives, and

nouns; the other indicates definiteness or referentiality, and attaches only to

verbs. The principal functions of suffixes are (1) to convey plurality when

attached to nouns, and (2) to express mood, modality, and evidentiality when

attached to verbs. They can also, however, attach to adjectives, where they may

denote either state or evidentiality.

The chapter is organized in the following manner. In section 3.1, a brief

introduction to the concepts of morpheme and word in Jarawa is presented.

Section 3.2 focuses on word classes in Jarawa, and section 3.3 discusses noun

morphology. A detailed examination of Jarawa verb morphology is presented in

section 3.4 and of Jarawa adjectives in 3.5. Section, 3.6 presents adverbs, and

section 3.7 postpositions. Section 3.8 is a discussion of copula, and section 3.9

focuses on word formation. A brief conclusion is presented in 3.10.

3.1 Morpheme/word

The criterion that has been followed throughout the dissertation to identify a

morpheme is: a phonologically defined set of segments having a constant range

73
74

of meaning in the overall structure of the language is a morpheme. Phonetic

criteria of the production of a defined set of segments in one breath, or with a

pause, have also been taken into consideration, particularly to decide whether

two morphemes are articulated as a word or two different words.

In Jarawa, a morpheme may be a free root which exists independently, as in the

case of /napo/ ‘fish’. It may be a bound root which occurs with obligatory

prefixes (possessive prefixes in the case of a noun) as in ən-oɖə ‘hair’ (where {ən-}

is used as generic possessive marker for humans). Sometimes, a verb may be a

bound root such as ən-ətəhə ‘sit’ (imperative) for humans. In both cases whether

it is the nominal root oɖə ‘hair’ or verbal root ətəhə ‘sit’, they cannot occur

independently if they refer to human beings. The nominal root oɖə ‘hair’ can be

used for the hair of humans as well as non-human animals while verbal root

ətəhə ‘sit’ is only used for human beings. In the case of non-humans, the third

person genitive marker would be used with such a root. For example, if the hair

of a dog is to be referred to, wᵼmə wa-oɖə ‘dog 3POSS-hair’.

Thus, a bound root has to be prefixed with some morpheme(s). There are two

kinds of prefixes which co-occur with nominals: one is the definite genitive

marker (personal possessive pronoun), and the other the generic human

possessive pronoun {ən-}. The morpheme {ən-} is a more obvious choice for

prefixation in case of humans. It has two interpretations in this language; one is a

generic interpretation for humans, and the other, a second person pronominal.

There are other kinds of morphemes which can attach to roots/stems — whether

free or bound — as prefixes or suffixes; these can be classified as bound

morphemes. For instance, the word bəᵼʈʰe-jə ‘go-VEVD’ contains two morphemes;

74
 75

one is bəᵼʈʰe ‘go’ and the other is {-jə} ‘VEVD’. Here bəᵼʈʰe ‘go’ is a free morpheme

and {-jə} ‘VEVD’ is a bound morpheme. A morpheme which occurs as a free root

or bound root, will, in the present study, be referred to as a lexical word, and a

morpheme which has to be attached to a root/stem as an affix, as a bound

morpheme.

3.2 Word classes

In order to identify word classes in Jarawa, it is necessary to go beyond notional

definitions that entail much overlapping. Therefore, a scrutiny of different words

is attempted through different diagnostic tests. Consequently, two more criteria,

morphological diagnosis, and syntactic distribution, have been employed to

identify different word classes in Jarawa. The morphological test is based on the

choice of the occurrences of prefixes and suffixes with the root form. Specifically,

the choice of a prefix or a suffix depends upon the lexical category of the root.

For instance, Tense Aspect and Mood (TAM) category only comes with a verbal

roots or a root used as the verbal predicate of a clause. In the syntactic

distribution test, the occurrences of lexical roots in certain syntactic

configurations have been evaluated. The distribution of lexical roots in a

clause/sentence depends upon the lexical category they belong to. For instance, a

lexical root may occur either as a subject, object, adjunct, or predicate, in a clause.

After scrutiny it was found that Jarawa has six kinds of word classes: noun, verb,

pronoun, adjective, adverb, and postposition. Nouns, verbs, adjectives and

adverbs are open classes, where nouns and verbs are abundant while adjectives

and adverbs are fewer in number; postpositions and pronouns are closed classes

with a small number of lexemes. In (3.1) these classes are arranged in decreasing

order of the number of words in each class.

75
76

(3.1) noun > verb > adjective > adverb > pronoun > postposition

3.3 Noun Morphology: Nouns and Noun Phrases

This section focuses on nouns and pronouns, and their morphological behaviour

in the language. The section also discusses other categories that could be part of a

Noun Phrase (NP), and their behaviour within the phrases. An NP can have

noun or pronoun as head. Section 3.3.1 discusses pronouns and their different

forms, section 3.3.2 focuses on nouns and their forms, and section 3.3.3 deals

with the noun phrase and elements which could be part of the noun phrase.

3.3.1 Pronouns

Pronouns are heads of NPs. They have no lexical content, and either refers to

some nouns which may have occurred within the sentence, or draw reference

from context. These types of NPs occur at the place of noun-headed NPs to avoid

repetition. There are three kinds of pronouns: personal pronouns (pronominals),

demonstrative pronouns, and interrogative pronouns.

[Link] Personal pronouns

Jarawa uses personal pronouns to refer to humans and non-humans. It employs

mi as first person, ŋi as second person, and hi as third person pronoun. Second

person has, however, three variants {ŋi, ni, and ən} and third person has two

variants {hi and əhi}. {ən} has two functions in this language. First, it is used as

second person pronominal (generally as subject and possessive forms). Second, it

is used as a pronominal that has generic interpretation for humans. Jarawa

personal pronouns are summed up in table 3.1 below.

76
 77

Table 3.1: Jarawa personal pronouns

1st Person mi

2nd Person ŋi ~ ni ~ ən

3rd Person hi ~ əhi

Of these pronouns, first person and second person pronouns are used to refer to

human beings while third person pronouns can be used to refer to both humans

and non-humans, as illustrated in examples (3.2–3.6).

(3.2) mi omohə

1 sleep

I/we sleep.

(3.3) ŋi bəᵼʈʰe-jə

2 go-VEVD

You are going.

(3.4) hi ma napo ɖ-ija

3 1ACC fish REF-give

He gave me the fish.

(3.5) mi ih-ita

1 3-eat

I ate that[-human].

77
78

(3.6) he h-ita

3 3-eat

He ate that[-human].

There are no plural counterparts of these pronominals in Jarawa. Thus, the same

personal pronouns can be used for both singular and plural. For example, the

personal pronoun mi could mean first person singular or first person plural

depending upon the context in which the communication is carried out. The

sentence (3.7) demonstrates that there is no plural form of these pronominals.

(3.7) mi bəᵼʈʰe-jə

1 go-VEVD

I/we are going.

The corpus collected in the field shows that this sentence (3.7) was used when a

person was talking about himself, and also when he was talking about a group of

people. As we have seen above, second person and third person pronouns have

variants which can be termed as allomorphs of the respective pronouns. Apart

from the allophonic variations of these pronouns, the pronouns have “ma” and

“mi” forms in all the persons (see § 2.7.1). These two forms of pronominals are

shown in table 3.2.

Table 3.2: ma and mi pronominal forms

Person mi form ma form

1st Person mi ma

78
 79

2nd Person ŋi ~ ni ~ ən ŋa ~ na

3rd Person hi ~ əhi hiwa

It is difficult to describe the exact behaviour of these two forms; at the level of

morphology (particularly possession) these forms seem to occur in free variation.

In the present study these pronominals are divided into two groups according to

their usage. At the level of syntax, there seems to be distinction: one form is used

as subject and the other form is used as object. Forms like mi, ŋi, and hi are

always used in subject position of a sentence/clause. ma, ŋa, and hiwa are used in

object position of a sentence/clause. This is illustrated in examples (3.8) and (3.9).

(3.8) ŋi ma əjojəba

2 1ACC see

You saw me.

(3.9) mi ŋa napo ɖ-ija

1 2 ACC fish REF-give

I gave you the fish.

These examples show “mi” forms in the subject position and “ma” forms in the

object position of the clause. However, examples like (3.10) show that mi forms

may also be used in the object position of a clause.

(3.10) mi napo ŋi h-ija

1 fish 2 3-give

I gave you a fish.

79
80

This raises the question as to whether the hypothesis that “mi” forms are subjects

and “ma” forms are objects is valid. It is also difficult to say whether it is an

idiolectal variation. In the present study there is no answer to this issue.

However, the usage of “mi” forms in object position is limited to a few instances

in the present corpus. Also, as far as “ma” forms are concerned, they can never

occur as subject of the clause. Thus, we assume in the present study that the

distinction between these forms is as hypothesised.

In the case of possessive constructions, both the forms are possible. Within the

scope of this study, it can only be said that the distinction between “mi” and

“ma” forms seems to be phonological to some extent, though not always (for

detailed discussion, see § 2.5.1).

[Link].1 Morphological nature of personal pronouns

Pronominals in Jarawa can occur independently or can be attached to nouns,

verbs, adjectives or some other words. They can attach to a noun as a possessive

prefix; they can be prefixed to a verb either as an S(ubject) argument of an

intransitive verb or as an O(bject) argument of a transitive verb; and they can

also attach to predicative adjectives as their S argument.

It has been observed that these pronominals are prefixed to verbs, nouns, or

adjectives which begin with a vowel. In such cases the forms that are prefixed are

phonologically similar shortened forms of the independent forms. The

independent forms and their prefixed forms are given in table 3.3.

80
 81

Table 3.3: Independent pronominals and their prefixed forms

Person Independent forms Prefixed forms

1st Person mi ma m-

2nd Person ŋi ~ ni ŋa ~ na ŋ- ~ n- ~ ən-

3rd Person hi hiwa h- ~ hi- ~ ih- ~ he- ~ əh-

The prefixed form for the first person is {m-}, and is used for both mi and ma.

Second person has three variant forms in the “mi” form and two in the “ma”

form. The prefix {ŋ-} is used for ŋi and ŋa, and {n-} for ni and na. {ən-} is used only

as the prefixed form. In third person category, there are many prefixed forms.

These different forms are generally used as co-referring elements in a clause. The

third person category co-refers to third person object NPs in the clause. The third

person prefixed forms seem to be lexically conditioned.

A pronominal S argument of an intransitive verb which begins with a vowel may

be prefixed to the verb. This is illustrated in example (3.11).

(3.11) m-aʈiba

1-say

I say/said.

This is optional. The above sentence can also be uttered without prefixing the

pronominal, as given in (3.12).

81
82

(3.12) mi aʈiba

1 say

I say/said.

In transitive constructions, a pronominal O argument of a transitive verb which

begins with a vowel is prefixed to the verb. (3.13) illustrates the prefixation of the

O argument to the verb.

(3.13) mi ŋ-əjojəba

1 2ACC-see

I saw you.

Note that in the case of a ditransitive construction, a pronominal IO (Indirect

Object) does not get prefixed to the verb; shown in (3.14)

(3.14) hi čonel ɖi ma h-ija

3 banana REF 1ACC 3-give

He gave me the banana.

One possible explanation involves the presence of prefix {h-} on the verb, which

prevents the indirect object pronominal to be prefixed with the verb. The verbal

prefix {h-} denotes the presence of a Direct Object (for detailed discussion, see §

[Link].2). In (3.15) referential marker {ɖ- ~ ɖi} is prefixed to the verb {ija} when

the DO is demoted but in (3.16) when {h-} is attached to the verb, {ɖi} stands

independently.

82
 83

(3.15) mi ŋa napo ɖ-ija

1 2ACC fish REF-give

I gave you the fish.

(3.16) hi ma čonel ɖi h-ija

3 1ACC banana REF 3-give

He gave me the banana.

This supports the hypothesis that the prefixation of {h-} to transitive or

ditransitive verbs replaces the attachment of any other elements including IO in

ditransitive constructions.

A similar structure is found in the possessive construction, where the attachment

of the possessive prefix to the possessed noun is optional (3.17 and 3.18).

(3.17) m-a:w

1-arrow

My arrow.

(3.18) mi a:w

1 arrow

My arrow.

In (3.17) there is a possessive prefix m- on the possessed noun a:w ‘arrow’, while

in (3.18) the possessive pronoun is not prefixed, but uttered as a different word.

This flexibility is not allowed with inalienable possessions such as body parts,

83
84

and kinship terms. In such cases prefixation is obligatory. This is illustrated in

(3.19).

(3.19) m-ikʰwa

1POSS-eye

My eye.

Pronominals can also be attached to the Interrogative Marker (INTM) ka. ka

generally occurs at the beginning of a yes/no question. In this construction, the

pronominal, used as the S(ubject)/A(gent) argument of the clause, is suffixed to

the interrogative marker. Consider example (3.20).

(3.20) ka-ŋi kekahəʈə olleg ɖə bəᵼʈʰe

INTM-2 tomorrow Olleg LOC go

Will you go to Olleg tomorrow?

This is not obligatory, as illustrated by (3.21).

(3.21) ka ŋi kekahəʈə olleg ɖə bəᵼʈʰe

INTM 2 tomorrow Olleg LOC go

Will you go to Olleg tomorrow?

Nonetheless, the occurrence of (3.20) is more frequent than (3.21) in daily usage.

The morphological behaviour of pronominals of this language raises the question

of whether what I have called prefixation is truly affixation, or these reduced

84
 85

pronominals are in fact, clitics. Further investigation is required to determine the

precise nature of these reduced forms.

[Link] Demonstrative pronoun

In Jarawa, there are two kinds of demonstrative pronouns: one indicating

proximate object(s) while another denoting distant object(s). The proximate

demonstrative is li and the distal demonstrative is luwə. In this section, we

discuss the use of demonstratives as pronouns. These demonstratives are also

used as adjectives and adverbs (discussed in sections 3.3.5 and 3.3.6

respectively).

In (3.22), li is used as a third person pronominal where it refers to a person who

is climbing the tree. In fact, li is employed more frequently for third person

references than hi in subject position. luwə is used for distant object(s). Sometimes

it creates problems in interpreting the data because the same lexeme is also used

as adjective or adverb. Sentence (3.23) demonstrates the use of luwə as

demonstrative pronoun.

(3.22) li taŋ ɖə čapʈʰe-jə

PDEM tree LOC climb-VEVD

He is climbing the tree.

(3.23) luwə paʈʰo

DDEM arrow

That is an arrow.

85
86

[Link] Interrogative pronoun

In Jarawa, there are two interrogative pronouns: onne ‘who’ and onəhə ‘what’.

Another question word onəhəle ‘what-VS’ seems to be derived from onəhə ‘what’. I

propose this due to the {-le} suffix, which is generally employed with nouns to

denominalise them (for detailed discussion, see § 3.9.2). Onəhəle may, here, mean

‘what are (you) doing?’ or ‘why are (you) doing?’ Other question words like

kahaje ‘where’, ʈʰučəgə ‘where’, nojčʰe ‘how many’ may have some differences

since these words are used with Interrogative Marker (INTM).

3.3.2 Nouns and noun classification

Nouns can be morphologically classified into two groups according to their

occurrence with possessive prefixes. Nouns which do not obligatorily take

possessive prefixes can be classified as free forms, and those which obligatorily

take possessive prefixes as bound forms. Nouns denoting body parts or kinship

terms exist as bound forms; the morphemes which stand for names of body parts

or kinship terms must be prefixed with a possessive pronoun. For instance, the

word ecepo ‘eye’ cannot occur in isolation — it must occur with a possessor, as in

wa-ecepo ‘his eye’, m-ecepo ‘my eye’, ŋ-ecepo ‘your eye’, ən-ecepo ‘eye {human

generic interpretation}’, and so on. Similarly, kinship terms also must occur with

possessive prefixes. There is, however, a slight difference between the usages of

possessive prefixes with these two types of bound forms. In the case of a body

part, the nominal component can take possessive prefix {ən-} which has generic

interpretation for humans but kinship terms never take this prefix. It may be

because body part nominal may be referred to in general, but in the case of a

kinship term it must be interpreted as belonging to a specific individual; it is

yours, his, or mine (for detailed discussion, see § [Link]).

86
 87

Accordingly, my classification of Jarawa nouns is generally based on whether

they take affixes obligatorily. There are thus two kinds of nouns: one which take

affixes obligatorily (Class I) and the other which does not (Class II).

I also make an attempt to classify Jarawa nouns on a semantic basis. This gives

rise to three categories: celestial bodies, cultural artefacts, and flora and fauna.

a) Noun Class I

Noun Class I consists largely of nouns denoting body parts and kinship terms. In

section (i) below I focus on nouns related to body parts, and in section (ii) I

present a discussion of nouns related to kinship terms.

(i) Body part nouns

As mentioned earlier, nouns which are names of human body parts are bound

roots in Jarawa; otherwise they may occur as free roots. They can occur with the

generic human possessive prefix or the definite possessive prefix (for details, see

§ [Link]). There are two body part nouns čeŋ ‘blood’ and hʷetʰo ‘nerve’ which

occur as free roots. Also attested are compound nouns, like ən-iɲabotalaŋ ‘nostril’,

which is composed of two nouns, iɲɲabo ‘nose’ and talaŋ ‘hole’, and an obligatory

prefix {ən-}.

Table 3.4: Names of body parts in Jarawa

Body part Jarawa word

1. nose ən-iɲapo

87
88

2. nostril ən-iɲapotalaŋ

3. ear ən-ikʰwa

4. ear hole ən-ikʰwagdalaŋ / ən-ikʰwatalaŋ

5. hair ən-oɖə

6. beard ən-imugə

7. eye ən-epo / ən-ečepo

8. eyeball ən-epohiʷugə

9. eyebrow ən-epoʈʰugə

10. head ən-otʰa:p

11. mouth ən-imun

12. forehead ən-ečʰemug

13. lips ən-imun

14. neck ən-ɨnʈʰug

15. neck bone ən-olawana

16. back of neck ən-amataŋ

17. hand ən-ipil

18. finger ən-ɔm

19. nail ən-opetaŋ

20. palm ən-opagna

21. arm ən-oŋgəjag

88
 89

22. armpit ən-i:pɨ:lə

23. shoulder ən-alaŋ

24. stomach ən-ajkʈʰə

25. navel ən-ehʷaɭ

26. waist ən-epoja

27. chest ən-akotaŋ

28. leg ən-ipo

29. thigh ən-ɲanapo

30. knee ən-oɭaj

31. sole ən-uhʷaŋnapo

32. tooth ən-ahoɖ

33. back ən-inon

34. backbone ən-etʰotaŋ

35. blood čeŋ

36. bone ən-ogjag

37. breast ən-aka:g

38. cheek ən-oʈahiŋbo

39. chin ən-ipitaŋ / ən-ipita͂

40. nerve hʷetʰo

41. skin ən-itəjaŋ

89
90

42. toe ən-uhʷetaŋ

43. tongue ən-aɖal

(ii) Kin nouns

Kin nouns are a smaller set; they denote relationships that exist in that

community. Kin nouns are bound roots which, when combined with the definite

possessive prefix, express kinship terms. For example, m-awela ‘my wife’s

brother’ is never articulated without one of the definite possessive prefixes {m-},

{ŋ-} or {wa-}. Table 3.5 presents some kin nouns found in Jarawa. In this table, I

have omitted the possessive prefix.

Table 3.5: Jarawa nouns used for kinship terms

Kin noun Jarawa word

1. father -ummə

2. mother -kaja

3. husband -a:gi

4. wife -əŋa:p

5. elder sister -a:mi

6. elder brother -a:pə

7. son / daughter -ajə

8. younger brother / sister -aikʰwaʈə

90
 91

Noun Class II

In Jarawa the majority of nouns exist as free roots. As a sample, in the following

sections I present a brief list of nouns related to flora and fauna, nature, humans,

cultural artefacts, place names, and so on.

(i) Flora and fauna

It would be an impossibly large task to present every word related to flora and

fauna in this section. The Andaman Islands (particularly the Jarawa Reserve

Area) contain certain animals such as wild boar, deer, wild goats, and elephants.

Elephants were introduced by the British. Wild boars are part of the food habits

of the Jarawa people whereas deer and wild goats are not. There are insects,

birds, and reptiles found in the jungle of Andaman. Some nouns related to these

groups are listed in table 3.6.

Table 3.6: Names of fauna

Animal name Jarawa word

1. snake topo

2. rat uᵼg

3. pig hʷəwə

4. piglet oloŋa

5. elephant ʈʰehu:ʈʰu6

6 The word ʈʰehu:ʈʰu ‘elephant’ is actually a compound in which ʈʰe seems to be copula hu:ʈʰu

means ‘fat or big’

91
92

6. crow wara

7. butterfly pe:ʈpe:l

8. crocodile əru:gəǰjə

9. deer ka:ʈʰopijalə

10. snake (cobra) čəla:ŋ

11. dog wᵼmə

12. mosquito hʷi:na, ɸi:na

Andaman has a dense forest with many species of plants and trees. Jarawa
people have a strong body of knowledge of the flora of their area. In table 3.7, I
list some representative items which are useful in the life of the Jarawas.

Table 3.7: Names of flora

Plant name Jarawa word

1. jackfruit a:p

2. a kind of jungle/wild tuber čə:w

3. banana čonel

4. coconut ɖa:g

5. a kind of jungle/wild tuber wa:h

6. a kind of jungle fruit əmmi:n

92
 93

(ii) Nature

Nouns which describe nature and the cosmos are few in Jarawa. These words are

free roots. A sample list of such words is in table 3.8.

Table 3.8: Jarawa words related to nature and the cosmos

English word Jarawa word

1. sun jehejə

2. moon tape

3. star čiləpe

4. earth pela

5. air bi:jiŋ

6. sky paŋnaŋ

7. rain wəwə

(iii) Cultural artefacts

This section deals with items which are keys to the daily activities of the Jarawa

people. These include different types of baskets, bows, arrows, ornaments, and

other objects. Table 3.9 presents some of these nouns.

93
94

Table 3.9: Names of cultural artefacts

English gloss Jarawa word

1. bow a:w

2. arrow pa:ʈʰo

3. wooden basket for honey collection u:hu

4. knife to:wa:ɖ

5. rope/ pado

bow string/

knitting fibre

6. fishing net patʰo

7. net basket potʰo

8. basket made out of cane ta:ɽka

9. cooking vessel bu:čʰu

[Link] Gender

Jarawa lacks grammatical gender; thus there is no inflection for gender, nor is

there any gender agreement within the Noun Phrase. However, the language has

different lexical items to indicate the (natural) sex of an object. The examples in

(3.24) demonstrate the occurrence of nouns in different genders. In (3.24), the

words əŋka:w ‘old man’ and əŋka:ja:b ‘old woman’ as well as ta:hwə ‘bachelor’ and

ta:hwəja:b ‘spinster’ indicate that the form {-ja:b} represents feminine gender.

94
 95

However, the use of ja:b was found in only these two pairs of words, preventing

any definitive statements on the matter of gender in Jarawa.

(3.24) man /əŋa:gi/ old man /əŋka:w/

woman /əŋa:b/ old woman /əŋka:ja:b/

boy /tawə/ bachelor /ta:hwə/

girl /do:jə/ spinster /ta:hwəja:b/

[Link] Number

Countable nouns in Jarawa are classified as singular and plural where singular

denotes a single entity and plural indicates more than two entities. To express

singular number, Jarawa makes use of zero marking while to denote plural, the

language employs the suffix {-le}. The examples in (3.25) and (3.26) illustrate

number marking in Jarawa.

Noun (stem) + suffix Plural Noun

(3.25) paʈʰo + le paʈʰole

arrow PL arrows

(3.26) do:jə + le do:jəle

girl PL girls

However, the plural marking for plural nouns is not obligatory. Thus, example

(3.25) can be reproduced without {-le}. Hence both the examples (3.27) and

(3.28) are correct (see also § [Link]).

95
96

(3.27) naja paʈʰo-le

two arrow-PL

two arrows

(3.28) naja paʈʰo

two arrow

two arrows

[Link] Case marking

There is no overt syntactic case marking in Jarawa to differentiate between nouns

in their different grammatical roles. Generally word order serves this purpose.

Nonetheless Jarawa expresses the semantic case of the noun by means of

postpositions, as shown in example (3.29). A detailed discussion of case marking

is presented in section 3.7.

(3.29) tenmej olleg ɖə bəᵼʈʰe

Tenmey Olleg LOC go

Tenmey is going to Olleg.

3.3.3 Possession

Possession in Jarawa is expressed by prefixation of pronouns to the possessed

noun. The possessive pronouns used in the language are similar to the personal

pronouns used in other types of constructions. The possessive forms of

pronominals are listed in table 3.10. The table 3.10 differs from the table 3.3 in

listing the phonologically conditioned variants (“mi” and “ma” forms), and the

96
 97

shortened forms (possessive prefixes) together (usage patterns of different forms

of pronominals discussed in detail in § 2.5).

Table 3.10: Possessive forms

Person Possessive Forms

1st Person mi~ma~m

2nd Person ŋi~ŋa~na~ŋ~n

3rd Person wa~w

[Link] Inalienable possession

Inalienable nouns constitute those nouns which cannot be separated from the

humans who possess them. These generally include the names of body parts.

However, kinship terms are also included in inalienable nouns in Jarawa.

Inalienable nouns exist as bound roots in this language and are prefixed with

possessive pronouns as illustrated in example (3.30).

(3.30) m-oɖə

1POSS-hair

my/our hair

Here oɖə stands for ‘hair’, but ‘hair’ cannot be expressed independently in this

language if it refers to the hair of humans; it must be prefixed with a possessive

prefix. Similarly, when kinship terms are to be expressed, they also have to be

prefixed with possessive pronouns. However, there is a difference in the

prefixation of possessive prefixes to the names of body parts and kinship terms.

Kinship terms are (a) human specific, and (b) individual specific. That is, the

97
98

relationship between two or more individuals in the community is specific to

those individuals and/or groups. Hence, they always refer to a definite

individual/group in relation to another definite individual/group. Thus, when a

kinship term is expressed in this language, it takes a definite possessive pronoun,

as in example (3.31).

(3.31) m-awela

1 POSS-brother-in-law

My brother-in-law (wife’s brother)

Kinship terms are also formed by apposition. These kinds of kinship terms are

used for address, not for reference terms. Thus, ‘father of someone’ will be

translated as ‘someone father’. The examples (3.32–3.33) exemplify the use of

juxtaposed NPs for address terms. Note that the use of these address terms as

reference term attracts the obligatory prefixation of possessive prefix as shown in

(3.32a–3.33b).

(3.32) čajwaj ummə Cayway’s father (address term)

Cayway father

(3.32a) čajwaj wa-ummə Cayway’s father

Cayway 3POSS-father

(3.33) enme kaja Enme’s mother (address term)

Enme mother

(3.33b) enme wa-kaja Enme’s mother

Enme 3POSS-mother

98
 99

In contrast, body parts may belong to a specific individual, an indefinite

individual, or humans in general. In Jarawa, ən- is used as indefinite possessive

prefix, which encodes indefinite generic interpretation related to humans. Thus,

body parts take both kinds of personal possessive pronoun prefixes: definite

when referring to a definite possessor, and generic when referring to humans in

general. This is illustrated in example (3.34).

(3.34) ən-oɖə

POSS[HUMAN]-hair

hair

Thus, ‘hair’ may be ən-oɖə for the hair of humans in general as in (3.34), or m-oɖə

‘my hair’ as given in (3.30).

[Link] Alienable

Nouns other than body parts and kinship terms are alienable nouns in Jarawa.

Possession of an alienable noun is expressed by prefixing a possessive pronoun,

as shown in (3.35–3.37).

(3.35) mi čaɖɖa

1POSS hut

My hut.

(3.36) ŋi čaɖɖa

2POSS hut

Your hut.

99
100

(3.37) wa čaɖɖa

3POSS hut

His/her hut.

If there are two nouns where one is possessor and the other is possessed, then the

structure is possessor + 3POSS + possessed as given in (3.38).

(3.38) innen wa čaɖɖa

Non-Jarawa 3POSS hut

Non-Jarawa’s hut

3.3.4 Noun Phrase

This section presents a detailed examination of the constituents of the Noun

Phrase which may include demonstratives, numerals, adjectives, and other kinds

of nominal modifiers.

[Link] Basic Noun Phrase

The noun phrase in Jarawa consists of a noun and optional modifiers like

genitives, numerals, descriptive adjectives, and adverbs. Numerals and genitives

precede the noun while descriptive adjectives follow the nouns. Adverbs follow

the adjectives. The schema of a noun phrase is presented in (3.39) and illustrated

in (3.40).

(3.39) [Demonstrative] [Possessor] (Numeral) {(POSS)-(Noun)} (Adj (Adv)]

100
 101

(3.40) mi naja paʈʰo-le uʈʰu ʈʰuhumə

1POSS two arrow-PL big very

My two very big arrows.

In this construction, mi functions as possessor which is also used as a possessive

marker, so {mi} serves for both possessor and possessive prefix. But if the

possessor and the possessed are both nominals, the possessive prefix will get

attached to the possessed noun and the order of noun phrase will be as

illustrated in (3.41).

(3.41) innen naja wa-paʈʰo-le huʈʰu ʈʰuhumə

Non-Jarawa two 3POSS-arrow-PL big very

Non-Jarawa’s two big arrows.

[Link] Demonstratives

Jarawa does not have any article, though the morpheme ɖi sometimes seems to

refer to definiteness; however it seems to function as a morpheme which

demonstrates grammatical relations as well (for detailed discussion see § 4.9).

As mentioned earlier in § [Link], li indicates proximity, while luwə denotes

distance in this language.

(3.42) li paʈʰo

this arrow

This arrow.

101
102

(3.43) luwə paʈʰo-le

that arrow-PL

Those arrows.

[Link] Numerals and number marking

Jarawas have a counting system up to three, though recently they have coined a

word for four. More than three or four suggests ‘many’. After some contact with

officials of the Indian government and local residents, they have started showing

their fingers to denote counting, and seem to be in the process of including the

Hindi counting system in their communication with outsiders. The counting

system of Jarawa is given in table 3.11.

Table 3.11: Numerals

English Gloss Jarawa

1 one waja

2 two naja

3 three kaŋitəǰile

4 four ənə itəǰile (malawə)

5 five mala ʈʰuhumə

Jarawa does not use the strategy of invoking numerals to count objects in

syntactic structure, unless asked to specify the number of objects. In such cases

numerals precede the nouns, unlike adjectives, which follow the nouns. Nouns

are classified as singular and plural (see § [Link] for discussion of number), with

optional plural marking. There are two nouns, do:jə-le ‘girls’ and tawə-le ‘boys’

102
 103

which are generally used in plural reference. This may be because girls and boys

live in groups rather than in isolation, and are thus considered a plural entity

rather than individuals. However, when a woman gives birth to a girl child, the

child is referred to as do:j əcəle ‘girl child’.

[Link] Adjectives and Modifiers

In Jarawa, there are adjectives, possessive pronouns/prefixes, numerals, and

quantifiers which function as nominal modifiers. Adjectives follow the noun

while possessive prefixes (genitives) and numerals precede the nouns. It is

illustrated in the examples (3.44–3.47).

(3.44) paʈʰo huʈʰu

arrow big

A big arrow.

(3.45) mi-paʈʰo

1POSS-arrow

My arrow.

(3.46) innen wa-čaɖɖa

Non-Jarawa 3POSS-hut

A non-Jarawa’s arrow.

103
104

(3.47) naya paʈʰo

two arrow

Two arrows.

3.4 Verbal morphology

This section discusses verbs and verb phrases. Section 3.4.1 focuses on the verb

and its root forms. Section 3.4.2 deals with inflection and affixation related to

verbs. Causative constructions are described in 3.4.7.

3.4.1 Verbs and verb roots

Like nouns, verbal roots can also be classified as free or bound. Free roots exist

independently while bound roots have an obligatory prefix. Verbs used

exclusively for human actions are generally bound roots. For example, ən-ətəhə

‘sit’ contains the prefix {ən-}, and the root ətəhə which has been recorded as m-

ətəhə ‘I sit’, he-təhə ‘s/he sits’, and ən-ətəhə ‘you sit’. Thus ənətəhə may mean (a)

‘you sit’, or (b) ‘sit[HUMAN]’. For non-humans, the verb for ‘sit’ is oho. The {ən-}

prefix is not obligatorily for verbs used for both humans as well as non-humans.

For instance, ahapela ‘run’ is used for both and therefore does not have the {ən-}

prefix.

3.4.2 Verbal affixes

In Jarawa, verbs take one prefix and one suffix. Generally, in the prefix category

are pronominals (PRO), referential/definiteness markers (REF) or co-referencing

element (COREF), while as suffix are evidentiality marker (EVD) modality ((MOOD)

104
 105

and negative marker (NEG). The structure of a verb with its affixes is presented in

(3.48).

(3.48) (PRO) (REF) (COREF) + Verb + (EVD) (MOOD) (NEG)

Only one prefix and one suffix may be attached to a verb in a clause. For

example, if a clause has both a referential/definiteness marker and co-refencial

marker, the verb may take only the co-referential marker as a prefix, and the

referential/definiteness marker occurs as a separate word, as illustrated in (3.49).

(3.49) ma čonel ɖi h-ija

1ACC banana REF 3-give

Give me the banana.

Alternatively, the referential marker {ɖi-} may attach to the verb, in which case

the co-referential marker gets deleted, as (3.50) demonstrates.

(3.50) ma čonel ɖ-ija

1ACC banana REF-give

Give me the banana.

Similarly in the case of suffixes, one of the suffixes denoting evidential marker,

modality marking or negative marking can attached to a verb. (3.51) and (3.52)

demonstrate the use of suffixes {-hə} and {-ma} and (3.53) and (3.54) show that

both suffixes cannot attach simultaneously.

105
106

(3.51) mi allema-hə

1 come-HYP

I will come.

(3.52) mi allema-ma

1 come-NEG

I did not come.

(3.53) * mi allema-hə-ma

1 come-HYP-NEG

I will not come.

(3.54) * mi allema-ma-hə

1 come-NEG-HYP

I will not come.

3.4.3 Agreement

Jarawa verbs do not exhibit agreement with nominals in the clause. Sometimes,

there is a co-referring element prefixed to the verb, but I refrain from calling it

agreement because this co-referring element also has other functions. The

examples (3.55–3.57) demonstrate that there is no agreement morphology in this

language.

(3.55) mi čonel t-ita

1 banana REF-eat

I eat banana.

106
 107

(3.56) ŋi čonel t-ita

2 banana REF-eat

You eat bananas.

(3.57) hi čonel t-ita

3 banana REF-eat

S/he eats.

3.4.4 TAM

Verbs in Jarawa do not carry tense and aspect information morphologically.

However, it is possible to indicate time periphrastically through temporal

adverbs. It seems that the Jarawas’ perception of time does not extend beyond

three days. If an action has happened four days earlier, the temporal adverb

palahe ‘long before’ is used. Verbs take some suffixes which indicate modality

and evidentiality (for detailed discussion see § [Link]). The examples (3.58–3.60)

illustrate that there is no morphological marking of tense and aspect in Jarawa.

(3.58) mi t-ita

1 REF-eat

I eat / am eating.

(3.59) kiʈaje mi čonel t-ita

yesterday 1 banana REF-eat

Yesterday, I ate banana.

107
108

(3.60) kahiunen mi čonel t-ita

tomorrow 1 banana REF-eat

Tomorrow I will eat.

3.4.5 Function of suffixes: evidentiality and modality

Modality in Jarawa is marked as a verbal inflection. The morphemes indicating

mood are verb-sensitive: a morpheme can attach to a specific verb or group of

verbs. The kinds of modality marked in Jarawa are assertive and hypothetical.

Jarawa expresses evidentiality morphologically. Evidentiality in Jarawa falls into

two types: verifiable and non-verifiable. In verifiable cases, the speaker presents

the event as truth, and the hearer can verify the truth of the statement by

examining. In non-verifiable cases, the hearer does not have this facility. The

Jarawas distinguish two types of evidentials in the non-verifiable category: one

where the event is recent, so its effect of the event is still present; and the other

where the event is not in the recent past, so its effect may have been lost.

[Link] Evidential Marker

Jarawa employs evidential markers. This language distinguishes verifiable

evidential marker and non-verifiable evidential marker. Verifiable evidential

markers imply that the event can be verified by hearers and non-verifiable

evidential markers signify that event cannot be verified. The suffix {-jə} ~ {-wa} is

the verifiable evidentiality marker. The suffixes {-ʈʰe} and {-b} are non-verifiable

evidentiality marker.

108
 109

I. Verifiable Evidential Marker

a. {-jə}

The presence of this suffix indicates that the statement made by the speaker can

be verified by the hearer or other participants of the communicative event. This

is used when the action/event is still in progress, or the object referred to in the

event is still available. The examples below include context to make the usage

clear.

Context for (3.61): Some Jarawas are constructing a hut. Someone comes in and asks

what they are doing. The reply would be (3.61).

(3.61) mi čaɖɖa-le-jə

1 hut-VS-VEVD

I/we am/are making the hut.

Context for (3.62): Someone is climbing a tree and someone else asks what he is doing.

The reply would be (3.62).

(3.62) li taŋg ɖə čapʈʰe-jə

PDEM tree LOC climb-VEVD

He is climbing the tree.

From examples (3.61) and (3.62), a possible inference is that {-jə} is marking

present time and progressive. But this hypothesis requires further investigation

here; there are other examples which indicate something else. Consider for

instance, the example in (3.63).

109
110

Context for (3.63): One person asks another when he will go. If the listener intends to go

on the same day, the reply would be (3.60).

(3.63) olla mi bəᵼʈʰe-jə

today 1 go-VEVD

I/we will go today.

From example (3.63) one might interpret the morpheme {-jə} as a marker of

future tense, or present tense with progressive aspect indicating future time. The

first option (future tense marker) is less likely because in (3.61) and (3.62) the

same morpheme is used for present tense with progressive aspect. As it happens,

in many languages of the world present tense with progressive aspect does

indicate future time. However, example (3.64) below contradicts these two

hypotheses.

Context for (3.64): Someone wanting to know the whereabouts of Cayway’s mother

might be told the following.

(3.64) čajwaja kaja lə:w ɖ-eŋge-jə

name mother honey REF-bring-VEVD

Cayway’s mother has gone to bring honey.

The example (3.64) demonstrates that the cayway’s mother has left for jungle and

will bring honey, but she has not come yet. From this interpretation one might

opt for a third option, that this morpheme indicates incomplete action. The

example (3.65) shows a different usage.

110
 111

Context for (3.65): An official who was expected is standing on the side of the road with

some bananas. A Jarawa man sees the bananas and calls out to his wife as below.

(3.65) majoəba allema babu čonel ɖ-eŋge-jə

name come officer banana REF-bring-VEVD

Mayoəba, come, the officer has brought the bananas.

The example in (3.65) demonstrates perfective aspect and thus refutes all the

three hypotheses. If we reconsider examples (3.61–3.65), we will find that except

in (3.64), wherever {-jə} is used, it represents visual evidence. Before we make a

final statement, let us look at some more examples.

Context for (3.66): The researcher reaches the place where some Jarawas are staying and

asks where the others have gone. The reply is as below.

(3.66) əŋ ɖi kᵼjə ɖə bəᵼʈʰe-jə

Jarawa REF forest LOC go-VEVD

Jarawa have gone to the forest.

This example does not rely on visual evidence because the others have already

left for the forest. This is an example of perfective aspect, and is therefore in

opposition to (3.61) and (3.62) which encoded progressive aspect. Before we

reach at a conclusive result, we may consider the example given in (3.67).

Context for (3.67): There is a drum inside the jungle. One of the Jarawas who is to pass it

on to a non-Jarawa brought to the sea-shore in a dinghy. Officially, he is not allowed to

give the drum to a non-Jarawa, and may, as happens sometimes, say he will do so but

111
112

later not comply. As he leaves for the place where the drum is, he is asked whether he will

bring the drum and he replies as below.

(3.67) mi olo-olo ɖ-eŋe-jə

1 drum REF-bring-VEVD

I will bring the drum.

In (3.67) this morpheme seems to represent the truth of the action/event. A

comparison of all the examples suggests that {-jə} is not only used where visual

evidence is available, but also where the action, event, or the object in the clause

can be verified by other participants (in this case, the hearer) of the

communicative event.

b. {-wa}

The suffix {-wa} is also used as verifiable evidential marker. The occurrences of

the suffixes {-jə} and {-wa} are mutually exclusive (see Appendix III). It seems,

therefore, that are allomorphs. The examples in (3.68–3.69) represent the use of

the suffix {-wa}.

(3.68) mi inčo-wa

1 drink-VEVD

I drank water.

(3.69) tenmej ta:ŋ ɖi ičilo-wa

name tree REF cut-VEVD

Tenmey is cutting the tree.

112
 113

The conditioning environment for occurrences of the allomorphs {-jə} and {-wa}

seems to be phonological. {-jə} can occur with verbs which end with front vowels,

and {-wa} with those which end with a back vowel.

II. Non-Verifiable Marker

a. {-ʈʰe}

As we have seen in § [Link].1, Jarawa uses a suffix that indicates that the action

which is happening, (has) happened, or will happen, can be verified by the

participants of the communicative events. The suffixes {-ʈʰe} is used to denote

that the action cannot be verified. The example (3.70) represents the use of this

morpheme.

(3.70) gəcʰəhə mi inčo-ʈʰe

before now 1 drink-NVEVD

I drank water earlier (before now).

[Link] Non-Evidential Marker

{-b}

The morpheme {-b} has two variants: {-b} and {-bə}. The meaning of this

morpheme is not clear. It appears only with ita ‘eat’, as in (3.71). In the present

work this morpheme is glossed as NEVD.

113
114

(3.71) mi hʷəwə t-ita-b

1 wild boar REF-eat-NEVD

I ate (the meat of) wild boar.

The corpus suggests that it also attaches to the adjective ɖomo ‘fine’ as given in

(3.72).

(3.72) mi ɖomo-bə mi ŋ-aʈiba

1 good-NEVD 1 2ACC-tell

When I am cured, I will tell you.

Since the data available is limited to these examples, further investigation is

required to arrive at its precise meaning.

[Link] Modality Marking

I. Assertive Mood

{-jag}

The assertive/emphatic marker {-jag} has two variant forms {-jag} and {-g}. The

conditioning environment for the occurrences of these two variants is

phonological. If a verb ends with the vowel /a/, the {-g} suffix will be attached to

it. {-jag} is used in all other cases. It generally indicates emphatic mood or shows

a strong assertion. For example, in sentences (3.73–3.75) it means that the speaker

emphasizes or asserts the action of giving and going.

114
 115

(3.73) mi ih-ija-g

1 3-give-ASS

I really gave that.

(3.74) hi bəᵼʈʰe-jag

3 go-ASS

He has gone.

(3.75) mi pᵼpə h-ija-g ʈorči ɖ-ija

1 money 3-give-ASS torch REF-give

I gave you money, give me the torch.

II. Hypothetical Mood

The hypothetical mood marker is generally used for actions which are doubtful

or uncertain if they can occur. The suffix {-hə} expresses hypothetical mood.

{-hə}

The suffix {-hə} indicates the hypothetical mood. It incorporates those

actions/events which are supposed to occur in future, but there is no certainty

whether it will take place. It also includes doubtful events/actions.

Context for (3.76): A group of Jarawas is to move to another place. At the time of moving,

one of the Jarawas asks the researcher whether he would come to the place they are

supposed to move, as shown in (3.75).

115
116

(3.76) ka ŋi allema-hə

INTM 2 come-HYP

Will you come?

The sentence (3.76) has two interpretations: (1) the speaker is uncertain whether

the addressee will come to that place, or (2) the speaker was certain that the

addressee will not come to that place.

Context for (3.77): Two Jarawas are together. A government official comes and informs

them that Talu’s father has gone to Baratang. The Jarawas gossip among themselves and

then utter (3.79) in surprise. Talu’s father had told them that he would go to Baratang,

but they were certain that he would not.

(3.77) talu-əba aʈiba mi baraʈaŋ ɖə bəᵼʈʰe-hə

name-father say 1 [Link] LOC go-HYP

Talu’s father said that he would go to Baratang (the

addressees were uncertain about his going).

3.4.6 Function of prefixes

There are three types of prefixes generally attached to verbs. One is the

pronominal, which, when attached to a verb as prefix, has two different

functions: (a) subject of intransitive verb or object of transitive verb, or (b) co-

referencing element. Only third person pronouns can be used as co-referencing

elements. These two different functions of pronominals are described here as two

different prefixes.

116
 117

Another element which can be prefixed to a verb is referential/definiteness

marker. Usually part of a noun phrase (S or O argument of a clause), it can attach

to a verb as a prefix when there is no intervening element between the

referentiality marker and the verb.

[Link] Pronominals

Pronominals in Jarawa are more likely to be prefixed to verbs that start with a

vowel. Subject pronominals attach to an intransitive verb while object

pronominals attach to a transitive verb. If the verb is ditransitive, neither subject

nor indirect object pronominals attach to the verb. The nature and distribution of

these pronominals have been discussed in detail in § [Link]. Therefore, in this

section we will concentrate only on the use of the definiteness/referential marker

and co-referencing marker.

[Link] Referential/definiteness marker {ɖi-}

It seems that there are different functions of {ɖi-}. Sometimes it denotes

referentiality and sometimes definiteness. Though referentiality and definiteness

are similar concepts, I discuss separately, to distinguish between references made

to the visual (within sight of speaker-hearer), and non-visual (indefinite object;

its reference is derived from the preceding context). The former uses the

referential marker and the latter the definiteness marker. For instance, in (3.78),

iŋ ‘water’ refers to water within sight of the speaker-hearer.

117
118

(3.78) ma iŋ ɖ-ija

1ACC water REF-give

Give me that water. (water is seen)

In (3.78), this morpheme is employed with the noun pᵼpə ‘money’. In this

example, the speaker refers to money which he has just given, that may not have

a visual reference; the reference for this noun is drawn from the previous context

(for detailed discussion, see § 4.9.2).

(3.78) mi pᵼpə ɖ-ija ʈorči h-ija

1 money REF-give torch light 3-give

I gave you money, give me a torch light.

[Link] Co-referencing element {hi-}

The third person pronominal is employed to co-refer with a third person object

nominal of the clause. This pronominal is prefixed to the verb. It has different

allomorphs, {hi-}, {ih-}, {ihi-}, {əhi-}, and {h-}. It can also be used to refer to

indefinite objects. In the sentence (3.79), {ih-} is used as 3rd person pronoun which

refers to an object. In (3.80), {h-} is employed to co-refer with the noun iŋ ‘water’.

iŋ is here indefinite because it is not within sight. In (3.81), {h-} co-refers with the

nominal conel ‘banana’ which has moved to second position. In (3.82), there is no

{h-}, because {ɖi-} has been prefixed to the verb.

118
 119

(3.79) mi ih-ita

1 3-eat

I ate that (food).

(3.80) ma iŋ-e h-ija

1ACC water-EM 3-give

Give me water (water is not seen).

(3.81) jugdʰa čonel ɖi ma h-iya

PN banana REF 1ACC 3-give

Yugdha gave me the banana.

(3.82) jugdʰa ma čonel ɖ-ija

PN 1ACC banana REF-give

Yugdha gave the banana to me.

[Link] Conflict between {ɖi-} and {hi-}

We have observed that one of the functions of {ɖi-} is to indicate definiteness, and

that {hi-} co-refers with the O argument of the clause as well as indefiniteness;

and they usually prefix to the verbs. Both the morphemes {ɖi-} and {hi-} cannot

prefix to the verb together. {ɖi} can stand alone while {h-} has to be prefixed to the

verb obligatorily. As a syntactic rule, {ɖi-} will be prefixed to the verb when there

is no intervening element between it and the verb. In such cases, {h-} is dropped.

In sentence (3.83) we have both {ɖi-} and {h-} but {ɖi} is not closer to the verb. In

119
120

sentence (3.84) {ɖi} is closer to the verb but stands alone. In (3.85) {ɖi-} is prefixed

to the verb and {h-} is dropped. The sentences (3.84) and (3.85) demonstrate the

conflict between {ɖi} and {h-}, and the resolution to that conflict: {h-} is dropped.

(3.83) ʈeʈe paʈʰo ɖi ma h-ija

PN arrow REF 1ACC 3-give

Tete gave me the arrow.

(3.84) ʈeʈe ma paʈʰo ɖi h-ija

PN 1ACC arrow REF 3-give

Tete gave the arrow to me.

(3.85) ʈeʈe ma paʈʰo ɖ-ija

PN 1ACC arrow REF-give

Tete gave the arrow to me.

3.4.7 Causatives

Jarawa uses two kinds of strategies to form causatives: morphological and

syntactic. Morphological causatives are formed by prefixing the causative

morpheme {wa-} to the verbs, while syntactic causatives are formed by using the

verb of cause {woǰoʈʰo} with the main verb.

[Link] Morphological causatives

Morphological causatives in Jarawa are part of the verbal morphology. The

causative morpheme {wa-} is prefixed to a verb to form the causative

120
 121

construction. For example, in (3.86), tʰulə ‘sleep’ is an intransitive verb which

takes {wa-} as a prefix, adding the meaning ‘cause’, to become ‘cause to sleep’.

(3.86) əččəle wa-tʰulə

child CAUS-sleep

Make the child sleep.

In such constructions, the causer is treated as the agent (A) argument of the verb;

and the causee as the patient or O argument. This is evident from the use of the

definiteness/referential marker {ɖi} which is used with the S or O argument of the

clause. The sentences (3.87–3.88) illustrate this.

(3.87) wa-mummə əččəle ɖi wa-tʰulə

3POSS-father child REF CAUS-sleep

His/her father made the child sleep.

(3.88) wa-kaja əččəle ɖi wa-le-jə

3POSS-mother child REF CAUS-eat-VEVD

His/her mother is making the child eat.

In (3.88) the verb {le} ‘cause to eat’ seems to be a suppletive morpheme of ita ‘eat’

which occurs only in causative constructions. In (3.88), əččəle ‘child’ is treated as

the patient (O argument) of the verb le. The direct object can also be used

constructions where the lexical verb is already transitive. In such condition, the

causee which is treated as the O argument of the verb will be coded as IO, and

121
122

the second argument becomes DO. For instance, in (3.89) čonel ‘banana’ is

introduced as the patient or O argument of the verb le, making əččəle ‘child’ an

IO. The construction in (3.89) is identical to that found with ‘give’.

(3.89) wa-kaja əččəle ɖi čonel ɖi wa-le-jə

3POSS-mother child REF banana REF CAUS-eat-VEVD

His/her mother is feeding the banana to the child.

Our corpus has only two verbs at present which represent morphological

causatives, but its productivity cannot be denied with other verbs in larger

context. Since there is another option to express causation, i.e. the syntactic

causative, the corpus lacks more examples for morphological causatives.

[Link] Syntactic causatives

Syntactic causatives are formed using the independent word for causation along

with the lexical verb. In the present study it is, however, difficult to establish

whether the independent word for causation is a causative verb. Hence in this

work the independent word is glossed as causative (CAUS). The

modality/evidentiality suffix is attached to the main verb rather than the

causative (verb).

In causative constructions, if the lexical verb is intransitive, the causer acts as A

argument of the clause and the causee becomes the O argument. According to

general syntactic rules, the O argument will take definiteness/referential marker

{ɖi}. Thus, the schema of syntactic causative constructions of intransitive verbs

122
 123

would be as given in (3.90). In (3.91), the lexical verb inɖə7 ‘take bath’ has been

causativised. Thus, the causer wami ‘his/her elder sister’ becomes the A argument

of the clause and hiwa ‘3ACC’ becomes the O argument of the clause.

(3.90) [causer]NP [cause]NP [cause verb]VP [lexical verb]VP

(3.91) w-a:mi hiwa woǰoʈʰo inɖə

3POSS-elder sister 3ACC CAUS bathe

His/her elder sister bathed him.

If there are peripheral arguments, like adjuncts, in such causative clauses, they

precede the lexical verb. Thus, the schema for causative constructions of

intransitive verbs with peripheral arguments would be as given in (3.92) and

illustrated by (3.93).

(3.92) [causer]NP [cause]NP [cause verb]VP [adjunct]PP [verb]VP

(3.93) wa-kaja əččəle ɖi woǰoʈʰo taŋ ɖə čapʈʰe-jə

3POSS-mother child REF CAUS tree LOC climb-VEVD

His/her mother is making the child climb up the tree.

Similarly, if the lexical verb is transitive, the O argument remains the O

argument of causative clause, but the A argument of the transitive clause

becomes IO of the causative clause. The O argument precedes the lexical verb

7 The verb inɖə ‘take bath’is an intransitive verb in this language.

123
124

while IO precedes the cause verb. In general, the schema of causative

construction of transitive verbs is presented in (3.94) and illustrated in (3.95).

(3.94) [causer]NP [causee]NP [cause verb]VP [object]NP [verb]VP]]]]

(3.95) wa-kaja əččəle ɖi woǰoʈʰo iŋ-e ɖi inčo-wa

3POSS-mother child REF CAUS water-EM REF drink-VEVD

His mother made the child drink water.

This kind of pattern for causative constructions indicates that Jarawa makes use

of two clauses to form causative constructions. This is presented in (3.96).

(3.96) [causer]NP [cause]NP [Verb[CAUSE]]VP]] [arguments]NP [adjuncts]PP Verb]VP]]

3.5 Adjectives and Adjective Phrases

Jarawa has descriptive adjectives or what can be called pure adjectives. It has

adjectives (a) related to colour terms, (b) which describe/define a noun, and (c)

which qualify a noun. Numerals and determiners/demonstratives are not

discussed in this section, they are talked about in detail in other sections (for

detailed discussion of numerals, see § [Link]; for determiners, see § [Link]).

Jarawa distinguishes between descriptive and other types of adjectives like

numerals, demonstrative adjectives. Syntactically, the word order of descriptive

adjectives is different from that of other types. Though mentioned briefly in the

section on noun phrase, it would be better to go into detail about these adjectives

below.

124
 125

3.5.1 Descriptive adjectives

Jarawa uses descriptive or pure adjectives attributively and predicatively. In both

cases the adjective follows the noun. When it is used attributively, it simply

modifies the immediate noun, but when it used predicatively, it is used as the

predicate of a verbless clause. Attributive use follows the pattern in (3.97). This is

illustrated by example (3.98.)

(3.97) Noun + Adjective

(3.98) paʈʰo huʈʰu

arrow big

big arrow

3.5.2 Demonstrative adjectives

Jarawa lacks the use of copula between demonstrative pronouns and nominals.

This causes demonstrative adjectives and demonstrative pronouns to appear the

same. (3.99) illustrates this.

(3.99) li paʈʰo-le

PDEM arrow-PL

These are arrows.

The sentence (3.99) can also mean only the phrase ‘these arrows’. It is difficult to

syntactically assess whether (3.99) is a clause or a phrase. If this phrase is

employed as a subject or an object of a clause, then it can be interpreted as a

demonstrative adjective. For instance, in example (3.100), luwə paʈʰo-le ‘those

125
126

arrows’ is an example of a demonstrative adjective as this entire noun phrase is

used as an object of the clause.

(3.100) luwə paʈʰo-le ɖ-eŋge

DEM-A arrow-PL REF-bring

Bring those arrows.

3.5.3 Degree of adjectives

Jarawa differentiates between degrees of comparison. The language displays

positive, comparative, and superlative degrees. The degree of comparison is not

expressed morphologically. Rather, syntactic strategies are employed to express

degree of comparison. Positive degree is formed by using noun and adjective,

illustrated by (3.101).

(3.101) tenmej ʈʰi čew

PN COP good

Tenmey is good.

In the comparative degree involving comparison of two nouns, the noun being

compared precedes, without any additional morpheme (it has zero case

marking). The standard of comparison follows, and appears with a morpheme

{wɛ} which is, in the present study, referred to as the comparative morpheme

(COMPM). This morpheme may also be referred to as a postposition, as it is used

after the standard of comparison, not with the adjective. In this work {wɛ} is

taken as a comparative morpheme because its use is limited to comparison in the

126
 127

present corpus, and there is no other comparison-denoting element in such

constructions.

(3.102) ugə momo wɛ ʈʰi učəhə

Ugə Momo COMPM COP short

Ugə is shorter than Momo.

The superlative degree is formed by combining the adjective with an adverb, or

by using a new word which is not morphologically similar to the positive form of

a descriptive adjective. In (3.103), there is an adverb ʈʰuhumə ‘very’ which is used

to indicate superlative degree.

(3.103) kopel-wa-a:w ʈʰi huʈʰu ʈʰuhumə

Kopel-POSS-bow COP big very

Kopel’s bow is the biggest one.

In example (3.104), ʈomono ‘shortest’, is morphologically different from the

positive form, ucəhə ‘short’.

(3.104) m-a:w ʈʰi ʈomono

1POSS-bow COP shortest

My bow is the shortest one.

In comparative and superlative degrees, the copula ʈʰi is used before the adjective

as linking verb which links the noun with the adjective.

127
128

3.5.4 Adjective Phrases

Unlike numerals and the possessive prefix, the Adjective Phrase follows the

nouns. They may contain adverbs (intensifiers).

3.6 Adverbs

Jarawa has adverbs as a distinct word class. There is no derivational process to

convert adjectives into adverbs in this language. The adverbs which exist in this

language are purely lexical items. In this work adverbs will be classified as

intensifiers, temporal adverbs, and spatial adverbs.

3.6.1 Intensifier

These are adverbs which modify adjectives or verbs. The number of adverbs in

this group is very small; only two or three such adverbs are attested. One of them

is ʈʰuhumə ‘very’. The adverb ʈʰuhumə is widely used to modify descriptive

adjectives and verbs. For instance, in example (3.105), ʈʰuhumə modifies an

adjective and in example (3.106) it modifies a verb. In both cases it follows the

element which it modifies.

(3.105) paʈʰo talu ʈʰuhumə

arrow long very

A very long arrow.

128
 129

(3.106) ahapela ʈʰuhumə

run very

Run (very) fast.

3.6.2 Temporal adverbs

Temporal adverbs are those adverbs which denote time, like today, tomorrow,

morning. Jarawa divides time as today, a day before today, a day after today. The

day is divided into morning before sunrise, morning after sunrise, noon,

evening, and night.

3.6.3 Spatial adverbs

Jarawa divides immediate space into two different categories: proximate and

distant. Some examples of spatial adverbs include ikʰə ‘here (more proximate)’,

lijə ‘here’, luwə ‘there’, buʈʰə ‘near’, and no:ɖa ‘far away’. The usage of two other

words, lihəle and luwhəle is not clear. The adverbs lijə ‘here’ and luwə ‘there’ take

suffix {-kʰə} and ikʰə ‘here’ takes {-gə}. Though the meaning of these suffixes is not

clear, these are glossed as PL ‘place marker’ in the present study.

3.7 Postposition

In the present study it is difficult to ascertain whether Jarawa has postpositions.

Only two morphemes are attested which could be termed as postpositions. These

are {ɖə} and {ɲačʰe}. {ɖə} is glossed as locative (LOC) and {ɲačʰe} is glossed as

comitative (COM). {ɖə} is a multifunctional morpheme. It is employed to indicate

place of location, source of location, goal of location, and sometime it is also used

129
130

with temporal and spatial adverbs. {ɲačʰe} marks the comitative. Note that use of

{ɖə} as a postposition appears to be vague. Sometimes this morpheme was found

as a separate word (particularly with longer words polysyllabic words) and

sometimes seemed to exist as a suffix (especially with monosyllabic and

disyllabic words). As for {ɲačʰe}, it is evident that this morpheme is used as a

postposition.

3.7.1 Use of {ɖə}

The morpheme {ɖə} has multiple functions in this language. It used with

nominals which indicate location of actions/events, and with temporal and

spatial adverbs. In location of actions/events it specifies place of location, source

of location and goal of location.

[Link] Place of location

The morpheme {ɖə} is used with nominals to express the place of location of the

action/event. In (3.107), {ɖə} has been employed with čaɖɖa ‘hut’. The

combination gives the place of location for the verb action čalahe ‘stay’.

(3.107) mi čaɖɖa ɖə čalahe-jə

1 hut LOC stay-VEVD

I will stay at the hut.

130
 131

[Link] Source of location

This morpheme is also used with nominals to indicate the source of location of

the action/events. For instance, in sentence (3.106) pᵼpə ‘leaf’ is falling from the

tree. In (3.108) ‘tree’ is the source of location of the action ‘falling’. Hence taŋ

‘tree’ takes {ɖə}.

(3.108) taŋ ɖə pɨpə huwaǰi-jə

tree LOC leaf fall-VEVD

The leaf is falling from the tree.

[Link] Goal of location

This morpheme is also employed with nominals to express the goal of location of

the action/event. For instance, in sentence (3.109) Olleg is the name of a place

which is the goal of the verb bəᵼʈʰe ‘go’. Hence it takes {ɖə}.

(3.109) uʈʰa olleg ɖə bəɨʈʰe

Utha Olleg LOC go

Utha is going to Lakralunda/Olleg.

[Link] Temporal adverbs

When temporal adverbs (like English morning, evening, yesterday), denote points

of time, they take the morpheme {ɖə} as shown in (3.110). If used for periods of

time, they do not take {ɖə}; see example (3.111).

131
132

(3.110) kahijunen ɖə mi napo t-ita-b

morning LOC 1 fish eat-NVEVD

In the morning, I ate the fish.

(3.111) kiʈaje mi olleg ɖə bəᵼʈʰe

yesterday 1 [Link] LOC go

Yesterday, I went to Olleg.

[Link] Spatial adverbs

When a speaker uses a spatial adverb to refer to an object which has an

attachment with a place, the adverb takes the morpheme {ɖə}. It can be illustrated

from the example (3.112).

(3.112) luwa ɖə mi-paʈʰo

there LOC 1POSS-arrow

There is my arrow.

The summary of the use of {ɖə} is presented in table 3.12.

Table 3.12: Functions of {ɖə} as postposition with reference to semantic roles

Descriptive Names Semantic roles

place of location locative

source of location ablative

132
 133

goal of location destination

temporal adverbs point in time

spatial adverbs place

3.7.2 Use of {ɲacʰe}

Another word that is used as postposition is {ɲačʰe}. This postposition is

employed when a person accompanies another person in an action/event.

Sentence (3.113) demonstrates the use of {ɲačʰe}. Note that in (3.113) the second

person pronoun which is the S argument of the clause is dropped. (3.114) is the

response for the question asked in (3.113). I have used the term comitative (COM)

for this postposition.

(3.113) onne ɖaŋ (ɖi-əŋ) ɲačʰe allema

who REF-Jarawa COM come

Which Jarawa did you come with?

(3.114) mi topo-ɖ ɲačʰe allema

1 Topo-REF COM come

I came with Topo.

133
134

3.8 Copula {ʈʰi}

The morpheme {ʈʰi} has various usages in this language. It is used for identifying

an entity, or to qualify a noun. This morpheme also comes with the negative

particle, if the negative particle is used for short responses. It is also used with

spatial adverbs in short responses. Sometimes this morpheme appears with verbs

when the verb is used in adverbial relation to the main verb. Based on the

different usages of {ʈʰi}, it can be termed a copula in this language.

Unlike the verb which follows the object, this copula precedes the nominal

predicates or adjectival predicates. Jarawa is an SOV language. The expected

word order is noun, then nominal predicate or adjectival predicate, followed by

copula. However, the copula is used between the subject and the nominal

predicate or adjectival predicate. This copula may be inflected for

modality/evidentiality, but we have few examples which demonstrate this.

Followings are the usage of {ʈʰi}.

Usage of {ʈʰi}

1. Identity: It is used to indicate the identity of a noun. For instance, in

example (3.115) Tenmey is a proper noun (the name of a person), əŋ means the

Jarawa community and {ʈʰi} connects both the nouns.

(3.115) tenmej ʈʰi əŋ

Tenmey COP Jarawa

Tenmey is Jarawa.

134
 135

However, if there is a pronoun instead of the noun, no copula will be used, as

shown in example (3.116).

(3.116) mi əŋ

1 Jarawa

I/we am/are Jarawa.

In (3.117) below, {ʈʰi} is employed with the interrogative pronoun onne

‘who/which’. Here it seems that there are two clauses [onne ʈʰi əŋ]i and [øi lə:w ɖ-

ahole]. The first clause, ‘it is which Jarawa’, becomes the A argument of the

second clause ‘collected the honey’ (for detailed discussion see § 4.10).

(3.117) onne ʈʰi əŋ lə:w ɖ-ahole

who COP Jarawa honey REF-collect

It is which Jarawa who collected the honey?

2. Quality: It is also used with a descriptive but predicative adjective. For

instance, in example (3.118) uʈʰa is a proper noun, the name of a person, uʈʰu

means ‘fat’, and {ʈʰi} comes in between the noun and adjective.

(3.118) uʈʰa ʈʰi uʈʰu

name COP fat

Uʈʰa is fat.

However, if there is a pronoun instead of the noun, {ʈʰi} will not be used, as in

the example (3.119).

135
136

(3.119) hi uʈʰu

3 fat

He is fat.

3. Negative particle: It is also used with the negative particle naɖem ‘not’. {ʈʰi}

is employed with negative particle when the negative particle is used for a short

response of a sentence or clause. This is illustrated by the example (3.120a and

3.120b).

(3.120a) ka pramod ʈʰi əŋ

INTM Pramod COP Jarawa

Is Pramod a Jarawa?

(3.120b) ʈʰi naɖem

COP not

He is not.

In (3.120a) there is a question and we get the response as in (3.120b). However, it

is not essential to get the question like (3.120a) everytime, and hence the meaning

of the sentence can be derived from the context in which the conversation is

carried out.

4. Spatial adverbs: It is used with place adverbials like noɖa ‘far’ and buʈʰə

‘near’. Generally, it is employed in short responses to a question. (3.121a) and

(3.121b) illustrate the use of {ʈʰi} with place adverbs.

136
 137

(3.121a) ka olleg ʈʰi noɖa

INTM Olleg COP far

Is Olleg far away?

(3.121b) ʈʰi noɖa

COP far

It is far away.

5. Sometimes it is also used with other verbs, especially when there are two

verbs, and one verb is used as an adverb. The verb which is used as an adverb

will take {ʈʰi}, as in the example (3.122)

(3.122) mi ʈʰi čawaja allema

1 COP walk come

I came walking.

6. Sometimes when the S argument of the intransitive verb is an adjective, {ʈʰi}

follows the adjective and precedes the verb (see 3.123 below).

(3.123) malawə ʈʰi bəᵼʈʰe

many COP go

Many (Jarawa) have gone.

137
138

3.9 Word formation

Jarawa uses four strategies to form new words. They are:

1. Compounding

2. Denominalization

3. Reduplication

4. Empty morpheme

3.9.1 Compounding

Jarawa uses compounding of two or more words to coin a new word. The

strategy to make compounds is to use two nouns to form a compound as given in

(3.124).

(3.124) ən-iɲapo-talaŋ = nostril

[+HUMAN]-nose-hole

Jarawa also uses compounding of a noun with an adjective to form new words as

given in (3.125).

(3.125) iŋ-tapo = fresh water (drinking water)

water-good

Similarly the compound Noun + Noun + Adjective is also attested, as shown in

(3.126).

(3.126) hʷa-iŋ-tapo = watercourse (stream with clear, gooddrinking water)

stream-water-good

138
 139

Table 3.13: Examples of Jarawa compound nouns

Jarawa English

ən-ikʰwag-talaŋ earlobe

[HUM-ear-hole]

ən-epo-hiʷugə eyeball

[HUM-eye-ball]

ən-epo-ʈʰugə eyebrow

[[Link]]

3.9.2 Denominalisation

Generally nouns can be changed to verbs by suffixation. The verbalising suffix

(VS) {-le} serves the function of changing nouns to verbs. This is shown in (3.127–

3.128) and table 3.14 presents more examples of denominalised verbs.

(3.127) ča:ɖɖa + le = ča:ɖɖale

hut + VS = to construct a hut

(3.128) uɖɖu + le = uɖɖule

cough + VS = to cough

139
140

Table 3.14: Examples of Jarawa denominalised verbs

Jarawa English

ʈʰuhəb-le to burn wood

fire-VS

wočewoče-le to put together

piling-VS

tʰui-le to swing

swing-VS

taŋg-le to ride bicycle

bicycle8-VS

ipo-le to remove the bark of a tree

bark-VS

3.9.3 Reduplication

Reduplication in Jarawa is the repetition of a syllable or a word. The repetition of

a syllable is shown in (3.128) and (3.129). The words eweewe ‘itching’ and

wočewoče ‘piling’ exhibit repetition of syllables ewe and woče respectively.

(3.128) eweewe ‘itching’

(3.129) wočewoče ‘piling’

8 This word is referred to bicycle or any type of vehicle; hence this verb refers to using any kind of
vehicle.

140
 141

However, an entire word can also be reduplicated. If a word like a verb (only

verbal category is present in my corpus) is reduplicated, it has the function of

adverbial modifier. The example in (3.129) demonstrates how the word cajwaj

‘walk’ is reduplicated.

(3.129) mi čawaja čawaja allema

1 walk walk come

I came walking.

3.9.4 Empty morpheme

An empty morpheme has been observed attached to nouns. The meaning of this

empty morpheme is not clear, hence it is classified as “empty”. It is generally

attached to personal names but also to other nouns. It is shown in the examples

(3.130–3.132).

(3.130) babu-i

official-EM

Official

(3.131) dippu-i

Dippu-EM

Dippu (Personal Name)

(3.132) iŋ-e

water-EM

Water

141
142

It has two allomorphs, /i/ and /e/. /i/ is attached to words which end with a high

vowel, and /e/ is attached elsewhere.

3.10 Conclusion

I have discussed the morphology of Jarawa in this chapter. This included noun

morphology, verbal morphology, and causatives. This chapter also dealt with

other word classes like adjectives, adverbs and postpositions, and discussed

copula constructions. In noun morphology it was observed that pronominal-

headed NPs behave differently from nominal-headed NPs.

Jarawa makes use of verifiable evidential, assertive mood, and hypothetical

mood. We have noticed a word order phenomenon in this language, where the

copula precedes the nominal predicates or adjectival predicates.

In the adjective class we found two types of adjectives: (a) descriptive adjectives,

and (b) demonstrative, numerals, and possessives. Descriptive adjectives follow

the noun whereas demonstrative, numerals, and possessives precede the noun.

Adverbs are classified into two groups: (a) intensifiers, and (b) participle forms of

verbs.

142
 143

CHAPTER 4: SYNTAX

4.0 Introduction

The preceding chapter was a detailed study of Jarawa morphology. The present

chapter attempts to describe the syntactic structure of Jarawa: syntactic strategies

(how the language conveys commands, requests; the formation of questions, the

use of Reported Speech, and so on) and clause structure. The chapter will

elaborate these structures and their variances.

The chapter is organized in the following manner. In section 4.1, I present a brief

introduction to Jarawa clause structure. Section 4.2 discusses the verbless clause

in which a nominal or an adjective may play the role of a predicate. In section 4.3

is a detailed discussion of the verbal clause with core argument. Section 4.4

discusses verbal clauses with peripheral arguments. Section 4.5 focuses on the

negative construction, and section 4.6 describes imperative structures (includes

commands, requests, and prohibition). Section 4.7 concentrates on how

interrogative constructions are formed: it further focuses on yes/no and wh-

question types. In section 4.8, complex sentences are discussed, and a close

examination of direct speech is presented. Section 4.9 attempts to depict the

function of definiteness/referential morpheme {ɖi} which seems to be used in the

ergative-absolutive system. Section 4.10 presents the conclusion of the chapter.

4.1 Clause Structure

Jarawa employs two types of clausal structures: verbless clauses where nominals

or adjectives function as head of the predicate; and verbal clauses where verbs

143
144

are the head of the predicate, with core arguments and/or peripheral arguments.

Both types of clauses have different morphological and syntactic structures.

Verbless clauses generally lack modality categories which are part of verbal

clauses.

Verbless clauses (nominal predicates and adjectival predicates) contain nouns or

adjectives as the head. Morphologically, adjectival predicates differ from

nominal predicates at least in terms of affixation, but cannot be classified as

verbal predicates for they combine with a marker {–g} which differentiates

adjectival predicates from nominal predicates. I have made an attempt to

investigate whether the morpheme {–g} is equivalent to the suffix {–g} which is

attached to verbal predicate. It has been observed that no other suffixes, which

can be attached to verbal predicates, can be attached to adjectival predicates. In

the case of nominal predicates there is no prefixation of S argument pronominals,

but in adjectival predicates such prefixation is possible in cases where adjectival

predicates begin with a vowel. Otherwise, adjectival predicates follow the

syntactic structure of nominal predicates.

The copula can be the head of a predicate; it takes modality suffix. Data in the

corpus, at this point, is limited to only one sentence. Thus it is difficult to

ascertain whether the copula should be considered as a predicate. Moreover, it

has another function in the syntactic structure of the language: it is sometimes

used with verbs used to indicate adverbial function (detailed discussion in § 3.8).

In this dissertation copula constructions are described under verbless clauses.

Clause structures are also defined in terms of the arguments. In verbal clauses,

the core arguments S (subject), A (agent), O (object) are defined in terms of the

144
 145

Subject of intransitive verb, Agent of transitive verb and Object of transitive verb

respectively.

4.2 Verbless clauses

The two types of verbless clauses — nominal predicates and adjectival predicates

— differ morphologically and syntactically. Nominal predicates do not bear any

morphological marking, whereas adjectival predicates can have morphological

marking.

4.2.1 Nouns as head of verbless clause

While verbal predicates take modality and evidential categories as suffixes, and

adjectival predicates may also take a suffix (for detailed discussion see § 3.5 and

§ [Link]), nominal predicates never take such suffixes. Also, unlike verbal and

adjectival predicates, pronominals are not prefixed to S argument of nominal

predicates.

Nominal predicates have two different kinds of structures depending on whether

the S argument is a noun or a pronominal. If a pronominal is the S argument of

the nominal predicates, the clause is formed by juxtaposition, i.e. the pronominal

followed by the nominal predicate. If a noun is the S argument, the copula ʈʰi is

employed in the clause.

[Link] Pronominal as S argument of nominal predicate

Nominal predicates can have a pronominal as the S argument of the clause.

When a pronominal is the S argument of the clause, the pronominals preceding

the nominal predicates are juxtaposed. The structure of a verbless clause in the

145
146

case of a pronominal being the S argument of a nominal predicate is given in

(4.1).

(4.1) [S]PRO [VL P]NP

It must be noted that the copula ʈʰi can never be used in such clauses.

Furthermore, the pronominal is never prefixed to the nominal predicate in such

constructions. For instance, in example (4.2), mi is the first person pronoun which

is the S argument of the nominal predicate əŋ, thus we have “mi əŋ” without

prefixing the pronoun to the predicate nominal. It seems that such distinction is

necessary to maintain since there is prefixation of the pronoun to the nouns in

possessive constructions (for detailed discussion see § [Link]).

(4.2) mi əŋ

1 Jarawa

I am/we are Jarawa.

[Link] Noun as S argument of nominal predicate

If the S argument of the nominal predicate is a nominal headed noun phrase, the

copula ʈʰi appears between the two nominals, i.e. the copula ʈʰi precedes the

nominal predicate. (4.3) represents the structure of a nominal headed noun

phrase as the S argument of a nominal predicate. For example, if we have a

proper name instead of mi in the sentence (4.2), we will get the sentence (4.4). The

examples in (4.5–4.6) support this claim. Note the absence of any kind of

suffixation on the nominal predicate in this type of structure.

(4.3) [S]NP [COP] [VLP]NP

146
 147

(4.4) tenmej ʈʰi əŋ

Tenmey COP Jarawa

Tenmey is Jarawa.

(4.5) uʈʰa ʈʰi tawə

Uʈʰa COP boy

Uʈʰa is a boy.

(4.6) noru ʈʰi əŋka:b

Noru COP old man

Noru is an old man.

4.2.2 Adjective as head of verbless clause

Another type of verbless clause uses adjectives as predicates of a clause. Like

nominal predicates, adjectival predicates also have two different structures: one

which has pronominal as the S argument of the adjectival predicate and another

where the noun is the S argument of the adjectival predicate. This type of

verbless clause also has one more different structure where the suffix {-g} is

attached to the adjectival predicate. Thus, we have three different kinds of

structures for adjectival predicates: pronominal as an S argument, noun as an S

argument, and the adjectival predicates with suffix. All the three types differ in

their structures.

147
148

[Link] Pronominal as S argument of adjectival predicate

Like nominal predicates, adjectival predicates also take pronominals as their S

arguments. Two different strategies are employed when pronominals are used as

the S arguments of adjectival predicates. The first is pronominals (full form) to be

be followed by adjectival predicates beginning with a consonant as given in (4.7)

and illustrated in (4.8).

(4.7) [S]PRO [VLP]ADJP

(4.8) mi čew

1 good

I am/we are fine.

If the adjectival predicate begins with a vowel, the pronominal is prefixed to it.

This is contrary to nominal predicates where the pronominal as S argument does

not get prefixed (see (4.1)). In (4.9) the pronominal mi ‘I’ is prefixed to the

adjective ipaɖa ‘injury’.

(4.9) m-ipaɖa

1-injury

I am/we are injured.

[Link] Noun as S argument of adjectival predicate

Another kind of construction, where a noun is the S argument of the adjectival

predicate, employs copula {ʈʰi} between the S argument and the adjectival

predicate, like the nominal predicate. (4.10) represents this structure and (4.11)

148
 149

illustrates it. It must be noted that when {ʈʰi} is used in the adjectival predicate

construction, there is no kind of suffixation to the adjectival predicates.

(4.10) [S]NP [COP] [VLP]ADJP

(4.11) uʈʰa ʈʰi uʈʰu

Uʈʰa COP fat

Uʈʰa is fat.

The copula {ʈʰi} may be prefixed to an adjective starting with a vowel. In (4.12),

ʈʰ-ero is the combination of ʈʰi ‘COP’ and ero ‘angry’.

(4.12) uʈʰa ʈʰ-ero = ʈʰi ero

Uʈʰa COP-angry

Uʈʰa is angry.

[Link] Adjectival predicate with suffix {-g}

The suffix {-g} appears in verbless clauses with adjectival predicate. This suffix

can be used with constructions having a pronominal as the S argument as well as

those having a noun as the S argument of the adjectival predicate. The structure

of such clauses is presented in (4.13). Examples (4.14–4.15) illustrate it.

(4.13) [S] [VLP-g]ADJP

149
150

(4.14) paʈʰo talu-gə

arrow big-EVDV

The arrow is big.

(4.15) ŋi telo-gə

2 thin-EVDV

You are thin.

The suffix {-g} has two allomorphs, {-g} and {-gə}. The meaning of {-g} suffix is not

very clear. The similar suffix {-g} is employed with verbs to indicate assertion (for

detailed discussion, see § [Link].2.b), but it is not clear whether it is the same

suffix with similar function or it is just a visual evidential marker. There are two

instances where they seem to be different. In example (4.16), {-gə} is attached to

adjective ero ‘angry’. But if we observe the paradigm of its use with ero, it can be

claimed that {-gə} is attached with only 2nd Person and 3rd Person pronominals

(4.164.17), not with 1st Person pronominals (4.18). In case of verbal predicates, {-

g} is attached to verbs with all three pronominals; this illustrated by examples

(4.19–4.21). In the present work the suffix {-g} is referred to as visual evidential

(EVDV) in case of an adjectival predicate.

(4.16) ən ero-gə

2 angry-EVDV

You are angry.

150
 151

(4.17) əh ero-gə

3 angry-EVDV

He is angry.

(4.18) m ero

1 angry

I am/we are angry.

(4.19) mi bəᵼʈʰe-jag

1 go-ASS

I/we went (emphatic).

(4.20) ŋi bəᵼʈʰe-jag

2 go-ASS

You went (emphatic).

(4.21) hi (li) bəᵼʈʰe-jag

3 go-ASS

S/he went (emphatic).

No semantic difference was observed between the copula construction of

adjectival predicate and the construction of adjectival predicates with the suffix.

What was encountered in the field was that they occur interchangeably in both

kinds of constructions. Hence the sentence (4.14) can be reproduced as (4.22).

151
152

(4.22) paʈʰo ʈʰi talu

arrow COP big

The arrow is big.

Further, it must be noted that when the {-g} suffix is used with adjectival

predicates, {ʈʰi} is not employed. Hence, if we employ both the copula {ʈʰi} and

the suffix {–g} in same clause like the sentence (4.22), it becomes ungrammatical.

(4.23) * paʈʰo ʈʰi talu-gə

arrow COP big-EVDV

The arrow is big.

4.3 Verbal Clauses – I: Core arguments

Apart from verbless clauses, Jarawa has verbal clauses. In a verbal clause, a verb

is the central part of predication. These verbal predicates generally contain core

arguments of the clause, essential express the activity/event, and peripheral

arguments which provide additional information about the activity/event.

As discussed in the section 4.1 as core arguments, a verbal clause can have

subjects (generally referred to S and A) and objects (referred to O). These S, A,

and O are defined as: S, the subject of intransitive verbs; A, the subject of

transitive verbs; and O, the object of transitive verbs. This distinction is very

significant, as these categories have different roles in the grammar of a language.

The category O can further be split into DO and IO. DO is the direct object of the

clause, also called the patient of the clause, which has the direct effect of the

152
 153

action. IO does not bear the direct effect of the action; it could be benefactor, goal

etc. The general schema of a clause in Jarawa is given below.

(4.24) [Subject9]NP {[IO]NP [DO]NP} Verb]]VP

Other than core arguments, there are peripheral arguments. These peripheral

arguments provide additional information. They are optional in the clause. These

include temporal and spatial adverbs and/or nouns. The schema given in (4.25)

represents the structure of a clause in Jarawa with peripheral arguments.

(4.25) [Temporal adverb/Nouns]ADVP [Subject]NP {[Spatial

Adverbs/Noun]NP} {[Object]NP} Verb]]VP

In addition to these broad categories are smaller categories like referential

marker (see § 4.9), locational marker (see § 3.7), and others. These categories

emerge in certain specific contexts in a clause.

4.3.1 Intransitive clauses

In intransitive clauses, Jarawa employs S arguments of intransitive verbs as core

arguments, and intransitive verbs as the head of predication. There are some

denominalised verbs which function as intransitive verbs. We may put these

denominalised verbs in two categories, one in which they form part of weather-

type constructions, and another in which they form part of transitive verbs

(semantically); however, syntactically, they behave as intransitive verbs of the

clause (for detailed discussion, see § 4.3.2). Generally, intransitive clauses in

Jarawa can be formed as given in (4.26).

9 Subject in all the schema refers to NPs which function as S and/or A of the clause.

153
154

(4.26) [Subject]NP {[REF10]} [VerbIN]VP

(4.27) is an example of an intransitive clause in Jarawa.

(4.27) mi bəᵼʈʰe

1 go

I am/we are going.

If the S argument of an intransitive verb is a noun with specific reference, then it

takes the morpheme ɖi. This ɖi specifies the S argument (NP) of the clause. It is

generally prefixed to the verb if the verb begins with a vowel and there is no

intervening element betweem ɖi and the verb. For instance, in sentence (4.28),

hʷəwə ‘wild boar’ is a common noun, but here if it refers to a specific wild boar

which is sleeping (and it may be seen), ɖi appears after the noun hʷəwə ‘wild

boar’.

(4.28) hʷəwə ɖi omohə

boar REF sleep

The wild boar is sleeping.

ɖi can be prefixed to the verb omohə ‘sleep’ because it starts with a vowel. See

eample (4.29).

10 REF is referred to the marker which is used for definiteness and referential interpretation. It
includes both, definite object and specific, and indefinite but specific object.

154
 155

(4.29) hʷəwə ɖ-omohə

boar REF-sleep

The wild boar is sleeping.

In case hʷəwə does not refer to a specific wild boar, it will not take ɖi, as given in

(4.30).

(4.30) hʷəwə omohə

boar sleep

A wild boar is sleeping.

If the S argument of an intransitive verb is a pronominal, and the verb start with

a vowel, the pronominal can be prefixed to the verb. For example, the

intransitive verb aʈiba ‘say’ has mi ‘I/we’ as the S argument of the intransitive

clause in (4.31–4.32); this clause can be articulated as mi aʈiba as given in (4.31) or

maʈiba ‘I/we said’ as given in (4.32).

(4.31) mi aʈiba

1 say

I/we said.

(4.32) m-aʈiba

1-say

I/we said.

155
156

4.3.2 Denominalised verbs

Denominalised verbs are derived from nouns by adding the verbaliser suffix (VS)

{-le} (for detailed discussion, see § 3.9.1). These denominalised verbs may fall into

two broad categories: weather-type verbs, and activity/event-type verbs. The

weather-type verbs are a limited set. In such clauses, there is no S argument and

denominalized verbs express the action/event. Example (4.33) illustrates

denominalised weather-type constructions.

(4.33) wᵼwə-le

rain-VS

It may rain.

In (4.33) wᵼwə ʹrain’ is a noun which is denominalised to a verb by adding

verbaliser suffix {-le}. Here wᵼwə functions as the head of the predicate as well as

the S argument of the predicate.

The other types of denominalised verbs, which are activity/event oriented verbs,

are also formed in a similar manner, but in such clauses the noun functions as the

head of the predicate as well as the O argument of the clause. For example, in

(4.34) čaɖɖa-le represents denominalised verb where čaɖɖa ‘hut’ is a noun. In this

construction, it can be said that čaɖɖa with the verbalizer suffix has become the

predicate, but čaɖɖa in itself becomes an O argument of the clause inherently.

156
 157

(4.34) mi čaɖɖa-le-jə

1 hut-VS-VEVD

I/we are constructing a hut.

Syntactically, these denominalised verbs behave like intransitive verbs. Hence,

the S argument of such clauses may take ɖi. Since the O argument of such clauses

itself becomes a part of the predicate, it does not take ɖi as it happens in the case

of transitive clauses. The sentence in (4.35) illustrates the use of ɖi in such clauses.

(4.35) əŋ ɖi čaɖɖa-le-jə

Jarawa REF hut-VS-VEVD

Jarawas are constructing a hut.

Unlike verbal predicates, in these types of clauses ɖi is not prefixed to the

denominalised verb even if it starts with a vowel. Thus, in sentence (4.36) ɖi is

not prefixed, and hence the prefixation of ɖi results in ungrammaticality.

(4.36) uʈʰa ɖi uɖɖu-le-jə

Uʈʰa REF cough-VS-VEVD

Uʈʰa is coughing.

(4.37) * uʈʰa ɖ-uɖɖu-le-jə

Uʈʰa REF-cough-VS-VEVD

Uʈʰa is coughing.

157
158

4.3.3 Transitive clauses

Transitive clauses are headed by transitive verbal predicates and have two core

arguments, A and O. In such clauses, A precedes the O, and O precedes the verb.

The structures of transitive clauses vary according to the object NP: it will have a

different structure if the object NP is headed by a noun, or headed by a

pronominal. The verbal predicates can bear different prefixes based on the object

NP. If the object NP is headed by a definite noun or indefinite but specific it will

take referential marker ɖi. If it refers to an indefinite noun, it will take prefix {h-}.

If a pronoun heads an object NP, there will be neither referential marker ɖi nor

prefix {h-} on the verb; rather, the pronoun may be prefixed to the verb. The

structure of a transitive clause is below (4.38) and illustrated by (4.39).

(4.38) [Subject]NP [Object]NP {[REF]}VerbTR]]VP

(4.39) mi hʷəwə t11-ita-b

1 boar REF-eat-NVEVD

I/we ate the wild boar.

If the object of a transitive clause is a pronominal and the transitive verb starts

with a vowel, the object pronominal is generally prefixed to the transitive verb

(for prefix forms see § [Link]). The example (4.40) demonstrates the prefixation of

pronominals with the transitive predicates.

11 The verb ita éat’ takes {t-} as prefix which seems to be the allomorph of ɖi. Since this prefix does
not appear with any other verb, I assume {t-} as the allomorph of ɖi. If we attach the pronominal
prefix, then it will be m-ita ‘I eat’. However there is no further evidence for the support of the
hypothesis that they are allomorph.

158
 159

(4.40) tenmej m-əjojəba

Tenmey 1-see

Tenmey saw me.

If the object of the transitive verb is a noun, it takes referential marker ɖi (for

detailed discussion see § [Link].4.4). It specifies the noun phrase which it follows.

In example (4.41) the nominal object čonel ‘banana’ takes ɖi which specifies some

particular bananas. This ɖi may come independently after the noun čonel

‘banana’ in a clause such as (4.41), or it may be prefixed to the verb eŋge ‘bring’

provided there is no intervening element between ɖi and the verb, as given in

(4.42). In both cases it specifies the noun it follows.

(4.41) mi čonel ɖi eŋge-jə

1 banana REF bring-VEVD

I/we have brought the banana.

(4.42) mi čonel ɖ-eŋge-jə

1 banana REF-bring-VEVD

I/we have brought the banana.

If the noun čonel ‘banana’ refers to some non-specific banana, the prefix {h-} is

attached to the verb, as shown in (4.43) (for detailed discussion, see § [Link].4).

159
160

(4.43) mi čonel h-eŋge-jə

1 banana 3-bring-VEVD

I/we have brought a banana.

4.3.4 Ditansitive clauses

In ditransitive clauses, a ditransitive verbal predicate heads the clause, and has

three core arguments, A, IO, and DO. The IO and DO are part of O, are divided

according to their functions and usage in the clause.

The order of A and ditransitive verbs is fixed. A precedes the O (IO and DO) and

the ditransitive verb follows the O, as in transitive clauses. However, the order of

the objects is not fixed. The IO can follow the DO or precede it. Nonetheless, the

position of ɖi is fixed. Since ɖi is employed to specify the Object NP, it moves

with the direct object. The clausal structure of a ditransitive construction is given

in (4.44). The example (4.45) demonstrates the ordering of A, O, and verb in

ditransitive clauses.

(4.44) [Subject]NP [Indirect Object]NP [Direct Object]NP {[REF]}] VerbDT]]]VP

(4.45) mi ŋa napo ɖ-ija

1 2ACC fish REF-give

I/we gave you the fish.

Since the position of IO and DO is not fixed in ditransitive clauses, we may get

the clausal structure of a ditransitive clause as in (4.45) below. (4.46) illustrates

the clausal structure of a ditransitive clause where the DO precedes the IO.

160
 161

(4.45) [Subject]NP [Direct Object]NP [REF] [Indirect Object]NP VerbDT]]]VP

(4.46) mi napo ɖi ŋa h-ija

1 fish REF 2 ACC 3-give

I/we gave you the fish.

Three properties are observed in cases when the indirect IO follows the DO: (a) ɖi

moves with the DO; in other words, ɖi comes just after the DO; (b) the verb may

take co-referencing prefix {h-} which refers to the DO (for detailed discussion, see

§ [Link].4); and (c) if a pronominal is the IO and happens to be adjacent to verb

beginning with a vowel, this pronominal never gets prefixed to the verb.

Prefixing the IO to the ditransitive verb results in ungrammaticality of the

sentence as shown in (4.47).

(4.47) * mi napo ɖi ŋ-ija

1 fish REF 2ACC-give

I/we gave you the fish.

If the indirect object is a lexical noun phrase, this noun phrase also takes ɖi, as

shown in (4.48).

(4.48) mi napo ɖi topo ɖi h-ija

1 fish REF Topo REF 3-give

I/we gave Topo the fish.

161
162

4.4 Verbal Clauses – II: Peripheral arguments

In addition to the core arguments of the verb, there may be peripheral arguments

in a clause. They are usually temporal adverbs, spatial adverbs, and nouns which

have spatial significance.

(4.49) mi olleg ɖə bəᵼʈʰe

1 Olleg LOC go

I/we will go to Lakralunda/Olleg.

In intransitive clauses, these peripheral categories have different positions. The

temporal adverbs come at the beginning of the clause and spatial adverbs occur

after the subject and before the verb. The sentence structure of the intransitive

clause with peripheral categories is as given in (4.50).

(4.50) {[Temporal adverbs/noun]NP {[LOC]}}PP [Subject]NP {[Spatial

Adverbs/noun]NP [LOC]]PP} [VerbIN]VP

It must be noted that here the locative marker is presented with curly brackets

when it comes with temporal adverb or noun. These curly brackets here

represent optionality. Examples are shown in (4.51) and (4.52).

(4.51) kekahəʈe ɖə mi bəᵼʈʰe

tomorrow LOC 1 go

I/we will go tomorrow.

162
 163

(4.52) kekahəʈe mi olleg ɖə bəᵼʈʰe

tomorrow 1 Olleg LOC go

Tomorrow I/we will go to Olleg/Lakralunda.

A clause with transitive verbs has temporal adverbs at the beginning of the

clause, and spatial adverbs follow the subject of the clause. The general schema

to arrange these arguments in a clause of Jarawa is given in (4.53) followed by

examples in (4.54) and (4.55).

(4.53) [Temporal adverbs/noun]NP {[LOC]]PP} [Subject]NP {[Spatial

adverb/noun]NP [LOC]]PP} [Object]NP VerbTR]]VP]]]]

(4.54) kiʈaje ɖə ŋi napo ɖi aikʰwa

yesterday LOC 2 fish REF kill

Yesterday, you killed the fish.

(4.55) kiʈaje ŋi olleg ɖə napo ɖi aikʰwa

yesterday 2 Olleg LOC fish REF kill

Yesterday you killed the fish at Olleg/Lakralunda.

4.5 Negative clauses

Negation in Jarawa can be expressed in two ways: either by using the

independent negative word naɖem or the suffix –ma. Both are used for clausal

negation. There is no semantic difference between the two, but the suffix –ma has

restricted use in the language while naɖem can be used across the board.

163
164

4.5.1 Negation with naɖem

The position of this negative element in a clause is after the predicate. To negate

a declarative clause, the structure in (4.56) is to be used.

(4.56) [subject]NP {[object]}NP Verb ]]VP [naɖem]NEG

(4.57) uʈʰa allema naɖem

Uʈʰa come NEG

Uʈʰa did not come.

(4.58) mi olleg ɖə bəᵼʈʰe naɖem

1 Olleg LOC go NEG

I/we will not go to Olleg/Lakralunda.

This is a general strategy to negate the clause whether it is a declarative, an

interrogative, or an imperative sentence. The examples (4.59) and (4.60) show

interrogative and imperative sentences, respectively.

(4.59) onne ɖi lə:w ɖ-eŋge-jə naɖem

who REF honey REF-bring-VEVD NEG

Who didn’t bring the honey?

(4.60) bəᵼʈʰe naɖem

go not

Do not go.

164
 165

naɖem is the only negative element which can be used in verbless clauses. The

structure of verbless clauses with naɖem is presented in (4.61). The examples

which follow demonstrate the use of naɖem in these environments. naɖem occurs

after the nominal or adjectival predicate.

(4.61) [Subject]NP [VLP]VLP [naɖem]]]NEGP

(4.62) mi innen naɖem

1 non-Jarawa NEG

I am/we are not non-Jarawa.

(4.63) tenmej ʈʰi do:je naɖem

Tenmey COP girl NEG

Tenmey is not a girl.

(4.64) m-ero naɖem

1-angry NEG

I am/we are not angry.

(4.65) tenmej ʈʰi ero naɖem

Tenmey COP angry NEG

Tenmey is not angry.

In sentence (4.65), the copula ʈʰi can alternatively be prefixed to the predicative

adjective ero ‘angry’.

165
166

(4.66) tenmej ʈʰ-ero naɖem

Tenmey COP-angry NEG

Tenmey is not angry.

4.5.2 Negation with suffix {-ma}

Another way to negate the sentence in Jarawa is to employ suffix {-ma} on the

verbal predicate. But in such constructions, the verbal predicate does not take

any other suffix (for instance, modality suffixes). Thus the {-ma} suffix replaces

any other suffixal markings. Sentence (4.67) demonstrates the use of the {-ma}

suffix for negation. In this sentence, allema ‘come’ does not have any other suffix.

(4.68) is ungrammatical because allema ‘come’ bears the suffix {-hə} which does

not allow the suffix {-ma} to negate the sentence. In such cases, one has to employ

naɖem to express negation.

(4.67) uʈʰa allema-ma

Uʈʰa come-NEG

Uʈʰa did not come.

(4.68) * uʈʰa allema-hə-ma

Uʈʰa come-HYP-NEG

Uʈʰa may not come.

166
 167

(4.69) uʈʰa allema-hə naɖem

Uʈʰa come-HYP NEG

Uʈʰa may not come.

4.5.3 Use of ɲag

For short responses or an answer to a yes/no question, Jarawa uses the negative

particle ɲag at the beginning of the negative response, followed by the negated

sentence. For instance, the negative response to the yes/no question given in

(4.70) would be (4.71).

(4.70) ka ŋa-kaja pečame

INTM 2POSS-mother die

Has your mother died?

(4.71) ɲag ma-kaja pečame naɖem

No 1POSS-mother die NEG

No, my/our mother has not died.

The suffix {-ma} can be used instead of naɖem.

(4.72) ɲag ma-kaja pečame-ma

No 1POSS-mother die-NEG

No, my/our mother has not died.

An affirmative response of (4.70) is presented in (4.73).

167
168

(4.73) je: ma-kaja pečame

yes 1POSS-mother die

Yes, my/our mother has died.

4.6 Imperative clauses

Jarawa does not use distinct expressions for requests and command, nor does the

language have any kind of morphological marking to show imperatives.

Syntactically, the predicate with its core argument expresses all types of

imperatives, whether request, command, advice, or order.

4.6.1 Simple imperative clauses

Simple imperative clauses in Jarawa use a verbal predicates with its core

argument; that is, (transitive) verb + object NP. Sometimes they make use of

vocative forms in such constructions, though there is no different morphological

form for vocatives; the name of the person referred to is used as it is. The schema

in (4.74) presents this structure, and is followed by examples.

(4.74) {[Vocative]NP} {[Object]NP} Verb]]VP]]]

(4.75) unnə

go back

Go back.

(4.76) iŋ-e ɖ-eŋge-jə

water-EM REF-give-VEVD

Give me/us water.

168
 169

(4.77) ma iŋ-e ɖ-ija

1ACC water-EM REF-give

Give me/us water.

OR

Give me/us my/our water.

The example in (4.78) below illustrates the use of the vocative where either a

person’s name or a form of address may be used. In these constructions, there is

no morphological marking for vocative.

(4.78) čajwaj ummə olo olo ɖ-ija

Cayway father (voc) plastic bottle REF-give

Cayway’s father, give me/us the plastic bottle.

OR

Cayway’s father, pass the plastic bottle.

4.6.2 Prohibitive clauses

Prohibitives are a kind of command, order, or advice to not to do something. In

Jarawa, these are formed by putting the negative element naɖem after the verb,

like it happens in simple negative sentences.

(4.79) ullel ɖə bəᵼʈʰe naɖem

sea LOC go NEG

Don’t go in the sea.

169
170

Morphological negation may also be used, i.e. the suffix {-ma}. (4.80)

demonstrates the use of negative suffix {-ma} in prohibitives.

(4.80) ullel ɖə bəᵼʈʰe-ma

sea LOC go-NEG

Don’t go in the sea.

4.7 Interrogative clauses

Interrogatives in Jarawa are of two types: ‘yes/no’ questions and wh-questions.

In first type, the interrogative marker ka is used at the beginning of the sentence,

while in the second type question words (interrogative pronouns) are used to

create the interrogative. However, there are some instances where the

interrogative marker is used even in wh-questions. All these structures are

described in the following sub-sections.

4.7.1 Yes/no questions

The schema of a yes/no question is presented in (4.81).

(4.81) ka + [subject]NP + {[object]NP} + Verb]]VP ?

In (4.81), object has been put within the curly bracket (to show optionality) since

intransitive verbs do not take an object. However, instead of an object some

peripheral arguments like a spatial NP may come between the subject NP and

the verb.

170
 171

(4.82) ka ŋa-kaja pečame?

INTM 2POSS-mother die

Did your mother die?

(4.83) ka ŋi čonel ɖi he-eŋge-jə?

INTM 2 banana REF 3-bring-VEVD

Will you bring the banana?

(4.84) ka ŋi olleg-ɖə bəᵼʈʰe-jə?

INTM 2 Olleg-LOC go-VEVD

Will you go to Olleg/Lakralunda?

Yes/no questions may also be formed out of verbless clauses by using the

interrogative marker ka as shown in (4.85), followed by some examples.

(4.85) [ka] [subject]NP [VL Predicate]VLP

(4.86) ka ŋi əŋ ?

INTM 2 Jarawa

Are you a Jarawa?

(4.87) ka uʈʰa ʈʰi əŋ?

INTM Uʈʰa COP Jarawa

Is Uʈʰa a Jarawa?

171
172

(4.88) ka əh-ero?

INTM 3-angry

Is he angry?

(4.89) ka tenmej ʈʰ-ero?

INTM PN COP-angry

Is Tenmey angry?

4.7.2 Wh-question

In wh-questions in Jarawa, some NPs in the clause are replaced corresponding

question words (interrogative pronouns/adverbs). The wh-words stay in situ, i.e.

the question word used for the subject NP would stay in canonical subject

position and the question word used for the object NP would stay in canonical

object position. Other question words (interrogative adverbs like when and

where) also appear in canonical positions. Interrogative sentences with wh-

words explored below.

1. onne ‘who’

The question word onne ‘who’ is used for humans. It can be employed for the S

argument of an intransitive verb, the A argument of a transitive verb, and an

indirect object (including benefactors).

(4.90) onne ɖi olleg-ɖə bəᵼʈʰe?

who REF Olleg-LOC go

Who went to Olleg/Lakaralunda?

172
 173

(4.91) onne ɖi hiwa čonel ɖi h-ija-g?

who REF 3ACC banana REF 3-give-ASS

Who gave the banana to him?

(4.92) hi onne ɖi aɪkȹwa?

3 who REF kill

Who did he kill?

(4.93) ŋi onne ɖi napo ɖi h-ija?

2 who REF fish REF 3-give

Who did you give the fish to?

2. onəhə ‘what’

onəhə ‘what’ is generally used for the object of the clause. This wh-word usually

refers to non-humans.

(4.94) hi ŋa onəhə ɖi h-ija-g?

3 2 ACC what REF 3-give-ASS

What did he give to you?

(4.95) ŋi titabə ɖi onəhə ɖi h-ija-g?

2 Titabə REF what REF 3-give-ASS

What did you give to Titabə?

173
174

3. onəhə-le ’why’

onəhə-le has two different meanings: one is ‘what is being done’ and another is

‘why’. It is composed of two morphemes, onəhə ‘what’ and the {-le} suffix. The

suffix {-le} is employed for two purposes in this language: either to indicate

plurality or to denote denominalisation. If we take the first meaning of onəhə-le, it

suggests that {-le} is a verbaliser suffix and it yields the meaning of ‘what is being

done’.

(4.96) ŋi onəhə-le?

2 what-VS

What are you doing?

The second meaning of onəhə-le is illustrated in the examples below.

(4.97) ŋi onəhə-le ɖ-ičʰe?

2 what-VS REF-make/do

Why did you do that?

(4.98) hi onəhə-le wanna?

3 what-VS weep

Why is s/he weeping?

4. nojčʰe ’how many/much’

The word nojcʰe is used for ‘how many’ and ‘how much’.

174
 175

(4.99) luwə-kʰə-ɖə ŋi nojčʰe omohə?

there-PM-LOC 2 how many sleep

How many days will you sleep there? (cf. How long will

you stay there?

There are some cases where the interrogative marker ka is employed in this type

of wh-question, as shown in (4.100). It was not ascertained why ka appears in

such cases. The ordering of the interrogative marker and other constituents of the

clause is as we have in other interrogative constructions, where we have

interrogative marker in the beginning of the clause followed by other

constituents.

(4.100) ka-ŋa paʈʰo-le nojčʰe?

INTM-2POSS arrow-PL how many

How many arrows do you have?

(4.101) ka-ŋi kahiunen onəhə ɖi h-ija-g?

INTM-2 morning what REF 3-give-PST

What did you give in the morning?

Some constructions involve the structure SUB + INTM + question word. These

constructions mostly contain the subject NP and the question word for ‘where’.

For example, in the sentence (4.102) there is a subject NP čajwaj and question

word ʈʰučəʈəgə, consequently we have ‘cajwaj ka ʈʰučəʈəgə’

175
176

(4.102) čajwaj ka ʈʰučəʈəgə

Cayway INTM where

Where is Cayway?

This sentence can also be formed without ka:

(4.103) čajwaj ʈʰučəʈəgə

Cayway where

Where is Cayway?

Putting the interrogative marker ka at the beginning of the sentence causes

ungrammaticality. For example:

(4.104) * ka čajwaj ʈʰučəʈəgə

INTM PN where

Where is Cayway?

4.8 Complex sentences

The present corpus shows evidence of coordination, and of direct and indirect

speech. There is no substantial evidence for complex sentences involving

subordination. Distinguishing between direct and indirect speech is difficult

since no complementiser or any particle is used in such constructions.

176
 177

4.8.1 Coordination

The conjunction ən ‘and’ is used to join two simple sentences. This conjunction is

also used when two nouns are joined. The examples below demonstrate the use

of the conjunction ən.

(4.105) tenmej hʷəwə t-ita ən bəᵼʈʰe

Tenmey boar REF-eat and go

Tenmey ate and went.

(4.106) tenmej ən uʈʰa bəᵼʈʰe

Tenmey and Uʈʰa go

Tenmey and went.

4.8.2 Complementation

In this section we will discuss two kinds of complement clauses: clause internal

complement clauses and clause external complement clauses. The first type

occurs within a clause, in the object position within the VP. Jarawa is an SOV

language and therefore any clause that fills the object position should precede

the verb. However, it is possible to have a complement clause which follows the

verb as well. This type of clause is clause external complement clauses. Direct or

indirect speech is included in the clause external complement clause.

[Link] Complement internal clauses

It is difficult to ascertain if Jarawa has clause internal complement clauses

because it is not evident whether these types of clauses are NPs or complement

177
178

clauses. I will make an attempt to uncover whether these types of constructions

are complement clauses or complement nominalisations. Examples like (4.107)

suggest both these possibilities.

(4.107) mi ujijə ɖ-ipale-jə ɖi bəᵼʈʰe-jə

1 fish REF-fish[V]-VEVD REF go-VEVD

I will go for fishing.

(4.108) [mii] [øi]A [ujijə]NP [ɖ-ipale-jə]]]VP]NP [ɖi] [bəᵼʈʰe-jə]]]]VP

Two verb phrases are employed here, a main verb phrase and a dependent verb

phrase. Semantically both the verbs carry their lexical meaning, and they follow

the syntactic rules of a simple clause.

Here there are two clauses: (1) [mi]NP [ujijə]NP [ɖ-ipale-jə]]VP ‘we will do fishing’

and (2) [mi]NP [ɖi [bəᵼʈʰe-jə]VP ‘ we will go’. In (1), the verb is ipale ‘fish[V]’ and in (2)

the verb is bəᵼʈʰe ‘go[V]’. ipale is a transitive verb; thus it takes NP ujiyə ‘a kind of

fish’ as an object NP. This results in the use of the referential marker ɖi with the

object ujiyə. mi ‘first person pronoun’ serves as a subject for both the clauses.

In example (4.109) below, itʰəgə ‘cut’ is a transitive verb, so it takes an object NP;

this object NP takes definiteness marker ɖi. This entire verb phrase can be

considered as a clause, although the example in (4.109) differs from (4.107). In

(4.107) both the verbs take the evidentiality suffix whereas in (4.109) only the

main verb takes the evidential suffix. However, the absence of the suffix in

(4.109) does not pose any problem to the analysis of the clause, instead the verb

178
 179

in (4.109) bəᵼʈʰe which is intransitive. In intransitive constructions, the subject NP

takes ɖi if headed by a noun; if headed by a pronoun, it does not take ɖi. ɖi

always follows the NP it refers to. Thus, the ɖi in (4.109) does not refer to the

subject NP; for the subject NP is headed by a pronoun, and it is too far away

from the subject NP (see § 4.9 for conditions for the use of ɖi). Here it means

refers to the dependent clause, or what we can call the clause internal

complement clause.

(4.109) taŋ ɖ-itʰəgə ɖ-əᵼʈʰe-jə (ɖi-bəᵼʈʰe-jə)

tree REF-cut REF-go-VEVD

I will go for cutting wood.

Two questions raised by this discussion are given in (4.110).

(4.110) (a) What is the reason for using ɖi with bəᵼʈʰe in this clause?

(b) bəᵼʈʰe is an intransitive verb, and hence it cannot take an object

NP or a complement clause.

In (4.107) there are clearly two clauses with two verbs, each verb takes evidential

marker. In (4.109) only the main verb takes evidential marker. They also seem to

be embeded within one another. If they are two clauses embeded within one

another, we need to answer the question raised in (4.110). As we will notice again

in section 4.9, ɖi is a marker of definiteness/referentiality and it also specifies

grammatical relations. This morpheme is generally used with nominals

functioning as either S argument or O argument. It is also used with the nominal

which is IO of the clause. Since this morpheme is definiteness marker, we cannot

179
180

rule out the use of this morpheme with the nominal in A argument position but

the corpus reveals a seldom use of ɖi with nominal in A argument position.

Based on these facts we may infer that ɖi is used for the clause [ujijə]NP [ɖ-ipale-

yə]]VP treating it as an NP. Is it a plausible answer? It is still open for further

research. The second problem poses a grave problem. Since bəᵼʈʰe is an

intransitve verb it cannot take an object NP. Dixon (2010) proposed three criteria

to identify a complement clause which is reproduced in (4.111).

(4.111)

(I) It has the internal structure of a clause, at least as far as core arguments

are concerned.

(II) A complement clause functions as a core argument of another clause. If

a complement clause is in O function, for instance, it should show at

least some of the syntactic properties of O in that language.

(III) A complement clause will always describe a proposition; this can be a

fact, an activity, or a state (it may not be a place or a time).

(Dixon 2009:129)

In Jarawa, clause internal clauses have the structure of a normal clause. They

denote a proposition. In the case of (4.107) it denotes an action. They also

demonstrate the syntactic properties of O (for detailed discussion see section 4.9),

it is, however, misleading to say that ujijə ɖ-ipale-jə functions as the O of bəᵼʈʰe,

since bəᵼʈʰe is an intransitive verb and it cannot take an O argument. I, still,

submit that such clauses are possibly made of two clauses: one is the main clause

and the other is the clause internal complement clause. This is based on the

180
 181

assumption that these clause internal complement clauses are peripheral

arguments of the clause in case of (4.107). It is still an open question whether they

are complement clauses or not.

[Link] Complement external clauses

Another kind of complement clause is the clause external complement clause.

These types of clauses are not internal to the verb phrase of the main clause;

rather they exist as a different (dependent) clause. Direct and indirect speech is

considered as a clause external complement clause.

It is difficult to distinguish direct speech from indirect speech; for there is no

marking in both types. There are some instances where the difference between

direct and indirect speech seems to be employed, but the distinction between the

two could not be established. As the deictic center for pronouns in the utterances

indicates that such constructions are instances of direct speech, hence we

categorize these types of constructions as direct speech.

In such constructions, there are two clauses: one which is called reporting speech

consisting of reporting verb and another which is referred to reported speech. In

the present study reporting speech is referred to as main clause while reported

speech is referred to as dependent clause. In Jarawa the reported speech follows

the reporting verb but there is no particle or marker to distinguish the two

clauses: main clause and dependent clause. Thus, the direct speech in Jarawa has

the structure as given in (4.112).

(4.112) [Reporting Verb]MCL [Reported Speech]DCL

181
182

The examples below illustrate the use of direct speech in declarative sentences,

interrogative sentences and imperative sentences.

a) Declarative sentences as reported speech

If the dependent clause is declarative, the syntactic structure remains the same.

The schema for this kind of construction is presented in (4.113) and illustrated in

(4.104-4.105).the examples that follow.

(4.113) [Reporting Verb]MCL [Reported Speech]DCL

(4.114) čew m-aheapa naru čonel igeǰi

Cew 1ACC-say PN banana cut

Cew told me that Naru would cut a banana.

(4.115) naru m-aheaba mi ləkəralunɖa-ɖə bəᵼʈʰe

PN 1ACC-say 1 Larkalunda-LOC go

Naru said to me, “I will go to Larkalunda.”

b) Interrogative sentences as reported speech

If the dependent clause is a yes/no question, its structure remains the same as it

woud normally be, even when used in clause external complement clause. The

schema is given in (4.116) and illustrated in (4.107-4.108).

(4.116) [Reporting Verb]MCL [Yes/No Question Clause]DCL

182
 183

(4.117) dipu m-aʈiba ka ŋi hʷatapo-ɖə bəᵼʈʰe

PN 1ACC-say INTM 2 Hwatapo-LOC go

Dipu asked me, “Are you going to Hʷatapo?”

(4.118) sapan hiwa aʈiba ka gəčʰehə ŋi napo ɖi i:tə:-hə

Swapan 3ACC ask INTM before now 2 fish REF eat-HYP

Swapan asked him, “Have you had fish today?”

Interrogative dependent clauses have the same syntactic structure as simple wh-

question clauses. Thus the schema of such clause external complement clause is

presented in (4.119), and (4.120) illustrates it.

(4.119) [Reporting Verb]MCL [wh-question clause]DCL

(4.120) dipu hiwa aʈiba onne hʷəwə ɖ-aɪkʰwa

Dipu 3ACC ask who boar REF-kill

Dipu asked him, “Who killed the wild boar?”

c) Imperative sentences as reported speech

Imperative dependent clauses behave as normal imperatives. The schema is

presented in (4.121) and illustrated in the following examples.

(4.121) [Reporting Verb]MCL [Imperative Clause]DCL

183
184

(4.122) dipu m-aʈiba unnə

Dipu 1-say go home

Dipu told me to go home.

(4.123) naru ŋ-aʈiba čonel t-ita

Naru 2-say banana REF-eat

Naru told you to eat a banana.

4.9 Function of {ɖi}

The uses of morpheme ɖi seem to be various. Generally, this morpheme appears

after a noun phrase to indicate definiteness. It can occur with a S(ubject) NP or

O(bject) NP but rarely occurs with A(gent) NP of a clause. This usage of ɖi

suggests that it is also employed to express grammatical relations in a clause.

This morpheme is also used with VPs which appear to be clause internal

complement clauses. It occurs after the constituents, it is refers to; hence, it

moves with the constituents in a clause if they move to other places in a clause.

However, there are certain conditions on the use of the morpheme ɖi. The

conditions to employ ɖi in a clause are as follows:

i) This morpheme never occurs with pronominals.

Pronominals never take ɖi to express definiteness/referentiality or grammatical

relation in a clause. mi ‘I’ in example (4.124) is attached to the verb but it can

occur independently in such clauses. In example (4.125) there are two

184
 185

pronominals, one as A and another as O, but neither bear the morpheme ɖi.

These examples suggest that pronominals never take ɖi.

(4.124) m-aʈiba

1-say

I say/said.

(4.125) mi ŋ- əjojəba

1 2-see

I saw you.

ii) This morpheme can occur with nominal headed NPs.

iii) This morpheme is generally employed with the S/O of a clause. This

morpheme may occur with an A, but this pattern is very rare in this

language.

In example (4.126), tenmej ‘name of a person’ is a noun and S of the verb allema

‘come’, so the noun tenmej takes ɖi. Similarly, in example (4.127) there are two

nominal headed NPs, innen ‘non-Jarawa’ as A, and əŋ ‘Jarawa’ as O of the verb

əjojəba ‘see’. In this clause the O takes ɖi, which is attached to the verb. However,

əŋ ‘Jarawa’ in example (4.128) takes ɖi. The example in (4.128) has two clauses.

The first is an intransitive clause in which the S takes ɖi, but in the second clause,

əŋ ‘Jarawa’ is the A of the verb aikʰwa ‘kill’ which takes ɖi.

185
186

(4.126) tenmej ɖi allema

Tenmey REF come

Tenmey has come.

(4.127) innen əŋ ɖ-əjojəba

non-Jarawa Jarawa REF-see

A non-Jarawa saw the Jarawa.

(4.128) hʷəwə ɖi-ahapela əŋ ɖi ihi-aikʰwa

wild boar REF-run Jarawa REF 3-kill

The wild boar was running and a Jarawa killed that.

iv) This morpheme may occur with interrogative pronoun onne ‘who’.

The resulting phrase (onne ɖi) has a different interpretation (for

detailed discussion, see § 4.10 on cleft/relativization).

(4.129) onne ɖi aʈiba

who REF speak/say

Who said that?

v) Post-posed O does not take ɖi, i.e. if the O comes after the verb, ɖi

does not appear.

186
 187

Jarawa is an SOV language but sometimes employs SVO pattern. If an O is

moved to final position, we do not get ɖi with O. As a result, in (4.130), the O

does not take ɖi. But if this O occupies the position before the verb in a clause

as given in (4.131), it takes ɖi. However, if the O is preposed, i.e. moved to the

place before A, ɖi is employed with O as given in (4.132).

( 4.130) babu ilijema tenmej

official [Link] Tenmey

The official did not see Tenmey.

(4.131) babu tenmej ɖ-ilijema

official Tenmey [Link]

The official did not see Tenmey.

(4.132) tenmej ɖi babu ilijema

Tenmey REF official [Link]

The official did not see Tenmey.

vi) Verb Phrases also take ɖi if the verb phrase constitutes the internal

part of the clause, i.e. when verb phrases are used as clause internal

complement clauses, the verb phrase takes ɖi.

vii) This morpheme is the part of the NP it refers to (see examples (4.131

and 4.132).

187
188

4.9.1 Grammatical relations

Different grammatical roles and the relations between them can be expressed by

various means at the syntactic level. Some languages employ case marking to

differentiate the different roles of the core arguments; some languages resort the

fixed word order for the same purpose. In the case of Jarawa, grammatical

relations between arguments are expressed through referential marking.

Jarawa does not demonstrate any kind of case marking on its core arguments S,

A and O. It does, however, use a referential marker, usually with S or O, rarely

with A. This rareness of its use with the A of a transitive verb suggests that this

referential marker ɖi is employed to indicate grammatical relations. These

grammatical relations entail that this language follows ergativity-absolutive

patterns; however, the context for the use of ergativity-absolutive is

referentiality/definiteness. The other strategy to differentiate S/O from A is

specific to pronominal arguments and deals with prefixation of S/O to the verb

which is not possible with A.

Let us look first at pronominals. Pronominals never take ɖi, but can be prefixed

to verbs if a verb begins with a vowel. It has been observed in section [Link] that

prefixation of pronominals to verbs is possible only for S or O pronominals, and

not A pronominals. For example, in (4.133) we have prefixation of the S

argument to the verb aʈiba ‘say’. Similarly in (4.133) the O argument of the verb

əjojəba ‘see’ is prefixed.

188
 189

(4.133) m-aʈiba

1-say

I say/said.

(4.134) mi ŋ- əjojəba

1 2ACC-see

I saw you.

The distinction between S/O and A is more apparent when nominals are used in

these positions. The morpheme ɖi is always used with the NPs in S or O position.

For instance, in examples (4.135–4.141) S takes ɖi since all the verbs in these

examples are intransitive verbs.

(4.135) tenmej ɖi allema

Tenmey REF come

Tenmey has come.

(4.136) wᵼwə ɖi allema

rain REF come

It will rain (lit. The rain will come).

(4.137) uʈʰa ɖi bəᵼʈʰe

Uʈʰa REF go

Uʈʰa will go.

189
190

(4.138) əŋ ɖi čalahe

Jarawa REF stay

Jarawa is staying.

(4.139) hʷəwə ɖi ahapela

wild boar REF run

The wild boar was running.

(4.140) tawale ɖi omohə

Tawale REF sleep

Tawale is sleeping.

(4.141) əŋ ɖi aʈiba

Jarawa REF say

Jarawa said.

Similarly, in examples (4.142–4.144) the morpheme ɖi is employed with the O of

transitive verbs. It is clear here that S arguments of intransitive verbs and O

arguments of transitive verbs are treated similarly, i.e. both S and O take

referential marker ɖi while A of transitive verbs does not take referential marker

ɖi.

(4.142) əŋ tenmej ɖi aʈiba

Jarawa Tenmey REF say

Jarawa told Tenmey.

190
 191

(4.143) tenmej əŋ ɖi aʈiba

Tenmey Jarawa REF say

Tenmey said to Jarawa.

(4.144) innen əŋ ɖ-əjojəba

non-Jarawa Jarawa REF-see

Non Jarawa saw Jarawa.

Based on the preceding discussion I conclude that Jarawa marks grammatical

relations through the use of the definiteness/referential marker ɖi. This language

employs a split system of ergative-absolutive to denote grammatical relations

and the context of the use of this ergative-absolutive is referentiality/definiteness.

4.9.2 Definiteness

The morpheme ɖi is also used to indicate definiteness. The sentence (4.145)

demonstrates that ɖi is employed with the noun iŋ ‘water’ when the water is

within sight of the speaker-hearer. Hence, iŋ ‘water’ in (4.145) refers to some

particular water or a specific container which has water. If the water is not seen,

the co-referencing prefix h- is used instead of ɖi as given in (4.146).

(4.145) ma iŋ ɖ-ija

1ACC water REF-give

Give me that water (water is seen).

191
192

(4.146) ma iŋ h-ija

1ACC water 3-give

Give me water (water is not seen).

Again, the sentence (4.147) shows that ɖi is employed with the noun pᵼpə ‘money’

which means ‘the money that I gave you’, but ʈorci ‘torch’ does not take ɖi, and

refers to any torch (the speaker cannot see the torch).

(4.147) mi pᵼpə ɖ-ija ʈorči h-ija

1 money REF-give torch 3-give

I gave you the money, give me a torch.

But the use of ɖi in sentence (4.148) raises a question about the criteria employed

to consider an NP to be definite. In this sentence, ɖəwaɖəwa ‘biscuits’ is neither

seen nor not does it refer to a definite packet of biscuits. It may be the case that

the reference here is to a specific kind of packets of biscuits which the researcher

(myself) generally used to bring during fieldwork.

(4.148) ɖəwaɖəwa ɖ-eŋge-jə

biscuits REF-bring-VEVD

Bring a packet of biscuits. (lit. Must bring a packet of biscuits).

4.10 Cleft sentences/relativisation

In this language, there is at present no clear cut evidence for relativisation. There

are, however, some constructions which suggest that relativisation may be

192
 193

employed in this language. These constructions are reminiscent of English cleft

constructions. The example (4.149) appears to be a simple interrogative clause

where there is an interrogative pronoun onne ‘who’ and a noun phrase əŋ

‘Jarawa’, connected with a copula ʈʰi. Literally it means ‘it is which Jarawa?’

(4.149) onne ʈʰi əŋ?

who COP Jarawa

Which Jarawa is there? (Who is there?)

In (4.149) above, the interpretation of the clause is ’one of the Jarawas, but not

certain which’. This interpretation suggests two things: (1) this clause denotes

some indefinite entity, and (2) it seems that it may be similar to ‘it is which

Jarawa who is there?’ Moreover, this entire clause could be the S or A argument

of a clause. For instance, in (4.150) onne ʈʰi əŋ functions as the A argument of the

verb odehe ‘cut’. The structure of (4.150) can be represented as (4.151).

(4.150) onne ʈʰi əŋ taŋ odehe

who COP Jarawa tree cut

It is which Jarawa who cut the tree?

(4.151) [onne ʈʰi əŋ]i [øi] [taŋ]NP [odehe]]] VP

In (4.151) there are two clauses. One clause is [onne ʈʰi əŋ] and the other clause is

[ø taŋ odehe]. In the second clause, [ø] refers to the first clause. The interpretation

of the entire clause would be “It is which Jarawa who cut the tree?”

193
194

There is another kind of construction in (4.152). In this example, ɖi is used

instead of ʈʰi. (4.150) offers an indefinite interpretation of the noun while (4.152)

presents a definite interpretation of the noun. In both the examples, we have a set

of Jarawas and in that set there is someone who performed the action. However,

(4.150) informs the hearer that the set in itself is indefinite. But (4.152) suggests

that there is a definite set of Jarawas and one of them performed the action. In

other words, they (Jarawas) are sure of the set or subset of Jarawas referred to.

Jarawas always live in a band and keep on moving from one place to another.

Suppose that a tree was cut when a particular band were staying at some

different place. This means that they are not aware of which Jarawa had cut the

tree. But if a tree was cut after they had reached the place, this means the tree

was cut by someone from the same band. The proposition ‘the tree was cut’

when they were staying away from the place refers to cutting done by an

indefinite set of Jarawas, whereas the proposition ‘the tree was cut after they had

reached the place’ refers to cutting done by a definite set of Jarawas. This is the

difference between the examples (4.150) and (4.152). It must be noted that these

two types of constructions onne ʈʰi and onne ɖi are mutually exclusive.

(4.152) onne ɖi əŋ taŋ odehe

who REF Jarawa tree cut

It is which Jarawa who cut the tree?

It must be noted that the language does not have entire clause all the time.

Sometimes there may be the part of the clause, i.e. only one phrase onne ɖi or

onne ʈʰi. These phrases have the same interpretation as we had in (4.150) and

194
 195

(4.152). For instance, in (4.153) the clause seems to be a simple interrogative

clause which has an interrogative pronoun onne ‘who’, the O lə:w ‘honey’ and a

verb eŋge ‘bring’. This clause has only one additional constituent ʈʰi ‘COP’. It was

unclear earlier why the interrogative pronoun onne ‘who’ in A position takes ʈʰi

or ɖi. Now it is clear that in these types of constructions the noun phrase has

been dropped but the interpretation of the clause is similar to that of (4.150) or

(4.152).

(4.153) onne ʈʰi lə:w ɖi-eŋge?

who COP honey REF-bring

It is which Jarawa who collected/brought honey?

The corpus contains a few examples of the type mentioned in (4.153). The

examples (4.154–4.156) exemplify the use of such constructions where noun

phrases have been dropped.

(4.154) onne ɖi hiwa čonel (ɖi) h-ija

who REF 3ACC banana (REF) 3-give

Who gave banana to him?

(4.155) onne ʈʰi hiwa wa-mummə?

who COP 3ACC 3POSS-father

Who is his father?

195
196

(4.156) onne ɖaŋ (ɖi-əŋ) ɲačʰe allema

who REF-Jarawa with come

It is which Jarawa who did you come with?

4.11 Conclusion

This chapter described the structures of verbless and verbal clause types.

Verbless clauses include structures which have two nouns in juxtaposition, or a

noun with an adjective. In verbal clauses, the focus was on clauses with verbs as

the head of predicates; they include intransitive verbs, transitive verbs,

ditransitive verbs and denominalised verbs. I also examined the various

structures forming simple clauses, such as interrogative sentences, imperative

sentences, negation, and so on. Complex sentences were also described, which

include coordinating structures and complementation. The language employs

clause internal complement clauses and clause external complement clauses.

Clause external complement clauses consist of direct speech. Relative clauses

were also discussed and the use of ʈʰi and ɖi with interrogative pronouns,

especially onne ‘who’, was clarified.

196
 197

Chapter 5: Jarawa and Typology

5.0 Introduction

The present study is a descriptive grammar of Jarawa. I would like to now

investigate the data in order to determine the position of Jarawa in a typological

framework. So far Chapter 2 described the phonetics and phonology of Jarawa;

Chapter 3 investigated the morphological properties of Jarawa; Chapter 4 was an

attempt to discuss the syntactic structures in the language. This chapter will

investigate which kinds of typological patterns this language demonstrates.

The chapter is organized as follows. Section 5.1 contains an overview of

typology, including morphological typology and word order typology

formulated by Greenberg, and the developments which followed (Hawkins

(1983), Dryer (1992, 2007)). Section 5.2 presents a general outline of the typology

of Jarawa. Section 5.3 gives the morphological typology of Jarawa. Section 5.4

aims at the word order typology of Jarawa with reference to Greenberg (1963). A

brief conclusion is presented in section 5.5.

5.1 Language typology

Typology literally means the study of types. In linguistics, typology means the

study and classification of languages according to their structural features.

However, typology differs from general classification and taxonomy. According

to Marradi (1990) classification in general means the attributes are considered

one at a time, and entails that the classification is built on a single attribute at

each level. Classification means division of attributes at one level; at another

197
198

level another attribute will be considered, to split the group further. In

taxonomy, the attributes are considered in succession, while in typology multiple

attributes are taken together to group languages.

The term “typology” was introduced in linguistics in the 19th century by the

Schlegel brothers, as an attempt to classify the languages of world based on a

morphological criterion, i.e. how morphemes combine into words. Based on this

morphological criterion, there were three types of languages: isolating,

agglutinating, and fusional. Sapir added a new type, called synthetic. Another

kind of classification emerged with Greenberg’s seminal work (1957) on word

order typology. In this work he took multiple attributes to define a type.

…the simultaneous application of a set of individual criteria, a process which can be

conveniently conceptualized in terms of a multidimensional attribute space. The

attributes will presumably be chosen in that they represent closely related aspects of

the same general class of phenomena which may reasonably be expected to reveal

significant mutual connections.

(Greenberg (1957:76)

Comrie (1988) stipulates that the main concern of typology is to find language

universals, but at the same time diversity is observed in languages; the objective

is also to explain why languages demonstrate such diversity. Comrie (1981) has

described three reasons for languages to have common structural properties, (a)

due to genetic relatedness, (b) due to contact, and (c) due to the general

characteristics of human language. In his opinion, typologists are concerned with

the third one. Bickel (2007) discusses the concern of the 21st century typologist.

He says that typologists in 21st century are interested in developing the variables

to capture crosslinguistic similarities and differences, to explore universal and

198
 199

local skewings in the distribution of these variables, and finally to propose and

explain why we have such skewing. He further states that our ultimate goal is to

describe what is where and why. Thus typological classification is not only

restricted to finding out the properties shared by languages but to explain why

languages differ from each other.

5.2 Phonological typology of Jarawa

In a discussion of typological features of the phonology of a language, the

following points are generally discussed:

i) Whether the language is tonal or non-tonal.

ii) Which kind of vowel system has been adopted?

iii) Which kind of distinction is present at consonantal level?

iv) What is the syllabic structure?

v) What kinds of syllabic structure are common in this language at word

level?

Jarawa is a non-tonal language. It employs a seven-vowel system with a three-

way distinction in height and a two-way distinction in the horizontal position of

the raised part of the tongue. Thus, this language has /i/, /e/, /a/, /o/, /u/, /ə/ and

/ᵼ/. This seven-vowel system also has contrast in length.

The consonant system in this language contains plosives in abundance, followed

by nasals. Plosives demonstrate a five-way distinction in place {bilabial, dental,

retroflex, palatal and velar} and two-way distinction of voicing. Thus, the

language has /p, b/, /t, d/, /ʈ, ɖ /č, ǰ/ and /k, g/. Aspiration is also distinctive in

voiceless plosives except /p/. Two fricatives {voiceless bilabial and pharyngeal}

199
200

are also attested. Two approximants, bilabial and palatal, with a trill, a tap, and a

lateral are also present. No voiced consonant has aspiration at all. Labialisation is

not common in this language but one labialised pharyngeal fricative and one

labialised aspirated voiceless velar plosive are also found.

Jarawa has a (C)V(C) syllable structure with no consonant cluster either in initial

position or in final position. In this language the common phonological structure

of words is [Link]. However, other kinds of phonological structures of words

like [Link] are also attested.

5.3 Morphological typology of Jarawa

5.3.1 Root and Affixes

This language exhibits a two-way classification of roots: free roots and bound

roots. Bound roots are human-specific, consisting of human body parts, kinship

terms and some verbs which are used only for humans; free roots are used for

other than non-humans.

Affixes fall into two categories, prefixes and suffixes. Possessive pronouns,

pronouns, definiteness/referential marker, and causative elements can occur as

prefixes. Plural marking, verbaliser, mood, modality and evidentiality marker

are used as suffixes. Sometimes the copula is also prefixed to adjectives. Among

the prefixes, only the causative prefix falls under affixes since other elements like

possessive prefix, definiteness/referentiality marker and pronouns can occur

independently in syntax. In the derivational category the causative element

occurs as a prefix whereas the verbaliser element (the affix which changes a noun

into a verb) is suffixed. Plural markings on nouns, and mood, modality and

evidentiality markers on verbs fall under inflection category.

200
 201

5.3.2 Noun morphology

Jarawa employs a three-way classification of nouns: human body parts, kinship

terms, and other nouns. The human body parts and kinship terms are bound

roots whereas other nouns are free roots. Human body parts obligatorily take

definite possessive prefix or indefinite human generific possessive prefixes while

kinship terms carry only the definite possessive prefix obligatorily. In example

(5.1) oɖə ‘hair’ cannot occur as a free root if it refers to human hair; it must take a

possessive prefix which may be the indefinite human generic possessive prefix

ən- as in (5.1), or the definite possessive prefix like in (5.2). Similarly, awela

‘brother-in-law’ never occurs independently; it must occur with a definite

possessive prefix like in (5.3). Other nouns like napo ‘fish’, noha ‘bird’ čaɖɖa ‘hut’

occur as free roots. These nouns take a possessive prefix only when semantics

requires it.

(5.1) ən-oɖə

POSS[HUMAN]-hair

human hair

(5.2). m-oɖə

1POSS-hair

my hair

(5.3) m-awela

1POSS-brother-in-law

my brother-in-law (wife’s brother)

201
202

Countable nouns are divided into two, singular and plural. Singular nouns bear

zero affixation while plural nouns take plural suffix /-le/, though plural marking

is not obligatory. Nouns do not bear grammatical gender markings. No syntactic

case marking is carried on nouns, i.e. subject, direct object and indirect object

have zero case marking. This language has definiteness/referential marking on

direct object and indirect object, which differentiates subject and object of a

clause. The language makes use of semantic case marking for location, temporal

and spatial nouns.

5.3.3 Pronominal system

Jarawa employs a three-way classification of personal pronouns based on person.

There are three persons: first person, second person, and third person. For third

person pronoun Jarawa also makes use of demonstrative pronouns. There is no

number and gender distinction in the pronominal system, neither in personal

pronouns, nor in demonstratives. Like nouns, pronouns also do not bear any

kind of case marking. One of the characteristic features of personal pronouns in

Jarawa is that they can be prefixed to nouns, adjectives, and verbs (in the case of

verbs, to only those beginning with a vowel, and with no other element

intervening between the verb and personal pronoun.

5.3.4 Adjective class

Adjectives can be divided into two broad categories: pure/descriptive adjectives

form one group and demonstratives and numerals constitute another group. The

division of adjectives is based on the use of adjectives in syntactic constructions.

Pure adjectives follow nouns while demonstratives and numerals precede nouns.

Pure adjectives consist of words related to dimension, age, colour, and value.

202
 203

This type of adjectives also includes physical attributes. Possession is expressed

by the use of pronominals.

5.3.5 Verbal morphology

Like nouns, verbs are also classified into two categories: bound roots and free

roots. Only forms which are exclusively used for humans are bound roots. For

instance, ənətəhə ‘sit [HUMAN]’ has two morphemes, ən and ətəhə, because this

form has other forms like mətəhə ‘I sit’, əhitəhə ‘s/he/they sit’, and ənətəhə ‘you sit’.

For other animals oho ‘sit [NON-HUMAN] is used.

Jarawa does not show tense or aspect marking on the verbs, nor does it have

agreement marked on the verbs. However, a verb takes suffixes and prefixes

with the proviso that only one prefix and one suffix can attach to the verb at a

time. The principle function of suffixes in this language is to convey mood,

modality, and evidentiality; the prefix indicates the presence of the direct object.

This prefix is always the 3rd person pronoun and referred to as co-referencing

element in the present work. Further, personal pronouns, used as S and O

arguments of a verb, also get prefixed if there is no intervening element between

pronoun and verb. A verb that starts with a vowel may also take the

definiteness/referential marker as a prefix, provided there is no intervening

element between the definiteness/referential marker and the verb.

5.4 Word order typology of Jarawa

Though the term “word order typology” was in use for a long time, it was

Greenberg who brought the discipline to the forefront. Greenberg (1963) refers to

linguistic typology as a field which discusses the linguistic patterns of language

vis-à-vis some attributes. For instance, if a language is VSO, it will have

203
204

prepositions. These kinds of attributes are discussed in length in Greenberg

(1963). Basically, linguistic patterns were discussed with features like the

canonical pattern of subject, verb and object with reference to adpositions, noun-

adjective relations, noun-genitive relations, and noun-numeral relations. This

section focuses on word order in clauses, the position of the adposition, order of

noun-adjective, noun-genitive, noun-numeral, adjective-intensifier, verb-adverb,

and so on. This section will also investigate syntactic and morphological features

correlating with the word order as proposed by Greenberg (1963).

5.4.1 Word order in clauses

Typologically, word order in clauses means the order of subject, object and verb.

The order of these three elements determines the ordering of other pairs of

elements in the language. For example, if a language is SOV, there is a greater

chance that the language will have postpositions. Thus the order of subject,

object, and verb may determine the order of noun and adposition.

Jarawa is an SOV language: the subject precedes the object and the object

precedes the verb. This occurs in simple sentences which have a subject, a sigle

object, and a verb. But when a simple sentence has two objects, the pattern may

be subject, direct object, indirect object, and verb. Or it may be subject, indirect

object, direct object, and verb. The difference between the two is that when the

direct object is moved to the position immediately after the subject, the verb

contains the information that there is a direct object inside the clause (for the

difference in representation of direct and indirect object, see § 4.3.4). Let us

consider the following examples.

204
 205

(5.4) hʷəwə ɖi omohə

boar REF sleep

The wild boar is sleeping.

(5.5) mi hʷəwə t-ita-b

1 boar REF-eat-NEVD

I ate the wild boar.

In the case of ditransitives, the direct object and indirect object exhibit free word

order between them, but follow the overall SOV in which the subject precedes

the direct object, and direct object precedes the verb.

(5.6) mi ŋa napo ɖ-ija

1 2ACC fish REF-give

I gave you the fish.

(5.7) mi napo ɖi ŋa h-ija

1 fish REF 2ACC 3-give

I gave you the fish.

5.4.2 Peripheral arguments

Like other SOV languages, Jarawa places peripheral arguments and/or

adpositional phrases before the verb. The peripheral arguments and/or

adpositional phrases are not syntactic arguments of a verb, but demonstrate the

same pattern as object and verb. The precise positions these arguments occupy in

205
206

syntactic structures could vary. In intransitive clauses these arguments will just

precede the verb while in transitive clauses they may occur before the object. The

example in (5.8) illustrates the use of a peripheral argument of an intransitive

verb, where the peripheral argument olleg ‘place name’ precedes the verb as

object. Example (5.9) shows the position of the peripheral argument olleg ‘place

name’ which precedes the object, and therefore precedes the verb.

(5.8) tenmej olleg ɖə bəᵼʈʰe

Tenmey Olleg LOC go

Tenmey will go to Olleg/Lakralunda.

(5.9) kiʈaje ŋi olleg ɖə napo ɖi aikʰwa

yesterday 2 Olleg LOC fish REF kill

Yesterday you killed the fish at Olleg/Lakralunda.

5.4.3 Complement clauses

Jarawa make use of two types of complement clauses: clause internal and clause

external complement clauses. Clause internal complement clauses can also be

referred to as predicate internal clauses since they occur within the VP. They

precede the verb, like an object or peripheral argument. (5.10) presents the use of

clause internal complement clauses.

(5.10) mi ujijə ɖ-ipale-jə ɖi bəᵼʈʰe-jə

1 fish REF-fish[V]-VEVD REF go-VEVD

I will go fishing.

206
 207

In (5.10) we have two clauses, presented in (5.10a) and (5.10b) below. The clause

in (5.10a) is a complement clause embedded in the main clause presented in

(5.10b). The embedding of (5.10a) is evident in the presence of the morpheme ɖi

which is generally employed with the S or O argument of a verb.

(5.10a) [mi]NP [ujijə]NP [ɖ-ipale-jə]]VP ‘I will do fishing.’

(5.10b) [mi]NP [ɖi bəᵼʈʰe-jə]]VP ‘I will go.’

The other kinds of complement clauses are clause external complement clauses,

since these types of clauses follow the verbs. These types of complement clauses

occur as the complement of the verb “say”, the reporting verb. The example in

(5.11) demonstrates the use of clause external complement clause.

(5.11) čew m-aheapa naru čonel igeji

PN 1ACC-say PN banana cut

Cew told me that Naru would cut banana.

5.4.4 Adpositions

Like other SOV languages, Jarawa also has postpositions, albeit very few of

them. For instance, in example (5.12) and (5.13) the noun olleg ‘name of a place’

takes the location marker ɖə.

(5.12) mi olleg ɖə bəᵼʈʰe

1 Olleg LOC go

I will go to Olleg/Lakralunda.

207
208

(5.13) kiʈaye ŋi olleg ɖə napo ɖi aikʰwa

yesterday 2 Olleg LOC fish REF kill

Yesterday you killed the fish at Olleg/Lakralunda.

5.4.5 Comparative adjective (standard of comparision)

A typical feature of an SOV language is to place the marker of comparison after

the standard of comparison. This is true of Jarawa. For instance, in (5.15) the

standard of comparison is momo ‘name of a person’ which is compared with

another noun ugə ‘name of a person’. The marker of comparison is wɛ. From the

example it is evident that the marker of comparison follows the standard of

comparison.

(5.15) ugə momo wɛ ʈʰi učəhə

Ugə Momo COMPM COP short

Ugə is shorter than Momo.

5.4.6 Noun-adjective order

Jarawa adjectives follow the noun. However, this language distinguishes

between descriptive adjectives and demonstratives/numerals. Descriptive

adjectives follow the noun. In example (5.14) huʈʰu ‘big’ follows the noun paʈʰo

‘arrow’.

(5.14) paʈʰo huʈʰu

arrow big

A big arrow.

208
 209

5.4.7 Demonstrative-numeral-noun

Unlike descriptive adjectives, numerals always precede the noun.

Demonstratives precede the numerals. This is illustrated in the example (5.16).

(5.16) luwə naja paʈʰo

that two arrow

Those two arrows.

5.4.8 Genitive-noun

Jarawa exhibits possessor-possessed order. The possessive forms of pronominals

are either prefixed to the possessed nouns or stand independently before the

possessed nouns. Thus, possessive pronominals always come before the noun. In

example (5.17), the 1st person pronominal is possessor and in example (5.18) the

word innen ‘non-Jarawa’ is possessor.

(5.17) mi-paʈʰo

1POSS-arrow

My arrow.

(5.18) innen wa-čaɖɖa

non-Jarawa 3POSS-hut

Non Jarawa’s arrow.

If a whole noun phrase is possessed, the order will be as presented in (5.19). In

this language demonstratives and possessors are in complementary distribution.

209
210

(5.19) [(Possessor) (Numeral) {(POSS)-(Noun)} (Adj (Adv))].

In (5.19) the possessive prefix has been placed before the possessed noun in

brackets which means it is optional. It is optional when there are pronominals

which function as possessor. The examples in (5.20) and (5.21) represent the

construction of the entire noun phrase with all the modifiers.

(5.20) mi naja paʈʰo-le uʈʰu ʈʰuhumə

1Possessor two arrow-PL big very

My two very big arrows.

(5.21) innen naja wa-paʈʰo-le huʈʰu ʈʰuhumə

non-Jarawa two 3POSS-arrow-PL big very

Non-Jarawa’s two big arrows.

5.4.9 Adjective-adverb (intensifier)

As adjectives follow nouns, degree adverbs follow adjectives in this language.

Example (5.22) illustrates the use of the degree adverb ʈʰuhumə ‘very’ after the

adjective talu ‘long’.

(5.22) paʈʰo talu ʈʰuhumə

arrow long very

A very long arrow.

210
 211

5.4.10 Verb-adverb (manner of adverbs)

Degree adverbs follow verbs in this language. For instance, in (5.23) the verb

ahapela ‘run’ precedes the degree adverb ʈʰuhumə ‘very’. This usually does not

happen in SOV languages.

(5.23) li ahapela ʈʰuhumə

PDEM run very

He runs very fast.

5.4.11 Copula

The use of the Jarawa copula deviates from the expected SOV pattern. Dixon

(2010) stated that copula constructions follow the pattern of either intransitive

constructions or transitive constructions of the given language. From that point

of view, the placing of the copula in Jarawa is interesting: the copula precedes

nouns, predicative adjectives and verbs. For instance, in (5.24) the copula

precedes the noun. Similarly attribution is expressed by predicative adjectives

preceded by copula in the example (5.25). In the example (5.26), the verb cawaja

‘walk’ is used as adverb of the predicate allema ‘come’. Unlike other modifiers

(particularly intensifiers) of the predicate, cawaja precedes the main verb allema

and copula is employed before the verb cawaja.

(5.24) tenmej ʈʰi əŋ

Tenmey COP Jarawa

Tenmey is Jarawa.

211
212

(5.25) uʈʰa ʈʰi uʈʰu

Uʈʰa COP fat

Uʈʰa is fat.

(5.26) mi ʈʰi čawaja allema

1 COP walk come

I came walking.

Interestingly, the copula verb never occurs with pronominals. The examples in

(5.27) and (5.28) demonstrate that when the subject of such constructions is a

pronoun, the constructions are formed by the juxtaposing the subject and the

nominal predicate. Thus, the copula construction is possible only when the

copula subject is a nominal headed noun phrase; see examples (5.29) and (5.30).

(5.27) mi əŋ

1 Jarawa

I am/we are Jarawa.

(5.28) mi čew

1 good

I am fine.

(5.29) uʈʰa ʈʰi huʈʰu

Uʈʰa COP fat

Uʈʰa is fat.

212
 213

(5.30) tenmej ʈʰi əŋ

Tenmey COP Jarawa

Tenmey is Jarawa.

5.4.12 Causative construction

SOV languages have a tendency to form morphological causatives by means of

suffixes. Contrary to this, Jarawa employs a prefix to form morphological

causative. In example (5.31) is an intransitive verb tʰulə ‘sleep’. The causative

form of the tʰulə is constructed by adding prefix {wa-}. The example in (5.34)

represents a declarative sentence while (5.31) is an imperative sentence.

(5.31) əččəle wa-tʰulə

child CAUS-sleep

Make the child sleep.

(5.32) wa-mummə əččəle ɖi wa-tʰulə

3POSS-father child REF CAUS-sleep

His/her father made the child sleep.

5.4.13 Negation

Jarawa employs two strategies to express negation: either putting a negative

word at the end of a clause, or using a negative suffix to the verbs. Though the

negative word can be used in any type of clause, whether verbal or verbless, the

negative suffix is used only in verbal clauses. Examples in (5.33–5.38) illustrate

213
214

the use of the negative word while example (5.39) demonstrates negation of the

verbal clause.

(5.33) ullel ɖə bəᵼʈʰe naɖem

Sea LOC go NEG

Don’t go into the sea.

(5.34) tenmej ʈʰ-ero naɖem

Tenmey COP-angry NEG

Tenmey is not angry.

(5.35) tenmej ʈʰi ero naɖem

Tenmey COP angry NEG

Tenmey is not angry.

(5.36) m-ero naɖem

1-angry NEG

I am not angry.

(5.37) tenmej ʈʰi do:je naɖem

Tenmey COP girl NEG

Tenmey is not a girl.

214
 215

(5.38) mi innen naɖem

1 non-Jarawa NEG

I am non-Jarawa.

(5.39) uʈʰa allema-ma

Uʈʰa come-NEG

Uʈʰa did not come.

5.4.14 Question particle

Jarawa employs the question particle in yes/no questions. However, this question

particle also has (limited) use in content questions (wh-questions). The examples

(5.40) and (5.41) demonstrate the use of the question particle in yes/no questions

of verbless clauses while examples (5.42) and (5.43) show the use in yes/no

questions of verbal clauses.

(5.40) ka ŋi əŋ?

INTM 2 Jarawa

Are you a Jarawa?

(5.41) ka uʈʰa ʈʰi əŋ?

INTM Uʈʰa COP Jarawa

Is Uʈʰa a Jarawa?

215
216

(5.42) ka ŋa-kaja pečame?

INTM 2POSS-mother die

Did your mother die?

(5.43) ka ŋi čonel ɖi he-eŋge-jə?

INTM 2 banana REF 3-bring-VEVD

Will you bring the banana?

As mentioned, the question particle may also be used in content questions. The

examples (5.44) and (5.45) show this.

(5.44) ka-ŋi kahiunen onəhə ɖi h-ija-g?

INTM-2 morning what REF 3-give-PST

What did you give in the morning?

(5.45) ka ŋa-paʈʰo-le nojčʰe?

INTM 2GEN- arrow-PL how many

How many arrows do you have?

However, the use of the question particle in content questions is limited. Mostly

content questions are formed with the question word in situ. The examples (5.46–

5.48) illustrate the usual way of forming content questions, where question

words are employed without the question particle.

216
 217

(5.46) luwə-kʰə-ɖə ŋi nojčʰe omohə?

there-PM-LOC 2 how many sleep

How many days will you sleep there? (cf. How long will

you stay there?

(5.47) ŋi titabə ɖi onəhə ɖi h-ija-g?

2 Titabə REF what REF 3-give-ASS

What did you give to Titabə?

(5.48) hi onne ɖi aɪkȹwa?

3 who REF kill

Who did he kill?

There is one example in the corpus which shows that the question particle may

be used in between the noun phrase and the question word. In example (5.49)

cajwaj ‘name of a girl’ is a noun phrase and ʈʰucəʈəgə ‘where’ is a question word,

and question particle comes in between the two. The example given in (5.49) can

be articulated without question particle as given in (5.50). It is surprising that

here the question particle cannot be used in the beginning of the sentence; the

use of question particle at the beginning of the clause in (5.51) makes the

sentence ungrammatical.

(5.49) čajwaj ka ʈʰučəʈəgə?

Cayway INTM where

Where is Cayway?

217
218

(5.50) čajwaj ʈʰučəʈəgə?

Cayway where

Where is Cayway?

(5.51) * ka čajwaj ʈʰučəʈəgə?

INTM Cayway where

Where is Cayway?

5.4.15 Cleft sentences/relativisation

Relative constructions are attested only in interrogative constructions. These

types of relative constructions can be referred to as condensed relative clauses.

They have two clauses: a main clause, followed by a relative clause. There is no

overt relative pronoun or marker in this language but it can be interpreted as a

relative clause. So far, it has been difficult to establish the exact nature and

distribution of relative clauses, further research is required.

(5.52) onne ʈʰi əŋ taŋ odehe?

who COP Jarawa tree cut

It is which Jarawa who cut the tree?

5.5 Conclusion

This language is SOV but it deviates from the SOV patterns in many respects. A

tendency of SOV languages is to have postpositions, and this language also has a

postposition. Like some SOV languages Jarawa employs noun-adjective and

adjective-adverb patterns, but demonstratives, numerals and genitives precede

218
 219

the nouns. In the comparative construction, the marker of comparison follows

the standard of comparison. Peripheral arguments/adpositional phrases precede

the verb. In complement clauses this language employs both the patterns,

complement clauses can occur before the verb or after the verb. The negative

particle or suffix appears at the end of clause.

219
220

Chapter 6: Conclusion

The Jarawas are a monolingual community living along the western coast of

Andaman Islands. The Jarawa people are primarily hunter-gatherers of Negrito

stock, with short stature, black skin, and frizzy hair. There is no evidence of

where they migrated from. Geneticists claim that these people were part of early

migrations of Palaeolithic colonisers of Southeast Asia (Thangaraj et al. 2003: 86-

93). Kasyap et al. (2004) commented that it appeared that the aboriginal

populations of Andaman Islands remained in isolation for a longer period than

any known ancient population of the world. The Jarawas came in contact with

the outside world at the end of 1998. After full contact was established at the end

of 1998, this is the first attempt to write a descriptive grammar of the language of

the Jarawas. For data elicitation, the method of participant observation was

adopted. During the initial phase, the pictorial method was also followed. In the

pictorial method, pictures of body parts, fish, animals, insects, etc. were shown to

the consultants to collect the data related to these pictures. Natural conversations

among participants were noted down. Since the community is monolingual,

learning the language of the participants was inevitable. The data collected in the

span of four years has been the only source for writing the present thesis.

Based on the initial description, the data was further investigated in order to

place Jarawa in the typological framework provided by Greenberg (1963) and

later supplemented by Hawkins (1984) and Dryer (2007).

In segmental phonology, the sound inventory, syllabic structure, and

phonological phenomena of the language were described. The discussion also

220
 221

included problems associated with the identification and establishment of

sounds as phonemes. In order to establish the phonemic status of a particular

sound, four methodological principles were adopted. These principles were

phonological contrast, variability among sounds, distribution of sounds and

statistical criterion.

The sound inventory contains 41 sounds, out of which 28 are consonants and 13

are vowels. The vowel system was classified into two groups based on

qualitative contrasts and quantitative difference. In qualitative difference, three-

way distinction in height, three-way distinction in tongue position, and two-way

distinction in the position of lips were found. One high-mid central unrounded

vowel /ᵼ/ was also attested. Length was found to be phonemic except for the

high-mid central unrounded vowel. In the consonant category, voiceless and

voiced plosives and voiced nasals were attested. These plosives and nasals were

classified based on five places of articulation, namely bilabial, dental, retroflex,

palatal, and velar. The aspirated counterparts of these plosives were also found

(except bilabial). In the consonantal group, trill, flap, lateral sounds were also

found. Two approximants, bilabial and palatal, were also attested. Bilabial

fricative and pharyngeal fricative were also found, though this study could not

establish bilabial fricative as a phoneme. Two labialised sounds, namely voiceless

aspirated velar plosive and voiceless pharyngeal fricative, were also found.

The thesis also investigates the distribution of sounds at syllable level and word

level. Most of the consonants are found both syllable-initially and -finally, but

the sounds /ʈʰ, č, čʰ, ǰ, ɲ, r, hʷ/ are not found syllable finally. Vowels were not

taken into consideration at syllable level because vowels form the nucleus of a

syllable whether it is initially or finally. At the word level short vowels were

221
222

found to occur at initial, medial and final position while long vowels were found

to occur only initially and medially.

Syllable structure was also investigated, and it was found that the syllable in

Jarawa has a nucleus with optional onset and optional coda; the language attests

all four variants of minimal possible syllable structures: V, CV, VC and CVC. It

was also found that the language allows only one consonant in onset position

and one consonant in coda position. Words in Jarawa were found to be

monosyllabic, disyllabic or polysyllabic.

In phonological processes, the use of mi forms and ma forms was discussed.

Jarawa demonstrates the use of reduced forms in possessive constructions which

are compound words consisting of a possessive prefix and an independent root.

The first syllable of the independent root gets truncated and the reduced form is

attached to the possessive prefix.

In the morphology it was found that this language is agglutinating in nature, and

employs simple morphology in terms of affixation. There were a few prefixes

and suffixes. In the prefix category, there were pronominals and the

referential/definiteness marker whereas in the suffix category, were mood,

modality or evidential marking with verbs, plural marking with nouns, and

state/evidential marking with adjectives. Six word classes were found: noun,

pronoun, verb, adjective, adverb, and postposition.

Two kinds of roots were attested: free root and bound root. Three kinds of

prefixes were confirmed, which were possessive prefixes, pronominals, and the

referential/definiteness marker. In the suffix category negative suffix, evidential

222
 223

marker, mood/modality marking, state/evidential marking, plural marking,

denominalisation suffix were attested.

Two kinds of Noun Phrases (NPs) were found: nominal headed NPs and

pronominal headed NPs.

Pronominals are divided into three persons: first person, second person and third

person. Plural counterparts of these were not found. Morphologically,

pronominals were found to behave differently from nouns. Pronominals attach

to nouns as possessive prefixes, adjectives as subjects of complement clauses, and

to verbs as subjects of intransitive verbs or objects of transitive verbs. Two forms

of pronominals are also attested. These two forms were the “mi” form and the

“ma” form. Phonologically these forms demonstrated vowel harmony in the case

of the possessive prefix. When the possessive prefixes are used as independent

words, “mi” form is used. In syntax the “mi” form was found to be used as

subject whereas the “ma” form was used as object. Though the morphological

behaviour of pronominals indicated that they could possibly be considered as a

clitic, in this study this phenomenon is left for further research.

Two kinds of demonstrative pronouns, proximate li and distant luwə, are

attested. onne ‘who’, onəhə ‘what’, kahaje ‘where’, ʈʰučəgə ‘where’, nojčʰe ‘how

many/much’ were observed as interrogative pronouns.

Nouns were classified into Class I (bound roots) and Class II (free roots). Class II

were further classified semantically into celestial bodies, cultural artifacts, and

flora and fauna.

223
224

Countable nouns were categorised as singular and plural. Singular nouns were

found to be zero marked while plural nouns had suffix {-le}. The use of plural

marking was found to be optional.

Possession is expressed by prefixation of pronominals. Inalienable nouns,

consisting of body parts and kinship terms, were found as bound roots. In

alienable possession both the nouns are free roots. In this kind of possession, the

possessive prefix (pronominal) is attached to the possessed noun.

The lexically headed noun phrase contains a noun as head, and optional

modifiers like possessive prefix, numerals, descriptive adjectives and adverbs.

The possessive prefix, demonstratives, and numerals precede the noun, while

descriptive adjectives and intensifier follow the noun.

Verbs in this language neither exhibit agreement nor carry tense and aspect

information morphologically. Time reference is expressed periphrastically by the

use of temporal adverbs. The Jarawas’ perception of time does not go beyond

three days. palahe ‘long before’ was found to be employed to express beyond

three days. Verbs were found to possess a few prefixes and suffixes.

Pronominals, referential/definiteness marking and co-referencing element were

seen to be used as prefixes, and mood, modality and evidentiality markers as

suffixes.

This language distinguishes between hypothetical and assertive moods, and

evidentiality. Evidential markers convey whether the action is verifiable or non-

verifiable. Hypothetical mood indicates future actions/events which are either

doubtful or uncertain. Assertive mood expresses assertion or emphasis of the

action/events.

224
 225

Two kinds of causative constructions, morphological and syntactic, were found.

Morphological causatives are formed by prefixing causative prefix {wa-} to verbs.

Limited data suggested that morphological causative is employed to a handful of

verbs, and the syntactic causative across the board. In syntactic causatives, the

independent word is used to express causation along with the lexical verb.

In the adjective category, this language was observed to demonstrate descriptive

adjectives, numerals, and demonstratives. Descriptive adjectives consist of words

related to colour terms or which define, describe, and qualify nouns. Also found

was marking of the degree of adjectives. In degrees of comparison positive,

comparative, and superlative degrees are attested. The positive degree is formed

by the noun phrase and adjective. The comparative degree was seen to be

expressed syntactically. In the comparative degree, the standard of comparison

was observed to precede the marker of comparison, and the adjective to follow

the marker of comparison. Superlative degree was seen as formed by employing

the intensifier or a completely new word which was not found to have any

resemblance with the adjective of positive degree.

Adverb was also found as a separate word class. These are lexical items used as

intensifiers, temporal adverbs, and spatial adverbs. No morphological

derivational process was confirmed to form adverbs from adjectives.

No syntactic case marking was observed in this language. However, two

morphemes were attested as postpositions to express semantic case. These

morphemes were {ɖə} and {ɲačʰe}. The morpheme {ɖə} has multiple functions in

this language. This morpheme was found to express place of location, source of

location, and goal of location. It was also frequently found with temporal and

225
226

spatial adverbs, but not always. The morpheme {ɲačʰe} expresses the

commitative.

The morpheme {ʈʰi} was identified as copula. This copula is used to identify an

entity or qualify a noun. It was also seen employed with negative particles and

with temporal and spatial adverbs in short responses. Sometimes this morpheme

was also seen with verbs which were performing the function of adverbs. The

corpus suggested that this copula does not take inflection, but one example was

found where it was inflected for mood.

Four word formation processes were confirmed. These word formation processes

were compounding, denominalisation, reduplication, and the use of empty

morpheme. In compounding, three combinations were substantiated: noun-

noun, noun-adjective, and noun-noun-adjective. In the denominalisation process,

we saw that nouns are denominalised to verbs with the help of a suffix. In

reduplication, Jarawa demonstrates syllable reduplication and word

reduplication. As for word reduplication, the corpus suggested that verbs were

reduplicated to serve the function of adverbial modifiers.

A notable phenomenon was also observed under word formation processes: a

morpheme used mostly with personal nouns (but also other nouns) was

discussed, but the meaning of this morpheme was not clear. Hence this

morpheme was referred to as an empty morpheme.

In syntax, the different syntactic structures of Jarawa, and their variances were

discussed. These syntactic structures included simple clause structures,

interrogative constructions, imperative sentences, negative constructions, and

226
 227

cleft constructions. It was also observed that this language follows the ergative-

absolutive system.

In declarative sentences, verbless and verbal clauses were discussed. Verbless

clauses consisted of non-verbal predicates like nouns and adjectives as the heads

of clauses. Copula constructions were also included in verbless clauses. Verbal

clauses contained verbs as heads of main clauses.

In verbless clauses we encountered two different types of clauses: one type had a

nominal as the head of the predicate, and another had an adjective as the head of

the clause. Verbless clauses also included copula constructions, since the use of

copula constructions were mostly observed in this type. In verbless clauses with

a nominal as head, two constructions were discussed: one where the S argument

was a pronominal headed NP and another where it was a lexically headed NP.

Lexical NPs were found to take copula constructions while pronominal headed

NPs never take copula constructions.

In verbless clauses with adjectival predicates, three kinds of constructions were

discussed. One was with prefixation of a pronominal S argument in adjectival

predicates beginning with a vowel, and which never takes a copula. The second

was copula constructions with a lexically headed NP as S argument which take a

copula. The third was adjectival predicates suffixed for state/evidentiality, which

also do not take the copula.

In verbal clauses there were four kinds of constructions discussed. These were

intransitive, transitive, ditransitive, and constructions with denominalised verbs.

All these constructions were found to have different structures.

227
228

We saw that in intransitive clauses with pronominal headed NP as the S

argument, the NP can be prefixed to the verb (if the verb starts with a vowel).

This is not allowed with a lexically headed NP. However, the lexically headed

NP, if specific, takes the referential/definiteness marker in intransitive clauses.

Apart from these intransitive constructions, we encountered denominalised

constructions. These were posited to include (purely intransitive) weather type

constructions, and another type of construction, which semantically appeared to

be transitive but, behaved as if they were intransitive clauses.

Transitive clauses were considered as a separate group, with a transitive verb as

the head of the predicate. In such constructions two core arguments were

confirmed. These core arguments are A(gent) and O(bject). In these constructions

A precedes O and O precedes the verb. If the O argument is a pronominal

headed NP, it generally attaches to the verb if the verb begins with a vowel. In

the case of a lexically headed NP, we had observed the insertion of the

referential/definiteness marker after it. This referential/definiteness marker was

seen to alway move with the D(irect) O(bject).

In ditrasitive clauses, the position of A and the verb was found to be fixed in

declarative sentences, i.e. A always occurs at the beginning of the clause, and the

verb at the end of clause. The order of I(ndirect) O(bject) and DO might change.

The referential/definiteness marker is used to specify the DO, hence this marker

moves along with it. However, IO may also take referential/definiteness marker.

Apart from the core arguments, peripheral arguments of the clause were shown

to have different positions in the clause. Temporal adverbs generally occur at the

beginning while spatial adverbs/nouns just before the verb.

228
 229

To express negation, this language makes use of an independent word naɖem

after a predicate (verbal or verbless), or a suffix -ma on a verb. There was no

semantic difference observed between the two. We saw that the independent

word naɖem could be used across the board while the suffix -ma had restricted

use.

Imperatives are formed by using the verb with its core argument. As we saw,

there is no morphological marking to differentiate imperatives from other kinds

of sentences.

In interrogatives, we looked at yes/no questions and wh-questions, which differ

syntactically. Yes/no questions are formed by placing the interrogative marker ka

at the beginning of a clause, while in wh-questions the wh-word occupied its situ

position. The use of the interrogative marker with some wh-questions was also

found.

In the category of complex sentences, I explored coordination and complement

clauses. A conjunction ən ‘and’ was found to be used to express coordination. In

coordinated clauses, it was found that conjunction ən ‘and’ can be employed at

phrase level and clausal level to form coordinated clauses.

Two different kinds of complement clauses were also discussed. These two types

of clauses were clause internal and clause external complement clauses. In a

clause internal complement clause, the complement clause occurs within the

main clause in the object position of the verb phrase. Clause external

complement clauses were described as instances of direct speech in this

language. Since the language was found to be SOV, meaning the object occupies

229
230

the slot before verb, these kinds of constructions were considered as clause

external complement clauses.

Direct speech and cleft constructions were established as types of complex

constructions found in Jarawa. It was suggested that there is only direct speech

in this language (based on pronominal correlation). Cleft constructions were

found to have two sub-types: ɖi-constructions and ʈʰi-constructions. The former

suggests referential/definiteness of the entity referred to, and the latter

indefiniteness of the entity referred to.

The morpheme ɖi was shown to have two different usages in this language. One

usage is to indicate definiteness/referentiality, and the second is to express

grammatical relation. Referentiality was treated in this dissertation as applicable

when the entity referred to is within sight of the speaker-hearer. Definiteness,

then, is applicable when the entity referred to is not within sight but the speaker-

hearer has shared knowledge. . However, some restrictions were found on the

use of this morpheme. It never occurs with pronominals and postposed Os. It can

occur only with lexically headed S or O arguments of a clause. This morpheme

was also seen employed with interrogative pronoun onne ‘who’ and clause

internal complement clauses. In the case of the interrogative pronoun onne ‘who’,

the cleft construction is formed.

The way this morpheme is employed in this language also indicated that this

morpheme expressed grammatical relations, i.e. ergative-absolutive system. The

language employs split ergative-absolutive system. In the case of

definiteness/referentiality this morpheme was usually employed with S or O. The

use of the morpheme with A was rare if S or O were found to be

230
 231

definite/referential. Thus, this morpheme conveyed grammatical relations in the

context of definiteness/referentiality. With an indefinite object this morpheme

was not employed with S, O, or A.

Typologically, this language was found to be SOV. Postpositions were found.

Descriptive adjectives and manner of adverbs follow the head while numerals

and demonstratives precede the head. In the comparative degree, the marker of

comparison follows the standard of comparison. The copula precedes non-verbal

predicates in copula constructions. Negative particles or suffixes appear at the

end of the clause.

231
232

Bibliography

Abbi, Anvita 2001. A Manual of Linguistic Fieldwork and Structures of Indian

Languages. Munchen: Lincom Europa.

Abbi, Anvita. 2006. The Endangered Languages of the Andaman Islands. Munchen:

Lincom Europa.

Abbi, Anvita et al (2007). Where have all the speakers gone? A Sociolinguistic

study of the Great Andamanese. Indian Linguistics 68: 325-343

Abbi, Anvita. 2009. ‘Is Great Andamanese genealogically and typologically

distinct from Onge and Jarawa?’ Language Sciences 31: 791-812

Abdulali, h. 1962. The Wild Pigs in the Andamans. Journal of the Bombay Natural

History Society 59.1: 281-283.

Acharya, S. and S. Datta. 1998. The Jarawas are Coming. Sunday 25.48: 36-39.

Agrawal, H. N. 1967. ABO blood groups and the sickle cell investigations among

the Shompens of Great Nicobar, Journal of Indian Anthropological Society, 1:

149-150.

Agrawal, H. N. 1967. Physical characteristics of the Shompen of Great Nicobar,

Bulletin of Anthropological Survey of India, 1967: 83-97.

Aiyangar, S. K. 1923. Rajendra, the Gangaikonda Chola. Journal of Indian History

2.3: 317-369.

Akerman, K. 1975. Baler Shell Implements from North-West Australia. Mankind

10: 16-19.

232
 233

Alexander, J. E. 1827. Notice regarding the Little Andaman Island, Bay of Bengal.

The Edinburgh New Philosophical Journal 2: 43-48.

Alexander, James Edward 1827. Travels from India to England: Comprehending a

Visit to the Burman Empire, and a Journey Through Persia, Asia Minor,

European Turkey, Etc., in the Years 1825-26. London: Parbury, Alien and Co.

Andaman and Nicobar Administration, 1909. Gazetteer: Calcutta.

Andaman and Nicobar Administration. 1977. Retrieval from Precipice, AAJVS, Port

Blair.

Anderson, C. W. B. 1905. Exploration of Great Nicobar. London

Anderson, J. 1890. The Selungs of the Mergui Archipelago. London: Trubner and Co.

Awaradi, S. A. 1990. Computerized Master Plan (1991-2021) for Welfare of Primitive

Tribes of Andaman and Nicobar Islands. Port Blair: Andaman and Nicobar

Administration.

Awasthi, A. K. 1988. Plants used as food items by the tribals of Andaman and

Nicobar Islands. Journal of Andaman Science Association 4.2: 128-131.

Baer, A. 1995. Human Genes and Biocultural History in Southeast Asia. Asia

Perspectives 34.1: 21-35.

Baer, A. 2000. The Genetic History of the Orang Asli: Uniting Patchwork Data.

Bulletin of the Indo-Pacific Prehistory Association 19.3: 3-10.

Bailey, R. C., G. Head, M. Jenike, B. Owen, R. Rechtman and E. Zechenter. 1989.

Hunting and Gathering in Tropical Rain Forest: Is it Possible? American

Anthropologist 91: 59-82.

233
234

Balfour, Patrick. 1935. Grand Tour: Diary of an Eastward Journey. New York:

Harcourt, Brace and Co.

Bamshad, Michael J. et al 1998. ‘Female Gene Flow Stratifies Hindu Castes.’

Nature 395 (15 October 1998): 651-652.

Bamshad, Michael J. 2001. ‘Genetic Evidence on the Origins of Indian Caste

Population.’ Genome Research, November 2001: 1-11.

Banerjea, Kanak Tilak. 1998. ‘Among the Jarawas.’ Sunday Statesman, 8 February

1998:8.

Banerjee, P. K. 1993. ‘Imprints of Late Quaternary Climatic and Sea Level

Changes on East and South Indian Coast.’ Geo-Marine Letters 13: 56-60.

Barbosa, D. 1921. The Book of Duarte Barbosa, 2. Cambridge: The Hakluyt Society.

Basu, A. 1969. The Pahira: A population genetical study. American Journal of

Physical Anthropology 21:229.

Basu, B. K. 1990. The Onge. Calcutta: Seagull Books.

Basu, D.N. 1952. A Linguistic Introduction to the Andamanese. Bulletin of the

Department of Anthropology, Vol. 1-2, July pp. 55-70.

Belcher, E.1867. Notes on the Andaman Islands. Transactions of the Ethnological

Society of London 5: 40-49.

Bellwood, P. 1978. Man’s Conquest of the Pacific: Prehistory of Southeast Asia and

Oceania. Aukland: Collins.

Bellwood, P. 1985. Prehistory of the Indo-Malaysian Archipelago. New York:

Academic Press.

234
 235

Béteille, A. 1998. The Idea of Indigenous People. Current Anthropology 19.2: 187-

191.

Bhargava, N. 1981. Plants in Folk Life and Folklore in Andaman and Nicobar

Islands. In S. K. Jain, ed. Glimpses of Indian Botany, pp. 229-334. New Delhi:

Oxford and IBH Publishing Co. Pvt. Ltd.

Bhattacharya, Asutosh. 1983. Andhakarer Andamanese (In the Dark Andamans).

Calcutta: A. Mukherjee and Co.

Bickel, Balthasar 2007. ‘Typology in the 21st Century: Major Current

Developments. Linguistic Typology 11: 239-251.

Binford, L. R. 1980. Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement

System and Archaeological Site Formation. American Antiquity 45.1: 4-20.

Bird-David, N. H. 1991. Hunter-gatherers and Other People: A Reexamination. In

T. Ingold, D. Riches, and J. Woodburn, ed., Hunters and Gatherers 1:

History, Evolution and Social Change, pp. 17-30. New York/Oxford: Berg.

Bird-David, N. H. 1992. Beyond ‘the Hunting and Gathering Mode of

Subsistence’: Culture Sensitive Observations on the Nayaka and Other

Modern Hunter-Gatherers. Man (N.S.) 27: 19-44.

Birdsell, J. B. 1950. Some implication of the genetic concept of the race in terms of

spatial analysis. Cold Spring Harbor Symposia on Quantitative Biology,

15:259-314.

Birdsell, J. B. 1977. The Recalibration of a Paradigm for the First Peopling of

Greater Australia. In Allen, J. Golson, and R. Jones, ed., Sunda and Sahul,

pp. 113-167. New York: Academic Press.

235
236

Blevins, Julliette 2007. A long lost sister of Proto-Austronesian? Proto-Ongan,

mother of Jarawa and Onge of the Andaman Islands. Oceanic Linguistics 46

(1): 155–198.

Boden Kloss, C. 1903. In the Andamans and Nicobars: The Narrative of a Cruise in the

Schooner ‘Terrapin’, with Notices of the Islands, Their Fauna, Ethnology, Etc.

London: John Murray.

Bonin, G. von. 1931. Beitrag zur kraniologie von Ost-Asien. Biometrica 24: 52-113.

Bonington, M. C. 1931. Extract from the Census Report on the Andaman and

Nicobar Islands, Census of India 1931, Vol. 2, Part III pp. 172-176.

Bonington, M. C. 1933. The Andaman and Nicobar Islands. Calcutta: Superintendent

Government Printing, India.

Bonnington, M. C. 1932. Census of India, 1931. Andaman and Nicobar Islands,

Vol II, Parts 1and 2. Calcutta: Government of India Publication.

Bose, S. 1964. Economy of the Onge of Little Andaman. Man in India 44: 289-310.

Bowdler, S. 1983. Seiving Seashells: Midden Analysis in Australian Archaeology.

In G. Connah ed., Australian Field Archaeology: A Guide to Techniques, pp.

135-144. Canberra: Australian Institute of Aboriginal Studies.

Bowrey, T. 1905. A Geographical Account of Countries round the Bay of Bengal

(1699 to 1679). R. C. Temple. (ed.) Cambridge: The Hakluyt Society.

Breuil, H. and Lantier, R. 1965. The men of the Old Stone Age. London: George G.

Harrap Co. Ltd.

Burenhult, N. 1996. Deep Linguistic Prehistory with Particular Reference to

Andamanese. Lund University, Dept. of Linguistics Working Papers 45: 5-24.

236
 237

Burkitt, M. C. 1963. The old stone Age. New Delhi: Rupa & Co.

Busk, George. 1865. ‘Description of Two Andamanese Skulls.’ Transactions of the

Ethnological Society 4: 205-211.

Campbell, Colonel Walter. 1864. My Indian Journal. Edinburgh: Edmonston and

Douglas.

Cappieri, M. 1954. Some Essential Features of Andamanese Anthropology and

Demogenetics. Proceedings of the 4th International Congress of Anthropological

and Ethnological Science, Vienna 1952, 1: 293-298.

Cappieri, M. 1970. The Racial Homogeneity of the Andamanese, Part III:

Osteological Examination and Anthropometry. Mankind Quarterly 11.2: 96-

113.

Census of India 1961. The Andaman and Nicobar Islands, General report, Part I, Vol.

xvii (17). Delhi: Government of India Press.

Census of India 1981. The Andaman and Nicobar Islands, General Economic

Tables and Social and Cultural Tables, Part III, A and B, and Part IV, A

Chakraborthy, D. K. 1977. Great Andamanese of Strait Islands. Calcutta:

Anthropological Survey of India.

Chakraborty, D. K. 1990. The Great Andamanese. Calcutta: Seagull Books.

Chanda, Sudhendu. 1976. The Hostile Jarawa of Andaman and Semi Hostile

Shompens of Great Nicobar Island. Journal of Social Research 19.2: 64-73.

237
238

Chaturvedi, Y. 1980. Mammals of the Andamans and Nicobars: Their

Zoogeography and Faunal Affinity. Records of the Zoology Survey of India

77: 127-139.

Chengapa, B. S. 1958. In the Land of Hostile Jarawas and Other Wild Tribes of

the Andaman Islands. Indian Forester, February 1958: 108-120.

Choudhary, N. C. 1967. Tribes of Andaman Islands, Indian Museum Bulletin,

January 1967, pp. 51-58.

Choudhary, N. C. 1976. The Onge of Little Andaman-20 years after. Journal of

Social Research. 19.2:51-63.

Cippieri, M. 1953. ‘Some essential features of the Andamanese anthropology and

demogenetics.’ Journals of Anthropology Society (Bombay) 7:1-39.

Cipriani, L. 1953. Report on a Survey of the Little Andaman during 1951-1953.

Bulletin of the Anthropological Survey of India 2.1: 61-82.

Cipriani, L. 1954. A Survey of Little Andaman during 1954. Bulletin of the

Anthropological Survey of India 3.2: 66-94.

Cipriani, L. 1955. Excavations in Andamanese Kitchen-middens. Actés du IVe

Congrèss International des Sciences Anthropologues et Ethnologiques, Vienna,

II: 280-283.

Cipriani, L. 1966. The Andaman Islanders. London: Wiedenfeld and Nicolson.

Clark, G. 1992. Space, Time and Man: A Prehistorian’s View. Cambridge: Cambridge

University Press.

238
 239

Clifton, Mrs Talbot. 1911. Pilgrims to the Isles of Penance: Orchid Gathering in the

East. London: John Long.

Colebrooke, R. H. 1795. On the Andaman Islands. Asiatick Researches 4: 385-394.

Comrie, Bernard 1981. Language Universals and Linguistic Typology: Syntax and

Morphology. Chicago: Chicago University Press.

Comrie, Bernard 1988. Linguistic Typology. Annual Review of Anthropology 17:

145-159.

Connor, Steve. 1998. DNA Links “Stone Age” Tribe to First Humans. Independent

31. (August 1998)

Cooper, Z. and H. Achyuthan. 1991. Archaeological and Geomorphological

Investigations in the Andaman Islands. Tropical Geomorphology Newsletter

11: 7-9.

Cooper, Z. and H. Raghavan. 1988. Analyses of Sendiments from a Cave in the

Andaman Islands. Man and Environment 12: 67-74.

Cooper, Z. and H. Raghavan. 1989. Petrographic Features of Andamanese

Pottery. Indo-Pacific Prehistory Association Bulletin 9: 22-32.

Cooper, Z. and S. Bowdler. 1998. Flaked Glass Tools from the Andaman Islands

and Australia. Asian Perspectives 37.1: 74-83.

Cooper, Z. 1985. Archaeological Explanation in the Andaman Islands. Indo-Pacific

Prehistory Association Bulletin 6: 27-39.

239
240

Cooper, Z. 1987. A Report on Archaeological Investigations in the Andaman Islands

from 1984 to 1986. Submitted to the Homi Bhabha Fellowships Council,

Bombay.

Cooper, Z. 1988. Shell Artefacts from the Andaman Islands. Australian

Archaeology 26: 24-41.

Cooper, Z. 1989. Analysis of the Nature of Contacts with the Andaman Islands

during the Last Two Millennia. South Asian Studies 5: 133-147.

Cooper, Z. 1990a. Archaeological Evidence for Resource Exploitation in the

Andaman Islands. Man and Environment 15.1: 73-81.

Cooper, Z. 1990b. The Problem of the Origins of the Andamanese. Bulletin of the

Deccan College Post-graduate and Research Institute 49: 99-104.

Cooper, Z. 1990c. The End of ‘Bibipoiye’ (Dog Not) Days in the Andamans. In B.

Meehan and N. White ed. Hunter-Gatherer Demography: Past and Present,

pp. 117-125. The University of Sydney, Oceania Monographs 39.

Cooper, Z. 1993a. The Origins of the Andaman Islanders: Local Myth and

Archaeological Evidence. Antiquity 67: 394-399.

Cooper, Z. 1993b. Perception of Time in the Andaman Islands. World Archaeology

25.2: 261-267.

Cooper, Z. 1994. The Relevance of the Study of Abandoned Onge Encampments

in Understanding the Archaeological Record in the Andaman Islands. In

B. Allchin ed., Living Traditions: Studies in the Ethnoarchaelogy of South Asia,

pp. 235-63. New Delhi: Oxford and IBH Publishing Co. Pvt. Ltd.

240
 241

Cooper, Z. 1996a. Archaeological Evidence for Human Settlement in the

Andaman Islands. Indo-Pacific Prehistory Association Bulletin 15.2: 139-142.

Cooper, Z. 1996b. Archaeological Evidence of Maritime Contacts: The Andaman

Islands. In H. P. Ray and J. –F. Salles, ed. Tradition and Archaeology: Early

Maritime Contacts in the Indian Ocean, pp. 239-245. New Delhi:Manohar.

Cooper, Z. 1997a. Salient Features of Site Location in the Andaman Islands. Asian

Perspectives 36.2: 220-231.

Cooper, Z. 1997b. Prehistory of the Chitrakot Falls. Pune: Ravish Publishers.

Cooper, Zarine, 2002. Archaeology and History: Early Settlement in Andaman Islands.

Delhi: Oxford University Press.

Cooper, Zarine. 1989. Analysis of the Nature of Contacts with the Andaman

Islands During the Last Two Millennia. South Asian Studies 5: 133-147.

Coxon, Stanley W. 1915. And That Reminds Me: Being Incidents of a Life Spent at Sea,

and in the Andaman Islands, Burma, Australia, and India. New York: John

Lane Co.

Critchfield, Jean. 1963. Car Nicobarese Vocabulary. Chicago: University of Chicago

(mimeo).

Cutting, Suydam. 1932. Natives of the Andaman Islands. Natural History 32.5:

521-530.

Danda, A. K. 1988. ‘Process of ethnic integration: Some theoretical and

methodological problems.’ Human Science 37:356-368.

Danda, A. K. 1991. Ethnicity in India. New Delhi: Inter-India Publications.

241
242

Danda, D. 1987. ‘The Onge of Little Andaman.’ Human Science, 24: 89-90.

Danda, G. Dipali. 1987. ‘Little Andaman and the Onge.’ Human Science 36: 66-91.

Das, A. R. 1977. A Study on the Nicobarese Language. Calcutta: Anthropological

Survey of India.

Das, V. and A. Kagal. 1990. Through the Prism of Clerkdom. The Times of India:

Sunday Review, 16 September: 1-2.

Dasgupta, D and S. R. Sharma. 1982. A Hand Book of Onge Language. Calcutta:

Anthropological Survey of India.

Devi, L. Dilly 1980. Man-Food Relationship among the Onge. in I. P. Singh and S.

C. Tiwari (ed) Man and Environment.

Director of Health Service, Andaman and Nicobar Administration 1982: Health

Care Services in Andaman and Nicobar Islands, (report), Port Blair.

Dixon, R. M. W. 2009. Basic Linguistic Theory I and II. Oxford: Oxford University

Press.

Dobson, G. E. 1875. On the Andamans and Andamanese. Journal of the Royal

Anthropological Institute, London 4: 457-467.

Dryer, M. S. 1992. The Greenbergian Word Order Correlations. Language 68: 81-

138.

Dryer, M. S. 2007. Word Order. In Shopen, T. (ed.) Language Typology and

Syntactic Description I. Cambridge: Cambridge University Press.

242
 243

Dunn, F. L. 1968. Epdemiological Factors: Health and Disease in Hunter-

Gatherers. In R. Lee and I. DeVore, ed. Man the Hunter, pp. 221-28.

Chicago: Aldine.

Dutta, P. C. 1962. A Trial Excavation of a Kitchen-midden at South Andaman. Wiener

Vokerkundliche Mittelungen 10: 17-29.

Dutta, P. C. 1978. The Great Andamanese: Past and Present. Calcutta:

Anthropological Survey of India.

Dutta, P. C. 1983. Research in Biological Anthropology of the Andaman Negritos:

A Critical Review. Anthropologie 21.3: 259-268.

Dutta, T. R., R. Ahmed, S. R. Abbas, and M. K. V. Rao. 1985. Plants used by

Andaman Aborigines in Gathering Rock-bee Honey. Economic Botany 39.2:

130-138.

Ehlers, Otto E. 1898. An Indichen Furstenhofen. Berlin: Allgemeiner Verein für

Deutsche Litteratur.

Eichstedt, E. V. 1929. Über fas Aussterben der Negritos auf den Andamanen.

Archiv für Rassen und Gesellschafts Biologie 21: 1-4.

Eichstedt, E. V. 1930. Unter den Andamaneischen Zwergnegern. Die Umschau 34:

246-268.

Eickstedt, E. V. 1928. Die Negritos der Andamanen. Anthropologischer Anzeiger

5.3: 259-268.

Ellis, A. J., E. H. Man, and R. C. Temple. 1882. Report of Researches into the

Language of the South Andaman Islands, arranged by A. J. Ellis from the

243
244

papers of E. H. Man and Lieut. R. C. Temple. Transactions of the Philological

Society of London, pp. 44-73.

Erickson, P. and S. Beckerman. 1975. Population Determinants in Andaman

Islands. Mankind 10.2: 105-107.

Flower, W. H. 1880. On Affinities of Natives of the Andaman Islands. Journal of

the Royal Anthropological Institute 9: 108-135.

Forest Statistics 1998. Andaman and Nicobar Island Forest Department. Port Blair:

Van Sadan.

Fox, A. G. 1995. Malayan Paleosociology: Implications for Patterns of Genetic

Variation among the Orang Asli. American Anthropologist 97.2: 313-323.

Fox, A. L. 1878. Observations on Mr. Man’s Collection of Andamanese and

Nicobarese Objects. The Journal of the Anthropological Institute of Great

Britain and Ireland 7: 434-451.

Fox, R. G. 1969. Professional Primitives: Hunter and Gatherers of Nuclear South

Asia. Man in India 49.2: 139-160.

Fytche, Albert. 1861. A note on Certain Aborigines of the Andamans. Journal of

the Asiatic Society of Bengal 30: 263-267.

Ganguli, P. 1961. Religious Beliefs of the Negritos of Little Andaman. Eastern

Anthropologist 14.3: 243-248.

Ganguli, P. 1966. Vocabulary of the Negritos of Little Andaman with

Grammatical Notes and Materials. Bulletin of Department of Anthropology

Vol. 15: 1-30.

244
 245

Gates, R. R. 1940. Blood Groups from Andamans. Man 140: 50-56.

Glover, I. C. 1996. Recent Archaeological Evidence for Early Maritime Contacts

between India and Southeast Asia. In H. P. Ray and J. F. Salles ed.

Tradition and Archaeology: Early Maritime Contacts in the Indian Ocean pp.

129-158. New Delhi: Manohar.

Gomes, A. G. 1990. Demographic Implications of Villigisation among the Semang

of Malaysia. In B. Meehan and N. White, ed. Hunter-Gatherer Demography:

Past and Present pp. 126-138. University of Sydney: Oceania Monograph,

39.

Greenberg, Joseph H. 1963. Some universals of grammar with particular

reference to the order of meaningful elements. In Joseph Greenberg (ed.),

Universals of Language. Cambridge, MA: MIT Press, 73–113.

Greenberg, J. H. 1971. The Indo-Pacific Hypothesis. Current Trends in Linguistics

8: 807-871.

Guha, B. S. 1952. Report of a Survey of the Inhabitants of the Andaman and

Nicobar Islands During 1948-49, Bulletin of the Department of Anthropology

Vol. 1:1-7

Guha, B. S. 1954. A comparative study of the somatic traits of the Onges of Little

Andaman. Bulletin of Department of Anthropology 3.2: 117-143.

Gupta, Bandana. 1976. The Andamans. Calcutta: Jijnasa Publication.

Gupta, P. and Basu, A. 1960. Dermatoglyphics of Onges of Little Andaman.

Bulletin, Anthropological Survey of India 9.2: 51-63.

Gupta, S. K. 1951. Census of India 1951: Andaman and Nicobar Islands.

245
246

Haughton, J. C. 1861. Papers Relating to the Aborigines of the Andaman Islands.

Journal of the Asiatic Society of Bengal 1861: 251-263.

Hawkins, J. A. 1983. Word Order Universals. New York: Academic Press.

Hill, S. C. 1923-1928. Notes on Indian Piracy. The Indian Antiquity 52: 1-52; 53: 53-

84; 54: 85-92; 55: 93-124; 56: 125-204; 57: 205-248.

Holland, T. H. 1904a. Ancient Kitchen-Midden in the Andamans. Records of the

Geological Survey of India 31.1: 45.

Holland, T. H. 1904b. Ancient Kitchen-Midden in the Andamans. Records of the

Geological Survey of India 31.2: 107-108.

Hutton, J. H. 1931. Notes on the Andamanese and the Nicobarese. Man in

India11:1-5.

Hutton, J. H. 1951. Less Familiar Aspects of Primitive Trade. Proceedings of the

Prehistoric Society 7: 171-176.

Jafri, S. H. and S. N. Charan. 1992. Quench Textures in Pillow Basalt from the

Andaman and Nicobar Islands, Bay of Bengal, India. Proceedings of the

Indian Academy of Science 101: 99-107.

John Mase Field 1908. The Travels of Marcopolo. London: Everyman’s Library.

Joshi, M. C. 1971. Census of India 1971, Series 23, Andaman and Nicobar Island

Part II-A.

Justin, A. 1990. The Nicobarese. Calcutta: Seagull Books.

Kennedy, K. A. R. 1999. Paleoanthropology of South Asia. Evolutionary

Anthropology 8.5: 165-185.

246
 247

Kloss, C. B. 1903. In the Andamans and Nicobars – the Narrative of a Cruise in the

Schooner ‘Terrapin’. London. (Reprint 1971, New Delhi: Vivek Publishing).

Kloss, C. Boden 1903. In the Andaman and Nicobars. London: John Murry.

Kumar, Umesh and Biswas, Samir. 2002. The Jarawas and Their Habitat: A Man-

Nature Relationship. In Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact:

Ours with them theirs with us. Kolkata: Anthropological Survey of India.

Lal, Parmanand 1976. Andaman Islands: A Regional Geography. Calcutta:

Anthropological Survey of India.

Lapicque, L. 1894. Ethnographie des iles Andaman. Bulletins de la Societe

d’Anthropologie de Paris 3 Mai: 359-369.

Leach, E. R. 1971. Kimil: A Category of Andamanese Thought. In P. Maranda and

E. K. Maranda, ed., Structural Analysis of Oral Tradition. pp. 22-48.

Philadelphia: University of Pennsylvania Press.

Lehmann, Herman. 1955. The Andaman Islands. St. Bartholomew’s Hospital

Journal: 99-112.

Lowis, R. F. 1912. Census of India 1911. The Andaman and Nicobar Islands. Calcutta:

Superintendent Government Printing, India.

Lowis, R. F. 1923. Census of India 1921. The Andaman and Nicobar Islands. Calcutta:

Superintendent Government Printing, India.

Majumdar, S. N. 1927. McCrindles Ancient India as Described by Ptolemy. Calcutta:

Chudervertty, Chatterjee and Co. Ltd.

247
248

Majumder, R. C. 1975. Penal Settlement in Andamans. New Delhi: Ministry of

Education and Social Welfare, Government of India.

Man, E. H. and R. C. Temple. 1880. Note on Two Maps of the Andaman Islands.

The Journal of the Royal Geographical Society 50: 255-259.

Man, E. H. and Temple, R. C. 1877. The Lord’s Prayer in the South Andaman

Language. Calcutta.

Man, E. H. 1878a. The Andaman Islands. The Journal of the Anthropological

Institute of Great Britain and Ireland 7: 105-109.

Man, E. H. 1878b. On the Arts of the Andamanese and Nicobarese. The Journal of

the Anthropological Institute of Great Britain and Ireland 7: 451-467.

Man, E. H. 1882. On Andamanese and Nicobarese Objects. The Journal of the

Anthropological Institute of Great Britain and Ireland 11: 268-294.

Man, E. H. 1883. On the Aboriginal Inhabitants of the Andaman Islands. The

Journal of the Anthropological Institute of Great Britain and Ireland 12: 69-175;

327-434.

Man, E. H. 1885. On the Aboriginal Inhabitants of the Andaman Islands. Reprinted

1932, Great Britain: Bibling and Sons Ltd.

Man, E. H. 1885. On the Andaman Islands and their Inhabitants. The Journal of the

Anthropological Institute of Great Britain and Ireland 14: 253-272.

Man, E. H. 1886. A Brief Account of the Nicobar Islands with Special Reference to

the Inland Tribe of Great Nicobar. The Journal of Anthropological Institute of

Great Britain and Ireland 15: 428-451.

248
 249

Man, E. H. 1888. A Dictionary of the Central Nicobarese Language. Reprinted 1975.

Delhi: Sanskaran Prakashak

Man, E. H. 1889. The Nicobar Islanders. The Journal of Anthropological Institute of

Great Britain and Ireland 18: 354-393.

Man, E. H. 1894. Nicobar Pottery. The Journal of the Anthropological Institute of

Great Britain and Ireland 23: 21-27.

Man, E. H. 1923. Dictionary of the South Andaman Language. Bombay: British India

Press.

Man, E. H. 1883 (Reprint 1975). The Aborginal Inhabitants of the Andaman Islands.

Delhi: Sanskaran Prakashak.

Mann, R. S. 1976. Depopulation among Andaman Aborigines – An Analysis.

Journal of Social Research 19: 37-50.

Mann, R. S. 1980. The Bay Islanders. Bidisa: Institute of Social Research and

Applied Anthropology.

Manoharan, S. 1978. A Typological study of Andamanese and Nicobarese

Language, 10th ICAES, Mysore.

Marradi, A. 1990. ‘Classification, Taxonomy, Typology’. Quality and Quantity 24:

129-157.

Mathews, E. 2002. Health Problems and Health Care of the Jarawas in the

Transition Phase. In Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact:

Ours with them theirs with us. Kolkata: Anthropological Survey of India.

249
250

Mathur, L. P. 1968. History of the Andaman and Nicobar Islands (1756-1966). Delhi:

Sterling Publishers Pvt. Ltd.

Mathur, L. P. 1985. Kalapani: History of Andaman and Nicobar Islands with a Study of

India’s Freedom Struggle. New Delhi: Eastern Book Corporation.

Mc Carthy, A. D. M. 1940. Notes on the Jarawa Language. Port Blair: Chief

Commissioner’s Press.

Mouat, Frederic John. 1863. Adventures and Researches Among the Andaman

Islanders. London: Hurst and Blackett.

Moyne, Lord 1936. Walkabout: A Journey in Lands between the Pacific and Indian

Oceans. London: William Heinemann Ltd.

Mukerjee, Madhusree. 1999. Out of Africa, Into Asia. Scientific American 280.1

(January 1999): 24.

Mukerjee, Madhusree. 2003. The Land of Naked People: Encounters with Stone Age

Islanders. New Delhi: Penguin Books India Ltd.

Mukhopadhyay, K. 1989. The tribal policy of Jawaharlal Nehru: Tribes and Tribal

Policy. In K.S. Singh (ed) Calcutta: Anthropological Survey of India, pp 12-

26.

Mukhopadhyay, K. 2002a. A Brief History of Relationship Between the Jarawas

and Others. In Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact: Ours with

them theirs with us. Kolkata: Anthropological Survey of India.

Mukhopadhyay, K. 2002b. Welfare of the Jarawas and Some Post-Colonial

Thoughts. In Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact: Ours with

them theirs with us. Kolkata: Anthropological Survey of India.

250
 251

Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact: Ours with Them, Theirs with Us.

Kolkata: Anthropological Survey of India.

Myka, F. 1993. Decline of Indigenous Populations: The Case of the Andaman Islanders.

New Delhi: Rawat Publications.

Myka, F. 1995. Of Adoption and Orphanages: The Biocultural Dynamics of

Population Decline among the Andaman Islanders. Man in India 75.1: 1-9.

Naidu, T. S. 1999. The Jarawa: An Anthropological Perspective to Conserve the

Endangered Stone Age Aborigines. Pondicherry: Centre for futures Studies,

Pondicherry University.

Nair, V. S. 1979. A note on the Language of Jarawa. Bulletin of Anthropological

Survey of India 28 (3-4): 17-38.

Nigam, R. C. 1963. The Onge of Little Andaman – Their Maintenance

Institutions. Vanyajati 11: 147-156.

Nigam, R. C. 1955-56. Report on Field Investigations on Onge Speech. Bulletin of

the Department of Anthropology Vol. 13: 127-146.

Oldham, R. D. 1885. Notes on the Geology of the Andaman Islands. Records of the

Geological Survey of India 28.3: 135-145.

Olschki, I. 1960. MarcoPolo’s Asia. Translated from Italian by J. A. Scott. Berkeley

and Los Angeles: The University of California Press.

Page, Michael F. 1952. A Sea with Many Islands. London: Robert Hale Ltd.

Pandit, T. N. and Lall, T. 1985. The Tribal and Non-Tribal in Andaman Islands: A

Historical Perspective. Journal of Indian Anthropological Society 20: 111-131.

251
252

Pandit, T. N. 1974. Jarawa: Story of a Century of Mutual Hostility. Daily

Telegrams, 16th and 18th March. Port Blair.

Pandit, T. N. 1976a. The Original Inhabitants of the Andaman and Nicobar

Islands. Yojan 15th of August, 1976.

Pandit, T. N. 1976b. The Ethnic Situation in the Bay Islands. Yojana, 15th of

August, 1976.

Pandit, T. N. 1989. Tribal policy in the Andaman and Nicobar Islands: The

impact of Nehru’s philosophy. In K. S. Singh (ed) Jawaharlal Nehru: Tribes

and Tribal Policy. Calcutta: Anthropological Survey of India, pp 83-92.

Pandit, T. N. 1990. The Sentinelese. Calcutta: Seagull Books.

Pandit, T. N. 1991. Close Encounters with Stone Age. Sunday Times of India, 28

April 1991: 1-2.

Pandya, V. Above the Forest: A study of Andamanese Ethnoanemology, and the Power

of Ritual. Delhi: Oxford University Press.

Pandya, V. 1991. From Photography to Ethnography: Andamanese Documents

and Documentation. Visual Anthropology 4: 379-413

Pandya, V. 1999. ‘To Contact or Not? The Jarawas of the Andaman Islands.’

Cultural Survival Quarterly (Winter 1999): 59-65.

Parkinson, C. E. 1923. A Forest Flora of the Andaman Islands. Simla: Government

Central Press.

Pelliot, P. 1959. Notes on Marco Polo 1. Paris: Imprimerie Nationale, Librairie

Adrien-Maisonneuve.

252
 253

Pelliot, P. 1963. Notes on Marco Polo 2. Paris: Impriimerie Nationale, Librairie

Adrien-Maisonneuve.

Planning Commission, 1988. The Interim Report of Expert Group on the

Primitive Communities of Andaman and Nicobar Islands, Delhi

Portman, M. V. 1889. (Reprint 1990). A History of Our Relations with the

Andamanese. 2 vols. New Delhi: Asian Educational Services.

Radcliffe-Brown, A. R. 1909. The Religion of the Andaman Islanders. Folklore

20.3: 257-271.

Radcliffe-Brown, A. R. 1914. Notes on the Languages of the Andaman Islands.

Anthropos Vol. 9: 36-52.

Radcliffe Brown, A. R. 1922 (Reprint 1948). The Andaman Islanders. Glencoe,

Illinois: The Free Press.

Radhakrishnan, R. 1978. ‘Nicobarese Numerals.’ 10th ICAES, Mysore.

Rajasingh, V. R. and Ranganatha, M. R. 2000. A Handbook of Jarawa Language.

Mysore: Central Institute of Indian Languages.

Rajasingh, V. R. 2002. Structure of the Jarawa Language. In Mukhopadhyay, K. et

al (ed) Jarawa Contact: ours with them, theirs with us. Kolkata:

Anthropological Survey of India.

Ramachandran, R. 2000. A People in Peril. Fronline 17.16: 84-86.

Renaudot, E. 1733. Ancient Accounts of India and China by Two Mohammedan

Travellers (9th Century), translated from Arabic. London: Sam Harding.

253
254

Report of the Expert Committee on the Jarawas of Andaman Islands. July, 2003. New

Delhi: MHA, Govt. of India.

Rizvi, S. N. H. 1990. The Shompen: A Vanishing Tribe of the Great Nicobar Island.

Calcutta: Seagull Books.

Rodolfo, K. S. 1969. Bathymetry and Marine Geology of the Andaman Basin, and

Tectonic Implications for Southeast Asia. Geological Society of America

Bulletin 80: 1203-1230.

Roychowdhury, Rabin. 2004. Black Days in Andaman and Nicobar Islands. New

Delhi: Manas Publications.

Safford, W. E. 1902. The Abbott Collection from the Andaman Islands. Annual Report

of the Smithsonian Institution 1901, pp. 475-492. Washington: Government

Printing Office.

Sangal, P. M. 1971. Forest Food of the Tribal Population of Andaman and

Nicobar Islands. Indian Forester 97.2: 646-650.

Sarkar, B. N. 1985. The Jarawas of Andaman Islands: An Anthropometric Study.

Human Science 34: 186-194.

Sarkar, B. N. 1987. Dermatoglyphic Study of the Jarawa: A Negrito Tribe of

Andaman Islands. Human Science 36: 346-358.

Sarkar, B. N. 1989. Anthropometric study of the Jarawa of Andaman Islands.

Journal of Indian Anthropological Society 24: 79-83.

Sarkar, J. K. 1987. Befriending the Jarawa: A Challenging Task. Human Science 36:

1-12.

254
 255

Sarkar, J. K. 1989. Endangered Tribes and their Development in Andaman and

Nicobar Islands. Journal of Indian Anthropological Society 24: 1-45.

Sarkar, J. K. 1990. The Jarawa. Calcutta: Seagull Books.

Sarkar, J. K. 1996. Tribal Development Administration for the Small Population:

The Andaman and Nicobar Situation. Journal of Indian Anthropological

Society 31: 57-61.

Sarkar, J. 2002. No More Hostility: A Critical Phase in Jarawa Society. In

Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact: Ours with them theirs

with us. Kolkata: Anthropological Survey of India.

Sarkar, S. J. 1974. Socio-economic Aspects of Onge Fishing. Anthropos 69.3/4: 568-

589.

Sarkar, S. S. 1933. Negrito racial strain in India. Nature 137: 3477.

Sarkar, S. S. 1952. Blood Groups from Andaman and Nicobar Islands. Bulletin

Department of Anthropology 1.1: 25-30.

Sarkar, S. S. 1954. Aboriginal Races of India. Calcutta: Book Land Ltd.

Sekhsaria, P. 1998. Jarawa excursions. Frontline, 17 July: 66-71.

Sekhsaria, Pankaj (ed.) 2007. Troubled Islands. Pune: Mudra

Sekhsaria, Pankaj and Pandaya, Vishvajit 2010. The Jarawa Tribal Reserve Dossier.

New Delhi: UNESCO.

Sen, P. K. 1962. Land and People of the Andamans. Calcutta: Post Graduate Book

Mart.

255
256

Senkuttuvam, R. 2000. The Language of the Jarawa. Calcutta: Anthropological

Survey of India.

Service, Elman. 1971. The Andaman Islanders, In G. Prakash Reddy and V.

Sudersen (ed) Island Cultures of India, Secunderabad.

Sewell, R. B. S. 1925. Geographic and Oceanographic Researches in Indian

Waters: The Geography of the Andaman Sea Basin, Part I. Memoirs of the

Asiatic Society of Bengal 9: 1-26.

Sharma, B. D. 1975. Development of Small Tribal Communities: A theoretical Frame

Workshop on Primitive Tribal Communities. New Delhi: Ministry of Home

Affairs.

Sharma, B. D. 1976. An Approach to the Development of Onges of Little

Andaman. Journal of Social Research 19.2: 25.

Singh, Bhaktawar. 2002. Befreinding the Jarawa: Some Personal Experiences. In

Mukhopadhyay, K. et al (ed) 2002. Jarawa Contact: Ours with them theirs

with us. Kolkata: Anthropological Survey of India.

Singh, K. S. 1985. Tribal Society in India. Delhi: Manohar.

Singh, N. I. 1978. The Andaman Story. New Delhi: Vikas Publishing House Pvt.

Ltd.

Singh, Raghubir. 1975. The Last Andaman Islanders. National Goegraphic 148: 66-9

Sinha, Surjeet C. 1982. Reports of the Committee of Anthropologists on the

Jarawa and Shompen Situations, Ministry of Home Affairs, Government

of India, Delhi

256
 257

Solomon, V. 1902. Extracts from Diaries kept in Car Nicobar. The Journal of the

Anthropological Institute of Great Britain and Ireland 32: 202-238.

Sreenathan, M. 1998. Rhetorical violence: Language ideology and Jarawa

hostility. Man in India 78: 355.

Sreenathan, M. 2001. The Jarawas: Language and Culture. Calcutta: Anthropological

Survey of India.

Srinivasan, M. S. 1978-79. Geology and Mineral Resources of the Andaman and

Nicobar Islands. Andaman and Nicobar Information. Port Blair: Government

Printing Press.

Stiles. D. 1993. Hunter-Gatherer Trade in Wild Forest Products in the Early

Centuries AD with the Port of Broach, India. Asian Persectives 32.2: 153-

167.

Stoliczka, F. 1870. Note on the Kjökkenmöddings of the Andaman Islands.

Proceedings of the Asiatic Society, January, 1870, pp. 13-23.

Sullivan, L. R. 1921. A few Andamanese skull with comparative notes on Negrito

Craniometry. Anthropological papers of the American Museum of Natural

History, 23: 174-199.

Suryanarayan, V. and V. Sudarsen (ed) 1994. Andaman and Nicobar Islands:

Challenges of Development. New Delhi: Konark Publishers.

Syam Chaudhuri, N. K. 1977. The Social Structure of Car Nicobarese Islanders.

Calcutta: Anthropological Survey of India.

Symes, Michael. 1800. An Account of an Embassy to the Kingdom of Ava, sent by

the Governor-General of India in the Year 1795. London: J. Debrett.

257
258

Taylor, Frederick. 1911. Native Life in the Andaman Islands. Century Magazine

82: 891-904.

Temple, R. C. 1900a. Extracts from the Bengal Consultations of the 18th Century

Relating to the Andaman Islands. The Indian Antiquary 29: 103-116; 125-

135, 167-171, 286-288; 32 (1903): 34-44, 86-91, 193-201.

Temple, R. C. 1900b. An Unpublished Document about the Nicobars. The Indian

Antiquary 29: 341-347.

Temple, R. C. 1901a. Census of India, Andaman and Nicobar Islands, Government of

India Publication, Calcutta.

Temple, R. C. 1901b. The Trade in Andamanese Slaves. The Indian Antiquary 30:

120.

Temple, R. C. 1901c. An Unpublished 18th Century Document about the

Andaman Islands. The Indian Antiquary 30: 232-138.

Temple, R. C. 1903. The Andaman and Nicobar Islands. Census of India 1901, vol. 3.

Calcutta: Office of the Superintendent of Government Printing (Reprint

1985. New Delhi: Usha).

Temple, R. C. 1900. Extracts from the Bengal Consultation of the XVIIIth Century

Relating to the Andaman Islands. Indian Antiquary April 1900: 103-116.

Temple, R. C. 1901. An Unpublished XVIIIth Century Document about the

Andamans. Indian Antiquary June 1901: 232-238.

Temple, R. C. 1929. Remarks on the Andaman Islanders and Their Country.

Indian Antiquary July 1929.

258
 259

Theobald, W. 1862. Notes on Some Celts found in Bundelkhand and Some Chert

Implements from the Andamans. Journal of the Asiatic Society of Bengal 31:

323-328.

Thomson, D. F. 1939. The Seasonal Factor in Human Culture. Proceedings of the

Prehistoric Society 5: 209-221.

Tikader, B. K. 1984. Birds of the Andaman and Nicobar Islands. Calcutta: Zoological

Survey of India.

Toussaint, A. 1966. History of the Indian Ocean. translated by J. Guicharnaud.

London: Routledge and Kegan Paul.

Venkatesan, D. 1988.’ Innovations and Their Impact on the Onge of Little

Andaman.’ Man in India 68: 413-423.

Venkateswar, Sita. 1997. Policing Power, Governing Gender, and Re-imagining

Resistance: A Perspective on the Andaman Islanders. Ph.D. dissertation,

Rutgers University. New Jersey.

Venkateswar, Sita. 1999. The Andaman Islanders. Scientific American May 1999:

82-88.

Verma, I. C. 1976. Health Problems of Tribal Communities in Andaman and

Nicobar Islands, Journal of Social Research 19.2: 16-24.

Vidyarthi, L. P. 1976. Development Plans of Tribes of Andaman and Nicobar

Islands: An Action Oriented Report. Journal of Social Research 19.2: 74-85.

Whitaker, R. and Z. 1986. The Andaman Tribes – Victims of Development.

Cultural Survival Quarterly 10.2: 13-18.

259
260

Whitaker, R. 1985. Endangered Andamans: Managing Tropical Moist Forests. The

Environmental Series Group. World Wildlife Fund, India, April 1985.

Yule, H. 1929. The Book of Ser Marco Polo. London: John Murray Albermarle

Street.

Zide, Norman H. and Stampe David L. 1963. Initial Consonants Proto-Munda-

Nicobarese: Some Tentative Correspondences. Chicago: University of Chicago

(mimeo).

260
 261

Appendix IA

JARAWA: WORD-LIST-NOUN

S No. English Words Jarawa Words

001. old man ənka:b

002. pot bučʰu

003. rock penon

004. rock (very big) oɖəgə

005. land / soil pela

006. root ʈʰucə

007. a kind of creeper ʈʰigul

008. branch / thin stem ʈʰobučə

009. smoke bənel / pənel

010. adze titʰoja͂

011. holding part of an adze he apitaja

012. cutting part of an adze toe / tɔ

013. rifle hupətaŋ

014. bullet ənə:ɖə

015. a kind of leaf tʰujijə

016. pipe ʈʰahə

261
262

017. butterfly peʈpel

018. a kind of tree itəŋ

019. string from bark of itəŋ opajəle

020. cane tree halog

021. wooden basket uhuwə

022. creek hʷətʰol

023. fishing hook ujijə

024. plastic bottle olo olo

025. cap of a bottle olo olo de dubə

026. dry wood nam / name

027. stone ulijə

028. rope pado

029. arrow paʈʰo

030. bow a:w

031. fishing net patʰo

032. place aɖəhə

033. cough uɖu

034. a kind of tree from which they make tʰaɖo

uhuʷə

035. flower ohag

036. helicopter čakogere / ŋeŋeŋe

262
 263

037. airplane talahʷə / tugənulə

038. looking glass ənohabag

039. swing tʰuji

040. flame dʰundʰul

041. wind / air biji:ŋ

042. moon tape

043. sun jehe

044. dog wəwəmə

045. girl do:jə

046. boy tawə

047. banana čonel

048. ant čaɖapa

049. canoe taŋ

050. pit oɖuəlleu͂

051. spade dojə

052. sand bi:lə / bi:ɫə / bilowə / tetʰal

053. a kind of flower ʈʰapabaɖ

054. friend mahema / mitaǰəle

055. a kind of leaf whose soft stem serves eboǰim

to prepare broom
056.

263
264

057. stick used to make fishing net čʰumijə

058. a kind of fruit a:min-e

059. pillar outside hut to extend it ʈʰukča̯ɖɖa:jejə

060. toa:ɖ
cutter

061. resin bone

062. forest kɨjə / gɨjə

063. pig hʷəwə

064. honey lə:w

065. road / path le:b

066. a kind of flower toʈkʰolo

067. sky paŋgaŋ / paŋnaŋ

068. coconut ɖag

069. string of leaf around head ənajipotəhə

070. string of wool around head ənotohab

071. string of leaf around waist kaŋgapo

072. string of leaf suspending around waist pato (all of these ornaments are
made of a leaf called tʰujijə

075. string of leaf suspending around neck ənajinoɖanɖo

over back

076. string of leaf suspending around neck ənahodəwə

over chest

264
 265

077. string of leaf neck ŋagitohabə

078. eggs in honeycomb ha:čimə

079. honeycomb čilemal

080. honeycomb found inside tree poɖə

081. zeep totkile

082. truck baɽaw

083. wave ʈʰomaje / ʈʰome ʈʰuhumə

084. shells ʈʰepʈʰew

085. hut / home čaɖɖa:

086. wild potato čə:w / ǰə:w

087. wild potato či:gijə

088. wild potato papa~b̥ab̥a

089. wild potato nadohətaŋ

090. hillock/mound ʈʰinon

091. crowbar/anvil otahətə

092. net basket ta:ɨkə / ta:ɽkə

093. pool of water əŋhowal

094. watercourse (nallah) hʷa-iŋ-tapo / hʷawə

095. time when no water is there ənʈohʷa

096. time when there is lot of water onojə ʈʰuw boʈʰejə

265
266

097. fresh water for drinking iŋ-tapo

098. bad water iŋ-piʈʰi

099. sea water ʈʰomajə

100. thirst iŋkə

101. coal nape:n

102. ash hʷi:jə

103. testis ənikʰʷəg

104. tooth ənahoɖ

105. a kind of small mosquito hʷikičin

106. a kind of fruit hʷa:tʰə:

107. pond inkopotatʰe

108. a kind of fish which looks like snake tʰugən

266
 267

Appendix IB
JARAWA: WORD-LIST-VERB

S No. English words Jarawa words

001. to fry očʰenə: / očʰe-gə

002. to drink inčo-wa

003. to eat ita

004. to walk čawaja

005. to run əh-apela

006. to sing/talk gəgapa / gəgaba

007. to fry fish ənane-hə / eʈʰaʈuhagle

008. to dig out (potato) ipine

009 to fry (potato) ənane

010. to boil (potato) oʈʰaj-le

011. to go for hunting tahilə

012. to go bəɨʈʰe

013. to sleep omohə

014. to sleep tʰulə

015. to see əjojəba / illiyema / iliŋ

016. to sound iʈʰəhəʈʰe

267
268

017. to hear the sound of bus uʈʰiʈʰigu iʈʰəhəʈʰe

018. to quarrel / fight əniɲa / ənija

019. to angry ero

020. to push əɖə / oɖə

021. to pull oj ͂i ͂ʈʰege / ehʷa

022. to pull [HUMAN] ən-ipəʈʰeŋe

023. to carry ənena̯ / ənəjale

024. to lift ətʰokagʈʰe / atʰokaʈʰe

025. to stay / assemble čalahe

026. to write ənobelijahije

027. to speak /talk / name aʈiba

028. to give iya

029. to cut ičilo

030 to cut hair ən-oʈʰa / ən-oɖə-kali

031. to cut firewood həjutʰu / nam oɖəhe

032. to construct hut čaɖɖa-le

033. to hide arrow hə-əja / ahowa

034. to hide fish ajigɖəwa

035. to hide biscuit əʃaʈʰa / əčʰaʈʰa

036. to fall huwaji

268
 269

037. to throw hogi / apine

038. to forget oʈʰa ən-iɲijačʰe

039. to dig with a crowbar ənohelə

040. to dig with a spade pelajitʰə

041. to dig ipine

042. to come here allema-gə

043. to come here (to call up) əne͂wa

044. to weep wanna

045. to laugh əniji: / ənijə

046. to smile ən-agulemelegune

047. to remember oʈʰa ən-ijelaŋ

048. to catch fish in sea ulleləɖa̯

049. to catch fish in stream hʷaɖə

050. to take eŋge

051. to hold / catch kʰo:

052. to sit ən-ə:təhə

053. to stand ɖokəkte / tokəʈʰe

054. to tie / fasten aɸle-yə / akʰle

055. to unfasten ilibua

056. to keep fish napo ɖ-aye / he-əya

269
270

057. to come closer buʈʰə tʰame / ən-uhuwə

058. to wet ən-aŋgtʰə / ənənʈʰəkaneha

059. to rain wəwə-le

060. to take bath inɖə

061. to smell bad aʃu / ačʰu

062. to be hungry aŋgjačʰu

063. to be thirsty əbulə:

064. to marry əna əŋga-le

065. to burn wood ʈʰuhəb-le

066. to get frighten / scared ən-ənpəʈʰa

067. to bring ekane

068. to bring eŋge

069. to climb čapʈʰe

070. to get ən-ičajmə

071. to get down ǰagi-jə

072. to scratch / itching eweewe

073. to fry banana ekka

074. to fish waʈʰepa-le

075. to fish (alone) napowə

076. to fish (alone but nearby) ikʰə-wə ɖə aikʰwəkʈʰe

270
 271

077. to fish (in a group) napo tahile

078. to go home (to come back) unnə

079. to hunt (in a group) pijalɛ

080. to hunt (one day tour) howa

081. to hunt (more than one day) kijə tahile

082. to kill with an arrow aikʰwa

083. to steal uɨgəle

084. to swim waɽa

085. to smell ihune

086. to wash hand ən-ahopaho-le

087. to know oʈʰa inijəla

088. to bark helahoə:

089. to finish / to end hojata / ičʰeba

090. to pull hair ənoɖə h-igəʈʰe

091. to jump down from tree ənotoha

092. to tell a lie əhujəba

093. to bop əɲetɔ / əŋgo

094. to pluck off pakǰawaǰi

095. to collect ehaʈo

096. to be quite ən-aholagʈʰi

271
272

097. to tie ən-ahotʰwag

098. to untie ilebuwə

099. to sound in stomach ən-ihʷeɖə

100. to throw ipinunu

101. tooth picking ən-ahugluhə

102. to spit tʰuwə

103. to push hə-əɖe-jə

104. to pull ən-ahela-le

105. to jump upwards cowaǰi

106. to become small boi͂ja-gə

107. to jump into water əhanekʰʷeɖaǰi

108. to massage əneyom

109. to shy ən-ahimudijəʈʰe

110. to be dull (standing) ən-howa

111. to put together hawočewoče-le

112. to bite (by ant) ənigine

113. to embrace tʰowagə / eneje

114. to sneeze ən-ohoʈʰahe

115. to pull a rope pado he-ehʷag

116. to hang hehehe

272
 273

117. to carry on back apa

118. to be stationary iwəčə

119. to blow əhubə

120. to make ičʰe / ičʰi

121. to lift by hand aʈʰoha-g

122. to cut a tree taŋg hi-itʰə

123. to fall down (tree) hewojabə

124. to hammer inɖə / kʰočaleluhə toa

125. to lift əpa

126. to swing tʰui-le

127. to ride bicycle taŋg-le

128. to rub ačʰile-le

129. to beat onodəhejə

130. to kiss ən-imunəji

131. to show untopine

132. to make arrows / bows ejaja

133. to make fishing net uluhə patʰo

134. to make net basket bod

135. to make necklace hagulume

136. to burn wood ʈʰuhəgle

273
274

137. to shoot pig / fish / dear hatitijə

138. to click ɲono

139. to pain ənihəd / ulleɖa / kʰamohə

140. to take this arrow paʈʰo ɖə-ekəwaja

141. to empty / vacant hut čaɖɖa ɖ-epoha

142. to make a move ən-əʈʰuči / ən-əʈʰučə

143. to fly ugikəɨʈʰe

144. to go up ʈʰugeǰa-jə

145. to remove the bark ipole

146. to cough əna-uɖɖu-le

147. to smell bad inəgə piʈʰi

148. to smell good inəgə čew

149. to tear leaf oʈʰahejə

150. to leak from the roof of hut huijahuija

151. to go to water source iŋ ičʰe

152. to bring water / fetch water iŋ əčʰetahile

274
 275

Appendix IC
JARAWA: WORD-LIST-ADJECTIVE-ADVERB

S No. English Word Jarawa Word

001. far no:ɖə

002. near buʈʰə

003. inside itəla

004. white haləŋda

005. black hiɽu

006. red hoɽgidu

007. blue ələŋda

008. many mala

009. one wə:ja

010. two naja

011. three kaŋ itəǰile

012. four ənə itəǰile / malawə

013. five oyaʈʰe / mala ʈʰuhumə:

014. hot (summer) hulug

015. bad piʈʰi

016. good če:w

275
276

017. morning (7-10 am) kahiunnen

018. noon hejauijə

019. evening ʈʰepale

020. night kiʈʰale

021. today olla

022. tomorrow kekahəʈə

023. yesterday kiʈaje

024. day after tomorrow jakeka

025. day before yesterday četali

026. heavy anponi

027. left kuhʷə

028. right toheja

029. mad ənanačʰegə

030. mine own ma-huəɖə

031. your own ŋa-huəɖə

032. mine ma:g

033. happy əmpoəwə

034. sad ijan-piʈʰi

035. weak gəro

036. bald hʷəʈkʰele

276
 277

Appendix - II

JARAWA: SENTENCES

001. bibʰuti bəɨʈʰe-jə allema-ma

PN go-VEVD come-NEG

Bibhuti, having gone, has not come back.

002. bibʰuti bəɨʈʰe-jə allema nadem

PN go-VEVD come NEG

Bibhuti, having gone, has not come back.

003. kahiunen allema čawaja čawaja

morning come walk walk

In the morning we came walking.

004. əŋ ɖi aʈiba he h-ita

Jarawa REF speak 3 3-eat

Jarawa says, “he eats as he hita.”

277
278

005. talu-əbba ɖi aʈiba kiʈaje miɖil sʈeʈ ɖə bəɨʈʰe-hə

PN-father REF speak yesterday [Link] LOC go-HYP

Talu’s father said yesterday, “We will go to Middle Strait.”

006. čajwaj ummə olo olo ɖ-ija

PN father plastic bottle REF-give

Cayway’s father, give the plastic bottle/pass the plastic bottle.

007. wa-kaja:-g totanahi

3POSS-mother-EVD PN

His/her mother is Totanahi.

008. čajwaj ummə a:pa ipo

PN father carry PN

Cayway’s father carries Ipo on his back.

009. čajwaj ummə ipo h-apa:-g

PN father PN 3-carry-ASS

Cayway’s father carries Ipo on his back.

278
 279

010. iskul ɖə innen waja čaɖɖa ɖi bəɨʈʰe-jə

school LOC non-Jarawa children home REF go-VEVD

Non-Jarawa children are going home from the school.

011. majoba ʈibi ɖ-əjojəba čajwaj ummə ɖ-eŋge

PN TV REF-see PN father REF-bring

Mayoba, see the TV, Cayway’s father has brought it.

012. əččəle-wə ɖi boi ͂ja

child REF small

The child is small.

013. jehejə ɖi ǰagijə-gə

sun REF go down-ASS

The sun sets/is setting.

014. kaŋgapo ʈʰuhumə ɖi əčʰeəʈʰaje-jə

cloth too much REF tear-VEVD

Too much cloth was torn/being torn.

279
280

015. innen ɖi kaŋgapo ɖ-əčʰeəʈʰaje-jə

non-Jarawa REF cloth REF-tear-VEVD

A non-Jarawa tore the cloth/was tearing the cloth.

016. kaŋgapo odaje / otajehe ʈʰuhumə-le-jə

cloth tear too much-VS-VEVD

The cloth was torn.

017. mala-wə ɖi bəɨʈʰe

many REF go

Many have gone.

018. li taŋg ɖə čapʈʰe-jə

PDEM tree LOC climb-VEVD

He is climbing the tree.

019. taŋg ɖə pɨpə ɖi huwaǰi-jə

tree LOC leaf REF fall-VEVD

Leaves are falling from the tree.

280
 281

020. mi napo ɖi ŋi h-ija

1 fish REF 2 3-give

I gave you the fish.

021. uʈʰa hi-inɖə-jə

PN 3-hammer-VEVD

Uʈʰa is hammering.

022. mi innen ɖ-ilijema

1 Non-Jarawa [Link]

I did not see the non-Jarawa.

023. babu ilijema tenmej

official [Link] PN

The official did not see Tenmey.

024. babu tenmej ɖ-ilijema

official PN [Link]

The official did not see Tenmey.

281
282

025. əŋ m-ilijema

Jarawa [Link]

The Jarawa did not see me.

026. hʷəwə ɖi-ahapela əŋ ɖi-aikʰwa

boar REF-run Jarawa REF-kill

The boar was running and the Jarawa killed it.

027. hʷəwə aikʰwa bəɨʈʰe ʈʰi noɖa

boar kill go COP far

To kill the boar we go far. (lit. Going to kill the boar is far)

028. mi hiwa napo ɖ-ija

1 3ACC fish REF-give

I gave him the fish.

029. mi ŋa napo ɖ-ija

1 2ACC fish REF-give

I gave you the fish.

282
 283

030. hi ma napo ɖ-ija

3 1ACC fish REF-give

He gave me the fish.

031. ŋi ma napo ɖ-ija

2 1ACC fish REF-give

You gave me the fish.

032. kekahəʈa̯ mi ŋa napo ɖ-ija

tomorrow 1 2ACC fish REF-give

Tomorrow, I will give you a fish.

033. kiʈaje mi ŋa napo ɖ-ija

yesterday 1 2ACC fish REF-give

Yesterday, I gave you a fish.

034. kekahəʈa̯ ɖə mi napo t-ita-hə

tomorrow LOC 1 fish REF-eat-HYP

Tomorrow, I will eat a fish.

283
284

035. mi napo t-ita

1 fish REF-eat

I ate a fish.

036. kiʈaje ɖə mi napo t-ita

yesterday LOC 1 fish REF-eat

Yesterday, I had eaten a fish.

037. kiʈaje ɖə mi gɨjə~kɨjə

Yesterday LOC 1 forest

Yesterday, I went hunting (Yesterday I went to the forest).

038. kiʈaje ɖə mi tahilə

yesterday LOC 1 go

Yesterday, I went hunting.

039. kekahəʈa̯ mi hʷəwə tahilə-hə

tomorrow 1 boar go-HYP

Tomorrow, I will go for hunting. (Tomorrow, I will go for a boar.)

284
 285

040. mi-mugə ʈʰi boi͂ja

1POSS-beard COP small

My beard is small.

041. mi-mugə ʈʰi uʈʰu-ʈʰuhumə

1POSS-beard COP big-very

My beard is very big.

042. mi ɖomo-bə mi aʈiba-hə

1 fine-NVEVD 1 speak-HYP

When I am fine, I will talk to you.

043. kahiunen ɖə ŋi onəhə t-ita-bə

morning LOC 2 what REF-eat-NEVD

What did you eat in the morning?

044. kahiunen ɖə mi napo t-ita-bə

morning LOC 1 fish REF-eat-NEVD

In the morning, I ate fish.

285
286

045. mi ŋ-əjojəba

1 2-see

I saw you.

046. hi m-əjojəba

3 1-see

He saw me.

047. ŋi m-əjojəba

2 1-see

You saw me.

048. mi əŋ ɖ-əjojəba

1 Jarawa REF-see

I saw a Jarawa.

049. innen əŋ ɖ-əjojəba

non-Jarawa jarawa REF-see

A non-Jarawa saw a Jarawa.

286
 287

050. innen m-əjojəba

non-Jarawa 1-see

A non-Jarawa saw me.

051. kiʈaje ɖə mi ŋ-əjojəba

yesterday LOC 1 2-see

Yesterday, I saw you.

052. čajwaj ummə li waja

PN father PDEM [Link]

This is Cayway’s father’s child.

053. mi palaʈʰəŋ ŋ-əjojəba

1 earlier 2-see

I saw you earlier.

054. palaʈʰəŋ mi ŋi əjojəba

earlier 1 2 see

I saw you earlier.

287
288

055. taŋ ɖ-ithəgə ɖ-ɨəʈʰe-jə

wood REF-cut REF-go-VEVD

I will go for cutting wood.

056. čə:w ɖ-ipine toto-hə

wild potato REF-dig out go-HYP

I will go for digging out wild potato.

057. ujijə ɖ-ipale toto-hə

fish REF-fish go-HYP

I will go for fishing.

058. ujijə ɖ-ipale-jə ɖi bəɨʈʰe-jə

fish REF-fish-VEVD REF go-VEVD

I will go for fishing.

059. babu ma čonel ɖ-ija

official 1ACC banana REF-give

The official gave me bananas.

288
 289

060. moktčale boji ͂ya ʈʰi-jə

[Link] small/less COP-VEVD

Don’t use even less abusive language.

061. innen ɖi aʈiba a:m əŋ ɖi aʈiba tʰɛitʰɛ

non-Jarawa REF speak mango Jarawa REF speak mango

Non-Jarawas call it mango, Jarawas call it tʰɛitʰɛ.

062. mi ih-ita

1 3-eat

I ate that.

063. mi hi-əjojəba

1 3-see

I saw him/her/that.

064. mi hʷəwə t-ita

1 boar REF-eat

I eat pig.

289
290

065. kiʈaje hʷəwə t-ita

yesterday boar REF-eat

I ate a pig yesterday.

66. kekahəʈə hʷəwə t-ita

tomorrow boar REF-eat

I will eat a pig tomorrow.

67. ənəhoʈəʈa̯ napo t-ita

all fish REF-eat

We all eat fish.

68. čə:w / ǰə:w babu t-ita

wild potato official REF-eat

The official ate wild potato.

69. mi napo ɖ-eʈʰaʈʰuhagle

1 fish REF-fry

I fried the fish.

290
 291

070. əŋ ɖi aʈiba he h-ita

Jarawa REF speak 3 3-eat

Jarawa speaks, “he h-ita”.

071. ka na-kaja pečame?

INTM 2POSS-mother died

Did your mother die?

072. ka na-mummə pečame?

INTM 2POSS-father died

Did your father die?

073. na:g ma-kaja pečame-ma~naɖem

no 1POSS-mother died-NEG

No, my mother has not died.

074. na:g ma-mummə pečame-ma~naɖem

no 1POSS-father died-NEG

No, my father has not died.

291
292

075. hi ma-čaɖɖa ɖə čalahe

3 1POSS-hut LOC stay

He stays at my home.

076. kekahəʈə mi ǰʰulanpətər ɖə bəɨʈʰe-jə

tomorrow 1 [Link] LOC go-VEVD

I will go to Jhulanpatra tomorrow.

077. əminə ɖi bičʰe-jag he-eŋge-jə

a kind of fruit REF forget-ASS 3-bring-VEVD

I left əminə, I will bring it / I left əminə, I want to bring it.

078. ka ŋi dʰaninala ɖə bəɨʈʰe-jə?

INTM 2 [Link] LOC go-VEVD

Will you go to Dhani Nallah?

079. na:g mi ikʰə-gə čalahe-jə

no 1 here-PM stay-VEVD

No, I will stay here.

292
 293

080. mi dʰaninala ɖə bəɨʈʰe-jə

1 [Link] LOC go-VEVD

I will go to Dhani Nallah.

081. onahə:-le

what-VS

What are you doing (why)?

082. mi luwə ɖə omohə

1 there LOC sleep

I will sleep there.

083. mi ʈʰi čawaja allema

1 COP walking come

I came walking.

084. moktčale boi͂ja ʈʰi-jə

[Link] small/less COP-VEVD

Don’t use even less abusive language.

293
294

085. mi ih-ita

1 3-eat

I ate that.

086. onne ʈʰi əŋ

who COP Jarawa

Which Jarawa is there? (Who is there?)

087. jehe ɖi nojčʰe unne ɖə ŋi bəɨʈʰe

sun REF how go LOC 2 go

When will you go? (How far will the sun go when you go?)

088. omohə nojčʰe unne ɖə ŋi bəɨʈʰe

sleep how go LOC 2 go

When will you go? (How long will you sleep and then go?)

089. ka ŋi ɖomo?

INTM 2 fine

Are you fine?

294
 295

090. ka ʈʰi ɖomo

INTM COP fine

Are you fine?

091. onne ɖi aʈiba

who REF speak/say

Who said that?

092. hi onəhə ɖi aʈiba

3 what REF speak/say

What did he say?

093. hiwa čonel onne ɖi h-ija

3ACC banana who REF 3-give

Who gave bananas to him?

094. onne ɖi hiwa čonel (ɖi) h-ija

who REF 3ACC banana (REF) 3-give

Who gave bananas to him?

295
296

095. hi ŋa onəhə ɖi h-ija

3 2ACC what REF 3-give

What did he give to you?

096. ŋi napo onne ɖi h-ija

2 fish who REF 3-give

Who did you give a fish to?

097. ŋi onəhə-le ~ onahə-le ɖ-ičʰe?

2 what-VS / what-VS REF-make /do

Why did you do that?

098. hi jeha͂kəɖə bəɨʈʰe

3 where go

Where did he go?

099. hi jeha͂kəɖə lə:w ɖi ahole?

3 where honey REF collect

Where did he collect honey?

296
 297

100. onne ʈʰi lə:w ɖ-eŋge?

who COP honey REF-bring

Who collected/brought honey?

101. ə:hi onahə-le lə:w ɖi-ahole?

Jarawa (3PL) what-VS honey REF-collect

Why did the Jarawas collect honey?

102. ŋa-ʈʰo (ŋa-paʈʰo) ʈʰi nojčʰe unne

2POSS-arrow COP how far

How long is your arrow?

103. ma-ʈʰo (ma-paʈʰo) ʈʰi talu

1POSS-arrow COP long

My arrow is long.

104. ka-ŋi lə:w wa-ʈʰ-ahole nojčʰe?

INTM-2 honey ??-COP-collect how

How do you collect honey?

297
298

105. luwa-kʰə ɖə ŋi omohə nojčʰe?

there-PM LOC 2 sleep how

How many days will you sleep there?

106. ka-ŋa paʈʰo-le nojčʰe?

INTM-2 arrow-PL how

How many arrows do you have?

107. ka-ŋa lə:w nojčʰe?

INTM-2 honey how

How much honey do you have?

108. ka-ŋi alug nojčʰe ɖ(i)-eŋge?

INTM-2 wild potato how REF-bring

How much wild potato have you got/brought?

109. onne ʈʰi hi-wa wa-mummə?

who COP 3ACC 3POSS-father

Who is his father?

298
 299

110. onne ʈʰi hi-wa wa-kaja?

who COP 3ACC 3POSS-mother

Who is his mother?

111. hi onne ɖi waja?

3 who REF [Link]

Whose child is he/she?

112. luwa-kʰə ɖə onne ɖi čonel-le?

there-PM LOC who REF banana-PL

Whose bananas are there?

113. luwa-kʰə ɖə onne ɖi paʈʰo-le?

there-PM LOC who REF arrow-PL

Whose arrows are there?

114. ŋi nojčʰe unne hʷatapo ɖə hilule bəɨʈʰe?

2 how go [Link] LOC [Link] go

How will you go from Hilule to Hʷatapo?

299
300

115. ŋi nojčʰe unne baɽabalu bəɨʈʰe?

2 how go [Link] go

How will you go to Baɽabalu?

116. hi-itatʰe onne ɖi ŋa napo h-ija

3-ask who REF 2 fish 3-give

He asked, “Who had given fish to you?”

117. ənətə:hə:

sit

Sit.

118. tokagʈʰe

stand

Stand.

119. ikʰə-gə ɖə ənətə:hə:

here-PM LOC sit

Sit here.

300
 301

120. čaɖɖa ɖə bəɨʈʰe-jə

hut LOC go-VEVD

Go to the hut?

121. mahə/mahe čonel-le ɖi oʈʰejə

[Link] banana-PL REF bring

Bring bananas for me.

122. kʰo paʈʰo

hold arrow

Take/hold the arrow.

123. ma napo ɖ-ija

1ACC fish REF-give

Give me a fish.

124. ullele ɖə bəɨʈʰe naɖem

sea LOC go NEG

Don’t go in sea water.

301
302

125. ullele ɖə bəɨʈʰe-ma

sea LOC go-NEG

Don’t go in sea water.

126. le:b ɖə bəɨʈʰe naɖem

road LOC go NEG

Don’t go on the road.

127. ma-ʈʰo (ma-paʈʰo) ʈʰe-eŋge naɖem

1POSS-arrow COP-bring NEG

Don’t take my arrow.

128. əččəle ɖi wa-tʰulə-wə

child REF CAUS-sleep-VEVD

Make the child sleep.

129. əččəlewə kag ɲoɲo-wa

child breast feed-VEVD

(Breast)feed the child.

302
 303

130. ikʰə-hʷə-ɖə bəɨʈʰe naɖem wəji-koji ərugǰəji

here-steam-LOC go NEG some crocodile

Don’t go in this stream, a crocodile is here.

131. hiwa ŋewo

3ACC call

Call him.

132. babu-i ŋewo

official-EM call

The official is calling you.

133. magʈʰe naɖem

[Link] NEG

Don’t wake me up.

134. dipu oʈʰejə

PN wake

Wake Dipu up.

303
304

135. ənəʈʰučə

go away

Go away from here.

136. əhapela ʈʰuhumə

run INTSF

Run fast/faster.

137. m-ilebua

1-unfasten

Unfasten me.

138. m-agle naɖem

1-fasten NEG

Don’t fasten me.

139. bəɨʈʰe

go

Let’s go.

304
 305

140. əha bəɨʈʰe

okay go

Okay, go.

141. ma iŋ-e h-ija

1ACC water-EM 3-give

Give me water/Would you mind passing water, please?

142. ma iŋ-e ɖ-eŋge-jə

1 water-EM REF-bring-VEVD

Bring water for me/Give me water.

143. dipu itatʰe čaɖɖa ɖə bəɨʈʰe

PN say hut LOC go

Dipu said to go to the hut.

144. dipu itatʰe baɽabalu ɖə bəɨʈʰe naɖem

PN say [Link] LOC go NEG

Dipu said not to go to Baɽabalu.

305
306

145. hi-wa waja ɖi aʈiba waja-ɖi kag ɲoɲo-wa

3ACC [Link] REF ask 3POSS-child-REF breast feed-VEVD

Ask his mother to feed her child.

146. čew m-aheapa naru čonel igeǰi

PN 1-say PN banana cut

Cew told me that Naru would cut banana.

147. naru m-aheaba mi ləkəralunɖa ɖə bəɨʈʰe

PN 1-say 1 [Link] LOC go

Naru said to me, “I will go to Ləkəralunɖa.”

148. dipu m-aʈiba ka ŋi hʷatapo ɖə bəɨʈʰe

PN 1-say INTM 2 [Link] LOC go

Dipu asked me, “Are you going to Hʷatapo?”

149. sapan hi-wa aʈiba ka gəčʰehə ŋi napo ɖi i:tə:-hə

PN 3ACC say INTM before now 2 fish REF eat-HYP

Swapan asked him, “Have you had fish today?”

306
 307

150. dipu hi-wa aʈiba onne hʷəwə ɖ-aɪkʰwa

PN 3ACC say who boar REF-kill

Dipu asked him, “Who killed the wild boar?”

151. ka ŋi ləkəralunɖa ɖə bəɨʈʰe?

INTM 2 [Link] LOC go

Are you going to Ləkəralunɖa?

152. ka-ŋi kekahəʈa olleg ɖə bəɨʈʰe?

INTM-2 tomorrow [Link] LOC go

Will you go to Olleg/Ləkəralunɖa tomorrow?

153. ka-ŋi jakeka howa?

INTM-2 day after tomorrow go for hunting

Will you go for hunting day after tomorrow?

154. ka-ŋi gəčʰehə lə:w ahole?

INTM-2 before now honey collect

Did you get honey today?

307
308

155. ka-ŋi gəčʰehə lə:w ɖ-iliŋ?

INTM-2 before now honey REF-see

Did you see honey today?

156. ka-ŋi gəčʰehə naɖo t-ita-tʰe?

INTM-2 before now food REF-eat-NVEVD

Have you had food today?

157. ka-ŋi inɖəla?

INTM-2 bathe

Do you take bath daily?

158. ka-ŋi hʷəwə t-ita?

INTM-2 wild boar REF-eat

Do you eat a wild boar?

159. ka-ŋi napo t-ita ʈʰi-na čew

INTM-2 fish REF-eat COP-2 good

Having had fish, is it good (are you feeling good)?

308
 309

160. ka n-aɪkʰwaʈʰa-le nojčʰe?

INTM 2POSS-brother-PL how

How many brothers do you have?

161. ka ŋ-anona͂ge?

INTM 2POSS-cloth

Do you have cloth?

162. ka ŋa-kaŋgapo

INTM 2POSS-cloth

Do you have cloth?

163. lijə paʈʰo

here arrow

This is an arrow/Here is an arrow.

164. luwahe a:w

there bow

There is a bow/That is a bow.

309
310

165. lijə ma-paʈʰo-le

here 1POSS-arrow-PL

These are my arrows.

166. luwa əŋ-ahaŋ wa-ʈʰo-le

there jarawa-all 3POSS-arrow-PL

Those are all Jarawas’ arrows/There are all Jarawas’ arrows there.

167. ka ŋa-ʈʰo?

INTM 2POSS-arrow

Where is your arrow?

168. ma-ʈʰo ma-čaɖɖa

1POSS-arrow 1POSS-hut

My arrow is at my hut.

169. kohəhoʈʰo biskuʈ h-ija

all biscuit 3-give

Give biscuits to all/each one/everyone.

310
 311

170. ma waŋ napo ɖi h-ija

1ACC some fish REF 3-give

Give me a few fish.

171. kəhəhoʈʰo ɖi čonel enoloja

all REF banana get

All/everyone should get bananas.

172. lijə paʈʰo ɖi boi͂ja

here arrow REF small

This is a short arrow.

173. luwahe paʈʰo ɖi talu

there arrow REF long

That is a long arrow.

174. wəhi napowə

many jarawa fish

Many Jarawas have gone for fishing.

311
312

175. wəhi napo tahile

many jarawa fish go

Many Jarawas have gone for fish/fishing.

176. paʈʰo ʈʰi boi͂ja

arrow COP small

The arrow is short.

177. paʈʰo ʈʰi talu

arrow COP long

The arrow is long.

178. lijə taŋ ʈʰi boi͂ja

this tree COP small

This tree is short.

179. luwa taŋ talu talu ʈʰuhuma

there tree long long very

That tree is long/very long.

312
 313

180. tawle ʈʰi orgontʰo

PN COP short

Tawle is short.

181. topo ʈʰi orkalaŋ

PN COP tall

Topo is tall.

182. talu ʈʰi huʈʰu

PN COP fat

Talu is fat.

183. tingle ʈʰi telo

PN COP thin

Tingle is thin.

184. ačərəwa palahe ʈʰi huʈʰu čʰedali ʈʰi aiɖəhʷame

PN earlier COP fat now COP thiner

Acharawa was earlier fat, but now she is thinner.

313
314

185. čuɽ ʈʰ-ipilhunu= ʈʰi ipilhunu

PN COP- COP

Cuɽ is one-handed.

186. raǰu ʈʰ-ukʰunu = ʈʰi ukʰunu

PN COP-lame COP lame

Raju is lame.

187. hi ʈʰ- iʈʰəŋgem = ʈʰi iʈʰəŋgem

3 COP- deaf COP deaf

He is deaf.

188. hi ʈʰ-ənkegili = ʈʰi ənkegili

3 COP- dumb COP dumb

He is dumb.

189. čaɖɖa ɖi ʈʰ-olkalaŋ

hut REF COP-high

The hut is high.

314
 315

190. ʈʰinon ʈʰuhumə

hillock INTSF

The hillock is high.

191. ʈʰinon ʈʰi boi͂ja/waja/ogčan

hillock COP small/child/short

The hillock is small.

192. le:b ʈʰi huʈʰu

road COP fat

The road is wide.

193. itʃel/le:b boi͂ja

path/road small

The road is narrow.

194. əh-ero-gə

3-angry-EVD

He is angry.

315
316

195. ən-ero-gə

2-angry-EVD

You are angry.

196. m-ero

1-angry

I am angry.

197. ma-ʈʰo (ma-paʈʰo) ɖi tingle wa-ʈʰo (wa-paʈʰo) ʈʰ-obijagune

1POSS-arrow REF PN 3POSS-arrow COP-long

My arrow is longer than Tingle’s arrow.

198. ma-ʈʰo (ma-paʈʰo) ɖi tingle wa-ʈʰo (wa-paʈʰo) ɖi-ičʰebaune

1POSS-arrow REF PN 3POSS-arrow REF-long

My arrow is as long as Tingle’s arrow.

199. ugə momo wɛ ʈʰ-učəhə

PN PN COMPM COP-short

Ugə is shorter than Momo.

316
 317

200. momo ugə wɛ ʈʰ-orkalaŋ

PN PN COMPM COP-tall

Momo is taller than Ugə.

201. topo orpeʈʰaʈʰihə

PN tallest

Topo is the tallest person.

202. m-a:w likʰə wɛ ʈʰ-obijagune

my-bow here COMPM COP-big

My bow is bigger than his.

203. kopel-a-a:w ʈʰi huʈʰuʈʰuhumə

PN-3POSS-bow COP biggest

Kopel’s bow is the biggest one.

204. luwə do:jə ʈʰa m-aja

there girl COP 1POSS-child

The girl who is there is my child. (That girl is my child.)

317
318

205. luwə əʈʰwəgune m-aikʰwaʈə

there sit 1POSS-brother

That who is sitting there is my younger brother.

206. m-a:w ʈʰi-tormono

1POSS-bow COP-shortest

My bow is the shortest one.

207. oʈʰoje-le ipale-jə

lady-PL fish-VEVD

Ladies are fishing with the fishing net.

208. hʷəwə ɖi-ahapela əŋ ɖi ihi-aikʰwa

wild boar REF-run Jarawa REF 3-kill

The wild boar was running and Jarawa killed that. (cf. 23)

209. əŋ ɖi-aʈiba čajwaj ummə kɨjəʈə unnə

Jarawa REF-say cayway father forest come back

Jarawa says, Cayway’s father came back from the forest.

318
 319

210. mi-mugə boi͂ja-g

1POSS-beard small-EVD

My beard is small.

211. ma iŋ-e ɖ-ija

1ACC water-EM REF-give

Give me that water (seen).

212. ma iŋ-e h-ija

1ACC water-EM 3-give

Give me some water (unseen).

213. ikʰə-gə ənətəhə

here-PM sit

Sit here.

214. čonel ɖi ma h-ija

banana REF 1ACC 3-give

Give me the banana.

319
320

215. ma čonel ɖ-ija

1ACC banana REF-give

Give me the banana.

216. ma čonel ɖi h-ija

1ACC banana REF 3-give

Give me the banana.

217. ma čonel ɖ-eŋge-jə

1ACC banana REF-bring-VEVD

Bring the banana to me.

218. čonel ɖi ma eŋge-jə

banana REF 1ACC bring-VEVD

Bring the banana to me.

219. babu-i ma čonel he-eŋge-jag

officer-EM 1ACC banana 3-bring-ASS

The official brought a banana for me.

320
 321

220. babu-i ma he-eŋge-jag

officer-EM 1ACC 3-bring-ASS

The official brought a banana for me.

221. kekahəʈə mi hʷəwə tahilə-hə

tomorrow 1 boar go-HYP

Tomorrow I will go for hunting.

222. kekahəʈə mi kɨjə tahilə-hə

tomorrow 1 forest go-HYP

Tomorrow I will go for hunting.

223. kekahəʈə mi allema-hə

tomorrow 1 come-HYP

I will come tomorrow.

224. əhi onahə-le lə:w ɖi-ahole

Jarawa what-VS honey REF-collect

Why did the Jarawas collect honey?

321
322

225. lə:w ahole ih-ita-b

honey collect 3-eat-NVEVD

We collected honey and eat that.

226. ma iŋ h-ija mi inčo-hə

1ACC water 3-give 1 drink-HYP

Give me water, I will drink.

227. mi inčo-hə ma iŋ ɖ-ija

1 drink-HYP 1ACC water REF-give

Give me water, I will drink.

228. gəčʰəhə mi inčo-ʈʰe

before now 1 drink-NVEVD

I took water earlier (before now).

229. ma-ʈʰo ʈʰučʰəʈəgə

1POSS-arrow where

Where is my arrow?

322
 323

230. ŋa-ʈʰo ʈʰučʰəʈəgə

2POSS-arrow where

Where is your arrow?

231. ma-ʈʰo ɖi ma-čaɖɖa ɖə

1POSS-arrow REF 1POSS-hut LOC

My arrow is at my hut.

232. mi napo ɖi ŋi h-ija

1 fish REF 2 3-give

I gave you the fish.

233. *əččəlewə ʈʰi boi͂ja

child COP small

The child is small.

234. jehe čokilanji

sun set

The sun set.

323
324

235. jehe ɖi čokilanji

sun REF set

The sun set.

236. mi-mugə huʈʰu-gə

1POSS-beard big-EVD

My beard is big.

237. * mi-mugə huʈʰu-gə ʈʰuhumə

1POSS-beard big-EVD very

My beard is very big.

238. *mi-mugə huʈʰu ʈʰuhumə-gə

1POSS-beard big very-EVD

My beard is very big.

239. hi ejagija bəɨʈʰe-jə

3 where go-VEVD

Where has he gone?

324
 325

240. hi ejagiya bəɨʈʰe-jag

3 where go-ASS

Where did he go?

241. ə-ʈʰi = əŋ-ʈʰi lə:w ɖ-eŋge-jə

Jarawa-COP honey REF-bring-VEVD

Who has brought honey?

242. onne ʈʰi lə:w ɖ-eŋge-jə

who COP honey REF-bring-VEVD

Who has brought honey?

245. ma-ʈʰo talu-gə

1POSS-arrow long-EVD

My arrow is long.

246. m-ajigǰija

1-leave

Leave me.

325
326

247. m-aɸle naɖem

1-fasten NEG

Don’t fasten me.

248. ma iŋ-e ɖ-eŋge-jə

1ACC water-EM REF-bring-VEVD

Bring my water.

249. napo ɖi waŋ ma h-ija

fish REF some 1ACC 3-give

Give me some fish.

250. mi ikʰə-gə m-ətəhə

1 here-PM 1-sit

I am sitting here.

251. mi-čʰe ənətəhə

1 sit

I will sit.

326
 327

252. hʷəwə ɖi omohə

wild boar REF sleep

The wild boar is sleeping.

253. hʷəwə ɖi oho

wild boar REF sit

The wild boar is sitting.

254. hʷəwə ɖi ahekana

wild boar REF hide

The wild boar is hiding.

256. enme ʈʰijənlem uwajaʈʰə kɨjə tahilə

PN PN both forest go

Enme and Thiyənlem have gone to the forest (for hunting).

257. kekahəʈə čajwaj ummə səmpa uwajaʈʰə ǰagʈʰujə-hə

tomorrow cayway father PN both come-HYP

Tomorrow Cayway’s father and Sampan will come.

327
328

258. ka-ŋi kɨjə tahilə nojčʰe?

INTM-2 forest go how many

How many times did you go for hunting?

259. raja ʈʰučʰəʈəgə?

PN where

Where is Raja?

260. raja ka ʈʰučʰəʈəgə?

PN INTM where

Where is Raja?

261. *ka raǰa ʈʰučʰəʈəgə?

INTM PN where

Where is Raja?

262. čajwaj ʈʰučʰəʈəgə

PN where

Where is Cayway?

328
 329

263. čajwaj ka ʈʰučʰəʈəgə

PN INTM where

Where is Cayway?

264. *ka čajwaj ʈʰučʰəʈəgə

INTM PN where

Where is Cayway?

265. mi enme ə͂hə͂-nja-ʈʰe

1 PN with-go

I went with Enme.

266. onne ɖaŋ (ɖi-əŋ) njačʰe allema

who REF-Jarawa with come

Who did you come with?

267. čaɖɖa ɖə italəhə

hut LOC inside

Inside the hut.

329
330

268. kaŋgapo kəʈʰijə

cloth tear

The cloth is/was torn.

269. dewa kaŋgapo ɖi ikʰə-ʈʰə

PN cloth REF cut-NVEVD

Dewa tore the cloth.

270. əminə he-epohə mijə allema

a kind of fruit 3-forget 1 come

I forgot the fruit, I am coming.

271. əminə he-epohə mi eŋge-jə

a kind of fruit 3-forget 1 bring-VEVD

I forgot the fruit, I am bringing (it).

272. onne ʈʰi əŋ taŋ odehe

who COP Jarawa tree cut

Which Jarawa cut the tree?

330
 331

273. əččəle wa-tʰulə

child CAUS-sleep

Make the child sleep.

274. əččəle čonel ɖi wa-le-jə

child banana REF CAUS-eat-VEVD

Feed the banana to the child.

275. əččəle woǰoʈʰo inčo-wa

child CAUS drink-VEVD

Make the child drink.

276. əččəle woǰoʈʰa taŋ ɖə čapʈʰe-jə

child CAUS tree LOC climb-VEVD

Make the child climb on the tree.

277. əhi ʈʰi bəɨʈʰe

Jarawa COP go

All the Jarawa have gone.

331
332

278. ə-ʈʰi (əŋ-ʈʰi) bəɨ:

Jarawa go

Let us go.

279. əččəle ɖi wa-le-jə

child REF CAUS-eat-VEVD

Feed the child.

332
 333

Appendix – III

Distribution of verbal suffixes on verbs

S. No. Jarawa verbs -jə -jag -bə -ʈʰe -wa -hə

1. očʰenə: / očʰegə X X X X X X

2. inčo X X X √ √ √

3. ita X X √ √ -- √

4. čawaja X X X X X X

5. əhapela X X X X X X

6. gəgapa / gəgaba X X X X X X

7. eʈʰaʈuhagle √ √ X X X --

8. ipine √ √ X X X --

9. ənane √ √ X X X --

10. oʈʰaj-le √ √ X X X X

11. tahilə √ X X X X √

12. bəɨʈʰe √ √ X X X √

13. omohə X X X X X √

14. tʰulə X X X X X X

15. əjojəba X X X X X X

16. iʈʰəhəʈʰe √ √ X X X --

333
334

17. uʈʰiʈʰigu iʈʰəhəʈʰe √ √ X X X X

18. ənija √ -- X X X X

19. ero X X X X X X

20. əɖə -- -- -- -- -- --

21. oj ͂i ͂ʈʰege √ -- X X X X

22. ənəjale √ -- X X X X

23. ətʰokagʈʰe √ -- X X X X

24. čalahe √ X X X X X

25. ənobelijahije √ -- X X X X

26. aʈiba X X X X X √

27. ija X √ X -- X X

28. ičilo X X X √ √ √

29. ən-oɖə-kali √ -- X X X √

30. nam ɖ-oɖəhe √ √ X X X --

31. čaɖɖa-le √ X X X X X

32. ahowa X X X X X X

33. aigɖəwa -- -- -- -- -- --

34. əčʰaʈʰajə -- -- -- -- -- --

35. huwaǰi √ X X X X X

36. apine √ -- X X X X

37. inijačʰe √ -- X X X --

334
 335

38. ən-ohelə √ -- X X X X

39. pela-jitʰə -- -- -- -- -- --

40. ipine √ X X X X X

41. allema X X X X X √

42. əne͂wa -- -- -- -- -- --

43. wanna X X X X X X

44. əniji: √ -- X X X X

45. ən-agulemelegune √ X X X X X

46. oʈʰa ən-ijelaŋgə -- -- -- -- -- --

47. ulleləɖa̯ X X X X X X

48. hʷaɖə X X X X X X

49. eŋge √ √ X X X --

50. kʰo: X X X X X X

51. ən-ə:təhə X X X X X X

52. tokəʈʰe √ √ X X X --

53. aɸle √ √ X X X --

54. ilibua X X X X X X

55. napo ɖ-aye √ -- X X X X

56. buʈʰə tʰame √ X X X X X

57. ən-aŋgtʰə -- -- -- -- -- --

335
336

58. wəwə-le √ X X X X X

59. inɖə X X X X X X

60. ačʰu √ √ X X X X

61. aŋgijačʰuə X X X X X X

62. əbulə: X X X X X X

63. əna əŋgale √ -- X X X X

64. ʈʰuhəb-le √ -- X X X X

65. ən-ənpəʈʰa √ √ X X X X

66. ekane √ -- X X X --

67. eŋge √ √ X X X --

68. čapʈʰe √ √ X X X --

69. əničajmə -- -- -- -- -- --

70. ǰagijə -- -- -- -- -- --

71. eweewe X X X X X X

72. ekka X X X X X X

73. waʈʰepa-le √ √ X X X X

74. napowə X X X X X X

75. ikʰə-wə ɖə aikʰwəkʈʰe -- -- -- -- -- --

76. napo tahile √ X X X X √

77. unnə X X X X X X

336
 337

78. pijalɛ X X X X X X

79. howa X X X X X X

80. kijə tahile √ X X X X √

81. aikʰwa X X X -- X X

82. uɨgəle -- X X X X X

83. waɽa X X X X X X

84. ihune √ X X X X X

85. ən-ahopaho-le √ X X X X X

86. oʈʰa inijəla -- -- -- -- -- --

87. helahoə: -- -- -- -- -- --

88. ičʰeba X X X X X X

89. ənoɖə h-igəʈʰe √ -- X X X X

90. ənotoha X X X X X X

91. əhujəba X X X X X X

92. əŋgogə -- -- -- -- -- --

93. pakǰawaǰi √ -- X X X X

94. ehaʈo-jə √ -- X X X --

95. ən-aholagʈʰi √ X X X X X

96. ən-ahotʰwag X X X X X X

97. ilebuwə X X X X X X

337
338

98. ənihʷeɖə X X X X X X

99. ipinunu X -- X X X --

100. ən-ahuglu X -- X X X √

101. tʰuwə X X X X X X

102. əɖe √ X X X X --

103. ən-ahela-le √ -- X X X --

104. čowaǰi √ X X X X X

105. boi͂ja-gə X X X X X X

106. əh-anekʰʷeɖaǰi √ -- X X X X

107. ənejom X X X X X X

108. ən-ahimudijəʈʰe √ -- X X X X

109. ənhowa X X X X X X

110. ha-wočewoče-le √ -- X X X X

111. ən-igine √ -- X X X X

112. eneje √ -- X X X X

113. ən-ohoʈʰahe √ √ X X X X

114. pado he-ehʷag X X X X X X

115. hehehe X X X X X X

116. apa X -- X X X X

117. iwəčə X X X X X X

338
 339

118. əhubə X X X X X X

119. ičʰe √ √ X X X X

120. aʈʰohag -- -- -- -- -- --

121. itʰə √ X X X X X

122. hewojabə X X X X X X

123. inɖə X X X X X X

124. əpa X -- X X X X

125. tʰui-le √ X X X X X

126. taŋg-le √ X X X X X

127. ačʰile-le √ -- X X X X

128. onodəhe √ X X X X X

129. ənimunəǰi √ X X X X X

130. untopine √ X X X X X

131. ejaja -- -- -- -- -- --

132. uluhə patʰo X X X X X X

133. bod X X X X X X

134. hagulume -- -- -- -- -- --

135. ʈʰuhəgle √ X X X X X

136. hatiti √ X X X X X

137. ɲono X X X √ √ --

339
340

138. ənihəd -- -- -- -- -- --

139. paʈʰo ɖə-ekəwaja -- -- -- -- -- --

140. čcaɖɖa ɖ-epoha X X X X X X

141. ənəʈʰuči √ √ X X X X

142. ugikəəʈʰe √ -- X X X X

143. ʈʰugeǰa √ -- X X X X

144. ipo-le √ -- X X X X

145. əna-uɖɖu-le √ -- X X X X

146. inə-gə piʈȹi -- -- -- -- -- --

147. inə-gə čew -- -- -- -- -- --

148. oʈʰahe √ √ X X X X

149. huiyahuija -- -- -- -- -- --

150. iŋ ičʰe √ √ X X X X

151. iŋ əčʰe tahile √ X X X X √

152. iliŋ X X X X X X

340
 341

Appendix IV: Maps

Map 1: Andaman and Nicobar Islands

'-

ANDAMAN &

Source: [Link]

341
342

Map 2: Jarawa Reserve Area

Source: Sekhasaria, Pankaj and Pandya, V. 2010. The Jarawa Tribal Reserve
Dossier. UNESCO

342