The decay of ergativity in Kurmanci
Margreet Dorleijn
Tilburg University Press
This study wants to give an impression of the influence of Turkisch on the Kurmanci dialect of Kurdish. More in particular, it intends to offer a treatment of split ergativity in Kurmanci.The aim is to show, on the basis of emerging patterns of variation, how varieties of Kurmanci are on the road to becoming nominative/accusative altogether. One of the main findings is that, besides internal tendencies, contact with Turkish may be a determing factor in this process.
Studies in Multilingualism is published under the editorship of
Prof. Rene Appel, University of Amsterdam
Prof. Guus Extra, Tilburg University
Prof. Koen Jaspaert, University of Leuven
Prof. Ludo Verhoeven, Nijmegen University
Table of contents
Acknowledgements / ix
List of abbreviations used / xi
Map I / xiii
Map II / XV
1. Introduction / 1
1.1. Structural change: the decay of ergativity / 3
1.2. The organisation of this study / 4
2. The present study / 6
2.1. Notes on the classification and distribution of Kurdish / 6
2.2. The Kurmanci speaking community / 7
2.2.1. The Kurdish community in Turkey / 8
2.2.2. Language contact in Turkey / 8
2.2.3. Turkish as the dominant language / 9
2.2.4. Domains of usage / 10
2.3. Defining a norm / 10
2.4. Diyarbakir / 12
2.5. Working hypotheses and research questions / 15
2.6. Types of data and how they were collected / 17
2.6.1. Spontaneous data / 18
2.6.2. Elicited data (translation tasks) / 20
2.6.2.1. Structural conditions / 20
2.6.2.2. Turkish as a contact language / 21
2.6.2.3. Why a translation task / 23
2.6.3. Informants and interviewers / 24
2.7. Data processing: transcription / 28
2.7.1. Spontaneous data / 28
2.7.2. Elicited data / 28
3. Kurmanci: a brief grammatical outline / 31
3.1. Alphabet and phonology / 31
3.1.1. Alphabet and phoneme inventory / 31
3.1.2. The phonological structure of syllables and words / 33
3.2. Word order / 33
3.2.1. Unmarked order of constituents / 34
3.2.2. Structure of constituents / 34
3.3. Nominal categories: gender, number, specific indefiniteness suffix, izafet, case / 35
3.4. Verbal complex / 39
3.4.1. Infinitives and dual verbal root system / 39
3.4.2. Nonfinite forms / 39
3.4.3. The inflected verb / 39
3.5. Grammatical relations / 42
3.5.1. Passive / 42
3.5.2. Causative / 43
3.5.3. Reflexive and reciprocal pronouns / 43
3.6. Subordinate clauses / 44
3.6.1. Relative clauses / 44
3.6.2. Complement clauses / 45
3.6.3. Adverbial clauses / 45
4. Turkish-Kurdish contact phenomena / 47
4.1. Lexical borrowing / 47
4.1.1. Content words / 48
4.1.2. The borrowing of verbs / 49
4.1.3. Function words and discourse markers / 52
4.2. Slight morphological interference / 54
4.2.1. m-doublets / 54
4.2.2. Conditional marker se / 54
4.2.3. Turkish possesive suffixes / 55
4.2.4. Overextension of Kurdish possessive constructions / 55
4.3. Slight syntactic interference / 56
4.4. Loss and leveling of morphology / 57
4.4.1. Loss of gender distinction / 57
4.4.2. Pronouns and demonstratives / 58
4.4.3. The izafet / 58
4.4.4. Case / 60
4.4.5. Morphological leveling / 61
4.5. Morphosyntactic change / 61
4.5.1. Directionals and IO’s / 62
4.5.2. Null-subjects / 64
4.5.3. Case-marked definite objects and indefinite objects without case-marking / 65
4.5.4. Breakdown of the ergative system / 67
4.6. Summary and conclusions / 69
5. Ergativity in Kurmanci / 71
5.1. Ergativity / 72
5.1.1. Syntactic vs. morphological ergativity / 72
5.1.2. Classification of ergative languages / 73
5.2. The diachronic development of ergativity / 74
5.2.1. Ergativity - a marked phenomenon / 75
5.2.2. The diachronic development of ergativity in Iranian languages / 75
5.3. Two proposals concerning the structural nature of ergativity / 77
5.3.1. The Obligatory Case Parameter / 78
5.3.2. Unaccusativity / 80
5.4. Ergativity in Kurmanci / 82
5.4.1. Split-ergativity / 82
5.4.2. Ergative case-marking / 83
5.4.3. Ergative verbal agreement / 85
5.4.4. Morphological ergativity / 85
5.4.5. The distribution of ergative morphology / 89
5.4.6. Agreement pattern with reflexives and reciprocals / 90
5.4.7. Indirect transitives / 91
5.4.8. Fronting of the direct object / 91
5.5. Analysis / 93
5.5.1. Structural position of the S-argument / 94
5.5.2. ABS = NOM morphologically / 94
5.5.3. ABS is obligatory in finite clauses / 95
5.5.4. ABS is assigned in a manner different from NOM / 97
5.5.5. Verbal agreement dependent on ABS? / 97
5.5.6. Is there evidence for the different nature of verb endings
in past transitive structures in Kurmanci? / 99
5.5.7. Summary / 103
5.6. The participial nature of past verbal morphology / 103
5.6.1. Is past-root morphology case-absorbing? / 104
5.6.2. Infinitives in Kurmanci / 107
5.6.3. Case-absorbing past root morphology.
Some evidence from other Iranian languages / 109
5.6.4. A speculative intermezzo on the role of infinitives
in the history of Iranian languages / 112
5.6.5. Summary / 113
5.7. How is OBL assigned to the subject and ABS to the object / 114
5.6. The participial nature of past verbal morphology / 103
5.6.1. Is past-root morphology case-absorbing? / 104
5.6.2. Infinitives in Kurmanci / 107
5.6.3. Case-absorbing past root morphology.
Some evidence from other Iranian languages / 109
5.6.4. A speculative intermezzo on the role of infinitives
in the history of Iranian languages / 112
5.6.5. Summary / 113
5.7. How is OBL assigned to the subject and ABS to the object / 114
6. From ergative-absolutive to nominative-accusative? / 116
6.1. Deviant patterns / 116
6.1.1. The unexpected case-agreement patterns / 117
6.1.2. Agreement with OBL subjects (OBL DIR SA and OBL OBL SA) / 118
6.1.3. Agreement with OBL objects (OBL OBL, OA and DIR OBL OA) / 119
6.1.4. Non-agreement with DIR objects in OBL DIR constructions
(OBL DIR NA and OBL DIR SA) / 120
6.1.5. Non-agreement with DIR subjects in DIR OBL constructions
(DIR OBL NA and DIR OBL OA) / 121
6.1.6. DIR DIR constructions (DIR DIR OA, DIR DIR NA, DIR DIR SA) / 121
6.1.7. Summary / 122
6.1.8. OBL OBL NA / 123
6.1.9. DIROBLSA / 126
6.2. Analysis / 127
6.2.1. Third person singular (3s) / 128
6.2.2. Loss of OBL case-marking on singular NP’s / 128
6.2.3. Case-marking of definite direct objects / 129
6.2.4. DIR 3s: unmarked for person / 130
6.2.5. Second person singular (2s) / 131
6.2.6. A unified form for 2s and loss of the verbal agreement morpheme / 132
6.2.7. OBL renders person features inaccessible / 133
6.2.8. Third person plural (3p) / 133
6.2.9. 3p verbal agreement: number only / 134
6.2.9.1. 3p objects in OBL OBL structures: avoidance of
ambiguity and direct object marking / 136
6.3. Summary / 137
6.4. Marking of definite direct objects, movement of
arguments, null-subjects and the consequences / 137
6.4.1. The marking of definite direct objects / 138
6.4.2. Movement of the direct object / 140
6.4.3. Placement of adverbial constituents between the verb and the direct object / 143
6.4.4. Null subjects / 146
6.4.5. Summary / 149
7. Conclusions and suggestions for further research / 151
7.1. Conclusions / 151
7.2. Further research / 153
Appendices / 155
1. The long translation task / 155
2. The short translation task / 159
3. A sample of Hasan’s text / 161
4. A sample of Izzet’s text / 165
5. A sample of Sadiks text / 170
References / 175
Samenvatting in het Nederlands / 181
ACKNOWLEDGEMENTS
I wish to thank a number of people and institutions that have contributed to this study.
The University of Amsterdam and the Netherlands Organisation for Scientific Research (NWO) have provided financial support that made a fieldwork trip to Turkish Kurdistan possible. The Faculty of Arts of the University of, Amsterdam moreover granted me a one-trimester sabbatical.
I am greatly indebted to my supervisor, Pieter Muysken, for his inspiration and encouragement and for his patience with my idiosyncratic way of working.
To Rik Boeschoten, Reineke Bok-Bennema, Joris Croon, Michiel Leezenberg, Yaron Matras, Francine Swets and Amoud Vrolijk I am grateful for valuable discussions and comments on earlier versions of this study.
Hans Broekhuis and Joop Veld have patiently tried to familiarize me with all sorts of Government and Binding Theory intricacies. I thank them both.
I have profited greatly from Ab de Jongs knowledge of Old and Middle Iranian.
I am very much indebted to Joris Croon, Rahime Henden, Zeynel Abidin Kizilyaprak, Michiel Leezenberg and in particular Sadik Ziypak for their help in collecting the speech data. Martin van Bruinessen gave me disposal over Kurmanci speech samples which he had collected earlier.
Frans Janssen has drawn the maps in this book, the first one of which has appeared earlier in Martin van Bruinessen’s Agha, Shaikh and State (1992), and appears here with the kind permission of the author. The second map was drawn for this book, and on very short notice indeed. I am greatly indebted for that.
The tedious and difficult task of transcribing the spontaneous speech data was made much lighter through the help of Halid and Ferat (who may not care to have their family name mentioned) and Joris Croon.
Most of all, I am indebted to the people that provided the speech data. Sadik Ziypak is one of them. The others prefer to remain anonymous. At times some risk was involved in co-operating in the data collection sessions, but the fact that their language was the object of scientific study induced people to take that risk, even though most of them would question the actual benefit of such a study. I hope they are not disappointed by the result.
Finally, I thank Amoud, Steven and Alexandra Vrolijk for supporting me and distracting me, respectively, and for complying with quite a few wifeless or motherless Sunday afternoons.
Chapter 1: Introduction
This study tries to give an impression of the influence of Turkish on the Kurmanci dialect of Kurdish. More in particular, it intends to offer a treatment of split ergativity in Kurmanci. The aim is to show, on the basis of emerging patterns of variation, how varieties of Kurmanci are on the way to becoming nominative/accusative altogether. One of the main findings is that, besides internal tendencies, contact with Turkish may be a determining factor in this process.
Kurmanci is spoken by approximately 12 million people living mainly within the boundaries of the Turkish Republic. Turks and Kurds have a long history of (not always peaceful) coexistence, and many Kurds are bilingual in Turkish and Kurdish.
The linguistic effects of contact with Turkish on Kurmanci have, to my knowledge, not yet been the subject of extensive and systematic research. Two studies on Turkish-Kurdish language contact are Bakaev (1977) and (of much smaller scope) Zajaczkowski (1983). These studies are concerned mainly with the borrowing of Turkish content words into Kurdish. Biasing (1995) offers a treatment of the borrowing of Kurdish lexical items in Turkish. Bulut (forthc.) and Dorleijn (1996 and forthc.) include some observations on borrowed Turkish function words and bound morphemes in Kurmanci and also, in a cursory fashion, a description of some morpho-syntactic phenomena in Kurmanci that may be attributed to language-contact.
The study of Turkish-Kurdish language-contact is interesting in more than one respect. Firsdy, Turkish in contact has, at least from a morpho-syntactic point of view, mainly been studied in (older or newer) immigrant situations (see, e.g., Johanson, 1992 and references there), where Turkish is the language of the immigrant community. In these situations Turkish is dominated by another language, mosdy an Indo-European language, and receives elements from this language through borrowing (or copying, in the terminology of Johanson, 1992). In Turkish-Kurdish language contact the situation is almost the reverse: Turkish is the dominant language, in particular since the advent of the Turkish Republic with its stem Turkification policies.
Kurdish, an Indo-European language, is the receiving party. The situation is not entirely the reverse, obviously, for neither the Turkish, nor the Kurdish community can be considered an immigrant community.
Secondly, the two languages involved are not only genetically but also typologically highly distinct. Turkish is, for example, transparent from a morphological point of view. The syntactic function of each element is …
Margreet Dorleijn
The decay of ergativity in Kurmanci
Tilburg University
Tilburg University Press
Studies in Multilingualism 3
The decay of ergativity in Kurmanci
Margreet Dorleijn
Studies in Multilingualism
Vol. 1: Ad Backus, Two In One.
Bilingual Speech of Turkish Immigrants in The Netherlands, 1996
Vol. 2: Jeroen Aarssen, Relating events in two languages.
Acquisition of cohesive devices by Turkish-Dutch bilingual children at school age, 1996
Vol. 3: Margreet Dorleijn, The decay ofergativity in Kurmanci.
Language internal or contact induced?, 1996
The decay of ergativity in Kurmanci
Language internal or contact induced?
Academisch Proefschrift
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam,
op gezag van de Rector Magnificus
prof.dr. P.W.M. de Meijer
ten overstaan van een door het college van dekanen
ingestelde commissie in het openbaar te verdedigen
in de Aula der Universiteit
op maandag 25 november 1996 te 15.00 uur
door
Margreet Dorleijn
geboren te Amersfoort
Tilburg University Press 1996
Promotor: Prof.dr. P.C. Muysken
Faculteit der Letteren
© Tilburg University Press 1996
No part of this publication may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopy,
recording, or any information storage and retrieval system,
without permission from the copyright owner.
ISBN 90-361-9567-5
Tilburg University Press
P.O. Box 90153
5000 LETilburg
The Netherlands
