Cataloguing the World's Endangered Languages 1138922080, 9781138922082

Cataloguing the World’s Endangered Languages brings together the results of the extensive and influential Catalogue of E

315 147 2MB

English Pages 316 [317] Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Half Title
Title Page
Copyright Page
Table of Contents
List of contributors
Acknowledgements
1 Introduction: Why catalogue endangered languages?
2 Naming convention and challenges
3 Language classification and cataloguing endangered
languages
4 Language extinction then and now
5 The Language Endangerment Index
6 New knowledge produced by the Catalogue of Endangered
Languages
7 The world’s endangered languages and their status
8 How the Catalogue of Endangered Languages serves communities
whose languages are at risk
9 Triage: Setting priorities for endangered language research
Index
Recommend Papers

Cataloguing the World's Endangered Languages
 1138922080, 9781138922082

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

CATALOGUING THE WORLD’S ENDANGERED LANGUAGES

Cataloguing the World’s Endangered Languages brings together the results of the extensive and influential Catalogue of Endangered Languages (ELCat) project. Based on the findings from the most extensive endangered languages research project, this is the most comprehensive source of accurate information on endangered languages. The book presents the academic and scientific findings that underpin the online Catalogue, located at www.endangeredlanguages.com, making it an essential companion to the website for academics and researchers working in this area. While the online Catalogue displays much data from the ELCat project, this volume develops and emphasizes aspects of the research behind the data and includes topics of great interest in the field, not previously covered in a single volume. Cataloguing the World’s Endangered Languages is an important volume of particular interest to academics and researchers working with endangered languages. Lyle Campbell is Professor of Linguistics at the University of Hawai‘i at Mānoa and Director of the Endangered Languages Project 2009–2016. Anna Belew is a PhD candidate at the University of Hawai‘i at Mānoa and has been Project Coordinator for the Endangered Languages Project.

CATALOGUING THE WORLD’S ENDANGERED LANGUAGES

Edited by Lyle Campbell and Anna Belew

First published 2018 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2018 selection and editorial matter, Lyle Campbell and Anna Belew, individual chapters, the contributors The right of Lyle Campbell and Anna Belew to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Names: Campbell, Lyle editor author. | Belew, Anna editor author. | Endangered Languages Project (Website) Title: Cataloguing the world's endangered languages / edited by Lyle Campbell and Anna Belew. Description: Milton Park, Abingdon, Oxon ; New York, NY : Routledge, 2018. | Collected research based on information from the Catalogue of endangered languages, available via the Endangered Languages Project website. | Includes bibliographical references and index. Identifiers: LCCN 2017032500| ISBN 9781138922082 (hardback : alk. paper) | ISBN 9781315686028 (ebook) Subjects: LCSH: Endangered languages—Electronic information resources. | Catalogue of endangered languages (Online) Classification: LCC P40.5.E53 C38 2018 | DDC 400—dc23 LC record available at https://lccn.loc.gov/2017032500 ISBN: 978-1-138-92208-2(hbk) ISBN: 978-1-315-68602-8(ebk) Typeset in Times New Roman by Florence Production Ltd, Stoodleigh, Devon, UK Visit the companion website at www.endangeredlanguages.com

CONTENTS

List of contributors Acknowledgements

vii viii

1

Introduction: Why catalogue endangered languages? Lyle Campbell and Anna Belew

2

Naming convention and challenges Yen-ling Chen and Lyle Campbell

3

Language classification and cataloguing endangered languages Russell Barlow and Lyle Campbell

1

15

23

4

Language extinction then and now Anna Belew and Sean Simpson

49

5

The Language Endangerment Index Nala H. Lee and John R. Van Way

66

6

New knowledge produced by the Catalogue of Endangered Languages Lyle Campbell and Eve Okura

7

The world’s endangered languages and their status Anna Belew, Yen-ling Chen, Lyle Campbell, Russell Barlow, Bryn Hauk, Raina Heaton, and Stephanie Walla v

79

85

Contents

8

9

How the Catalogue of Endangered Languages serves communities whose languages are at risk Raina Heaton and Sean Simpson Triage: Setting priorities for endangered language research Bryn Hauk and Raina Heaton

Index

250

259

305

vi

CONTRIBUTORS

Russell Barlow, University of Hawai‘i at Mānoa [email protected] Anna Belew, University of Hawai‘i at Mānoa [email protected] Lyle Campbell, University of Hawai‘i at Mānoa [email protected] Yen-ling Chen, University of Hawai‘i at Mānoa [email protected] Bryn Hauk, University of Hawai‘i at Mānoa [email protected] Raina Heaton, University of Oklahoma [email protected] Nala H. Lee, National University of Singapore [email protected] Eve Okura, University of Hawai‘i at Mānoa [email protected] Sean Simpson, Georgetown University [email protected] John R. Van Way, University of Hawai‘i at Mānoa [email protected] Stephanie Walla, Eastern Michigan University [email protected]

vii

ACKNOWLEDGEMENTS

The research for the Catalogue of Endangered Languages project was funded by a grant from the National Science Foundation: “Collaborative Research: Endangered Languages Catalog (ELCat),” BCS-1058096 to the University of Hawai‘i at Mānoa (Principal Investigator Lyle Campbell) and BCS-1057725 to Eastern Michigan University (Principal Investigators Helen Aristar-Dry and Anthony Aristar), and by a grant from the Henry Luce Foundation: “Research on Endangered Languages in China and Mainland Southeast Asia for the Catalogue of Endangered Languages” (Lyle Campbell, PI.) The goals and basic organization of the Catalogue were established in an international workshop with some 50 specialists from around the world supported by National Science Foundation grant “Collaborative Research: Endangered Languages Information and Infrastructure Project” (NSF 0924140). The Endangered Languages Project website, where the Catalogue of Endangered Languages is housed, was built in partnership with Google.org.

viii

1 INTRODUCTION Why catalogue endangered languages? Lyle Campbell and Anna Belew

This chapter introduces cataloguing the world’s endangered languages. The principal aims of this book are to report and make known the findings of the Catalogue of Endangered Languages (available online at www.endangeredlanguages.com). This introductory chapter outlines the history of the Catalogue of Endangered Languages project, describes the Catalogue database, and highlights the Catalogue’s achievements in producing new knowledge, filling gaps in information, correcting errors, and updating information. In addition, it addresses how endangered languages are identified (together with their degree of endangerment), the causes of language endangerment, and the many reasons why loss of languages matters to us all.

Introduction: cataloguing endangered languages Awareness of the plight of endangered languages has become more widespread in recent years, and accounts of the world’s languages in crisis are not uncommon. For example, media reports of “language obituaries” – where the death1 of a language is reported with the death of its last known speaker – are increasingly common. Some recent examples include: Tommy George, the last fluent speaker of Awu Laya, an aboriginal language [of Australia], died August 12, 2016, at the age of 87. He had been awarded an honorary doctorate by James Cook University for help in documenting language and traditional fire management of land. (The Australian, Aug. 13, 2016)2 The family of worldwide indigenous languages lost another fluent speaker in the death of Doris Jean Lamar McLemore. Considered the last fluent speaker of the Wichita language, she walked on August 30, 2016. (Indian Country Today, October 6, 2016)3 Edwin Benson, last known native speaker of Mandan, died in December of 2016 at 85 years of age. (KFYR News, December 9, 2016)4 1

Lyle Campbell and Anna Belew

Some widely reported cases are quite famous. For example, Ubykh, a Northwest Caucasian language, was reported to have “died at daybreak, Oct. 8, 1992” when its last speaker, Tevfik Esenç, passed away (Crystal 2000:2). Eyak of Alaska lost its last speaker when Marie Smith Jones died at the age of 89 on January 21, 2008. Language obituaries such as these lend a personal touch to the alarm over language extinction and help to make the crisis of language endangerment more understandable. However, for most languages that have ceased to be spoken, we do not know who the last speaker may have been. While language obituaries such as these highlight individual cases of language death, they also help to cast a spotlight on the broader global concern of language endangerment: an endangered language, by definition, is one at risk of extinction, and many of the world’s languages are at risk of extinction, many of them in the very near future. The endangered languages crisis is widely acknowledged as one of the pressing problems facing humanity today, posing moral, practical, and scientific issues of enormous proportions. Previously, there was no single fully reliable source dedicated to the endangered languages of the world, and a catalogue of the world’s endangered languages was called for to help address these problems. Responding to this often-repeated call, Lyle Campbell, Helen Aristar-Dry, and Anthony Aristar set out to create just such a catalogue: a comprehensive database of the endangered languages of the world. Thus, the Catalogue of Endangered Languages (ELCat) was born. In 2012, it was made available to the public on the Endangered Languages Project (www.endangeredlanguages.com), a website dedicated to promoting endangered languages. The Catalogue informs users about the plight of endangered languages and encourages efforts to slow the loss. It provides information on the endangered languages of the world as a cost-free resource for the public, scholars, individuals and groups facing language loss, and funding agencies to help them make the best decisions in how to deploy limited resources most effectively (see Hauk and Heaton, Chapter 9). This website serves both as a resource for information on the endangered languages of the world, and as an arena to collaborate with others working to document, revitalize, and promote endangered languages. This book’s main objective is to make known the results from the comprehensive Catalogue of Endangered Languages, the most complete resource on the endangered languages of the world. It provides up-to-date information involving language identification, speaker numbers, locations, and language classification. It identifies which languages are most in danger and what impact their loss will have. It reports the Catalogue of Endangered Languages’ achievements in producing new knowledge, filling gaps in information, correcting errors, and updating information (for example, on number and age of speakers, locations of languages, and degree of endangerment). This book seizes the opportunity to present academic and scientific material that is not available on the Catalogue of Endangered Languages website or that is much abbreviated there. It emphasizes those aspects of endangered languages and the research on them that are of particular relevance and interest to scholars, students, and the more engaged members of the public. In addition, we hope this volume may be of use to practitioners of language documentation and conservation who are interested in global trends in language endangerment and revitalization, and in the methods used to produce this kind of data.

Language endangerment in context Languages have been going extinct for as far back in history as we are able to see. There are many well-known cases from antiquity: Etruscan, Hittite, Sumerian, and many others. Given 2

Introduction

this, some might ask, why the current alarm? The answer is that languages are becoming extinct today at a strikingly accelerated rate; the magnitude of language extinction is much greater now than ever before (see Belew and Simpson, Chapter 4). The evidence of the sharp increase in language loss is seen in the history of language extinction and in the current status of languages everywhere. For example, California had some 100 American Indian languages at the time of the Gold Rush, c.1850, but only 18 are still spoken today; none of them is being learned by children through intergenerational transmission; all are highly endangered. Of the some 314 languages spoken in what is now the US and Canada when Europeans first arrived, 152 no longer have native speakers (48%). Of about 280 languages at the time of first European contact in territory now belonging to the US, only 76 are still spoken (27%). However, these figures are misleading – all of the remaining 76 are endangered, most of them critically so, and only about a dozen are still being passed on to children. Many of them will soon become dormant if language revitalization efforts are not successful. The statistics for languages no longer spoken and for those critically endangered are similarly very high in Australia, much of Latin America, and stretches of northern Eurasia; in fact, no inhabited region of the world is exempt from language endangerment. The Catalogue of Endangered Languages lists 427 languages as “critically endangered,” with 262 languages having ten or fewer speakers. Altogether, the Catalogue of Endangered Languages lists 3,138 currently endangered languages5; that is 45.6% of the 6,879 living languages in the world as listed by Ethnologue (Simons & Fennig 2017).6 The number of extinct language families offers another perspective on the crisis. Of the world’s c.407 independent language families (including language isolates, language families that have only one member), already 96 are extinct – no language belonging to any of these families has any remaining native speakers. This means that 24% of the linguistic diversity of the world, calculated in terms of language families, has been lost. Of all the millennia in which languages could have disappeared, two-thirds of these language families became extinct in only the last 60 years, dramatically underscoring the accelerating rate of language extinction in recent times. Worse, many more languages and language families are on the brink of losing their last native speakers and will soon follow, portending a drastic change in the world’s linguistic diversity. What does this mean? The loss of a specific language may be likened to the loss of a single species, say the Bengal tiger or the Right whale. However, the extinction of whole families of languages is a tragedy analogous to the loss of whole branches of the animal kingdom, say to the loss of all felines or all cetaceans. Just imagine the distress of biologists attempting to understand the animal kingdom with major branches missing. Yet this is what confronts us: a staggering quarter of the linguistic diversity of the world is gone forever. Endangered languages are often likened to endangered species in the public discourse (though we make no claims as to the validity of this comparison), but actual comparisons show that losses to the world’s linguistic diversity are even greater than losses in the diversity of biological species. Figures reported for endangered biological species vary widely, some elevated no doubt in part due to well-meaning attempts to raise public sympathy for the cause and in part due to the fact that about many species very little is known. A representative count lists as “Critically Endangered” or “Endangered” about 12% of mammals, 13% of birds, and 21% of amphibians (IUCN Red List 2017). Clearly the threat to biological species is smaller in numerical scale than that to human languages: as noted above, 46% of all living languages are currently listed as endangered (Catalogue of Endangered Languages) – the rate of endangerment for the most highly endangered kind of animals, the amphibians, is less than half of that of endangered languages. This underscores the seriousness of the language endangerment crisis. 3

Lyle Campbell and Anna Belew

How do we determine that a language is endangered? How does a language come to have an entry in the Catalogue of Endangered Languages? The main criteria used by the Catalogue to determine whether a language is endangered are: • •

• •

The absolute number of speakers – in very general terms, the fewer the number of speakers, the less likely the language’s long-term survival. Intergenerational transmission – if a language is not being learned by children in the traditional way, passed on from one generation to the next, it is likely doomed to extinction unless revitalization efforts prove successful. The greater the degree of intergenerational transmission, the more likely the language’s survival. Trends in the number of speakers – the more rapidly the number of speakers decreases, the more endangered the language is. Decrease in domains of use – the more the domains in which the language is used are reduced, the greater its endangerment becomes.

The Catalogue of Endangered Languages reports the vitality of the world’s endangered languages based on these criteria. The Catalogue’s Language Endangerment Index (LEI) gives a score for the degree of endangerment of each endangered language. In general, for a language to receive an entry in the Catalogue, it needs to score higher than 0% on the LEI. Non-endangered languages such as French or Thai would score a 0% on all factors used by the LEI; higher percentages indicate higher levels of endangerment (for example, a language that scored 92% on the LEI would be “Critically Endangered”). See Lee and Van Way, Chapter 5, for details of how the LEI calculates levels of endangerment.

Causes of language endangerment It bears repeating that “the factors determining obsolescence of language are non-linguistic” (Swadesh 1948:235). Numerous causal factors of language endangerment have been identified, and are sometimes grouped under higher-order categories such as the following: •

• •



Economic factors: lack of economic opportunity, rapid economic transformation, shifts in work patterns, resource depletion, forced changes in subsistence patterns, communication with outside regions, resettlement, destruction of habitat, globalization, etc. Political and social factors: discrimination, repression, official language policies, level of education available, population dispersal, etc. Attitudes: attitudes of the speakers towards the languages under threat and towards the official national language(s) and dominant languages that surround them, attitudes of members of mainstream society towards minorities and their languages, prestige or stigma associated with the endangered language and dominant language(s), etc. Lack of institutional support: as represented in the roles of the languages in education, government, churches, and the media, and even recreational activities (sports events, popular culture, music, etc.).

Causes of language endangerment are complex, however, and differ widely depending upon the language(s) involved; it is impossible to provide any simple list of factors, and their effects upon a given language may vary greatly. For additional discussion of causes of language endangerment, see Belew and Simpson, Chapter 4. 4

Introduction

Why should we care? The endangered languages literature provides many answers to this question, and many reasons for why language endangerment matters to us all. Some of the main ones follow.

Social justice and human rights Language loss is often not voluntary; it frequently involves violations of human rights, pushed by political or social repression, oppression, aggression, prejudice, violence, and at times by ethnic cleansing and genocide. Cases such as these are a matter of right and wrong, and that is important to everyone. Globally, the concern for language endangerment connects with concern for various types of social injustice. Language loss is often experienced as a crisis of social identity. For many communities, work towards language revitalization is not about language in isolation, but is part of a “larger effort to restore personal and societal wellness” (Pfeiffer and Holm 1994:35). Some scholars and community activists insist that ongoing language loss leads to damaged communities and dysfunctional behaviors. They argue that one’s psychological, social, and physical well-being is connected with one’s native language; it shapes one’s values, self-image, identity, relationships, and ultimately success in life. (Indeed, the 5th International Conference on Language Documentation and Conservation was organized around the theme of “Language and Wellbeing”; see http://icldc5.icldc-hawaii.org). Indigenous voices testify to the crucial role language plays in their cultural and personal identity. The following are representative: How can I believe the foolish idea That my language is weak and poor If my mother’s last words Were in Evenki? (Alitet Nemtushkin, Evenki poet)7 For centuries our languages have been a reflection of those cultural distinctions that have made us who we are as a people, and in a sense have been an element of the many things that have made us strong. (Stephen Greymorning [Arapaho] 1999:6) Why save our languages, since they now seem to have no political, economic, or global relevance? That impression is exactly the reason why we should save our languages, because it is the spiritual relevance deeply embedded in our own languages that makes them relevant to us as American Indians today. (Richard Littlebear [Northern Cheyenne] 2000). Each language still spoken is fundamental to the personal, social and – a key term in the discourse of indigenous peoples – spiritual identity of its speakers. (Zepeda [Tohono O’odham nation] and Hill 1991:1) I can’t stress enough the importance of retaining our tribal languages, when it comes to the core relevance or existence of our people . . . You could argue that when a tribe loses its language, it loses a piece of its inner-most being, a part of its soul or spirit . . . When it comes to native languages, the situation is simple: Use it or lose it. (Sonny Skyhawk [Sicangu Lakota, Hollywood actor] 2012) 5

Lyle Campbell and Anna Belew

Human concerns Languages are the treasure houses of information for history, literature, philosophy, art, and the wisdom and knowledge of humankind. The stories, ideas, and words contained in a language help us make sense of our own lives and of the world around us – of the human experience, and of the human condition in general. When a language goes extinct without documentation, we lose incalculable amounts of human knowledge. We illustrate this by mention of only two areas, literature and history. Literature: As many scholars often assert, by studying literature, we learn what it means to be human. This is equally true of the oral literatures of the indigenous peoples of the world – they, too, have grappled with the complexities of their world and the problems of life, and the insights and discoveries represented in their literatures – whether written or oral – are of no less value to us all. When a language becomes extinct without documentation, taking all its oral literature, oral tradition, and oral history with it into oblivion, all of humanity is diminished. History: We study history “to gain access to the laboratory of human experience” (Stearns 1998). Great reservoirs of historical information are contained in languages. The classification of related languages teaches us about the history of human groups and how they are related to one another, and we gain understanding of contacts and migrations, the original homelands where languages were spoken, and past cultures from the comparison of related languages and the study of language change – all irretrievably lost when a language is lost without adequate documentation. Because languages encompass the world’s knowledge and wisdom, the loss of the literature and historical information, and much more, means loss in the potential ways of experiencing and understanding the world.

Loss of knowledge The world’s linguistic diversity is one of humanity’s most valuable treasures. This means that the loss of the many hundreds of languages that have already become extinct is a cataclysmic intellectual disaster, on many different levels. To cite a single example, encoded in each language is knowledge about the natural and cultural world it is used in. This knowledge is often not known outside of the small speech communities where the majority of endangered languages are spoken. When a language dies without adequate documentation it takes with it this irreplaceable knowledge. It is argued that in principle, loss of such knowledge could have devastating consequences even for humankind’s very survival. Reduction of language diversity diminishes the adaptational strength of the human species because it lowers the pool of knowledge from which we can draw. A telling example comes from the Seri (of Sonora, Mexico, with c.700 speakers). The Seri have knowledge of “eelgrass” (Zostera marina L.) and “eelgrass seed,” which they call xnois. It is “the only known grain from the sea used as a human food source . . . eelgrass has considerable potential as a general food source . . . Its cultivation would not require fresh water, pesticides, or artificial fertilizer” (Felger and Moser 1973:355–356). Seri has a whole set of vocabulary items dealing with eelgrass and its use. According to the argument, it is all too plausible to imagine a future in which some natural or human-caused disaster might compromise land-based crops, leaving human survival in jeopardy because of the loss of knowledge of alternative food sources such as the knowledge of eelgrass reflected in the Seri language. Speculation aside, it is clear that documentation of the languages of small-scale societies and of the knowledge they hold has significantly benefitted humanity. 6

Introduction

Other examples come from medicine. It has been reported that 75% of plant-derived pharmaceuticals were discovered by examining traditional medicines, where the language of curers often played a key role (see, for example, Bierer et al. 1996). If these languages had become extinct and knowledge of the medicinal plants and their uses had been lost, all of humanity would be impoverished and human survival would be left less secure. For these and other reasons, it can be argued that reduction of language diversity diminishes the adaptive strength of humans as a species because it lowers the pool of knowledge from which we can draw for survival.8

Consequences for understanding human language A major goal of linguistics is to understand human cognition and human language capacity through the study of what is possible and impossible in human languages. Discovery of previously unknown linguistic features and traits as we document languages contributes to achieving this goal and advances knowledge of how the human mind works. Conversely, language extinction is horrendous, impeding achievement of this goal. The following example illustrates this well. The discovery of the existence of languages with OVS (Object–Verb–Subject) and OSV (Object–Subject–Verb) basic word orders forced abandonment of a previously postulated language universal. Greenberg (1978:2) had proposed that “whenever the object precedes the verb the subject does likewise.” However, it was discovered that Hixkaryana (a Cariban language of Brazil, with only 350 speakers) has OVS basic word order, seen in the sentence: toto yonoye kamura man ate jaguar “The jaguar ate the man.” We now know that several languages have OVS or OSV basic word order; most of them are spoken in small communities in the Amazon. Discovery of languages with these basic word orders not only forced abandonment of the postulated universal, but also required revision of numerous other theoretical claims about language. It is all too plausible, given what has happened to indigenous languages at the hands of unscrupulous loggers, miners, and ranchers, that the few languages with these word orders could have become extinct before they were documented, leaving us forever with erroneous assumptions about what is possible and impossible in human language and how that reflects on understanding of human cognition. Documentation of endangered languages has frequently and repeatedly demonstrated the importance of obtaining adequate descriptions of these languages; the discovery of previously unknown linguistic traits is helping linguists to comprehend the full range of what is possible and impossible in human language (for numerous examples, see Palosaari & Campbell 2011).

Peace through loss of languages? A final reason for why we should care about language loss involves setting straight erroneous views about the role of languages in geopolitical conflicts. It is often said that if we had fewer languages, we would understand each other better and live in greater harmony. But this is far from true. That monolingualism does not guarantee nor even foster greater “understanding” is attested throughout history. As David Crystal (2000:27) reminds us, “all of the large monolingual countries of the world have had their civil wars.” It is also shown by the many recent and ongoing armed conflicts among groups speaking the same language: in Darfur, Egypt, 7

Lyle Campbell and Anna Belew

Iraq, Libya, Syria, Yemen, Colombia, Northern Ireland, Thailand, or the 1994 genocide in Rwanda involving Hutu and Tutsi, both speakers of the same language, Kinyarwanda. This contrasts with lack of such conflicts in relatively peaceful, officially multilingual Belgium, Canada, Finland, Luxembourg, Singapore, Switzerland, Tanzania, etc. Multilingual and multicultural countries need to recognize that national unity and understanding are not fostered by monolingualism or ethnic cleansing, but that recognition of minority language rights can bring about mutual trust, peace, and ultimately, national stability. We need to expose the erroneous assumption that people and countries cannot be both multilingual and successful, and show, rather, that there are significant benefits from multilingualism. In short, there is no evidence that fewer languages might lead to greater harmony.

A brief history of ELCat In 2009, a group of prominent linguists and language activists gathered for the Endangered Languages Information and Infrastructure Workshop (ELIIP) at the University of Utah, funded by the US National Science Foundation. The workshop’s goal was to lay the groundwork for a comprehensive digital database of information about the world’s endangered languages, and for the infrastructure for disseminating and updating this fast-changing data. The workshop produced recommendations for what types of data should be collected about endangered languages, and how to structure this information in a database; recommendations for assessing linguistic vitality in a more thorough, accurate, and universally applicable way than existing endangerment metrics; and recommendations for constructing a website that would allow the public to view, search, and contribute to this database. In 2011, the US National Science Foundation awarded funding to construct the database designed at the ELIIP meeting. The project, now called the Catalogue of Endangered Languages (abbreviated ELCat), was undertaken in collaboration between teams at the University of Hawai‘i at Mānoa (UHM) Department of Linguistics and The LINGUIST List at Eastern Michigan University (EMU). UHM was responsible for providing data on the languages of Europe, the Americas, Asia, and the Pacific, while EMU was responsible for the technical infrastructure of the Catalogue’s database, as well as for data on the languages of Africa and Australia. The initial version of the Catalogue database was compiled during 2011 and early 2012 by the UHM and EMU teams, with input and supervision from the project’s team of Regional Directors (see Appendix 1.1 for the list of Regional Directors). Also in 2011, the project’s principal investigators were approached by representatives from Google.org, the nonprofit arm of the technology company Google. Google.org contributed by building the web platform, in collaboration with the web design firm Vizzuality and the ELCat teams, and providing fixed-term funding for the maintenance and development of the project. On June 21, 2012, the Endangered Languages Project (ELP) website was launched at www.endangeredlanguages.com, featuring the initial version of the Catalogue of Endangered Languages. Among the other features of the ELP site, the web platform allowed the public to suggest additional information or changes to the information contained in the Catalogue. Over the next four years, the ELCat team continued adding and updating the language information in the Catalogue (see Appendix 1.2 for a complete list of staff, research assistants, and volunteers who have contributed to the ELCat project), and they continued vetting and inputting site users’ suggestions for language data improvements and updates. The ability for users who have firsthand experience with a language to submit information directly, and for the ELCat team to reply quickly, ask for additional data, and implement these changes, is of major benefit to ELCat’s goal of providing the most accurate and up-to-date language information possible 8

Introduction

(see also Heaton and Simpson, Chapter 8, for discussion of how the ELP website’s functionality serves language communities). In 2013, the Henry Luce Foundation awarded ELCat a grant (2013–2016) supporting research on the endangered languages of China and mainland Southeast Asia (Lyle Campbell, PI), an area much in need of this concentrated effort. In 2015, the NSF-funded period for the construction of the ELCat database came to a close. In order to ensure the continued maintenance and accuracy of the Catalogue, a workshop was held at UHM in February 2015 titled Sustaining the Catalogue of Endangered Languages. The personnel involved in the ELCat and ELP projects came together to devise a long-term governance structure for the Catalogue, procedures for adding and changing language data after grant funding expired, and roles and responsibilities for the continued upkeep of ELCat. As a result of this workshop, ELCat’s International Advisory Board was formed, with Dr. Bill Palmer as the initial chair. (See Appendix 1.3 for the current membership of the International Advisory Board). In 2016, Google.org’s period of funding the ELP website came to a close, and hosting and technical support for the ELCat and ELP projects moved to the University of Hawai‘i at Mānoa’s Office of Information Technology and Support, where it remains today.

The ELCat database Below we provide a brief description of the ELCat database, which may be useful to those interested in the technical aspects of the project, as well as clarifying certain aspects of how the Catalogue’s language data are presented. The Catalogue of Endangered Languages is stored and hosted on the same servers, and within the same database, as the Endangered Languages Project. The site is written in Django (a Python-based web development framework), which encompasses both the front end (website) and back end (database), and is hosted at the University of Hawai‘i at Mānoa’s Office of Information Technology and Support. The integration of ELCat and ELP means that each language entry in the Catalogue generates its own webpage, where all information about the language is displayed, along with any user-submitted materials pertaining to that language. Only ELCat staff has access to the back end of the database, and only authorized members of the research team and International Board of Directors are able to edit or add language information directly. Users may suggest changes or additions to the Catalogue’s information directly through the language web pages (see Heaton and Simpson, Chapter 8), but these changes are carefully reviewed by the research team before being implemented – the Catalogue itself is strictly academically moderated, rather than being crowd-sourced. ELCat is not a fieldwork initiative; it does not collect information on endangered languages directly. Rather, it compiles published and unpublished data from a wide variety of sources, including journal articles, books, news reports, conference presentations, and direct communications with scholars and community language experts. The structure of the ELCat database allows for the incorporation of information from multiple bibliographic sources regarding a single language. This is key to ELCat’s goal of providing a comprehensive database of language information: rather than limiting the data presented to that published in a single source, the Catalogue’s users can access all available information from all available sources. This is especially important in cases where conflicting information exists, or where published information has changed over time. Users are able to view all of the information from a single bibliographic source at once, and can scroll to display information from other sources. For example, the language page for Southern Selkup (www.endangeredlanguages.com/lang/8555), a language of Russia, includes information from five bibliographic sources. By scrolling through the information from different sources, users can see that a publication from 2005 provided a 9

Lyle Campbell and Anna Belew

count of 15–20 speakers of Southern Selkup, another source from 2013 estimated there were fewer than five speakers, and a 2015 source listed one remaining speaker. All information is presented with a citation of its original source, so users may follow up by reading the original source or make judgments about its reliability. The ability to see and compare information from multiple publications is useful for tracking changes in a language’s vitality, as well as allowing users to see if there is conflicting information about a language’s vitality and context (e.g. one source may consider a language severely endangered, while another considers it mildly endangered). While ELCat strives to present all available information on a language, the design of the website necessitates designating a single source of information that will appear first when a user arrives at a language’s page. This is the “preferred” source, and is selected by the ELCat research team based on recency, reliability, and completeness of information. In the case of Southern Selkup, the 2015 source citing one speaker is “preferred,” both because it reflects the most recent information about the language, and because it is deemed highly reliable (a direct communication from a linguist who is conducting firsthand research on the language). All statistics and information presented in this volume are based on the “preferred” sources of information in the ELCat database as of April 4, 2017; because the database is updated regularly as new data becomes available, information presented here may differ from what is found in the Catalogue at any point in the future. For more information about the structure and functionality of the Catalogue and the Endangered Languages Project, please visit www.endangeredlanguages.com/about.

The book’s chapters and how they contribute to the book’s goals This volume aims to present the knowledge gained from the construction of the Catalogue of Endangered Languages: not only the data gathered about endangered languages themselves, but the lessons learned from the process of researching, compiling, and maintaining a database of information about the world’s endangered languages. While the ELP website allows users to contribute language data and materials, it is limited in its ability to present supplementary material explaining the rationale, processes, and complications involved in curating language information on a large scale. This volume therefore provides a comprehensive overview of the compilation of, and lessons learned from, the Catalogue of Endangered Languages. These lessons are broadly applicable: many of the issues that the ELCat project has addressed are also faced by researchers working with endangered languages, language workers in revitalization and revival initiatives, language activists and policymakers, learners and teachers of endangered languages, academics working with large datasets in other social sciences and humanities, and others doing language-related work. For example, the issue of what to call a language is often a complicated one, and one that is probably familiar to anyone who has worked with endangered or under-documented languages. In Chapter 2, Chen and Campbell outline potential complications involved in the names representing given languages. In some cases, use of autonyms (the name of the language in the language itself, or what its speakers call it) are preferred by language communities over exonyms (names given by outsiders). However, Chen and Campbell discuss numerous complications involved in names for languages, with a description of how the Catalogue has chosen to resolve these complicated issues. In Chapter 3, Barlow and Campbell provide an overview of language classification and its importance in particular to endangered languages. They discuss the challenges posed by attempts to classify under-documented or poorly attested languages, and explain ELCat’s 10

Introduction

approach to contested or unclear classifications, as well as the importance of endangered languages to understanding the world’s linguistic diversity, and the benefits that accurate classification can provide to speakers and language communities. In this context, they provide an updated list of the world’s language families (including language isolates), which incorporates recent consensus knowledge about the classification of the languages in the Catalogue. In Chapter 4, Belew and Simpson present an overview of language death, dormancy, and revival in the Catalogue of Endangered Languages. The concept and terminology of language “death” is discussed, and the authors outline ELCat’s policies regarding languages with no native speakers, including those which are undergoing revival efforts. They then discuss what is known about language extinction in the past, and how this compares to rates of global language loss today, as seen in the Catalogue’s data. Finally, they provide an overview of the Catalogue’s information on previously dormant languages that are now being revived, concluding that while language loss is currently occurring at an unprecedented pace, the growing rate of language revival is cause for cautious optimism. ELCat’s metric for measuring language endangerment is a central feature of the project, and in Chapter 5, Lee and Van Way describe the Language Endangerment Index (LEI), its rationale and workings, and its benefits as compared with other vitality-assessment tools. They discuss the need for flexibility in such a tool when dealing with a wide range of endangerment scenarios and levels of documentation. They outline the LEI’s approach to accounting for all available and unavailable information about a language’s vitality. Finally, they discuss the use of the LEI within ELCat, and provide a number of case studies to illustrate how the LEI may be applied to a variety of language situations. While much of the volume’s first half focuses on the Catalogue’s development and how the project addresses the challenges of language cataloguing, the second half of the book focuses on applications of the information compiled by the ELCat project. In Chapter 6, Campbell and Okura present the new knowledge gained from the Catalogue’s data. They provide updated calculations of the current pace of language loss, comparing it to the oft-repeated claim that “one language dies every two weeks,” and they issue a corrective to this figure. Chapter 7 presents the Catalogue’s full listing of the world’s endangered languages, their endangerment levels, number of speakers, and locations. The lists are organized by major region, and brief overviews of language endangerment in each region are provided, along with useful references to further work on language endangerment in specific parts of the world. While the Catalogue is a scholarly resource, its target audience is not limited to academics. In Chapter 8, Heaton and Simpson discuss how the Catalogue and the Endangered Languages Project website can serve the needs of language communities. They present the Catalogue’s goals for serving language stakeholders, and provide several case studies of community organizations making use of ELCat and ELP to support their language work. They outline some of the features of the Catalogue and the Endangered Languages Project that may be of additional benefit for language communities in the future. Concluding the volume is Chapter 9, in which Hauk and Heaton address the issue of “triage,” how to utilize the information contained in the Catalogue of Endangered Languages to prioritize language documentation work. One of the Catalogue’s objectives is to help granting bodies, policymakers, and community organizations make informed decisions about how to deploy limited resources for language documentation, by providing accurate information about endangered languages. Hauk and Heaton discuss a number of potential criteria for setting priorities in language documentation, such as degree of current endangerment, typological uniqueness, and amount of existing documentation, as well as extralinguistic factors such as the language’s accessibility to outside researchers. 11

Lyle Campbell and Anna Belew

Conclusions This book, Cataloguing the World’s Endangered Languages, aims to present the results of the Catalogue of Endangered Languages Project and to make its findings better-known to its varied audience, to scholars, members of communities whose languages are endangered, funding agencies, teachers and students, the media, language policy decision makers, and the general public. The Catalogue of Endangered Languages is the most comprehensive resource on the endangered languages of the world, and this book reports its achievements in producing new knowledge, filling gaps in information, correcting errors, and updating information: for example, on number of speakers, degree of endangerment, and much more. The book goes beyond this, presenting also academic and scientific information about endangered languages that is not available on the website that hosts the Catalogue of Endangered Languages (at www.endangeredlanguages.com) or is not sufficiently elaborated there. There was previously no single fully reliable source dedicated exclusively to the endangered languages of the world; the Catalogue of Endangered Languages (ELCat) meets its objective of filling this gap. The Catalogue and this volume have broader goals, as well: they are dedicated to promoting endangered languages, to informing users and readers about the plight of endangered languages, and to encouraging efforts to turn the tide and reduce language loss. It is hoped that this book will contribute to these ends and prove to be a useful resource for all with interest in and concern for the languages of the world, their diversity, and for the language endangerment crisis facing us all.

Notes 1

2 3 4 5 6 7 8

Terms such as “death” and “extinction” are still common in the linguistic literature and popular media, and used here with the caveat that “death” of a language need not be permanent; see Belew and Simpson, Chapter 4, for more on the terminology of language “death.” www.theaustralian.com.au/news/nation/language-lost-with-the-passing-of-great-elder-tommygeorge/news-story/3fca836f8f19e249e18437bc2732415f https://indiancountrymedianetwork.com/news/native-news/doris-mclemore-last-fluent-wichita-speakerwalks-on/ www.kfyrtv.com/content/news/Edwin-Benson-last-known-fluent-speaker-of-Mandan-passes-awayat-85-405723515.html The total number of languages listed in the Catalogue of Endangered Languages is 3,394, but this includes 256 dormant or awakening languages, that is, languages with no known native speakers. Ethnologue’s total number of languages is 7,099; adjusted for the 220 with no known speakers and one constructed language, their number of living languages is 6,879. www.unesco.org/new/en/culture/themes/endangered-languages/faq-on-endangered-languages/ In principle it is possible for a society to give up its language and shift to another, and find ways to talk about this sort of knowledge in the new language. What we observe, however, in case after case, is that when a language is not passed on to the next generation, the knowledge of the natural and cultural world encoded in that language fails to be transmitted as well.

References Bierer, Donald E. & Carlson, Thomas J. & King, Steven R. 1996. Shaman pharmaceuticals: Integrating indigenous knowledge, tropical medicinal plants, medicine, modern science, and reciprocity into a novel drug discovery approach. Saluda, NC: Network Science Corporation. Crystal, David. 2000. Language death. Cambridge: Cambridge University Press. Felger, Richard & Moser, Mary Beck. 1973. Eelgrass (Zostera marina L.) in the Gulf of California. Science 181(4097). 355–56. Greenberg, Joseph H. 1978. Introduction. In Greenberg, Joseph H. & Ferguson, Charles A. & Moravcsik, Edith A. (eds.), Universals of human language, vol. 2, 1–6. Palo Alto, CA: Stanford University Press.

12

Introduction Greymorning, Stephen. 1999. Running the gauntlet of an indigenous language program. In Reyhner, Jon & Cantoni, Gina & St. Clair, Robert N. & Yazzie, Evangeline Parsons, Revitalizing indigenous languages, 6–16. Flagstaff, AZ: Northern Arizona University. International Union for Conservation of Nature and Natural Resources. 2017. Summary statistics, Table 3a. (www.iucnredlist.org/about/summary-statistics#Tables_3_4) Littlebear, Richard. 1999. Some rare and radical ideas for keeping Indigenous languages alive. In Reyhner, Jon & Cantoni, Gina & St. Clair, Robert N. & Yazzie, Evangeline Parsons, Revitalizing indigenous languages, 1–5. Flagstaff, AZ: Northern Arizona University. Palosaari, Naomi & Campbell, Lyle. 2011. Structural aspects of language endangerment. In Austin, Peter K. & Sallabank, Julia (eds.), The Cambridge handbook of endangered languages, 100–119. Cambridge: Cambridge University Press. Pfeiffer, Anita & Holm, Wayne. 1994. Laanaa Nisin: Diné education in the year 2004. Journal of Navajo Education 11. 35–43. Simons, Gary F. & Fennig, Charles D. 2017. Ethnologue: Languages of the world, twentieth edition. (www. ethnologue.com) Skyhawk, Sonny. 2012. Why should we keep tribal languages alive? Indian Country, April 6, 2012. (http://indiancountrytodaymedianetwork.com/2012/04/06/why-should-we-keep-tribal-languages-alive99182) Stearns, Peter. 1998. Why study history? (www.historians.org/about-aha-and-membership/aha-historyand-archives/archives/why-study-history-(1998)) Swadesh, Morris. 1948. Sociologic notes on obsolescent languages. International Journal of American Linguistics 14. 226–235. Zepeda, Ofelia & Hill, Jane. 1991. The condition of Native American languages in the United States. In Robins, R. H. & Uhlenbeck, E. M. (eds.), Endangered languages, 135–155. Oxford: Berg.

Appendix 1.1 Regional Directors for the Catalogue of Endangered Languages, 2011–2015 Region1

Director(s)

Africa Australia Caucasus East Asia Europe Indonesia Mexico, Central America, and the Caribbean Near East and Central Asia

Matthias Brenzinger, University of Cape Town Claire Bowern, Yale University Alice C. Harris, University of Massachusetts Amherst David Bradley, La Trobe University Brian Joseph, Ohio State University I. Wayan Arka, Australian National University Lyle Campbell, University of Hawai‘i at Mānoa

North America Northern and Central Eurasia Pacific Romance languages Sign languages South America South Asia Southeast Asia

Habib Borjian, Columbia University Charles C. Häberl, Rutgers University Keren Rice, University of Toronto Lyle Campbell, University of Hawai‘i at Mānoa Juha Janhunen, University of Helsinki Bill Palmer, University of Newcastle Martin Maiden, Oxford University James Woodward, University of Hawai‘i at Mānoa Willem F. Adelaar, Leiden University Lyle Campbell, University of Hawai‘i at Mānoa Gregory Anderson, Living Tongues Institute for Endangered Languages David Solnit

1.

While the Catalogue is divided into 12 primary geographic regions, some specific language families (e.g. Romance) or sub-areas (e.g. Indonesia) have supplementary supervision by additional Directors.

13

Lyle Campbell and Anna Belew

Appendix 1.2 Staff, research assistants, and volunteers for the Catalogue of Endangered Languages project Eastern Michigan University

University of Hawai‘i at Mānoa

Anthony Aristar (Co-Principal Investigator, 2011–2013)

Lyle Campbell (Principal Investigator and project director, 2011–2016) Gary Holton (Catalogue Director, 2016–Present) Carolina Aragon (volunteer) Russell Barlow (RA) Anna Belew (RA, co-project coordinator) Yen-ling Chen (RA) Katie Butler Gao (RA) Bryn Hauk (RA, co-project coordinator) Raina Heaton (RA) Joelle Kirtley (volunteer) Nemanja Komar (RA) Nala Huiying Lee (RA) Stephanie Locke (volunteer) Colleen O’Brien (volunteer) Christianne Ono (RA) Melody Ann Ross (volunteer) Sean Simpson (RA, project coordinator) Alexander D. Smith (RA, project coordinator) Kaori Ueki (volunteer) John Van Way (RA, project coordinator)

Helen Aristar-Dry (Co-Principal Investigator, 2011–2013) Verónica Grondona (Principal Investigator, 2013-2015) Anna Belew (project manager, 2011–2013) Amy Brunett (RA) Jacob Collard (RA) Lesley Dennison (RA) Kristen Dunkinson (RA) Bryn Hauk (project manager, 2013–2014) Uliana Kazagasheva (RA) Lwin Moe (lead programmer) Marcus Nero (assistant project manager, 2015-2016) Sara Oldaugh (assistant project manager, 2015-2016) Stephanie Walla (project manager, 2014–2015) Brent Woo (RA)

Appendix 1.3 Membership of the Catalogue of Endangered Languages International Board of Directors, 2015–present Bill Palmer (International Board of Directors Chair), University of Newcastle Gary Holton (Catalogue Director), University of Hawai‘i at Mānoa Willem Adelaar, Leiden University (Regional Director for South America) Greg Anderson, Living Tongues Institute for Endangered Languages (Regional Director for South Asia) Habib Borjian, Columbia University (Regional Director for Near East) David Bradley, LaTrobe University (Regional Director for East Asia) Matthias Brenzinger, University of Cape Town (Regional Director for Africa) Lyle Campbell, University of Hawai‘i at Mānoa (Regional Director for the Americas) Verónica Grondona, Eastern Michigan University Tracey Herbert, First Peoples’ Cultural Council Brian Joseph, The Ohio State University (Regional Director for Europe) Mary Linn, Smithsonian Institute Keren Rice, University of Toronto (Regional Director for North America) David Solnit (Regional Director for East and Southeast Asia)

14

2 NAMING CONVENTIONS AND CHALLENGES Yen-ling Chen and Lyle Campbell

How languages should be named presents numerous challenges for the Catalogue of Endangered Languages. This chapter identifies the difficulties and explains how these are dealt with and how language names are handled in the Catalogue. A guiding principle with important consequences for naming languages is to make the Catalogue of Endangered Languages maximally useful and maximally accessible to the widest possible range of users and audiences, recognizing their different needs and respecting their preferences.

Introduction This chapter is about the language names in the Catalogue of Endangered Languages, and it explains how decisions about language names are handled in the Catalogue. The Catalogue has identified 3,394 endangered languages. All of them are catalogued with a name, and some with several alternative names, which may be either autonyms or exonyms. (Not all of them have a corresponding ISO 639-3 code, but the ISO code is also provided wherever one exists for a given language.) An autonym is the name used by a group of people to refer to themselves or their language. In contrast, an exonym is a name given by outsiders to refer to a group of people or their language; it is not the name that a group uses to refer to itself or its language. Note that different speech communities that speak a single language may employ different names to refer to themselves or their language, especially when different varieties of the language are involved. The naming conventions used in the Catalogue of Endangered Languages attempt to represent all the salient names by which a language is (or has ever been) known. Therefore, these names can be either autonyms or exonyms. The name by which a language is most widely known in the literature is typically adopted as the primary name (that is, the name that appears first and most prominently in a language page’s heading) in the Catalogue. However, where community members1 express interest in displaying a different name as the primary name, the Catalogue of Endangered Languages attempts where possible to display these preferred names as the primary identifier for the language. For the purposes of the Catalogue, it is not so important which of the alternate names may be listed first, and alternative names are generally not listed in any specific order; what the Catalogue holds as 15

Yen-ling Chen and Lyle Campbell

most important is the possibility for users to search for and find the language by any of its alternate names. It is worth mentioning that the Catalogue of Endangered Language is presented on the upto-date, open-access website of the Endangered Languages Project (ELP), at www.endangered languages.com. Hence, the searchability of ELP is an important criterion for being a user-friendly website. Primary names of all the language entries in the Catalogue are represented in large font. Alternative names provided in Catalogue of Endangered Languages are represented in the “also known as” line on the website. Language names written in non-Latin scripts are also included in the database wherever available, and are searchable on the site. However, given the inability of mainstream web browsers to display certain characters, names written in scripts that do not have Unicode-compatible fonts are not eligible for inclusion in ELCat. The Catalogue of Endangered Languages seeks the best solution to issues involving language names based on some general principles. The decision for what entities are given an independent entry as a separate language in the Catalogue is based on strictly linguistic criteria (i.e. mutual intelligibility). Experience has made it clear that the Catalogue’s decisions must be free of political and other non-linguistic agendas – the Catalogue will not include (or delete) an entry in order, for example, to justify aggression or repression, to gain access to government funding, to get government or other recognition, to provide support for political independence, to underwrite land rights claims, to attempt to justify petitions for extra representatives in governing bodies, to seek exemption from restrictions that apply to others, etc. Decisions in unclear cases in the Catalogue were reached based on standard linguistic criteria, in consultation with the Regional Directors, community members, and others. In cases where there is significant disagreement, the different opinions and the reasons behind them, where known, should be represented in public comments in the language entries of the Catalogue of Endangered Languages.

The problem of a single name applied to two or more languages One difficulty is that it is not uncommon for a single name to apply to two or more separate languages. For example, the name “Tuha” applies to both Dukha [no ISO 639-3] and Uighur Uryangkhai [no ISO 639-3], two distinct Turkic languages. “Nanai” can refer to Nanai [gld],2 Kilen [no ISO 639-3], and Kili [no ISO 639-3], which are separate languages all belonging to the Tungusic language family. Also, sometimes languages within the same language family, although in different subgroups, bear names that are similar – for instance, Mulam [mlm] and Mulao [giu], both of which are Tai-Kadai languages, and Lizu [no ISO 639-3] and Lisu [lis], both of which are Tibeto-Burman, and Kadu [zkd] and Kaduo [ktp], both being TibetoBurman as well. These often cause confusion and misidentification. There are also languages with the same name but in different parts of the world. For example, there are four languages named “Mono.” Such cases are distinguished by placing an indication of the location in parentheses after the name, for example in this case, “Mono (Cameroon)” [mru] and “Mono (Democratic Republic of the Congo)” [mnh] (both Niger-Congo languages) vs. “Mono (Solomon Islands)” [mte] (an Austronesian language) vs. “Mono (United States)” [mnr] (a UtoAztecan language). A particular name can be associated with different languages in the same region, for example, “Solon” is associated with Evenki [evn], Ongkor Solon [no ISO 639-3], and Solon [no ISO 639-3], three different Tungusic languages, and “Khamnigan” is associated with Khamnigan Ewenki [no ISO 639-3], which is Tungusic, and Khamnigan Mongol [no ISO 639-3], which is Mongolic. 16

Naming conventions and challenges

The problem of what to call a language is not, of course, confined to endangered languages. For example, linguistically speaking, there is no single language within the Chinese/Sinitic subgroup (of the Sino-Tibetan family) named “Chinese.” Chinese is a generic term for a large grouping of languages. Nowadays, the term “Chinese” is typically associated with Mandarin (the language of the bureaucrats, originally from Sanskrit mantrin “counselor”), or more precisely with the Mandarin (官話) subgroup, which includes Standard Mandarin, Pekingese, Sichuanese, Shandonghua, etc., of the Sinitic branch. Because speakers of Mandarin have been politically dominant in the history of China, Standard Chinese [cmn], also known as Putonghua (普通 話), literally “Common Language,” or Kuoyu (國語), literally “National Language,” has been selected as the only official language in China and Taiwan. Using “Chinese” as equivalent to “Standard Mandarin” leads people for political reasons to consider other Sinitic languages as not independent languages but as mere dialects3 of Mandarin (see Chapter 3 for discussion of the problem of distinguishing dialects from distinct languages). In the Catalogue, “Chinese” is used as a generic term for the Sinitic branch or things written in Chinese characters (Traditional Chinese or Simplified Chinese), or in reference to an unspecified Chinese language; it is never used to refer only to Standard Mandarin. In the following sections, four major challenges for the naming conventions used in the Catalogue of Endangered Languages are addressed.

The problem of a multiple names applied to a single language The first challenge is that of multiple names associated with a single language, and how these are to be represented in the Catalogue of Endangered Languages. A single language often has multiple names. Given that the Catalogue seeks to provide the most accurate and comprehensive information on endangered languages, the inclusion of multiple language names cannot and should not be avoided, both conventional names (exonyms, xenonyms) and community-recommended names, i.e. self-designations, community-preferred names, or the name of the language in the language itself (autonyms, endonyms). Briefly, a xenonym is nearly the same as an exonym; it is the name of a language given by outsiders, “foreigners,” not the name the language community itself calls their language. An endonym, in contrast, is essentially the same as an autonym, the name members of a language group call their language. For example, Nivaclé [cag] is also widely known by the names Ashluslay and Chulupí (each with several variant spellings), but has also been called at one time or another by several other names: Etehua, Guentusé, Mathlela, Sogciagay, Sowa, and Suhin (most of these with multiple spellings) (see Campbell et al. in preparation; Fabre 2016). “Nivaclé” is the most recent and now the most common of these names, an autonym, from the native word niwakle “man, person, Nivaclé person, Nivaclé language.” “Chulupí” was very common as the name for the language and people, but is now being abandoned because chulupí means “cockroach” in the Spanish of Bolivia and parts of Paraguay, and that is taken as negative (see below for the problem of offensive names). Many languages of South America, for example, have multiple names; often several names for the same language (see Campbell 2012 for numerous examples).

The problem of mismatches between exonyms and autonyms The second challenge is that the names for languages in the academic or popular literature and the names preferred by speakers of the languages often do not match. Some language group members have, in recent times, strongly advocated adoption of their autonyms, and a good 17

Yen-ling Chen and Lyle Campbell

number of these autonyms have come to be accepted in the academic literature: for example, “O’odham” for Papago (and Pima) [ood], “Ohlone” for Costanoan [cst, css], “Kwak’wala” for Kwakiutl [kwk], “Purépecha” for Tarascan [tsz, pua], “Tlahuica” for Ocuilteco [ocu], “Mapudungun” (or “Mapadungu”) for Araucanian (or Mapuche) [arn], among many others. On the other hand, in some cases exonyms have become the common name used by (and sometimes preferred by) community members to refer to their languages; the exonym can become part of the language’s and the community’s cultural heritage, gaining such importance that it in effect becomes their autonym. In a few situations, an exonym has become preferred in order to avoid disagreements among different groups of speakers over competing claims about autonyms. Puyuma [pyu] is an Austronesian language spoken in southeastern Taiwan, specifically in Taitung City (臺東市) and Peinan Township (卑南鄉) in Taitung County (臺東縣). According to Teng (2009:822), Puyuma has three dialects: Nanwang (the most conservative variety), Katipul-Kasavakan, and the rest of the varieties. Puyuma, meaning “unity, concord” or “send to the field,” is an autonym used by the Nanwang community (Teng 2008). However, “Puyuma” is now used as a generic term for all the dialects together. The language Puyuma is also referred to in Mandarin Chinese by the general public and by scholars in Taiwan as Peinan/Beinan (卑南), after the name of the township where the Nanwang variety is spoken. Given the fact that Puyuma is the most recognized name in the academic literature and that it is an autonym per se, this name is maintained (with an explanation of the alternate names) in the Catalogue. Nadouhua is an unclassified Hlai language (Tai-Kadai) spoken in Hainan, China. It was first reported in Z. Fu (1990) as “Nadouhua,” which had around 3,000 speakers. The speakers called themselves /thɛn¹¹ lai¹¹/, dwelling in two villages, Nadou village (那斗村) and Yue village (月村). Both Nadouhua (那斗話) and Laiyu (來語) are names used in the academic literature, with the former originating from the village name and the latter from the autonym. Nadouhua is preferred in scholarly work written in English (for example, Norquest 2007; Ostapirat 2008), whereas Laiyu (來語) is preferred in the Chinese literature (for example, C. Fu 2011; 2012). Both names are listed in the Catalogue, with Nadouhua as a primary name, considering that this name is used more often in the academic literature. In addition, Nadouhua is chosen to avoid confusion with Bolyu (倈語) [ply], an Austroasiatic language spoken in Guangxi, China, because in Mandarin 倈語 is pronounced as Laiyu, which is homophonous with 來語 Laiyu. In a few cases, exonyms are given preference where the autonyms involve difficult or unfamiliar scripts and spellings, making searching for these names difficult if not impossible. For example, Comanche is preferred over native nʉmʉ tekwapʉ; Inuktitut is given with its spelling in Latin letters as the primary name, with its spelling in syllabics, ᐃᓄᒃᑎᑐᑦ, as an alternative name. Names with special characters can be exceptionally difficult to search for, particularly if the users do not have an idea of what to look for, how to find such scripts on their computers, are unable to display these scripts in their web browsers, etc.

The problem of multiple autonyms The third challenge is that there can be more than one preferred name for a language among its speakers, even though linguistically it is the same language. Often names representing just one dialect or region (or clan) are championed to represent the language as a whole for some non-linguistic reason, especially by some faction in a group. This can lead to conflict among different groups who speak the language. Some, insisting on an autonym that actually only 18

Naming conventions and challenges

refers to a particular variety (dialect), have in this way attempted to get recognition for the dialect as a separate language, often for political motives. One example is a user-submitted recommendation that the Catalogue change the current primary name of the language identified by code [str] from “Senćoŧen” to “Northern Straits Salish,” since Senćoŧen is the name of just one of several dialects of Northern Straits Salish. In addition to Saanich (Senćoŧen), other dialects involved include Samish, Lummi, Songish, Sooke, and Semiahmoo, and motivation for the recommended name change was to encourage the people associated with these other dialects also to feel included. Jejueo [jje] (제주어; 濟州語) “Jeju Language” has several names in the literature, i.e. Jejumal (제주말) or Jejutmal (제줏말) “Jeju Speech,” Jeju Saturi (제주 사투리) “Jeju dialect,” Saturi (사투리) “dialect,” Jeju Bangeon (제주 방언; 濟州方言) or Jejudo Bangeon (제주도 방언; 濟州島方言) “Jeju dialect,” and Jeju Jiyeokeo (제주 지역어; 濟州地域語) “Jeju regional language” (Sejung Yang, personal communication 2016). Taking language preservation and revitalization into consideration, Jejueo (제주어) “Jeju language” was chosen by O’Grady et al. (2014) when applying for an ISO 639-3 code for the language, and Jejueo was adopted as its primary name in the Catalogue. The local government in Jeju also uses the name Jejueo (제주어) in official documents. Pazeh-Kaxabu [uun] serves as another example where different speakers of the same language prefer different names and have separate group identities. Pazeh-Kaxabu, a critically endangered Austronesian language in Taiwan, has two dialects, namely Pazeh and Kaxabu. The Pazeh variety has received more attention in the academic literature since the late 1800s (see Ino 1897; Ogawa 1923; Asai 1937), and study of this variety has been taken as representative of the whole language (see Blust 1999; Lin 2000; Li & Tsuchida 2001; 2002, among others). The Kaxabu variety, on the other hand, was first mentioned in Ferrell (1970), together with Pazeh. Ferrell (1970:73) mentioned that the term Pazeh “ripe, matured” was probably derived from the Chinese term 熟番, literally “cooked savages,” which was used to refer to those Sinicized aborigines in Taiwan, whereas Kaxabu is a place name, the region where the Kaxabu live.4 After the death of the last fluent speaker of the Pazeh dialect, Tata Pan Jin-yu/Phoann Kim-giok (潘金玉) in 2010, Ethnologue mistakenly classified PazehKaxabu as extinct (Lewis et al. 2014, 2015). However, there remain 12 fluent elderly speakers of the Kaxabu dialect, with the oldest being 92 in 2016 (Tatauan Kaxabu a ahan, n.d.; Elizabeth Zeitoun, personal communication 2016).5 Given that the name “Pazeh,” representing the Pazeh group, has occurred more frequently in the academic literature but that the Kaxabu group still has fluent speakers and they prefer “Kaxabu,” the name Pazeh-Kaxabu is used in the Catalogue. (Speakers prefer the spelling “Kaxabu” instead of “Kahabu.”)

Non-linguistic issues associated with language names The fourth challenge is how to handle the socio-political concerns often associated with language names. It is not uncommon for pejorative names to be given to languages by outsiders. Despite Shakespeare’s opinion that “that which we call a rose by any other name would smell as sweet,” many do not find these names sweet but offensive, and may want the offensive names eliminated. Nevertheless, some languages are best known by names that some now consider offensive, and it is for these names that many users of the Catalogue will search in order to find these languages. Therefore, for the Catalogue to be effective, searchable, and comprehensive, the names cannot be removed altogether; however, whenever a name is known to be offensive, the Catalogue includes it with an explanation of the issue, along with a recommendation that this name should not be used. For example, the Nadahup languages of Brazil 19

Yen-ling Chen and Lyle Campbell

and Colombia were previously known as “Makú” languages, from an Arawakan exonym meaning “without speech.” The term Makú is now strongly dispreferred, and the Catalogue’s entries for the Nadahup languages alert readers to this fact. The Berber languages that are spoken in northern Africa (a branch of Afro-Asiatic) present another example. The term “Berber” is an exonym meaning “barbarous” (cf. Greek Barbaroi and Latin barbarous, Brett & Fentress 1997); now many prefer to call these languages “Tamazight” or “Amazigh” languages, based on an autonym from the root mazigh, meaning “free man” or “noble man.” In a similar case, the names “Lolo(ish)” (used in the English literature) and “Yi” (used in the Chinese literature) both refer to the same Tibeto-Burman branch of SinoTibetan.6 Bradley (2004) has proposed “Ngwi” to replace “Lolo(ish)” as the name for that subgroup, since Lolo is pejorative: both characters in the Chinese spelling of the name “Lolo” have the dog radical, 猓 luo and 玀 luo. Ngwi originally meant “silver” but now is used as a widespread autonym within the Ngwi subgroup, which included Sa’nguie [ysy], Samei [smh], Nosu [iii], Sani [ysn], Hani [hni], Akha [ahk], Lisu [lis], Lipo [lpo], etc. Chinese scholars have also recommended the use of 倮 luo, which contains the human radical, as a replacement for 猓 luo and 玀 luo, both with the dog radical. As for the Chinese name “Yi,” Bradley (2004) explains that the government of the People’s Republic of China have adopted the new Chinese spelling, Yi (彞), to refer to the Yi Nationality, replacing the old name Yi (夷), a derogatory term used in Chinese documents during the Ming and Qing dynasties to refer to certain minority groups in the South. Despite the fact that the etymologies of Berber and Lolo/Yi involve derogatory terms, given the widespread use of Berber and Lolo(ish)/Yi in the literature, both names are employed in the Catalogue, especially as the names for subgroups, although also with the newer proposals as alternative names. The exonym/autonym distinction is not always clear, and an autonym may change so much that it ends up becoming an exonym. Wichí [mzh, wlv, mtp] is a Matacoan language of Argentina and Bolivia. The name previously used for the language, “Mataco,” originally was not offensive, but some were uncomfortable with it because it sounded to them like Spanish matar “to kill.” As a result, the name was changed to Wichí, in principle an autonym, but actually closer to an exonym; it is based on the Wichí word /wikyi/, which is the name for the group (an ethnonym). The Wichí name for their language is lhamtes “the language” or wikyi lhamtes “the language of the Wichí” (Terraza 2008:8; Fabre 2016). The names “preferred by” some speakers sometimes are not generally accepted or recognized by most members of the community, but rather are favored (even invented) for particular nonlinguistic purposes. For example, some felt that Kaqchikel (Cakchiquel) [cak] (a Mayan language of Guatemala) was not a true autonym and suggested instead Kaqchikel ch’ab’al, whereas the true autonym is qach’ab’al (literally “our language”), and the truest autonym would be [qač’aɓɨɬ], the phonetic pronunciation of what is spelled qach’ab’al. Sometimes there is also disagreement about which spelling of a name to use, or which script or orthography – Kaqchikel vs. Cakchiquel is a good example. A mismatch between names for linguistic groups and for ethnic groups is not uncommon. It is not uncommon for speakers to adopt the ethnonym given by local authorities as the name for their language. Speakers of Lakkia [lbc] and Bunu [bwn, buh] are classified as belonging to the Yao (瑤) Nationality in China (Ratliff 2010:3), and consider themselves Yao/ Mienic speakers, although Lakkia linguistically belongs to the Tai-Kadai family and Bunu linguistically is Hmongic. In cases such as this, both the linguistic and ethnic names are given in the Catalogue when clarification is needed. Following the above-mentioned case, to avoid 20

Naming conventions and challenges

such confusion, the term Hmong-Mien is adopted in the Catalogue for the language family, instead of Miao-Yao (苗瑤), which is often associated with ethnic, not linguistic classification in China. In sum, naming is not purely linguistically driven, as illustrated by the case studies under discussion. Choices about language names often need to be adjusted accordingly, with sociopolitical factors taken into consideration. Also, it is important to note that the classification and categorization of one’s linguistic community and one’s ethnicity might not always be a one-to-one match.

Conclusion As seen here, decisions about how to list the names of languages in the Catalogue of Endangered Languages must confront several challenges. The overarching principles that ELCat follows are to make the Catalogue as user-friendly and comprehensive as possible, to facilitate the ability to search for and find languages, and to respect the communities whose languages are involved. In following these principles, ELCat attempts to list all the salient names by which a language is or has been known. ELCat identifies where possible and displays the names preferred by the majority of community members (representing the language as a whole rather than any particular variety of the language or subgroup of the overall community). ELCat identifies, where known, names that are considered offensive and recommends against their use. ELCat also displays both external and conventional names (exonyms, xenonyms) and communityrecommended names (self-designations, community-preferred names, or the name of the language in the language itself – autonyms, endonyms).

Notes 1 2

3

4

5

6

By community members, we here loosely mean speakers, people whose heritage language is involved, or people who are otherwise stakeholders in decisions about a language. Throughout this volume, three-letter codes in square brackets following language names represent language codes in the ISO 639-3 standard. Brackets containing multiple codes, e.g. [abc, xyz] indicate that the Catalogue considers two or more ISO 639-3-designated varieties to belong to a single language, one with multiple ISO codes. The term 方言 fangyan used in the Chinese academic literature is often mistranslated as “dialect.” However, fangyan, literally “a speech variety of a place,” should often be regarded as a vernacular or a language, instead of a dialect of Mandarin. Ferrell (1970:73) states that “Pazeh informants at present no longer recognize the derivation of pazeh and state that it is ‘just our tribal name.’ Kahabu is the name of the area where the Kahabu villages are now located, and may possibly be attributed to a different ethnic group living in the area when the Pazeh-Kahabu moved into the P’uli Basin in the early 1800s.” Ethnologue (Simons & Fennig 2017) calls this Kulon-Pazeh and notes: “2 [speakers] (Jen-kuei Li 2013). Last speaker of the Pazeh dialect died in 2010. Only known speakers use Kaxabu dialect (Jenkuei Li 2013).” However, their source, Paul Jen-kuei Li (2013), does not mention that there are only two remaining speakers – the population of the Pazeh-Kaxabu is also larger than two. The name of the Tai-Kadai language family is also controversial, although not derogatory. Ostapirat (2000:19–20) points out that Tai-Kadai is homophonous with “Tai ladder” in Thai, the official language of Thailand. He further explains that “The term /thai kadai/ has often elicited smiles or funny looks from non-linguists (sometimes from linguists as well!) when they first hear it.” For this reason, he proposed the use of “Kra-Dai” where both morphemes are autonyms (see Ostapirat 2000 for more detail). Both “Tai-Kadai” and “Kra-Dai” have been in use in the academic literature.

21

Yen-ling Chen and Lyle Campbell

References Asai, Erin (淺井惠倫). 1937. Basai. (Field notes.) Blust, Robert. 1999. Notes on Pazeh phonology and morphology. Oceanic Linguistics 38(2). 321–365. Bradley, David. 2004. Endangered Central Ngwi languages of northwestern Yunnan. (Paper presented at the 37th International Conference on Sino-Tibetan Languages and Linguistics, Lund University, Sweden, October 2004.) Brett, Michael & Fentress, Elizabeth. 1997. The Berbers. Oxford and Cambridge: Blackwell. Campbell, Lyle. 2012. The classification of South American indigenous languages. In Campbell, Lyle & Grondona, Verónica (eds.), The indigenous languages of South America: A comprehensive guide, 59–166. Berlin: de Gruyter. Campbell, Lyle & Díaz, Luis & Ángel, Fernando. In preparation. Nivaclé Grammar. Fabre, Alain. 2016. Gramática de la lengua nivacle. Munich: LINCOM. Ferrell, Raleigh. 1970. The Pazeh-Kahabu language. 考古人類學刊 [Journal of Archaeology and Anthropology] 31/32. 73–97. Fu, Changzhong (符昌忠). 2011. 來語概況 [On the Lai language]. 民族語文 [Minority Languages of China] 2011(3). 67–81. Fu, Changzhong (符昌忠). 2012. 來語的系屬問題 [The genetic position of the Lai language]. 民族語 文 [Minority Languages of China] 2012(1). 51–55. Fu, Zhennan (符鎮南). 1990. 黎語的方言島—那斗話 [A dialect island of Li - Natouhua]. 民族語文 [Minority Languages of China] 1990(4). 14–18. Ino, Kanori (伊能嘉矩). 1897. 番語集 [Pazeh language]. (Field notes.) Lewis, M. Paul & Simons, Gary F. & Fennig, Charles D. (eds.). 2014. Ethnologue: Languages of the world, seventeenth edition. Dallas, Texas: SIL International. (www.ethnologue.com/17/) Lewis, M. Paul & Simons, Gary F. & Fennig, Charles D. (eds.). 2015. Ethnologue: Languages of the world, eighteenth edition. Dallas, Texas: SIL International. (www.ethnologue.com/18/) Li, Paul Jen-kuei. 2013. The last text of the last Pazih speaker. In Guangshun, Cao & Chappell, Hilary & Djamouri, Redouane & Wiebusch, Thekla (eds.), 《綜古述今 鉤深取極》 [Breaking down the barriers: Interdisciplinary studies in Chinese linguistics and beyond], 1135–1147. Taipei: Institute of Linguistics (Preparatory Office), Academia Sinica. Li, Paul Jen-kuei (李壬癸) & Tsuchida, Shigeru (土田滋). 2001. Pazih dictionary. Taipei: Institute of Linguistics (Preparatory Office), Academia Sinica. Li, Paul Jen-kuei (李壬癸) & Tsuchida, Shigeru (土田滋). 2002. Pazih texts and songs. Taipei: Institute of Linguistics (Preparatory Office), Academia Sinica. Lin, Ying-chin (林英津). 2000. 巴則海語 [On Pazeh]. 2000. Taipei: 遠流出版社 (Yuan-Liou Publishing Co., Ltd.). Norquest, Peter. 2007. A phonological system of Proto-Hlai. University of Arizona. (Doctoral dissertation.) Ogawa, Naoyoshi (小川尚義). 1923. 台灣の蕃語に就て [On the Formosan aboriginal languages]. 台 灣時報 [Taiwan Jiho] 49. 6–23. O’Grady, William & Yang, Changyong & Yang, Sejung. 2014. ISO 639-3 language code request for Jejueo. (www-01.sil.org/iso639-3/cr_files/2014-004_jje.pdf) Ostapirat, Weera. 2008. The Hlai language. In Diller, Anthony V. N. & Edmondson, Jerold A. & Luo, Yongxian (eds.), The Tai-Kadai languages, 623–652. London and New York: Routledge. Ratliff, Martha. 2010. Hmong-Mien language history. Canberra: Pacific Linguistics. Simons, Gary F. & Fennig, Charles D. 2017. Ethnologue: Languages of the world, twentieth edition. (www.ethnologue.com) Tatauan Kaxabu a ahan (噶哈巫語全球資訊網) [The Kaxabu]. (n.d.). (http://kaxabu.weebly.com/ 221342170424043354862448925391.html) Teng, Stacy Fang-Ching. 2008. A reference grammar of Puyuma, an Austronesian language of Taiwan. Pacific linguistics. 595. Canberra: Pacific Linguistics, Research School of Pacific and Asian Studies, Australian National University. Teng, Stacy Fang-ching. 2009. Case syncretism in Puyuma. Language and Linguistics 10(4). 819–844. Terraza, Jimena. 2008. Gramática del wichí: Fonología y morfosintaxis. Montreal: Université du Québec à Montréal. (Doctoral dissertation.)

22

3 LANGUAGE CLASSIFICATION AND CATALOGUING ENDANGERED LANGUAGES Russell Barlow and Lyle Campbell

Introduction A major contribution to result from the creation of the Catalogue of Endangered Languages (ELCat) is a more definitive understanding of the language families of the world and their classification. Surprisingly, most people, including most linguists, have little idea how many independent language families exist. In this chapter we answer the following questions: why is an accurate classification of languages important for endangered languages?; how many language families are there?; why is it so difficult to determine this?; and how have language endangerment and extinction affected the linguistic diversity of the world?

Why is an accurate classification of the world’s language families so important for endangered languages? To understand the linguistic diversity of the world it is necessary to have an accurate understanding of the classification of the world’s languages. Unfortunately, the rapid loss of endangered languages is quickly reducing that diversity. The loss of any language constitutes a monumental loss of scientific information and cultural knowledge; however, the extinction of a whole family of languages is a tragedy far greater in magnitude. Yet loss not only of individual languages but of whole families of languages is what we must contend with. Since languages are dying at a rapid rate, and since the resources available to academics and activists are limited, sadly, it is highly probable that not every language will be documented or have the chance to see revitalization efforts before disappearing. Scholars and granting agencies alike need an accurate classification of the world’s language in order to make judicious decisions about how to invest efforts and where to deploy limited funds in order to achieve the maximum positive results. Although there are a number of factors – social, economic, and political, as well as linguistic – that can enter into decisions about which languages deserve priority in language documentation (see Hauk and Heaton, Chapter 9), language affiliation is an extremely important one, and is the topic of this chapter. A central goal of linguistics is to comprehend the full range of what is possible and impossible in human languages, charting the extent to which natural languages can and do differ while revealing any linguistic universals or general typological tendencies that exist among them. Therefore, it is important to document 23

Russell Barlow and Lyle Campbell

all extant language families before they become extinct. A sample of languages that excludes entire language families (or even primary branches of larger families) is, needless to say, a poor sample for achieving this goal. In order for linguists to make and test hypotheses about what is possible in human languages, and to determine which relationships among traits and constructions are to be expected, attention must be paid to documenting the world’s undescribed language families, isolates, and primary subgroups. Genetic affiliation therefore becomes a very significant criterion on the basis of which to document as much of the range of human languages as is possible, and it is for this reason that ELCat has paid particular attention to providing accurate classification, creating what we believe is the most accurate overall phylogenetic classification of the world’s endangered languages. There are, of course, a number of other compelling reasons for why endangered languages need a proper phylogenetic classification. Peoples whose languages are at risk can benefit from knowing who their linguistic relatives are. For pragmatic purposes, knowledge of a language’s closest relatives can be of very significant benefit in language revitalization. For example, if a speaker community needs specific vocabulary for a revitalization or language education project, but that lexical domain was lost before it could be documented, suitable lexemes can sometimes be reconstructed from knowledge of closely related languages. Not only that, but people in general want to know about their languages’ relationships to other languages – it is invaluable for understanding the history of humankind and the ways different peoples are related through their languages. Each independent language family (including each language isolate) constitutes an independent window on what linguistic traits may be found in the languages of the world, on what is possible in human language, and – through their study – on the potentials and limitations of human cognition. Language isolates constitute an extremely rich laboratory for striving to discover the full scope of what is possible in human language. Similarly, a sole surviving member of a language family whose other languages, now extinct, are undocumented or only poorly attested deserves high priority. Also, documentation of an endangered language of a relatively undocumented subfamily that is a major branch of a larger language family deserves some priority. Conversely, a language that is a very close relative of other well-documented languages or is one of many closely related languages, some of which are relatively unthreatened, is of less urgency for documentation, though of course they too should be documented. For example, if a choice must be made between dedicating documentation efforts and resources to another Grassfields Bantu language (of the large Niger-Congo family) or to an additional Tahitic Polynesian language (of the large Austronesian family) versus an endangered language isolate, then it is incumbent upon us to give documentation priority to the language isolate. Its loss without documentation would have a greater negative impact on understanding the world’s linguistic diversity and of the full range of what is possible in human languages than the loss of a language that has well-documented close relatives. In short, accurate knowledge of the genealogical classification of the world’s languages is crucially important to endangered languages and their documentation. Put in context, there are c.96 language families (including language isolates) that are known to be extinct (i.e., all the languages belonging to each of these families is dormant, none having any known native speakers). That c.96 of the world’s c.407 independent language families constitutes 24% of the world’s linguistic diversity – calculated in terms of language families – lost forever. As just argued, endangered language isolates deserve high priority in decisions about which languages to document and how to deploy resources and research efforts (see Okura 2018). Already, 59 of the world’s c.159 known language isolates are extinct (37%). Moreover, nearly two-thirds of all extinct languages families are language isolates (59 of the 24

Language classifications

aforementioned c.96, i.e., 61%). Most of the remaining language isolates are endangered, many extremely so (see Okura 2018). When ELCat was conceived in 2009 (see Campbell and Belew, Chapter 1), it was charged by its designers and funders with providing more accurate information on the genealogical affiliations of endangered languages, to reflect the consensus views of most historical linguistic specialists within various regions of the world. ELCat has achieved this goal, providing understanding of which languages and language families are likely to become extinct and what impact their loss will have on linguistic science.

The language families of the world The following is the listing of the world’s language families (including language isolates) identified for ELCat. Families in the list with an asterisk (*) following the name are extinct, that is, no language that is a member of these families has any known native speakers. The Catalogue of Endangered Languages does not include information directly on all of these families, but only on those that have endangered languages. The discussion that follows below is based on this list. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

Abinomn (isolate) Abun (isolate) (Wapuan?) Adai* (isolate [unclassified?]) Afra (Usku) (isolate) Afro-Asiatic Aikanã (isolate) Ainu* (isolate) Algic (family) Alsea* (isolate) Amto-Musan (family, two languages) Andaquí* (isolate) Andoque (isolate) Anêm (isolate) Angan (family) Anim (family) Ap Ma (Botin, Kambot, Kambrambo) (isolate) Arafundi (family, three languages) Arara do Rio Branco* (Arara do Beiradão, Mato Grosso Arara) (isolate) Arawakan (family) Arawan (family) Asaba (Suarmin, Duranmin) (isolate) Atacameño (Cunza, Kunza)* (isolate) Atakapa* (isolate, possibly a small family) Athapaskan-Eyak-Tlingit (Na-Dene [narrow sense]) (family) Atlantic-Congo (core of Niger-Congo minus unproven “branches”) (family) Austroasiatic (family) Austronesian (family) Awaké (Uruak, Arutani)* (isolate) Awin-Pa (family, 2 languages) Aymaran (family) 25

Russell Barlow and Lyle Campbell

31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

Bachamal* (isolate, possibly a member of the North Daly family) Baibai-Fas (family, two languages) Baining (family) Baiyamo (isolate) Banaro (isolate) Bangi Me (isolate) Barbacoan (family) Basque (isolate) Bayono-Awbono (family, two languages) Beothuk* (isolate) Berta (family, three languages) Betoi-Jirara* (isolate) Bilua (isolate) Bogaya (isolate) Boran (family) Border (family) Bororoan (family) Bosavi family Bulaka River (family, two languages) Bunaban (family) Burmeso (isolate) Burushaski (isolate) Busa (Odiai) (isolate) Caddoan (family) Cahuapanan (family) Camsá (isolate) Candoshi (Canndoshi-Sharpa) (isolate) Canichana* (isolate) Cariban (family) Cayuse* (isolate) Cayuvava* (Cayubaba) (isolate) Central Sudanic (family) Chapacuran (family) Charruan* (family) Chibchan (family) Chimakuan* (family) Chimariko* (isolate) Chinookan* (family) Chipaya-Uru (Uru-Chipaya) (family) Chiquitano (isolate) Chitimacha* (isolate) Chocoan (family) Cholonan* (family, two languages) Chonan (family) Chono* (isolate) Chukotko-Kamchatkan (family) Chumashan* (family) Coahuilteco* (isolate) 26

Language classifications

79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126.

Cochimí-Yuman (family) Cofán (A’ingaé) (isolate) Comecrudan* (family) Coosan* (family, two languages) Cotoname* (isolate) Cuitlatec* (isolate) Culli (Culle)* (isolate) Dagan (family) Daju (family) Damal (Uhunduni, Amung) (isolate) Dem (isolate) Dibiyaso (isolate) Dizoid (family) Dogon (family) Doso-Turumsa (family, two languages) Dravidian (family) Duna (isolate) East Bird’s Head (family, three languages) East Kutubu (family, two languages) East Strickland (family) Eastern Daly* (family) Eastern Jebel (family) Eastern Trans-Fly (family) Elamite* (isolate) Eleman (family) Elseng (Morwap) (isolate) Eskimo-Aleut (family) Esmeralda* (Atacame) (isolate) Esselen* (isolate) Fasu (isolate) Fulnio (Yate) (isolate) Furan (family) Gaagudju * (isolate) Garrwan (family) Geelvink Bay (Cenderawasih Bay) (family) Giimbiyu *(family) Gimojan (North Omotic?) (family) Goilalan (family) Great Andamanese (family) Guachí* (isolate) Guaicurian* (family) Guaicuruan (family) Guajiboan (family) Guamo* (isolate) Guató* (isolate) Gunwinyguan (family) Guriaso (isolate) Hadza (isolate) 27

Russell Barlow and Lyle Campbell

127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174.

Haida (isolate or small family?) Harákmbut-Katukinan (family) Hatam-Mansim (family, two languages) Hattic* (isolate) Heiban (family) Hruso (Hruso-Aka) (isolate?) Huarpean* (family) Huave (isolate) Hurrian-(Hurro-Urartean)* (family) Ijoid (family) Inanwatan (family, two languages) Indo-European (family) Irantxe (Münkü) (isolate) Iroquoian (family) Itonama (isolate) Iwaidjan (family) Jabutían (family) Jalaa* (isolate) Japonic (family) Jarrakan (family) Jêan (Jê family) Jeikó* (Jeicó, Jaiko) Brazil (isolate) [Macro-Jêan?] (isolate) Jicaquean (Tol) (family) Jirajaran* (family) Jivaroan (family) Jotí (Yuwana) Kadu (Kadugli-Krongo) (family) Kaki Ae (isolate) Kakua-Nukak (family) Kalapuyan* (family) Kamakanan* (family) Kamula (isolate) Kapauri (isolate) (Kapori) Kapixaná (Kanoé (isolate) Karajá language area (family of two languages, possibly an isolate) Karami* (isolate) Karankawa* (isolate) Karirían* (family) Kartvelian (family) Karuk (Karok) (isolate) Kassite* (isolate) Kaure-Narau (family of two languages, possibly an isolate) Kaweskaran (Qawasqaran, Alacalufan) (family, three languages?) Kayagar (family, three languages) Kehu (isolate) Keresan (family, two languages?) Khoe (Khoe-Kwadi?) (Kwadi may be an isolate) (family) Kibiri-Porome (Kibiri) (isolate) 28

Language classifications

175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222.

Kimki (isolate) Kiowa-Tanoan (family) Kiwaian (family) Koiarian (family) Kol (isolate) Kolopom (family, three languages) Koman (family) Konda-Yahadian (family, two languages) Kootenai (Kutenai) (isolate) Koreanic (family) Kosare (isolate) Krenákan (Botocudan) (family) Kresh-Aja (Family) Kuliak (family) Kunama (family) Kungarakany* (isolate) Kuot (isolate) Kusunda (isolate) Kwalean (family, three languages) Kwaza (Koayá) (isolate) Kwerbic (family) Kwomtari (family, two languages) Kx’a (Ju + ‡Huan) (family) Laal (isolate) Lakes Plain (family) Lavukaleve (isolate) Leco* (isolate) Left May (Arai) (family) Lencan* (family, two languages) Lepki-Murkim (family, 2 languages) Limilngan* (family) Lower Sepik-Ramu (family) Lule-Vilelan* (family) Maban (family) Maiduan (family) Mailuan (family) Mairasi (family, three languages) Máko* (Maku) (isolate) Mande (family) Mangarrayi* (isolate) Maningrida (family, three languages) Manubaran (family, two languages) Mao (family) (Africa) Mapudungun (Araucanian) (family, two languages) Maran (family) Marori (Moraori) (isolate) Marrku-Wurrugu (family) Mascoyan (family) 29

Russell Barlow and Lyle Campbell

223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270.

Masep (isolate) Matacoan (family) Matanawí* (isolate) Mawes (isolate) Maxakalían (family) Mayan (family) Maybrat (isolate) Miao-Yao (Hmong-Mien) (family) Mirndi (Mindi) (family) Misumalpan (family) Mixe-Zoquean (family) Mochica (Yunga)* (isolate) Mombum (family, two languages) Ulmapo (family, three languages) (formerly also called “Mongol-Langam”) Mongolian (family) Monumbo (family, two languages) Mor (isolate) Morehead-Wasur (family) Mosetén-Chinamé (isolate) Movima (isolate) Mpur (isolate) Munichi* (Muniche) isolate) Muskogean (family) Nadehup (Nadahup, “Makúan”) (family) Nakh-Dagestanian (family) Nambiquaran (family) Namla-Tofanma (family, two languages) Nara (isolate?) Narrow Talodi (family) Natchez* (isolate) Ndu (family) Niger-Congo (family) Nihali (isolate) Nilotic (family) Nimboran (family) Nivkh (isolate, possibly a small family of two languages) North Bougainville (family) North-Eastern Tasmanian* (family) North Halmahera (family) Northern Daly (Daly) (family) Northwest Caucasian (family) Nubian (+Meroitic) (family) Nyimang (family, two languages) (Africa) Nyulnyulan (family) Ofayé (Opayé) (isolate) Omurano* (isolate) Onge-Jarawa (Jarawa-Onge) (family, two languages, Andaman Islands) Ongota (isolate?) 30

Language classifications

271. 272. 273. 274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318.

Otomacoan* (small family) Otomanguean (family) Oyster Bay* (small family?) Paez(an) (isolate; possibly a small family?) Pahoturi (family, two languages) Palaihnihan (family) Pama-Nyungan (family) Pano-Tacanan (family) Pauwasi (family) Pawaia (isolate) Payaguá* (isolate) Pele-Ata (isolate) Piawi (family, two languages) Pirahã (Muran) (isolate, possibly a small family of closely related languages) Plateau (Plateau Penutian) (family) Pomoan (family) Powle-Ma (“Molof”) (isolate) Puinave (isolate) Puquina* (isolate) Purari (formerly “Namau”) (isolate) Purí-Coroado* (isolate) Pyu (isolate) Quechuan (family) Rashad (family, two languages) Rikbaktsá (Canoeiro) (isolate) Saharan (family) Sáliban (family) Salinan* (family, two languages) Salishan (family) Sandawe (isolate?, disputed) Sapé (Kaliana)* (isolate) Sause (isolate) Savosavo (isolate) Sechura-Catacaoan* (family) Senagi (family, two languages) Sentani (family, four languages) Sepik (family) Seri (isolate) Shastan* (family) Sino-Tibetan (family) Siouan-Catawban (family) Siuslaw* (isolate) Sko (Skou) (family) Somahai (Momuna) (family, two languages) Songhay (family) South Bird’s Head (family) South Bougainville (family) Southern Daly (family) 31

Russell Barlow and Lyle Campbell

319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332. 333. 334. 335. 336. 337. 338. 339. 340. 341. 342. 343. 344. 345. 346. 347. 348. 349. 350. 351. 352. 353. 354. 355. 356. 357. 358. 359. 360. 361. 362. 363. 364. 365. 366.

South-Eastern Tasmanian* (family, two languages?) South Omotic (Aroid?) (family) Suki-Gogodala (family) Sulka (isolate) Sumerian (isolate)* Surmic (family) Tabo (Waia) (isolate) Tai-Kadai (family) Taiap (isolate) Takelma* (isolate) Tama (Taman) (family) Tambora (isolate)* Tanahmerah (isolate) Tangkic (family) Ta-Ne-Omotic (family) (?) Tarascan (Purépecha) (isolate) Taruma (Taruamá)* (isolate) Taulil-Butam* (family, two languages) Taushiro (Pinchi) (isolate) Teberan (family, two languages) Tegem (Lafofa) (Unclassified?, isolate?, family, or member of Niger-Congo?) Temein (family, two languages) Tequiraca* (isolate) Tequistlatecan (family, three languages) Tikuna-Yurí (family) Timor-Alor-Pantar (family) Timotean* (small family) Timucuan* (small family) Tiniguan* (small family) Tiwi (isolate) Tonkawa* (isolate) Tor-Orya (family) Torricelli (family) Totonacan (family) Touo (isolate) Trans New Guinea (family) Trumai (isolate) Tsimshianic (family) Tukanoan (family) Tungusic (family) Tunica* (isolate) Tupían (family) Turama-Kikori (family) Turkic (family) Tuu (family) Tyrsenian (Etruscan + Lemnian)* (family) Umbugarla/Ngurmbur* (isolate or small family?) Uralic (family) 32

Language classifications

367. 368. 369. 370. 371. 372. 373. 374. 375. 376. 377. 378. 379. 380. 381. 382. 383. 384. 385. 386. 387. 388. 389. 390. 391. 392. 393. 394. 395. 396. 397. 398. 399. 400. 401. 402. 403. 404. 405. 406. 407.

Urarina (isolate) Utian (Miwok-Costanoan) (family) Uto-Aztecan (family) Wagiman (Wageman)* (isolate) Wakashan (family) Walio (family) Waorani (isolate) Warao (isolate) Wardaman* (Yangmanic?) (isolate or small family) Washo (isolate) West Bird’s Head (family) West Bomberai (family, three languages) Western Daly (family) Wintuan (family) Wiru (isolate) Witotoan (family) Worrorran (family) Xinkan (Xincan)* (family, four languages) Xukurúan* (family) Yagan (Yámana) (isolate) Yaguan (Peba-Yagua) Yale (Yalë, Nagatman) (isolate) Yana* (isolate) Yanomaman (family) Yareban (family) Yaruro (Pumé) (isolate) Yawa (family, two languages) Yele (Yélî Dnye) (isolate) Yeniseian (family) Yerakai (isolate) Yetfa-Biksi (isolate) Yokutsan (family) Yuat (family) Yuchi (isolate) Yukaghir (family) Yukian* (family) Yuracaré (isolate) Yurumanguí* (isolate) Zamucoan (family) Zaparoan (family) Zuni (isolate)

(For discussion of the uncertain or disputed cases, see Campbell 2013; 2018; in press a; and Campbell et al. in press.) The answer, then, to the question of how many language families there are in the world is c.407. However, this is far from an absolute, concrete number; it is just the best we can do in current circumstances, for several reasons (see below). 33

Russell Barlow and Lyle Campbell

Currently (as of April 4, 2017), ELCat includes languages representing 219 language families. This figure includes 121 language isolates and 163 multiple-language families. (Language isolates are language families composed of only a single member language.) This list does not include sign languages, mixed languages, or pidgin and creole languages – ELCat lists these languages in their own categories, separate from the language families. The Catalogue includes 67 sign languages, 6 mixed languages (one of them dormant), and 25 pidgins and creoles (two of them dormant) among the world’s endangered languages. Finally, there are 25 languages that, due to insufficient data, are listed as “unclassified.” (More on these below.) Perhaps unsurprisingly, the families with the most endangered and dormant languages listed are also some of the largest families in the world: Austronesian (444 endangered languages listed in ELCat), Niger-Congo (370 listed), Sino-Tibetan (251 listed), Pama-Nyungan (255 listed), Indo-European (186 listed), Afro-Asiatic (172 listed), Austro-Asiatic (96 listed), and Trans-New Guinea (94 listed). We hasten to point out that there are numerous extinct languages in these families that do not show up in the Catalogue as dormant or awakening because they became extinct before c.1960.1 The only comparable attempt to determine the language families of the world is that of Glottolog (Hammarström et al. 2016). While Glottolog’s classification (with a total of 422 language families and isolates) is similar to ELCat’s classification, there are significant differences. Some families differ in the two only by the names used to designate them. Some entities listed as families in Glottolog are separated into more than one family in ELCat, and a few families that are separate in Glottolog are joined into single larger families in ELCat. There are also a few differences involving some languages considered unclassified in ELCat but classified as belonging to a given family in Glottolog. On the whole, however, the two classifications are similar, though arrived at mostly independently. Recent editions of Ethnologue (www.ethnologue.com/browse/families) have improved their classification markedly over that of previous editions. When ELCat was launched in 2012, the then-current edition of Ethnologue had a total of only 78 families. The current edition (Simons & Fennig 2017) lists 232 (146 language families and 86 language isolates). There are numerous differences between this classification and that of ELCat, not just the difference in total number of language families (including isolates). The language “genealogy” given in World Atlas of Linguistic Structures (WALS, http://wals.info/languoid/genealogy) is somewhat similar to that of Ethnologue, with only 257 total groupings, again with numerous differences from the ELCat classification.

Unknowns in language classification Let us turn now to the matter of why ELCat’s total number of c.407 language families (including isolates) must be considered relative rather than absolute.

Unclassified languages There are a good number of unclassified, indeed unclassifiable, languages whose genetic affiliation is unknown. There are two sorts of unclassified languages. The first are those extinct languages that are too poorly attested to be grouped with any other language or language family. Most of these have been extinct for so long that they are not registered in ELCat. A few examples include: 34

Language classifications

Aranama-Tamique, Texas Baenan, Brazil Camunico, Northeast Italy (survived to 2nd half of 1st millennium BCE) Eteocretan, Crete, 7–3 centuries BCE Gamela, Brazil Gule, Sudan Kaskean, Northeast Anatolia 2nd millennium BCE Maratino, Northeast Mexico Minkin, Australia Mure, Bolivia Naolan, Tamaulipas, Mexico Pictish, Scotland 7–10 centuries CE, few inscriptions Solano, Texas, Northeast Mexico Sorothaptic, Iberian Peninsula, pre-Celtic, Bronze Age Tarairiú, Brazil (For several other unclassified extinct languages of Asia and Europe, see Michalowski 2018.) The second kind of unclassified languages consists of extant languages that cannot be classified for lack of data, i.e., languages not yet described sufficiently to compare them meaningfully with other languages in order to determine whether they may have relatives. We mention a few examples. In Africa: Bung, Lufu, Kujargé, and perhaps Mpre (Mpra) (Blench 2018). In Asia and the Pacific: Sentinelese, Bhatola, Waxianghua, Turumsa, Kembra, and Lepki. In South America there are many, for example: Ewarhuyana, Kaimbé, Kambiwá, Kapinawá, Pankararé, Truká, Tremembé, Wakoná, Wasu, etc. (See Campbell 2012; Zamponi in press; cf. Seifart and Hammarström 2018.)

Dialect vs. language Another reason that the total number of language families is uncertain has to do with the vexed question of separate languages vs. dialects of single languages. An entity with dialects (variants of a single language) is a single language, and can be a language isolate if it has no known relatives. However, if there is sufficient diversification that the variants are not mutually intelligible, and hence are considered distinct languages, then what is at stake is a family of related languages rather than a single language with multiple dialects. Since the boundary between relatively divergent dialects of a single language and closely related languages is sometimes difficult to determine, there are instances of disagreement among specialists, where some see a single language with dialects but others see a family of related languages. Clearly differences of opinion about language vs. dialect can influence interpretations of the number of language families if the language in question is an isolate, and also can determine how the languages in these families relate to one another. ELCat attempts to determine whether an entity deserves independent language status or dialect status, but in cases where opinion varies and the answer is unclear, ELCat gives the disputed element its own entry with a discussion of the range of opinion about its status. Some of these disputed entities may be language isolates.

“Uncontacted” groups Another reason that the exact number of distinct language families is not known has to do with the “uncontacted” peoples around the world, especially in Amazonia in South America. It was 35

Russell Barlow and Lyle Campbell

estimated in 2013 that there were some 100 uncontacted groups around the world, most in South America, but some also in Papua New Guinea, Central Africa, and elsewhere (Holmes 2013). For example, in Brazil alone there are at least 40 “uncontacted” isolated indigenous groups; some estimate upwards of 70 in Brazil, and around 90 in South America as a whole (see, for example, Holmes 2013; see also https://en.wikipedia.org/wiki/Uncontacted_peoples# South_America, accessed 5-20-2017). For some of these uncontacted groups, it is not known whether they speak an already identified language, a language currently unknown but which may belong to a known language family, or a language that represents an as yet unknown language family (a language isolate). Since the languages of these uncontacted groups are unknown, obviously most of them do not show up in the entries in the Catalogue of Endangered Languages. Another reason we will never know for sure the exact number of language families has to do with the languages of undeciphered scripts, for example the language Linear A (Minoan). (For a list of undeciphered writing systems, see https://en.wikipedia.org/wiki/Undeciphered_ writing_systems.) These scripts could represent known languages, or unknown languages that nevertheless belong to known language families, or unknown languages that belong to unknown language families. Until and unless they are deciphered, we will not know whether additional, previously uncounted language families are involved.

Language isolates Each language isolate is in fact an individual language family, albeit a family of but a single member language. The endangered language isolates are listed in ELCat under the classification heading “isolate,” followed by their geographical region. Nevertheless, each is considered an independent language family. (See Campbell 2018 for discussion of why language isolates are not very different from language families with multiple member languages.) There are vastly more languages isolates than most people, including most linguists, are aware of – c.159. That is, 39% of the world’s c.407 independent language families are language isolates. The list of the world’s language isolates follows. ELCat includes only 86 of these, since not all isolates are endangered. The extinct and dormant ones are indicated with an asterisk (*) following the name.

Africa [6?] Bangi Me Hadza Jalaa* Laal Ongota (?) (often classified as an isolate, though Blench (2018) considers its status undecided, with a possible Afroasiatic affiliation as reasonable) Sandawe (?) (now usually considered an isolate, Blench (2018) considers its status undecided) Blench (2018) lists the first four of these as definite isolates; he lists 12 languages that have been reported or suggested as isolates, but which are controversial: Bēosi (Madagascar), Dompo, Guanche (Canary Islands), Gumuz, Kujargé (unclassified), Kwadi (unclassified)2, Meroitic (now generally seen as a relative of Nubian), Mpra (perhaps unclassified), Ongota, Oropom (unclassified), Sandawe, and Shabo. 36

Language classifications

Asia [10] (see Georg 2018, Michalowski 2018) Ainu Burushaski Elamite* Kassite* (unclassified?) Hattian* (Hattic) Hruso(?) Kusunda Nihali Nivkh (Gilyak)3 Sumerian* Greater Andamanese (if it is a single language) may be another isolate, but this is uncertain (Georg 2018). Sentinelese, on North Sentinel Island, Andaman Islands, is sometimes listed as an isolate, but its status is unknown and so it is better considered unclassified (Georg 2018).

Australia [7] (see Bowern 2018) Bachamal* Gaagudju* Kungarakany* Mangarrayi* Tiwi Umbugarla/Ngurmbur* Wagiman*

Europe [1] Basque For discussion of several unclassified languages of Europe, see Michalowski (2018).

Pacific [c.54] (see Hammarström 2018) Abinomn Abun (Wapuan?) Afra (Usku) Anêm Ap Ma (Botin, Kambot, Kambrambo) Asaba (Suarmin) Baiyamo Banaro Bilua Bogaya Burmeso Busa (Odiai) 37

Russell Barlow and Lyle Campbell

Damal (Uhunduni, Amung) Dem Dibiyaso Duna Elseng (Morwap) Fasu Guriaso Kaki Ae Kamula Kapauri Karami* Kehu Kibiri-Porome Kimki Kol Kosare Kuot Lavukaleve Marori (Moraori) Masep Mawes Maybrat Mor Mpur Pawaia Pele-Ata Powle-Ma (“Molof”) Purari (formerly “Namau”) Pyu Sause Savosavo Sulka Tabo (Waia) Taiap Tambora* Tanahmerah Touo Wiru Yale (Yalë, Nagatman) Yele (Yélî Dnye) Yerakai Yetfa-Biksi Some of these languages could possibly be considered unclassified, rather than true isolates, given the limited information available on them.

38

Language classifications

North America [23] (cf. Mithun 2018; Campbell et al. in press) Adai*(?) Alsea* Atakapa* Beothuk* Cayuse* Chimariko* Chitimacha* Coahuilteco* Cotoname* Esselen* Haida (?) Karankawa* Karuk Kootenai Natchez* Siuslaw* Takelma* Tonkawa* Tunica* Washo Yuchi (Euchee) Yana* Zuni The number of languages listed as isolates in North America varies considerably in different publications, and the differences illustrate some of the difficulties in determining whether something is or is not a language isolate, with consequences for what the total number of language isolates in the world is considered to be. (See Campbell 2018 for details.)

Mexico and Central America [4] (see Heaton 2018) Cuitlatec* Huave Purhépecha (Tarascan) Seri

South America [54] (see Seifart and Hammarstöm 2018; Zamponi 2018; Campbell in press b) Aikanã (Brazil) Andaquí* (Colombia) Andoque (Brazil, Peru) Arutani (Awaké, Uruak)* (Venezuela, Brazil) Atacameño* (Cunza) (Chile) Betoi* (Betoi-Jirara) (Colombia) 39

Russell Barlow and Lyle Campbell

Camsá (Sibundoy) (Colombia) Candoshi (Candoshi-Shapra) (Peru) Canichana* (Bolivia) Cayubaba (Cayuvava)* (Bolivia) Chiquitano (Bolivia) Chono* (Chile) Cofán (A’ingaé) (Colombia, Ecuador) Culli (Culle)* (Peru; unclassified?) Esmeralda (Atacame)* (Ecuador) Fulniô (Yaté) (Brazil) Guachí* (Brazil) Guamó* (Venezuela) Guató* (Brazil) Irantxe (Iranche, Mynky) (Brazil) Itonama (Bolivia) Jeikó* (Brazil) Jotí (Yuwana) (Venezuela) Kanoé (Kapixaná) (Brazil) Kwaza (Koayá, Koaiá) (Brazil) Leco* (Bolivia, extinct?) Máku* (Máko) (Brazil) Matanawí* (Brazil) Mato Grosso Arara (Arara de Beiradão)* (Brazil) Mochica* (Yunga, Chim) (Peru) Mosetén-Chimané (Bolivia) Movima (Bolivia) Munichi* (Otanabe) (Peru) Muran (Pirahã)(?) (Brazil) Ofayé (Opayé) (Brazil Omurano* (Peru) Paez (Colombia) Payaguá* (Paraguay) Puinave (Colombia, Venezuela) Puquina* (Bolivia) Purí-Coroado* (Brazil) Rikbaktsá (Brazil) Sapé (Kaliana)(*?) (Venezuela) Taruma* (Brazil, Guyana?) Taushiro (Pinche) (Peru) Tequiraca* (Aewa, Auishiri) (Peru) Trumai (Trumaí) (Brazil) Urarina (Peru) Waorani (Auca, Sabela) (Ecuador) Warao (Guyana, Surinam, Venezuela) Yámana* (Yagan) (Chile) Yaruro (Pumé) (Venezuela) Yuracaré (Bolivia) Yurumanguí* (Colombia) 40

Language classifications

However, again, our figure of c.159 is far from an absolutely concrete answer to the question of how many language isolates there are. The reasons for this uncertainty are the same ones as those for uncertainty about the total number of language families, e.g. unclassified languages, the question of separate language vs. dialects of a single language, “uncontacted” languages, and so on. (See Campbell 2018; Okura 2018, for details.) We note that c.59 of the isolates are dormant (no longer spoken), and that 119 of them are endangered.

Other, non-language-family classificatory categories in ELCat Sign languages Unfortunately, sign languages are usually left out of lists of language families. In ELCat, sign languages are given under the classification heading “sign language,” followed by their geographical region. In general, the classification of sign languages into families does not follow the same principles as those for classification of spoken languages. Rather than being based on the criterion of shared innovations among more closely related languages, sign languages tend to be classified instead according to which major sign languages may have influenced them most, thus not distinguishing what is inherited from what is borrowed as is done in the classification of spoken languages. Since the comparative method (or some equivalent thereof) has yet to be systematically applied to sign languages, their proper genealogical classification, if there ever should be one, remains largely unknown. There are 67 endangered sign languages in ELCat. There are 29 endangered sign languages in the Americas, 15 in Europe, 13 in Southeast Asia, 4 in Africa, 2 in South Asia, 1 each in Australia, the Near East, and the Pacific, and none in the Caucasus.

Pidgins and creoles Pidgins and creoles are treated in ELCat as their own category in the classification schema, since – according to the predominant view of their genesis – they do not belong to a single language family. They are also treated together as a single category in ELCat because it is not always a simple matter deciding whether a given language is a pidgin or is a creole. Instead of listing subfamilies under the heading “pidgin or creole,” ELCat indicates the dominant lexifier language. This reveals information of the sort that, of 25 endangered pidgins and creoles, 6 of these have English as the primary lexifier language, 4 have Dutch, 2 French, 2 Portuguese, 2 Hausa, 2 Arabic, and a separate language for each of the others. Perhaps even more interesting is the fact that not all pidgin or creole languages in the Catalogue have a European lexifier language (contrary to some earlier claims that were made about creoles). There are, for example, the Arabic-based creole Nubi [kcn], Chinook/Nootka-based Chinook Wawa (Chinook Jargon) [chn], and several others.

Mixed languages ELCat has a classification heading “mixed language” for the endangered mixed languages of the world, which are not numerous (there are few mixed languages in general, endangered or otherwise). Mixed languages also do not fit neatly into any single family, as they are the result of the fusion of two (or more) identifiable source languages. Accordingly, they are treated 41

Russell Barlow and Lyle Campbell

as a separate category in the Catalogue. There are five endangered mixed languages in ELCat (plus one dormant one). Included in their classification are the languages (or language families) contributing to their makeup. For example, Michif [crg] is classified as “mixed language, French+Cree” and Tagdal [tda] is classified as “mixed language, Songhay+Tuareg.” The others are: Copper Island Aleut (Mednyj Aleut) (Russian+Aleut), Kallawaya (Quechua+ Puquina), Mbugu (Ma’a) (Bantu+Cushitic), and Wutunhua (Wutun) (Chinese+TibetanMongolian).

Unclassified languages Another non-family classificatory category in ELCat is that of “unclassified languages.” Like isolates, these are languages that have not been shown to be related to any other known languages. Unlike isolates, however, unclassified languages have insufficient information about them for them to be classified. In other words, to qualify as an isolate, a language must have enough documentation to be compared meaningfully with other languages but where the evidence supports no genetic affiliation. To say that two languages are “related” means that they are demonstrably related; to say that two languages are “not related,” however, only means that there is not sufficient evidence to prove genetic affiliation, not that they have definitively not descended from a common proto-language. Included for each unclassified language is mention of any leading hypotheses for possible genetic affiliation (followed by a question mark). These are presented regardless of the paucity of evidence. When no such proposal (contested or not) has been put forth, included in the classification is the geographically closest language family (again followed by a question mark). This enables – at the very least – some basic reference for where the unclassified language is spoken. In no case, however, does the Catalogue make any claim assigning the unclassified language to the family name given with a following question mark. For example, Ongota [bxe] is classified in ELCat as “unclassified; Afro-Asiatic?”. Ethnologue classifies the language as “Afro-Asiatic” (but as “unclassified” within this family) (Simons & Fennig 2017). While a number of linguists do suspect that Ongota belongs to the Afro-Asiatic family (the South Omotic branch according to Ehret, cited in Savà & Tosco 2007; the Cushitic branch according to Bender, cited in Savà & Tosco 2007; the East Cushitic branch according to Savà & Tosco 2003; or an independent branch of the family according to Fleming 2006), there is much debate about this language, and other proposals have been made, including a Nilo-Saharan affiliation (Blažek 2007) and that Ongota is a language isolate (Mous 2003). Thus, it is most prudent to list the language as “unclassified,” but to provide the leading hypothesis for its possible genetic affiliation (Afro-Asiatic). Himarimã [hir], on the other hand, is so poorly known to linguists that it is impossible to speculate on its genetic affiliation (Ethnologue classifies it simply as “unclassified”). Indeed, the Himarimã people of the Brazilian Amazon have remained mostly uncontacted. In ELCat the language is classified as “unclassified; Arawan?”. While it is not now possible to prove Arawan affiliation (or even that Himarimã is distinct from other known languages), this designation at least reveals something of the geographical position of the Himarimã language – spoken between two Arawan languages (Paumarí [pad] and Jamamadí [jaa]).

42

Language classifications

ELCat’s classificatory practices Much care and attention has been paid to making the classifications in the ELCat entries as informative, concise, accurate, and consistent as possible. ELCat’s classification keeps to the following principles, some touched on above. ELCat seeks to reflect consensus classifications, where there is general agreement about the status of particular language families and about the family membership of particular languages. The Catalogue’s overarching approach in classifying the languages of the world into families and subgroups has been to follow the most current and most defensible claims in the linguistic literature specialized for that region or family. The failure to consider region- or familyspecific scholarship has been a major flaw in many other attempts at broad-scale genetic classification, resulting in oft-repeated classifications in the general linguistic literature that are not supported (and that have been refuted in some cases) by the best linguists conducting specialized research in the region. ELCat attempts to avoid controversial claims about classification wherever possible, but does allow for comments about uncertain or disputed cases, following the best region/familyspecific scholarship, where possible. As mentioned above, Ethnologue (Simons & Fennig 2017) “lumps” together a number of entities that ELCat finds to be distinct language families, whereas Glottolog separates a number of entities that ELCat finds to be supported language family classifications. An illustrative case is the spurious “Altaic” family, which was postulated in the nineteenth century, primarily based on the presence of typological features shared among the putatively related languages (Campbell & Poser 2008:235–243). Modern comparative linguists now know, of course, that typological traits can and do spread to unrelated families in linguistic areas and that systematic correspondences are necessary to establish that languages are genetically related. Thus the consensus among linguistic specialists in central Eurasian languages is to reject the hypothesized “Altaic” family, with its the putative grouping of Turkic, Mongolic, and Tungusic, to which some add also Korean and Japanese (Japonic) (Campbell & Poser 2008; Pereltsvaig 2012, among others). Nevertheless, “Altaic” has persisted in much of the general linguistic literature. For example, the 16th edition of Ethnologue (2009), which was the most recent version when ELCat started in 2011, includes an “Altaic” family consisting of 66 languages. This erroneous classification persisted into the 17th edition (2014); it was corrected in the 19th edition (2016), though a search of the 2016 Ethnologue still turned up several places where “Altaic” appeared in particular contexts; this has been corrected in the current (20th) edition (see Simons & Fennig 2017). However, by accurately categorizing these languages as belonging to three separate families – Turkic, Tungusic, and Mongolic – ELCat not only does justice to the linguistic facts of central Eurasian languages, it allows attention to be drawn to the actual levels of endangerment of entire language families. Thus, it can be seen that all members of the Tungusic family (16 languages in ELCat) are endangered. Were Tungusic to be grouped with Turkic and Mongolic, however, this picture of language-family-level endangerment would be obscured, because not all presumed “Altaic” languages are endangered – the Turkic languages Turkish [tur] and Azerbaijani [azj, azb], for example, are quite vibrant, as is the Mongolic language Mongolian [khk, mvf]. Thus the proper genetic classification of languages can reveal which families are more endangered, enabling documentation or revitalization efforts to be directed accordingly. Another telling example is the proposed “Nilo-Saharan” family. While some (e.g., Bender 1996, 2000; Ehret 2001) do support a Nilo-Saharan unity, it is generally recognized as “the least widely accepted” of Joseph Greenberg’s four African phyla (Bender 2000:43). As Fleming (1987:168–169) describes it: 43

Russell Barlow and Lyle Campbell

It [Nilo-Saharan] has also been called “Greenberg’s waste basket,” hence a collection of hard-to-classify languages and a very unreliable entity as a phylum. Vis-à-vis AA [Afroasiatic] or N-K [Niger-Kordofanian], N-S [Nilo-Saharan] is widely viewed as the more shaky of the three. (See also Mikkola 1998:71; 1999:130; Campbell & Poser 2008:132–136). By some counts, the languages putatively comprising “Nilo-Sarahan” belong to no fewer than 20 different families (including three large languages often considered isolates that are not endangered: Gumuz [guk], Kunama [kun], and Nara [nrb]). The grouping of these languages together, however, creates the false appearance of diminished linguistic diversity in the region, potentially to the detriment of research efforts. Even if it were someday revealed that some or all of these families and isolates do belong to a single higher-order grouping, the fact that this possible classification has proven so contentious indicates great internal diversity. Besides, our best hopes of reaching a satisfactory conclusion regarding the classification of these languages would be greatly improved by better documenting many of putative Nilo-Saharan’s lesser-studied languages. By maintaining a relatively conservative classification (compared to, say, that of Ethnologue, which includes a “Nilo-Saharan” family of 207 members), ELCat helps to point attention to these small yet crucial endangered language groups. Great attention has also been paid to the proper designation of subgroups to which a language may belong. This is especially important for some of the larger families (such as IndoEuropean or Austronesian) that have a great time depth and whose primary branches represent great linguistic (and even geographic) diversity. It has thus been deemed essential to include primary subgroups for language families in the Catalogue. That said, there was a principled decision made not to include every possible level of subfamily designation. While such detailed subcategorization is often informative for the historical development of different branches (and subbranches) within a family tree, it can obscure the picture of which are the major nodes in that tree. Also, these finely detailed groupings are often contentious, and it is difficult (if not impossible) to obtain a consensus classification. (Finally, from a stylistic standpoint, such lengthy lists of levels of subgrouping for larger families create a daunting mess on a website.) Therefore, a limit has been placed, such that usually no more than three nodes (levels of inclusion), occasionally four levels, between the proto language and the language in question are included in any language’s classification in ELCat. These are selected to be the most important (and widely accepted) branches in the family. Many smaller families do not have this many levels in their subgrouping; but even for the larger families (which may indeed have many levels in some classifications), three is sufficient to convey the clearest and most important information about a language’s genetic affiliations and family-internal relationships. Thus, for example, Mócheno [mhn] is categorized in ELCat as “Indo-European, Germanic, West Germanic, High German)”. The website Glottolog, on the other hand, which seeks to provide more detail in its classifications, categorizes Mócheno as: Indo-European, Germanic, Northwest Germanic, West Germanic, High German, Middle-Modern High German, Modern High German, Alpine Germanic, Bayerisch While such a detailed classification may provide more historical information, it also runs the risk of including nodes that are not established (such as “Northwest Germanic”). The ELCat categorizations strive for a balance between detail of information on the one hand and practicality and consensus on the other. Rapa [ray] is another example to illustrate the point. Rapa is categorized in Glottolog as: 44

Language classifications

Austronesian, Nuclear Austronesian, Malayo-Polynesian, Central-Eastern MalayoPolynesian, Eastern Malayo-Polynesian, Oceanic, Central Pacific linkage, TokelauFijian, Polynesian, Nuclear Polynesian, Northern Outlier Polynesian-East Polynesian, Solomons Northern Outlier Polynesian-East Polynesian, Central Northern Outlier Polynesian-East Polynesian, East Polynesian, Central Eastern Polynesian, MangaiaOld Rapa While this classification is certainly impressive (if accurate), it is nevertheless more useful for our purposes to categorize Rapa in ELCat as: “Austronesian, Malayo-Polynesian, Oceanic, Polynesian.” But the issue of subclassification is not merely a stylistic one. The proper classification of the primary subgroups of large families is crucial in shaping our understanding of endangered languages. We take as examples of the perils of improper subgrouping a few cases from Austronesian, the world’s second-largest family (after Niger-Congo) in terms of number of languages. The largest and most diverse subgroup of Austronesian is Malayo-Polynesian, with over 1,200 languages, containing every Austronesian language not spoken on the island of Taiwan. Of this subgroup, ELCat includes among its endangered Austronesian languages four which each constitute the only member of a primary branch of the immense Malayo-Polynesian subfamily. For the reasons cited above, these four languages can thus be singled out as critic-ally important for our understanding – both diachronic and synchronic – of the Austronesian family and, in turn, of the world’s languages. For comparison, we turn again to Ethnologue (Simons & Fennig 2017), to see that “over-classification” can obfuscate the linguistic picture. Of these four languages – Ati [atk], Chamorro [cha], Enggano [eno], and Kowiai [kwh] – only Chamorro is classified by Ethnologue as constituting its own primary branch of Malayo-Polynesian. Ati is classified as “Austronesian, Malayo-Polynesian, Philippine, Greater Central Philippine, Central Philippine, Bisayan, Central, Peripheral,” giving the false appearance of closer affinity with a large number of other Austronesian languages. Likewise, Kowiai is classified as “Austronesian, Malayo-Polynesian, Central-Eastern Malayo-Polynesian, South Bomberai.”4 Ethnologue misses the mark in the other direction in its classification of Enggano, which is clearly Austronesian, but has not (as of yet) been shown to subgroup with any other Malayo-Polynesian languages (Edwards 2015). Ethnologue mistakenly categorizes Enggano as “unclassified.”

Conclusions In this chapter we attempted to answer several questions: why is an accurate classification of languages important for endangered languages?; how many language families are there in the world?; why is it so difficult to determine this?; and how have language endangerment and extinction affected the linguistic diversity of the world? The research on which the Catalogue of Endangered Languages (ELCat) is based provides answers to these questions, answers presented in this chapter. We found that there are (or were) c.407 independent language families (including language isolates) in the world, and that c.96 of them are extinct, i.e., that no language belonging to any of these families has any known native speakers. This means that essentially a quarter (24%) of the world’s linguistic diversity – calculated in terms of language families – has been lost already. We also found that c.59 of the world’s c.159 language isolates are already extinct (37%). In ELCat, 219 language families are represented because they have at least one endangered language. This figure includes 119 language isolates and 163 multiple-language families. This means that in 58% of the world’s total of c.407 language families have one 45

Russell Barlow and Lyle Campbell

or more endangered languages. Put in current terms, by removing the 96 already completely extinct language families, 219 of the remaining 311 language families have endangered languages: that’s 70% of existing language families that have endangered languages among their members. The actual answer to the question of how many language families there are in the world is complicated by several theoretical and methodological questions fundamental to historical linguistics. We only mention these here; for discussion and details, see Campbell and Poser (2008). These questions include: 1.

2. 3. 4. 5. 6.

Can linguistic diffusion be a serious challenge to linguistic genealogical relationships?, or, how successful in difficult cases can we be at distinguishing inheritance from borrowing? Is there a temporal threshold beyond which genetic relationship among languages is no longer demonstrable? Why have typological comparisons sometimes led to erroneous hypotheses of language families? Why is it that human genetics and other non-linguistic fields cannot help to settle questions of phylogenetic relationships among languages? Is the family tree model flawed? What is the prognosis for discovering new family relationships among languages? What recent progress has been made?

The principal conclusion to this chapter is that ELCat provides a much clearer picture of the language families of the world and their current status. All findings support the observations made throughout this volume that language loss is happening at an alarming rate, with major negative impact on the linguistic diversity of the world.

Notes 1

2 3

4

ELCat continues to list languages that lost their last known speaker after 1960. The reason for including these dormant languages is because it is very difficult in many circumstances to know that in fact there are no longer any speakers; this allows for the benefit of doubt. Several cases are known where a language was declared extinct but then later other speakers turned up. Kwadi is extinct and “its affiliation cannot be resolved” (Blench 2018), although it is treated by some as Khoe. Nivkh (Gilyak) has two fairly divergent varieties, and opinion has varied concerning whether they should be viewed as separate languages or as dialects of a single language. Under the view that multiple languages are involved, the putative small family has been called Amuric (Georg 2018). It should be noted, however, that alternative classifications in the specialized literature (cf. Blust 1993) include Kowiai as an unclassified member of Central-Eastern Malayo-Polynesian. This subgroup, however, is itself vast (over 700 members), and thus any unclassified member should warrant serious linguistic attention.

References Bender, M. Lionel. 1996. The Nilo-Saharan languages: A comparative essay. Munich: Lincom Europa. Bender, M. Lionel. 2000. Nilo-Saharan. In Heine, Bernd & Nurse, Derek (eds.), African languages: An introduction, 43–73. Cambridge: Cambridge University Press. Blažek, Václav. 2007. Nilo-Saharan stratum of Ongota. In Reh, Mechthild & Payne, Doris L. (eds.), Advances in Nilo-Saharan linguistics. Proceedings of the 8th Nilo-Saharan Linguistics Colloquium, University of Hamburg, August 22–25, 2001, 1–10. Köln: Rüdiger Köppe Verlag.

46

Language classifications Blench, Roger. 2018. African language isolates. In Campbell, Lyle (ed.), Language isolates, 162–192. Abingdon and New York: Routledge. Blust, Robert. 1993. Central and Central-Eastern Malayo-Polynesian. Oceanic Linguistics 32. 241–294. Bowern, Claire. 2018. Language isolates of Australia. In Campbell, Lyle (ed.), Language isolates, 323–343. Abingdon and New York: Routledge. Campbell, Lyle. 2012. The classification of South American indigenous languages. In Campbell, Lyle & Grondona, Verónica (eds.), The indigenous languages of South America: A comprehensive guide, 59–166. Berlin: De Gruyter. Campbell, Lyle. 2013. Historical linguistics: An introduction. 3rd edn. Cambridge, MA: MIT Press. Campbell, Lyle. 2018. Language isolates and their history. In Campbell, Lyle (ed.), Language isolates, 1–18. Abingdon and New York: Routledge. Campbell, Lyle. In press a. North America. In Moseley, Christopher J. & Asher, Ronald E. (eds.), Encyclopedia of the world’s endangered languages. London: Routledge. Campbell, Lyle. In press b. Languages of South America. In Moseley, Christopher J. & Asher, Ronald E. (eds.), Atlas of the world’s languages. 2nd edn. London: Routledge. Campbell, Lyle & Golla, Victor & Goddard, Ives & Mithun, Marianne. In press. Languages of North America. In Moseley, Christopher J. & Asher, Ronald E. (eds.), Atlas of the world’s languages. 2nd edn. London: Routledge. Campbell, Lyle & Poser, William. 2008. Language classification: History and method. Cambridge: Cambridge University Press. Edwards, Owen. 2015. The position of Enggano within Austronesian. Oceanic Linguistics 54. 54–109. Ehret, Christopher. 2001. A historical-comparative reconstruction of Nilo-Saharan. Köln: Rüdiger Köppe Verlag. Fleming, Harold. 1987. Review article: Towards a definitive classification of the world’s languages (review of A guide to the world’s languages, by Merritt Ruhlen). Diachronica 4. 159–223. Fleming, Harold. 2006. Ongota: A decisive language in African prehistory (Aethiopistische Forschungen, 64.) Wiesbaden: Otto Harrassowitz. Georg, Stefan. 2018. Other isolated languages of Asia. In Campbell, Lyle (ed.), Language isolates, 139–161. Abingdon and New York: Routledge. Hammarström, Harald. 2018. Language isolates in the New Guinea region. In Campbell, Lyle (ed.), Language isolates, 260–286. Abingdon and New York: Routledge. Hammarström, Harald & Forkel, Robert & Haspelmath, Martin & Bank, Sebastian. 2016. Glottolog 2.7. Jena: Max Planck Institute for the Science of Human History. (http://glottolog.org) Heaton, Raina. 2018. Language isolates of Mesoamerica and Northern Mexico. In Campbell, Lyle (ed.), Language isolates, 229–259. Abingdon and New York: Routledge. Holmes, Bob. 2013. How many uncontacted tribes are there left in the world? New Scientist, August 22, 2013. (www.newscientist.com/article/dn24090-how-many-uncontacted-tribes-are-left-in-the-world/) Simons, Gary F. and Charles D. Fennig (eds.). 2017. Ethnologue: Languages of the world, twentieth edition. Dallas, Texas: SIL International. (www.ethnologue.com) Michalowski, Piotr. 2018. Ancient Near Eastern and European isolates. In Campbell, Lyle (ed.), Language isolates, 19–58. Abingdon and New York: Routledge. Mikkola, Pertti. 1998. Random coincidence in mass comparison: Preliminary analysis of the NiloSaharan lexicon. Nordic Journal of African Studies 7. 63–92. Mikkola, Pertti. 1999. Nilo-Saharan revisited. Nordic Journal of African Studies 8. 108–138. Mithun, Marianne. 2018. Language isolates of North America. In Campbell, Lyle (ed.), Language isolates, 193–228. Abingdon and New York: Routledge. Mous, Maarten. 2003. Loss of linguistic diversity in Africa. In Janse, Mark & Tol, Sijmen (eds.), Language death and language maintenance: Theoretical, practical and descriptive approaches, 157–170. Amsterdam: John Benjamins. Okura, Eve. 2018. Endangered language isolates. In Campbell, Lyle (ed.), Language isolates, 344–371. Abingdon and New York: Routledge. Pereltsvaig, Asya. 2012. Languages of the world: An introduction. Cambridge: Cambridge University Press. Savà, Graziano & Tosco, Mauro. 2003. The classification of Ongota. In Bender, M. Lionel & Takács, Gábor & Appleyard, David L. (eds.), Selected comparative-historical Afrasian linguistic studies in memory of Igor M. Diakonoff, 307–316. Munich: Lincom.

47

Russell Barlow and Lyle Campbell Savà, Graziano & Tosco, Mauro. 2007. Review article of Ongota: A decisive language in African prehistory, by Harold C. Fleming. Aethiopica 10. 223–232. Seifart, Frank & Hammarström, Harald. 2018. Language isolates in South America. In Campbell, Lyle (ed.), Language isolates, 260–286. Abingdon and New York: Routledge. Zamponi, Raoul. In press. Extinct isolates, unclassified languages, and families. In Epps, Patience & Michael, Lev (eds.), Amazonian languages. Berlin: De Gruyter.

48

4 LANGUAGE EXTINCTION THEN AND NOW Anna Belew and Sean Simpson

In this chapter we present an overview of language death, dormancy, and revival in the Catalogue of Endangered Languages. First, we present the Catalogue’s terminology and policies regarding languages with no native speakers. We then discuss what is known about language death in the past, as compared with language loss in recent decades, finding that languages are now being lost at rates unparalleled in known history. Finally, we provide an overview of formerly dormant languages that are being revived, concluding that the steep increase in language revival programs is cause for hope, and we present a listing of dormant and awakening languages in the Catalogue.

Introduction: Language loss at unprecedented rates The introductory material for the Catalogue of Endangered Languages1 notes that today, languages are becoming extinct more rapidly than at any known point in history. This idea is found in much of the current literature and publicity surrounding language endangerment: the current wave of language endangerment and extinction is accelerated, alarming, and unprecedented. In order to sustain these claims, however, it is necessary to compare global language endangerment today with what is known of language loss in the past. An unfortunately common response to the language endangerment crisis, by lay people and linguists alike, is that language death has always happened – that the current loss of global linguistic diversity is neither new nor concerning. Languages have come into existence, been transmitted for a time, and disappeared since humans have been talking; they have always given way to other languages as a result of social, environmental, economic, political, and other pressures. If this is the case, what could be the basis of linguists’ recent concern for endangered languages? Are scholars and language advocates unnecessarily hand-wringing over a natural, inevitable, and age-old process, as if biologists were issuing urgent calls to halt evolution? Is language loss today a normal continuation of human history, wherein languages have been in a constant flux of birth and death? To grasp the urgency of language endangerment today, we must briefly address what is known about language loss in the remote past. We must then consider the known cases of language death and dormancy in more recent history, and how these compare with language death in the past, as well as to the overall amount of linguistic diversity remaining in the world. 49

Anna Belew and Sean Simpson

The patterns emerging today, as seen in the Catalogue’s data, are alarming in their scope and pace: language loss is occurring today at an unprecedented rate.

Dormancy vs. extinction: the rhetoric of language “death” First, it is necessary to clarify the terms used when discussing language loss, and outline the Catalogue’s approach to languages that no longer have fluent speakers. In keeping with much of the literature on language endangerment and loss, throughout this chapter we use the terms “extinction” and “death” to refer to languages that ceased to be spoken long ago and that are not being revived, and “endangered” to refer to languages that are at risk of losing all of their speakers. These terms reflect a view that languages are analogous to biological species – an analogy much used in documentary and historical linguistics, and one that in many ways is quite useful. New languages are typically born via language diversification, a process akin to speciation in biology; languages can be said to occupy certain ecological niches (the destruction of which may threaten their vitality); they may seem to compete with one another for space within their respective environments; and so on. A language’s “life” consists of its use and transmission among a group of people; it seems to compete with other languages for “resources” in attracting speakers; and it may “evolve” by undergoing language change. A language metaphorically “lives” within the minds of fluent speakers and in its use in a speaker population; a natural extension of the biological metaphor is to say that once a language is no longer spoken, it has died – has become extinct. Associating language loss with death, however, introduces a harmful, erroneous assumption: namely, that this state is irreversible. Metaphors aside, languages are not living organisms, and language loss (the metaphorical “death” of a language), thankfully, is not necessarily permanent. With the recognition of the language endangerment crisis and the recent movements for language revitalization, there are a growing number of cases in which languages that have been labeled as “dead” or “extinct” are actually once again gaining new speakers. However, because the rhetoric of “death” is so entrenched within discussion of language endangerment and revitalization, before attempting to revive a so-called “dead” language, it is necessary first to break down the fallacy that, like with dead biological organisms, nothing can be done to bring a “dead” language back to life. A quote from a young participant of a Myaamia (Miami) culture and language camp is a poignant illustration of the sort of psychological barriers that must be overcome for language revival efforts: “If Myaamia was a dead language, how would we be able to speak it?” (Leonard 2011:135). Note that there have been no fluent first-language speakers of Myaamia since the 1960s; however, based on materials in and descriptions of the language from before that time, a very active language recovery program is underway with a number of second-language speakers. Modern Hebrew is another dramatic and often-cited example of a formerly metaphorically “extinct” language that has been successfully revived, today said to have as many as 9 million speakers. A further problem with terming a language that has lost its last fluent speakers “dead” is that it overlooks the fact that there are often semi-speakers, heritage learners, passive bilinguals, etc. with varying levels of competence in the language who may remain after the last truly fluent speaker has passed away. Insofar as language “lives” in the minds of those who speak, use, and understand it, the language still “survives” in the minds of semi-speakers, even if they do not have native-like proficiency. To ignore this group of language users is to erase (or deny) the last links a community might have to their heritage language, further perpetuating the idea of language “death” as permanent. 50

Language extinction then and now

The implication of permanence embodied in the terms “extinction” and “language death,” the ideological and emotional barriers to language revival that they create, and the erasure of a community’s links with their heritage language have caused many in minority language communities and academic circles to avoid such terms. Rather, academics and language stakeholders now generally prefer a new metaphor – that of “sleeping.” Here, languages that no longer have any fluent speakers are said to be “dormant.” The metaphor of dormancy carries the opposite implication of that of “death” – the implication that such languages are waiting to, at some point in the future, “wake up.” This has the advantage of presenting language revival as an option, rather than creating a barrier of perceived impossibility. The Catalogue of Endangered Languages has therefore adopted the term “dormant” to describe those languages for which we cannot be certain the last native speakers have passed on; languages that have lost their last native speakers recently enough that some number of passive or semi-speakers may still remain; and/or languages with no remaining native speakers, but sufficient documentation to undertake revitalization. In this chapter, we follow suit, describing languages that may be revived as “dormant.” Likewise, the Catalogue adopts the metaphor of “awakening” to describe those languages around the world that have at some point lost their last fluent speaker, but that are currently undergoing concerted efforts at revival.2 We reserve the terms “extinction” and “death” for languages that lost their last speakers in the remote past, have no living rememberers or semi-speakers, and are not associated with heritage groups attempting to awaken the language.

The Catalogue’s approach to dormant and awakening languages The Catalogue of Endangered Languages includes two categories of languages with no fluent3 speakers: “Dormant” and “Awakening” (See Lee & Van Way, Chapter 5, on the Language Endangerment Index). The latter, “Awakening,” designates languages that no longer have fluent speakers but that are undergoing targeted revival efforts with the express purpose of creating new speakers. “Dormant” languages included in the Catalogue are languages that have lost, or are believed to have lost, their last fluent speaker since approximately 1960 (roughly 50 years before the ELCat project began in 2011). Languages with remaining semi-speakers, rememberers, or second-language speakers, but no fluent speakers and no targeted revitalization efforts, are considered “Dormant” as well. For some languages, there is little or no information available regarding when their last fluent speakers passed on. If it is likely that the language was still spoken after 1960, even if the precise date of dormancy is unknown, it is included in the Catalogue. The reason for including languages that were lost since 1960 is partly that speakers may still be found: there have been cases in which speakers were found of languages previously thought “dead”, some years or decades after the passing of what was then thought to be the last speaker. However, languages that are reported to have lost their last speaker before 1960 are highly unlikely to have remaining speakers. In the case of more recently dormant languages, there may also be conflicting reports about their status, as when one source describes a language as having “very few speakers left,” another reports it as “likely extinct,” and so on. It is often difficult to determine whether a language still has remaining speakers, and in many cases the available information may be decades old: a language reported to have just a few elderly speakers in the 1990s is likely to now be dormant, but could conceivably still be spoken. The Catalogue errs on the side of caution and includes languages of uncertain status, but that have been reported as potentially or probably being dormant, in the list of “Dormant” languages. 51

Anna Belew and Sean Simpson

Language death in the remote past: knowns and unknowns What is known about language extinction in the remote past, and how is it known? Unlike inquiries into early pottery or animal domestication, the archaeological record is largely silent on matters of language – spoken language does not leave bones, shards, or other physical traces. Written language may be preserved and transmitted over thousands of years, but the invention of writing is relatively recent in human history (roughly 5,000 years ago); the majority of human languages have never been represented in writing. While extinct languages may leave some traces in the form of loanwords or other contact-induced changes in surviving languages, without written records, we can know very little about the languages that became extinct in the remote past. If no written record of a language was ever produced (or found), it cannot be reconstructed based on data from descendant languages or areal influences, and it died so long ago that no oral history recalls it, it is difficult to know that it ever existed. However, there do exist some records that indicate that some particular language existed, even if they provide little or no information about its structure. The Palaic language, for example, was probably spoken somewhere in the northwest of present-day Turkey, and ceased to be spoken around the 15th century BC after the area was conquered by another Anatolian people. Palaic is referenced in a number of texts by the neighboring Hittites. While the amount of actual Palaic language in the texts is minimal (though there is enough data to identify the language as Indo-European), there are references to people speaking it – there is evidence that the language existed (Carruba 1970). This is more than is known about the vast majority of extinct languages; our awareness of Palaic, and many other long-extinct languages, is only possible through the written records of neighboring (or conquering) peoples. However, there were many now-extinct languages spoken in areas without written traditions. What can be known about patterns of ancient language death in the absence of written records? The comparative method can tell us a good deal about past languages that have surviving descendants, and in some cases shed light on roughly when and where they were spoken. But it is generally assumed that many, many languages and language families have died out, without modern descendant languages and with no attestation, over the course of human history. Just how many is a matter of speculation; some estimates of how many languages have ever existed range from 130,000 to 500,000 (Pagel 2000), but a reliable count is both impossible and beyond the scope of this chapter. Regardless, the situation in which we find ourselves today is one in which 46% of languages are endangered: nearly half of all living languages are at risk of loss within the next few generations, representing a potentially drastic decrease in linguistic diversity. Languages have died throughout history; however, barring unknown catastrophes in the distant past, it is extremely unlikely that there has ever been a time when such a high proportion of linguistic diversity stood to be lost in such a short time.

Accelerating causes of language endangerment Languages cease to be spoken for a wide variety of reasons, and under a wide variety of circumstances. The pressures that cause language loss do not operate uniformly; under similar conditions, one language may be maintained while another falls silent. For more in-depth discussion of the causes and mechanisms of language death, see Rehg and Campbell (in press), Thomason (2015), Dorian (1989), Austin and Sallabank (2011), Brenzinger (2007), Crystal (2000), Grenoble and Whaley (1998), Janse and Tol (2003), Tsunoda (2006), inter alia. However, there are broad and overlapping categories of pressures that often impact linguistic vitality, which we briefly outline below; all of these factors may be considered to be accelerating today. 52

Language extinction then and now

First, economic factors often play a role in language loss (see Harbert et al. 2008; Nettle & Romaine 2000; Robinson 1996, etc.). This can include language shift for the purposes of accessing economic opportunities, rapid economic transformations in a region or speaker community, industrialization and urbanization, destruction of the resources and habitat in which a speaker population traditionally made their living, resettlement or migration for labor purposes, increasing integration into national and global economic systems, and so on. Changes in language use as a result of economic factors are sometimes voluntary on the part of speakers, sometimes forced, and often somewhere in the middle. Today, rapid rates of urbanization around the world, accelerating destruction of ecosystems inhabited by indigenous or minority groups, globalization, etc. pose increasingly severe threats to linguistic diversity. Second, demographic factors often impact language vitality (see Maffi 2001, Williams & Comfort 2007, etc.). These can include declining populations in a speaker group, dispersal of speakers (including refugee populations), disruptions in cultural transmission (as in cases where children are forcibly separated from parents or caretakers), war, genocide, natural disaster, population collapse as a result of disease, intermarriage with other linguistic groups, invasion or settlement by outsiders in a group’s traditional territory, lack or disruption of social cohesion, and so on. These factors have been particularly salient in the Americas and Australia, where European contact and colonization caused widespread loss of life and disruption of languages and cultures; however, demographic factors are also driving language loss in many other parts of the world today, particularly those experiencing rapid population dispersal due to armed conflict and the effects of climate change. Demographic factors that threaten languages can certainly be considered to be accelerating. Third, cultural and political factors can drive language endangerment (see Jones 2015; Romaine 2002; Connell 2015). These include official suppression or outlawing of languages, stigmatization of a language by broader society, lack of institutional support for languages, language policies unfavorable to small or endangered languages, the role of language in the religion(s) of a group, lack of autonomy or self-determination among a language’s speakers, the imposition of colonial languages, the political status of an ethnicity or tribe associated with a language, the role of language in national or regional identity, and so on. While policies aimed at sustaining endangered languages have been adopted by a number of countries and international bodies within the past several decades, there are many other areas of the world where no such policies exist and/or suppression or stigmatization of languages is ongoing. Fourth, aspects of language use, language choice, and language transmission can result in language loss (see Lüpke & Storch 2013; Reiter 2010; Rosendal 2016, etc.). This includes changes in the domains of use of a language, changes in the manner or degree of language transmission in families or communities, formal education (or its absence) in a language, changing patterns of multilingualism (such as growing bilingualism in a formerly monolingual population, or conversely, shrinking linguistic repertoires in a formerly multilingual population), speakers’ changing or declining proficiency in a language, the availability and prevalence of literacy in a language, the role of a language in media, the ability to use a language on computers, the internet, and text messages, etc. Patterns of language use and transmission are changing rapidly in many parts of the world, often in ways interlinked with the other factors outlined above. Finally, attitudes and ideologies can play a major role in language endangerment (see Sallabank 2013; Kroskrity & Field 2009; Austin & Sallabank 2014, etc.). Linguistic pride or shame, the degree of social prestige associated with a language, the relationships between language and ethnic or tribal identity, attitudes towards a language by non-speakers, ideologies about the suitability of languages for certain purposes, the ideas or values symbolically 53

Anna Belew and Sean Simpson

associated with a language (e.g., modernity, tradition, ruralness, urbanity, globalism, localism, “backwardness,” “coolness,” etc.), linguistic purism, the desire to belong (or not) to groups or social categories affiliated with a language, individual and group motivation to learn or maintain a language, and many other attitudinal factors, can help determine whether a language is maintained or lost. While attitudes favorable to language maintenance and revitalization are increasing in some communities and individuals, there are also many communities in many parts of the world where attitudes towards endangered languages, both on the part of speakers and non-speakers, are increasingly indifferent or negative. While every scenario of language loss is shaped by different combinations of factors, and each of the above pressures may have a different (or no) effect on a given language, it can be broadly stated that the factors which cause language endangerment are accelerating on a global scale.

The current situation of language endangerment and dormancy These changes in global linguistic contexts and in the socioeconomic, cultural, and environmental positions in which endangered language speakers find themselves have led to a situation more dire for global linguistic diversity than ever before. Of the approximately 700 languages known to have ever gone extinct,4 roughly 216 of them fell silent in the past 57 years (see Appendices 4.1 and 4.2): 30% of all known cases of language death have occurred since 1960. See Campbell and Okura, Chapter 6, for discussion of the overall pace of language loss seen in the Catalogue’s data. Here, we focus on those languages that have already become dormant. If we examine the 256 dormant and awakening languages in the Catalogue (Appendices 4.1 and 4.2), it is clear that not all areas of the world are experiencing language loss in the same way. Table 4.1 shows the number of known dormant and awakening languages in each region of the world. Australia and the Americas are home to the highest numbers (108 and 100, respectively) of dormant and awakening languages. This is, in part, a result of there being more available information about languages in these regions than other parts of the world (see Belew & Simpson in press). There are likely numerous dormant languages in other regions that have not been reported. However, it is also uncontroversial that these areas have experienced the most severe language loss in recent history. Note that Table 4.1 only includes languages that became dormant after 1960 (per the Catalogue’s criteria) and languages which are actively being revived. Table 4.1 Number of known dormant and awakening languages per region Region

Number of recently dormant and awakening languages

Australia North America South America Africa Europe Mexico, Central America, and the Caribbean Pacific Southeast Asia Near East South Asia East Asia

108 48 44 16 9 8 8 6 5 3 1

54

Language extinction then and now

The accelerating pace of language loss globally can be seen in the dormant and awakening languages in Appendices 4.1 and 4.2 for which we have some information about when their last fluent speaker passed away. Of the 256 dormant and awakening languages listed, 34 are known to have become dormant between 1960 and 1979; another 37 became dormant between 1980 and 1999; and 69 have become dormant since the year 2000, in the past 17 years alone. (The dates of loss of the last speakers of the remaining 116 languages are either unknown but believed to be after 1960, pre-1960 in the case of a few awakening languages, or imprecise.5) While more information may have been reported about newly dormant languages than those that fell silent several decades ago, these numbers nevertheless do illustrate an increase in rate of language loss with each passing decade. While it is impossible to predict exactly how many languages will fall silent as time passes, this rapid increase is of immense concern. Table 4.2 Known dormant languages since 1960 by decades in which last speaker passed away Timespan

Number of known languages which lost their last speaker

1960–1979 1980–1999 2000–2017 Unknown or other

34 37 69 116

Reasons to be hopeful: The rising tide of language revival Despite the alarming pace of language endangerment today, there are reasons to be hopeful. Language revitalization movements around the world are growing, and there already exist success stories that demonstrate the possibility of maintaining currently viable languages, of revitalizing endangered languages, and even of reviving formerly dormant languages. Cornish, Hawaiian, Hebrew,6 Kaurna, Māori, Wampanoag, Welsh, Yurok, and a good number of other languages have, in the past few decades, made remarkable strides in increasing the number and proficiency of their speakers. In some cases (like Cornish, Hebrew, Wampanoag, and Kaurna), languages that had previously been dormant, with no native speakers at all, have awakened with new learners. Each of these languages has unique circumstances, approaches, and results, but each is a success story in its own right. Hebrew is perhaps the best-known success story in terms of sheer numbers of new speakers: for roughly 1,700 years prior to the nineteenth century, Hebrew was no one’s native language, and was used exclusively for liturgical purposes and occasionally as a lingua franca. However, starting in the 1880s, scholars began to revive Hebrew for literary and conversational purposes; by 1948, with the founding of Israel, it had become a national language. Today, the revived form of Modern Hebrew7 is spoken by millions of native speakers in all domains of life. Similarly, the Hawaiian language has experienced an enormous reawakening. The number of native speakers of Hawaiian had dropped from at least 800,000 in the 1770s to only a few hundred in the 1970s (Nettle & Romaine 2000:95). In the 1960s and 1970s, however, the Hawaiian language and culture renaissance began bringing attention to the precarious state of the language and devising solutions for language revitalization. In 1978, Hawaiian gained status as an official language of Hawai‘i, along with English, and in the 1980s the first pūnana leo, or Hawaiian immersion schools, were opened. Today, there are Hawaiian immersion schools on every inhabited Hawaiian island (except Lana‘i and Ni‘ihau) helping children grow up as speakers of Hawaiian, and there are undergraduate and graduate programs for Hawaiian 55

Anna Belew and Sean Simpson

language learning and scholarship at the University of Hawai‘i, with an estimated 3,000–8,000 second-language speakers and learners of Hawaiian (Wurm 2007; NeSmith 2016). However, most languages have no official status in their nations or states; even so, concerted efforts by communities or individuals can give languages new “life.” This is true even of longdormant languages, for example, Wampanoag (Wôpanâak), which ceased to be spoken in the mid-nineteenth century. Thanks to thorough existing documentation and enormous effort on the part of contemporary linguists and community members, the language is now awakening with immersion camps, courses for adult learners, and even a child being raised as a native speaker.8 Similarly, Myaamia (Miami) lost its last speaker in the 1960s; today, there are new speakers and many learners, thanks to available documentation and to linguists, teachers, and learners with a concerted will to bring the language back into use. As seen in Appendix 4.1, at least 69 languages around the world that had previously lost their last fluent speakers are now awakening. The revitalization of an endangered language, or the revival of a dormant one, has at a minimum two requirements for success: first, there must be people who have the desire, the time, and the resources to undertake the work involved. Second, there must be enough available information about the language, be it in the form of living speakers, or, if there are no living speakers, sufficient documentation of the language. We believe that, despite the growing threats to linguistic diversity and accelerated pace of language loss today, there is also a growing will to conserve linguistic diversity, and a growing ability to create documentation for the purposes of language revitalization. In the future, it is likely that an ever-greater number of languages will meet these minimum requirements for revitalization or revival. The past 25 years have seen an enormous increase in interest and concern about language endangerment among scholars. The 1992 special issue of Language on language endangerment is considered the seminal moment for recognition of the endangered-languages crisis in linguistics (though of course many linguists were concerned with and studying language endangerment prior to then). Since 1992, there has been an explosion in publications, conferences, dedicated journals, and courses on language documentation and revitalization (see, for example, the journals Language Documentation & Conservation and Language Documentation & Description, and the conferences International Conference on Language Documentation & Conservation and Institute on Collaborative Language Research, among many other initiatives). Similarly, outside of academic linguistics, community-based programs for language maintenance, revitalization, and revival have grown steadily. Funding for language revitalization has become more available in the US, Canada, Australia, and the EU through private foundations and government funding initiatives. Some private foundations and local organizations also provide funding for language revitalization work in Central and South America, Asia, the Pacific, and Africa. There are a growing number of publications to guide community- and home-based initiatives for language revitalization (e.g., Hinton & Hale 2001, Hobson et al. 2010, Hinton 2013, etc.), workshops to aid communities in language work (e.g., the Breath of Life Archival Institute, the Northwest Indian Language Institute, the biennial Puliima conference, American Indian Language Development Institute, Canadian Indigenous Languages and Literacy Development Institute, Indigenous Language Institute, and others), and each year, new communities undertake the task of revitalizing or reviving their languages. The Catalogue of Endangered Languages currently lists 69 awakening languages (languages being revived after having lost their last speakers) (see Appendix 4.1). For the majority of these awakening languages (60%), revival initiatives began in the past 30 years. Indeed, the number of revival programs has risen astronomically since the turn of the century; 31 revival programs have begun since the year 2000, as compared with 13 that 56

Language extinction then and now

are known to have begun between 1980 and 1999. As revival programs begin creating new speakers and bringing formerly dormant languages back into use, communities in similar situations can find inspiration and practical lessons in existing programs, creating a “snowballing” of language revival. The majority of these revival initiatives are taking place in North America or Australia, reflecting not only the disproportionate rates of language loss in these areas, but also the availability of documentation, and of resources for language revival (see e.g. Amery & Gale 2008). However, there are also some revival initiatives in other regions; languages are beginning to awaken worldwide. We are hopeful that this upward trend will continue, and that the geographic spread of revival initiatives will increase. It is also worth noting with respect to the second necessity for language revival, sufficient documentation is available for more languages now than ever before. The majority of the world’s languages remain under-documented. Using grammars as a rough metric for documentation, approximately 1,708 grammars have been published since 1967 (Thieberger 2017) – if each grammar were of a separate language, this would represent contemporary descriptions of 24% of the world’s languages, at most. However, the body of documentary and descriptive data on small languages is growing rapidly. Of the 1,708 grammars published in the past 50 years, 683 of them, or 40%, have been published since the year 2000, indicating an accelerated pace of language documentation in response to the accelerated pace of language loss. Not only is the number of languages with some documentation growing, but with the advent of consumer audio and video recording equipment, the prevalence of smartphones in many (though certainly not all) endangered-language contexts, high-capacity digital storage media, and high-speed internet, it is now possible to collect, store, and disseminate massive amounts of language data. While long-term durability of digital data is of course a concern (see Bird & Simons 2003 and Barwick & Thieberger 2006), the fact remains that the capacity to create and preserve records of language is better than ever before, and will likely to continue to improve as new recording technology and documentary methodologies develop. There are many reasons to be hopeful about the current and future prospects for revival of dormant languages, on both the cultural and technological fronts.

Conclusions The world’s linguistic diversity faces far greater threats today than in the past, and the rate of language loss has accelerated to an alarming degree. Contrary to some claims, the sharp decline in the world’s linguistic diversity is not a normal continuation of an age-old pattern of language death. About 46% of the world’s languages are in danger of being lost or have recently fallen silent. This is a rapid reduction in linguistic diversity unlikely to have occurred at any previous point in history, and the rate of language loss is highly accelerated in recent decades. As seen in the Catalogue’s data and discussed on p. 55 and p. 80, 69 languages have become dormant in the past 17 years alone; nearly as many languages are known to have lost their last speaker since the year 2000 as during the previous four decades combined. However, new efforts to slow or reverse these losses are cause for hope. Language documentation and revitalization programs are on the rise, as are revival initiatives to re-awaken previously dormant languages: roughly 45% of known revival initiatives have begun since the year 2000 alone. Academics, language stakeholders, and the general public are finally becoming aware of the urgency of conserving languages. And although language loss is occurring at a pace unparalleled in history, so now are efforts towards language conservation and revival. It may not be possible at this point to prevent major reduction in the world’s linguistic diversity; the factors causing language endangerment are complex, growing, and beyond the 57

Anna Belew and Sean Simpson

reach of most people to solve. However, it is possible to maintain or revitalize languages, if a community has the will and dedication to do so, and it is certainly possible to create adequate records for future language revival. We hope that future generations will see the current era not just as one of the alarmingly accelerated loss of languages, but as the point when the concerted efforts of language communities, linguists, and other stakeholders began to turn the tide of language loss.

Notes 1 2

3

4

5

6 7

8

www.endangeredlanguages.com/about/ While terminology used for language renewal differs between authors and regions, in this chapter, we use the term “revitalization” for initiatives to expand the use and speaker population of a language that has not yet lost its last fluent speakers. “Revival” and “awakening” are used to refer to efforts to create new speakers and expand the use of a language that has become dormant. While it can be difficult to determine who is a “native” or “fluent” speaker in cases of language endangerment, the Catalogue follows primary sources in their determination of how many “fluent” speakers there are. So, if a source reports 10 native speakers and 20 semi-speakers, the Catalogue reports the speaker numbers in this way. It is generally assumed that thousands of languages have become extinct over the span of human history; however, the figure of c.698 represents named extinct languages for which we have some linguistic information (see Campbell and Okura, Chapter 6). Languages with very broad ranges of dates during which they may have become dormant, e.g., 1975–2007, are not included in these counts, as it is unclear in which decade they lost their last fluent speaker. Modern Hebrew is not included in the Catalogue of Endangered Languages, as it can be considered to be fully revived and not at all endangered, as a national language with millions of native speakers. It should be noted that revived varieties of languages may differ significantly from their previous forms; however, this should not necessarily be seen as a failure of revival efforts. See NeSmith (2002), Zuckermann and Walsh (2011), Sayers and Renko-Michelsen (2015), and Hinton and Ahlers (1999) for discussion of hybridity and change in revived languages. See www.wlrp.org, the website of the Wôpanâak Langauge Reclamation Project, for news and updates on the revival of Wôpanâak.

References Amery, Rob & Gale, Mary-Anne. 2008. But our language was just asleep: A history of language revival in Australia. In McGregor, William B. (ed.), Encountering Aboriginal languages: Studies in the history of Australian linguistics, 339–382. Canberra: Pacific Linguistics. Austin, Peter K. & Sallabank, Julia (eds.). 2011. The Cambridge handbook of endangered languages. Cambridge: Cambridge University Press. Austin, Peter K. & Sallabank, Julia (eds.). 2014. Endangered languages: Beliefs and ideologies in language documentation and revitalization. Oxford: Oxford University Press. Barwick, Linda & Thieberger, Nick (eds.). 2006. Sustainable data from digital fieldwork: From creation to archive and back. Sydney: Sydney University Press. Belew, Anna & Simpson, Sean. In press. The status of the world’s endangered languages. In Rehg, Kenneth L. & Campbell, Lyle (eds.). Handbook of endangered languages. Oxford: Oxford University Press. Bird, Steven & Simons, Gary. 2003. Seven dimensions of portability for language documentation and description. Language 79. 557–582. Brenzinger, Matthias (ed.). 2007. Language diversity endangered. Berlin: De Gruyter. Carruba, Onofrio. 1970. Das Palaische. Texte, Grammatik, Lexikon. Wiesbaden: Harrassowitz. Connell, Bruce. 2015. The role of colonial languages in language endangerment in Africa. In McLaughlin, Fiona & Henderson, Brent & Essegbey, James (eds.). Language documentation and endangerment in Africa. Amsterdam: John Benjamins. Crystal, David. 2000. Language death. Cambridge: Cambridge University Press.

58

Language extinction then and now Dorian, Nancy (ed.). 1989. Investigating obsolescence: Studies in language contraction and death. Cambridge: Cambridge University Press. Grenoble, Lenore A. & Whaley, Lindsay J. (eds.). 1998. Endangered languages: Language loss and community response. Cambridge: Cambridge University Press. Harbert, Wayne & McConnell-Ginet, Sally & Miller, Amanda & Whitman, John (eds.). 2008. Language and poverty. Clevedon: Multilingual Matters. Hinton, Leanne. 2013. Bringing our languages home: Language revitalization for families. Berkeley: Heyday. Hinton, Leanne & Ahlers, Jocelyn. 1999. The issue of “authenticity” in California language restoration. Anthropology and Education Quarterly 30(1). 56–67. Hinton, Leanne & Hale, Kenneth. 2001. The green book of language revitalization in practice. Leiden: Brill. Hobson, John & Lowe, Kevin & Poetsch, Susan & Walsh, Michael (eds.). 2010. Re-awakening languages: Theory and practice in the revitalisation of Australia’s Indigenous languages. Sydney: Sydney University Press. Janse, Mark & Tol, Sijmen (eds.). 2003. Language death and language maintenance: Theoretical, practical and descriptive approaches. Amsterdam: John Benjamins. Jones, Mari C. (ed.). 2015. Policy and planning for endangered languages. Cambridge: Cambridge University Press. Kroskrity, Paul & Field, Margaret C. 2009. Native American language ideologies: Beliefs, practices, and struggles in Indian country. Tucson: University of Arizona Press. Leonard, Wesley. 2011. Challenging “extinction” through modern Miami language practices. American Indian Culture and Research Journal 35(2). 135–160. Lüpke, Friederike & Storch, Anne. 2013. Repertoires and choices in African languages. Berlin: de Gruyter. Maffi, Luisa (ed.). 2001. On biocultural diversity: Linking language, knowledge, and the environment. Washington, DC: Smithsonian Institution Press. NeSmith, Keao. 2002. Tūtū’s Hawaiian and the emergence of a Neo-Hawaiian language. Honolulu: University of Hawai‘i at Mānoa. (Master’s thesis.) NeSmith, Keao. 2016. Proposal to establish policies and guidelines for the County of Kaua‘i regarding the use of Hawaiian language. MS. Nettle, Daniel & Romaine, Suzanne. 2000. Vanishing voices: The extinction of the world’s languages. Oxford: Oxford University Press. Pagel, Mark. 2000. The history, rate, and pattern of world linguistic evolution. In Knight, Chris & StuddertKennedy, Michael & Hurford, James (eds.), The evolutionary emergence of language, 391–416. Cambridge: Cambridge University Press. Rehg, Kenneth L. & Campbell, Lyle (eds.). In press. Handbook of endangered languages. Oxford: Oxford University Press. Reiter, Sabine. 2010. Linguistic vitality in the Awetí indigenous community: A case study from the Upper Xingu multilingual area. In Flores Farfán, José Antonio & Ramallo, Fernando (eds.), New perspectives on endangered languages. Amsterdam: John Benjamins. Robinson, Clinton. 1996. Language use in rural development. Berlin: De Gruyter. Romaine, Suzanne. 2002. The impact of language policy on endangered languages. International Journal of Multicultural Studies 4(2). 194–212. Rosendal, Tove. 2016. Language transmission and use in a bilingual setting in rural Tanzania: Findings from an in-depth study of Ngoni. In Filipović, Luna & Pütz, Martin (eds.), Endangered languages and languages in danger: Issues of documentation, policy, and language rights, 361–386. Amsterdam: John Benjamins. Sallabank, Julia. 2013. Attitudes to endangered languages: Identities and policies. Cambridge: Cambridge University Press. Sayers, Dave & Renko-Michelsen, Zsuzsanna. 2015. Phoenix from the ashes: Reconstructed Cornish in relation to Einar Haugen’s four-step model of language standardisation. Sociolinguistica: Internationales Jahrbuch für europäische Soziolinguistik 29(1). 17–38. Thieberger, Nick. 2017. LD&C possibilities for the next decade. Language Documentation & Conservation 11. 1–4. Thomason, Sarah. 2015. Endangered languages: An introduction. Cambridge: Cambridge University Press.

59

Anna Belew and Sean Simpson Tsunoda, Tasaku. 2006. Language endangerment and language revitalization: An introduction. Berlin: De Gruyter. Williams, Robert S. & Comfort, Jade. 2007. Language documentation and description among refugee populations. In Austin, Peter K. & Bond, Oliver & Nathan, David (eds.), Proceedings of Conference on Language Documentation and Linguistic Theory, 261–268. London: SOAS. Wurm, Stephen. 2007. Australasia and the Pacific. In Moseley, Christopher (ed.). Encyclopedia of the world’s endangered languages. Abingdon: Routledge. Zuckermann, Ghil’ad & Walsh, Michael. 2011. Stop, revive, survive: Lessons from the Hebrew revival applicable to the reclamation, maintenance and empowerment of Aboriginal languages and cultures. Australian Journal of Linguistics 31(1). 111–127.

Appendix 4.11 Awakening languages listed in the Catalogue of Endangered Languages, with approximate date of the death of the last native speaker (if known) and approximate date revival efforts began2 (if any) Language name

ISO 639-3 Country spoken

Year language became dormant

Year revival efforts began

Awabakal Barbareño Barngarla Batyala Biri Boonwurrung Boruca Buandig Catawba Chiquimulilla Xinka Chitimacha Chiwere Cornish Cupeño Darkinyung Dhanggati Dharug Dharumbal Dhurga Eora Eyak Githabul Goreng Gros Ventre Gudjal Gudungurra Janday Kaniyang Kanza Kaurna Klallam Klamath-Modoc

awk boi bjb xby bzr, xgi (none) brn xbg chc (none) ctm iow cor cup xda dyn xdk xgm dhu (none) eya gih xgg ats wrg xrd jan (none) ksk zku clm kla

Before 1890 1965 c.1960 c.1987 Unknown Unknown 2004 Unknown 1959 1970s 1940 1996 c.1777 1987 Early 1900s Unknown Early 1900s? Unknown Unknown Unknown 2008 Unknown Unknown 1981 Unknown Unknown Unknown Unknown Early 1980s 1929 2014 2003

2001 2010 2012 Unknown Unknown Unknown 2004 2011 Unknown c.2000 Early 1990s Unknown c.1904 Unknown 2003 c.1993 1990s Unknown c.1976 Unknown 2011 Unknown Unknown c.2006 c.2005 Unknown Unknown c.1990s 2001 1989 2014 2003

Australia United States Australia Australia Australia Australia Costa Rica Australia United States Guatemala United States United States United Kingdom United States Australia Australia Australia Australia Australia Australia United States Australia Australia United States Australia Australia Australia Australia United States Australia United States United States

60

Language extinction then and now

Language name

ISO 639-3 Country spoken

Year language became dormant

Year revival efforts began

Kurnai Lipan Luiseño Makah Manx Mathi-Mathi Myaamia Natchez Ngarrindjeri Ngunawal Osage Pinjarup Quapaw San Francisco Bay Costanoan Serrano Siraya Sîshëë Southeastern Pomo Soyot Taungurung Tolowa Tunica Umutina Wampanoag Wangkumara Wappo Wardandi Wathawurrung Wemba-Wemba Wirangu Wiyot Woiwurrung Wotjobaluk Yabula-Yabula Yagara Yorta Yorta Yurok

unn apl lui myh glv dmd mia ncz nay xul osa pnj qua cst ser fos sih pom (none) dgw tol tun umo wam xwk wao wxw wth xww wgu wiy wyi xwt yxy yxg xyy yur

Unknown c.1980s Unknown 2002 1974 Unknown c.1962 1965 1960s? Unknown 2005 Unknown Unknown 1920s 2002 1905 Early 2000s? Unknown Unknown Unknown Before 1990 1948 2004 Late 1800s c.2005 1990 Unknown Unknown Unknown Unknown 1962 Unknown Unknown Unknown Unknown c.2000 2013

c.1995 c.2008 c.2002 Unknown c.1992 Unknown c.1997 Early 2000s 1985 Unknown 2005 Unknown c.2012 c.2002 c.2005 c.1999 Unknown 2002 Early 2000s 2004 1980s 2010 2004 1993 c.2005 2012 Unknown 1998 c.1992 2004 Unknown Unknown 2005 Unknown Unknown 2010 2013

Australia United States United States United States United Kingdom Australia United States United States Australia Australia United States Australia United States United States United States Taiwan New Caledonia United States Russia Australia United States United States Brazil United States Australia United States Australia Australia Australia Australia United States Australia Australia Australia Australia Australia United States

1

Our sincere thanks to the following people for providing assistance in finding some of the dates for language dormancy and awakening: Sharon Atkinson, Lyle Campbell, Kris Eira, John Hobson, Kevin Lowe, Doug Marmion, Daryn McKenny, and Nick Thieberger. All errors are our own. Any language revival programs not listed here are not deliberate omissions; if you know of a program to revive a formerly dormant language which is not included in the Catalogue of Endangered Languages, please contact feedback@endangered languages.com.

2

In cases where language revitalization programs predate the passing of the last native speaker, the date given for beginning of revival efforts is the same as the date of dormancy, as revival of a formerly dormant language is by definition not possible before the language has become dormant.

61

Anna Belew and Sean Simpson

Appendix 4.2 List of dormant languages in the Catalogue of Endangered Languages not known to be undergoing active revival efforts, with best available approximate year that the last native speaker died (if known)1 Language name

ISO 639-3

Country spoken

Year language became dormant

Aasáx Adithinngithigh Agavotaguerra Aghu-Tharnggala Aka-Bo Aka-Kora Akuriyo Amanayé Anambé of Cairarí Andoa Aore Apiaká Arabana-Wangkangurru Arapaso Arára of Mato Grosso Atsugewi Auré Awaké Awngthim Awu Laya Ayerreyenge Badimaya Baygo Berbice Creole Dutch Bibbulman Bigambal Bung Cahuarano Canichana Cayuvava Central Pomo Chiapanec Coast Miwok Copper Island Aleut Demushbo Dhungaloo Djangun Duungidjawu Eastern Pomo Eel River Athabaskan El Molo Gagudju Garlali Gayiri

aas dth avo gtu akm ack ako ama aan anb aor api ard arj axg atw aux atx gwm typ axe bia byg brc xbp xbe bqd cah caz cyb poo cip csi mud (none) dhx djf (none) peb (none) elo gbu gll (none)

Tanzania Australia Brazil Australia India India Suriname Brazil Brazil Peru Vanuatu Brazil Australia Brazil Brazil USA Brazil Brazil and Venezuela Australia Australia Australia Australia Sudan Guyana Australia Australia Cameroon Peru Bolivia Bolivia USA Mexico USA Russia Bolivia Australia Australia Australia USA USA Kenya Australia Australia Australia

c.1976 Unknown After 1973 Unknown 2010 2009 2002 c.2001 After 2006 2012 1982–2001 2012 1988–2003 c.1990 Before 2001 1988 Before 2009 c.1970 Unknown 2016 Unknown 1966–1982 1978–1992 2005 Unknown 1996 Before 1997 Early 1990s 1991–2000 After 1962 1997–2016 1990–1997 1970s 1995–2008 2010–2014 After 2012 Unknown c.1970 Before 2008 Before the 1970s 1999 2002 Early 2000s Unknown

62

Language extinction then and now

Language name

ISO 639-3

Country spoken

Year language became dormant

Guana Guazacapán Xinka Gubbi Gubbi Gugu-Warra Gününa-Küne Guramalum Guwa Hanis Holikachuk Honduran Lenca Hpon Ikaranggal Jabirr-Jabirr Jiwarli Jumaytepeque Xinka Jurruru Kaixana Kalkatungu Kamas Katawixi Kiowa Apache Kok-Nar Koko Babangk Kolakngat Kubi Kukatj Kungarakany Kungkari Kuruaya Kuuk-Yak Kuyani Kw’adza Laghu Lake Miwok Lapachu Leco Linngithigh Livonian Loun Lower Southern Aranda Lushi Maidu Makolkol Malkana Malyangapa Mandahuaca Mandan Mapia Mayi-Kulan

gqn qda gbw wrw pue grz xgw csz hoi len hpo ikr dyb dze qhq tju (none) ktg, wwb xas xat apk gko okg (none) kof ggd ggk lku kyr uky gvy wka lgb lmw (none) lec lnj liv lox axl (none) nmu zmh vml yga mht mhq mpy xyk

Brazil Guatemala Australia Australia Argentina Papua New Guinea Australia USA USA Honduras Myanmar Australia Australia Australia Guatemala Australia Brazil Australia Russia Brazil USA Australia Australia Australia Nigeria Australia Australia Australia Brazil Australia Australia Tanzania Solomon Islands USA Bolivia Bolivia Australia Latvia Indonesia Australia Nigeria USA Papua New Guinea Australia Australia Brazil, Venezuela USA Indonesia Australia

c.2014 c.2010 Unknown Before 2003 1960–1970 1987–2007 1938–1990 1972 2012 Before 1974 1959–2007 Unknown 1970s 1986 Before 2010 1967–1986 After 1950 Unknown 1989 1986–2011 2008 Early 1970s Unknown Unknown Before 1995 Before 2003 1989 Unknown Before 2007 After 1975 Unknown 1976–1999 1984 1994–2008 1975–2005 2001–2009 Late 1970s 2011–2014 2007–2015 c.2011 2006–2011 1994–2007 1988–2003 1970s 1976 1990–2012 2016 2000–2007 1975–2007

63

Anna Belew and Sean Simpson

Language name

ISO 639-3

Country spoken

Year language became dormant

Mayi-Thakurti Mayi-Yapi Mbabaram Mbiywom Mesmes Minjungbal Mirití Mithaka Mlaḥsô Mogholi Mondé Mpra Munichi Nari Nari Ndyuka-Trio Pidgin Ngalakgan Nganyaywana Ngarigu Ngumbarl Nhirrpi Nimbari Nisenan Nonuya Northern Pomo Nukini Nukunu Oblo Ona Opata-Eudeve Pawnee Phalok Pinikura Pirriya Pitta-Pitta Plains Miwok Poyanawa Punthamara Purduna Quileute Sabüm Salvadoran Lenca Sapé Saraveca Sawknah Sirenikski Siuslaw Ski:xs (Southern Tsimshian) Southern Khanty Sowa

xyt xyj vmb (none) mys xjb mmv (none) lhs mhj mnd (none) myr rnr njt nig nyx xni (none) hrp nmr nsz noj pej nuc nnv obl ona opt paw (none) pnv xpa pit, lnw, wky pmw pyn xpt bxn qui sbo (none) spc sar swn ysr sis (none) (none) sww

Australia Australia Australia Australia Ethiopia Australia Brazil Australia Turkey, Syria Afghanistan Brazil Ghana Peru Australia Suriname Australia Australia Australia Australia Australia Cameroon USA Peru, Colombia USA Brazil Australia Cameroon Argentina Mexico USA Thailand Australia Australia Australia USA Brazil Australia Australia USA Malaysia El Salvador Venezuela Bolivia Libya Russia USA USA, Canada Russia Vanuatu

1975–2007 Before 2011 1979 Unknown After 2001 Unknown Before 1998 Unknown c.1998 1980s 2006–2014 1963–2007 1988–2008 Unknown 1960s 2004 Unknown 2004–2011 Unknown Before 2011 1992–2009 Unknown 1990s–2015 2005 1960s–1970s c.2000 Before 2002 1970s or 1980s 1993–2010 c.2008 Before 2007 1981–2004 1969 Late 1970s 1960s 2009–2013 Unknown 1988–1994 1999 Unknown Before 1997 c.2004 1960s Before 2012? 1997 1970s 2008–2014 Before 1993 2000

64

Language extinction then and now

Language name

ISO 639-3

Country spoken

Year language became dormant

Tequiraca Teteté Thiin Tholp Tirahi To Torá Totoró Tule Umbugarla Umiida Unami Unggarranggu Unggumi Uru Uruava Veddah Vilela Wadi-Wadi Wajuk Warndarrang Warriyangga Warrnambool Wasco-Wishram Western Mansi White Lachi Wichita Wintu Worrorra Wotapuri-Katarqalai Xipaya Yahuna Yalarnnga Yaminawa Yanda Yaraldi Yardliyawarra Yarluyandi Yavitero Yawijibaya Yevanic Yingkarta Yitha-Yitha Zari Zeem

ash teb iin (none) tra toz trz ttk (none) umr xud unm xun xgu ure urv ved vil xwd xwj wnd wri gjm wac (none) lwh wic wnw wro wsv xiy ynu ylr yaa yda (none) yxl yry yvt jbw yej yia xth zaz zua

Peru Ecuador Australia Vanuatu Afghanistan Cameroon Brazil Colombia Nigeria Australia Australia USA Australia Australia Bolivia Papua New Guinea Sri Lanka Argentina Australia Australia Australia Australia Australia USA Russia Vietnam USA USA Australia Afghanistan Brazil Colombia Australia Brazil, Peru, Bolivia Australia Australia Australia Australia Venezuela Australia Israel Australia Australia Nigeria Nigeria

1930s–2008 c.1975 Before 2004 2000s Unknown Unknown 1990–2006 1998–2007 Before 2005 1987–2001 2009–2011 2002 1984–2011 1988–2009 2004–2012 c.1973 Before 1997 1981–2011 Unknown Unknown 1974 1973–1988 Unknown 2008–2014 Before 1993 1990–2008 2016 2003 1990–2011 Before 2003 2003–2010 Before 1988 1980 2006–2013 Before the 1970s 1960s 1987–2004 Before 2000 2000–2005 2009–2011 1987–1996 1981–1987 2005–2015 1995–2000 Early 2000s

1

Dates given in ranges indicate the latest date at which fluent speakers were reported, and the earliest date at which the language was reported to have no more fluent speakers; the date the language became dormant is presumed to be between these dates.

65

5 THE LANGUAGE ENDANGERMENT INDEX Nala H. Lee and John R. Van Way

The Language Endangerment Index (LEI) is a tool developed by ELCat researchers for assessing and quantifying how endangered a language is. This tool allows linguists, community members, and the general public a way to easily compare the vitality of languages across the globe. It is derived from four factors (intergenerational transmission, number of speakers, speaker number trends, and domains of use) and includes a certainty scale dependent on how many factors are used. This chapter explains how the LEI was developed and how to use it, with reference to specific languages as examples.

What is the Language Endangerment Index (LEI)? Decidedly, “endangered” languages are the primary concern of the Catalogue of Endangered Languages (ELCat). Yet, what constitutes an “endangered” language? An even more complicated question is, what makes one language more endangered than another? Bearing in mind that ELCat aims to provide reliable and up-to-date information on the endangered languages of the world, including the extent of endangerment for all its listed languages, a key question naturally emerged: how should the level of endangerment of any language be assessed? While some researchers in the field prefer to rely on their instincts and experience when deciding which language is more endangered than another, and while others may find the issue of enumerating endangered languages contentious (Grenoble 2016), it is necessary to provide an accurate quantitative measure of the world’s endangered languages for several reasons (Lee & Van Way 2016). Quantitative measures of language endangerment have the following advantages: (i) they allow a lay audience to understand the scope of the problem of language endangerment from a bird’s-eye view; (ii) they raise awareness of language endangerment in communities where it is an issue; (iii) they give researchers and language activists a tool to assess the urgency of attention needed for certain languages; (iv) they enable language endangerment assessments to be compared across languages in different contexts and, in that way, allow funding agencies to compare and assess both the urgency and likelihood of success for proposals to document and/or conserve endangered languages; and (v) they allow better understanding of whether or not linguistic diversity can be correlated with biodiversity. Given the many reasons for providing an accurate measure of language endangerment, the Language Endangerment Index 66

The Language Endangerment Index

(LEI) was developed as a tool for ELCat and as a resource for linguists, community members, and others interested in language-endangerment issues. Why is LEI necessary when other quantitative measures of language endangerment (or vitality) exist? The other quantitative measures of language endangerment include the Graded Intergenerational Disruption Scale or GIDS (Fishman 1991), UNESCO’s nine factors for assessing language vitality (UNESCO 2003), Krauss’ schema for assessing language viability (Krauss 2007), and the Expanded Graded Intergenerational Disruption Scale or EGIDS (Lewis & Simons 2010). These various measures of language endangerment have their advantages and disadvantages, and these have been discussed extensively in, for example, Dwyer (2011), Gao (2015), and Lee and Van Way (2016). Essentially, none of these tools are suitable for use on ELCat for one important reason: they all require complete and total information in order to generate a language endangerment assessment. That is, all of these tools require particular types of information about each language and, when there are gaps in knowledge about one area (e.g., domains of use), the tool cannot be confidently utilized. For example, EGIDS assumes that researchers have information regarding intergenerational transmission and its disruption across domains of language use. In reality, most languages and their social conditions are simply not well-documented. Lehmann (1999) suggests that there is no reliable estimate of the number of languages that have received linguistic description, and that it is highly probable that nothing is known about half of the world’s languages aside from their names. For this reason, LEI is designed to mitigate this problem of imperfect knowledge. LEI is a tool that can be used to assess the endangerment of any language of the world, regardless of how much or how little information is actually available on it. Based on the four factors of intergenerational transmission, absolute number of speakers, speaker number trends and domains of use, LEI can be used to generate an overall level of endangerment. The different levels of endangerment are “Critically Endangered,” “Severely Endangered,” “Endangered,”1 “Threatened,” “Vulnerable,” “At Risk,”2 and “Safe.” Consequently, based on the number of factors (out of those four) that users have information about, a level of certainty can also be generated. The following is an example of a LEI assessment: Miqie Yi [ISO 639-3: yiq], a Ngwi (Loloish) language spoken in Yunnan, China, is “Threatened” at 100% certainty (Gao 2015).3 The 100% level of certainty generated here is based on the fact that all four factors (intergenerational transmission, absolute number of speakers, speaker number trends and domains of use) were considered for the assessment of Miqie Yi. Compare this with the assessment of Bih [ibh], a Malayo-Chamic language spoken in Vietnam (Ngyuen 2013), which is “Critically Endangered” at 60% certainty.4 The lower level of certainty with which we can understand the level of endangerment of Bih is derived from fewer factors being used for the assessment since less information is available. The exact mechanisms of LEI are elaborated in “How do we apply the LEI?”. For now, it suffices to say that LEI is different from its predecessors in that it can be used to generate an overall score of language endangerment regardless of how much or how little information researchers may have about the language in question, and that it provides a level of certainty regarding the level of endangerment it generates.

How do we apply the LEI? LEI comprises four individual scales, each concerning a factor associated with language loss: intergenerational transmission, absolute number of speakers, speaker number trends and domains of use. On each factor, a language can be ranked from zero to five, where each number is associated with a particular description. On each scale, researchers evaluate multiple sources and for each source they determine which description best fits the language, and assign the 67

Nala H. Lee and John R. Van Way

language the number that corresponds with the description. In general, the larger the number, the more endangered the language is with regards to that particular factor. All descriptions are written in a clear and straightforward manner, so that these scales are accessible to anyone interested in attaining an assessment of endangerment, be they community members, linguists, or the general public. Results on the individual scales are aggregated to attain an overall level of endangerment. Importantly, it is not obligatory for all factors to be used, considering that perfect information is not always available. The number of factors that are actually used then forms the basis for the level of certainty we have regarding the endangerment assessment attained. The formulae for deriving overall levels of endangerment and certainty are presented in “Deriving overall level of endangerment and level of certainty.”

Four factors of LEI The following subsections provide information regarding the individual scales of intergenerational transmission, absolute number of speakers, speaker number trends, and domains of use.

Intergenerational transmission Among the four factors, intergenerational transmission is weighted most heavily when determining the overall assessment. While the three other factors (absolute number of speakers, speaker number trends, and domains of use) are equally weighted, intergenerational transmission carries twice the weight of each of the other three factors on LEI. Simply stated, intergenerational transmission is most essential to ensuring the continued viability of any language. In an almost tautological manner, if the language is not being passed down to the younger generations, it cannot survive into the future. Doubling the weight of intergenerational transmission score thus reflects the relative importance of language acquisition for the future survival of any language. LEI’s scale of intergenerational transmission is presented in Table 5.1. The two polar ends of the scale represent two extreme scenarios. A language that is considered to be “Critically Endangered” on the scale of intergenerational transmission has only a few elderly speakers, and no one else speaks the language. An example of such a language would be Dusner [dsn], a Malayo-Polynesian language spoken in Papua, Indonesia. The

Table 5.1 Scale of intergenerational transmission 5

4

3

2

1

0

Critically Endangered

Severely Endangered

Endangered

Threatened

Vulnerable

Safe

There are only a few elderly speakers.

Many of the grandparent generation speak the language, but the younger people generally do not.

Some adults in the community are speakers, but the language is not spoken by children.

Most adults in the community are speakers, but children generally are not.

Most adults and some children are speakers.

All members of the community, including children, speak the language.

68

The Language Endangerment Index

language has only three speakers left, all elderly (Dalrymple & Mofu 2012), and it is clearly not being learned by the younger generations (Wurm 2007). Compared with Dusner, which is “Critically Endangered” on the scale of intergenerational transmission, a “Safe” language would be one that is spoken by all members of the community, including children.

Absolute number of speakers The next three factors comprising the LEI do not appear in any order of significance. The first of these is the scale of absolute number of speakers. It is necessary to consider the number of speakers of a language for an endangerment assessment, even if numbers and cut-offs between numbers are inherently arbitrary. In many instances, an estimate of speaker numbers may be the only information available for an assessment of language endangerment. An endangerment index that does not take into account speaker numbers would be quite unsuitable for evaluating languages for which nothing is known except for speaker numbers. For this factor, we are interested in the number of native speakers, given the various notions of speakers that exist. Note that the ranges on LEI’s speaker number scales are designed to be comparable across levels within this factor, as well as with LEI’s other factors. The different ranges are also reasonable in their approximate influence on language vitality or endangerment. The scale of absolute number of speakers is presented in Table 5.2. Again, the two polar ends of the scale represent two extreme scenarios. A language that is “Critically Endangered” on the scale of absolute number of speakers would have between one and nine speakers. Dusner, as previously mentioned, has three speakers at the time of writing, and would be assessed as “Critically Endangered” on the scale of absolute number of speakers. Similarly, Akuntsú [aqz], a Tupian language spoken in Brazil, has only four speakers (Aragon 2014, Aragon personal communication) at the time of writing, and would constitute a “Critically Endangered” language on this scale. In contrast, a “Safe” language on the scale of absolute number of speakers would have at least 100,000 speakers. Table 5.2 Scale of absolute number of speakers 5

4

3

2

1

0

Critically Endangered

Severely Endangered

Endangered

Threatened

Vulnerable

Safe

1–9 speakers

10–99 speakers

100–999 speakers

1,000–9,999 speakers

10,000–99,999 speakers

≥ 100,000 speakers

Speaker number trends With regard to speaker numbers, it is necessary and possible to go beyond absolute number of speakers when relevant information is available. LEI’s scale of speaker number trends is designed to capture a more dynamic view of language shift by providing information on the direction and rate of shift (whether numbers are increasing or decreasing and at what pace). The scale of speaker number trends is presented in Table 5.3. A “Critically Endangered” language on this scale is one that is spoken by a small percentage of the community and whose speaker numbers are decreasing very rapidly. Two examples of such critically endangered languages are Thao and Bih. Thao [ssf], a Formosan language spoken in Taiwan, has only 15 speakers “out of a considerably larger population which claims ancestry” (Blust 2003:1). Bih, 69

Nala H. Lee and John R. Van Way Table 5.3 Scale of speaker number trends 5

4

3

2

1

0

Critically Endangered

Severely Endangered

Endangered

Threatened

Vulnerable

Safe

A small percentage of the community speaks the language, and speaker numbers are decreasing very rapidly.

Less than half of the community speaks the language, and speaker numbers are decreasing at an accelerated pace.

Only about half of community members speak the language. Speaker numbers are decreasing steadily, but not at an accelerated pace.

A majority of community members speak the language. Speaker numbers are gradually decreasing.

Most members of the community speak the language. Speaker numbers may be decreasing, but very slowly.

Almost all community members speak the language, and speaker numbers are stable or increasing.

a Malayo-Chamic language spoken in Vietnam, is spoken by only a small percentage of the community and numbers are decreasing rapidly (Ngyuen 2013). In contrast, a “Safe” language on the scale of speaker number trends would be spoken by almost all community members, and have speaker numbers that are stable or increasing.

Domains of use The last scale on the LEI is that of domains of use. This scale is based on Fishman’s (1965; 1991) notion of domains of use, which refers to “interactions that are rather unambiguously related (topically and situationally) to one or another of the major institutions of society” (Fishman 1991:44). Examples of these various institutions would include government, media, education, religion, and family, among others. Underlying the scale of domains of use is an Table 5.4 Scale of domains of use 5

4

3

2

1

0

Critically Endangered

Severely Endangered

Endangered

Threatened

Vulnerable

Safe

Used only in a few very specific domains, such as in ceremonies, songs, prayer, proverbs, or certain limited domestic activities.

Used mainly just in the home and/or with family, and may not be the primary language even in these domains for many community members.

Used mainly just in the home and/or with family, but remains the primary language of these domains for many community members.

Used in some nonofficial domains along with other languages, and remains the primary language used in the home for many community members.

Used in most domains except for official ones such as government, mass media, education, etc.

Used in most domains, including official ones such as government, mass media, education, etc.

70

The Language Endangerment Index

implicit cline of language shift. If the language is used in official domains, such as education, it is inferred that the same language would also be popularly used in informal domains, such as the family. At the “Critically Endangered” end of the scale, the language is used only in a few very specific domains, such as in ceremonies, songs, prayers, proverbs, or certain limited domestic activities. These domains basically represent the last bastions of a language. Examples of languages that are “Critically Endangered” on the scale of domains of use include languages such as Ge’ez, or Ancient Ethiopic, which is used only for liturgical purposes.5 At the opposite end of the scale, the “Safe” languages are used in most, if not all domains, including both official and non-official ones.

Deriving overall level of endangerment and level of certainty After assessing a language on these four scales (and noting whether any factor is lacking information), the individual scores are aggregated to attain an overall score of language endangerment. The scores assigned for each factor are summed up, with the score for intergenerational transmission multiplied by two, since intergenerational transmission has been identified to be the most essential factor for the continued survival of any language. The total score for all factors used is converted to a percentage of the highest attainable score based on the actual number of factors used. If all four factors are used, the highest attainable score is 25, since each factor has a maximum score of 5, except for intergenerational transmission, which is doubly weighted and has a maximum possible score of 10. Applying the same logic, when two factors are used (excluding intergenerational transmission), the highest attainable score is 10. However, when the two factors used include intergenerational transmission, the highest attainable score is 15. Hence, to attain the overall level of endangerment for any language, the total sum of scores for all factors used is divided by the highest attainable score based on the number of factors used, and converted to a percentage (by multiplying the output by 100). The exact method for deriving the overall level of endangerment is represented in the following formula: Level of endangerment = {[(intergenerational transmission score × 2) + absolute number of speaker score + speaker number trends score + domains of use score]/total possible score based on number of factors used} × 100 Table 5.5 Language Endangerment Index (LEI) and levels of certainty Language Endangerment Index

Level of certainty

100–81% = Critically Endangered

25 points possible = 100% certain, based on the evidence available 20 points possible = 80% certain, based on the evidence available. 15 points possible = 60% certain, based on the evidence available 10 points possible = 40% certain, based on the evidence available 5 points possible = 20% certain, based on the evidence available

80–61% = Severely Endangered 60–41% = Endangered 40–21% =Threatened 20–1% =Vulnerable 0% with less than 100% certainty = At Risk 0% with 100% certainty = Safe

71

Nala H. Lee and John R. Van Way

The percentage that is generated by the formula is to be interpreted using the index on the left in Table 5.5. The LEI pairs ranges of scores with discrete overall levels of endangerment. For example, if a language scores between 41% to 61% using the given formula, we would consider it to be “Endangered” on the LEI. The scale on the right of Table 5.5 represents the different levels of certainty, that is, it tells us how certain we can be about the overall level of endangerment generated. The basis for level of certainty is the number of factors known and used to derive the overall level of endangerment for the language. While overall level of endangerment and level of certainty are two separate pieces of information, they should be considered together for a more accurate overview of any language’s degree of endangerment.

Examples of LEI assessments The use of the LEI is demonstrated by the assessments of three languages in this section, namely Sentinelese, Dupaningan Agta, and Miqie Yi. These languages are chosen because they represent different language groups and have different amounts of vitality information available for each of them at the time of writing. Sentinelese [std] is an unclassified language that is spoken on North Sentinel Island, which is located among the Andaman Islands. There is very little known about the language – the Sentinelese community prefer to have minimal or no contact with the outside world, and have been hostile to outsiders who have attempted to land on their island (Singh 1978). Because of the isolation of the Sentinelese speakers, it is not possible to rate the language on all four LEI factors. If information from a Census of India is taken for the basis of an assessment, we would be able to rate Sentinelese on at least one factor, that being absolute number of speakers. In 2001, the Census of India recorded 39 individuals on the island in a survey that supposedly took place from a distance.6 Based on this information, ELCat rates the language as being “Severely Endangered” on the scale of absolute number of speakers (see Table 5.2). Since the language is only assessed on one factor (since no other factors are known due to the language’s isolation), the language is also rated as being “Severely Endangered” on the overall LEI scale of endangerment, with the understanding that this is only 20% certain because information is lacking regarding the other three factors in the LEI (see Table 5.5). Dupaningan Agta [duo] is a Malayo-Polynesian language spoken in the north-eastern part of Luzon in the Philippines. Information regarding the social context of Dupaningan Agta can be found in Robinson’s (2008) reference grammar of the language. The language is said to have about 1,400 speakers, and is beginning to lose child speakers. In 5 out of 35 Dupaningan Agta communities, children no longer speak the language. Speakers are bilingual in Ilokano, and even in places where children know Dupaningan Agta, they will often reply in Ilokano when spoken to in Dupaningan Agta. Based on this information, it is possible to assess the language for its overall level of endangerment based on absolute number of speakers and intergenerational transmission, but not speaker number trends and domains of use. Dupaningan Agta is thus assessed to be “Threatened” on the scale of absolute number of speakers since it has an estimated 1,400 speakers (see Table 5.2). It is also assessed to be “Threatened” on the scale of intergenerational transmission, since most adults are speakers, but children generally are not (see Table 5.1). The level of endangerment of the language is therefore attained by this equation: [(2×2)+2]/[(5×2)+5]×100=6/15×100= 40% (threatened), where 6 is the total number of points accorded on both scales of intergenerational transmission and absolute number of speakers, and 15 is the total number of attainable points on both scales, since intergenerational transmission is doubly weighted. As the language is scored on the basis of 15 possible points out of 72

The Language Endangerment Index

25 (5×2 for intergenerational transmission and 5 for absolute number of speakers), we can be 60% certain that the language is “Threatened” on the LEI (see Table 5.5). A language for which we have all LEI assessment factors is Miqie Yi, a Ngwi Loloish language spoken in Yunnan, China. The social circumstances of the language are recorded by Gao (2015), in a paper that compares different methods of assessing the linguistic vitality of Miqie Yi. Most Miqie Yi villages are found in clusters together with other ethnic groups, including the Han, Lipo, Lolo, Nasu, Hani, and Miao. Most elders and some middle-aged Miqie Yi speakers also speak the languages of the other groups near the village, as well as Mandarin. Gao (2015) also notes that intermarriage with the Han Chinese and with other ethnic groups (such as the Lipo) has had a significant impact on the language’s intergenerational transmission. In general, Miqie Yi is spoken by most adults and some children, and it has 8,000 speakers. While a majority of the community speaks the language, speaker numbers are gradually decreasing. The language is also used mainly in the home and may not be the primary language of communication there for many. Based on the information available, Miqie Yi is assessed as “Vulnerable” on the scale of intergenerational transmission (see Table 5.1), “Threatened” on the scales of absolute number of speakers and speaker number trends (see Tables 5.2–5.3), and “Severely Endangered” on the scale of domains of use (see Table 5.4). The level of endangerment for the language can hence be attained with the following equation: [(1×2)+2+2+4]/25×100=10/25×100=40% (threatened); all four factors are used so the total score that the language is assigned is divided by 25, which is the maximum number of points possible on all four scales, with intergenerational transmission doubly weighted. As all factors are used, we can be 100% certain that Miqie Yi is “Threatened” on LEI (see Table 5.5).7 The assessments of all three languages are presented in Table 5.6 for a clearer demonstration of how LEI works. Beyond LEI, ELCat also utilizes two other categories for languages that are thought to have recently lost their last speaker, or those that have experienced a period of broken transmission but are currently being revived. We recognize that languages are often vital parts of culture for speakers, rememberers and their descendants. Recognizing that the topic of “dead” or “extinct” languages is a sensitive one, we utilize the term “Dormant” for languages that are thought to have recently lost their last speakers (see also Belew & Simpson, Chapter 4). This also encourages revitalization efforts, more than a label such as “extinct” would. ELCat also recognizes the efforts of communities that are undertaking efforts to awaken their previously dormant languages, and encourages these efforts with “Awakening” assessments. A good example of an “Awakening” language is Wampanoag [wam], an Algonquian language spoken mainly in Massachusetts. Wampanoag had been dormant for over a century before community members began earnest revitalization methods, and the language now has second language learners (Fenelon and Hall 2008; Hinton 2011). ELCat thus provides an “Awakening” label for such languages that were once dormant, but where there exists within the community a form of targeted language revitalization undertaken and overseen by a coherent organized group of interested parties, with the expressed goal of creating new speakers of the language.

How was the Language Endangerment Index devised and why does it work? LEI was devised with the successes and shortcomings of other methods of vitality assessments in mind. These methods include GIDS (Fishman 1991), UNESCO’s (2003) nine factors for assessing language vitality, Krauss’ (2007) schema for assessing language viability, and 73

2 THREATENED Most adults in the community are speakers, but children generally are not.

1 VULNERABLE Most adults and some children are speakers.

Miqie Yi

2 THREATENED 1000–9999 speakers

2 THREATENED 1000–9999 speakers

2 THREATENED A majority of community members speak the language. Speaker numbers are gradually decreasing.

Used mainly just in the home and/or with family, and may not be the primary language even in these domains for many community members.

ENDANGERED

4 SEVERELY

10/25×100=40% THREATENED

6/15×100=40% THREATENED

ENDANGERED

ENDANGERED

10–99 speakers

4/5×100=80% SEVERELY

4 SEVERELY

Domains of use

(Language Endangerment Index)

Speaker number trends

Intergenerational transmission

Absolute number of speakers

Level of endangerment

Factors of endangerment

Dupaningan Agta

Sentinelese

Language

Table 5.6 Assessment of levels of endangerment of Sentinelese, Dupaningan Agta, and Miqie Yi using LEI

100% certain (25/25 possible points, based on four factors including IT, which is doubly weighted.

60% certain (15/25 possible points, based on two factors, including IT, which is doubly weighted.

20% certain (5/25 possible points, based on one factor)

Level of certainty

The Language Endangerment Index

EGIDS (Lewis & Simons 2010). While these methods are not suitable for use on ELCat for the shared reason that we want to be able to generate an overall level of endangerment regardless of how much or how little information is available on any language, they were built on useful assumptions and lessons, some of which were incorporated into LEI. LEI’s approach is based on the study of individual factors associated with language loss. Krauss (2007:2) states that “it remains a major study. . . to consider factors detracting from language ‘SAFETY’ ” (emphasis in original), although his own schema for assessing language viability is focused mostly on intergenerational transmission. This is similar to GIDS and EGIDS, which are measures of endangerment that assume intergenerational transmission is gradually disrupted across domains until the language is no longer viable. On these measures of language vitality, languages are usually assessed by being matched to descriptors that correspond with particular labels. For example, a level 5 Developing language would be one where “the language is in vigorous use, with literature in a standardized form being used by some though this is not yet widespread or sustainable.”8 Inevitably, descriptors such as these conflate multiple factors such as intergenerational transmission and domains of use. Also, it should be noted that when endangerment assessment relies on the match between a language and one specific descriptor, assessment cannot take place when one lacks very specific types of information. Conversely, an approach that is based on individually delineated factors, such as UNESCO’s nine factors for assessing language endangerment, is more comprehensive regarding the factors associated with language loss. However, it is not clear how UNESCO’s nine factors combine to give an overall score of endangerment, much less what happens when information regarding some of these factors is lacking. In reality, the overall score that is reflected for each language on the UNESCO’s Atlas of the world’s languages in danger (Moseley 2010) appears to be based on one factor alone: intergenerational transmission. Hence, while LEI takes a leaf from UNESCO in examining individual factors associated with language endangerment separately, it also assigns an overall score to the language based on combining individual scores on different factors. LEI then further differentiates itself from the other measures of language endangerment by providing a score of certainty based on the number of factors used. The four factors that appear on LEI have been selected after much deliberation. They are factors that are deemed important enough to have appeared in various forms in other methods of vitality assessment. For example, intergenerational transmission underlies GIDS, EGIDS, Krauss’ schema for assessing language viability, and also UNESCO’s nine factors; again, even though the nine factors are meant to be considered together, the endangerment ratings presented in UNESCO’s Atlas of the world’s languages in danger (Moseley 2010) appear to be based mainly on intergenerational transmission. The developers of these other measures have identified, and rightfully so, the necessity of intergenerational transmission for the continued survival of any language. Again, when a language is not passed on to the next generation, it will consequently fall out of use entirely. Intergenerational transmission is thus deemed to be twice as important as all other factors used on LEI, and it receives double the weight of each of the other factors. The other three factors utilized are also components that are found in other vitality measures, albeit in different forms. For example, while speaker number trends are not one of UNESCO’s nine factors, the proportion of speakers within the total population is. Absolute number of speakers is also a component of UNESCO’s nine factors, and domains of use exist across different factors such as trends in existing language domains and response to new domains and media in this vitality measure. In fact, domains of language use also appear to underlie GIDS and EGIDS, since both describe levels at which particular domains experience language shift, hence viewing intergenerational transmission in a more complex way. LEI, on the other 75

Nala H. Lee and John R. Van Way

hand, chooses to simplify these factors, particularly since they should be factors that are universal and easily comparable across different languages. The four LEI factors of intergenerational transmission, absolute speaker numbers, speaker number trends, and domains of use thus represent basic factors that are usually associated with language loss, and are factors that are universal, comparable, and also easily understood by those with or without training in linguistics. These four factors also involve information that is reasonably attainable. While information such as language attitudes are undoubtedly important and can inform our views on language vitality, they are difficult to obtain and compare. Hence, while the four LEI factors are based in part on the most vital components of LEI’s predecessors, they also meet LEI’s criteria of being universal, comparable, and to some extent, more obtainable than other factors. In designing LEI, these were important considerations because LEI must be utilized for all languages of the world. A final lesson that LEI takes from frameworks such as Krauss’ and EGIDS is that there should be more threatened than safe categories. Given the immensity of the language endangerment problem, many languages cannot be considered safe. The categories and ranges that LEI utilizes may come across as being very conservative with regard to what it deems “Safe.” For example, on the scale of absolute number of speakers, the “Vulnerable” category includes any language that has between 10,000 and 99,999 speakers. We take a stance that is similar to Krauss (2007), who states that a very large proportion of what he deems to be “Safe” languages are spoken by a million or more people, including children, or are languages that have official state or regional support. We are also of the opinion that it is better to err on the side of caution, given the indisputable threat of language death. For these reasons, LEI can be viewed as an index of endangerment rather than an index of vitality.

Conclusions The LEI, designed as a central part of ELCat, has given the field a finely tuned tool for assessing the level of endangerment of languages all over the world. Taking the best of what other previous assessment tools offered, LEI first identifies and then separates the four most crucial individual factors contributing to endangerment (intergenerational transmission, speaker number trends, domains of use, and absolute number of speakers), then recombines these four factors for an overall quantitative measure of endangerment. LEI also recognizes that because researchers often have information about only some of the endangerment factors, treating each factor separately allows the user to quickly ascertain the level of certainty for the language’s overall endangerment score. This certainty index has been built into the LEI, and is a part of all ELCat language entries. It is hoped that any and all stakeholders in the field of endangered languages will find LEI’s strengths and flexibility to be of value in supporting their work to understand and protect the world’s fragile linguistic diversity.

Notes 1 2

The LEI level labeled “Endangered” specifically indicates a moderate level of endangerment, rather than a general indication that a language is to some degree endangered. A language is assigned the label “At Risk” if the LEI score is 0, but the certainty rating is less than 100%. We chose this category because some languages score low on certain factors (e.g., a language with more than 100,000 speakers would receive a 0 or “Safe” endangerment rating if other factors were unknown), but additional research on other factors would be required to determine whether the language is no longer “At Risk.” If additional research shows all four factors are consistent with the “Safe” rating, it gets designated as “Safe” and is no longer shown in the Catalogue.

76

The Language Endangerment Index 3 4 5

6

7 8

Also see www.endangeredlanguages.com/lang/3191. See www.endangeredlanguages.com/lang/1468. Note that the domain of religion is not regarded as an official one, considering that speakers of threatened languages may practice non-dominant religions and traditions. Fishman (1991:99) states that the remaining few speakers of a language include “reciter,” “blessers,” “cursers,” and “prayers.” Note that it is controversial as to whether the survey of the Sentinelese was actually conducted (see Gregory Anderson’s comments on the ELCat language page: www.endangeredlanguages.com/lang/ 5420). Gao (2015) provides not only an overall view of endangerment, but also vitality assessment of individual Miqie Yi villages using LEI. EGIDS (Lewis & Simons 2010) has been updated since its inception, and this description appears in the latest edition on Ethnologue (Simons & Fennig 2017, at www.ethnologue.com/about/languagestatus).

References Aragon, Carolina. 2014. A grammar of Akuntsú, a Tupían language. Honolulu: University of Hawai‘i at Mānoa. (Doctoral dissertation.) Blust, Robert. 2003. Thao dictionary. Taipei: Academica Sinica, Institute of Ethnology. Dalrymple, Mary & Mofu, Suriel. 2012. Language of the world/Materials 487: Dusner. Munich: Lincom. Dwyer, Arienne M. 2011. Tools and techniques for endangered-language assessment and revitalization. In Vitality and viability of minority languages, October 23–24, 2009. New York: Trace Foundation Lecture Series Proceedings. Preprint. Fenelon, James V. & Hall, Thomas D. 2008. Revitalization and indigenous resistance to globalization and neoliberalism. American Behavioral Scientist 51. 1867–1901. Fishman, Joshua A. 1965. Who speaks what language to whom and when? La Linguistique 1. 67–88. Fishman, Joshua. 1991. Reversing language shift. Clevedon: Multilingual Matters. Gao, Katie B. 2015. Assessing the linguistic vitality of Miqie: An endangered Ngwi (Loloish) language of Yunnan, China. Language Documentation & Conservation 9. 164–191. Grenoble, Lenore A. 2016. A response to “Assessing levels of endangerment in the Catalogue of Endangered Languages (ELCat) using the Language Endangerment Index (LEI),” by Nala Huiying Lee & John Van Way. Language in Society 45. 293–300. Hinton, Leanne. 2011. Language revitalization and language pedagogy: New teaching and learning strategies. Language and Education (Special Issue: “Applied linguist needed”: Cross-disciplinary networking for revitalization and education in endangered language contexts), 25. 307–318. Krauss, Michael. 2007. Classification and terminology for degrees of language endangerment. In Brenzinger, Matthias (ed.), Language diversity endangered, 1–8. Berlin: De Gruyter. Lee, Nala H. & Van Way, John. 2016. Assessing levels of endangerment in the Catalogue of Endangered Languages (ELCat) using the Language Endangerment Index (LEI). Language in Society 45. 271–292. Lehmann, Christian. 1999. Documentation of endangered languages: A priority task for linguistics. Arbeitspapiere des Seminars für Sprachwissenschaft der Universität Erfurt 1. 1–15. Erfurt: Universität Erfurt. (www2.uni-erfurt.de/sprachwissenschaft/ASSidUE/ASSidUE01.pdf) Lewis, M. Paul & Simons, Gary F. 2010. Assessing endangerment: expanding Fishman’s GIDS. Revue Roumaine de Linguistique 55. 103–120. Michalowski, Piotr. 2018. Ancient Near Eastern and European isolates. In Lyle Campbell (ed.), Language isolates, 19–58. Abingdon and New York: Routledge. Moseley, Christopher (ed.) 2010. Atlas of the world’s languages in danger. 3rd edn. Paris: United Nations Educational, Scientific and Cultural Organization. (www.unesco.org/culture/languages-atlas/en/ atlasmap.html) Ngyuen, Tam. 2013. A grammar of Bih. Eugene: University of Oregon. (Doctoral dissertation.) Robinson, Laura C. 2008. Dupaningan Agta. Honolulu: University of Hawai‘i at Mānoa. (Doctoral dissertation.) Simons, Gary F. & Fennig, Charles D. 2017. Ethnologue: Languages of the world, twentieth edition. Dallas, Texas: SIL International. (www.ethnologue.com) Singh, N. Iqbal. 1978. The Andaman story. New Delhi: Vikas.

77

Nala H. Lee and John R. Van Way UNESCO Ad hoc expert group on endangered languages (Matthias Brenzinger, Arienne M. Dwyer, Tjeerd de Graaf, Collette Grinevald, Michael Krauss, Osahito Miyaoka, Nicholas Ostler, Osamu Sakiyama, María E. Villalón, Akira Y. Yamamoto, Ofelia Zapeda). 2003. Language vitality and endangerment. Document submitted to the International Expert Meeting on UNESCO Programme Safeguarding of Endangered Languages, Paris, 10–12 March 2003. (www.unesco.org/culture/ich/doc/src/00120-EN. pdf) Wurm, Stephen A. 2007. Australia and the Pacific. In Moseley, Christopher (ed.), Encyclopedia of the world’s endangered languages, 424–557. London: Routledge.

78

6 NEW KNOWLEDGE PRODUCED BY THE CATALOGUE OF ENDANGERED LANGUAGES Lyle Campbell and Eve Okura

One objective of the Catalogue of Endangered Languages (ELCat) was to produce new knowledge. This chapter reports discoveries from ELCat that have resulted in new knowledge about the world’s languages. Among others, these include massive amounts of new information on 3,394 endangered languages, concrete confirmation of the highly accelerated rate of language extinction in recent times, correction of claims about the rate of loss (not one language dying every two weeks, as often claimed, but rather a language becoming dormant about every three months), and clear identification of “unknowns” (the languages of uncontacted groups, unclassified languages, and missing information).

Introduction One of the objectives of the Catalogue of Endangered Languages (ELCat), set out in the workshop that designed the project and by the grants that supported its construction, was to produce new knowledge, and this objective has been fulfilled. Our purpose in this chapter is to describe and call attention to discoveries from ELCat that have resulted in new knowledge about the world’s languages.1 The Catalogue of Endangered Languages contains massive amounts of new information on 3,394 endangered languages. It records information on each language’s location, numbers of speakers, degree of endangerment, classification (genetic affiliation), alternate names, major variants (dialects), resources about the language, etc. It has corrected many errors concerning the very existence of numerous languages. On the one hand, many languages included in other reference works proved spurious, either non-existent or not distinct from other languages that were reported as separate languages. These were removed from the list of languages. On the other hand, ELCat found it necessary to add some 269 languages that are generally missing from other sources; these are the ones lacking ISO codes (in the ISO 639-3 standard for language codes). ELCat also provides for the first time a clear overall view of the status of the world’s languages. ELCat currently lists a total of 3,138 endangered languages.2 That is, 46% of the 79

Lyle Campbell and Eve Okura

world’s 6,879 living languages based on Ethnologue’s figure of languages currently in the world (Simons & Fennig 2017) are endangered.3 This 46% is near to the oft-cited 50% scenario of the earth’s languages expected to become extinct or doomed by the century’s end, but far from the 90% worst-case scenario of languages expected to be extinct by then (cf. Krauss 1992:7). ELCat makes no prediction about dates, about when, if ever, endangered languages will become extinct. Many of the languages at the less endangered end of the endangerment scale will probably survive, though many of the surviving languages will become more endangered than they are now. Nevertheless, this finding that 46% of all languages are endangered does underscore the seriousness of the endangered languages crisis.

Confirmation of accelerated language loss ELCat has provided clear empirical confirmation of the commonly repeated claim that languages are becoming extinct at a highly accelerated rate. The evidence of the sharp increase in language loss is seen in the history of language extinction and in the current status of languages everywhere. We will never know a true number of how many languages have become “extinct” (dormant) throughout history. For now, let’s take 700 as a working figure. This number contains: (1) 484 languages from LINGUIST List’s inventory of 573 extinct languages, minus 89 cases of “old” or “classical” languages in their list that did not actually become extinct but survived in a changed modern form, e.g. Old English, Classical Nahuatl, and so on (http://linguistlist.org/forms/langs/ get-extinct.cfm); and (2) the dormant and awakening languages in ELCat that are not duplicated in LINGUIST List’s directory of extinct languages. There are currently a total of 256 languages that have no known native speakers listed in ELCat, 187 dormant languages and 69 awakening languages. Some of these languages are included also in the LINGUIST List’s inventory of extinct languages, though c.205 languages in the Catalogue are not included in the LINGUIST List tally. ELCat includes entries for languages reported to have lost their last native speaker since 1960 (because it is often difficult to know when a language truly has no remaining speakers) and for languages that no longer have native speakers but for which revival programs exist (called “awakening” languages; see Belew and Simpson, Chapter 4). Together, this makes a total of c.700 extinct languages that we actually know something about. Obviously this number is unsatisfying in many respects and could be expanded greatly. For example, the number of extinct languages in South America that we know something about is 217 (c.162 long-extinct languages and c.55 that lost their last known speaker after 1950), but this is misleading since there are an additional c.381 names of unverified languages from the past that have disappeared and mostly lack attestation, so together possibly c.598 extinct languages in South America alone (see Campbell 2012; in press b), and many other such languages could be added from other regions. If we include the unconfirmed extinct languages, then the list of extinct South American languages alone becomes very large indeed in view of the c.700 figure of extinct languages cited above. Furthermore, it is also generally believed that many other languages became extinct without their names ever being recorded; of these we can say nothing. Clearly, we will never know how many languages have disappeared throughout history, and this challenges any statistics about extinct languages we may hope to present. (See Belew and Simpson, Chapter 4, for additional discussion.) Nevertheless, of the extinct languages that we actually know something of, it is clear nearly 30% became extinct since 1960, again confirming the rapid extinction of languages in recent times. 80

New knowledge produced by ELCat

The number of extinct language families offers additional perspective on the loss of linguistic diversity crisis. Of the world’s c.407 independent language families (including language isolates, language families that have only one member) (Barlow & Campbell Chapter 3, Campbell 2018), already 96 are extinct – no language belonging to any of these families has any remaining native speaker. This means that 24% of the linguistic diversity of the world, calculated in terms of language families, has been lost. Of these extinct language families, 28 became extinct since 1960 (29.2%). This confirms the common claim of significant on-going loss of language diversity, as well as of the highly accelerated rate of language loss in recent times. Of all the millennia in which languages could have disappeared, nearly a third of these language families have become dormant in only the last 60 years, dramatically underscoring the accelerating rate of language extinction in recent times. Unfortunately, many more languages and language families are on the brink of losing their last native speakers, foreshadowing a drastic change in the overall picture of the world’s linguistic diversity. ELCat finds that 269 languages have fewer than 10 speakers, 427 are “Critically Endangered,” 371 are “Severely Endangered,” and together 798 are “Critically” or “Severely” Endangered – that is 25.4% of living endangered languages, 11.6% of all living languages, are at high risk of loss. Unless revitalization efforts are successful, most of these will become extinct in the near future.4

An important corrective Very importantly, the frequently repeated claim that one language goes extinct every two weeks is not supported. Contrary to this often-cited claim, ELCat findings show that on average about one language each three months becomes extinct, or 4.3 per year (Campbell et al. 2013). Understandably, people are concerned with the crisis of endangered languages, and the claim that a language goes extinct each two weeks dramatically called attention to the plight of these languages, which explains, we assume, why the claim has been repeated so often. There is, however, no need to repeat this inaccurate and exaggerated claim as propaganda for the cause – the truth about what is happening to endangered languages is sufficiently shocking for anyone, as seen, for example, in the list of recently extinct languages, in the list of languages with fewer than ten speakers, probably all of which will soon be extinct (if revitalization efforts do not succeed), and in case after case of rapid decline in endangered languages represented in ELCat. The evidence is more than sufficient for purposes of alarm, and dramatically confirms the ongoing crisis of accelerated language loss.

Known unknowns To this point we have been talking about “known knowns” (to borrow from Donald Rumsfeld), about newly discovered things, new knowledge derived from ELCat. The discoveries from ELCat also include, however, “known unknowns,” things about which we now have come to realize more clearly that we do not know and need to dedicate more attention to. The “known unknowns” include uncontacted groups, unclassified languages, languages that have scant if any documentation, and missing information, among other things.

Uncontacted groups One reason that we do not know the exact number of endangered languages, or of languages in general, or of distinct language families, or of language isolates is because of the “uncontacted” 81

Lyle Campbell and Eve Okura

peoples around the world, especially in Amazonia in South America. It was estimated in 2013 that there were some 100 uncontacted “tribes” around the world, with some 90 in South America, but with some also in New Guinea, Central Africa, and elsewhere (Holmes 2013). In Brazil alone there are at least 40 “uncontacted” isolated indigenous groups, and some count upwards of 70 such groups there. (https://en.wikipedia.org/wiki/Uncontacted_peoples#South_America, accessed 9-6-2016). Brazil’s FUNAI (Fundação Nacional do Índio) mentions 107 “isolated groups,” some previously contacted, but many not yet contacted and mostly unknown (www. funai.gov.br/index.php/nossas-acoes/povos-indigenas-isolados-e-de-recente-contato? limitstart=0#, accessed 3-23-2017). For many of these cases, it is not known whether the people speak a variety of an already identified language, a language currently unknown but which belongs to a known language family, a language that represents an as yet unknown language family, or a language with no other relatives in the world, a language isolate. Where the languages of these uncontacted groups are not known, they do not show up in the entries in the Catalogue of Endangered Languages, although it is assumed that most of them are endangered.

Unclassified languages There are a good number of unclassified languages whose genetic affiliation is unknown. Many of these are long extinct languages that are too poorly attested to be grouped with any other language or language family. These are of less direct relevance to us as we catalogue today’s endangered languages. Of more immediate interest are the extant languages that cannot be classified for lack of data, languages not yet described sufficiently to compare them meaningfully with other languages in order to determine whether they may have relatives. ELCat list 25 endangered unclassified languages. A few examples include: Himarimã (South America), Kembra (Indonesia [Western New Guinea]), Kujargé (Africa), Larrakiya (Australia), Lepki (Indonesia [Western New Guinea]), Mpre (Mpra) (Africa), Ngomburr (Australia), Sentinel(ese) (India [Andaman Islands]), Turumsa (Papua New Guinea), etc. It is important not to confuse unclassified languages with language isolates. Languages that have sufficient documentation for them to be compared meaningfully with other languages but where it has not been possible to demonstrate any genetic relationship with any other language or language family are language isolates, of which a large number are endangered and are thus listed in ELCat. Language isolates are not “unclassified”; they are just not related (relatable) to other languages (see Campbell 2018).

Missing information About the unclassified languages and languages of uncontacted groups we have very little information, if any. However, there are also cases of better-known languages about which we may have information of some sorts but where we still lack important information of other sorts. These are also known unknowns – things that we know we do not know – that we should investigate so that we can fill in the missing information. The missing information can be of various sorts. For a number of languages, the best available sources are out of date or in some ways inaccurate. For many endangered languages we do not have reliable information in categories such as the number of speakers, the age of the youngest speakers, the domains in which the language is used, whether children are learning the language, whether the number of speakers is declining, or the location and distribution of the language, among others. (See Belew and Simpson, in press, and Hauk & Heaton, Chapter 9, for further discussion of missing information about endangered languages.) 82

New knowledge produced by ELCat

Conclusions We side with the oft-repeated quote from Nicolaus Copernicus (1473–1543): “To know that we know what we know, and to know that we do not know what we do not know, is true knowledge.” The Catalogue of Endangered Languages indeed has provided new knowledge, knowledge that allows us now to know things that were previously unknown, new “known knowns.” Its goals include the search for true knowledge by obtaining information that we know that we do not currently have on many endangered languages, known unknowns, and through this new information, produce new knowledge relevant to ELCat’s several audiences, contributing to understanding of endangered languages generally. Carrying the “knowns/unknowns” terminology further, there are almost certainly “unknown unknowns,” unanticipated discoveries yet to be made. We do not know what ELCat research will discover as we go forward, though we are confident that continued discovery does lie ahead of us. The goal is to fill in as much of the missing information as possible, to correct errors in existing sources, and to push back the frontiers of knowledge about endangered languages. ELCat’s user interface invites submissions that recommend improvements, information about the unknowns, and additions and improvements on the knowns.

Notes 1

2 3

4

The goals and basic organization of the Catalogue of Endangered Languages were established in an international workshop with some 50 specialists from around the world sponsored by a National Science Foundation grant, and grants that supported the construction of the Catalogue were from the National Science Foundation and the Henry Luce Foundation (see Campbell and Belew, Chapter 1, for details). The total number of languages listed in the Catalogue of Endangered Languages is 3,394, but this includes 256 dormant or awakening languages, that is, languages with no known native speakers. Ethnologue lists a total of 7,099 languages (Simons & Fennig 2017); however, when adjusted for the 220 languages with no known speakers and one constructed language, Ethnologue’s number of living languages is 6,879. These figures may be even somewhat higher, since there are 37 languages for which ELCat has no information on their vitality status or number of speakers – we presume many of these have very few speakers and are endangered, since there is some information, if very limited, on most languages with larger numbers of speakers. (Ethnologue reports 119 for which speaker numbers are unknown, c.3.1% of their total number of languages, Simons & Fennig 2017.)

References Belew, Anna & Simpson, Sean. In press. The status of the world’s endangered languages. In Rehg, Kenneth L. & Campbell, Lyle (eds.), Handbook of endangered languages. Oxford: Oxford University Press. Campbell, Lyle. 2012. The classification of South American indigenous languages. In Campbell, Lyle & Grondona, Verónica (eds.), The indigenous languages of South America: A comprehensive guide, 59–166. Berlin: De Gruyter. Campbell, Lyle. In press a. Languages of South America. In Moseley, Christopher J. & Asher, Ronald E. (eds.), Atlas of the world’s languages. London: Routledge. Campbell, Lyle. 2018. Language isolates and their history. In Campbell, Lyle (ed.), Language isolates, 1–18. Abingdon and New York: Routledge. Campbell, Lyle & Heaton, Raina & Lee, Nala & Okura, Eve & Simpson, Sean & Ueki, Kaori & Van Way, John. 2013. New knowledge: Findings from the Catalogue of Endangered Languages (“ELCat”). (Paper presented at the 3rd International Conference on Language Documentation and Conservation, Honolulu, February 28–March 3, 2013.) (https://scholarspace.manoa.hawaii.edu/bitstream/10125/ 26145/2/26145.pdf)

83

Lyle Campbell and Eve Okura Holmes, Bob. 2013. How many uncontacted tribes are there left in the world? New Scientist, August 22, 2013. (www.newscientist.com/article/dn24090-how-many-uncontacted-tribes-are-left-in-the-world/) Krauss, Michael E. 1992. The world’s languages in crisis. Language 68. 4–10. Simons, Gary F. & Fennig, Charles D. (eds.). 2017. Ethnologue: Languages of the world, 20th edn. Dallas: SIL International.

84

7 THE WORLD’S ENDANGERED LANGUAGES AND THEIR STATUS Anna Belew, Yen-ling Chen, Lyle Campbell, Russell Barlow, Bryn Hauk, Raina Heaton, and Stephanie Walla

The Catalogue of Endangered Languages has compiled data on the current vitality of 3,394 languages, improving upon the available information in many other sources. This chapter presents an abbreviated version of the Catalogue’s data, providing a complete list of the endangered, dormant, and awakening languages of the world, organized by geographic region. In addition to the Catalogue’s data, each section is introduced by a brief overview of language endangerment in that region. This chapter serves as a useful reference for the world’s endangered languages, as well as pointing readers to additional resources regarding language endangerment in each region. This chapter lists the endangered languages of the world and provides an overview of their current status, based on the data in the Catalogue of Endangered Languages. The Catalogue is primarily a digital database of language information, designed so it can be updated regularly as new information becomes available. In this chapter we provide a summary of this information for any who prefer to see it in print or in abbreviated form, for those who have unreliable internet access, and for any administrators who still trust only hard-copy, peer-reviewed publications and distrust internet sources. The abbreviated information presented here from the Catalogue is intended to serve as a snapshot of the world’s endangered languages in early 2017. The online Catalogue is constantly being updated, rather than being periodically released as new editions. This means that this printed version can serve any in the future who wish to compare the status of endangered languages and the information available about them in the Catalogue with the data available in the past. Here, the world’s endangered languages are listed by regions, with a short discussion of causes and rates of language endangerment in each region. The geographic divisions used by the Catalogue of Endangered Languages are similar to, for example, the United Nations geoscheme, which divides the world into five primary regions; however, ELCat’s 12 regional divisions are a reflection of both language areas, and the particular expertise of its personnel. For example, Australia is a “subregion” of the Pacific in the UN geoscheme, but a primary region in ELCat; it constitutes a distinct language area and is under the supervision of a regional director who is an expert in Australian languages, not in languages of the wider Pacific. 85

Anna Belew et al.

It should also be noted that some languages are spoken in multiple countries, some of which straddle regional divisions. In these cases, for the purposes of this chapter, each language is assigned to a single region according to where a majority of its speakers are found. For example, Laqua [laq] has speakers in both China (which is part of East Asia within ELCat’s regional divisions) and Vietnam (part of ELCat’s Southeast Asia region). As it is reported to have more speakers in Vietnam than in China, it is listed here in the Southeast Asia section. If a language spoken across national borders is not listed in the expected regional section, such as Laqua in East Asia, users should look for it in the section covering a neighboring region. The listings of endangered languages provided in the following sections are drawn from the Catalogue’s database as of April 4, 2017. They reflect the data provided by the “preferred” source for each language (see Campbell and Belew, Chapter 1). The information in each listing is structured as follows: Primary language name (Up to three alternate names) [ISO 639-3 code, if one exists]: LEI endangerment rating, number of speakers (if available); language family affiliation;1 country/countries where spoken.

Africa Anna Belew Linguistic diversity in Africa The African continent is home to extensive cultural and linguistic diversity. More than 2,000 African languages have been identified; the 20th edition of the Ethnologue provides a figure of 2,144 living languages in Africa (Simons & Fennig 2017), to which can be added another 39 living languages in the Catalogue of Endangered Languages that are not yet recognized by ISO 639-3. Given an estimate of 7,099 living and recently dormant languages (ibid.), Africa is thus home to 30% of the world’s languages. Africa is sometimes misperceived as lacking in linguistic diversity. Joseph Greenberg’s (1963) classification, which assigned nearly all of Africa’s languages to one of only four families, still seems to haunt popular conceptions of Africa as relatively linguistically homogenous (see Mous 2003). However, this perception of homogeneity is mistaken: Greenberg's four large groupings are disputed (Khoisan no longer accepted, Nilo-Saharan rejected by many, and even some of the membership of Niger-Congo and Afroasiatic is contested), and some ten smaller families have been identified, as well as at least six and perhaps up to 20 linguistic isolates (Blench 2018; Dimmendaal 2015). Many of Africa’s endangered languages contain typologically rare features, as well; for example, an endangered Cushitic language, Dahalo, is one of the only known human languages to make use of all four airstream mechanisms (clicks, ejectives, implosives, and pulmonics) (Maddieson et al. 1993). The degree of linguistic diversity – and accompanying cultural, scientific, and historical knowledge – that stands to be lost in Africa is enormous.

Causes of language endangerment in Africa The endangered languages of Africa face many of the same pressures as endangered languages across the world: namely, reductions in speaker numbers, contraction of domains of use, disrupted intergenerational transmission, negative attitudes, and/or socioeconomic and cultural 86

The world’s endangered languages

incentives for language shift. In most African countries (42 of 54 UN-recognized nations in Africa), English, French, Arabic, Spanish and/or Portuguese are the sole official languages, with limited or no official status for indigenous languages; however, the role of “official” languages varies widely, and in many contexts they may have only a minor impact on endangered languages (see Connell 2015). While common threads run through language endangerment scenarios across the globe, any overview of language endangerment in Africa must take into account sociolinguistic factors which are more prevalent in African contexts than elsewhere. For one, as noted in Mous (2003), many accounts of language endangerment focus on imported or colonial languages as the targets of language shift (e.g. widespread shift to English, French, Portuguese, Spanish, or a creolized version of the colonial language). However, in Africa, shift to indigenous languages of wider communication may be more typical: for example, shift to Fulfulde varieties across the Sahel, Swahili in Kenya and Tanzania, Wolof in Senegal, or Sango in the Central African Republic. Similarly, language endangerment is often imagined to result in a speaker population that is monolingual in a dominant language (e.g. shift to English in Australia); in most African contexts, though, multilingualism is the norm, and shift away from an endangered language generally means dropping it from a repertoire of multiple languages and varieties, not reduction of one’s repertoire to a single language (see Lüpke 2015). These factors – shift to other indigenous languages and widespread multilingualism – can render language shift and endangerment “harder to see” for outside scholars, and certainly more complicated to assess, than scenarios in which speakers transition to monolingualism in an official or national language. “External” endangerment factors – that is, factors that impact the stability or safety of speaker communities themselves – may also have severe repercussions for language endangerment in Africa. The Second Sudanese Civil War, from 1983 through 2005, caused the deaths of an estimated two million people and displaced perhaps five million more. The conflict had a severely detrimental impact on the languages of the Republic of South Sudan, where an estimated 70% of languages of the former Sudan are found (Abu-Manga 2010). Other ongoing armed conflicts, such as the Boko Haram insurgency in Nigeria and neighboring countries, can destabilize or displace speaker populations and sociolinguistic contexts, and preclude the conduct of language surveys or documentation work in affected regions. There are likely many languages that are yet unknown to outside linguists in areas which have sustained prolonged conflict, such as the Republic of South Sudan, parts of the Democratic Republic of the Congo, and the Central African Republic. (Indeed, these countries are among those for which the Catalogue of Endangered Languages’ data is most sparse.) Similarly, desertification, drought, natural disasters, and other effects of climate change which impact economic activity and agriculture are likely to increase migration into urban centers in coming decades. The so-called “rural exodus” into cities has frequently been identified as a major factor in disturbing the social, political, geographic, and economic systems which sustain language maintenance, and often contributes to language shift.2

The current status of endangered languages in Africa Below is presented a brief overview of the endangered languages of Africa, intended to provide a statistical snapshot rather than a comprehensive treatment; for further discussion of language endangerment in Africa see Brenzinger (1992, 2008), Batibo (2005), Kandybowicz & Torrence (2017) and Essegbey et al. (2015). At the time of writing, the Catalogue of Endangered Languages identifies 604 endangered languages and 16 dormant languages in 49 countries and territories of Africa.3 Given an estimate 87

Anna Belew et al.

of 2,183 living languages in Africa (see p. 86), 27.6% of the continent’s languages are currently endangered – a figure significantly lower than the global rate of roughly 46% (see Campbell and Okura, Chapter 6). Of the endangered and dormant languages in Africa, 1.9% are “At Risk,” and a plurality (40.1%) fall into the “Threatened” category, while the second most common endangerment rating is “Vulnerable” (25.4%) – the three least severe endangerment ratings account for 67.4% of endangered and dormant African languages. Only 13.5% of endangered African languages are rated “Severely” or “Critically Endangered.” This stands in notable contrast to global averages, where 25.4% of endangered languages are at least “Severely Endangered.” Similarly, “Threatened,” “Vulnerable,” and “At Risk” languages together account for only 45.4% of languages in the Catalogue worldwide. Overall, statistically speaking, endangered African languages appear to be slightly less at risk than endangered languages globally. Geographically, the highest rates of language endangerment correspond with areas of high linguistic diversity. The Nigeria-Cameroon borderland and surrounding area is famously linguistically diverse, and unsurprisingly, Nigeria and Cameroon are the African nations with the largest numbers of endangered and dormant languages (171 and 75, respectively). Endangered languages also cluster in other nations at roughly 10°N latitude: Chad (50), Republic of Sudan (45), and Ethiopia (45) are home to the next highest numbers of endangered and dormant languages. Below is an alphabetical listing of the languages of Africa currently included in the Catalogue of Endangered Languages.

Endangered languages of Africa !Xun (!Kung-Ekoka, !Xũ, North-Central Ju) [knw]: Vulnerable, 14,000–18,000 speakers; Kx’a; spoken in South Africa, Namibia, and Angola. ||Ani (//Ani, /Anda, Handá) [hnh]: Endangered, 1,000 speakers; Khoe; spoken in Botswana. ||Gana (G//ana, Dxana) [gnk]: Threatened, 1,030 speakers; Khoe; spoken in Botswana. |Gui (|Gwi, G/wi, /Gwi) [gwj]: Threatened, 1,470 speakers; Khoe; spoken in Botswana. ǂHoan (ǂHuã, ǂHuan, ≠Hû) [huc]: Severely endangered, fewer than 40 speakers; Kx’a; spoken in Botswana. ǂKx’au||’ein (Gobabis Kung, Auen) [aue]: Threatened; Kx’a; spoken in Botswana and Namibia. Aasáx (Asax, Asá, Aasá) [aas]: Dormant; Afro-Asiatic; formerly spoken in Tanzania. Acheron (Garme, Asheron, Aceron) [acz]: Vulnerable, 70,000 speakers; Niger-Congo; spoken in Sudan. Adamorobe Sign Language [ads]: Threatened, 35 speakers; sign language; spoken in Ghana. Afitti (Ditti, Unietti) [aft]: Endangered, approximately 4,000 speakers; Nyimang; spoken in Sudan. Aguna (Awuna, Agunaco) [aug]: Threatened, 3,470 speakers; Niger-Congo; spoken in Benin. Àhàn (Ahaan) [ahn]: Endangered, 300 speakers; Niger-Congo; spoken in Nigeria. Aja (Ajja, Adja) [aja]: Endangered, 200 speakers; Kresh-Aja; spoken in Sudan. Ajumbu (Mbu’) [muc]: Endangered, 200–300 speakers; Niger-Congo; spoken in Cameroon. Aka (Sillok, Jebels Sillok, Jebel Silak) [soh]: Endangered, a few hundred speakers; Eastern Jebel; spoken in Sudan. Ake (Aike, Akye) [aik]: Vulnerable, fewer than 2,000 speakers; Niger-Congo; spoken in Nigeria. Akum (Anyar, Okum) [aku]: Endangered, 600 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Alago (Aragu, Arago, Argo) [ala]: Vulnerable, 15,000 speakers; Niger-Congo; spoken in Nigeria. 88

The world’s endangered languages

Alagwa (Alagwaisi, Alagwase, Alawa) [wbj]: Threatened, fewer than 10,000 speakers; AfroAsiatic; spoken in Tanzania. Ambele (Tinta, Ambala, Bata’o) [ael]: Vulnerable, 3,600 speakers; Niger-Congo; spoken in Cameroon. Amdang (Mimi, Mima, Mututu) [amj]: Threatened; Furan; spoken in Chad. Ampari (Ampari Dogon, Ambeenge) [aqd]: Threatened, 5,200 speakers; Niger-Congo; spoken in Mali. Anaang [anw]: Vulnerable, 733,040 speakers; Niger-Congo; spoken in Nigeria. Anfillo (Mao, Southern Mao) [myo]: Endangered, approximately 500 speakers; Afro-Asiatic; spoken in Ethiopia. Anii (Gisida, Basila, Baseca) [blo]: Vulnerable, fewer than 25,000 speakers; Niger-Congo; spoken in Togo and Benin. Animere (Anyimere, Kunda) [anf]: Severely endangered, 30 speakers; Niger-Congo; spoken in Ghana. Arbore (Arbora, Erbore, Irbore) [arv]: Threatened, 4,441 speakers; Afro-Asiatic; spoken in Ethiopia. Argobba (Argoba) [agj]: Vulnerable, 10,860 speakers; Afro-Asiatic; spoken in Ethiopia. Assangori (Sungor, Soungor, Assoungor) [sjg]: Vulnerable, 23,500 speakers; Taman; spoken in Sudan and Chad. Atong (Etoh) [ato]: Vulnerable, 4,200 speakers; Niger-Congo; spoken in Cameroon. Aweer (Boni) [bob]: Endangered, fewer than 200 speakers; Afro-Asiatic; spoken in Kenya and Somalia. Awjilah (Awjila, Aujila, Augila) [auj]: Threatened, fewer than 2,000 speakers; Afro-Asiatic; spoken in Libya. Awngi (Awiya, Awi, Agaw) [awn]: At risk, 500,000 speakers; Afro-Asiatic; spoken in Ethiopia. Ayu (Aya) [ayu]: Severely endangered, a few hundred speakers; Niger-Congo; spoken in Nigeria. Baale (Kacipo-Baalesi) [koe]: Threatened, approximately 9,000 speakers; Surmic; spoken in Sudan and Ethiopia. Baan (Ogoi) [bvj]: Threatened, fewer than 5,000 speakers; Niger-Congo; spoken in Nigeria. Babalia Creole Arabic [bbz]: Threatened, 3,940 speakers; pidgin or creole; spoken in Chad. Bade (Bede, Gidgid) [bde]: At risk, approximately 250,000 speakers; Afro-Asiatic; spoken in Nigeria. Badyara (Badara, Badian) [pbp]: Vulnerable, at least 12,000 speakers; Niger-Congo; spoken in Senegal, Guinea, and Guinea-Bissau. Baga Binari (Binareg) [bcg]: Threatened, 3,000 speakers; Niger-Congo; spoken in Guinea. Baga Koga (Barka) [bgo]: Threatened, fewer than 5,000 speakers; Niger-Congo; spoken in Guinea. Baga Manduri (Mandari, Maduri) [bmd]: Threatened, 4,000 speakers; Niger-Congo; spoken in Guinea. Baga Mboteni [bgm]: Threatened, fewer than 4,900 speakers; Niger-Congo; spoken in Guinea. Baga Sitemu (Barka, Sitemuú, Tchitem) [bsp]: Threatened, 4,000 speakers; Niger-Congo; spoken in Guinea. Bago-Kusuntu (Koussountou) [bqg]: Threatened, 7,500 speakers; Niger-Congo; spoken in Togo. Bai (Bari) [bdj]: Threatened, 2,500 speakers; Niger-Congo; spoken in South Sudan. Bainouk-Gunyaamolo (Niamone) [bcz]: Threatened, approximately 6,000 speakers; NigerCongo; spoken in Senegal and Gambia. 89

Anna Belew et al.

Bainouk-Samik [bcb]: Threatened, 1,850 speakers; Niger-Congo; spoken in Senegal. Baïnounk Gubëeher [no ISO 639-3 code]: Threatened, 1,100 speakers; Niger-Congo; spoken in Senegal. Baïnounk Gujaher [no ISO 639-3 code]: Vulnerable; Niger-Congo; spoken in Senegal. Baka (East Region, Cameroon) (Bayaka, Bibaya) [bkc]: Vulnerable, 30,000–50,000 speakers; Niger-Congo; spoken in Cameroon and Gabon. Baka (Far North Region, Cameroon) [no ISO 639-3 code]: Severely endangered, 140 speakers; Afro-Asiatic; spoken in Cameroon. Bakole (Bamusso) [kme]: Endangered, 300 speakers; Niger-Congo; spoken in Cameroon. Bakpinka (Begbungba, Uwet, Iyongiyong) [bbs]: Threatened, 4,000 speakers; Niger-Congo; spoken in Nigeria. Baldemu (Balda, Mbazla) [bdn]: Critically endangered, fewer than three speakers; Afro-Asiatic; spoken in Cameroon. Bali (Bibaali) [bcn]: Threatened, 1,000 speakers; Niger-Congo; spoken in Nigeria. Bana (Koma, Banaeg) [bcw]: Vulnerable, 23,000 speakers; Afro-Asiatic; spoken in Cameroon. Bandial (Banjaal, Banjal) [bqj]: Threatened, approximately 7,000 speakers; Niger-Congo; spoken in Senegal. Bangandu (Temne, Bagando, Bangando) [bgf]: Threatened, 2,700 speakers; Niger-Congo; spoken in Cameroon and Congo. Bangime (Bangi Me, Bangeri Me) [dba]: Threatened, 1,200–3,000 speakers; isolate; spoken in Mali. Bankagooma (Bankagoma, Bankagoroma) [bxw]: Threatened, 6,000 speakers; Niger-Congo; spoken in Mali. Bankan Tey (Walo, Oualo) [dbw]: Threatened, 1,320 speakers; Niger-Congo; spoken in Mali. Bankon (Bo, Abaw) [abb]: Vulnerable, 12,000 speakers; Niger-Congo; spoken in Cameroon. Barain (Barein, Guilia) [bva]: Threatened, 4,100 speakers; Afro-Asiatic; spoken in Chad. Barambu (Barambo, Amiangba, Amiangbwa) [brm]: Vulnerable, 25,600 speakers; NigerCongo; spoken in Democratic Republic of the Congo. Barikanchi [bxo]: Unknown vitality; pidgin or creole; spoken in Nigeria. Barombi (Lombi, Lambi, Rombi) [bbi]: Threatened, approximately 3,000 speakers; NigerCongo; spoken in Cameroon. Basketo (Basketto, Baskatta, Mesketo) [bst]: Vulnerable, approximately 60,000 speakers; AfroAsiatic; spoken in Ethiopia. Bassa-Kontagora (Basa, Basa-Kontagora) [bsr]: Critically endangered, ten speakers; NigerCongo; spoken in Nigeria. Bassari (Basari, Tenda Basari, Biyan) [bsc]: Vulnerable, 10,000–25,000 speakers; Niger-Congo; spoken in Senegal, Guinea, and Guinea-Bissau. Bati (Bati Ba Ngong, Bati de Brousse) [btc]: Endangered, 500–1,000 speakers; Niger-Congo; spoken in Cameroon. Baygo (Baigo, Bego, Beko) [byg]: Dormant; Daju; formerly spoken in Sudan. Bayot (Baiote, Baiot, Bayotte) [bda]: Vulnerable, 17,150 speakers; Niger-Congo; spoken in Senegal, Gambia, and Guinea-Bissau. Bayso (Gidicho, Baiso, Alkali) [bsw]: Threatened, approximately 3,200 speakers; Afro-Asiatic; spoken in Ethiopia. Beeke (Beke, Ibeeke) [bkf]: Endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Beezen [bnz]: Endangered, 400 speakers; Niger-Congo; spoken in Cameroon. 90

The world’s endangered languages

Belanda Bor [bxb]: Threatened, 8,000 speakers; Nilotic; spoken in South Sudan. Bele (Beele, Àbéélé, Bellawa) [bxq]: Endangered, 120 speakers; Afro-Asiatic; spoken in Nigeria. Beli (Behli, Beili, Jur Beli) [blm]: Threatened, 6,600 speakers; Central Sudanic; spoken in South Sudan. Ben Tey (Beni) [dbt]: Threatened, 3,000 speakers; Niger-Congo; spoken in Mali. Bende (Sibhende, Kibende, Si’bende) [bdp]: Endangered, 27,000–41,290 speakers; NigerCongo; spoken in Tanzania. Benga (Qaqet) [bng]: Severely endangered; Niger-Congo; spoken in Gabon and Equatorial Guinea. Beni Iznassen (Beni Snassen, Iznacen) [no ISO 639-3 code]: Severely endangered; Afro-Asiatic; spoken in Morocco. Berakou (Babalia, Bubalia) [bxv]: Critically endangered, two speakers; Central Sudanic; spoken in Chad. Besme (Huner, Hounar, ‘Unar) [bes]: Threatened, 1,230 speakers; Niger-Congo; spoken in Chad. Bete [byf]: Severely endangered, a few speakers; Niger-Congo; spoken in Nigeria. Beti (Eotile) [eot]: Endangered, 200 speakers; Niger-Congo; spoken in Côte d’Ivoire. Biafada (Beafada, Biafar, Bidyola) [bif]: Vulnerable, 44,900 speakers; Niger-Congo; spoken in Guinea-Bissau. Bikya (Furu) [byb]: Critically endangered, one speaker; Niger-Congo; spoken in Cameroon. Bilen (Bilin, Bogo, Bogos) [byn]: Vulnerable, approximately 90,000 speakers; Afro-Asiatic; spoken in Eritrea. Bina (Bogana, Binawa) [byj]: Threatened, 7,000 speakers; Niger-Congo; spoken in Nigeria. Birri (Biri, Viri, Bviri) [bvq]: Endangered, 200 speakers; Unclassified; spoken in Central African Republic. Birwa [brl]: Vulnerable; Niger-Congo; spoken in South Africa and Botswana. Biseni (Amegi, Northeast Central Ijo) [ije]: Threatened, 4,800 speakers; Niger-Congo; spoken in Nigeria. Bishuo (Biyam, Furu) [bwh]: Critically endangered, one speaker; Niger-Congo; spoken in Cameroon. Bodi (Podi) [no ISO 639-3 code]: Threatened, 4,570 speakers; Surmic; spoken in Ethiopia. Bodo [boy]: Severely endangered, 15 speakers; Niger-Congo; spoken in Central African Republic. Boguru (Koguru, Kogoro, Buguru) [bqu]: Endangered, fewer than 500 speakers; NigerCongo; spoken in South Sudan and Democratic Republic of the Congo. Bolo (Haka, Libolo, Lubolo) [blv]: Threatened, 2,630 speakers; Niger-Congo; spoken in Angola. Boloki (Baloki, Buluki, Boleki) [bkt]: Threatened, 4,200 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Bolondo [bzm]: Threatened, 3,000 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Bom (Bome, Bum, Bomo) [bmf]: Critically endangered, a few hundred speakers; Niger-Congo; spoken in Sierra Leone. Bomboli (Bombongo) [bml]: Threatened, 2,500 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Bon Gula (Êeni, Gula Guera, Taataal) [glc]: Threatened, approximately 1,500 speakers; NigerCongo; spoken in Chad. 91

Anna Belew et al.

Bondei (Kibondei, Bonde) [bou]: Vulnerable, 80,000 speakers; Niger-Congo; spoken in Tanzania. Bondu So (Bondum Dom, Najamba-Kindige) [dbu]: Vulnerable, 24,700 speakers; Niger-Congo; spoken in Mali. Bongo (Bungu, Dor) [bot]: Vulnerable, 10,100 speakers; Central Sudanic; spoken in Sudan. Bonjo [bok]: Endangered, 2,000–3,000 speakers; Niger-Congo; spoken in Congo. Boon (Af-Boon, Boni) [bnl]: Severely endangered, 59 speakers; Afro-Asiatic; spoken in Somalia. Boor (Bwara, Damraw) [bvf]: Severely endangered, fewer than 100 speakers; Afro-Asiatic; spoken in Chad. Boró (Shinassha, Borna, Bworo) [bwo]: Threatened, approximately 4,000 speakers; AfroAsiatic; spoken in Ethiopia. Boto (Boot, Bibot) [no ISO 639-3 code]: Endangered, approximately 500 speakers; Afro-Asiatic; spoken in Nigeria. Bu (Abu) [jid]: Threatened, approximately 6,000 speakers; Niger-Congo; spoken in Nigeria. Bubia (Wovea, Bobe, Bobea) [bbx]: Severely endangered, 600 speakers; Niger-Congo; spoken in Cameroon. Buduma (Yidena, Yedima) [bdm]: Vulnerable, 54,800 speakers; Afro-Asiatic; spoken in Nigeria, Cameroon, and Chad. Buli (Bili, Bɨlɨ, Bələ) [no ISO 639-3 code]: Endangered; Afro-Asiatic; spoken in Nigeria. Bung [bqd]: Dormant; Niger-Congo; formerly spoken in Cameroon. Bunoge (Korandabo, Budu-Tagu) [dgb]: Threatened, 1,000 speakers; Niger-Congo; spoken in Mali. Bure (Bubburè) [bvh]: Critically endangered, approximately ten speakers; Afro-Asiatic; spoken in Nigeria. Burji (Bambala, Bembala, Daashi) [bji]: Vulnerable, 46,100 speakers; Afro-Asiatic; spoken in Kenya and Ethiopia. Burunge (Mbulugwe) [bds]: Endangered, a few hundred speakers; Afro-Asiatic; spoken in Tanzania. Buso [bso]: Severely endangered, 40–50 speakers; Afro-Asiatic; spoken in Chad. Bussa (Muusiye, Mossiya, Mossittaata) [dox]: Threatened, approximately 1,500 speakers; AfroAsiatic; spoken in Ethiopia. Busuu (Awa, Furu) [bju]: Critically endangered, eight speakers; Niger-Congo; spoken in Cameroon. Buu [no ISO 639-3 code]: Endangered, 100–200 speakers; Niger-Congo; spoken in Cameroon. Buu (Zaranda) [no ISO 639-3 code]: Severely endangered, approximately 600 speakers; AfroAsiatic; spoken in Nigeria. Bwisi (Ibwisi, Mbwisi) [bwz]: Threatened, 4,250 speakers; Niger-Congo; spoken in Gabon and Congo. C’ara (Chara, Ciara) [cra]: Threatened, 6,932 speakers; Afro-Asiatic; spoken in Ethiopia. Cara (Chara, Fachara, Nfachara) [cfd]: Threatened, fewer than 3,000 speakers; Niger-Congo; spoken in Nigeria. Centúúm (Cen Tuum, Jalaa) [cet]: Critically endangered; isolate; spoken in Nigeria. Chaari (Chari, Caari, Cààrí) [no ISO 639-3 code]: Endangered, approximately 500 speakers; Afro-Asiatic; spoken in Nigeria. Chakali [cli]: Threatened, approximately 5,000 speakers; Niger-Congo; spoken in Ghana. Chala (Tschala, Tsa, Cãla) [cll]: Threatened, 3,000 speakers; Niger-Congo; spoken in Ghana. 92

The world’s endangered languages

Chenoua (Shenowa, Sheliff Basin) [cnu]: Vulnerable, 76,300 speakers; Afro-Asiatic; spoken in Algeria. Cicipu (Acipanci, Achipa, Sagamuk) [awc]: Vulnerable, 20,000 speakers; Niger-Congo; spoken in Nigeria. Daasanach (Dhaasanac, Dasenech, Daasanech) [dsh]: Vulnerable, 32,064 speakers; AfroAsiatic; spoken in Kenya and Ethiopia. Daba (Dabba, Kola) [dbq]: Vulnerable, 25,000 speakers; Afro-Asiatic; spoken in Nigeria and Cameroon. Dahālík (Dahlak, Dahalik) [dlk]: Threatened, approximately 2,500 speakers; Afro-Asiatic; spoken in Eritrea. Dahalo (Sanye, Guo Garimani) [dal]: Endangered, fewer than 400 speakers; Afro-Asiatic; spoken in Kenya. Daho-Doo [das]: Threatened, 4,000 speakers; Niger-Congo; spoken in Côte d’Ivoire. Dair (Daier, Thaminyi, Hill Nubian) [drb]: Threatened, 1,000 speakers; Nubian; spoken in Sudan. Dama [dmm]: Critically endangered, 50 speakers; Niger-Congo; spoken in Cameroon. Damakawa (Tidama’un) [dam]: Critically endangered; Niger-Congo; spoken in Nigeria. Dangaleat (Dangla, Danal, Dangal) [daa]: Vulnerable, 45,000 speakers; Afro-Asiatic; spoken in Chad. Dass (Barawa, Dot, Zodi) [dot]: Threatened, 8,830 speakers; Afro-Asiatic; spoken in Nigeria. Defaka (Afakani) [afn]: Severely endangered, fewer than 200 speakers; Niger-Congo; spoken in Nigeria. Deg (Degha, Dɛ, Mo) [mzw]: Vulnerable, 27,500 speakers; Niger-Congo; spoken in Ghana and Côte d’Ivoire. Denya (Anyang, Agnang, Anyan) [anv]: Vulnerable, at least 11,200 speakers; Niger-Congo; spoken in Cameroon. Dewoin (De, Dey, Dei) [dee]: Threatened, 8,100 speakers; Niger-Congo; spoken in Liberia. Dhaiso (Kidhaiso, Daiso, Daisu) [dhs]: Threatened, 5,000 speakers; Niger-Congo; spoken in Tanzania. Dibole (Babole, Southern Bomitaba) [bvx]: Threatened, 3,000–4,000 speakers; Niger-Congo; spoken in Congo. Dilling (Delen, Warki, Warkimbe) [dil]: Threatened, 5,300 speakers; Nubian; spoken in Sudan. Dimbong (Bumbong, Kalong, Kaalong) [dii]: Threatened, fewer than 150 speakers; NigerCongo; spoken in Cameroon. Dimé (Dime, Dima, Dimme) [dim]: Threatened, approximately 5,000 speakers; Afro-Asiatic; spoken in Ethiopia. Dir (Dììr, Dra, Baram Dutse) [no ISO 639-3 code]: Endangered, approximately 500 speakers; Afro-Asiatic; spoken in Nigeria. Doe (Dohe, Kidoe) [doe]: Vulnerable, approximately 24,000 speakers; Niger-Congo; spoken in Tanzania. Dogul Dom (Dogon, Dogul-Dom, Dogulu) [dbg]: Vulnerable, 15,700 speakers; Niger-Congo; spoken in Mali. Doka [dbi]: Vulnerable, 11,400 speakers; Niger-Congo; spoken in Nigeria. Doko-Uyanga (Uyanga, Dosanga, Basanga) [uya]: Endangered, 200 speakers; Niger-Congo; spoken in Nigeria. Dombe [dov]: Threatened, 5,430 speakers; Niger-Congo; spoken in Zimbabwe. 93

Anna Belew et al.

Dompo (Guang, Dumpo, Ndmpo) [doy]: Severely endangered, 60–70 speakers; Niger-Congo; spoken in Ghana. Dong (Donga, Dɔ) [doh]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Dulbu [dbo]: Severely endangered, 80 speakers; Niger-Congo; spoken in Nigeria. Dungu (Dungi, Dingi, Dwingi) [dbv]: Threatened, 1,100 speakers; Niger-Congo; spoken in Nigeria. Dwang (Guang, Dwan, Nchumunu) [nnu]: Threatened, 8,200 speakers; Niger-Congo; spoken in Ghana. Dyarim (Dyàrìm Tə, Kaiwari, Kaiyorawa) [no ISO 639-3 code]: Severely endangered, approximately 100 speakers; Afro-Asiatic; spoken in Nigeria. Dzando [dzn]: Threatened, 6,000 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Ebughu (Oron) [ebg]: Threatened, at least 5,000 speakers; Niger-Congo; spoken in Nigeria. Effutu (Efutu) [no ISO 639-3 code]: Threatened, fewer than 20,000 speakers; Niger-Congo; spoken in Ghana. Ega (Diés, Egwa) [ega]: Severely endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Côte d’Ivoire. Ejamat (Ediamat, Fulup, Feloup) [eja]: Vulnerable, 25,000–50,000 speakers; Niger-Congo; spoken in Senegal and Guinea-Bissau. Ekit (Eket, Ekid) [eke]: Vulnerable, approximately 200,000 speakers; Niger-Congo; spoken in Nigeria. El Hugeirat (El Hagarat, Hill Nubian) [elh]: Severely endangered, 50 speakers; Nubian; spoken in Sudan. El Molo (Elmolo, Fura-Pawa, Ldes) [elo]: Dormant; Afro-Asiatic; formerly spoken in Kenya. Eman (Emane) [emn]: Endangered, 800 speakers; Niger-Congo; spoken in Cameroon. Enwan (Enwang) [enw]: Vulnerable, 15,000 speakers; Niger-Congo; spoken in Nigeria. Eruwa (Erohwa, Erakwa, Arokwa) [erh]: Vulnerable, 64,000 speakers; Niger-Congo; spoken in Nigeria. Etebi [etb]: Vulnerable, 15,000 speakers; Niger-Congo; spoken in Nigeria. Evant (Evand, Avand, Avande) [bzz]: Vulnerable, 11,000 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Fadashi (Fedashe) [no ISO 639-3 code]: Threatened, 8,715 speakers; Berta; spoken in Ethiopia. Fali of Baissa [fah]: Critically endangered, a few speakers; Niger-Congo; spoken in Nigeria. Fam [fam]: Endangered, fewer than 500 speakers; Niger-Congo; spoken in Nigeria. Fang [fak]: Threatened, 4,000–6,000 speakers; Niger-Congo; spoken in Cameroon. Fania (Fagnia, Fanya, Fanyan) [fni]: Threatened, 1,100 speakers; Niger-Congo; spoken in Chad. Figuig (Figuig Berber) [no ISO 639-3 code]: Threatened, 14,280 speakers; Afro-Asiatic; spoken in Morocco. Firan (Fəran, Frisian, Faran) [fir]: Vulnerable, approximately 3,000 speakers; Niger-Congo; spoken in Nigeria. Fongoro (Gele, Kole) [fgr]: Critically endangered, a few speakers; Central Sudanic; spoken in Chad. Fungom (Northern Fungom, We) [no ISO 639-3 code]: Threatened, 1,000 speakers; NigerCongo; spoken in Cameroon. Fyem (Fyam, Pyem, Paiem) [pym]: Endangered, fewer than 14,000 speakers; Niger-Congo; spoken in Nigeria. Gaa (Tiba, Gà) [ttb]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Gadang [gdk]: Threatened, 2,500 speakers; Afro-Asiatic; spoken in Chad. 94

The world’s endangered languages

Ganjulé (Ganjule, Ganjawle) [no ISO 639-3 code]: Severely endangered, fewer than 50 speakers; Afro-Asiatic; spoken in Ethiopia. Ganza (Ganzo, Koma) [gza]: Threatened, 5,400 speakers; Afro-Asiatic; spoken in Ethiopia. Ganzi [gnz]: Threatened, 1,400 speakers; Niger-Congo; spoken in Central African Republic. Garre (Af-Garre) [gex]: Vulnerable, 57,500 speakers; Afro-Asiatic; spoken in Somalia. Gats’amé (Kachama-Ganjule, Gats’ame, Get’eme) [kcx]: Threatened, approximately 1,000 speakers; Afro-Asiatic; spoken in Ethiopia. Gbanziri (Gbanzili-’Bolaka, Gbanzili, Banziri) [gbg]: Vulnerable, 17,500 speakers; NigerCongo; spoken in Central African Republic and Democratic Republic of the Congo. Gbaya (Kresh, Kreish, Kreich) [krs]: Vulnerable; Kresh-Aja; spoken in Sudan, Central African Republic, and South Sudan. Gbayi (Kpasiya) [gyg]: Threatened, approximately 5,000 speakers; Niger-Congo; spoken in Central African Republic. Gbii (Gbi-Dowlu, Gbee) [ggb]: Threatened, 5,600 speakers; Niger-Congo; spoken in Liberia. Geji (Gejawa, Kayauri) [gji]: Threatened, 1,000 speakers; Afro-Asiatic; spoken in Nigeria. Geme (Jeme, Ngba Geme, Gueme) [geq]: Endangered, 550 speakers; Niger-Congo; spoken in Central African Republic. Geruma (Gerumawa, Gerema, Germa) [gea]: Threatened, 4,700 speakers; Afro-Asiatic; spoken in Nigeria. Geviya (Eviya, Gevia, Avia) [gev]: Severely endangered, fewer than 400 speakers; Niger-Congo; spoken in Gabon. Ghadamès [gha]: Threatened, 4,000 speakers; Afro-Asiatic; spoken in Tunisia and Libya. Ghomara (Ghumāra, Ghmara, Ghomara Berber) [gho]: Threatened, approximately 10,000 speakers; Afro-Asiatic; spoken in Morocco. Gibanawa (Gembanawa, Gimbanawa, Jega) [gib]: Unknown vitality; pidgin or creole; spoken in Nigeria. Goundo [goy]: Critically endangered, 20–30 speakers; Niger-Congo; spoken in Chad. Gundi (Ngundi, Ngondi) [gdi]: Threatened, 9,000 speakers; Niger-Congo; spoken in Central African Republic. Gurmana [gvm]: Threatened, at least 3,000 speakers; Niger-Congo; spoken in Nigeria. Guruntum (Guruntum-Mbaaru, Gurdung, Tala) [grd]: Vulnerable; Afro-Asiatic; spoken in Nigeria. Gusilay (Kusiilaay, Gusilaay, Gusiilay) [gsl]: Vulnerable, 10,000–25,000 speakers; NigerCongo; spoken in Senegal. Guus (Sigidi, Sugudi, Sigdi) [no ISO 639-3 code]: Threatened, 7,000 speakers; Afro-Asiatic; spoken in Nigeria. Gwamhi-Wuri (Lyase, Lyase-Ne, Gwamhyə-Wuri-Mba) [bga]: Vulnerable, 16,000 speakers; Niger-Congo; spoken in Nigeria and Benin. Gweno (Kigweno, Ghonu, Kighonu) [gwe]: Threatened, a few thousand speakers; Niger-Congo; spoken in Tanzania. Gyele (Bakola, Bagyeli, Giele) [gyi]: Endangered, 2,200–5,000 speakers; Niger-Congo; spoken in Cameroon and Equatorial Guinea. Gyem (Gyemawa, Gema, Gemawa) [gye]: Endangered, 100 speakers; Niger-Congo; spoken in Nigeria. Hadza (Hatsa, Hadzabi, Wakindiga) [hts]: Threatened, 950 speakers; isolate; spoken in Tanzania. Hamer-Bana-Kara (Hamer, Hamer-Banna, Hamar-Koke) [amf]: Vulnerable, 42,800 speakers; Afro-Asiatic; spoken in Ethiopia. 95

Anna Belew et al.

Hanga (Anga) [hag]: Threatened, 6,800 speakers; Niger-Congo; spoken in Ghana. Hasha (Yashi) [ybj]: Endangered, approximately 400 speakers; Niger-Congo; spoken in Nigeria. Hausa Sign Language (Hausa Sign Language, Maganar Hannu, Maganar Bebaye) [hsl]: Unknown vitality; sign language; spoken in Nigeria. Herero (Ochiherero, Otjiherero) [her]: Vulnerable, approximately 24,500 speakers; NigerCongo; spoken in Botswana, Namibia, and Angola. Hijuk [hij]: Endangered, 400 speakers; Niger-Congo; spoken in Cameroon. Himba (Dhimba, Dimba, Otjidhimba) [dhm]: Vulnerable, 30,000–35,000 speakers; NigerCongo; spoken in Namibia and Angola. Holoholo (Kiholoholo, Horohoro, Guha) [hoo]: Vulnerable; Niger-Congo; spoken in Botswana and Zimbabwe. Hõne (Hone, Wapan Gwana, Gaateri) [juh]: Severely endangered, approximately 7,000 speakers; Niger-Congo; spoken in Nigeria. Horom (Kaleri) [hoe]: Vulnerable, fewer than 1,500 speakers; Niger-Congo; spoken in Nigeria. Hozo (Begi-Mao) [hoz]: Threatened, 3,000 speakers; Afro-Asiatic; spoken in Ethiopia. Hungworo (Ngwoi, Nkwoi, Ngwe) [nat]: Threatened, approximately 5,000 speakers; NigerCongo; spoken in Nigeria. Hwarasa (Qwara, Qwarina) [no ISO 639-3 code]: Critically endangered, a few speakers; AfroAsiatic; spoken in Ethiopia. Hwela (Hwela-Numu, Huela, Weila) [no ISO 639-3 code]: Threatened, fewer than 1,500 speakers; Niger-Congo; spoken in Ghana. Ibani (Bonny, Ubani, Ịjọ) [iby]: Vulnerable, 60,000 speakers; Niger-Congo; spoken in Nigeria. Ibino (Ibeno, Ibuno) [ibn]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Ibuoro [ibr]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Idon (Idong) [idc]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Igo (Ahlõ, Ahlon, Achlo) [ahl]: Threatened, 6,000 speakers; Niger-Congo; spoken in Togo. Ik (Icietot, Teuso, Teuth) [ikx]: Vulnerable, approximately 16,000 speakers; Kuliak; spoken in Uganda. Ikizu (Ikikizo, Ikikizu, Kiikizu) [ikz]: At risk, 132,000 speakers; Niger-Congo; spoken in Tanzania. Ikoma (Nata, Ikinata, Ekinata) [ntk]: Vulnerable, 36,000 speakers; Niger-Congo; spoken in Tanzania. Ikorom [no ISO 639-3 code]: Threatened, 5,000–8,000 speakers; Niger-Congo; spoken in Nigeria. Ilue (Idua) [ilv]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Indri (Yanderika, Yandirika) [idr]: Severely endangered, approximately 700 speakers; NigerCongo; spoken in South Sudan. Isanzu (Kinyihanzu, Kinyisanzu) [isn]: Vulnerable, 32,400 speakers; Niger-Congo; spoken in Tanzania. Isu [isu]: Vulnerable, 10,400 speakers; Niger-Congo; spoken in Cameroon. Isubu (Su, Isu, Isuwu) [szv]: Endangered, 800 speakers; Niger-Congo; spoken in Cameroon. Itu Mbon Uzo (Itu Mbon Uso, Itu Mbuzo, Ito Mbonuso) [itm]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Iyasa (Yasa) [yko]: Threatened, 2,000–3,000 speakers; Niger-Congo; spoken in Cameroon and Equatorial Guinea. 96

The world’s endangered languages

Iyive (Uive, Yiive, Ndir) [uiv]: Endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Izon (Ijo, Izo, Uzo) [ijc]: At risk, 1,000,000 speakers; Niger-Congo; spoken in Nigeria. Izora (Chokobo, Cokobo, Cikobu) [cbo]: Endangered, 425 speakers; Niger-Congo; spoken in Nigeria. Jalkunan (Dyala, Dyalanu, Jalkuna) [bxl]: Endangered, 600 speakers; Niger-Congo; spoken in Burkina Faso. Janji (Anafejanzi, Jenji, Tijanji) [jni]: Threatened, 1,150 speakers; Niger-Congo; spoken in Nigeria. Jaya [jyy]: Threatened, 2,200 speakers; Central Sudanic; spoken in Chad. Jeri Kuo (Jeli Kuo, Celle, Jeri) [jek]: Severely endangered, approximately 1,500 speakers; Niger-Congo; spoken in Côte d’Ivoire. Jiba (Kona, Jukun Kona, Jibi) [juo]: Threatened, 1,000–2,000 speakers; Niger-Congo; spoken in Nigeria. Jilbe (Zoulbou) [jie]: Threatened, fewer than 2,000 speakers; Afro-Asiatic; spoken in Nigeria. Jimi (Cameroon) (Djimi, Jimjimen, ‘Um Falin) [jim]: Threatened, 2,500–3,500 speakers; AfroAsiatic; spoken in Cameroon. Jimi (Nigeria) (Bi-Gimu) [jmi]: Severely endangered, 1,000 speakers; Afro-Asiatic; spoken in Nigeria. Jola-Fonyi (Kújoolaak Kati Fooñi, Kujamataak, Jola-Fogny) [dyo]: At risk, 413,490 speakers; Niger-Congo; spoken in Senegal, Gambia, and Guinea-Bissau. Jonkor Bourmataguil (Djongor Bourmataguil, Dougne, Karakir) [jeu]: Threatened, fewer than 1,500 speakers; Afro-Asiatic; spoken in Chad. Ju (Sho) [juu]: Endangered, 150 speakers; Afro-Asiatic; spoken in Nigeria. Ju|’hoan (Ju/’hoan, Agau, Kung-Tsumkwe) [ktz]: Vulnerable, 11,000 speakers; Kx’a; spoken in Botswana and Namibia. Judeo-Egyptian Arabic [no ISO 639-3 code]: Unknown vitality; Afro-Asiatic; spoken in Egypt, Israel, and the United States of America. Judeo-Moroccan Arabic [aju]: At risk, 258,930 speakers; Afro-Asiatic; spoken in Morocco, Israel, and the United States of America. Judeo-Tripolitanian Arabic [yud]: Vulnerable; Afro-Asiatic; spoken in Libya, Israel, and the United States of America. Judeo-Tunisian Arabic [ajt]: At risk, 352,500 speakers; Afro-Asiatic; spoken in Tunisia, Israel, and the United States of America. K’emant (Qimant, Kimanteney, Western Agaw) [ahg]: Threatened, 1,500 speakers; AfroAsiatic; spoken in Ethiopia and Eritrea. Kabalai (Kaba-Lai, Kabalay, Kabalaye) [kvf]: Vulnerable, fewer than 17,885 speakers; AfroAsiatic; spoken in Chad. Kabwa [cwa]: Threatened, 8,500 speakers; Niger-Congo; spoken in Tanzania. Kadaru (Kadaro, Kadero, Kaderu) [kdu]: Threatened, 7,000 speakers; Nubian; spoken in Sudan. Kagoro (Kakolo) [xkg]: Vulnerable, 15,000 speakers; Niger-Congo; spoken in Mali. Kaivi (Kaibi) [kce]: Threatened, 2,320 speakers; Niger-Congo; spoken in Nigeria. Kalabari (Ịjọ) [ijn]: At risk, 258,000 speakers; Niger-Congo; spoken in Nigeria. Kalamsé (Kalemsé, Kalenga, Sàmòmá) [knz]: Threatened; Niger-Congo; spoken in Mali and Burkina Faso. Kamara (Mara) [jmr]: Vulnerable, 3,000 speakers; Niger-Congo; spoken in Ghana. Kamda (Kamdang) [no ISO 639-3 code]: Threatened, approximately 3,000 speakers; Kadu; spoken in Sudan. 97

Anna Belew et al.

Kami (Nigeria) [kmi]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Kami (Tanzania) (Kikami) [kcu]: Vulnerable, 16,400 speakers; Niger-Congo; spoken in Tanzania. Kande (Kanda, Okande, Okandé) [kbs]: Severely endangered, fewer than 100 speakers; NigerCongo; spoken in Gabon. Kanga (Krongo Abdalla, Kufa-Lima, Chiroro-Kursi) [kcp]: Threatened, approximately 8,000 speakers; Kadu; spoken in Sudan. Kantosi (Kantonsi, Yare, Yarsi) [xkt]: Threatened, 2,300 speakers; Niger-Congo; spoken in Ghana. Kanu (Likanu, Kaanu, Kano) [khx]: Threatened, 3,500 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Kapya [klo]: Endangered, 200 speakers; Niger-Congo; spoken in Nigeria. Kara (Fer, Dam Fer, Fertit) [kah]: Threatened, 4,800 speakers; Central Sudanic; spoken in Central African Republic. Karang (Kareng, Mbum, Mbum-East) [kzr]: Vulnerable, 18,000 speakers; Niger-Congo; spoken in Cameroon and Chad. Karanga (Kurunga) [kth]: Threatened, 10,000 speakers; Maban; spoken in Chad. Kari (Kare, Li-Kari-Li) [kbj]: Endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Kariya (Kariyu, Kauyawa, Lipkawa) [kil]: Threatened, 3,000 speakers; Afro-Asiatic; spoken in Nigeria. Karko (Garko, Kithonirishe) [kko]: Endangered, fewer than 8,000 speakers; Nubian; spoken in Sudan. Karo (Kerre, Cherre, Kere) [kxh]: Threatened, 1,000 speakers; Afro-Asiatic; spoken in Ethiopia. Kasanga (Cassanga, Kassanga, I-Hadja) [ccj]: Critically endangered, a few speakers; NigerCongo; spoken in Guinea-Bissau. Katla (Akalak, Kalak) [kcr]: Vulnerable, 14,200 speakers; Niger-Congo; spoken in Sudan. Keiga (Yega, Keiga-Timero, Keiga-Al-Kheil) [kec]: Threatened, 6,070 speakers; Kadu; spoken in Sudan. Kelo (Tornasi, Kelo-Beni Sheko, Ndu-Faa-Keelo) [xel]: Endangered, 200 speakers; Eastern Jebel; spoken in South Sudan. Kendeje (Yaali, Kindeje) [klf]: Endangered, fewer than 2,000 speakers; Maban; spoken in Chad. Kendem (Bokwa-Kendem) [kvm]: Threatened, 1,000 speakers; Niger-Congo; spoken in Cameroon. Kenuzi (Kenuz, Kunuzi, Kenzi) [xnz]: Threatened, 50,000 speakers; Nubian; spoken in Libya and Egypt. Kgalagadi (Khalagari, Khalakadi, Kxhalaxadi) [xkv]: Vulnerable, 10,000–15,000 speakers; Niger-Congo; spoken in Botswana and Namibia. Khe (Kheso, Bambadion-Kheso) [kqg]: Threatened, 1,300 speakers; Niger-Congo; spoken in Burkina Faso and Côte d’Ivoire. Khisa (Komono, Khi Khipa, Kumwenu) [kqm]: Threatened, 8,000 speakers; Niger-Congo; spoken in Burkina Faso and Côte d’Ivoire. Khoekhoe (Khoekhoegowap, Khoekhoegowab, Khoe) [naq]: Vulnerable, 190,000–210,000 speakers; Khoe; spoken in South Africa, Botswana, and Namibia. Khwe (Kxoe, Khoe, Buga) [xuu]: Threatened, 7,000 speakers; Khoe; spoken in South Africa, Botswana, Namibia, Angola, and Zambia. Kim (Masa, Kosop, Kwasap) [kia]: Vulnerable, 15,400 speakers; Niger-Congo; spoken in Chad. 98

The world’s endangered languages

Kiong (Akoiyang, Akayon, Okonyong) [kkm]: Severely endangered, fewer than 100 speakers; Niger-Congo; spoken in Nigeria. Kirike (Okrika, Ịjọ) [okr]: At risk, 248,000 speakers; Niger-Congo; spoken in Nigeria. Kiwilwana (Ilwana) [mlk]: Threatened, 8,000 speakers; Niger-Congo; spoken in Kenya. Ko (Kau, Fungor, Fungur) [fuj]: Threatened, 2,680 speakers; Niger-Congo; spoken in Sudan. Kobiana (Cobiana, Uboi, Buy) [kcj]: Endangered, approximately 400 speakers; Niger-Congo; spoken in Senegal and Guinea-Bissau. Koenoem (Kanam, Korenoem) [kcs]: Threatened, 3,000 speakers; Afro-Asiatic; spoken in Nigeria. Kofa (Kota, Gbwata, Batta) [kso]: Vulnerable; Afro-Asiatic; spoken in Nigeria. Koke (Khoke) [kou]: Endangered, 600 speakers; Niger-Congo; spoken in Chad. Kolbila (Kolbilari, Kolbilla, Kolena) [klc]: Endangered, a few hundred speakers; NigerCongo; spoken in Cameroon. Komo (Koma of Daga, Como, Central Koma) [xom]: Threatened, 1,435 speakers; Koman; spoken in Ethiopia and South Sudan. Konni (Koma, Koni, Komung) [kma]: Vulnerable, 3,800 speakers; Niger-Congo; spoken in Ghana. Kono (Konu, Kwono, Kowono) [klk]: Threatened, 5,520 speakers; Niger-Congo; spoken in Nigeria. Korana (Kora,!Ora) [kqz]: Critically endangered, one speaker; Khoe; spoken in South Africa. Koro (Cote d’Ivoire) (Koro Jula) [kfo]: Vulnerable, 40,000 speakers; Niger-Congo; spoken in Côte d’Ivoire. Korop (Ododop, Durop, Kurop) [krp]: Vulnerable, 17,640 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Koshin (Kosin, Kaw) [kid]: Threatened, 3,000–3,500 speakers; Niger-Congo; spoken in Cameroon. Kota (Ikota, Ikuta, Kotu) [koq]: Vulnerable, 43,460 speakers; Niger-Congo; spoken in Gabon and Congo. Kpan (Eregba, Yorda, Ibukwo) [kpk]: Vulnerable, 11,400 speakers; Niger-Congo; spoken in Nigeria. Kpatili (Kpatiri, Kpatere, Ngindere) [kym]: Threatened, 4,500 speakers; Niger-Congo; spoken in Central African Republic. Kpessi (Kpesi, Kpétsi) [kef]: Vulnerable, 2,358–3,100 speakers; Niger-Congo; spoken in Togo. Krim (Kim, Kittim, Kirim) [krm]: Critically endangered, a few hundred speakers; Niger-Congo; spoken in Sierra Leone. Krongo (Korongo, Kurungu, Kadumodi) [kgo]: Vulnerable, 21,700 speakers; Kadu; spoken in Sudan. Kubi (Deno, Denawa, Denwa) [kof]: Dormant; Afro-Asiatic; formerly spoken in Nigeria. Kudu-Camo (Kudawa, Kuda-Chamo) [kov]: Severely endangered, a few speakers; NigerCongo; spoken in Nigeria. Kugbo [kes]: Threatened, 2,000 speakers; Niger-Congo; spoken in Nigeria. Kujarge (Kujarke) [vkj]: Endangered, approximately 1,000 speakers; Unclassified; spoken in Chad. Kuk [kfn]: Threatened, 3,000 speakers; Niger-Congo; spoken in Cameroon. Kulfa (Kulfe, Kurmi, Kurumi) [kxj]: Threatened, 2,200 speakers; Central Sudanic; spoken in Chad. Kulung (Nigeria) (Bambur, Kuluno, Bakulung) [bbu]: Vulnerable, 15,000 speakers; NigerCongo; spoken in Nigeria. 99

Anna Belew et al.

Kung [kfl]: Endangered, 600–800 speakers; Niger-Congo; spoken in Cameroon. Kupto (Kúttò) [kpa]: Threatened, fewer than 3,000 speakers; Afro-Asiatic; spoken in Nigeria. Kuteb (Ati, Kutep, Kutev) [kub]: At risk, at least 100,000 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Kuturmi (Ada) [khj]: Vulnerable, 10,500 speakers; Niger-Congo; spoken in Nigeria. Kuwaataay (Kwatay) [cwt]: Threatened, 6,210 speakers; Niger-Congo; spoken in Senegal. Kw’adza (Qwadza, Kwadza, Ng’omvia) [wka]: Dormant; Afro-Asiatic; formerly spoken in Tanzania. Kwama (Takwama, Gwama, Goma) [kmq]: Vulnerable; Koman; spoken in Ethiopia. Kwarandzyey (Korandje, ‫[ )ةيلابلبلا‬kcy]: Endangered, approximately 3,000 speakers; Songhay; spoken in Algeria. Kwegu (Koegu, Kwegi, Koyego) [xwg]: Endangered, approximately 300 speakers; Surmic; spoken in Ethiopia. Laal (Gori) [gdm]: Endangered, fewer than 500 speakers; isolate; spoken in Chad. Labir (Jaku, Jakun, Jakanci) [jku]: Vulnerable, 13,000 speakers; Niger-Congo; spoken in Nigeria. Lafofa [laf]: Endangered, 600 speakers; Niger-Congo; spoken in Sudan. Lagwan (Logone, Lagwan Kotoko, Kotoko-Logone) [kot]: Vulnerable, 10,000 speakers; AfroAsiatic; spoken in Nigeria, Cameroon, and Chad. Landoma (Landouman, Landuma, Tyapi) [ldm]: Vulnerable, 14,400 speakers; Niger-Congo; spoken in Guinea. Langa [no ISO 639-3 code]: Endangered, fewer than 500 speakers; Niger-Congo; spoken in Cameroon. Langas (Laŋas, Nyámzax, Lundur) [no ISO 639-3 code]: Endangered; Afro-Asiatic; spoken in Nigeria. Laro (Laru, Aaleira, Yillaro) [lro]: Vulnerable, 40,000 speakers; Niger-Congo; spoken in Sudan. Lehar (Lala, Laalaa) [cae]: Threatened, 5,000–10,000 speakers; Niger-Congo; spoken in Senegal. Lere [gnh]: Critically endangered, few speakers; Niger-Congo; spoken in Nigeria. Logba [lgq]: Threatened, 7,500 speakers; Niger-Congo; spoken in Ghana. Logol (Lukha) [lof]: Threatened, 2,600 speakers; Niger-Congo; spoken in Sudan. Logorik (Liguri) [liu]: Threatened, 2,000 speakers; Daju; spoken in Sudan. Lufu [ldq]: Threatened, 3,200 speakers; Niger-Congo; spoken in Nigeria. Lumun (Lomon, Kuku-Lumun) [lmd]: Vulnerable, 30,000 speakers; Niger-Congo; spoken in Sudan. Luo [luw]: Critically endangered, one speaker; Niger-Congo; spoken in Cameroon. Luri (Lúr) [ldd]: Critically endangered, two speakers; Afro-Asiatic; spoken in Nigeria. Lushi (Lukshi, Dokshi) [no ISO 639-3 code]: Dormant; Afro-Asiatic; formerly spoken in Nigeria. Mabaan (Maaban, Meban, Southern Burun) [mfz]: Severely endangered, 25 speakers; Nilotic; spoken in South Sudan. Mabire [muj]: Critically endangered, approximately five speakers; Afro-Asiatic; spoken in Chad. Mághdì (Tala, Widala, Maghdi) [gmd]: Threatened, 2,000 speakers; Niger-Congo; spoken in Nigeria. Maha (Maaka, Maka, Maga) [mew]: Vulnerable, 10,000 speakers; Afro-Asiatic; spoken in Nigeria. 100

The world’s endangered languages

Mahongwe [mhb]: Endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Gabon. Mahwa (Mawa, Mahoua, Mahoura) [mcw]: Threatened, 6,560 speakers; Afro-Asiatic; spoken in Chad. Mak (Panyam, Panya, Leemak) [pbl]: Threatened, 5,690 speakers; Niger-Congo; spoken in Nigeria. Makwe (Kimakwe, Palma, Macue) [ymk]: Vulnerable, 32,000 speakers; Niger-Congo; spoken in Tanzania and Mozambique. Mala (Rumaya, Rumaiya, Amala) [ruy]: Threatened, 6,630 speakers; Niger-Congo; spoken in Nigeria. Malé (Maalé) [mdy]: Vulnerable, 53,800 speakers; Afro-Asiatic; spoken in Ethiopia. Malian Sign Language (Mali Sign Language, Langue des Signes Malienne, LaSiMa) [bog]: Endangered; sign language; spoken in Mali. Malo (Melo, Mello) [mfx]: Vulnerable, 20,151 speakers; Afro-Asiatic; spoken in Ethiopia. Mambai (Mangbei, Mangbai, Manbai) [mcs]: Vulnerable, approximately 15,000 speakers; Niger-Congo; spoken in Cameroon and Chad. Manda (Kimanda, Kinyasa, Nyasa) [mgs]: Vulnerable, 22,000 speakers; Niger-Congo; spoken in Tanzania. Mangas [zns]: Endangered, 180 speakers; Afro-Asiatic; spoken in Nigeria. Mangayat (Mangaya, Mongaiyat, Bug) [myj]: Endangered, 400 speakers; Niger-Congo; spoken in South Sudan. Mani (Bullom So, Northern Bullom, Bolom) [buy]: Severely endangered, fewer than 200 speakers; Niger-Congo; spoken in Guinea and Sierra Leone. Mansoanka (Sua, Mansoanca, Maswanka) [msw]: Vulnerable, 10,000–25,000 speakers; NigerCongo; spoken in Gambia and Guinea-Bissau. Mao of Bambeshi (Bambassi, Bambeshi, Siggoyo) [myf]: Threatened, 5,000 speakers; AfroAsiatic; spoken in Ethiopia. Maore Comorian (Comorian, Comores Swahili, Komoro) [swb]: Threatened; Niger-Congo; spoken in Comoros, Mayotte, Réunion, and Madagascar. Maslam (Mandage, Mendage, Mandagué) [msv]: Threatened, 5,000–6,000 speakers; AfroAsiatic; spoken in Cameroon and Chad. Massalat [mdg]: Severely endangered, ten speakers; Maban; spoken in Chad. Mbara (Massa de Guelengdeng, Guelengdeng, G’kelendeng) [mpk]: Endangered, fewer than 1,000 speakers; Afro-Asiatic; spoken in Chad. Mbonga (Mboa, Mboŋa) [xmb]: Threatened, 1,490 speakers; Niger-Congo; spoken in Cameroon. Mbongno (Bungnu, Bunu, Bungun) [bgu]: Threatened; Niger-Congo; spoken in Nigeria and Cameroon. Mbowe (Esimbowe) [mxo]: Threatened, 2,690 speakers; Niger-Congo; spoken in Zambia. Mbre (Bre, Bere, Pre) [mka]: Endangered, 200 speakers; Niger-Congo; spoken in Côte d’Ivoire. Mbugu (Ma’a, Mbougou, Wama’a) [mhd]: Threatened, approximately 7,000 speakers; mixed language; spoken in Tanzania. Mbulungish (Baga Foré, Baga Monson, Monshon) [mbv]: Threatened, 5,000 speakers; NigerCongo; spoken in Guinea and Guinea-Bissau. Ménik (Budik, Bedik, Tandanke) [tnr]: Threatened; Niger-Congo; spoken in Senegal. Menka (Wando Bando, Mamwoh) [mea]: Vulnerable, 5,200 speakers; Niger-Congo; spoken in Cameroon. Mesmes (Mäsmäs) [mys]: Dormant; Afro-Asiatic; formerly spoken in Ethiopia. 101

Anna Belew et al.

Midah (Majera, Mida’a, Da’a) [xmj]: Threatened, 1,000 speakers; Afro-Asiatic; spoken in Cameroon and Chad. Miltu (Miltou) [mlj]: Endangered, fewer than 300 speakers; Afro-Asiatic; spoken in Chad. Mingang Doso (Munga Doso, Ngwai Mungàn, Doso) [mko]: Threatened, 3,000 speakers; NigerCongo; spoken in Nigeria. Miyobe (Soruba, Mi yɔbɛ, Bijobe) [soy]: Threatened, 8,700 speakers; Niger-Congo; spoken in Togo and Benin. Mlomp (Gulompaay) [mlo]: Threatened, 5,400 speakers; Niger-Congo; spoken in Senegal. Mo’da (Gberi, Gweri, Gbara) [gbn]: Endangered, 600 speakers; Central Sudanic; spoken in South Sudan. Molo (Malkan, Tura-Ka-Molo) [zmo]: Endangered, approximately 100 speakers; Eastern Jebel; spoken in Sudan. Mombo (Kolu So, Kolu, Kolum So) [dmb]: Vulnerable, 19,000 speakers; Niger-Congo; spoken in Mali. Mono (Mon-Non) [mru]: Severely endangered, fewer than 300 speakers; Niger-Congo; spoken in Cameroon. Mono (Amono) [mnh]: Vulnerable, 62,102 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Monzombo (Monjombo, Mondjembo, Munzombo) [moj]: Vulnerable, 12,600 speakers; NigerCongo; spoken in Congo and Central African Republic. Moro (Dhimorong) [mor]: Vulnerable, 30,000 speakers; Niger-Congo; spoken in Sudan. Morokodo (Ma’di) [mgc]: Threatened, 3,400 speakers; Central Sudanic; spoken in South Sudan. Mpra (Mpre) [no ISO 639-3 code]: Dormant; Unclassified; formerly spoken in Ghana. Mser (Kotoko-Kuseri, Kuseri, Kouseri) [kqx]: Severely endangered, 500 speakers; AfroAsiatic; spoken in Cameroon and Chad. Mundabli-Mufu (Ji, Bu) [boe]: Endangered, 430–600 speakers; Niger-Congo; spoken in Cameroon. Mungbam (Mijong, Missong, Abar) [mij]: Threatened, 1,850–2,150 speakers; Niger-Congo; spoken in Cameroon. Murle (Murelei, Merule, Mourle) [mur]: Vulnerable, 60,200 speakers; Surmic; spoken in Ethiopia and South Sudan. Mursi (Murzi, Murzu, Merdu) [muz]: Threatened, 3,278 speakers; Surmic; spoken in Ethiopia. Mushungulu (Kimushungulu, Mushunguli) [xma]: Vulnerable, 23,000 speakers; Niger-Congo; spoken in Somalia. Mvanip (Mvanöp, Mvanon, Mvanlip) [mcj]: Threatened, 100 speakers; Niger-Congo; spoken in Nigeria. Mwenyi [no ISO 639-3 code]: Threatened, 6,000 speakers; Niger-Congo; spoken in Zambia. Mwini (Mwiini, Chimwiini, Af-Chimwiini) [no ISO 639-3 code]: Endangered; Niger-Congo; spoken in Tanzania. N||ng (N|uu, N|u, N/u) [ngh]: Critically endangered, five speakers; Tuu; spoken in South Africa. Naki (Mekaf, Munkaf, Nkap) [mff]: Threatened, 3,000 speakers; Niger-Congo; spoken in Cameroon. Nalu (Nalou) [naj]: Vulnerable, fewer than 22,000 speakers; Niger-Congo; spoken in Guinea and Guinea-Bissau. Nambya (Chinambya, Nanzva, Nambzya) [nmq]: Threatened; Niger-Congo; spoken in Botswana and Zimbabwe. Nanga (Nanga Dama Dogon, Nanga Dogon) [nzz]: Threatened, 3,000 speakers; Niger-Congo; spoken in Mali. 102

The world’s endangered languages

Narim (Larim, Larimo, Lariim) [loh]: Threatened, 3,620 speakers; Surmic; spoken in South Sudan. Naro (Nharo, Naron, /Ai San) [nhr]: Vulnerable, approximately 10,000 speakers; Khoe; spoken in Botswana and Namibia. Natioro (Koo’ra, Natyoro, Natjoro) [nti]: Vulnerable, approximately 4,000 speakers; NigerCongo; spoken in Burkina Faso. Nayi (Naya, Na’o, Nao) [noz]: Severely endangered, 7,188 speakers; Afro-Asiatic; spoken in Ethiopia. Nchumbulu (Guang) [nlu]: Threatened, 1,800 speakers; Niger-Congo; spoken in Ghana. Ndai (Galke, Pormi) [gke]: Critically endangered, a few speakers; Niger-Congo; spoken in Cameroon. Ndam (Dam, Ndamm, Gulei) [ndm]: Threatened, 6,500 speakers; Afro-Asiatic; spoken in Chad. Nding (Eliri) [eli]: Threatened, 3,513 speakers; Niger-Congo; spoken in Sudan. Ndoola (Ndola, Ndoro, Ndooro) [ndr]: Vulnerable, at least 10,000 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Ndunda [nuh]: Endangered, 300–400 speakers; Niger-Congo; spoken in Nigeria. Ndut (Ndoute) [ndv]: Vulnerable, 25,000–50,000 speakers; Niger-Congo; spoken in Senegal. Ngamo (Ngamawa, Gamo, Gamawa) [nbh]: Vulnerable, approximately 60,000 speakers; AfroAsiatic; spoken in Nigeria. Ngasa (Shaka, Ongamo) [nsg]: Endangered, 200–300 speakers; Nilotic; spoken in Kenya and Tanzania. Ngbinda (Bungbinda, Bangbinda) [nbd]: Critically endangered, few speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Nggwahyi (Ngwaxi, Ngwohi) [ngx]: Threatened, 2,000 speakers; Afro-Asiatic; spoken in Nigeria. Ngile (Masakin, Mesakin, Daloka) [jle]: Vulnerable, 38,000 speakers; Niger-Congo; spoken in Sudan. Ngizim (Ngizmawa, Ngezzim, Ngódṣin) [ngi]: Vulnerable, 80,000 speakers; Afro-Asiatic; spoken in Nigeria. Ngombe (Ngombe-Kaka, Bagando-Ngombe, Bangando-Ngombe) [nmj]: Threatened, 1,400 speakers; Niger-Congo; spoken in Central African Republic. Ngong (Gong, Puuri, Nagumi) [nnx]: Critically endangered, two speakers; Niger-Congo; spoken in Cameroon. Ngwaba (Gombi, Goba) [ngw]: Vulnerable, 10,000 speakers; Afro-Asiatic; spoken in Nigeria. Niellim (Lua, Nielim, Mjillem) [nie]: Threatened, 5,160 speakers; Niger-Congo; spoken in Chad. Nimbari (Niamniam, Nyamnyam, Bari) [nmr]: Dormant; Niger-Congo; formerly spoken in Cameroon. Ningye [nns]: Threatened, fewer than 4,000 speakers; Niger-Congo; spoken in Nigeria. Ninzo (Ninzam, Nunzo) [nin]: Critically endangered, two speakers; Niger-Congo; spoken in Nigeria. Njalgulgule (Nyolge, Nyoolne, Ngulgule) [njl]: Endangered, 900 speakers; Daju; spoken in South Sudan. Njerep (Njerup) [njr]: Critically endangered, four speakers; Niger-Congo; spoken in Nigeria. Nkari [nkz]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Nkoroo (Kirika, Nkoro, Ịjọ) [nkx]: Threatened, fewer than 5,000 speakers; Niger-Congo; spoken in Nigeria. Noon (None, Non, Serer-Noon) [snf]: Vulnerable, 25,000–50,000 speakers; Niger-Congo; spoken in Senegal. 103

Anna Belew et al.

Notre (Nõtre, Bulba, Nootre) [bly]: Vulnerable, 2,368 speakers; Niger-Congo; spoken in Benin. Noy (Loo) [noy]: Severely endangered, 36 speakers; Niger-Congo; spoken in Chad. Nubaca (Bango, Bongo, Baca) [baf]: Endangered, 800 speakers; Niger-Congo; spoken in Cameroon. Nubi (Kinúbi) [kcn]: Threatened, approximately 25,000 speakers; pidgin or creole; spoken in Kenya, South Sudan, Democratic Republic of the Congo, and Uganda. Nyamusa-Molo [nwm]: Threatened, 1,200 speakers; Central Sudanic; spoken in South Sudan. Nyang’i (Nuangeya, Nyuangia, Nyangiya) [nyp]: Critically endangered, fewer than ten speakers; Kuliak; spoken in Uganda. Nyanjang (Njang, Njanga) [no ISO 639-3 code]: Critically endangered, four speakers; NigerCongo; spoken in Cameroon. Oblo [obl]: Dormant; Niger-Congo; formerly spoken in Cameroon. Obulom (Abuloma, Obulum) [obu]: Threatened, 3,420 speakers; Niger-Congo; spoken in Nigeria. Odut [oda]: Severely endangered, 20 speakers; Niger-Congo; spoken in Nigeria. Okiek (Akiek, Akie, Ogiek) [oki]: Severely endangered, 50 speakers; Nilotic; spoken in Kenya and Tanzania. Okobo [okb]: Vulnerable, 50,000 speakers; Niger-Congo; spoken in Nigeria. Okodia (Okordia, Akita, Ịjọ) [okd]: Threatened, 3,600 speakers; Niger-Congo; spoken in Nigeria. Omotik [omt]: Severely endangered, 50 speakers; Nilotic; spoken in Kenya. Ongota (Birale, ‘Ongota, Birelle) [bxe]: Critically endangered, eight speakers; Unclassified; spoken in Ethiopia. Opuuo (Opo-Shita, Opo, Opuo) [lgn]: Endangered, 301 speakers; Koman; spoken in Ethiopia and South Sudan. Oruma (Ịjọ) [orr]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Otoro (Utoro, Dhitoro, Litoro) [otr]: Endangered, 10,000 speakers; Niger-Congo; spoken in Sudan. Ouargli (Tagargrent, Teggargrent, Ouargla) [oua]: Threatened, 5,000 speakers; Afro-Asiatic; spoken in Algeria. Oued-Righ Berber (Temacine Tamazight, Righ, Tougourt) [tjo]: Threatened, 6,000 speakers; Afro-Asiatic; spoken in Algeria. Oyda (Oida) [oyd]: Vulnerable, 15,000 speakers; Afro-Asiatic; spoken in Ethiopia. Pa’a (Pa’anci, Afa, Foni) [pqa]: Threatened, 8,000 speakers; Afro-Asiatic; spoken in Nigeria. Paloor (Falor, Palar, Sili) [fap]: Threatened, 5,000–10,000 speakers; Niger-Congo; spoken in Senegal. Pam [pmn]: Severely endangered, 30 speakers; Niger-Congo; spoken in Cameroon. Pana (Sama) [pnq]: Threatened, 7,800 speakers; Niger-Congo; spoken in Burkina Faso. Pana (Pani) [pnz]: Vulnerable, 85,980 speakers; Niger-Congo; spoken in Nigeria, Cameroon, Chad, and Central African Republic. Pande (Ipande) [bkj]: Threatened, 8,870 speakers; Niger-Congo; spoken in Central African Republic. Penange [no ISO 639-3 code]: Unknown vitality; Niger-Congo; spoken in Mali. Pimbwe (Ichipimbwe, Cipimbwe, Kipimbwe) [piw]: Vulnerable, 29,000 speakers; Niger-Congo; spoken in Tanzania. Pinji (Gapinji, Apinji, Apindji) [pic]: Threatened, 5,000 speakers; Niger-Congo; spoken in Gabon. Piti (Pitti, Abisi, Bisi) [pcn]: Threatened, 5,530 speakers; Niger-Congo; spoken in Nigeria. 104

The world’s endangered languages

Plapo Krumen (Plapo, Krumen, Plapo) [ktj]: Severely endangered, 100 speakers; Niger-Congo; spoken in Côte d’Ivoire. Polci (Posə, Pəlci, Palchi) [plj]: Vulnerable, 50,000 speakers; Afro-Asiatic; spoken in Nigeria. Pongu (Pongo, Pangu, Arringeu) [png]: Vulnerable, at least 20,000 speakers; Niger-Congo; spoken in Nigeria and Burkina Faso. Putai (Marghi West, West Marg) [mfl]: Severely endangered, 50 speakers; Afro-Asiatic; spoken in Nigeria. Reel (Atuot, Atwot, Thok Cieng Reel) [atu]: Vulnerable, 50,000 speakers; Nilotic; spoken in Sudan. Reshe (Tsureshe, Tsureja, Bareshe) [res]: Vulnerable, fewer than 44,000 speakers; Niger-Congo; spoken in Nigeria. Rukul [no ISO 639-3 code]: Endangered, 500–1,000 speakers; Niger-Congo; spoken in Nigeria. Ruma (Ruruma, Rurama, Turama) [ruz]: Threatened, 5,090 speakers; Niger-Congo; spoken in Nigeria. Saba (Jelkung) [saa]: Threatened, 1,340 speakers; Afro-Asiatic; spoken in Chad. Sagalla (Kisagala, Kisagalla, Sagala) [tga]: At risk, 100,000 speakers; Niger-Congo; spoken in Kenya. Sake (Asake, Shake) [sak]: Threatened, 8,000 speakers; Niger-Congo; spoken in Gabon. Sambe [xab]: Critically endangered, two speakers; Niger-Congo; spoken in Nigeria. Sandawe (Sandaui, Sandaweeki, Sandwe) [sad]: Vulnerable, 60,000 speakers; isolate; spoken in Tanzania. Sanye (Waata, Sanya, Wasanye) [ssn]: Vulnerable, 17,400 speakers; Afro-Asiatic; spoken in Kenya. Sarwa (Sarua, Saroua) [swy]: Threatened, 2,000 speakers; Afro-Asiatic; spoken in Chad. Sawknah (Sokna, Sawkna) [swn]: Dormant; Afro-Asiatic; formerly spoken in Libya. Segeju (Kisegeju, Sageju, Sengeju) [seg]: Threatened, 7,000 speakers; Niger-Congo; spoken in Tanzania. Seki (Sekyani, Sekiani, Sekiyani) [syi]: Vulnerable, 14,690 speakers; Niger-Congo; spoken in Gabon and Equatorial Guinea. Sekpele (Likpe, Mu, Bosele) [lip]: Vulnerable, 23,400 speakers; Niger-Congo; spoken in Ghana. Selee (Santrokofi, Sentrokofi, Bale) [snw]: Threatened, 6,000 speakers; Niger-Congo; spoken in Ghana. Sengwer (Cherengany, Mei) [no ISO 639-3 code]: Threatened, fewer than 6,000 speakers; Nilotic; spoken in Kenya. Senhaja de Srair (Sanhãja of Srair, Senhajiya, Sanhaja de Srair) [sjs]: Vulnerable, 50,000 speakers; Afro-Asiatic; spoken in Morocco. Settla (Kisettla, Kisetla) [sta]: Unknown vitality; pidgin or creole; spoken in Zambia. Sezo (Seze) [sze]: Threatened, 3,000 speakers; Afro-Asiatic; spoken in Ethiopia. Sha [scw]: Threatened, 3,000 speakers; Afro-Asiatic; spoken in Nigeria. Shabo (Chabu, Shabu, Sabu) [sbf]: Severely endangered, fewer than 600 speakers; Unclassified; spoken in Ethiopia. Shanga (Shangawa, Shonga, Shongawa) [sho]: Threatened, 5,000 speakers; Niger-Congo; spoken in Nigeria. Shatt (Caning) [shj]: Vulnerable, 30,000 speakers; Daju; spoken in Sudan. Shau (Sho, Lìsháù) [sqh]: Unknown vitality; Niger-Congo; spoken in Nigeria. Shé (She, Sce, Kaba) [no ISO 639-3 code]: Vulnerable, 13,116 speakers; Afro-Asiatic; spoken in Ethiopia. 105

Anna Belew et al.

Sheko (Shekko, Shekka, Tschako) [she]: Vulnerable; Afro-Asiatic; spoken in Ethiopia. Shen (Laru, Larawa, Laranchi) [lan]: Threatened, approximately 1,000 speakers; Niger-Congo; spoken in Nigeria. Sheni (Shenanci, Seni, Shani) [scv]: Critically endangered, six speakers; Niger-Congo; spoken in Nigeria. Shiki (Gubi, Guba, Gubawa) [gua]: Endangered, fewer than 1,000 speakers; Niger-Congo; spoken in Nigeria. Shirawa (Shira) [no ISO 639-3 code]: Unknown vitality; Afro-Asiatic; spoken in Nigeria. Shua (Mashuakwe, Tshumakwe, Shua-Khwe) [shg]: Endangered, 4,100 speakers; Khoe; spoken in Botswana. Shwai (Shirumba, Shuway, Ludumor) [shw]: Threatened, 3,500 speakers; Niger-Congo; spoken in Sudan. Sichela [no ISO 639-3 code]: Vulnerable, fewer than 20,000 speakers; Niger-Congo; spoken in Tanzania. Sighu (Lesighu, Mississiou) [sxe]: Endangered, several hundred speakers; Niger-Congo; spoken in Gabon. Simba (Nsindak) [sbw]: Threatened, 3,000 speakers; Niger-Congo; spoken in Gabon. Sininkere (Silinkere, Silanke) [skq]: Threatened, 6,000 speakers; Niger-Congo; spoken in Burkina Faso. Siphuthi (Seputhi, Sephuthi, Phuthi) [no ISO 639-3 code]: Vulnerable, approximately 20,000 speakers; Niger-Congo; spoken in South Africa and Lesotho. Siri (Sirawa) [sir]: Endangered, 3,800 speakers; Afro-Asiatic; spoken in Nigeria. Siwi (Siwi, Siwa, Sioua) [siz]: Vulnerable, 10,000 speakers; Afro-Asiatic; spoken in Egypt. Sokoro (Bedanga) [sok]: Threatened, 5,000 speakers; Afro-Asiatic; spoken in Chad. Somyev (Somyewe, Somyɛwɛ, Kila) [kgt]: Critically endangered, fewer than 20 speakers; NigerCongo; spoken in Nigeria and Cameroon. Soo (So, Tepeth, Tepes) [teu]: Severely endangered, approximately 100 speakers; Kuliak; spoken in Uganda. Southeast Ijo (Ijaw, Brass Ijo, Ịjọ) [ijs]: Vulnerable, 71,500 speakers; Niger-Congo; spoken in Nigeria. Suba (Olusuba) [sxb]: At risk, 174,000 speakers; Niger-Congo; spoken in Kenya and Tanzania. Subiya (Echisubia, Subia, Supia) [sbs]: Vulnerable, 24,500 speakers; Niger-Congo; spoken in Botswana, Namibia, and Zambia. Sukur (Sakun, Sugur, Adikimmu Sukur) [syk]: Vulnerable, approximately 15,000 speakers; Afro-Asiatic; spoken in Nigeria. Sur (Tapshin, Tapshinawa, Suru) [tdl]: Vulnerable, fewer than 4,000 speakers; Niger-Congo; spoken in Nigeria. Suri (Surma, Shuri, Churi) [suq]: Vulnerable, approximately 28,000 speakers; Surmic; spoken in Ethiopia and South Sudan. Surubu (Srubu, Fiti, Skrubu) [sde]: Threatened, 7,170 speakers; Niger-Congo; spoken in Nigeria. Taa (!Xóõ, Tâa, !Xuun) [nmn]: Threatened, 2,600 speakers; Tuu; spoken in Botswana and Namibia. Tagdal [tda]: Vulnerable, 10,000–20,000 speakers; mixed language; spoken in Niger. Tagoi (Tumale, Tagoy, Umāle) [tag]: Vulnerable, 13,000 speakers; Niger-Congo; spoken in Sudan. Tahaggart Tamahaq (Tamachek, Tamashekin) [thv]: Vulnerable, 62,000 speakers; AfroAsiatic; spoken in Algeria, Niger, and Libya. 106

The world’s endangered languages

Tala [tak]: Threatened, 1,000 speakers; Afro-Asiatic; spoken in Nigeria. Talinga-Bwisi (Talinga, Kitalinga, Lubwisi) [tlj]: Vulnerable, 75,000–85,000 speakers; NigerCongo; spoken in Democratic Republic of the Congo and Uganda. Talodi (Jomang, Gajomang, Ajomang) [tlo]: Threatened, 1,500 speakers; Niger-Congo; spoken in Sudan. Tamki (Temki) [tax]: Endangered, 500 speakers; Afro-Asiatic; spoken in Chad. Tasawaq (Ingelshi, Tasawa) [twq]: Threatened, 2,000–10,000 speakers; Songhay; spoken in Niger. Taznatit (Zenati, Zenatia, Znatiya) [grr]: Vulnerable, 40,000 speakers; Afro-Asiatic; spoken in Algeria. Tchitchege [tck]: Threatened, 2,000 speakers; Niger-Congo; spoken in Gabon. Tchumbuli (Basa, Tshummbuli, Chombulon) [bqa]: Severely endangered, 1,838 speakers; Niger-Congo; spoken in Benin. Tebul Sign Language [tsy]: Endangered, 500 speakers; sign language; spoken in Mali. Tebul Ure (Oru Yille, Tew Tegu) [dtu]: Vulnerable, 3,000–4,000 speakers; Niger-Congo; spoken in Mali. Tedaga (Teda, Toda, Todaga) [tuq]: Vulnerable, 42,500 speakers; Saharan; spoken in Nigeria, Niger, Chad, and Libya. Tegali (Orig, Tagale, Tegele) [ras]: Vulnerable, 35,700 speakers; Niger-Congo; spoken in Sudan. Temein (Temainian, Rone, Ronge) [teq]: Vulnerable, 13,000 speakers; Temein; spoken in Sudan. Tennet (Tenet) [tex]: Threatened, 4,000–10,000 speakers; Surmic; spoken in South Sudan. Tese (Teis-Umm-Danab, Keiga Jirru, Keiga Girru) [keg]: Threatened, 1,400 speakers; Temein; spoken in Sudan. Tetserret [no ISO 639-3 code]: Severely endangered, fewer than 2,000 speakers; Afro-Asiatic; spoken in Niger. Tha (Joole Manga, Kapawa, Joole Marga) [thy]: Threatened, 1,000 speakers; Niger-Congo; spoken in Nigeria. Tidikelt Tamazight (Tidikelt, Tamazight, Tidikelt Berber) [tia]: Threatened, 9,000 speakers; Afro-Asiatic; spoken in Algeria. Tiéfo (Foro, Tyefo, Tyeforo) [tiq]: Endangered, approximately 1,000 speakers; Niger-Congo; spoken in Burkina Faso. Tima (Lomorik, Lomuriki, Tamanik) [tms]: Threatened, fewer than 5,000 speakers; Niger-Congo; spoken in Sudan. Tiranige Diga (Tiranige, Duleri, Duleri Dom) [tde]: Threatened, 4,200 speakers; Niger-Congo; spoken in Mali. To [toz]: Dormant; Niger-Congo; formerly spoken in Cameroon. Tocho (Toicho, Tacho) [taz]: Threatened, 3,800 speakers; Niger-Congo; spoken in Sudan. Tondi Songway Kiini (TSK) [tst]: Threatened, 3,000 speakers; Songhay; spoken in Mali. Toram (Torom, Torum) [trj]: Threatened, fewer than 4,000 speakers; Afro-Asiatic; spoken in Chad. Ts’amay (Tsamai, S’amai, Tamaha) [tsb]: Threatened, 8,621 speakers; Afro-Asiatic; spoken in Ethiopia. Ts’ixa (Ts’exa) [no ISO 639-3 code]: Endangered, approximately 180 speakers; Khoe; spoken in Botswana. Tshwa (Cuaa, Hiechware, Chware) [hio]: Threatened, 4,100 speakers; Khoe; spoken in Botswana and Zimbabwe. 107

Anna Belew et al.

Tsuvadi (Kambari, Avadi, Abadi) [tvd]: At risk, 150,000 speakers; Niger-Congo; spoken in Nigeria. Tswapong (Setswapong) [two]: Threatened, 2,000 speakers; Niger-Congo; spoken in Botswana. Tule (Tulai) [no ISO 639-3 code]: Dormant; Afro-Asiatic; formerly spoken in Nigeria. Tulishi (Tulesh, Thulishi, Kuntulishi) [tey]: Threatened, approximately 2,500 speakers; Kadu; spoken in Sudan. Tumi (Tutumi, Kitimi) [kku]: Threatened, 2,270 speakers; Niger-Congo; spoken in Nigeria. Tumtum [tbr]: Threatened, 7,300 speakers; Kadu; spoken in Sudan. Tunia (Tounia, Tunya, Tun) [tug]: Threatened, 2,260 speakers; Niger-Congo; spoken in Chad. Tunni (Af-Tunni) [tqq]: Vulnerable, 23,000 speakers; Afro-Asiatic; spoken in Somalia. Tunzu (Dugusa, Duguza) [dza]: Threatened, approximately 2,500 speakers; Niger-Congo; spoken in Nigeria. Tuotomb (Ponek, Bonek, Tuotom) [ttf]: Endangered, fewer than 1,000 speakers; NigerCongo; spoken in Cameroon. Tuwuli (Bowili, Bowiri, Liwuli) [bov]: Vulnerable, approximately 11,000 speakers; NigerCongo; spoken in Ghana. Twendi (Cambap, Northern Tiwa) [twn]: Severely endangered, approximately 30 speakers; Niger-Congo; spoken in Cameroon. Ubi (Oubi) [ubi]: Vulnerable, approximately 1,500 speakers; Afro-Asiatic; spoken in Chad. Uda [uda]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Nigeria. Ukhwejo (Benkonjo) [ukh]: Threatened, fewer than 2,000 speakers; Niger-Congo; spoken in Central African Republic. Ukwa [ukq]: Endangered, 100 speakers; Niger-Congo; spoken in Nigeria. Uncunwee (Ghulfan, Gulfan, Wunci) [ghl]: Endangered, fewer than 16,000 speakers; Nubian; spoken in Sudan. Vidunda (Chividunda, Kividunda, Ndunda) [vid]: Threatened, fewer than 10,000 speakers; Niger-Congo; spoken in Tanzania. Vinza [vin]: Vulnerable, 10,000 speakers; Niger-Congo; spoken in Tanzania. Vumbu (Yivoumbou) [vum]: Endangered, several hundred speakers; Niger-Congo; spoken in Gabon. Wali (Walari, Walarishe) [wll]: Threatened, 9,000 speakers; Nubian; spoken in Sudan. Wamey (Wamay, Wamei, Konyagi) [cou]: Vulnerable, 10,000–25,000 speakers; Niger-Congo; spoken in Senegal and Guinea. Wanda (Iciwanda, Ichiwanda, Wandia) [wbh]: Vulnerable, fewer than 42,810 speakers; NigerCongo; spoken in Tanzania. Wapan (Jukun, Juku, Wakari) [juk]: At risk, 100,000 speakers; Niger-Congo; spoken in Nigeria. Wïpha (Wase, Wase Tofa, Jukun Wase) [juw]: Threatened, 1,610 speakers; Niger-Congo; spoken in Nigeria. Wara (Wára, Ouara, Ouala) [wbf]: Threatened, 4,500 speakers; Niger-Congo; spoken in Burkina Faso. Warji (Sirzakwai, Warja, Warjawa) [wji]: Vulnerable, 77,700 speakers; Afro-Asiatic; spoken in Nigeria. Warnang (Werni) [wrn]: Threatened, 1,100 speakers; Niger-Congo; spoken in Sudan. Wawa [www]: Threatened, 2,500 speakers; Niger-Congo; spoken in Nigeria and Cameroon. Yaaku (Mukogodo, Mogogodo, Mukoquodo) [muu]: Severely endangered, approximately 50 speakers; Afro-Asiatic; spoken in Kenya. Yaka (Aka, Nyoyaka, Beká) [axk]: Vulnerable, approximately 30,000 speakers; Niger-Congo; spoken in Congo and Central African Republic. 108

The world’s endangered languages

Yanda Dom (Yanda Dom Dogon) [dym]: Threatened, 2,000 speakers; Niger-Congo; spoken in Mali. Yangkam (Yankam, Basharawa, Bashiri) [bsx]: Critically endangered, 400 speakers; NigerCongo; spoken in Nigeria. Yango (Gbendere) [yng]: Threatened, 3,000 speakers; Niger-Congo; spoken in Democratic Republic of the Congo. Yauma [yax]: Vulnerable, 22,200 speakers; Niger-Congo; spoken in Angola and Zambia. Yem (Yemsa, Yemma, Janjero) [jnj]: Threatened; Afro-Asiatic; spoken in Ethiopia. Yeyi (Shiyeyi, Yeei, Yei) [yey]: Severely endangered, 20,000 speakers; Niger-Congo; spoken in Botswana and Namibia. Yukuben (Boritsũ, Nyikuben, Nyikobe) [ybl]: Vulnerable, fewer than 16,000 speakers; NigerCongo; spoken in Nigeria and Cameroon. Yulu (Youlou) [yul]: Threatened, 3,000–5,000 speakers; Central Sudanic; spoken in Central African Republic, South Sudan, and Democratic Republic of the Congo. Zaar (Saya, Sayanci, Zaar) [say]: Vulnerable, approximately 150,000 speakers; Afro-Asiatic; spoken in Nigeria. Zakshi (Zaksi, Zaksə) [no ISO 639-3 code]: Endangered, approximately 500 speakers; AfroAsiatic; spoken in Nigeria. Zan Gula (Gula Guera, Goula, Moriil) [zna]: Threatened, 5,000 speakers; Niger-Congo; spoken in Chad. Zangwal (Zwangal, Twar) [zah]: Endangered, approximately 100 speakers; Afro-Asiatic; spoken in Nigeria. Zaramo (Zalamo, Kizaramo, Dzalamo) [zaj]: Critically endangered, a few speakers; NigerCongo; spoken in Tanzania. Zari (Zariwa, Kopti, Kwapm) [zaz]: Dormant; Afro-Asiatic; formerly spoken in Nigeria. Zay (Zway, Lak’i, Laqi) [zwa]: Threatened, fewer than 4,000 speakers; Afro-Asiatic; spoken in Ethiopia. Zaysé-Zergulla (Zaysse) [zay]: Vulnerable, 17,797 speakers; Afro-Asiatic; spoken in Ethiopia. Zeem [zua]: Dormant; Afro-Asiatic; formerly spoken in Nigeria. Zenaga [zen]: Endangered, at least 200 speakers; Afro-Asiatic; spoken in Mauritania and Western Sahara. Zialo (Ziolo) [zil]: Vulnerable, approximately 25,000 speakers; Niger-Congo; spoken in Guinea. Zirenkel [zrn]: Threatened, approximately 2,500 speakers; Afro-Asiatic; spoken in Chad. Zizilivakan (Ziziliveken, Ziliva, Àmzírív) [ziz]: Endangered; Afro-Asiatic; spoken in Nigeria and Cameroon. Zul (Bi Zule, Mbarmi, Zulawa) [no ISO 639-3 code]: Threatened, approximately 4,000 speakers; Afro-Asiatic; spoken in Nigeria. Zumaya [zuy]: Critically endangered; Afro-Asiatic; spoken in Cameroon.

Australia Stephanie Walla Linguistic diversity in Australia The languages in this region include those that are indigenous to Australia, including those of the Torres Strait. While Australia is located geographically close to islands in the Pacific, the languages spoken in these areas are discussed in the regional overview for the Pacific. 109

Anna Belew et al.

The Australian continent is home to a number of endangered language communities, with the majority of the endangered languages belonging to the Pama-Nyungan language family. As of April 2017, the Catalogue of Endangered Languages lists roughly 71% (255) of the endangered and dormant languages of this region as belonging to this family. The other language families in this region include Gunwinyguan (14), Worrorran (13), Western Daly (10), Nyulnyulan (9), Mirndi (7), Isolate (6), Iwaidjan (5), Maningrida (4), Tangkic (4), Giimbiyu (3), Jarrakan (3), Southern Daly (3), Bunaban (2), Eastern Daly (2), Garrwan (2), Limilngan (2), Northern Daly (2), Yangmanic (2), and Eastern Trans-Fly (1). Additionally, ELCat lists one creole for this region (Torres Strait Creole) and one sign language (Australian Sign Language). There are also four languages that remain “unclassified,” which touches on the issues of under-documentation of the languages in this region. Simply, many of the languages are critically endangered with very few speakers, making them difficult to document or classify.

Causes of language endangerment in Australia The largest contributor to language loss in Australia can be attributed to colonization. Thieberger (2013) estimates that there were between 250 and 600 languages spoken in Australia when the Europeans first settled in Sydney in 1788. Unfortunately, the large range in language estimates is due to historically poor documentation and recording of the indigenous languages. The consequences of British settlement were devastating, and Australian languages fell victim first to causes that put the language population at risk, and then to causes that discouraged or prevented speakers from using their languages. Indigenous languages in this region were affected by four major causes of language endangerment: disease, genocide, political repression, and cultural hegemony. A wave of epidemic Old World diseases like smallpox and measles caused a severe decline in the Aboriginal population, especially to those in areas of greatest population density, which was followed by conflicts over land resulting in widespread death of Aboriginal peoples. With the colonial invasion came forcible removal of Aboriginal people from their land, which resulted in a loss of traditional food, resources, and cultural and spiritual practices, which had a major and detrimental impact on Aboriginal languages as speaker populations were dispersed. Assimilation policies and the removal of Aboriginal children from their homes, which continued into the 1970s, further perpetuated language loss. The Stolen Generation of children were placed in institutions that forced their assimilation into Anglo-Australian culture and discouraged indigenous language use. The concentration of Aboriginal people in missions and reserves from different language groups resulted in the development of lingua francas between the groups, and eventually pidgins and creole languages, which remain important for communication today. Thieberger (2013) suggests that both the lack of one dominant indigenous language and the multilingualism of Aboriginal people, which aided English acquisition by indigenous people while the dominant society favored monolingualism, may have contributed to current language endangerment. As of January 2017, ELCat lists 356 endangered, dormant, and awakening languages in this region at varying degrees of vitality.

The current status of endangered languages in Australia Considering the historical effects on languages in this region, language endangerment remains a significant problem for Aboriginal communities in Australia. As of April 4, 2017, ELCat lists a total of 356 endangered and dormant languages for this region. Alarmingly, of those, 110

The world’s endangered languages

32% (116) are listed as having ten or fewer speakers. Eight of the languages in this region are listed without any known information about number of speakers. In some cases, we have not been able to confirm speaker numbers for these languages; for others there is no one who grew up speaking the language, but there are people who identify with it and are working to revitalize it. For these reasons, we have included these languages in our analysis of the languages in the region. As for languages known to have no native speakers in this region, there are 73 dormant languages and 35 awakening. Of the non-dormant and non-awakening languages that have sufficient information for a vitality rating, 229 (64%) are considered at least “Endangered.” Of these, about 60% (139) fall into the category of “Critically Endangered,” making this the most common vitality rating for languages in the region. The most stable languages in the region are classified as either “Threatened” (14) or “Vulnerable” (5) and consist of only 5% of the languages of Australia, making this the region of the world with the highest threat of language loss and lowest language stability. Australia’s status of having some of the most endangered languages in the world has spawned a great number of language maintenance and revitalization programs. Currently, there are many efforts in this region to support and revitalize endangered languages that have included collaboration between speakers, traditional language owners, and academics. The IAD Press, for instance, has produced numerous language resources like learner’s guides and language dictionaries, as well as cultural texts capturing oral histories and short stories. For a list of more revitalization efforts, see Purdie et al.’s (2008) summary of educational programs involving endangered languages; for further reading on language revitalization in Australia, see Hobson et al. (2011). Below is an alphabetical listing of all languages of Australia included in the Catalogue of Endangered Languages.

Endangered languages of Australia Adithinngithigh (Adetingiti) [dth]: Dormant; Pama-Nyungan; formerly spoken in Australia. Adnyamathanha (Adynyamathanha, Wailpi, Wailbi) [adt]: Endangered, 100 speakers; PamaNyungan; spoken in Australia. Aghu-Tharnggala (Aghu Tharnggalu, Kuku-Mini, Ikarranggali) [gtu]: Dormant; PamaNyungan; formerly spoken in Australia. Agwamin (Ak Waumin, Ak-Waumin, Egwamin) [wmi]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Alawa (Alaua, Allaura, Allawa) [alh]: Critically endangered, 17 speakers; Gunwinyguan; spoken in Australia. Alngith (Alngidh, Alingid, Laynngithy) [aid]: Critically endangered, three speakers; PamaNyungan; spoken in Australia. Alyawarr (Alyawarra, Alyawarre, Aljawara) [aly]: Vulnerable, 1,664 speakers; Pama-Nyungan; spoken in Australia. Amurdak (Wureidbug, Amurag, A’moordiyu) [amg]: Critically endangered, three to five speakers; Iwaidjan; spoken in Australia. Andajin (Andidja) [ajn]: Critically endangered, two speakers; Worrorran; spoken in Australia. Anindilyakwa (Amakurupa, Andilagwa, Andiljangwa) [aoi]: Threatened, approximately 1,500 speakers; Gunwinyguan; spoken in Australia. 111

Anna Belew et al.

Anmatyerre (Anmatjirra, Anmatjera, Unmatjera) [amx]: Threatened, 501–1,000 speakers; Pama-Nyungan; spoken in Australia. Antakirinya (Kadjilaranda, Madutara, Mbenderinga) [ant]: Severely endangered, 50 speakers; Pama-Nyungan; spoken in Australia. Antekerrepenh (Andigibinha, Antekerrepinhe, Andakerebina) [adg]: Critically endangered, very few speakers; Pama-Nyungan; spoken in Australia. Arabana-Wangkangurru (Arbana, Ngarabana, Arabuna) [ard]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ariba (Kurtjar, Araba, Rib) [aea]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Atampaya [amz]: Critically endangered, four speakers; Pama-Nyungan; spoken in Australia. Australian Sign Language (Auslan) [asf]: Threatened, 6,500 speakers; Sign Language; spoken in Australia. Awabakal (Arwarbukarl, Awabagal, Hunter River and Lake Macquarie) [awk]: Awakening; Pama-Nyungan; spoken in Australia. Awngthim (Angutimi, Mamangidigh) [gwm]: Dormant; Pama-Nyungan; formerly spoken in Australia. Awu Laya (Kuku-Thaypan, Gugu Dhayban, Kuku Thaypan) [typ]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ayapathu (Abadja, Ai ebadu, Aiabadu) [ayd]: Critically endangered, six speakers; PamaNyungan; spoken in Australia. Ayerreyenge (Jaroinga, Jarionga, Yaroinga) [axe]: Dormant; Pama-Nyungan; formerly spoken in Australia. Badimaya (Widimaya, Parti-Maya, Bidungu) [bia]: Dormant; Pama-Nyungan; formerly spoken in Australia. Badjiri (Budjari, Badyidi, Badjidi) [jbi]: Critically endangered, approximately two speakers; Pama-Nyungan; spoken in Australia. Bandjalang (Bandjelang, Bogganger, Bundala) [bdy]: Severely endangered, a few speakers; Pama-Nyungan; spoken in Australia. Barababaraba (Baraparapa, Burraburburaba, Barrababarraba) [rbp]: Critically endangered, five speakers; Pama-Nyungan; spoken in Australia. Bardi (Baadi, Bard, Baardi) [bcj]: Critically endangered, fewer than five speakers; Nyulnyulan; spoken in Australia. Barngarla (Parnkala, Parnkalla, Banggarla) [bjb]: Awakening; Pama-Nyungan; spoken in Australia. Barrow Point [bpt]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Batjamalh (Batyamal, Wadjiginy, Wogaity) [wdj]: Severely endangered, fewer than 12 speakers; isolate; spoken in Australia. Batyala (Badtjala, Batjala, Ngulungbara) [xby]: Awakening; Pama-Nyungan; spoken in Australia. Bibbulman (Pibelmen, Pepelman, Peopleman) [xbp]: Dormant; Pama-Nyungan; formerly spoken in Australia. Bidjara (Bitjara, Bithara, Bidjera) [bym]: Severely endangered, 20 speakers; Pama-Nyungan; spoken in Australia. Bigambal (Bigambul, Bigumbil) [xbe]: Dormant; Pama-Nyungan; formerly spoken in Australia. Bilinarra (Pilinara, Bunara, Boonarra) [no ISO 639-3 code]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. 112

The world’s endangered languages

Binbinka (Binbinga) [no ISO 639-3 code]: Critically endangered, two speakers; Mirndi; spoken in Australia. Bininj Gun-wok (Mayali, Kunwinjku) [gup]: Threatened, approximately 1,600 speakers; Gunwinyguan; spoken in Australia. Biri (Biria, Birri Gubba, Birigaba) [bzr, xgi]: Awakening; Pama-Nyungan; spoken in Australia. Boonwurrung (Bunurong, Boon Wurrung, Bun wurrung) [no ISO 639-3 code]: Awakening; Pama-Nyungan; spoken in Australia. Buandig (Buwandik, Booandik, Buandic) [xbg]: Awakening; Pama-Nyungan; spoken in Australia. Bunuba (Punapa, Bunapa, Punaba) [bck]: Critically endangered; Bunaban; spoken in Australia. Burarra (Anbara, Anbarra, Barara) [bvr]: Endangered, 400–600 speakers; Maningrida; spoken in Australia. Cocos Islands Malay (Cocos, Kokos) [coa]: Threatened, 5,000 speakers; Austronesian; spoken in Australia and Malaysia. Dadi Dadi (Dadidadi, Dardidardi, Dardi-Dardi) [dda]: Severely endangered, ten speakers; PamaNyungan; spoken in Australia. Dalabon (Ngalkbun, Ngalkbon, Buin) [ngk]: Critically endangered, five speakers; Gunwinyguan; spoken in Australia. Darkinyung (Darginjang, Darginyung, Darkinjang) [xda]: Awakening; Pama-Nyungan; spoken in Australia. Dhanggati (Djan-Gadi, Dainggati, Boorkutti) [dyn]: Awakening; Pama-Nyungan; spoken in Australia. Dhangu-Djangu (Dhangu, Dhaŋu, Djangu) [dhg]: Endangered, fewer than 350 speakers; PamaNyungan; spoken in Australia. Dharawal (Tharawal, Darawal, Carawal) [tbh]: Critically endangered, four speakers; PamaNyungan; spoken in Australia. Dharug (Daruk, Dharruk, Dharuk) [xdk]: Awakening; Pama-Nyungan; spoken in Australia. Dharumbal (Baiali, Biyali, Darambal) [xgm]: Awakening; Pama-Nyungan; spoken in Australia. Dhay’yi (Dayi, Dha’i, Daii) [dax]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Dhungaloo [dhx]: Dormant; Pama-Nyungan; formerly spoken in Australia. Dhurga (Dhu’rga, Durga, Thoorga) [dhu]: Awakening; Pama-Nyungan; spoken in Australia. Dhuwal (Duala, Dual, Wulamba) [duj, djr]: Threatened, 3,044 speakers; Pama-Nyungan; spoken in Australia. Dhuwala (Gupapuyngu, Gobabingo, Kupapuyngu) [guf, gnn]: Endangered, 321 speakers; PamaNyungan; spoken in Australia. Diyari (Dieri, Dayerrie, Deerie) [dif, dit]: Severely endangered, approximately 12 speakers; Pama-Nyungan; spoken in Australia. Djabugay (Dyabugay, Tjapukai, Bulum-Bulum) [dyy]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Djabwurrung (Tjapwurong, Tyapwurru, Chaapwurru) [tjw]: Critically endangered, seven speakers; Pama-Nyungan; spoken in Australia. Djangun (Adho-Adhom, Butju, Changunberries) [djf]: Dormant; Pama-Nyungan; formerly spoken in Australia. Djinang (Jandjinung, Yandijinang, Yandjinung) [dji]: Endangered, approximately 200 speakers; Pama-Nyungan; spoken in Australia. 113

Anna Belew et al.

Djinba (Jinba, Djimba, Outjanbah) [djb]: Severely endangered, 60–90 speakers; PamaNyungan; spoken in Australia. Djuwarliny (Juwaliny, Tjuwalinj, Tjiwarliñ) [no ISO 639-3 code]: Critically endangered, a few speakers; Pama-Nyungan; spoken in Australia. Duungidjawu (Dundijau) [no ISO 639-3 code]: Dormant; Pama-Nyungan; formerly spoken in Australia. Dyirbal (Chirpa, Chirpalji, Choolngai) [dbl]: Critically endangered, six speakers; PamaNyungan; spoken in Australia. Eastern and Central Arrernte (Mparntwe Arrernte) [aer]: Threatened, 3,820 speakers; PamaNyungan; spoken in Australia. Emmi (Merranunggu, Warrgat, Maranungku) [zmr, amy]: Critically endangered, 15–20 speakers; Western Daly; spoken in Australia. Eora (Iyora, Iora, Eaora) [no ISO 639-3 code]: Awakening; Pama-Nyungan; spoken in Australia. Erre (Ere, Ari, Erei) [err]: Critically endangered, one speaker; Giimbiyu; spoken in Australia. Flinders Island (Yalgawarra, Wurima, Mutumui) [fln]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Gagudju (Abedal, Abiddul, Arwur) [gbu]: Dormant; isolate; formerly spoken in Australia. Gajirrabeng (Gadjerong, Kajirrawung, Gadyerong) [gdh]: Critically endangered, fewer than three speakers; Jarrakan; spoken in Australia. Gamberre (Gamberre, Gambere, Gambre) [gma]: Critically endangered, six speakers; Worrorran; spoken in Australia. Gamilaraay (Camileroi, Gamilaroi, Kamilaroi) [kld]: Critically endangered; Pama-Nyungan; spoken in Australia. Ganggalida (Engarilla, Eugoola) [gcd]: Critically endangered, one speaker; Tangkic; spoken in Australia. Garig (Gaari, Gari, Garik) [no ISO 639-3 code]: Critically endangered, four speakers; Iwaidjan; spoken in Australia. Garlali (Kullila, Kalali, Galali) [gll]: Dormant; Pama-Nyungan; formerly spoken in Australia. Garrwa (Karrwa, Garawa, Karawa) [wrk]: Severely endangered, 59 speakers; Garrwan; spoken in Australia. Gayiri (Kairi, Khararya) [no ISO 639-3 code]: Dormant; Pama-Nyungan; formerly spoken in Australia. Girramay (Keramai, Kiramay, Kiramai) [no ISO 639-3 code]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Githabul (Gidhabal, Gidabal, Kidabal) [gih]: Awakening; Pama-Nyungan; spoken in Australia. Giyug [giy]: Critically endangered, two speakers; Unclassified; spoken in Australia. Golpa (Kolpa, Golbu, Gorlba) [lja]: Severely endangered; Pama-Nyungan; spoken in Australia. Gooniyandi (Guniyandi, Guniandi, Gunian) [gni]: Severely endangered, approximately 100 speakers; Bunaban; spoken in Australia. Goreng (Koreng, Korengi, Kuriny) [xgg]: Awakening; Pama-Nyungan; spoken in Australia. Gubbi Gubbi (Gabi-Gabi, Gabi, Kabikabi) [gbw]: Dormant; Pama-Nyungan; formerly spoken in Australia. Gudjal (Warrungu, Warungu, Gugu-Badhun) [wrg]: Awakening; Pama-Nyungan; spoken in Australia. Gugu-Badhun (Kokopatun, Patun, Koko Padun) [gdc]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. 114

The world’s endangered languages

Gugu-Warra (Kokowarra, Kokowara, Laia) [wrw]: Dormant; Pama-Nyungan; formerly spoken in Australia. Gumbaynggir (Gumbaingari, Kumbainggar, Kumbaingeri) [kgs]: Severely endangered, 40 speakers; Pama-Nyungan; spoken in Australia. Gundungurra (Burragorang, Gan dangara, Gandangara) [xrd]: Awakening; Pama-Nyungan; spoken in Australia. Gungabula (Kongabula, Ongabula, Khungabula) [gyf]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Gunggari (Kunggari, Congaro, Coongurri) [kgl]: Severely endangered, approximately ten speakers; Pama-Nyungan; spoken in Australia. Gunin (Kwini, Gunan, Gwini) [gww]: Severely endangered, at least 50 speakers; Worrorran; spoken in Australia. Gunya (Kunja, Gunja, Kurnja) [gyy]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Gureng Gureng (Curanggurang, Curang-gurang, Goorang-goorang) [gnr]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Gurindji (Wurlayi, Coorinji, Garundji) [gue]: Endangered, 175 speakers; Pama-Nyungan; spoken in Australia. Gurr-goni (Gurrgoni, Gurrogoni, Gungurugoni) [gge]: Critically endangered, fewer than 20 speakers; Maningrida; spoken in Australia. Guugu-Yimidhirr (Kukuyimidir, Koko Imudji, Gugu Yimijir) [kky]: Severely endangered, fewer than 20 speakers; Pama-Nyungan; spoken in Australia. Guwa (Goa, Goamulgo, Coa) [xgw]: Dormant; Pama-Nyungan; formerly spoken in Australia. Guwamu (Koamu, Kooma, Kuamu) [gwu]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Ikaranggal (Ikarranggali, Aghu Tharnggalu) [ikr]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ilgar (Garig-Ilgar,Garig) [ilg]: Critically endangered, fewer than five speakers; Iwaidjan; spoken in Australia. Iwaidja (Iwaydja, Iwaidji, Ibadjo) [ibd]: Endangered, 150 speakers; Iwaidjan; spoken in Australia. Jabirr-Jabirr (Jabirrjabirr, Dyaberdyaber, Djaberdjaber) [dyb]: Dormant; Nyulnyulan; formerly spoken in Australia. Jaminjung (Baj Subgroup, Dyaminydyung, Djamadjong) [djd]: Severely endangered, at least 100 speakers; Mirndi; spoken in Australia. Janday (Jandai, Koenpel, Goenpul) [jan]: Awakening; Pama-Nyungan; spoken in Australia. Jaru (Djaru, Jaroo, Tjaru) [ddj]: Severely endangered, 250 speakers; Pama-Nyungan; spoken in Australia. Jawi (Djaui, Towahi, Tohawi) [djw]: Critically endangered, fewer than three speakers; Nyulnyulan; spoken in Australia. Jawoyn (Adowen, Charmong, Chauan) [djn]: Critically endangered, fewer than 100 speakers; Gunwinyguan; spoken in Australia. Jingulu (Djingili, Djingulu, Jingali) [jig]: Severely endangered, approximately ten speakers; Mirndi; spoken in Australia. Jiwarli (Djwarli, Tjiwarli, Djiwali) [dze]: Dormant; Pama-Nyungan; formerly spoken in Australia. Jurruru (Tjuroro, Tjururu, Tjororo) [tju]: Dormant; Pama-Nyungan; formerly spoken in Australia. 115

Anna Belew et al.

Kaanju (Kandju, Kaantyu, Gandju) [kbe]: Critically endangered, fewer than 50 speakers; PamaNyungan; spoken in Australia. Kalaw Kawaw Ya (Kala Yagaw Ya, Yagar Yagar, Mabuiag) [mwp]: Endangered, approximately 4,000 speakers; Pama-Nyungan; spoken in Australia. Kalkatungu (Calcadoon, Galgadung, Galgadungu) [ktg, wwb]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kamu (Gamor, Kamor, Ngangigamor) [xmu]: Critically endangered, two speakers; Eastern Daly; spoken in Australia. Kaniyang (Nyunga, Kaniyan, Kaneang) [no ISO 639-3 code]: Awakening; Pama-Nyungan; spoken in Australia. Karajarri (Karrajarri, Garadjiri, Garadyari) [gbd]: Critically endangered, 12 speakers; PamaNyungan; spoken in Australia. Kariyarra (Kariera, Karriara, Kariyara) [vka]: Critically endangered, fewer than two speakers; Pama-Nyungan; spoken in Australia. Kartujarra (Kartutjara, Kardutjara, Kadaddjara) [no ISO 639-3 code]: Critically endangered, five speakers; Pama-Nyungan; spoken in Australia. Kaurna (Kaura, Coorna, Koornawarra) [zku]: Awakening; Pama-Nyungan; spoken in Australia. Kayardild (Gayardild, Gaiardilt) [gyd]: Critically endangered, five speakers; Tangkic; spoken in Australia. Kaytetye (Kaiditj, Kaititj, Gaididj) [gbb]: Endangered, 200 speakers; Pama-Nyungan; spoken in Australia. Kija (Gidja, Kidja, Ku:tji) [gia]: Severely endangered, approximately 100 speakers; Jarrakan; spoken in Australia. Kok-Nar (Gundara, Gog Nar, Kundara) [gko]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kokatha (Aluna, Cocotah, Cookutta) [ktd]: Critically endangered, three speakers; PamaNyungan; spoken in Australia. Koko Babangk (Kok Peponk, Kok-Babonk, Koko Kunaniy) [okg]: Dormant; Pama-Nyungan; formerly spoken in Australia. Koko-Bera (Koko Bera, Kukaberra, Kungkara) [kkp]: Severely endangered, fewer than 15 speakers; Pama-Nyungan; spoken in Australia. Kolakngat (Kolacgnat, Coligan, Kolijon) [no ISO 639-3 code]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kugu-Nganhcara (Wik Nganychara, Wik Ngencherr, Wik-Ngencherra) [wua]: Severely endangered, fewer than 50 speakers; Pama-Nyungan; spoken in Australia. Kukatj (Kukatji, Konggada, Kalibamu) [ggd]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kukatja (Nambulatji , Panara, Pardoo) [kux]: Vulnerable, 1,000 speakers; Pama-Nyungan; spoken in Australia. Kuku-Mangk (Kugu-Mangk) [xmq]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Kuku-Mu’inh (Kugu-Mu’inh) [xmp]: Critically endangered, seven speakers; Pama-Nyungan; spoken in Australia. Kuku-Muminh (Kugu-Muminh, Wik Muminh, Wik-Mumin) [xmh]: Critically endangered, fewer than 31 speakers; Pama-Nyungan; spoken in Australia. Kuku-Ugbanh (Kugu-Ugbanh) [ugb]: Critically endangered, six speakers; Pama-Nyungan; spoken in Australia. 116

The world’s endangered languages

Kuku-Uwanh (Kugu-Uwanh, Kugu Nganhcara) [uwa]: Severely endangered, fewer than 40 speakers; Pama-Nyungan; spoken in Australia. Kuku-Yalanji (Guguyalanji, Koko-Yalanji, Kuku-Yalangi) [gvn]: Endangered, 700 speakers; Pama-Nyungan; spoken in Australia. Kunbarlang (Gunbalang, Walang, Warlang) [wlg]: Severely endangered, 50–100 speakers; Gunwinyguan; spoken in Australia. Kungarakany (Gungaragan, Gunerakan, Kangarraga) [ggk]: Dormant; isolate; formerly spoken in Australia. Kunggara (Koonkurri, Ungorri, Gunggara) [kvs]: Critically endangered, ten speakers; PamaNyungan; spoken in Australia. Kungkari (Gunggari, Kuungkari, Koonkerri) [lku]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kunjen (Guguminjen, Kukumindjen, Kun’djan) [kjn]: Critically endangered, 20–25 speakers; Pama-Nyungan; spoken in Australia. Kurnai (Gunnai, Gunai, Ganai) [unn]: Awakening; Pama-Nyungan; spoken in Australia. Kurrama (Gurama, Janari, Jawunmara) [vku]: Critically endangered, fewer than ten speakers; Pama-Nyungan; spoken in Australia. Kurtjar (Kurtjjar, Kurrtyar, Gurdyar) [gdj]: Critically endangered, one to two speakers; PamaNyungan; spoken in Australia. Kuthant (Gudhand, Gudanda, Gudhanda) [xut]: Critically endangered, three speakers; PamaNyungan; spoken in Australia. Kuuk Thaayorre (Thayore, Thaayorre, Taiol) [thd]: Endangered, 150 speakers; PamaNyungan; spoken in Australia. Kuuk-Yak (Koko Yak, Kuku Yak) [uky]: Dormant; Pama-Nyungan; formerly spoken in Australia. Kuuku-Ya’u (Ya’o, Koko-Ja’o, Kokoyao) [kuy]: Severely endangered, fewer than 100 speakers; Pama-Nyungan; spoken in Australia. Kuwema (Tyaraity, Dyeraidy, Daktjerat) [woa]: Severely endangered, ten speakers; Northern Daly; spoken in Australia. Kuyani (Guyani, Kijani, Kwiani) [gvy]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ladji Ladji (Latjilatji, Laitchi-Laitchi, Litchy-Litchy) [llj]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Lamalama (Lamulamul, Lamu-Lamu, Mba Rumbathama) [lby]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Lardil (Engarilla, Eugoola, Ganggalida) [lbz]: Critically endangered, fewer than five speakers; Tangkic; spoken in Australia. Larrakiya (Laragiya, Larakia, Larakiya) [lrg]: Critically endangered, six speakers; Unclassified; spoken in Australia. Limilngan (Limil, Minitjja, Manadja) [lmc]: Critically endangered, three speakers; Limilngan; spoken in Australia. Linngithigh (Linngithig, Leningitij, Linngidhigh) [lnj]: Dormant; Pama-Nyungan; formerly spoken in Australia. Liyagalawumirr [no ISO 639-3 code]: Severely endangered, 50 speakers; Pama-Nyungan; spoken in Australia. Liyagawumirr [no ISO 639-3 code]: Severely endangered, 50 speakers; Pama-Nyungan; spoken in Australia. 117

Anna Belew et al.

Lower Southern Aranda (Aldolanga, Aldolinga, Alitera) [axl]: Dormant; Pama-Nyungan; formerly spoken in Australia. Malak Malak (MullukMulluk, Ngolok-Wangar) [mpb]: Critically endangered, 11 speakers; Northern Daly; spoken in Australia. Malkana (Maljanna, Maldjana, Malgana) [vml]: Dormant; Pama-Nyungan; formerly spoken in Australia. Malyangapa (Maljangaba, Malya-napa, Mulya-napa) [yga]: Dormant; Pama-Nyungan; formerly spoken in Australia. Manda (Manhthe) [zma]: Critically endangered, fewer than 25 speakers; Western Daly; spoken in Australia. Mandandanyi (Mandandanjnjdji, Mandandanji, Mundaeinbura) [zmk]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Mandjalpingu (Manydjalpingu, Manydjalpuyŋu, Mandalpuy) [no ISO 639-3 code]: Severely endangered, 30 speakers; Pama-Nyungan; spoken in Australia. Mangala (Mangalaa, Mangarla, Manala) [mem]: Severely endangered, fewer than 20 speakers; Pama-Nyungan; spoken in Australia. Mangarrayi (Mangarayi, Mungerry, Mangarai) [mpc]: Critically endangered, fewer than 50 speakers; Gunwinyguan; spoken in Australia. Mangerr (Mangerri, Mangerei, Mennagi) [zme]: Critically endangered, one speaker; Giimbiyu; spoken in Australia. Manjiljarra (Mantjiltjara, Mandijildjara, Manyjilyjara) [no ISO 639-3 code]: Endangered, fewer than 500 speakers; Pama-Nyungan; spoken in Australia. Margany (Marrganj, Maranganji, Mardigan) [zmc]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Maridan (Marridan, Meradan, Maredan) [zmd]: Critically endangered, fewer than 20 speakers; Western Daly; spoken in Australia. Maridjabin (Marrityabin, Maretyabin, Maridyerbin) [zmj]: Critically endangered, fewer than 20 speakers; Western Daly; spoken in Australia. Marimanindji (Marrimanindji, Maramarandji, Maramanandji) [zmm]: Critically endangered, fewer than 15 speakers; Western Daly; spoken in Australia. Mariyedi (Marijedi, Murijadi, Marijadi) [zmy]: Critically endangered, fewer than 20 speakers; Western Daly; spoken in Australia. Marra (Leelalwarra, Leelawarra, Mala) [mec]: Critically endangered, five speakers; Gunwinyguan; spoken in Australia. Marrgu (Maraura, Maroura, Marowera) [mhg]: Critically endangered, one speaker; unclassified; spoken in Australia. Marriammu (Marri Ammu, Mare Ammu, Mareammu) [xru]: Critically endangered, six to seven speakers; Western Daly; spoken in Australia. Marringarr (Marenggar, Maringa, Marengar) [zmt]: Critically endangered, 30–40 speakers; Western Daly; spoken in Australia. Marrithiyel (Berinken, Berringen, Berringin) [mfr]: Critically endangered, fewer than 25 speakers; Western Daly; spoken in Australia. Martu Wangka (Mardo, Targudi, Jigalong) [mpj]: Endangered, 800–1,000 speakers; PamaNyungan; spoken in Australia. Martuthunira (Martuyhunira, Mardudjungara, Mardudhunera) [vma]: Critically endangered, five speakers; Pama-Nyungan; spoken in Australia. Mathi-Mathi (Madhi Madhi, Madhimadhi, Madi Madi) [dmd]: Awakening; Pama-Nyungan; spoken in Australia. 118

The world’s endangered languages

Mati Ke (Madige, Magati-Ge, Magadige) [zmg]: Critically endangered, two to three speakers; Western Daly; spoken in Australia. Matngele (Madngele, Matngala, Warat) [zml]: Critically endangered, 15–20 speakers; Eastern Daly; spoken in Australia. Mawng (Gunmarung, Mawung, Gun-Marung) [mph]: Vulnerable, approximately 300 speakers; Iwaidjan; spoken in Australia. Mayi-Kulan (Maykulan, Wunamara, Mayi-Thakurti) [xyk]: Dormant; Pama-Nyungan; formerly spoken in Australia. Mayi-Kutuna (Maigudina, Maigudung, Maikudun) [xmy]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Mayi-Thakurti (Maidhagudi, Maitakudi, Mayatagoorri) [xyt]: Dormant; Pama-Nyungan; formerly spoken in Australia. Mayi-Yapi (Miappe, Miubbi, Myabi) [xyj]: Dormant; Pama-Nyungan; formerly spoken in Australia. Mbabaram (Mbara, Barbarem, Bar-barum) [vmb]: Dormant; Pama-Nyungan; formerly spoken in Australia. Mbariman-Gudhinma (Rimang-Gudinhma, Port Stewart Lamalama, Parimankutinma) [zmv]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Mbiywom (Mbeiwum, Kok Mbewan, Kok Mbewam) [no ISO 639-3 code]: Dormant; PamaNyungan; formerly spoken in Australia. Meriam (Miriam, Meryam Mir, Miriam-Mir) [ulk]: Endangered, 320 speakers; Eastern TransFly; spoken in Australia. Minjungbal (Minyangbal, Minyung, Minyowa) [xjb]: Dormant; Pama-Nyungan; formerly spoken in Australia. Miriwoong (Mirung, Merong, Miriwun) [mep]: Critically endangered; Jarrakan; spoken in Australia. Mirniny (Mirning-Ngadjunma-Kalarko, Mirning, Mirniny) [no ISO 639-3 code]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Mithaka (Mithaka, Midhaga) [no ISO 639-3 code]: Dormant; Pama-Nyungan; formerly spoken in Australia. Mudburra (Madbara, Moodburra, Mootburra) [dmw]: Critically endangered, approximately 50 speakers; Pama-Nyungan; spoken in Australia. Muluridyi (Binjara, Kokomoloroij, Kokomoloroitji) [vmu]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Murrinh-patha (Garama, Garrama, Murrinh-Patha) [mwf]: Threatened; Southern Daly; spoken in Australia. Muruwari (Marawari, Muruwarri, Murawarri) [zmu]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Nakara (Kokori, Nagara, Na-kara) [nck]: Severely endangered, 75–100 speakers; Maningrida; spoken in Australia. Nari Nari (Narri-Narri, Narinari, Nari-Nari) [rnr]: Dormant; Pama-Nyungan; formerly spoken in Australia. Narungga (Nanunga, Naranga, Narranga) [nnr]: Severely endangered, 24 speakers; PamaNyungan; spoken in Australia. Ndjebbana (Gunibidji, Gunavidji, Ndjébbana) [djj]: Endangered, 241 speakers; Maningrida; spoken in Australia. Ngaanyatjarra (Nyanganyatjara, Ngaanjatjarra, Ngadawongga) [ntj]: Threatened, 1,300 speakers; Pama-Nyungan; spoken in Australia. 119

Anna Belew et al.

Ngaatjatjara (Jabungadja, Ku:rara, Nadadjara) [no ISO 639-3 code]: Severely endangered, 12 speakers; Pama-Nyungan; spoken in Australia. Ngalakgan (Ngalakant, Hongalla, Ngalangan) [nig]: Dormant; Gunwinyguan; formerly spoken in Australia. Ngaliwurru (Baj Subgroup, Ngaliwerra, Ngaliwuru) [no ISO 639-3 code]: Endangered, approximately 100 speakers; Mirndi; spoken in Australia. Ngamini (Yarluyandi, Yarluyandji, Karangura) [nmv]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Ngan’gikurunggurr (Tyemeri, Marityemeri, Ngangikurrunggurr) [no ISO 639-3 code]: Endangered, 240 speakers; Southern Daly; spoken in Australia. Ngan’gityemerri (Ngan’gityemeri, Ngan’gikurunggurr, Nangikurrunggurr) [nam]: Endangered, approximately 150 speakers; Southern Daly; spoken in Australia. Ngandi (N’gundi, Ngalbon, Ngalgbun) [nid]: Critically endangered, nine speakers; Gunwinyguan; spoken in Australia. Nganyaywana (Anaiwan, Anaywan, Anewan) [nyx]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ngardi (Ngarti, Ngadi, Ngari) [rxd]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Ngarigu (Ngarigu, Ngarigo, Ngaruku) [xni]: Dormant; Pama-Nyungan; formerly spoken in Australia. Ngarinyin (Ungarinyin, Ngarinjin, Ungarinjin) [ung]: Critically endangered, approximately 12 speakers; Worrorran; spoken in Australia. Ngarinyman (Ngarinmany, Agarinman, Airiman) [nbj]: Endangered, fewer than 170 speakers; Pama-Nyungan; spoken in Australia. Ngarla (Nga:la, Ngala, Ngerla) [nlr]: Critically endangered, very few speakers; Pama-Nyungan; spoken in Australia. Ngarluma (Ngalama, Ngaluma, Ngallooma) [nrl]: Severely endangered, fewer than 20 speakers; Pama-Nyungan; spoken in Australia. Ngarnka (Kundanja, Anga, Angee) [nji]: Critically endangered, approximately three speakers; Mirndi; spoken in Australia. Ngarrindjeri (Narrinyeri, Ngarinyeri, Yaralde) [nay]: Awakening; Pama-Nyungan; spoken in Australia. Ngawun (Ngaun, Ngaon, Nouun) [nxn]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Ngiyambaa (Wongaibon, Ngiyampaa, Wangaaybuwan) [wyb]: Critically endangered; PamaNyungan; spoken in Australia. Ngomburr (Ngormbur, Gnormbur, Gnumbu) [nrx]: Critically endangered, one speaker; Unclassified; spoken in Australia. Ngumbarl (Ngombaru, Ngormbal) [no ISO 639-3 code]: Dormant; Nyulnyulan; formerly spoken in Australia. Ngunawal (Burragorang, Ngunuwal, Ngunnawal) [xul]: Awakening; Pama-Nyungan; spoken in Australia. Nhanta (Amangu, Champion Bay, Jau) [nha]: Critically endangered, two speakers; PamaNyungan; spoken in Australia. Nhirrpi (Nhirrpi) [hrp]: Dormant; Pama-Nyungan; formerly spoken in Australia. Nhuwala (Nuala, Ngoala, Noella) [nhf]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. 120

The world’s endangered languages

Nimanburru (Nimanbur, Nimanboru, Nimanboro) [nmp]: Critically endangered, two speakers; Nyulnyulan; spoken in Australia. Nukunu (Nugunu, Nukuna, Doora) [nnv]: Dormant; Pama-Nyungan; formerly spoken in Australia. Nungali (Ilngali, Jilngali) [nug]: Critically endangered, two speakers; Mirndi; spoken in Australia. Nyamal (Njamat, Wanbarda, Widagari) [nly]: Critically endangered, fewer than ten speakers; Pama-Nyungan; spoken in Australia. Nyangumarta (Ibarga, Ibarrga, Ibargo) [nna]: Endangered, approximately 520 speakers; Pama-Nyungan; spoken in Australia. Nyawaygi (Yi, Nawagi, Nyawigi) [nyt]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Nyikina (Njigina, Njikena, Nyigini) [nyh]: Severely endangered, approximately 50 speakers; Nyulnyulan; spoken in Australia. Nyiyaparli (Njiabali, Njijabali, Iabali) [xny]: Severely endangered, approximately 50 speakers; Pama-Nyungan; spoken in Australia. Nyulnyul (Nyul-Nyul, Niol-Niol, Nyolnyol) [nyv]: Critically endangered, one speaker; Nyulnyulan; spoken in Australia. Olkol (Koka-ollugul, Koko Olkol, Koko Olkolo) [olk]: Critically endangered, five speakers; Pama-Nyungan; spoken in Australia. Paakantyi (Darling, Kula) [drl]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Pakanh (Bakanha, Bakanh, Ajabakan) [pkn]: Severely endangered, ten speakers; PamaNyungan; spoken in Australia. Palyku (Nijadali, Bailko, Balgu) [nad]: Critically endangered, very few speakers; PamaNyungan; spoken in Australia. Panyjima (Pandjima, Panjtjima, Bandjima) [pnw]: Severely endangered, approximately 50 speakers; Pama-Nyungan; spoken in Australia. Payungu (Baijungo, Paiunggu, Baiong) [bxj]: Critically endangered, six speakers; PamaNyungan; spoken in Australia. Pini (Bandjanu, Banjanu, Bini) [pii]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Pinikura (Binnigoora, Biniguru, Binnigora) [pnv]: Dormant; Pama-Nyungan; formerly spoken in Australia. Pinjarup (Pinjarra, Bangoula, Banyoula) [pnj]: Awakening; Pama-Nyungan; spoken in Australia. Pintiini (Bindinini, Bindunda, Pindiini) [pti]: Endangered, 200–300 speakers; Pama-Nyungan; spoken in Australia. Pintupi-Luritja (Panika, Bindubu, Pi:ntubi) [piu]: Threatened, 2,080 speakers; Pama-Nyungan; spoken in Australia. Pirriya (Birria, Biria, Bidia) [xpa]: Dormant; Pama-Nyungan; formerly spoken in Australia. Pitjantjatjara (Bidjandja, Bidjandjadjara, Bidjandjara) [pjt]: Vulnerable, 3,000 speakers; Pama-Nyungan; spoken in Australia. Pitta-Pitta (Pita Pita, Pitha-Pitha, Bidhabidha) [pit, lnw, wky]: Dormant; Pama-Nyungan; formerly spoken in Australia. Punthamara (Bundhamara, Bunthomarra, Buntamara) [xpt]: Dormant; Pama-Nyungan; formerly spoken in Australia. 121

Anna Belew et al.

Purduna (Buduna, Boordoona, Bootena) [bxn]: Dormant; Pama-Nyungan; formerly spoken in Australia. Putijarra (Budidjara, Putujara, Pawututjara) [no ISO 639-3 code]: Critically endangered, approximately four speakers; Pama-Nyungan; spoken in Australia. Rembarrnga (Rembarranga, Rembarunga, Rembarrunga) [rmb]: Severely endangered, 69 speakers; Gunwinyguan; spoken in Australia. Ritharrngu (Ritarungo, Ridarngo, Rittarungo) [rit]: Endangered, at least 300 speakers; PamaNyungan; spoken in Australia. Tagalaka (Dagalang, Da:galag, Takalak) [tgz]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Taungurung (Daungwurrung, Daung Wurrung, Dhagung-wurrung) [dgw]: Awakening; PamaNyungan, Kulin; spoken in Australia. Thalanyji (Dalandji, Dalendi, Djalendi) [dhl]: Critically endangered, a few speakers; PamaNyungan; spoken in Australia. Tharrkari (Targari, Dal’gari, Tarkarri) [dhr]: Critically endangered, approximately two speakers; Pama-Nyungan; spoken in Australia. Thiin (Te:n, Teen, Deen) [iin]: Dormant; Pama-Nyungan; formerly spoken in Australia. Tiwi (Tunuvivi, Diwi, Wonga:k) [tiw]: Threatened, 1,500 speakers; isolate; spoken in Australia. Tjungundji [no ISO 639-3 code]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Tjupany (Tjupany) [no ISO 639-3 code]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Torres Strait Creole (Torres Strait Pidgin, Lockhart Creole) [tcs]: Vulnerable, approximately 6,000 speakers; pidgin or creole; spoken in Australia. Umbugarla (Mbakarla, Mbukarla, Bugula) [umr]: Dormant; isolate; formerly spoken in Australia. Umbuygamu (Moroba-Lama, Baka, Banjigam) [umg]: Critically endangered, seven speakers; Pama-Nyungan; spoken in Australia. Umiida (Umida, Umi:da, Oomida) [xud]: Dormant; Worrorran; formerly spoken in Australia. Umpila (Ombila, Ompeila, Ompela) [ump]: Critically endangered, a few speakers; PamaNyungan; spoken in Australia. Umpithamu (Baka, Banjigam, Barungguan) [umd]: Critically endangered, at least two speakers; Pama-Nyungan; spoken in Australia. Unggarranggu (Unggarrangu, Unggarangi, Oonggarrangoo) [xun]: Dormant; Worrorran; formerly spoken in Australia. Unggumi (Ungkami, Wunggumi, Oonggoomi) [xgu]: Dormant; Worrorran; formerly spoken in Australia. Uradhi (Yinwum, Jinwum, Yeemwoom) [urf]: Critically endangered, two speakers; PamaNyungan; spoken in Australia. Urningangga (Uningangk, Wuningak, Wuningangk) [urc]: Severely endangered, ten speakers; Giimbiyu; spoken in Australia. Waanyi (Wanji, Wainyi, Wanyu) [wny]: Critically endangered, one speaker; Garrwan; spoken in Australia. Wadi-Wadi (Wadi Wadi, Wathiwathi, Watthiwatthi) [xwd]: Dormant; Pama-Nyungan; formerly spoken in Australia. Wadjigu (Wadja, Wainjago, Wainjigo) [wdu]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. 122

The world’s endangered languages

Wagiman (Wogeman, Wageman, Wagaman) [waq]: Severely endangered, fewer than 50 speakers; isolate; spoken in Australia. Wajarri (Watjari, Watjarri, Wadjari) [wbv]: Severely endangered, fewer than 50 speakers; Pama-Nyungan; spoken in Australia. Wajuk (Beelia, Beeloo, Derbal) [xwj]: Dormant; Pama-Nyungan; formerly spoken in Australia. Waka-Waka (Enibura, Mungar, Nukunukubara) [wkw, wur]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Wakaya (Akaja, Arkiya, Leewakya) [wga]: Severely endangered, ten speakers; Pama-Nyungan; spoken in Australia. Walmajarri (Juwaliny, Tjuwalinj, Tjiwaliny) [wmt]: Endangered, 860 speakers; PamaNyungan; spoken in Australia. Wambaya (Wombaia, Wambaja) [wmb]: Critically endangered, two to three speakers; Mirndi; spoken in Australia. Wanggamala (Tharlimanha, Wanggamanha, Wangkamala) [wnm]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Wangganguru (Wangkangurru, Wangka-Jutjuru, Wangkadjera) [wgg]: Critically endangered, one to two speakers; Pama-Nyungan; spoken in Australia. Wangkajunga (Wangkajungka) [no ISO 639-3 code]: Severely endangered, 20 speakers; PamaNyungan; spoken in Australia. Wangkumara (Wongkumara, Wangumarra, Karenggapa) [xwk]: Awakening; Pama-Nyungan; spoken in Australia. Wardaman (Wartaman, Warduman, Wadaman) [wrr]: Severely endangered, 30 speakers; Yangmanic; spoken in Australia. Wardandi (Wadarandee, Wardandie, Kardagur) [wxw]: Awakening; Pama-Nyungan; spoken in Australia. Warlmanpa (Wolperi, Wolpirra, Albura) [wrl]: Severely endangered, 36 speakers; PamaNyungan; spoken in Australia. Warlpiri (Albura, Alpira, Alpiri) [wbp]: Threatened, approximately 3,000 speakers; PamaNyungan; spoken in Australia. Warluwarra (Didjadidja, Elookera, Maula) [wrb]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Warndarrang (Wandaran, Wandarang, Wandarung) [wnd]: Dormant; Gunwinyguan; formerly spoken in Australia. Warnman (Mardu, Nanidjara, Nanid-jara) [wbt]: Severely endangered, fewer than 30 speakers; Pama-Nyungan; spoken in Australia. Warray (Awarai, Awarra, Awarrai) [wrz]: Critically endangered, four speakers; Gunwinyguan; spoken in Australia. Warrgamay (Warakamai, Wargamay, Waragamai) [wgy]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Warriyangga (Wariyangga, Warriyangka, Wariengga) [wri]: Dormant; Pama-Nyungan; formerly spoken in Australia. Warrnambool (Dhauwurd Wurrung, Gunditjmara, Dhauhurtwurru) [gjm]: Dormant; PamaNyungan; formerly spoken in Australia. Warrwa (Warwa, Warrawai, Wa:rwa) [wwr]: Critically endangered, two speakers; Nyulnyulan; spoken in Australia. Warumungu (Waramanga, Warramunga, Aira-manga) [wrm]: Severely endangered, 50 speakers; Pama-Nyungan; spoken in Australia. 123

Anna Belew et al.

Wathawurrung (Wudjawuru, Wadawio, Wadourer) [wth]: Awakening; Pama-Nyungan; spoken in Australia. Wemba-Wemba (Wambawamba, Wambo Wambo, Wemba) [xww]: Awakening; PamaNyungan; spoken in Australia. Wergaia (Meintangk, Meintank, Painbali) [weg]: Critically endangered, five speakers; PamaNyungan; spoken in Australia. Western Arrarnte (Western Arrarnta) [are]: Threatened, 1,500 speakers; Pama-Nyungan; spoken in Australia. Wik-Epa (Wikepa, Wik Epa, Wik-Eppa) [wie]: Critically endangered, three speakers; PamaNyungan; spoken in Australia. Wik-Iiyanh (Wik-Iyanh, Wikianji, Wikianyi) [wij]: Severely endangered, fewer than 40 speakers; Pama-Nyungan; spoken in Australia. Wik-Keyangan (Wik-Keyenganh) [wif]: Critically endangered, three speakers; PamaNyungan; spoken in Australia. Wik-Me’anha (Wik-Em’an, Wik, Wikmean) [wih]: Severely endangered, fewer than 12 speakers; Pama-Nyungan; spoken in Australia. Wik-Mungkan (Wik-Munkan, Wik-Mungkhn, Munkan) [wim]: Threatened, 1,060 speakers; Pama-Nyungan; spoken in Australia. Wik-Ngathana (Wikngatara, Wik Ngathara, Wik Ngathana) [wig]: Endangered, approximately 100 speakers; Pama-Nyungan; spoken in Australia. Wik-Ngatharr (Wik-Ngathara, Wik-Ngatharra, Wik-Ngathrr) [wik]: Severely endangered, fewer than 86 speakers; Pama-Nyungan; spoken in Australia. Wilawila (Wila-Wila, Northern Language, Walmbi) [wil]: Critically endangered, two speakers; Worrorran; spoken in Australia. Wiradjuri (Wuradjeri, Wordjerg, Woradjerg) [wrh]: Critically endangered, three speakers; Pama-Nyungan; spoken in Australia. Wirangu (Nhawu, Njangga, Nyangga) [wgu]: Awakening; Pama-Nyungan; spoken in Australia. Woiwurrung (Bayry-Binip, Coraloon, Djadjuwuru) [wyi]: Awakening; Pama-Nyungan; spoken in Australia. Worla (Ola, Walajangarri, Wula) [no ISO 639-3 code]: Critically endangered, five speakers; Worrorran; spoken in Australia. Worrorra (Worora, Worrora, Wurora) [wro]: Dormant; Worrorran; formerly spoken in Australia. Wotjobaluk (Guli-ballaiuk, Watjo, Wattyabullak) [xwt]: Awakening; Pama-Nyungan; spoken in Australia. Wubuy (Nunggubuyu, Nangabuya, Ningburia) [nuy]: Endangered, 300 speakers; Gunwinyguan; spoken in Australia. Wudjari (Widjara, Warangoo, Warranger) [no ISO 639-3 code]: Critically endangered, approximately one speaker; Pama-Nyungan; spoken in Australia. Wulna (Woolna, Beriguruk, Berrigurruk) [wux]: Critically endangered, one speaker; Limilngan; spoken in Australia. Wunambal (Unambal, Wunambul, Northern Language) [wub]: Severely endangered, fewer than 20 speakers; Worrorran; spoken in Australia. Yabula-Yabula (Jabulajabula, Narinari, Yabala-Yabala) [yxy]: Awakening; Pama-Nyungan; spoken in Australia. Yagara (Yuggera, Jagara, Jagarabal) [yxg]: Awakening; Pama-Nyungan; spoken in Australia. Yalarnnga (Jalanga, Wonganja, Yalarrnga) [ylr]: Dormant; Pama-Nyungan; formerly spoken in Australia. 124

The world’s endangered languages

Yan-nhangu (Yarnango, Yan-Nhangu, Nangu) [jay]: Endangered, 265 speakers; PamaNyungan; spoken in Australia. Yanda (Janda, Yunda) [yda]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yandruwandha (Endawarra, Innamouka, Jandruwanta) [ynd]: Critically endangered, two speakers; Pama-Nyungan; spoken in Australia. Yangkaal (Jangga, Jang-Kala, Janjula) [nny]: Critically endangered, one speaker; Tangkic; spoken in Australia. Yangman (Jungman, Dagoman, Yungman) [jng]: Critically endangered, fewer than ten speakers; Yangmanic; spoken in Australia. Yankunytjatjara (Alinjera, Ankundjara, Everard Range) [kdd]: Critically endangered, 50–70 speakers; Pama-Nyungan; spoken in Australia. Yanyuwa (Aniula, Anjula, Anula) [jao]: Severely endangered, 52 speakers; Pama-Nyungan; spoken in Australia. Yaraldi (Jarildekald, Yarrildie, Yaralde) [no ISO 639-3 code]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yardliyawarra (Yardliwarra, Yadliura, Arabatura) [yxl]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yarluyandi (Jeljujendi, Yelyuyendi, Yarleeyandee) [yry]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yawarrawarrka (Yawarrawarka, Jauraworka, Jaurorka) [yww]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. Yawijibaya (Jawdjobara, Yaudjibara, Yawjibarra) [jbw]: Dormant; Worrorran; formerly spoken in Australia. Yawuru (Yaoro, Jaudjibara, Jawdjibaia) [ywr, dyd]: Critically endangered, fewer than five speakers; Nyulnyulan; spoken in Australia. Yidiny (Yidini, Idinji, Boolboora) [yii]: Severely endangered, 12 speakers; Pama-Nyungan; spoken in Australia. Yiiji (Jeidji, Yeidji, Northern Language) [no ISO 639-3 code]: Severely endangered, fewer than 50 speakers; Worrorran; spoken in Australia. Yindjibarndi (Binjiebandie, Indjiban, Indjibandje) [yij]: Severely endangered, 600–700 speakers; Pama-Nyungan; spoken in Australia. Yindjilandji (Bularnu, Dhidhanu, Injdjiladji) [yil]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Yingkarta (Yinggarda, Ingarda, Ingada) [yia]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yinhawangka (Inawangga, Inawongga, Ngaunmardi. Nalawonga) [ywg]: Critically endangered; Pama-Nyungan; spoken in Australia. Yir-Yoront (Jir Joront, Yirʔ Yoront, Kokomindjin) [yiy]: Severely endangered, fewer than 15 speakers; Pama-Nyungan; spoken in Australia. Yitha-Yitha (Yitha Yitha, Eetha-eetha, Eethee Eethee) [xth]: Dormant; Pama-Nyungan; formerly spoken in Australia. Yorta Yorta (Arramouro, Boonegatha, Echuca) [xyy]: Awakening; Pama-Nyungan; spoken in Australia. Yugambal (Jugumbir, Jukamba, Jukambal) [yub]: Critically endangered, one speaker; PamaNyungan; spoken in Australia. Yulparija (Yulparrija, Yilparitja, Yulbaridja) [no ISO 639-3 code]: Severely endangered, approximately 100 speakers; Pama-Nyungan; spoken in Australia. Yuwaalaraay (Yualyai, Ualari, Juwalarai) [no ISO 639-3 code]: Critically endangered, one speaker; Pama-Nyungan; spoken in Australia. 125

Anna Belew et al.

The Caucasus Bryn Hauk Linguistic diversity in the Caucasus The Caucasus region, defined by the Caucasus mountain range and flanked by the Black Sea and Caspian Sea, includes the entire territory of three nations, Armenia, Azerbaijan, and Georgia, as well as an area of Russia comprising ten federal subjects: the republics of Adygea, Chechnya, Dagestan, Ingushetia, Kabardino-Balkaria, Kalmykia, Karachay-Cherkessia, and North OssetiaAlania, and the krais of Krasnodar and Stavropol. This relatively small area, somewhat larger than the state of California, is home to dramatic linguistic diversity. Ethnologue (Simons & Fennig 2017) lists a total of 58 living languages4 in the Caucasus, belonging to seven language families: Northeast Caucasian (NakhDagestanian; 29 languages), Indo-European (eight languages), Turkic (six languages), South Caucasian (Kartvelian) (five languages), Northwest Caucasian (Abkhaz-Adyghean) (four languages), Afro-Asiatic (two languages: Assyrian Neo-Aramaic and Bohtan Neo-Aramaic), and Mongolic (one language: Kalmyk-Oirat). Additionally, Ethnologue recognizes an Armenian-Romani mixed language, Lomavren, as well as two sign languages,5 Armenian Sign Language and Russian Sign Language (Simons & Fennig 2017). Of these language families, three are endemic to the Caucasus: Northeast Caucasian, Northwest Caucasian, and South Caucasian. Due to the sharing of some areal linguistic features and their geographic proximity, some attempts have been made to unite these families into a larger macro-family. However, as Nichols (1992) notes, the time depth on Northeast Caucasian alone is comparable to that of Indo-European, meaning that any hypothetical genetic relationship among these families is most likely unsupportable in terms of linguistic evidence, given that the large time depth that would have to be assumed is so long ago that these languages could have changed so much that truly inherited features, even if any were preserved, would be extremely difficult if not impossible to determine. Thus, the shared features perceived today are considered to be due to areal diffusion rather than evidence of a genetic relationship. Such areal features – some of which are even shared among Caucasian languages belonging to families not indigenous to the region – include large consonant inventories, small vowel inventories, ejective consonants, and ergativity (Comrie 2008). The 58 living languages of the Caucasus can also be broken down according to the political entities in which they are found today, although a large number of them are spoken across national borders. Ethnologue lists seven languages in Armenia, 18 in Azerbaijan, 19 in Georgia, and 36 in Caucasian Russia (Simons & Fennig 2017). Within the Caucasus, the areas of greatest linguistic diversity are western Dagestan and northern Azerbaijan. A little more than half of all languages in the Caucasus are presently endangered.

Causes of language endangerment in the Caucasus Although many Caucasian languages are spoken by just a few hundred or a few thousand speakers, speaker numbers alone cannot explain the growing threat of language endangerment in the Caucasus. For many of the smaller languages, historically spoken in small areas or even in single villages in the mountains, it is unlikely that speakers ever numbered more than a few thousands at most. Nevertheless, these languages persisted for centuries (or more likely, millennia) with relatively little threat of endangerment until relatively recent times (Gippert 126

The world’s endangered languages

2008). Rather, the main causes of language endangerment in the Caucasus are urbanization, modernization, and the lasting effects of political ideologies that served to undermine the unity of indigenous communities – factors which together have disrupted intergenerational transmission for the most vulnerable languages (Friedman 2010; Kazakevich & Kibrik 2007; Grenoble 2003). In early Caucasian history, geography correlated with the size and standing of speaker communities, such that highland languages most likely had larger numbers of speakers than they do today, and their languages tended to spread down the mountain to lowland villages. However, during the Little Ice Age (late middle ages through nineteenth century), lowland pastures and farms fared better than the highlands, where global cooling lead to transhumance among the highland populations. This reversed the dynamic of language contact, and it became the trend that highland communities knew the language of the lowlands, but not vice versa. Thus, in the last few centuries, languages have spread uphill, exerting pressure on highland languages (Nichols 1998, as cited in Friedman 2010). Following the Bolshevik revolution in 1917, the entire territory of the Caucasus came under Soviet rule, increasing linguistic pressures from Russian in a region previously only partially under Russian control. At this stage, the three titular languages of the Caucasus had had a long literary tradition, each written in a different script: Armenian in the Armenian alphabet, Azerbaijani in Arabic, and Georgian in mkhedruli, which had by then become the dominant orthography among three uniquely Georgian writing systems. Thus, the smallest Caucasian languages, which had persisted with low speaker numbers for many centuries, came under heavy pressure not only from Russian, but also from larger Caucasian languages, especially the three titular languages (Grenoble 2003). In the 1920s, when policy was informed under a rubric of indigenization, Soviet authorities surveyed the languages of the Caucasus and decided which languages to develop by adapting an orthography, preparing teaching materials, and establishing mother tongue primary and secondary schools. Many languages were not selected for development, ensuring their continued competition with neighboring Caucasian languages in addition to Russian. Well-intentioned initiatives to develop the indigenous Caucasian languages suffered from inconsistent execution of policies, most recognizably in terms of orthography development. This stems from a remarkably successful Soviet directive to increase literacy throughout the Soviet Union, which in early years was very low. However, some policies building toward this goal required languages to abandon old orthographies abruptly in favor of an entirely new script, in some cases multiple times for a single language. For instance, in the mid and late 1920s, several languages with a pre-Soviet tradition of using the Arabic script (e.g., Azerbaijani, Chechen, Avar), followed later by the Mongolicbased Kalmyk-Oirat in 1931, were abruptly converted to using Latin-based orthographies. In 1938, policy shifted again, favoring Russification (or, in Georgia, Georgianization) over indigenization. Orthographies were again replaced, this time by Cyrillic script. By this point, nearly all remaining mother-tongue school programs were closed down in favor of Russianlanguage schools. These sudden conversions of orthography served not only to cut off speakers of literary languages from a long textual tradition, but also to interrupt literacy in minority languages, compromising efforts to maintain the mother tongues of minority groups and ultimately promoting Russian and Georgian as the languages of literacy (Grenoble 2003; Kazakevich & Kibrik 2007). The subsequent period of Russification featured numerous policies that were destructive for native languages and communities. Starting in the 1940s, several ethnic minorities of the Caucasus were selected for mass exile in Central Asia, which had a devastating impact on 127

Anna Belew et al.

minority languages. Russians were also sent in large numbers to the territories of ethnolinguistic minorities, further disrupting these communities and promoting the use of Russian as a lingua franca. When exiled speakers were finally permitted to return to the Caucasus in the late 1950s and early 1960s, the influx of exiles again displaced other linguistic minorities who had remained in the homeland (Grenoble 2003; Kazakevich & Kibrik 2007). Overall, the policies of the Soviet Union were broadly inconsistent in their treatment of indigenous languages, ranging from officially supportive to deliberately repressive. This history serves as an important background to the language endangerment story of the Caucasus, which is today also subject to the same globalization pressures that affect languages throughout the connected world. Throughout the region, a breakdown of endogamy has also favored switching to a lingua franca over the minority languages (Comrie 2008). Many minority languages that were displaced by larger Caucasian languages, such as Chechen and Kabardian, have suffered a total disruption of literacy and intergenerational transmission, with younger generations shifting entirely to Russian or another official language. Languages with official status in the Caucasian Russian republics that also have educational programs remain mostly safe, but could soon be threatened by Russian or Georgian, as well as by a new interest in learning global languages such as English. Finally, disputed territory and violent conflicts also present challenges for language maintenance and development in this region (Friedman 2010).

The current status of endangered languages in the Caucasus As of April 4, 2017, the Catalogue of Endangered Languages identifies 32 languages in the Caucasus at some level of endangerment. Given the total count of 58 languages based on Ethnologue (Simons & Fennig 2017), 55% of the languages in the Caucasus are currently considered endangered to some degree, which is slightly higher than the global average of 46%. Of the 32 Caucasian languages under threat, nine are considered “Vulnerable,” 14 are “Threatened,” five are “Endangered,” three are “Severely Endangered,” and none is “Critically Endangered” by the most current available information. In terms of language families, some level of endangerment is ascribed to 23 Northeast Caucasian languages, three Indo-European languages, two South Caucasian languages, one Northwest Caucasian language, one Afro-Asiatic language, and one Mongolic language. The three endemic families face differing levels of endangerment as a whole family. Specifically, 79.3% of the 29 languages in the Northeast Caucasian family are endangered, which is an alarmingly high endangerment rate for the family. In comparison, 50% of the four South Caucasian languages and 25% of the four Northwest Caucasian languages are endangered. Below is an alphabetic listing of the languages of the Caucasus currently included in the Catalogue of Endangered Languages.

Endangered languages of the Caucasus Abaza (Abazintsy, Abazin, Ahuwa) [abq]: Vulnerable, 47,880 speakers; Northwest Caucasian; spoken in Russia. Aghul (Agul, агъул чӀал, Ağul ç’al) [agx]: Threatened, fewer than 18,000 speakers; Northeast Caucasian; spoken in Russia and Azerbaijan. Akhvakh (Axvax, Ашвaлъи мицIи, авахский язык) [akv]: Threatened, approximately 5,000 speakers; Northeast Caucasian; spoken in Russia. Andi (Qwannab, qwavannab micci, андийский язык) [ani]: Vulnerable, 10,000–20,000 speakers; Northeast Caucasian; spoken in Russia. 128

The world’s endangered languages

Archi (арчинский язык, аршаттен чIат, Archintsy) [aqc]: Threatened, 1,237 speakers; Northeast Caucasian; spoken in Russia. Bagvalal (Bagulal, Kvanadin, багвалинский язык) [kva]: Threatened, 4,000–6,000 speakers; Northeast Caucasian; spoken in Russia. Batsbi (Bats, Tsova-Tush, batsba motjiti) [bbl]: Severely endangered, 800 speakers; Northeast Caucasian; spoken in Georgia. Bezhta (Bezheta, Kapucha, бежитинский язык) [kap]: Threatened, 6,200 speakers; Northeast Caucasian; spoken in Russia. Bohtan Neo-Aramaic (Sôreth, Hértevin, севернобохтанский язык) [bhn, hrt]: Severely endangered, fewer than 500 speakers; Afro-Asiatic; spoken in Russia and Georgia. Botlikh (Botlix, Bujxałi mic’c’i, ботлихский язык) [bph]: Threatened, 5,500 speakers; Northeast Caucasian; spoken in Russia. Budukh (Budugh, Budux, budad mez) [bdk]: Threatened; Northeast Caucasian; spoken in Azerbaijan. Chamalal (Camalal, чамалинский язык, çamalaldub miçʿçʿ) [cji]: Threatened, 4,000–6,000 speakers; Northeast Caucasian; spoken in Russia. Ghodoberi (Godoberi, годоберинский язык, ღოდობერიული ენა) [gdo]: Endangered, 2,500–4,000 speakers; Northeast Caucasian; spoken in Russia. Hinukh (Hinuq, Hinux, гинухский язык) [gin]: Endangered, 200–500 speakers; Northeast Caucasian; spoken in Russia. Homshetsi (Hamshen; Homshetsma; Hemşince) [no ISO 639-3 code]: Vulnerable, “relatively large numbers” of speakers; Indo-European; spoken in Turkey, Russia, and Georgia. Hunzib (Gunzib, гунзибский язык, ჰუნზიბური ენა) [huz]: Endangered, 600–800 speakers; Northeast Caucasian; spoken in Russia. Judeo-Tat (Hebrew Tat, Jewish Tat, Juhuri) [jdt]: Vulnerable, approximately 25,000 speakers; Indo-European; spoken in Russia, Azerbaijan, and Israel. Kalmyk-Oirat (Kalmyk, West Mongolian, Xal’mg keln) [xal]: Threatened; Mongolic; spoken in Russia, China, and Mongolia. Karata (Karatai, каратинский язык, კარატაული ენა) [kpt]: Threatened, 5,000 speakers; Northeast Caucasian; spoken in Russia. Khinalugh (Xinalug, Xinaliq, Khinalug) [kjj]: Threatened, 3,000 speakers; Northeast Caucasian; spoken in Azerbaijan. Khvarshi (Khwarshi, Xvarshi, хваршинский язык) [khv]: Threatened, approximately 300 speakers; Northeast Caucasian; spoken in Russia. Kryts (Kryz, Kryc, Dzek) [kry]: Endangered, fewer than 5,000 speakers; Northeast Caucasian; spoken in Azerbaijan. Laz [lzz]: Vulnerable, 22,000–33,000 speakers; South Caucasian (a.k.a. Kartvelian); spoken in Turkey and Georgia. Muslim Tat (Tati, татский язык, თათური ენა) [ttt]: Severely endangered, approximately 5,000 speakers; Indo-European; spoken in Russia, Azerbaijan, and Iran. Noghay (Nogai, Noghai, ногайский язык) [nog]: Threatened, approximately 70,000 speakers; Turkic; spoken in Russia. Rutul (Rutuly, Mukhad, Mykhanidy) [rut]: Vulnerable, 19,000 speakers; Northeast Caucasian; spoken in Russia and Azerbaijan. Svan (lušnu nin სვანური ენა, ლუშნუ ნინ) [sva]: Vulnerable, 30,000 speakers; South Caucasian (a.k.a. Kartvelian); spoken in Georgia. Tabasaran (Tabassaran, табасаранский язык, თაბასარანული ენა) [tab]: Vulnerable, 93,630 speakers; Northeast Caucasian; spoken in Russia and Azerbaijan. 129

Anna Belew et al.

Tindi (Tindal, Tindin, тиндийский язык) [tin]: Threatened, 4,000–6,000 speakers; Northeast Caucasian; spoken in Russia. Tsakhur (Tsaxur, Caxur, цахурский язык) [tkr]: Vulnerable, approximately 19,000 speakers; Northeast Caucasian; spoken in Russia and Azerbaijan. Tsez (Dido, Cez, цезский язык) [ddo]: Threatened, 7,000–8,000 speakers; Northeast Caucasian; spoken in Russia. Udi (удин муз, удинский язык, უდიური ენა) [udi]: Endangered, approximately 5,500 speakers; Northeast Caucasian; spoken in Russia, Azerbaijan, and Georgia.

East Asia Yen-ling Chen Linguistic diversity in East Asia The six East Asian countries in the Catalogue of Endangered Languages are China, Japan, Mongolia, South Korea, and Taiwan.6 Most languages of Russia (those not spoken in border regions with China or Mongolia) and all languages of the Caucasus are treated in their own respective sections of this book. Ethnologue lists around 350 living languages (not necessarily endangered) in these five countries (Simons & Fennig 2017). The Catalogue of Endangered Languages lists an additional 27 languages not identified in Ethnologue; these languages have no ISO 639-3 codes, and for this reason the languages identified in the Catalogue of Endangered Languages do not necessarily have one-to-one correspondences to those listed in Ethnologue. There is then a total of c.370 living languages in East Asia. These languages are classified in 14 language families (including one isolate): Austroasiatic, Austronesian, Hmong-Mien (Miao-Yao), Japonic, Koreanic, Tai-Kadai (Kra-Dai), Mongolic, Sino-Tibetan, Tungusic, Turkic, Uralic, Indo-European (Sarikoli [srh] and Wakhi [wbl]), and Ainu [ain] (a language isolate), as well as a number of pidgins and creoles, mixed languages, and sign languages.7 This does not include the number of East Asian languages spoken in Russian territory and treated in the section on Russia. Clearly, East Asia is an area of rich linguistic diversity.

Causes of language endangerment in East Asia Political factors and population size have played an important role in language shift and endangerment in East Asia. Social status and economic pressures also correlate with language shift and endangerment here. National language policies tend to favor certain linguistic varieties, disadvantaging minority languages and increasing language shift. Because language policies promote the use of some languages, their speakers gain higher social status. This prestige impacts economic opportunities, with negative consequences for minority languages. Chinese culture has been influential in this region for centuries. For example, both Japanese and Korean have integrated numerous elements from the Chinese languages and the Chinese writing system as the result of long-term contact. (Languages of Southeast Asia have also been heavily influenced by Chinese.) However, Japanese and Korean have never been endangered, in spite of the Chinese influence. However, Ainu [ain] and the Ryukyuan languages of Okinawa in Japan, and Jejueo [jje] in Korea, are not this lucky – all are endangered. The Ainu and the Ryukyuans have de facto been administered by the Japanese government since the nineteenth century (Shibatani 1990). Japan’s “assimilation policy” (同化政策) promoting a standard 130

The world’s endangered languages

language and a standard writing system had a huge impact on these minorities. This policy fostered linguistic homogeneity in Japan, leading towards linguistic and cultural assimilation among minorities. The case of Jejueo in Korea is similar, where Korean and Jejueo are two separate languages, though the medium of instruction for Jeju children is Korean. Economic pressure also plays a role in language shift in Jeju (Sejung Yang personal communication 2016). As a famous Korean tourist spot, on Jeju Island ability to speak Korean is favored for good business. Shift from Jejueo to Korean is thus advanced. As mentioned earlier, Chinese culture has been dominant in East Asia for 2,000 years. None of the endangered languages in China or Taiwan belong to the Chinese (Sinitic) subfamily of Sino-Tibetan.8 The Han Chinese began immigrating to Taiwan (formerly known as Formosa) in the seventeenth century (Chou 2009). A majority of them spoke varieties of Southern Min and Hakka. By contrast, the indigenous peoples, who have been in Taiwan for c.6000 years,9 all spoke Austronesian languages. Chinese immigrants to Taiwan gradually came to outnumber the speakers of Austronesian languages. Riots between different ethnic and linguistic groups were not uncommon. Due to the Chinese expansion, the Formosan aboriginals, especially those of the plain such as the Siraya [fos] and the Kavalan [ckv], lost lands and were forced to live among the Han Chinese and to follow the rules established by the Chinese-speaking rulers. The second wave of assimilation took place during the Japanese occupation (1895–1945), during which the Japanese promoted “Japanization,” encouraging the Taiwanese people (regardless of their ethnicity) to adopt the Japanese language and culture. Moreover, after the defeat of the Chinese Nationalists (Kuomintang) by the Chinese Communists in 1949, the Nationalists fled to Taiwan from China. During 1949–1987, the Chinese Nationalists imposed a series of policies to “de-Japanize” the Taiwanese people, prohibiting the use of non-Mandarin languages. Given the socio-political status of the speakers of Austronesian languages, it is not surprising that in the past 300 years, they have gradually abandoned their heritage languages in favor of Chinese (Southern Min, Hakka, and Mandarin, especially Mandarin among younger generations born after the 1980s) or Japanese (prior to 1949). Since 2001, heritage language learning has been implemented in elementary education; however, it remains symbolic rather than effective, especially among diasporic groups. The influence of the Chinese languages and culture on the non-Chinese speaking population of China proper is well known, with Mandarin speakers politically dominant and making up the largest sector of the population. Janhunen (2005:43) has proposed the term “ethnic inflation,” which he describes as “a government tool for the rapid assimilation of the minorities under a minimal danger of ethnic conflict.” That is, the government of China welcomes the autonomy of officially recognized “nationalities” (ethnicities) and their ethnic equality, while at the same time expecting these ethnic groups to be “loyal to the majority language and culture.” Bradley (2005:5) mentions that “according to the law on national autonomy, all cadres, including members of the Han Chinese majority, are expected to study local language and writing of the area where they work.” Nevertheless, the need to preserve minority languages rarely caught the attention of local governments. Bradley has also pointed out that in some regions, local governments allowed people to change their registered nationality. However, the alteration of one’s nationality is never correlated with one’s ability to speak the language of their reacquired ethnicity. Moreover, in 2000 the government of China passed a language law that reaffirms the importance of Standard Mandarin and the use of simplified Chinese. In China, Manchu [mnc] serves as a counterexample to the assertion that languages of the politically dominant groups survive: “Manchu is notoriously one of those rare state languages 131

Anna Belew et al.

that have declined to the verge of extinction in spite of maximal opportunities of official support in the past” (Janhunen 1997:126). The Qing Dynasty (1644–1911) was the last dynasty in the Chinese history; it was ruled by the Manchu, a Tungusic language, unrelated to Chinese. Despite the Manchus having been the ruling class in pre-modern China, their language is now classified as “critically endangered” by the Catalogue of Endangered Languages. Janhunen (1997:126–127) ascribed the endangerment of Manchu to the Sinicization of the Manchu ruling class and to the acculturation of Manchuria due to the influx of the Han Chinese.10 He mentioned that Manchu survives in the twenty-first century only among the Xibe [sjo], a diaspora group of Manchu living in Xinjiang/East Turkestan since 1763. Mongolia is a relatively young country, politically. In the past three centuries, both China and Russia have had great impact on its people. After the establishment of Mongolia, the government promoted Khalkha Mongolian as the standard and prestigious variety. Endangered languages in Mongolia tend to represented in groups that have smaller total populations, and the speakers of these languages have never gained political power. According to Janhunen (1997; 2005), following the changing relationships between China and Russia, forced relocations of these linguistic minorities were not unheard of, as illustrated by the cases of Dagur [dta] and Khamnigan Mongol. These groups broke up into smaller communities after relocation. Janhunen (personal communication 2016) mentions that because most minority languages in Mongolia are closely related to Khalkha Mongolian, “language shift takes place via diglossia.” He also points out that Written Mongol leads to the general impression that there is only one Mongolian language in the country. As for speakers of non-Mongolic languages, they have been highly assimilated by the Khalkha Mongolians, due to their small size.

The current status of endangered languages in East Asia The Catalogue of Endangered Languages registers 139 endangered languages in China, Japan, Mongolia, South Korea, and Taiwan. These include one language isolate (Ainu [ain]), one sign language (Hong Kong Sign Language [hks]), two creoles (Yilan Creole [no ISO 639-3] and Patuá [mzs]), and one mixed language (Wutunhua [wuh]). Of the 139 endangered languages, none is dormant and only one language (Siraya [fos]) is classified as “Awakening.” There are 21 languages that fall into the “Critically Endangered” category, and another 23 are rated as “Severely Endangered.” In addition, 30 languages are identified as “Endangered” (the level of vitality between “Severely Endangered” and “Threatened”) in the Catalogue, 39 as “Threatened,” 23 as “Vulnerable,” and two as “At Risk.” Of the c.370 living languages in East Asia, 38.6% of them are endangered and are thus included in the Catalogue of Endangered Languages. Considering its landmass and population, it is unsurprising that China is home to the highest number of endangered languages of the East Asia region, most of which are scattered in southwestern China, known for its linguistic and cultural diversity. A majority of these endangered languages in China belong to the Sino-Tibetan language family, although none of them falls on the Sinitic side. The country with the second highest number of endangered languages is Taiwan, with 12 endangered Austronesian languages and one Japanese-based creole. Japan has seven. By contrast, South Korea has the lowest number of endangered languages, with only one, Jejueo, listed as “Endangered.” Below is an alphabetical list of all languages of East Asia included in the Catalogue of Endangered Languages.

132

The world’s endangered languages

Endangered languages of East Asia Ainu (China) (Aynu, 艾努語) [aib]: Threatened, fewer than 12,000 speakers; Turkic; spoken in China. Ainu (Japan) (Ainu Itak, アイヌ イタㇰ) [ain]: Critically endangered, two speakers; isolate; spoken in Japan. Akeu (Aki, Akui, 阿克) [aeu]: Threatened, approximately 5,000 speakers; Sino-Tibetan; spoken in China, Myanmar, and Thailand. Aluo (阿羅語) [yna]: Endangered, fewer than 40,000 speakers; Sino-Tibetan; spoken in China. Amami-North Okinawan (奄美方言, 国頭方言) [kzg, xug, ryn, okn, ams, tkn, yox]: Endangered, approximately 100,000 speakers; Japonic; spoken in Japan. Ayizi (阿夷子語) [yyz]: Critically endangered; Sino-Tibetan; spoken in China. Baha Buyang (Paha Buyang, 巴哈布央語) [yha]: Severely endangered; Tai-Kadai; spoken in China. Baima (Bai Ma, Pe, 白馬語) [bqh]: Threatened, 10,000 speakers; Sino-Tibetan; spoken in China. Bolyu (倈語, Palyu) [ply]: Severely endangered, approximately 500 speakers; Austro-Asiatic; spoken in China. Bugan (布賡語, 布甘語) [bbh]: Vulnerable, approximately 2,700 speakers; Austro-Asiatic; spoken in China. Bumang (布芒語, 曼仗傣) [bvp]: Threatened, approximately 200 speakers; Austro-Asiatic; spoken in China. Buyuan Jinuo (基諾語補遠方言) [jiy]: Endangered, approximately 1,000 speakers; SinoTibetan; spoken in China. Central Lalo (中臘羅語) [ywt]: Vulnerable, 300,000 speakers; Sino-Tibetan; spoken in China. Chesu [ych] (車蘇): Endangered, approximately 3,300 speakers; Sino-Tibetan; spoken in China. Choni (Cone, 卓尼話) [cda]: At risk, 154,000 speakers; Sino-Tibetan; spoken in China. Dagur (Daur, 達斡爾) [dta]: Endangered, at least 50,000 speakers; Mongolic; spoken in China and Mongolia. Darkhat (達爾扈特) [drh (retired)]: Vulnerable, 20,400 speakers; Mongolic; spoken in Mongolia. Dongxiang (Santa, 東鄉語) [sce]: Vulnerable, 200,000 speakers; Mongolic; spoken in China. Drung (T’rung, Rawang, 獨龍語) [duu]: Vulnerable, approximately 7,000 speakers; SinoTibetan; spoken in China. Dukha (Tuha) [no ISO 639-3 code]: Endangered, 100–200 speakers; Turkic; spoken in Mongolia. Duoxu (多緒, 多須) [no ISO 639-3 code]: Critically endangered, nine speakers; Sino-Tibetan; spoken in China. E’ma Buyang (峨馬布央) [yzg]: Endangered, 300 speakers; Tai-Kadai; spoken in China. East Yugur (東部裕固語, Shira Yughur) [yuy]: Vulnerable, 4,000 speakers; Mongolic; spoken in China. Ersu (爾蘇語) [ers]: Threatened, approximately 13,000 speakers; Sino-Tibetan; spoken in China. Gansu Bonan (甘肅保安語, 積石山保安語) [peh]: Threatened, 10,000–15,000 speakers; Mongolic; spoken in China. Gao Gelao (Gao, 仡佬語稿方言) [gqu]: Severely endangered, fewer than 6,000 speakers; TaiKadai; spoken in China. Green Gelao (仡佬語哈給方言, Hoki Gelao, 青仡佬) [giq]: Severely endangered, fewer than 6,000 speakers; Tai-Kadai; spoken in China and Vietnam. 133

Anna Belew et al.

Guiqiong (Guichong, 貴瓊) [gqi]: Endangered, 9,677 speakers; Sino-Tibetan; spoken in China. Hachijo (八丈方言, 青ヶ島方言) [no ISO 639-3 code]: Endangered, fewer than 8,000 speakers; Japonic; spoken in Japan. Hlersu (Shansu, Sansu, 山蘇) [hle]: Vulnerable, approximately 15,000 speakers; Sino-Tibetan; spoken in China. Hong Kong Sign Language (Heung Kong Sau Yue, 香港手語) [hks]: Threatened, 9,000 speakers; sign language; spoken in China. Horpa (Ergong, 爾龔語; 道孚語) [ero]: Vulnerable, 45,000 speakers; Sino-Tibetan; spoken in China. Hu (Angku, 戶語) [huo]: Endangered, approximately 700 speakers; Austro-Asiatic; spoken in China. Huzhu (Mongghul, 互助土家語) [no ISO 639-3 code]: Endangered, 50,000 speakers; Mongolic; spoken in China. Ili Turki (T’urk, Tuerke, 伊犁突厥語) [ili]: Critically endangered; Turkic; spoken in China. Jejueo (제주어, 濟州語) [jje]: Endangered, 5,000–10,000 speakers; Koreanic; spoken in South Korea. Jiongnai Bunu (炯奈語, 花籃瑤) [pnu]: Threatened; Hmong-Mien; spoken in China. Kaduo (卡多話) [ktp]: Vulnerable, approximately 200,000 speakers; Sino-Tibetan; spoken in China. Kanakanavu (Kanakanabu, 卡那卡那富語) [xnb]: Critically endangered, six speakers; Austronesian; spoken in Taiwan. Kangjia (康家語) [kxs]: Severely endangered, 377 speakers; Mongolic; spoken in China. Kathu (Gasu) [ykt]: Threatened, 5,000 speakers; Sino-Tibetan; spoken in China. Kavalan (噶瑪蘭語) [ckv]: Critically endangered; Austronesian; spoken in Taiwan. Kazhuo (卡卓語) [no ISO 639-3 code]: Threatened, fewer than 6,341 speakers; Sino-Tibetan; spoken in China. Khamnigan Ewenki (Khamnigan Evenki, Khamnigan Tungus, 通古斯鄂溫克) [no ISO 639-3 code]: Threatened, approximately 1,000 speakers; Tungusic; spoken in Russia, China, and Mongolia. Khamnigan Mongol (Khamnigan, 哈姆尼蒙古語, хамниганский язык) [no ISO 639-3 code]: Severely endangered, approximately 300 speakers; Mongolic; spoken in Russia, China, and Mongolia. Kilen (Sungari Nanay, килен, 赫哲語奇楞方言) [no ISO 639-3 code]: Severely endangered, 20 speakers; Tungusic; spoken in Russia and China. Kuanhua (寬話, 寬語) [xnh]: Threatened, 1,000 speakers; Austro-Asiatic; spoken in China. Kucong (Cosung, 苦聰話) [lkc]: Vulnerable, approximately 45,000 speakers; Sino-Tibetan; spoken in China and Vietnam. Laji (Lachi) [no ISO 639-3 code]: Critically endangered, 200 speakers; Tai-Kadai; spoken in China. Lakkia (Lakkja, 茶山瑤) [lbc]: Threatened, 9,000 speakers; Tai-Kadai; spoken in China. Lama (Laemae, Northern Bai) [bfc]: Endangered, 15,000 speakers; Sino-Tibetan; spoken in China. Lamu [llh]: Endangered, 120 speakers; Sino-Tibetan; spoken in China. Langnian Buyang (郎念布央) [yln]: Severely endangered, 300 speakers; Tai-Kadai; spoken in China. Laomian (老緬語) [lwm]: Threatened; Sino-Tibetan; spoken in China and Myanmar. 134

The world’s endangered languages

Lizu (里汝語) [no ISO 639-3 code]: Endangered, approximately 4,000 speakers; Sino-Tibetan; spoken in China. Lopnor Uighur (Lopluk, 維吾爾語羅布話) [no ISO 639-3 code]: Critically endangered, fewer than 24,000 speakers; Turkic; spoken in China. Mak (莫話, 莫語) [mkg]: Endangered, 5,000 speakers; Tai-Kadai; spoken in China. Man Met (Manmit, 克蔑語) [mml]: Threatened, approximately 1,000 speakers; AustroAsiatic; spoken in China. Manchu (滿語) [mnc]: Critically endangered, fewer than 20 speakers; Tungusic; spoken in China. Manchurian Kirghiz (富裕吉爾吉斯語; 滿洲柯爾克孜語) [no ISO 639-3 code]: Critically endangered, fewer than ten speakers; Turkic; spoken in China. Maonan (毛南話, 毛難語) [mmd]: Threatened, 70,000 speakers; Tai-Kadai; spoken in China. Minhe Monguor (Minhe Mangghuer, 民和土家語) [no ISO 639-3 code]: Vulnerable, 25,000 speakers; Mongolic; spoken in China. Miqie Yi (Micha, 密岔) [yiq]: Threatened, 8,000 speakers; Sino-Tibetan; spoken in China. Miyako (Southern Ryukyuan, 宮古方言) [mvi]: Severely endangered, 10,000–15,000 speakers; Japonic; spoken in Japan. Mo’ang (Maang, 麼昂語, 末昂話) [no ISO 639-3 code]: Threatened, at least 5,000 speakers; Sino-Tibetan; spoken in China and Vietnam. Moji (Flathead Phulai, Muji, Phula) [ymi]: Vulnerable, at least 400 speakers; Sino-Tibetan; spoken in China. Muda (Muta) [ymd]: Threatened, 2,000 speakers; Sino-Tibetan; spoken in China. Mulam (仫佬語) [mlm]: Endangered, 30,000 speakers; Tai-Kadai; spoken in China. Mulao (木佬語) [giu]: Critically endangered, a few speakers; Tai-Kadai; spoken in China. Muya (Minyag, 木雅語) [mvm]: Vulnerable, fewer than 15,000 speakers; Sino-Tibetan; spoken in China. Muzi (Muji) [ymz]: Threatened, 10,000 speakers; Sino-Tibetan; spoken in China. Nadouhua (那斗話, 來語) [no ISO 639-3 code]: Severely endangered; Tai-Kadai; spoken in China. Naluo (那羅, 巧武土語) [ylo]: Vulnerable, at least 15,000 speakers; Sino-Tibetan; spoken in China. Namuyi (Namuzi, 納木依, 納木茲) [nmy]: Endangered, 5,000 speakers; Sino-Tibetan; spoken in China. Nanai (赫哲語, Нанай) [gld]: Severely endangered, approximately 180 speakers; Tungusic; spoken in Russia and China. New Bargut (新巴爾虎土語) [no ISO 639-3 code]: Vulnerable; Mongolic; spoken in China. Northern Muji (Bokha, Phula) [ymx]: Endangered, 9,000 speakers; Sino-Tibetan; spoken in China. Northern Pumi (Prinmi, Northern Prinmi, 北普米語) [pmi]: Vulnerable, 32,300 speakers; Sino-Tibetan; spoken in China. Northern Qiang (羌語北部方言) [cng]: Threatened; Sino-Tibetan; spoken in China. Northern Tujia (北部土家語) [tji]: Severely endangered, approximately 65,000 speakers; SinoTibetan; spoken in China. Nyagrong Minyag (Minyag, 新龍木雅語) [no ISO 639-3 code]: Threatened, fewer than 1,000 speakers; Sino-Tibetan; spoken in China. Old Bargut (陳巴爾虎土語) [no ISO 639-3 code]: Threatened; Mongolic; spoken in China. Ongkor Solon (Onkor Solon, Turkestan Solon) [no ISO 639-3 code]: Critically endangered, approximately one speaker; Tungusic; spoken in Kazakhstan and China. 135

Anna Belew et al.

Ordos (Urdus, 鄂爾多斯) [no ISO 639-3 code]: Vulnerable, fewer than 100,000 speakers; Mongolic; spoken in China. Oroqen (Orochen, 鄂倫春語) [orh]: Severely endangered, fewer than 2,500 speakers; Tungusic; spoken in China. Patuá (Macao Creole Portuguese, Macaense, 澳門土生葡語) [mzs]: Critically endangered, approximately 50 speakers; pidgin or creole; spoken in China. Pazeh-Kaxabu (巴宰語, 巴則海語, 噶哈巫語) [uun]: Critically endangered, 12 speakers; Austronesian; spoken in Taiwan. Pela (Bola, 波拉語) [bxd]: Vulnerable, approximately 600 speakers; Sino-Tibetan; spoken in China and Myanmar. Phana’ (Pana, Bana, Phana) [phq]: Endangered; Sino-Tibetan; spoken in China and Laos. Puyuma (卑南語) [pyu]: Severely endangered, approximately 1,000 speakers; Austronesian; spoken in Taiwan. Qila Muji (Phula, Muji) [ymq]: Vulnerable, 1,500 speakers; Sino-Tibetan; spoken in China. Qinghai Bonan (Bao’an Tu, 青海保安語) [peh]: Threatened, 4,000 speakers; Mongolic; spoken in China. Queyu (Choyo, 卻隅話) [qvy]: Threatened, 7,000 speakers; Sino-Tibetan; spoken in China. Red Gelao (Voa Dê, A-ou, 紅仡佬) [gir]: Critically endangered, fewer than 6,000 speakers; Tai-Kadai; spoken in China and Vietnam. Saaroa (La’alua, 拉阿魯哇語) [sxr]: Critically endangered, ten speakers; Austronesian; spoken in Taiwan. Sadu (撒都語) [no ISO 639-3 code]: Endangered, fewer than 1,505 speakers; Sino-Tibetan; spoken in China. Saisiyat (賽夏語) [xsy]: Severely endangered, approximately 3,000 speakers; Austronesian; spoken in Taiwan. Sakizaya (撒奇萊雅語, 南勢阿美語) [ais]: Endangered, approximately 500–600 speakers; Austronesian; spoken in Taiwan. Salar (撒拉語, саларский) [slr]: Threatened, 95,000 speakers; Turkic; spoken in China. Samatao (Samaduo, 撒慕, Eastern Samadu) [ysd]: Critically endangered, approximately 400 speakers; Sino-Tibetan; spoken in China. Samatu (Samadu) [no ISO 639-3 code]: Critically endangered, fewer than 7,500 speakers; SinoTibetan; spoken in China. Samei [smh]: Threatened, approximately 18,667 speakers; Sino-Tibetan; spoken in China. Sangkong (桑孔話, 布下土語) [sgk]: Threatened, fewer than 2,000 speakers; Sino-Tibetan; spoken in China. Sanie [ysy]: Severely endangered, approximately 8,000 speakers; Sino-Tibetan; spoken in China. Sarikoli (中國塔吉克語, 色勒庫爾語) [srh]: Vulnerable, 16,000 speakers; Indo-European; spoken in China. Seediq (Truku, 賽德克語, 太魯閣語) [trv]: Critically endangered; Austronesian; spoken in Taiwan. She (畬語) [shx]: Severely endangered, 1,000 speakers; Hmong-Mien; spoken in China. Shixing (Xumi, 史興話, 旭米語) [sxg]: Threatened, approximately 1,000 speakers; SinoTibetan; spoken in China. Sinkiang Dagur (Turkestan Dagur, Daur, 新疆達斡爾) [no ISO 639-3 code]: Threatened; Mongolic; spoken in Kazakhstan and China. Siraya (西拉雅語) [fos]: Awakening; Austronesian; spoken in Taiwan. Solon (Manchurian Solon, 鄂溫克語輝方言, 索倫鄂溫克) [no ISO 639-3 code]: Threatened, at least 15,000 speakers; Tungusic; spoken in China. 136

The world’s endangered languages

South-Central Okinawan (シ マ ク ト ゥ バ , 沖 縄 語 , Uchinaaguchi) [ryu]: Severely endangered, 95,000 speakers; Japonic; spoken in Japan. Southern Muji (Phula) [ymc]: Vulnerable, 26,000 speakers; Sino-Tibetan; spoken in China. Southern Pumi (Prinmi, 南普米語) [pmj]: Vulnerable, 60,000 speakers; Sino-Tibetan; spoken in China. Southern Qiang (羌語南部方言) [qxs]: Threatened; Sino-Tibetan; spoken in China. Southern Tujia (南部土家語) [tjs]: Severely endangered, 2,000 speakers; Sino-Tibetan; spoken in China. T’en (Then, Yanghuang, 佯僙語) [tct]: Threatened, 20,000 speakers; Tai-Kadai; spoken in China. Talu (他留話) [yta]: Vulnerable, 10,138 speakers; Sino-Tibetan; spoken in China. Tanglang (堂郎語) [ytl]: Threatened, 947 speakers; Sino-Tibetan; spoken in China. Tao (Yami, 雅美語, 達悟語) [tao]: Severely endangered, approximately 3,000 speakers; Austronesian; spoken in Taiwan. Tazhi [no ISO 639-3 code]: Endangered; Sino-Tibetan; spoken in China. Thao (邵語) [ssf]: Critically endangered, four speakers; Austronesian; spoken in Taiwan. Tsat (Utsat, Hainan Cham, 回輝話) [huq]: Threatened, fewer than 4,000 speakers; Austronesian; spoken in China. Tsou (鄒語) [tsu]: Endangered, approximately 1,000 speakers; Austronesian; spoken in Taiwan. Tuha (Uighur Uryangkhai, Khövsgöl Uryangkhai) [no ISO 639-3 code]: Critically endangered, fewer than 20 speakers; Turkic; spoken in Mongolia. U (烏語) [uuu]: Threatened, 40,000 speakers; Austro-Asiatic; spoken in China. Wa (佤語, Va, 佤語) [wbm]: At risk; Austro-Asiatic; spoken in China and Myanmar. White Gelao (仡佬語多羅方言, 白仡佬) [giw]: Severely endangered, fewer than 6,000 speakers; Tai-Kadai; spoken in China and Vietnam. Wunai Bunu (布努語唔奈方言) [bwn]: Threatened; Hmong-Mien; spoken in China. Wutunhua (Wutun, 五屯話) [wuh]: Endangered, 4,000 speakers; mixed language; spoken in China. Xiandao (仙島話) [acn]: Endangered, 40 speakers; Sino-Tibetan; spoken in China. Xibe (Sibo, 錫伯語) [sjo]: Vulnerable, approximately 27,000 speakers; Tungusic; spoken in China. Yaeyama (八重山語) [rys]: Endangered, a few thousand speakers; Japonic; spoken in Japan. Yellow Uyghur (Yellow Uighur, West Yugur, 西部裕固語) [ybe]: Threatened, 4,600 speakers; Turkic; spoken in China. Yerong (雅郎布央, 耶容布央, 土瑤) [yrn]: Endangered, at least 300 speakers; Tai-Kadai; spoken in China. Yilan Creole (寒溪話, 宜蘭克里奧爾) [no ISO 639-3 code]: Endangered, approximately 3,000 speakers; pidgin or creole; spoken in Taiwan. Yonaguni (与那国語) [yoi]: Severely endangered, fewer than 1,000 speakers; Japonic; spoken in Japan. Youle Jinuo (攸樂基諾語) [jiu]: Severely endangered, fewer than 10,450 speakers; SinoTibetan; spoken in China. Younuo Bunu (優諾語) [buh]: Threatened; Hmong-Mien; spoken in China. Zauzou (柔若語) [zal]: Endangered, fewer than 2,100 speakers; Sino-Tibetan; spoken in China. Zhaba (Zaba, nDrapa, 扎巴語) [zhb]: Threatened, fewer than 7,700 speakers; Sino-Tibetan; spoken in China.

137

Anna Belew et al.

Europe Bryn Hauk On the macro-region of Europe and Russia outside the Caucasus In this section, the status of the endangered languages of Europe and Russia is presented. While, this section may be conceived of being principally about Europe, to which Russia belongs, considerable space is dedicated to delineating the territory of Russia and its languages that are involved. This is hardly surprising, given the number of endangered languages located in Russia. For the purpose of this summary, Europe includes Russia, except Russia in this discussion excludes Russian territory involving languages of the Caucasus. The federal subjects of Caucasian Russia11 are discussed instead together with the rest of the Caucasus region in the section on the Caucasus. Throughout this subsection, all mentions of Russia refer specifically to Russia outside the Caucasus. It remains somewhat incongruous to treat Europe and Russia together as a single region for any linguistic purposes. Spanning 12 time zones, Russia’s territory is nearly as great in eastto-west expansion as the Austronesian language territory, but it lacks compelling common linguistic or cultural traits like the family relationship that unites peoples from Madagascar to Rapa Nui. However, it is common practice to treat Russia (or at least European Russia) as part of Europe, owing in large part to the fact that, although only a quarter of Russia’s territory falls geographically within Europe, roughly three quarters of its population lives in that region.12 Furthermore, throughout Russia, the present-day crisis of language endangerment is the result of the same political history that has endangered minority languages in the European former Soviet republics. Thus, it is recognized that treating Europe and Russia together as a macroregion has very little linguistic justification, but is rather a conventional convenience.

Linguistic diversity in Europe Given the territory as delineated here, another issue must be addressed before we attempt to characterize the region’s linguistic diversity. In Europe more so than in other regions, the decision to consider some entity a separate language as opposed to a dialect of some language or to consider an entity merely a dialect of some national language in a number of cases has been determined on the basis of non-linguistic factors. Some nations have (or have had in the past) policies favoring a single national language for the sake of cultural unity or for the promotion of a single cultural identity to distinguish themselves from their neighbors. In such cases, separate but related languages within the national borders have been labeled as “dialects” or “patois” rather than independent languages, as is the case with Tsakonian in Greece, for instance. On the other hand, some entities have been designated separate languages for nationalistic motives though they may be mutually intelligible and therefore on linguistic criteria alone are not truly separate languages, as in the case of Swedish and Norwegian, of Serbian and Croatian. In Europe it has also frequently been the case that speaker communities separate spoken languages nevertheless use a single literary language. For instance, Occitan has traditionally served as the literary language for the several Occitan-Romance languages in France, and as a result, these varieties in the present day are often treated as dialects of Occitan, in spite of their distinctive differences (Salminen 2007a). The opposite case also occurs, where multiple literary traditions are used for a single language, as for instance, with Serbian and Croatian (formerly 138

The world’s endangered languages

officially considered a single language, Serbo-Croatian) and in the case of Norwegian, with Nynorsk along side Bokmål. Such multiple literary traditions tend to favor viewing entities as different languages, in cases that should perhaps better be treated as involving dialectal differences. For these reasons, it is challenging to count accurately the number of languages in Europe and thereby faithfully characterize linguistic diversity in the region. Since opinions will vary in these cases, the numbers in this section should be considered relative, not concrete counts. Based on a count of languages from the Ethnologue (Simons & Fennig 2017), the total number of living languages in Europe and Russia is 266. Of these, 61 are spoken in Russian outside of the Caucasus, leaving 205 in Europe west of Russia. These 266 languages belong to 13 different language families: Indo-European (127), Uralic (44), Turkic (19), Tungusic (8), Chukotko-Kamchatkan (5), Eskimo-Aleut (3), Mongolic (2), Yeniseian (2), Yukaghir (2), Afro-Asiatic (2: Maltese and Cypriot Spoken Arabic), Gilyak (Nivkh) (disputed as either a language isolate or as a small family of two languages: Amur Nivkh and Sakhalin Nivkh), and one language isolate, Basque. There are also 41 sign languages in the region (Simons & Fennig 2017). All members of four of these language families are spoken entirely in the territory of Russia: Chukotko-Kamchatkan, Yeniseian, Yukaghir, and Gilyak (Nivkh) (if it is a separate family and not a language isolate), plus the language isolate Ainu. The Uralic family is located entirely within Europe and Russia, and the Tungusic family is split only between Russia and China. Given the physical expansiveness of the territory covered in this section, the overall linguistic diversity level seems to be low. In fact, Juaristi et al. (2008) estimate that, while 7% of the world’s total population lives in Europe proper, only 3% of the world’s total languages can be considered “languages of Europe.” The reasons for this are discussed in the next section. However, this observation does not convey the astounding diversity of immigrant languages that now have speakers located in Europe, where in some cases immigrant communities are established in remarkably high numbers. For instance, Kabyle (an Algerian Berber language) is spoken by more than 500,000 people in France, while in London alone there is a community of some 143,000 Punjabi speakers (Juaristi et al. 2008). Nichols (1992) also notes that a fair amount of linguistic diversity has been lost from this region in historical times. We know of many extinct languages around the Mediterranean (Aquitanian, Iberian, Etruscan, Lemnian, etc.) as well as whole branches of Indo-European that have been lost (Anatolian, Phrygian, and Illyrian).

Causes of language endangerment in Europe In Europe, the root causes of language endangerment may be different from in other regions, and may be more driven by pragmatic and economic realities and population movements, as well as by government policies. European minority language communities have not experienced the same level of stigmatization or political subjugation as indigenous communities on other continents (Salminen 2007a). However, language policies toward minority languages and peoples certainly factor into the endangerment for many languages in this region. One such detrimental ideology enacted on a political level in various places and at different time periods in Europe is the idea that a nation should be equated with a language, and vice versa. As mentioned earlier, such policies serve to erase minority languages within a nation’s borders by dismissing them as “dialects” or “patois,” and can create unnecessary splits in a single language if spoken across national boundaries. This ideology was on full display in Russia’s annexation of Crimea from Ukraine in 2014. At least part of the Russian 139

Anna Belew et al.

administration’s justification for taking control of this territory is that a majority of people in the Crimea are speakers of Russian, not Ukrainian – a sort of linguistic manifest destiny. This of course glosses over endangered languages in Crimea, specifically Krymchak and Karaim, as well as the minority language Crimean Tatar and the many speakers of Ukrainian. Bolstered in some cases by single-language ideologies and policies, but also in many cases by economic and international expediency, the majority of languages (especially official national languages) of Europe and Russia exert the most pressure on the endangered languages of the region. Indeed, one cannot help but observe that the vast majority of colonial languages that have endangered minority languages around the globe originated in Europe and are not only still spoken there, but enjoy official status both nationally and in the European Union. Of course, the many indigenous minority languages within Europe and Russia are not immune to the immense international economic and political capital of the traditional colonial languages (English, French, German, Russian, Portuguese, and Spanish). In fact, pressures from these languages factor into national ideologies favoring singlelanguage policies throughout Europe. For instance, all three Baltic republics (Estonia, Latvia, and Lithuania) by law have only one official language (Estonian, Latvian, and Lithuanian, respectively), which represents a backlash against the Russian language from Soviet times. Although these policies once again foreground the Baltic’s national languages, the region’s minority languages (Võro, Livonian, and Karaim, for instance) are the unintended casualties of such policy, being forced to compete not only with a national language, but also with Russian and with other global languages like English and German. Language policy in the Soviet Union deserves special attention (summarized also for the languages of the Caucasus in that section). Throughout Soviet territory, minority languages first benefitted from the Soviet policy of indigenization, and later suffered from the Stalin-era rubric of Russification. In the earlier era, minority languages were selected for development, and Soviet authorities worked extensively to cultivate literacy throughout the region, with measurable success. However, several changes in policy forced languages to abruptly switch orthographies throughout the 1920s and early 1930s. Many indigenous languages received their first alphabets in the early Soviet Union in Latin script (or other scripts appropriate to their language family and region, such as Classical Mongolian near the border with Mongolia), but were later Cyrillicized, interrupting the indigenous language literacy that was attained in the previous alphabet. A subsequent shutdown of most mother tongue schools toward the late 1930s ensured that minority peoples did not regain literacy in the new orthographies and cemented Russian as the language of education, literacy, and intercultural communication (Kazakevich & Kibrik 2007). During the era of Russification, roughly from the 1940s–50s, minority peoples were forcibly deported from their ancestral homes and sent to other Soviet Republics, chiefly Central Asia. At the same time, ethnic Russians were sent to indigenous territories to develop large agricultural communes, such that when indigenous peoples were eventually permitted to return to their homelands, sometimes decades later, a massive cultural and environmental shift had turned them into foreigners in their own homes. Of all Soviet policy, these deportations had the greatest negative impact on the indigenous languages of the Soviet Union (Kazakevich & Kibrik 2007). In recent years, the cultural climate has shifted throughout Europe and the former Soviet Union to be friendlier toward indigenous and minority languages. In some countries, linguistic diversity and multilingualism are actively cultivated. ELDIA (European Language Diversity for All)13 is an example of a recent project aimed at documenting multilingualism, specifically varieties of Uralic languages. Several languages in Europe have also benefitted 140

The world’s endangered languages

from revitalization programs in recent years. North Saami is one of the most visible and successful of these programs. Of course, even the best intentions can result in language programs that are cosmetic only, or that were developed too late to have the desired impact. Even in the case of the admirable North Saami revitalization programs, there have been some issues with this language overshadowing neighboring Saami languages, thereby benefitting North Saami at other Saami languages’ expense. Such issues are resolvable, if the current climate persists, and if adequate resources are allotted to vulnerable languages (Salminen 2007a). Finally, sign languages should be mentioned separately, because the context of their survival and endangerment is unique. Some 15 sign languages are endangered across Europe and Russia. These are threatened at least in part by medical advancements: cochlear implants and vaccines for German measles are reducing the number of deaf people. However, since Deaf communities represent a very different kind of language community from those of spoken languages, standard measures of language endangerment may be less appropriate for sign languages. It remains the case that sign languages are often under-researched, meaning that there is less data available to give an accurate assessment of these languages’ vitality.

The current status of endangered languages in Europe As of April 4, 2017, the Catalogue of Endangered Languages identifies 168 languages in Europe and Russia that are to some degree endangered. Given the total count of 266 languages based on Ethnologue (Simons & Fennig 2017), roughly 64.4% of languages in this region are currently considered endangered to some degree, which is markedly higher than the global average of 46%. Of the 169 languages listed in the Catalogue, 15 are are considered to be “At Risk,” 30 are “Vulnerable,” 38 are “Threatened,” 29 are “Endangered,” 24 are “Severely Endangered,” and 23 are “Critically Endangered.” There are six recently dormant languages (Livonian, Copper Island Aleut, Kamas, Sirenikski, Southern Khanty, and Western Mansi), and three awakening (Soyot, Cornish, and Manx). In terms of language families, some level of endangerment is ascribed to 65 Indo-European languages, 41 Uralic languages (all except Finnish, Hungarian, and Estonian), all 19 Turkic languages, 15 sign languages, all eight Tungusic languages, all five Chukotkto-Kamchatkan languages, four Afro-Asiatic languages, two Eskimo-Aleut languages, both Yeniseian languages, both Yukaghir languages, both Nivkh languages, and one mixed language (Copper Island Aleut). This means that the rough endangerment rates by language family for Europe are follows: Chukotko-Kamchatkan – 100%, Tungusic – 100%, Yeniseian – 100%; Yukagir – 100%, Nivkh – 100%, Turkic – 100%, Uralic – 93%, Eskimo Aleut – 67%, Indo-European – 52%, Afro-Asiatic 50%, and Sign languages – 37%. For Chukotko-Kamchatkan, Yeniseian, Yukaghir, Nivh, and Tungusic, every single member of these families is endangered. Below is an alphabetic listing of the languages of Europe and Russia currently included in the Catalogue of Endangered Languages.

Endangered languages of Europe Akkala Saami (Ahkkil, Babino Saami, Akkalansaame) [sia]: Critically endangered, one speaker; Uralic; spoken in Russia. Alabugat Tatar (Nogai-Tatar, Tatar-Noghay, Alabugat Tatar-Nogai) [no ISO 639-3 code]: Threatened, approximately 400 speakers; Turkic; spoken in Russia. Algherese Catalan [no ISO 639-3 code]: Vulnerable, 20,000–30,000 speakers; Indo-European; spoken in Italy. 141

Anna Belew et al.

Alutor (Alyutor, Алюторский язык, nəməlʔu) [alr]: Severely endangered, fewer than 2,000 speakers; Chukotko-Kamchatkan; spoken in Russia. Amur Nivkh (Nivx, Gilyak, амурский нивхский язык) [no ISO 639-3 code]: Severely endangered, fewer than 500 speakers; Nivkh; spoken in Russia. Aragonese (Aragoieraz, Altoaragonés, Aragonés) [arg]: Endangered, fewer than 10,000 speakers; Indo-European; spoken in Spain. Arbëreshë (Italo-Albanian, Albanian, Arber) [aae]: Vulnerable, fewer than 100,000 speakers; Indo-European; spoken in Italy. Aromanian (Armãneashti, Armãneascã, Armãneshce) [rup]: Threatened, 350,000 speakers; Indo-European; spoken in Bulgaria, Romania, Greece, Macedonia, Albania, Bosnia and Herzegovina, and Serbia. Arvanitika (Albanian, Arber, Arbresh) [aat]: Vulnerable, 50,000 speakers; Indo-European; spoken in Greece. Asturian (Asturiano, Leonese, Leonés) [ast]: At risk, 110,000 speakers; Indo-European; spoken in Spain and Portugal. Auvergnat (Auvergnat Occitan, Auvernhas, Auverne) [no ISO 639-3 code]: Threatened, 150,000 speakers; Indo-European; spoken in France. Balkan Gagauz Turkish (Balkan Turkic, Gagauz) [bgx]: Threatened, 331,000 speakers; Turkic; spoken in Turkey, Bulgaria, and Macedonia. Balkan Romani (Romany) [rmn]: At risk, 611,800 speakers; Indo-European; spoken in Russia, Bulgaria, Romania, Macedonia, Serbia, Slovakia, and Hungary. Baltic Romani (Romany, Zigeuner) [rml]: Vulnerable, 58,460 speakers; Indo-European; spoken in Russia, Ukraine, Lithuania, Poland, Estonia, Latvia, and Belarus. Baraba Tatar (барабинский язык, Параба, 巴拉巴鞑靼语) [no ISO 639-3 code]: Critically endangered, fewer than 8,000 speakers; Turkic; spoken in Russia. Breton (Brezhoneg) [bre]: At risk, 206,000 speakers; Indo-European; spoken in France. Burgenland Croatian [no ISO 639-3 code]: Threatened, 28,000 speakers; Indo-European; spoken in Austria. Campidanese Sardinian [sro]: Vulnerable, 500,000 speakers; Indo-European; spoken in Italy. Cappadocian Greek (Asia Minor Greek) [cpg]: Endangered; Indo-European; spoken in Turkey and Greece. Carpathian Romani (Romany, Cigány) [rmc]: At risk, 472,470 speakers; Indo-European; spoken in Hungary, Romania, Slovakia, and Ukraine. Catalan Sign Language (Llengua de Signes Catalana, Llengua Catalana de Signes, Lengua de Signos Catalana) [csc]: Vulnerable, 18,000 speakers; sign language; spoken in Spain. Central Selkup (Narym, Tym Selkup, čumyľ qumyt əty) [no ISO 639-3 code]: Critically endangered, two speakers; Uralic; spoken in Russia. Chukchi (Luoravetlan, Chukchee, чукотский язык) [ckt]: Threatened, fewer than 11,000 speakers; Chukotko-Kamchatkan; spoken in Russia. Chulym (Chulym Tatar, Chulim, Ös til) [clw]: Severely endangered, 35 speakers; Turkic; spoken in Russia. Cimbrian (Tzimbro, Zimbrisch) [cim]: Threatened, approximately 2,000 speakers; IndoEuropean; spoken in Italy. Copper Island Aleut (Mednyj Aleut, Copper Island Attuan, алеутско-медновский язык) [mud]: Dormant; mixed language; formerly spoken in Russia. Cornish (Kernowek, Kernewek, Curnoack) [cor]: Awakening; Indo-European; spoken in the United Kingdom. 142

The world’s endangered languages

Corsican (Corsu, Corso, Corse) [cos]: At risk, 160,000 speakers; Indo-European; spoken in France and Italy. Cypriot Spoken Arabic (Kormakiti Arabic) [acy]: Endangered, fewer than 2,000 speakers; Afro-Asiatic; spoken in Cyprus. Czech Sign Language [cse]: Vulnerable, 12,000 speakers; sign language; spoken in Czech Republic. Dalecarlian (Dalska, Dalmaal) [dlc]: Threatened, approximately 13,000 speakers; IndoEuropean; spoken in Sweden. Danish Sign Language [dsl]: Threatened, 5,000 speakers; sign language; spoken in Denmark. Dimli (Dimili, Dimlī, Zazaki) [diq]: Vulnerable, 1,000,000 speakers; Indo-European; spoken in Turkey. Dolgan (Dulğan, Haka, долганский язык) [dlg]: Threatened, 5,500 speakers; Turkic; spoken in Russia. Dutch Sign Language (Sign Language of the Netherlands, SLN, Nederlandse Gebarentaal) [dse]: Vulnerable, 20,000 speakers; sign language; spoken in the Netherlands. Eastern Buryat (Mongolian Buriat, бурят-монгольский язык, buryaad xelen) [bxm]: Vulnerable, at least 300,000 speakers; Mongolic; spoken in Russia, China, and Mongolia. Eastern Khanty (Eastern Khant, Ostyak, восточно-хантыйский язык) [no ISO 639-3 code]: Critically endangered, approximately 480 speakers; Uralic; spoken in Russia. Eastern Mansi (Eastern Vogul, восточно-мансийский язык, кондинский-мансийский язык) [no ISO 639-3 code]: Severely endangered, fewer than 500 speakers; Uralic; spoken in Russia. Eastern Mari (Meadow Mari, Cheremis, Grassland Mari) [mhr]: Vulnerable, 500,000 speakers; Uralic; spoken in Russia. Eastern Yiddish (Central Yiddish, Judeo-German) [ydd]: At risk, at least 1,200,000 speakers; Indo-European; spoken in the United States of America, Israel, Ukraine, Canada, and Belarus. Erzya (Mordvin, Erza, эрзянский язык) [myv]: Threatened, approximately 250,000 speakers; Uralic; spoken in Russia. Estonian Sign Language (Viipekeel) [eso]: Threatened, 4,500 speakers; sign language; spoken in Estonia. Even (Lamut, Orich, Ilqan) [eve]: Severely endangered, approximately 7,000 speakers; Tungusic; spoken in Russia. Evenki (Solon, Ewenki, эвенкский язык) [evn]: Endangered, approximately 10,000 speakers; Tungusic; spoken in Russia, China, and Mongolia. Finnish Sign Language (Viittomakieli) [fse]: Threatened, 5,000 speakers; sign language; spoken in Finland. Finnish-Swedish Sign Language (Finland-Swedish Sign Language, Suomenruotsalainen Vittomakieli, Finlandssvenskt Teckenspråk) [fss]: Endangered, 150 speakers; sign language; spoken in Finland. Flemish Sign Language (South Belgium Sign Language, North Belgium Sign Language, Vlaamse Gebarentaal) [vgt]: Threatened, 6,000 speakers; sign language; spoken in Belgium. Forest Enets (Bai Enets, Pe-Bae, лесной энецкий) [enf]: Severely endangered, approximately ten speakers; Uralic; spoken in Russia. Forest Nenets (Forest Yurak, metsänenetsi, ненецкий лесной язык) [no ISO 639-3 code]: Endangered; Uralic; spoken in Russia. 143

Anna Belew et al.

Forest Yukagir (Southern Yukagir, Kolyma Yukaghir, южноюкагирский язык) [yux]: Critically endangered, fewer than 50 speakers; Yukaghir; spoken in Russia. Francoprovençal (Arpitan, Romand) [frp]: Endangered, 130,000 speakers; Indo-European; spoken in France, Italy, and Switzerland. Friulian (Furlan, Frioulan, Frioulian) [fur]: Vulnerable, 350,000–500,000 speakers; Indo-European; spoken in Italy. Gagauz (Gagauzi, Gaugausen) [gag]: Vulnerable, approximately 200,000 speakers; Turkic; spoken in Ukraine, Bulgaria, Romania, Greece, Macedonia, and Moldova. Gallo (Galo, Gallot, Galoo) [no ISO 639-3 code]: Threatened, 200,000–400,000 speakers; IndoEuropean; spoken in France. Gallurese Sardinian (Northeastern Sardinian) [sdn]: At risk, 100,000 speakers; Indo-European; spoken in Italy. Gardiol [no ISO 639-3 code]: Severely endangered, 300 speakers; Indo-European; spoken in Italy. Gascon (Gascon Occitan, Aranés, Aranese) [no ISO 639-3 code]: At risk, 250,000 speakers; Indo-European; spoken in France and Spain. Guernésiais (Dgèrnésiais, Guernsey French, Guernésiais) [no ISO 639-3 code]: Endangered, 200-300 speakers; Indo-European; spoken in the United Kingdom. Inari Saami (инари-саамский язык, inarinsaame, anarâškielâ) [smn]: Endangered, approximately 300 speakers; Uralic; spoken in Finland. Ingrian (Izhorian, ижорский язык, inkeroisen kieli) [izh]: Severely endangered, approximately 130 speakers; Uralic; spoken in Russia. Irish (Gaeilge, Irish Gaelic) [gle]: Endangered, 20,000–40,000 speakers; Indo-European; spoken in the United Kingdom and Ireland. Istriot [ist]: Severely endangered, approximately 400 speakers; Indo-European; spoken in Croatia. Istro Romanian [ruo]: Endangered, approximately 300 speakers; Indo-European; spoken in Croatia. Itelmen (ительменский язык, Itənmən, Kamchadal) [itl]: Severely endangered, fewer than 20 speakers; Chukotko-Kamchatkan; spoken in Russia. Jèrriais (Guernésiais, Jersey French, Jersiais) [nrf]: Threatened, 2,874 speakers; Indo-European; spoken in the United Kingdom. Judeo-Italian (Italkian) [itk]: Severely endangered, fewer than 50 speakers; Indo-European; spoken in Italy. Kamas (Kamassian, камасинский язык, калмажи шəкə) [xas]: Dormant; Uralic; formerly spoken in Russia. Karaim (Karaite, Karaïm) [kdr]: Critically endangered, 500 speakers; Turkic; spoken in Ukraine and Lithuania. Karelian (karjala, карельский язык) [krl]: Threatened, 63,000 speakers; Uralic; spoken in Russia and Finland. Kashubian (Cassubian, Kaszubski, Cashubian) [csb]: Vulnerable, 50,000 speakers; Indo-European; spoken in Poland. Kerek (керекский язык) [krk]: Critically endangered, fewer than ten speakers; ChukotkoKamchatkan; spoken in Russia. Ket (Imbak, кетский язык, Yenisei Ostyak) [ket]: Endangered, approximately 500 speakers; Yeniseian; spoken in Russia. Khakas (Khakhas, Abakan Tatar, хакасский язык) [kjh]: Threatened, 6,000 speakers; Turkic; spoken in Russia. 144

The world’s endangered languages

Kildin Saami (Kildin Sámi, Eastern Saami, Kola Saami) [sjd]: Severely endangered, approximately 300 speakers; Uralic; spoken in Russia. Kili (Kur-Urmi Nanay, Kur-Urmi Nanai, кили) [no ISO 639-3 code]: Endangered, approximately 100 speakers; Tungusic; spoken in Russia. Komi-Permyak (Komi, Permyak, Komi-Perm) [koi]: Vulnerable, 110,000 speakers; Uralic; spoken in Russia. Koryak (Nymylan, Korak, корякский язык) [kpy]: Endangered, approximately 3,000 speakers; Chukotko-Kamchatkan; spoken in Russia. Krymchak (Judeo-Crimean Tatar, крымчакский язык, кърымчах тыльы) [jct]: Critically endangered, 200 speakers; Turkic; spoken in Georgia, Ukraine, Bulgaria, and Romania. Kumandin (Qumanda, Northern Altai, Куванды) [no ISO 639-3 code]: Endangered, 1,044 speakers; Turkic; spoken in Russia. Kven Finnish (North Fennic, North Finnish, Kven) [fkv]: Severely endangered, 1,500–10,000 speakers; Uralic; spoken in Norway. Ladin (Ladino, Dolomite, Rhaeto-Romance) [lld]: Threatened, approximately 20,000 speakers; Indo-European; spoken in Italy. Languedocien (Languedocien Occitan, Languedocian, Lengadoucian) [no ISO 639-3 code]: Vulnerable, at least 1,000,000 speakers; Indo-European; spoken in France. Limousin (Limousin Occitan, Lemosin) [no ISO 639-3 code]: At risk, 400,000 speakers; IndoEuropean; spoken in France. Livonian (Liv, Liivõ Keelj, rānda kēļ) [liv]: Dormant; Uralic; formerly spoken in Latvia. Livvi (Livvi Karelian, Olonets, Aunus) [olo]: Endangered, approximately 27,000 speakers; Uralic; spoken in Russia and Finland. Logudorese Sardinian [src]: Threatened, approximately 500,000 speakers; Indo-European; spoken in Italy. Lower Silesian (Schlesisch, Silesian German) [sli]: Vulnerable, 22,900 speakers; IndoEuropean; spoken in Germany, Poland, and Czech Republic. Lower Sorbian (Lusatian) [dsb]: Threatened, 7,240 speakers; Indo-European; spoken in Germany. Ludian (Ludic, Ludic Karelian, людиковский язык) [lud]: Critically endangered, 300 speakers; Uralic; spoken in Russia. Lule Saami (Lulesamisk, Lulesamiska, luulajansaame) [smj]: Endangered, 1,000–2,000 speakers; Uralic; spoken in Sweden and Norway. Manx (Gaelg, Gailck, Manks) [glv]: Awakening; Indo-European; spoken in the United Kingdom. Mardin Sign Language (Dilsizce, Eski Işaretler) [no ISO 639-3 code]: Severely endangered, approximately 40 speakers; sign language; spoken in Turkey. Megleno-Romanian (Megleno-Rumanian, Meglenitic, Meglenite) [ruq]: Threatened, 5,000 speakers; Indo-European; spoken in Greece and Macedonia. Middle Chulym (Melets Tatar, Chulim, Ös til) [no ISO 639-3 code]: Severely endangered, fewer than 50 speakers; Turkic; spoken in Russia. Miranda do Douro (Mirandês, Mirandesa, Mirandese) [mwl]: Vulnerable, 15,000 speakers; Indo-European; spoken in Spain and Portugal. Mócheno [mhn]: Threatened, 1,900 speakers; Indo-European; spoken in Italy. Moksha (Mokša, Mordvin, мокшанский язык) [mdf]: At risk, 200,000 speakers; Uralic; spoken in Russia. Naukan Yupik (Naukanski, East Cape Yupik) [ynk]: Endangered, approximately 200 speakers; Eskimo-Aleut; spoken in Russia. 145

Anna Belew et al.

Negidal (Neghidal, Nigidal, Ilkan Beye) [neg]: Severely endangered, approximately 150 speakers; Tungusic; spoken in Russia. Nganasan (Tavgi Samoyed, Tavgy, Ŋanasan) [nio]: Severely endangered, approximately 500 speakers; Uralic; spoken in Russia. Norman (Northern French, Normand) [no ISO 639-3 code]: Severely endangered, approximately 11,100 speakers; Indo-European; spoken in France. North Saami (Sami, North Sámi, Northern Saami) [sme]: Vulnerable, 16,500 speakers; Uralic; spoken in Russia, Sweden, Norway, and Finland. Northern Altai (Teleut, Telengut, северноалтайский язык) [atv]: Endangered, fewer than 10,000 speakers; Turkic; spoken in Russia. Northern Frisian [frr]: Threatened, approximately 8,000 speakers; Indo-European; spoken in Germany. Northern Khanty (Northern Khant, Ostyak, северно-хантыйский язык) [no ISO 639-3 code]: Threatened, approximately 10,000 speakers; Uralic; spoken in Russia. Northern Mansi (Northern Vogul, Sos’va, северно-мансийский язык) [no ISO 639-3 code]: Endangered, approximately 3,000 speakers; Uralic; spoken in Russia. Northern Selkup (Taz Selkup, šöľqumyt əty, северноселькупский язык) [no ISO 639-3 code]: Endangered, fewer than 600 speakers; Uralic; spoken in Russia. Norwegian Sign Language [nsl]: Threatened, 4,000 speakers; sign language; spoken in Norway. Oroch (Orochi, орочский язык, Sprache der Oročonen) [oac]: Severely endangered, fewer than 150 speakers; Tungusic; spoken in Russia. Orok (Ulta, Ujlta, орокский язык) [oaa]: Critically endangered, fewer than 60 speakers; Tungusic; spoken in Russia and Japan. Picard (Rouchi, Chtimi) [pcd]: Vulnerable, approximately 200,000 speakers; Indo-European; spoken in Belgium and France. Piemontese (Turinese Italian, Piedmontese) [pms]: Vulnerable, approximately 2,000,000 speakers; Indo-European; spoken in Italy. Pite Saami (Arjeplog Saami, Pitesamisk, Pitesamiska) [sje]: Critically endangered, approximately 42 speakers; Uralic; spoken in Sweden and Norway. Poitevin [no ISO 639-3 code]: Critically endangered; Indo-European; spoken in France. Pontic (Pontic Greek, Romeyka, Rumca) [pnt]: Threatened, 100,000–200,000 speakers; IndoEuropean; spoken in Greece. Portuguese Sign Language (Lingua Gestual Portuguesa) [psr]: Threatened, fewer than 8,000 speakers; sign language; spoken in Portugal. Provençal (Provençal Occitan, Provenzale, Prouvençau) [no ISO 639-3 code]: Threatened, approximately 200,000 speakers; Indo-European; spoken in France. Romagnol (Emiliano-Romagnolo, Rumagnol) [rgn]: Threatened, 1,000,000–2,000,000 speakers; Indo-European; spoken in Italy. Romansch (Grishun, Rumauntsch, Rheto-Romance) [roh]: Vulnerable, approximately 60,000 speakers; Indo-European; spoken in Switzerland. Sakhalin Nivkh (Nivx, Gilyak, сахалинский нивхский язык) [niv]: Severely endangered, at least 500 speakers; Nivkh; spoken in Russia. Sassarese Sardinian (Northwestern Sardinian) [sdc]: At risk, 100,000 speakers; Indo-European; spoken in Italy. Saterfriesisch (East Frisian, Saterländisch) [stq]: Threatened, 5,000 speakers; Indo-European; spoken in Germany. 146

The world’s endangered languages

Scottish Gaelic (Gàidhlig, Scots Gaelic) [gla]: Threatened, 20,000–30,000 speakers; IndoEuropean; spoken in the United Kingdom. Shor (Kuznets Tatar, Kondoma Tatar, Mras-Su Tatar) [cjs]: Severely endangered, fewer than 10,000 speakers; Turkic; spoken in Russia. Sinte Romani (Manuche Romany) [rmo]: At risk, 318,920 speakers; Indo-European; spoken in Germany, France, Austria, Italy, Switzerland, and Serbia. Sirenikski (Sirenik Yupik) [ysr]: Dormant; Eskimo-Aleut; formerly spoken in Russia. Skolt Saami (Koltta, Eastern Saami, Kola Saami) [sms]: Critically endangered, approximately 300 speakers; Uralic; spoken in Russia, Norway, and Finland. South Saami [sma]: Endangered, 600 speakers; Uralic; spoken in Sweden and Norway. Southern Altai (Altai Kalmuck, Oirot, южноалтайский язык) [alt]: Threatened, fewer than 20,000 speakers; Turkic; spoken in Russia. Southern Khanty (Southern Khant, Ostyak, южно-хантыйский язык) [no ISO 639-3 code]: Dormant; Uralic; formerly spoken in Russia. Southern Selkup (Ket Selkup, южноселькупский язык, tüj qumyt əty) [no ISO 639-3 code]: Critically endangered, one speaker; Uralic; spoken in Russia. Soyot (Soyote, Tuva, сойотский язык) [no ISO 639-3 code]: Awakening; Turkic; spoken in Russia. Swedish Sign Language [swl]: Threatened; sign language; spoken in Sweden. Swiss-French Sign Language (Langage Gestuelle, Langue des Signes Suisse Romande, LSF-SR) [ssr]: Endangered, 1,700 speakers; sign language; spoken in Switzerland. Swiss-German Sign Language (Natürliche Gebärde, Deutschschweizerische Gebärdensprache, DSGS) [sgg]: Endangered, 5,500 speakers; sign language; spoken in Switzerland and Liechtenstein. Swiss-Italian Sign Language (Lingua dei Segni Italiana, LIS-SI) [slf]: Endangered, 300 speakers; sign language; spoken in Switzerland. Ter Saami (Ter Sami, Kola Saami, Eastern Saami) [sjt]: Severely endangered, 30 speakers; Uralic; spoken in Russia. Tofa (Karagas, Tofalar, тофаларский язык) [kim]: Critically endangered, approximately 40 speakers; Turkic; spoken in Russia. Tsakonian (Tsaconia, Tsakonia) [tsd]: Severely endangered, 2,000–4,000 speakers; IndoEuropean; spoken in Greece. Tundra Enets (Somatu Enets, Madu, тундровый энецкий) [enh]: Critically endangered, approximately 30 speakers; Uralic; spoken in Russia. Tundra Nenets (Tundra Yurak, Bol’shezemel’sk Nenets, tundranenetsi) [no ISO 639-3 code]: Endangered, 20,000 speakers; Uralic; spoken in Russia. Tundra Yukagir (Northern Yukagir, Vadul Yukaghir, cеверноюкагирский язык) [ykg]: Critically endangered, fewer than 50 speakers; Yukaghir; spoken in Russia. Tuva (Tuvin, Tyva, Tuvan) [tyv]: At risk, 242,754 speakers; Turkic; spoken in Russia, China, and Mongolia. Udege (Udihe, Udeghe, удэгейский язык) [ude]: Critically endangered, fewer than 100 speakers; Tungusic; spoken in Russia. Udmurt (Votyak, удмурт кыл, удмуртский язык) [udm]: Threatened, 324,338 speakers; Uralic; spoken in Russia. Ulch (Olcha, Ulcha, ульчский язык) [ulc]: Critically endangered, fewer than 1,000 speakers; Tungusic; spoken in Russia. Ume Saami (Southern Saami, Ume Sami, Umesamiska) [sju]: Critically endangered, fewer than 20 speakers; Uralic; spoken in Sweden. 147

Anna Belew et al.

Upper Silesian (Silesian, ślōnskŏ gŏdka) [szl]: Vulnerable, 56,000 speakers; Indo-European; spoken in Poland. Upper Sorbian (Eastern Sorbian) [hsb]: Vulnerable, 18,240 speakers; Indo-European; spoken in Germany. Urum (урумский язык, урум тыль, Urumca) [uum]: Endangered; Turkic; spoken in Russia, Georgia, Ukraine, and Greece. Veps (Vepsian, vepsä, вепсский язык) [vep]: Severely endangered, 3,613 speakers; Uralic; spoken in Russia. Vivaro-Alpine (Vivaro-Alpine Occitan, Alpine Provençal, Vivaroalpenc) [no ISO 639-3 code]: Endangered, 200,000 speakers; Indo-European; spoken in France and Italy. Vlax Romani (Romany, Romenes) [rmy]: At risk, 885,970 speakers; Indo-European; spoken in Bosnia and Herzegovina, Germany, Italy, and Slovakia. Võro (Võru, Võro kiil, Werro) [vro]: Vulnerable, approximately 50,000 speakers; Uralic; spoken in Estonia. Votic (Vod, Vot, водский язык) [vot]: Critically endangered, fewer than ten speakers; Uralic; spoken in Russia. Walloon (Walon) [wln]: Threatened, approximately 600,000 speakers; Indo-European; spoken in Belgium, France, and Luxembourg. Welsh (Cymraeg, y Gymraeg, Kymrisch) [cym]: Vulnerable, 562,016 speakers; Indo-European; spoken in Argentina and United Kingdom. West Frisian (Frysk, Fries, Frisian) [fry]: Vulnerable, approximately 300,000 speakers; IndoEuropean; spoken in the Netherlands. Western Buryat (Russian Buriat, бурят-монгольский язык, buryaad xelen) [bxr]: Endangered, fewer than 50,000 speakers; Mongolic; spoken in Russia. Western Mansi (Western Vogul, Pelym, западно-мансийский язык) [no ISO 639-3 code]: Dormant; Uralic; formerly spoken in Russia. Western Mari (Hill Mari, Cheremis, High Mari) [mrj]: Vulnerable, fewer than 50,000 speakers; Uralic; spoken in Russia. Western Yiddish (Central Yiddish) [yih]: Threatened, 5,400 speakers; Indo-European; spoken in Germany, Israel, Belgium, France, Netherlands, Switzerland, and Hungary. Yakut (Sakha, якутский язык, Saxa tıla) [sah]: Vulnerable, 456,288 speakers; Turkic; spoken in Russia. Yazva (Komi-Yazva, Eastern Permyak, коми-язьвинский язык) [no ISO 639-3 code]: Critically endangered, approximately 200 speakers; Uralic; spoken in Russia. Yug (Yugh, Southern Ket, Sym Ket) [yuu]: Critically endangered, fewer than five speakers; Yeniseian; spoken in Russia.

Mexico, Central America, and the Caribbean Lyle Campbell Linguistic diversity in Mexico, Central America, and the Caribbean14 The Americas are notorious for their linguistic diversity.15 For example, it has been claimed that “in one small portion of the area, in Mexico just north of the Isthmus of Tehuantepec, one finds a diversity of linguistic type hard to match on an entire continent in the Old World” (McQuown 1955:501). The parts of the region that ELCat calls Mexico, Central America, and 148

The world’s endangered languages

the Caribbean are not united in any way other than vaguely by geography; that is, the Caribbean, Mexico north of Mesoamerica, and Central America below Mesoamerica are culturally and linguistically unconnected for the most part. “Mesoamerica,” a well-known area within this region, refers to a culture area defined by a number of cultural traits shared by the indigenous cultures of the region that extends from the Pánuco River in northern Mexico to the Lempa River in El Salvador, continuing along the Pacific coast of Nicaragua and Costa Rica. Mesoamerica is also a linguistic area, which coincides approximately with the territory of the Mesoamerican culture area, defined by linguistic traits shared among languages of the area (cf. Campbell et al. 1986). Mesoamerica contains the largest number of distinct language families and individual languages in this region, with seven language families – Mayan, Mixe-Zoquean, Tequistlatecan, Totonacan (Totonac-Tepehua), Otomanguean, Uto-Aztecan, and Xinkan – and three language isolates – Cuitlatec, Huave, and Tarascan (Purépecha). Some of the language families have representatives that extend beyond Mexico, Central America, and the Caribbean. Uto-Aztecan languages extend from Oregon to Nicaragua. Algonquian (of the larger Algic family), mostly in North America north of Mexico, has one representative that extends into Mexico, Kickapoo. The Chibchan family extends from Honduras to Colombia and on into Venezuela. The Arawakan family of South America, with representatives from Venezuela and the Guianas to Argentina, also laps over into the Central America with Garífuna, that reached Central America from Caribbean islands. The countries of this region, together with the number of endangered languages in each, are: Belize: four Costa Rica: nine (three are sign languages, one of which is non-indigenous) El Salvador: three (one is dormant; one non-indigenous sign language) Guatemala: 24 (one awakening, two dormant; one a non-indigenous sign language) Honduras: eight (one dormant, one a non-indigenous sign language) Jamaica: one (the non-indigenous Jamaica Country Sign Language) Mexico: 121 (two dormant, two sign languages, one of which is non-indigenous) Nicaragua: five (one a non-indigenous sign language, one a creole) Panama: 13 (two sign languages, one creole) Puerto Rico: one (the non-indigenous Puerto Rican Sign Language) Trinidad and Tobago: one (the non-indigenous Trinidad and Tobago Sign Language) Cuba and Haiti, also in this region, are among the few countries in the world that have no endangered languages.

Causes of language endangerment in Mexico, Central America, and the Caribbean In particular diseases introduced from Europe but also slavery, resettlement, and warfare against indigenous groups in early colonial times resulted in a general population collapse in this region. In Hispaniola (today’s Dominican Republic and Haiti) the culture and language of the Taino, whom Columbus encountered upon his arrival to the New World, had been lost by 1650. In Mexico and Central America, disease primarily reduced the population to a bare shadow of what it was in the time before European contact. The imprecise estimates of Mexico’s 149

Anna Belew et al.

population in 1519, when the Spaniards arrived, range from six million to 25 million, but it is agreed that by 1600 no more than 2 million remained (Stevenson 2007). A figure of 90% decline by 1700 is commonly cited for Mexico (see Denevan 1992). The encomienda system also contributed to population decline. The policies of the dominant colonial powers favored and even forced language shift, to Spanish in Mexico and Central America, and to Dutch, English, and French in different parts of the Caribbean. Economic factors are thought to have the greatest impact today motivating ongoing language shifts, including resource depletion, forced changes in subsistence patterns, lack of economic opportunities, on-going industrialization, and migration for economic advantage. Political and social factors, often intertwined with economic ones, are also important. Speakers of the indigenous languages in many cases have accepted the dominant society’s negative evaluations of their languages, though fortunately this has changed to a more positive image of their languages and cultural identities in many places in more recent times. There is, however, still much discrimination, repression, and stigmatization of indigenous peoples and their languages in this region. As in other regions, the attitudes towards what the indigenous language symbolizes versus what the dominant language symbolizes contribute to language shift. The dominant language has associations with nationhood, civilization, education, economic advantage, progress, and the future, as opposed to images of the minority languages as representing the opposite, associated with the past, economic disadvantage, being rural, and having the low general prestige. Absence of institutional support (the minority language not being represented in schools, government, judicial systems, academies, texts, the media, etc.) intersects with this. Civil wars and natural disasters (earthquakes, hurricanes) have displaced many indigenous people here, bringing many to large urban centers. Gradual assimilation of indigenous groups bringing them into the mainstream campesino culture of Mexico and Central America and globalization are strong factors in the endangerment of the languages of these regions.

The current status of endangered languages in Mexico, Central America, and the Caribbean As elsewhere in the Americas, all the indigenous languages of this region are endangered. The Catalogue of Endangered Languages (ELCat) has 166 endangered languages in this region; six are dormant and two are awakening, so 158 are still living. There are 14 sign languages, three of them indigenous (Bribri Sign Language, Brunca Sign Language, and Yucatec Maya Sign Language). There are two endangered pidgin and creole languages (San Miguel Creole French and Nicaraguan Creole English). Of these endangered languages, five have ten or fewer speakers, 14 fewer than 100 speakers. The number of still spoken languages at each level of vitality in ELCat in this region follows, together with the percentage this amounts to of all these 158 living languages of this region: 10 6 17 57 51 13 4

“Critically Endangered” (6.5%) “Severely Endangered” (3.9%) “Endangered” (11%) “Threatened” (37%) “Vulnerable” (33.1%) “At Risk” (8.4%) Unknown vitality (2.6%) 150

The world’s endangered languages

Below is an alphabetical listing of all languages of Mexico, Central America, and the Caribbean currently included in the Catalogue of Endangered Languages.

Endangered languages of Mexico, Central America, and the Caribbean Achi’ (Rabinal K’iche’, Achí) [acr]: Vulnerable, 51,593 speakers; Mayan; spoken in Guatemala. Akateko (Acatec, Western Kanjobal) [knj]: Threatened, 5,572 speakers; Mayan; spoken in Guatemala and Mexico. Albarradas Zapotec (Santo Domingo Albarradas Zapotec, Zapotec, Albarradas) [zas]: Threatened, 5,500 speakers; Otomanguean; spoken in Mexico. Asunción Mixtepec Zapotec (North Central Zimatlan Zapotec, Zapoteco de Asunción Mixtepec, Central Zapotec) [zoo]: Severely endangered, 100 speakers; Otomanguean; spoken in Mexico. Awakateko (Aguacatec, Aguacateco, Awakatek) [agu]: Vulnerable, 16,272 speakers; Mayan; spoken in Guatemala. Ayapanec Zoque (Ayapanec, Ayapaneco, Ayapa) [zoq]: Critically endangered, four speakers; Mixe-Zoquean; spoken in Mexico. Ayoquesco (Ayoquesco Zapotec) [zaf]: Endangered, 880 speakers; Otomanguean; spoken in Mexico. Bocotá (Buglere, Bokota, Bogota) [sab]: Threatened, 3,251 speakers; Chibchan; spoken in Panama. Boruca (Boruka, Borunca, Borunka) [brn]: Awakening; Chibchan; spoken in Costa Rica. Bribri (Talamanca, Viceita, Duit) [bzd]: Threatened, 7,475 speakers; Chibchan; spoken in Costa Rica and Panama. Bribri Sign Language [no ISO 639-3 code]: Endangered, approximately 400 speakers; sign language; spoken in Costa Rica. Brunca Sign Language [no ISO 639-3 code]: Endangered, 808 speakers; sign language; spoken in Costa Rica. Cabécar (Kabékar, Cabecar, Tucurrique) [cjp]: Threatened, 8,441 speakers; Chibchan; spoken in Costa Rica. Central Baja Mixtec (Silacayoapan Mixtec, Northwest Oaxaca Mixtec, Mixteco de Yucuná) [mks, mxa]: Vulnerable, 24,130 speakers; Otomanguean; spoken in Mexico. Central Chinantec (jajme dzä mii, jmiih kia’ dzä mii, Chinanteco Central) [cuc, cvn, csa, cle, cpa]: Vulnerable, 34,852 speakers; Otomanguean; spoken in Mexico. Ch’orti’ (Chortí, Apay) [caa]: Threatened, 9,105 speakers; Mayan; spoken in Honduras and Guatemala. Chiapanec (Chiapaneco, Chiapaneko) [cip]: Dormant; Otomanguean; formerly spoken in Mexico. Chichimeca-Jonaz (Chichimeco, Chichimeca, Chichimec) [pei]: Threatened, 1,582 speakers; Otomanguean; spoken in Mexico. Chiquihuitlán Mazatec (Masateko, Mazateco) [maq]: Threatened, 1,502 speakers; Otomanguean; spoken in Mexico. Chiquimulilla Xinka (Chiguimuliya) [no ISO 639-3 code]: Awakening; Xincan; spoken in Guatemala. Chiriqui Sign Language (Lengua de Señas de Chiriqui, LSCH) [no ISO 639-3 code]: Threatened; sign language; spoken in Panama. Choapan (Choapan Zapotec) [zpc]: Vulnerable, 12,000 speakers; Otomanguean; spoken in Mexico. 151

Anna Belew et al.

Chocho (Chocholteco, Chuchona, Chochoteco) [coz]: Vulnerable, 12,554 speakers; Otomanguean; spoken in Mexico. Chol (Ch’ol, Tumbalá, Ch’ol de Sabanilla) [cti, ctu]: At risk, 128,240 speakers; Mayan; spoken in Mexico. Chontal de Tabasco (Chontal Maya) [chf]: Vulnerable, 36,267 speakers; Mayan; spoken in Mexico. Chuj [cac, cnm]: Vulnerable, 38,253 speakers; Mayan; spoken in Guatemala and Mexico. Coast Mixtec (Amoltepec Mixtec, Western Sola de Vega Mixtec) [mbz, mih, mio, mjc, mtu, mxt, mza, vmj]: Vulnerable, 63,180 speakers; Otomanguean; spoken in Mexico. Coatlán Zapotec [zps, zpt, ztp, zao, zam, zpr, zap, ztg, ztl, zpo, zpb]: Vulnerable, 75,000 speakers; Otomanguean; spoken in Mexico. Cora [crn, cok]: Vulnerable, 17,086 speakers; Uto-Aztecan; spoken in Mexico. Cuicatec (Cuica, Cuicateco) [cux, cut]: Threatened, 1,641 speakers; Otomanguean; spoken in Mexico. Eastern Alta Mixtec (Yutanduchi Mixtec, Mixteco de Yutanduchi de Guerrero) [mab, mil, mqh, mtx, mxs, mxy, vmm, vmx, xtd, xtp, xts]: Vulnerable, 24,244 speakers; Otomanguean; spoken in Mexico. Eastern Nahua (Highland Puebla Nahuatl, Náhuat de la Sierra de Puebla, Sierra Puebla Náhuatl) [azz]: At risk, 125,000 speakers; Uto-Aztecan; spoken in Mexico. Etla (Tejalapan Zapotec) [zpy]: Threatened, 1,276 speakers; Otomanguean; spoken in Mexico. Garífuna (Garifuna, Garfuna, Black Carib) [cab]: At risk, approximately 100,000 speakers; Arawakan; spoken in Honduras, Guatemala, and Belize. Guarijío (Warihío, Huarijio, Varohio) [var]: Threatened, 2,136 speakers; Uto-Aztecan; spoken in Mexico. Guatemalan Sign Language (Lensegua) [gsm]: Threatened, 28,000–256,000 speakers; sign language; spoken in Guatemala. Guatuso (Maléku Jaíka, Watuso-Wétar, Guetar) [gut]: Endangered, 276 speakers; Chibchan; spoken in Costa Rica. Guaymí (Ngäbere, Chiriqui, Ngobere) [gym]: At risk, 147,623 speakers; Chibchan; spoken in Costa Rica and Panama. Guazacapán Xinka (Xinca de Guazacapán) [qda]: Dormant; Xincan; formerly spoken in Guatemala. Guerrero Amuzgo (Amusgo, Nomndaa) [amu]: Vulnerable; Otomanguean; spoken in Mexico. Guerrero Mixtec (Alacatlatzala Mixtec, Central Misteko, Highland Guerrero Mixtec) [mim, mxv, xta, xty]: Vulnerable, 97,200 speakers; Otomanguean; spoken in Mexico. Highland Chontal (Chontal de Oaxaca) [chd]: Threatened, 3,413 speakers; Tequistlatecan; spoken in Mexico. Honduran Lenca (Lenka, Lenca) [len]: Dormant; Lencan; formerly spoken in Honduras. Honduras Sign Language (LESHO) [hds]: Threatened; sign language; spoken in Honduras. Huastec (Wasteko, Huasteco, Te:nek) [hsf, hus, hva]: Threatened, 1,700 speakers; Mayan; spoken in Mexico. Huave (Huave) [hve, hue, huv, hvv]: Vulnerable, 10,000–15,000 speakers; isolate; spoken in Mexico. Huehuetla Tepehua (Huehuetla Tepewa, Tepehua de Hidalgo) [tee]: Threatened, 3,000 speakers; Totonacan; spoken in Mexico. Huichol (Vixaritari Vaniuqui, Vizaritari Vaniuki) [hch]: Vulnerable, 35,724 speakers; UtoAztecan; spoken in Mexico. Isthmus Zapotec [zai]: Vulnerable, 80,000 speakers; Otomanguean; spoken in Mexico. 152

The world’s endangered languages

Itza’ (Itz, Itzá, Itzaj) [itz]: Severely endangered, 12 speakers; Mayan; spoken in Guatemala. Ixcatec (Ixcateco, Ixcatec, Iskateko) [ixc]: Critically endangered, nine speakers; Otomanguean; spoken in Mexico. Ixil [ixi, ixj, ixl]: Vulnerable, 69,137 speakers; Mayan; spoken in Guatemala. Ixtenco Otomí (Yühmü) [otz]: Endangered, 740 speakers; Otomanguean; spoken in Mexico. Ixtlán (Sierra de Juárez Zapotec) [zaa, zpd, zae]: Vulnerable, 12,900 speakers; Otomanguean; spoken in Mexico. Jakalteko (Jacaltec, Jakalteco, Popti’) [jac, jai]: Vulnerable, 38,350 speakers; Mayan; spoken in Guatemala and Mexico. Jamaican Country Sign Language (Konchri Sain, Country Sign) [jcs]: Threatened, 2,500 speakers; sign language; spoken in Jamaica. Jumaytepeque Xinka (Jumay Xinka, Xinca de Jumaytepeque) [qhq]: Dormant; Xincan; formerly spoken in Guatemala. K’iche’ (Kiché, Quiché, Quiche) [quc]: Vulnerable, 922,378 speakers; Mayan; spoken in Guatemala and Mexico. Kaqchikel (Kakchiquel, Cakchiquel) [cak, ckk, cke, ckc, cki, ckj, ckd, ckf, ckw, cbm]: Vulnerable, 475,889 speakers; Mayan; spoken in Guatemala. Kiliwa (Kiliwi, Quiligua) [klb]: Critically endangered, fewer than eight speakers; CochimiYuman; spoken in Mexico. Kumeyaay (Central Diegueño, Kamia, Cuchimí) [dih]: Critically endangered, 40–50 speakers; Cochimi-Yuman; spoken in Mexico and the United States of America. Kuna (Cuna) [cuk, kvn]: Vulnerable, 44,100 speakers; Chibchan; spoken in Panama and Colombia. Lacandon (Lacandón, Lakantún, Lacandon Maya) [lac]: Endangered, 563 speakers; Mayan; spoken in Mexico. Lower Central Chinantec (Chinanteko) [cvn]: Threatened, 1,500 speakers; Otomanguean; spoken in Mexico. Lower Southeastern Chinantec (jujmi dsa m+ta’o, jujmi dsa maji’i, Chinanteco del sureste bajo) [cnl]: Vulnerable, 10,700 speakers; Otomanguean; spoken in Mexico. Lower West-Central Chinantec (jajmi dzä kï ï ‘, jejmei, jejmi) [cuc, cnt]: Threatened, 9,260 speakers; Otomanguean; spoken in Mexico. Lowland Chontal (Huamelultec, Lowland Oaxaca Chontal, Huamelula Chontal) [clo]: Severely endangered, 100 speakers; Tequistlatecan; spoken in Mexico. Lowland Mixe (Isthmus Mixe, Eastern Mixe, Guichicovi Mixe) [mco, mir, mzl, pxm]: Vulnerable, 50,900 speakers; Mixe-Zoquean; spoken in Mexico. Mam (Qyo:l) [mam]: Vulnerable, 519,664 speakers; Mayan; spoken in Guatemala and Mexico. Matlatzinca (Bot’una) [mat]: Threatened, 1,134 speakers; Otomanguean; spoken in Mexico. Mayo (Cahita, Yaqui-Mayo, Kahita) [mfy]: Vulnerable, 32,702 speakers; Uto-Aztecan; spoken in Mexico. Mexican Sign Language (Lenguaje de Signos Mexicano, Lenguaje Manual Mexicana, Lenguaje de las Manos) [mfs]: Vulnerable, 87,000–100,000 speakers; sign language; spoken in Mexico. Mezquital Otomí (Othomí) [ote]: At risk, 100,000 speakers; Otomanguean; spoken in Mexico. Middle Southeastern Chinantec (jujmi del sureste medio, Chinanteco del sureste medio, Tepinapa) [cte]: Vulnerable, 77,087 speakers; Otomanguean; spoken in Mexico. Midland Mixe (South Central Mixe, Atitlán Mixe) [mxq, neq]: Vulnerable, 21,000 speakers; Mixe-Zoquean; spoken in Mexico. 153

Anna Belew et al.

Misantla Totonac (Southeastern Totonac, Yecuatla Totonac, Laakanaachiwiin) [tlc]: Endangered, 500 speakers; Totonacan; spoken in Mexico. Mískito (Mosquito, Mísquito, Mískitu) [miq]: At risk, 150,000 speakers; Misumalpan; spoken in Honduras and Nicaragua. Mixtepec Mixtec (Eastern Juxtlahuaca Mixtec, Mixteco de Oeste Central, Mixteco de San Juan Mixtepec) [mix]: Threatened, 9,170 speakers; Otomanguean; spoken in Mexico. Mixtepec Zapotec [zpm]: Threatened, 7,000 speakers; Otomanguean; spoken in Mexico. Mocho’ (Motozintlec, Motozintleco, Mochó) [mhc]: Critically endangered, fewer than 30 speakers; Mayan; spoken in Mexico. Mopán [mop]: Vulnerable, 10,000–15,000 speakers; Mayan; spoken in Guatemala and Belize. Nicaragua Creole English (Mískito Coast Creole, Nicaraguan Creole English) [bzk]: Vulnerable, 35,000–50,0000 speakers; pidgin or creole; spoken in Nicaragua. Nicaraguan Sign Language (Idioma de Senas de Nicaragua) [ncs]: Threatened, 3,000 speakers; sign language; spoken in Nicaragua. Northeastern Alta Mixtec (Apasco-Apoala Mixtec, Apasco Mixtec, Apoala Mixtec) [mip, vmq]: Threatened, 10,220 speakers; Otomanguean; spoken in Mexico. Northern Alta Mixtec (Coatzospan Mixtec, Mixteco de Coatzóspan, Mixteco de San Juan Coatzospan) [miz, xtu]: Threatened, 3,970 speakers; Otomanguean; spoken in Mexico. Northern Baja Mixtec (Chigmecatitlán Mixtec, Central Puebla Mixtec, Mixteco de Santa María Chigmecatitlán) [mii, mit, xtb]: Threatened, 8,320 speakers; Otomanguean; spoken in Mexico. Northern Chinantec (Chinanteco del norte, Ojitlán Chinantec, Chinantec) [chj]: Vulnerable, 37,900 speakers; Otomanguean; spoken in Mexico. Northern Pame (Xi’iuy, Pame, Pamean) [pmq]: Threatened, 5,620 speakers; Otomanguean; spoken in Mexico. Northern Tepehuán (Northern Tepehuan) [ntp]: Threatened, approximately 8,000 speakers; Uto-Aztecan; spoken in Mexico. Northern Totonac (Xicotepec de Juárez, Totonaco de Villa Juárez) [too]: Threatened, 3,000 speakers; Totonacan; spoken in Mexico. Northwestern Chinantec (jau jmai, Chinanteco del noroeste, Tlacoatzintepec Chinantec) [ctl]: Threatened, 1,720 speakers; Otomanguean; spoken in Mexico. Oaxaca Zoque (Chimalapa Zoque) [zoh]: Threatened, 5,216 speakers; Mixe-Zoquean; spoken in Mexico. Ocotlán (Ocotlán Zapotec, Zapotec, Zapoteco del Poniente de Ocotlán) [zac, zpv, zpn]: Vulnerable, 15,000 speakers; Otomanguean; spoken in Mexico. Ocuiltec (Ocuilteco, Okwilteko, Atzinca) [ocu]: Endangered, 737 speakers; Otomanguean; spoken in Mexico. Oluta Popoluca (Oluteco) [plo]: Severely endangered, 12 speakers; Mixe-Zoquean; spoken in Mexico. Opata-Eudeve (Opata, Endeve, Eudeve) [opt]: Dormant; Uto-Aztecan; formerly spoken in Mexico. Original Costa Rican Sign Language (Old Costa Rican Sign Language) [no ISO 639-3 code]: Critically endangered; sign language; spoken in Costa Rica. Paipai (Pai’pai, Akwa’ala, Cuñeil) [ppi]: Severely endangered, fewer than 100 speakers; Cochimi-Yuman; spoken in Mexico. Panamanian Sign Language (Lengua de Señas Panameñas) [lsp]: Threatened; sign language; spoken in Panama. 154

The world’s endangered languages

Papabuco (Papabuco, Zapoteco de San Juan Elotepec, Northern Zapotec) [zte, zpw, zpz]: Threatened, 5,070 speakers; Otomanguean; spoken in Mexico. Papantla Totonac (Lowland Totonac, Lowland Totonako) [top]: Vulnerable, 80,000 speakers; Totonacan; spoken in Mexico. Pech (Paya, Seco, Bayano) [pay]: Endangered, 600 speakers; Chibchan; spoken in Honduras. Petapa Zapotec (Petapa Zapotec) [zpe]: Threatened, 8,000 speakers; Otomanguean; spoken in Mexico. Pima Bajo (Lower Piman, Mountain Pima, Névome) [pia]: Endangered; Uto-Aztecan; spoken in Mexico. Pipil (Nahuat, Nawat, Nahuate) [ppl]: Critically endangered, approximately 97 speakers; UtoAztecan; spoken in El Salvador. Pisaflores Tepehua [tpp]: Threatened, 4,000 speakers; Totonacan; spoken in Mexico. Popoloca [pbf, pbe, pow, poe, pps, pls, pca]: Vulnerable, 17,964 speakers; Otomanguean; spoken in Mexico. Poqomam (Pokomam, Pocomám) [poc]: Threatened, 9,548 speakers; Mayan; spoken in Guatemala. Poqomchi’ [poh, pob]: Vulnerable, 69,716 speakers; Mayan; spoken in Guatemala. Puerto Rican Sign Language (PRSL) [psl]: Vulnerable, 8,000–40,000 speakers; sign language; spoken in Puerto Rico. Purepecha (Tarascan, Tarasco, Phorhépecha) [tsz, pua]: Threatened, 117,221 speakers; isolate; spoken in Mexico. Q’anjob’al (Eastern Kanjobal) [kjb]: Vulnerable, 99,211 speakers; Mayan; spoken in Guatemala. Q’eqchi’ (K’ekchí, Kekch, Kekchi) [kek]: Vulnerable, 726,723 speakers; Mayan; spoken in Guatemala and Belize. Rama (Melchora, Voto, Boto) [rma]: Critically endangered, 29 speakers; Chibchan; spoken in Nicaragua. Rincón (Sierra de Villa Alta, Yalálog) [zar, zsr]: Threatened, 5,000 speakers; Otomanguean; spoken in Mexico. Sakapulteko (Sacapultec, Sakapulteco) [quv]: Threatened, 3,940 speakers; Mayan; spoken in Guatemala. Salvadoran Lenca (Chilanga) [no ISO 639-3 code]: Dormant; Lencan; formerly spoken in El Salvador. Salvadoran Sign Language (El Salvadoran Sign Language, Lengua de Señas Salvadoreñas, LESSA) [esn]: Vulnerable; sign language; spoken in El Salvador. San Agustín Mixtepec Zapotec [ztm]: Severely endangered, 59 speakers; Otomanguean; spoken in Mexico. San Miguel Creole French [scf]: Critically endangered, three speakers; pidgin or creole; spoken in Panama. Sayula Popoluca (Popoluca de Sayula, Sayulateco) [pos]: Threatened, 3,025 speakers; MixeZoquean; spoken in Mexico. Seri [sei]: Threatened, 716–1,000 speakers; isolate; spoken in Mexico. Sierra Chinantec (jmiih dzä mo’, Chinanteco de la Sierra, Valle Nacional Chinantec) [chq, cco, cvn]: Vulnerable, 10,440 speakers; Otomanguean; spoken in Mexico. Sierra Popoluca (Highland Popoluca, Soteapanec, Soteapan Zoque) [poi]: Vulnerable, 26,017 speakers; Mixe-Zoquean; spoken in Mexico. Sierra Totonac (Highland Totonac, Highland Totonako, Totonaco de la Sierra) [tos]: At risk, 120,000 speakers; Totonacan; spoken in Mexico. 155

Anna Belew et al.

Sipakapa (Sipacapa, Sipacapense, Sipakapense) [qum]: Threatened, 6,344 speakers; Mayan; spoken in Guatemala. Southern Baja Mixtec (Western Juxtlahuaca Mixtec, Coicoyán Mixtec, Mixteco del Oeste de Juxtlahuaca) [jmx, miy, vmc]: Vulnerable, 47,500 speakers; Otomanguean; spoken in Mexico. Southern Highland Mixe (Tlahuitoltepec Mixe, West Central Mixe) [mxp]: Threatened, 3,617 speakers; Mixe-Zoquean; spoken in Mexico. Southern Tepehuán (Southern Tepehuan, Southeastern Tepehuan, Southwestern Tepehuan) [stp]: Vulnerable, 28,917 speakers; Uto-Aztecan; spoken in Mexico. Sumo (Sumo-Mayangna, Sumu, Ulwa) [sum, yan, ulw]: Threatened, approximately 10,000 speakers; Misumalpan; spoken in Honduras and Nicaragua. Tarahumara (Rarámuri, Ralámuli, Tubare) [tar, thh, tcu, twr, tac]: Vulnerable, 85,018 speakers; Uto-Aztecan; spoken in Mexico. Teco (Teko, Tectiteco, Tectitec) [ttc]: Threatened, 5,900 speakers; Mayan; spoken in Guatemala and Mexico. Tejalapan Zapotec (Zapoteco de San Felipe Tejalapan, Zapoteco de Tejalápam) [ztt]: Endangered, 120 speakers; Otomanguean; spoken in Mexico. Teribe (Tiribí, Tirub, Terraba) [tfr]: Threatened, 3,000 speakers; Chibchan; spoken in Costa Rica and Panama. Texistepec (Popoluca de Texistepec, Texistepec Zoque) [poq]: Critically endangered, one speaker; Mixe-Zoquean; spoken in Mexico. Tezoatlan Mixtec (Cacaloxtepec Mixtec, Huajuapan Mixtec, Mixteco de Cacaloxtepec) [miu, mxb]: Threatened, 5,540 speakers; Otomanguean; spoken in Mexico. Tilapa Otomí (Otomí, Othomí) [otl]: Endangered, 400 speakers; Otomanguean; spoken in Mexico. Tlachichilco Tepehua (Tlachichilco Tepewa) [tpt]: Threatened, 3,000 speakers; Totonacan; spoken in Mexico. Tlacolula [zab, ztt, ztj, zaw, zaq, zpf]: No estimate available; Otomanguean; spoken in Mexico. Tlacolulita Zapotec (Southeastern Yautepec Zapotec, Zapoteco de Asunciùn Tlacolulita) [zpk]: Endangered, 140 speakers; Otomanguean; spoken in Mexico. Tlapanec (Tlappanec, Tlapaneco, Me’pa Tlapanec) [tpx, tpc, tcf, tpl]: At risk, 120,072 speakers; Otomanguean; spoken in Mexico. Tojolabal (Tojolab’al) [toj]: Threatened; Mayan; spoken in Mexico. Tol (Eastern Jicaque, Tolpan, Jicaque) [jic]: Endangered, 250 speakers; Jicaquean; spoken in Honduras. Totonac [toc, tlp, tos, top, tcw, tku, tqt, too, tlc]: At risk, 244,033 speakers; Totonacan; spoken in Mexico. Totontepec Mixe (Northern Highland Mixe, Ayuk, Northwestern Mixe) [mto]: Threatened, 5,200 speakers; Mixe-Zoquean; spoken in Mexico. Trinidad and Tobago Sign Language (Trinidad Sign Language, TTSL, TSL) [lst]: Threatened, 2,000 speakers; sign language; spoken in Trindad and Tobago. Triqui [trs, trc, trq]: Vulnerable, 25,883 speakers; Otomanguean; spoken in Mexico. Tz’utujil (Tzutuhil, Tzutujil) [tzj]: Vulnerable, 47,669 speakers; Mayan; spoken in Guatemala. Tzeltal (Tseltal, Tzendal) [tzb, tzh]: At risk, 261,084 speakers; Mayan; spoken in Mexico. Tzotzil (Ts’ots’il) [tzc, tze, tzu, tzs, tzo, tzz]: At risk, 229,203 speakers; Mayan; spoken in Mexico. Upper Southeastern Chinantec (jumi dsa mojai, Chinanteco del sureste alto, Ozumacín Chinantec) [chz]: Threatened, 3,140 speakers; Otomanguean; spoken in Mexico. 156

The world’s endangered languages

Upper West-Central Chinantec (jaú jm, jmiih kia’ dzä jii’, Chinanteco del oeste central alto) [cuc]: Threatened, 7,410 speakers; Otomanguean; spoken in Mexico. Uspanteko (Uspanteco, Uspantec) [usp]: Threatened, 1,231 speakers; Mayan; spoken in Guatemala. Villalta [zad, zav, zpu, zpq, ztc, zat]: No estimate available; Otomanguean; spoken in Mexico. Western Alta Mixtec (Itundujia Mixtec, Eastern Putla Mixtec, Mixteco de Santa Cruz Itundujia) [mce, mdv, meh, mib, mie, mig, mpm, mvg, xti, xtj, xtl, xtm, xtn, xtt]: Vulnerable, 61,530 speakers; Otomanguean; spoken in Mexico. Western Baja Mixtec (Mixteco del Oeste Alto) [no ISO 639-3 code]: No estimate available; Otomanguean; spoken in Mexico. Western Chinantec (Jujma, Chinanteco del oeste, Sochiapám Chinantec) [cso]: Threatened, 3,590 speakers; Otomanguean; spoken in Mexico. Western Highland Chatino (Sierra Occidental Chatino, Chatino de la Zona Alta Occidental) [ctp]: Vulnerable, 12,000 speakers; Otomanguean; spoken in Mexico. Western Huasteca Nahuatl (Western Nahuatl) [nhw]: At risk, 400,000 speakers; Uto-Aztecan; spoken in Mexico. Yaqui (Yaki, Cahita, Yaqui-Mayo) [yaq]: Vulnerable, 21,000 speakers; Uto-Aztecan; spoken in Mexico and the United States of America. Yautepec Zapotec (Zapoteco de San Bartolo Yautepec) [zpb]: Endangered, 310 speakers; Otomanguean; spoken in Mexico. Yucatec (Yukateko) [yua]: At risk, 706,000 speakers; Mayan; spoken in Mexico and Belize. Yucatec Maya Sign Language (Nohya Sign Language) [msd]: Endangered, approximately 500 speakers; sign language; spoken in Mexico. Zaachila Zapotec (Oaxaca Zapotec) [ztx]: Endangered, 550 speakers; Otomanguean; spoken in Mexico. Zacatepec Chatino (Chatino de San Marcos Zacatepec) [ctz]: Threatened, 1,000 speakers; Otomanguean; spoken in Mexico. Zimatlán [zph, zpp, zpl]: No estimate available; Otomanguean; spoken in Mexico.

The Near East and Central Asia Raina Heaton Linguistic diversity in the Near East and Central Asia The region here called the “Near East and Central Asia” includes those languages spoken in the Middle East, i.e. in Iran, Iraq, Afghanistan, Israel, Syria, Oman, Kuwait, Yemen, Jordan, the United Arab Emirates, Lebanon, and Saudi Arabia, as well as those languages in the neighboring Central Asian countries of Tajikistan, Kazakhstan, Kyrgyzstan, and Uzbekistan. This region also includes several languages that are spoken in the border areas between Iran and Turkey, as well as Afghanistan and Pakistan. There are a number of endangered language communities in Central Asia (primarily in Kazakhstan, but also Uzbekistan) which are diaspora communities that migrated there from parts of the former Soviet Union, and as such are discussed in the regional overview for Europe and Russia. Additionally, many of the endangered languages that have communities of speakers in Kazakhstan and Kyrgyzstan are also spoken in China. Those languages that are spoken in both Central Asia and China are discussed in the regional overview of East Asia. As such, this 157

Anna Belew et al.

section primarily includes endangered languages of the Near East, based on where they are currently (as opposed to historically) spoken in the region. The majority of the endangered languages of the Near East and Central Asia belong to two genetic groups: Indo-Iranian languages (a branch of Indo-European) and Semitic languages (a branch of Afro-Asiatic). The region is also home to one endangered sign language (Al-Sayyid Bedouin Sign Language). A few languages in the region belong to other language families, namely Khalaj, a Turkic language spoken primarily in Iran, and Ladino (also known as JudeoSpanish), a Romance (Indo-European) language spoken in Turkey and Israel. There are also two other dormant languages in the region, Yevanic (Greek, Indo-European) which was spoken primarily in Israel after the population was devastated by the Holocaust, and Mogholi (Mongolic), formerly spoken by the members of Genghis Khan’s army stationed in Afghanistan. Additionally, there are large numbers of Yiddish speakers in Israel, and pockets of various Romani-speaking communities can be found in the Near East and Central Asia. All Yiddish and Romani language groups are covered in the overview of Europe and Russia. The issues of languages vs. dialects and appropriate classification were particularly important in a region where tribal/ethnic identities, as well as religious affiliation, are tied into language politics. For example, with respect to religious affiliation, there are Jewish and Christian varieties of Northeastern Neo-Aramaic that can be quite distinct, despite being spoken in the same area. In terms of ethnic identities, the Šuḡnī group of East Iranian Pamir languages, despite being at least somewhat mutually intelligible, do not consider themselves to speak a dialect of any other group’s language, and as such they have often been treated as a group of closely related languages (Buddruss 1988). Given these issues, the Catalogue includes a number of new divisions and languages in the region to best reflect the current scholarly consensus. Most notably, we have included a number of recently recognized Iranian languages (e.g. Garma’i, Habib Borjian p.c. 2017), and classified the Northeastern Neo-Aramaic varieties into language groups under more accurate labels. We also necessarily had to deal with Arabic dialects that are not all mutually intelligible, and that are becoming endangered due to the spread of national languages. We have attempted to accurately represent these non-mutually intelligible endangered varieties of Arabic in the Catalogue (e.g. Cypriote Spoken Arabic, Uzbeki Spoken Arabic, and Judeo-Yemeni Arabic). Those Arabic varieties in North Africa are discussed in the regional overview of Africa.

Causes of language endangerment in the Near East and Central Asia The region between Russia, India, Europe, and Northern Africa has been the location of intense conflict and territorial disputes, which have devastated and/or displaced many communities and contributed significantly to the endangerment of local languages. For example, speakers of Aširat Northeastern Neo-Aramaic were massacred by the Ottomans during World War I and fled from the mountains of Hakkari in Turkey to Iran, Iraq, and Syria (Häberl forthcoming:29), and the language is currently endangered. More recently, due to the war in Afghanistan, speakers of Savi (Sawi, Indo-Iranian) were dispersed into refugee camps in Chitral and Dir districts of Pakistan and their current situation is unknown (Bashir 2003). Additionally, many speakers of Jewish languages and varieties in the region have migrated to Israel, where their children often do not grow up learning their parents’ native language. Urbanization, as well as the promotion of national languages in school, in government, and in the media has also contributed to the endangerment of many of the smaller languages of the region. There is very little support for local languages, and mandatory military service, universal education (which has led to rising literacy rates), and assimilationist policies all contribute 158

The world’s endangered languages

to bilingualism/multilingualism and language shift throughout the region (Häberl forthcoming:2). The largest regional languages which exert pressure on smaller languages are Arabic and Persian, as well as Hebrew in Israel. Speakers of endangered languages also often must know other national languages for their country of residence, including Russian, Tajik, Uzbek, Pashto/Urdu, and Kurdish in Kurdish areas. Other languages that speakers of endangered languages in the region have been reported to speak include Shina (Indo-Iranian, Pakistan), Khowar (Indo-Iranian, Pakistan), Dari (Indo-Iranian, Afghanistan), Lori (Indo-Iranian, Iran), Bandari (Indo-Iranian), Iran), Qashqa’i (Turkic, Iran), Qeshmi (Indo-Iranian, Iran), and Pashai (Indo-Iranian, Afghanistan). With respect to urbanization, not only have speakers migrated in search of economic opportunity, but in Iran, those communities located near urban centers as well as large thoroughfares have shifted more quickly to larger languages, and the local languages in those areas are more endangered as a result. Additionally, ethnic identity and attitude towards one’s mother tongue influence language maintenance and vitality. Communities that take pride in their native languages and value their tribal identities are more likely to continue to speak their languages than communities that do not gain social prestige from their linguistic and ethnic affiliations (Borjian 2009:75–76). However, this often serves to enforce existing power dynamics, where speakers of larger languages like Balochi value their language, whereas languages like AlviriVidari which lack social capital are not highly valued by their speakers and are more endangered (Alviri-Vidari, according to the LEI, is “Severely Endangered”).

The current status of endangered languages in the Near East While there are not a huge number of languages, endangered or otherwise, in the Near East and Central Asia in comparison to other regions, language endangerment is a significant problem among the remaining minority groups in the region. There are 103 languages in danger in the region, most of which are spoken primarily in Iran and Afghanistan. There are five dormant or likely dormant languages in the region (Yevanic, Mogholi, Mlaḥsô, and likely also WotapuriKatarqalai and Tirahi), which have ceased to be spoken in the past ~60 years. There are also ten “Critically Endangered” and “Severely Endangered” languages, including Central Jewish Neo-Aramaic, Neo-Mandaic, Kuhpayi, Jowshaqani, Garma’i, Inter-Zab Jewish Neo-Aramaic, Kho’ini, Alviri-Vidari, Baṭḥari, and Shumashti. This likely also includes four Judeo-Median languages,16 which have been described as “on the verge of extinction” (Borjian 2015). An additional 15 languages are “Endangered” (i.e., moderately endangered, the medium rank on the Language Endangerment Index). Most are either “Threatened” or “Vulnerable.” However, a pervasive problem in this region is a lack of comprehensive documentation (Owens 2007:264–265). Even if a language is well documented in terms of its grammar, often basic information about the vitality of the language was not published alongside it. This is partly due to the aims of the researchers (e.g. with focus on comparative work as opposed to comprehensive documentation), and partly due to the fact that historically, researchers in the region often worked with speakers in cities outside the villages for logistical reasons. It is therefore not surprising that the vitality assessments for over half of the endangered languages in the region are only rated by the Catalogue as 20% certain (i.e. based on a single factor [most commonly the absolute speaker number]). Additionally, there are 14 languages in the region for which we have no current vitality estimate,17 and there are an additional ~10 languages for which we know that the most recent primary vitality information is at least 20 years old (some as old as the early 1960s, e.g., for Koh’ini). On the other hand, there are 12 languages for which we have 100% of the information necessary to calculate vitality, 159

Anna Belew et al.

which range from “Severely Endangered” to “Vulnerable.”18 See Häberl (forthcoming) and also Owens (2007) for additional overviews of language endangerment in the Near East and Central Asia. Below is an alphabetical list of languages of the Near East and Central Asia included in the Catalogue of Endangered Languages.

Endangered languages of the Near East and Central Asia Aftari [no ISO 639-3 code]: Endangered, fewer than 500 speakers; Indo-European; spoken in Iran. Al-Sayyid Bedouin Sign Language (ABSL) [syy]: Endangered, at least 150 speakers; sign language; spoken in Israel. Alviri-Vidari [avd]: Severely endangered; Indo-European; spoken in Iran. Ashkun (Ashkund, Ashkuni, Wamayi) [ask]: Threatened, fewer than 7,000 speakers; IndoEuropean; spoken in Afghanistan. Ashtiani (Astiani, Ashtiyani) [atn]: Vulnerable, 21,100 speakers; Indo-European; spoken in Iran. Aširat Northeastern Neo-Aramaic (Asirat Northeastern Neo-Aramaic, Assyrian, NENA) [no ISO 639-3 code]: Endangered; Afro-Asiatic; spoken in Turkey, Iraq, and Syria. Bajelani (Bajalani, Gurani, Chichamachu) [bjm]: Vulnerable, 20,000 speakers; Indo-European; spoken in Iran and Iraq. Bartangi (Bartang) [no ISO 639-3 code]: Vulnerable, 2,425 speakers; Indo-European; spoken in Tajikistan. Bashkardi (Bashaka) [bsg]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Baṭḥari (Boṭaḥārī, Bathari, Bautahari) [bhm]: Severely endangered, fewer than 100 speakers; Afro-Asiatic; spoken in Yemen and Oman. Bukhori (Bukharian, Bukhari, Bukharin) [bhh]: At risk, 110,000 speakers; Indo-European; spoken in Israel and Uzbekistan. Central Jewish Neo-Aramaic (Barzani Jewish Neo-Aramaic, NENA) [no ISO 639-3 code]: Critically endangered; Afro-Asiatic; spoken in Israel. Eshtehardi [esh]: No estimate available; Indo-European; spoken in Iran. Farvi (Farvigi) [no ISO 639-3 code]: Endangered, approximately 500 speakers; Indo-European; spoken in Iran. Garma’i [no ISO 639-3 code]: Severely endangered, ten speakers; Indo-European; spoken in Iran. Gawar-Bati (Gawar-bātī, Narisātī, Narsātī) [gwt]: Vulnerable, 8,000–10,000 speakers; IndoEuropean; spoken in Afghanistan and Pakistan. Gazi [gzi]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Gozarkhani [goz]: No estimate available; Indo-European; spoken in Iran. Grangali (Nangalami) [nli]: Threatened, 5,000 speakers; Indo-European; spoken in Afghanistan. Gurani (Howrami, Hawrami, Hewrami) [hac]: Vulnerable, fewer than 100,000 speakers; Indo-European; spoken in Iran and Iraq. Ḥarsusi (Ḥarsūsī, Harsusi, Hersyet) [hss]: Endangered, approximately 700 speakers; AfroAsiatic; spoken in Oman. Harzani (Harzandi) [hrz]: Vulnerable, 28,100 speakers; Indo-European; spoken in Iran. Hobyot (Habyot, Hobyót, Hobi) [hoh]: Endangered, 400 speakers; Afro-Asiatic; spoken in Yemen and Oman. 160

The world’s endangered languages

Hulaulá (Judeo-Aramaic, Lishana Noshan, Lishana Axni) [huy]: Threatened, approximately 9,500 speakers; Afro-Asiatic; spoken in Israel and Iran. Inter-Zab Jewish Neo-Aramaic (Lishanid Noshan, Arbel Jewish Neo-Aramaic, Lishana Didán) [aij]: Severely endangered; Afro-Asiatic; spoken in Israel and Iraq. Ishkashimi (Ishkashmi, Ishkashim, Eshkashimi) [isk]: Threatened, 2,500 speakers; IndoEuropean; spoken in Tajikistan and Afghanistan. Jarqu’i [no ISO 639-3 code]: Threatened, fewer than 14,000 speakers; Indo-European; spoken in Iran. Jewish Azerbaijani Neo-Aramaic (Lishán Didán, Lishanán, ‫[ )نادد ناشل‬trg]: Threatened, 4,450 speakers; Afro-Asiatic; spoken in Azerbaijan, Georgia, Israel, and Iran. Jibbali (Jibbālī, Shehri, Śḥeri) [shv]: Threatened, approximately 5,000 speakers; Afro-Asiatic; spoken in Oman. Jowshaqani [no ISO 639-3 code]: Severely endangered, approximately 400 speakers; IndoEuropean; spoken in Iran. Judeo-Hamadani [no ISO 639-3 code]: Endangered, fewer than 1,000 speakers; IndoEuropean; spoken in Israel and Iran. Judeo-Iraqi Arabic (Judeo-Baghdadi) [yhd]: At risk, 151,820 speakers; Afro-Asiatic; spoken in Israel and Iraq. Judeo-Isfahani (Jidi) [no ISO 639-3 code]: No estimate available; Indo-European; spoken in Iran. Judeo-Kashani [no ISO 639-3 code]: No estimate available; Indo-European; spoken in Iran. Judeo-Shirazi (Jidi) [no ISO 639-3 code]: No estimate available; Indo-European; spoken in Iran. Judeo-Yazdi [no ISO 639-3 code]: No estimate available; Indo-European; spoken in Iran. Judeo-Yemeni Arabic [jye]: Vulnerable, 51,000 speakers; Afro-Asiatic; spoken in Israel and Yemen. Kabatei [xkp]: No estimate available; Indo-European; spoken in Iran. Kajali [xkj]: No estimate available; Indo-European; spoken in Iran. Karingani [kgn]: Vulnerable, 17,600 speakers; Indo-European; spoken in Iran. Kati (Kata-vari, Bashgali, Kativiri) [bsh]: Threatened, 3,700–5,100 speakers; Indo-European; spoken in Afghanistan and Pakistan. Kesha’i (Keša’i, Keši) [no ISO 639-3 code]: Endangered, 200 speakers; Indo-European; spoken in Iran. Khalaj (Turkic Khalaj, Chaladsch, Arghu) [klj]: Vulnerable, 20,000 speakers; Turkic, Southern; spoken in Iran. Kho’ini [xkc]: Severely endangered; Indo-European; spoken in Iran. Khufi (Khuf, Chuf) [no ISO 639-3 code]: Vulnerable, approximately 2,380 speakers; IndoEuropean; spoken in Afghanistan. Khunsari [kfm]: Vulnerable, 21,100 speakers; Indo-European; spoken in Iran. Koroshi [ktl]: Endangered, 180 speakers; Indo-European; spoken in Iran. Kuhpayi (Kuhpāʾi) [no ISO 639-3 code]: Severely endangered; Indo-European; spoken in Iran. Kumzari (Kumzāri, Komzāri) [zum]: Vulnerable, approximately 4,000 speakers; IndoEuropean; spoken in Oman. Ladino (Judeo-Spanish, Sephardic, Hakitia) [lad]: At risk, 400,000 speakers; Indo-European; spoken in Israel, Greece, Turkey, and other countries. Lari (Larestani, Achomi, Fars) [lrl]: Vulnerable, 80,000 speakers; Indo-European; spoken in Iran. 161

Anna Belew et al.

Lasgerdi [lsa]: Threatened, 1,000 speakers; Indo-European; spoken in Iran. Lishana Deni (Judeo-Aramaic, Lishan Hudaye, Lishan Hozaye) [lsd]: Threatened, 7,500 speakers; Afro-Asiatic; spoken in Israel and Iraq. Luwati (Lawatiyya, Lawatiya, Khojki) [luv]: Vulnerable, 5,000 speakers; Indo-European; spoken in Oman. Maraghei [vmh]: No estimate available; Indo-European; spoken in Iran. Mehri (Mehrī, Mahri, Meḥri) [gdq]: At risk, 100,000 speakers; Afro-Asiatic; spoken in Yemen, Oman, and Kuwait. Meymai (Meyma’i) [no ISO 639-3 code]: Threatened, approximately 2,000 speakers; IndoEuropean; spoken in Iran. Mlaḥsô (Mlahsö, Mlaḥso, Suryoyo) [lhs]: Dormant; Afro-Asiatic; formerly spoken in Turkey and Syria. Mogholi (Moghol, Mogul, Mogol) [mhj]: Dormant; Mongolic; formerly spoken in Afghanistan. Munji (Munjani, Munjhan, Munjiwar) [mnj]: Threatened, 3,770 speakers; Indo-European; spoken in Afghanistan. Natanzi [ntz]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Nayini (Biyabanak) [nyq]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Neo-Mandaic (Mandaean, Modern Mandaic) [mid]: Severely endangered, 100–200 speakers; Afro-Asiatic; spoken in Iran and Iraq. Northern Northeastern Neo-Aramaic (Assyrian, NENA, Hakkari) [no ISO 639-3 code]: Vulnerable, 10,000–20,000 speakers; Afro-Asiatic; spoken in Turkey. Ormuri (Oormuri, Urmuri, Ormur) [oru]: Threatened, approximately 3,000 speakers; IndoEuropean; spoken in Afghanistan and Pakistan. Oroshor (Oroshori, Roshorvi) [no ISO 639-3 code]: Vulnerable, 1,950 speakers; IndoEuropean; spoken in Tajikistan. Parachi [prc]: Threatened, 6,900 speakers; Indo-European; spoken in Afghanistan. Parya (Asiatic Romany, Afghana-Yi Nasfurush, Afghana-Yi Siyarui) [paq]: Threatened, fewer than 7,500 speakers; Indo-European; spoken in Uzbekistan, Tajikistan, and Afghanistan. Prasuni (Wasi-Wari, Prasun, Veruni) [prn]: Vulnerable, 8,000 speakers; Indo-European; spoken in Afghanistan. Razajerdi [rat]: No estimate available; Indo-European; spoken in Iran. Razihi (Jabal Rāziḥ) [no ISO 639-3 code]: Vulnerable, approximately 24,000 speakers; AfroAsiatic; spoken in Yemen. Rudbari [rdb]: No estimate available; Indo-European; spoken in Iran. Rushani (Rushani, Rushan, Oroshani) [no ISO 639-3 code]: Vulnerable; Indo-European; spoken in Tajikistan and Afghanistan. Sangisari (Sangesari, Mahdi Shahri, Mehdishahri) [sgr]: Vulnerable, 36,000 speakers; IndoEuropean; spoken in Iran. Savi (Sawi, Sauji, Sau) [sdg]: No estimate available; Indo-European; spoken in Afghanistan and Pakistan. Sedehi (Varnosfaderani) [no ISO 639-3 code]: Vulnerable, fewer than 200,000 speakers; IndoEuropean; spoken in Iran. Semnani (Simnani) [smy]: Vulnerable, 60,000 speakers; Indo-European; spoken in Iran. Senaya (Christian Neo-Aramaic, Shan Sray) [syn]: Endangered, fewer than 200 speakers; AfroAsiatic; spoken in Iran. Shahmirzadi [srz]: No estimate available; Indo-European; spoken in Iran. Shughni (Shugnan-Rushan, Shighni, Khugni) [sgh]: Vulnerable, 80,000–100,000 speakers; Indo-European; spoken in Tajikistan and Afghanistan. 162

The world’s endangered languages

Shumashti (Šumāštī, Dardu) [sts]: Severely endangered; Indo-European; spoken in Afghanistan. Sivandi [siy]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Soi (Soh, Sohi, So’i) [soj]: Threatened, 7,030 speakers; Indo-European; spoken in Iran. Soqoṭri (Saqatri) [sqt]: Endangered, fewer than 50,000 speakers; Afro-Asiatic; spoken in Yemen and United Arab Emirates. Sorkhei [sqo]: Vulnerable, 10,000 speakers; Indo-European; spoken in Iran. Southern Northeastern Neo-Aramaic (Assyrian, NENA) [no ISO 639-3 code]: Threatened; Afro-Asiatic; spoken in Turkey. Tajiki Spoken Arabic (Tajiki Arabic, Jugari, Bukhara Arabic) [abh]: Threatened, 6,000 speakers; Afro-Asiatic; spoken in Uzbekistan, Tajikistan, and Afghanistan. Takestani (Takistani, Southern Tati) [tks]: At risk, 220,000 speakers; Indo-European; spoken in Iran. Talysh (Talish, ‫یشلات نابز‬, Talış dili) [tly]: At risk, 912,000 speakers; Indo-European; spoken in Azerbaijan and Iran. Tirahi (Tirāhī, Dardu) [tra]: Dormant; Indo-European; formerly spoken in Afghanistan. Tregami (Trigami, Tregāmī, Gambīrī) [trm]: Threatened, 1,000 speakers; Indo-European; spoken in Afghanistan. Turoyo (Ṭurōyō, Suryoyo, Syryoyo) [tru]: Vulnerable, 84,000 speakers; Afro-Asiatic; spoken in Turkey, Iraq, Syria, and Lebanon. Upper Taromi [tov]: No estimate available; Indo-European; spoken in Iran. Urmia Northeastern Neo-Aramaic (Assyrian, NENA) [no ISO 639-3 code]: Vulnerable, fewer than 20,560 speakers; Afro-Asiatic; spoken in Iran. Uzbeki Spoken Arabic (Central Asian Arabic) [auz]: Endangered, 700 speakers; Afro-Asiatic; spoken in Uzbekistan. Vafsi [vaf]: Vulnerable, approximately 20,000 speakers; Indo-European; spoken in Iran. Waigali (Kalasa-alā, Waigalī, Wai-alā) [wbk]: Threatened, 1,500 speakers; Indo-European; spoken in Afghanistan. Wakhi (Guhjali, Wakhani, Wakhigi) [wbl]: Vulnerable, 37,570 speakers; Indo-European; spoken in Tajikistan, Afghanistan, Pakistan, and China. Western Neo-Aramaic (Maalula, Siryon, Loghtha Siryanoytha) [amw]: Endangered; AfroAsiatic; spoken in Syria. Wotapuri-Katarqalai (Wotapūrī-Katāqalāī, Wotapuri, Dardu) [wsv]: Dormant; IndoEuropean; formerly spoken in Afghanistan. Yaghnobi (Yagnobi, Yaghnob) [yai]: Vulnerable, approximately 13,500 speakers; IndoEuropean; spoken in Tajikistan. Yazgulyami (Iazgulem, Yazgulam, Yazghulami) [yah]: Vulnerable, 6,000–9,000 speakers; IndoEuropean; spoken in Tajikistan. Yevanic (Judeo-Greek, Romaniyot, Romaniote) [yej]: Dormant; Indo-European; formerly spoken in Israel. Zefra’i [no ISO 639-3 code]: Threatened, fewer than 3,000 speakers; Indo-European; spoken in Iran. Zemiaki (Zamyaki, Jamlám basa) [no ISO 639-3 code]: Endangered, 400–500 speakers; IndoEuropean; spoken in Afghanistan. Zoroastrian Dari (Behdināni) [gbz]: Endangered; Indo-European; spoken in Iran.

163

Anna Belew et al.

North America Lyle Campbell Linguistic diversity in North America19 Linguistic diversity in the Americas is profound, with over 40% of all the language families (including isolates) of the world located here.20 North America is characterized not only by extensive linguistic diversity, but also by the extremely precarious state of most of its languages. At the time of first European contact, there were more than 300 indigenous languages here. The North American Indian languages are so diverse that there is no feature or complex of features shared by all. They are classified into some 57 language families, including 14 larger language families, 18 smaller language families, and 25 language isolates (languages with no known relatives, thus language families with but a single member language). Geographically, too, the diversity of some areas is notable. Some 37 of the 57 language families lie west of the Rocky Mountains, and 20 of them solely in California. California alone thus shows more linguistic diversity than all of Europe (Golla et al. 2008; Campbell 1997.) When Europeans arrived in North America they encountered 312 languages that we are sure about, of which 150 have no known remaining native speakers (48%). Of 280 languages at that time in territory now belonging to the US, only 76 are still spoken (27%), all of which are endangered (Golla et al. 2008; Campbell 1997).

Causes of language endangerment in North America It is a combination of the causes of language endangerment known from elsewhere that has caused languages of North America to become endangered, though with greater impact here and in Australia than anywhere else in the world. The attitudes, prejudices, and policies of the dominant colonial powers fanned language shift, predominantly to English, but also to French in eastern portions of Canada, and in earlier times, Russian, Spanish, Dutch, German, and Swedish all had their impact in certain regions. US and Canadian policy towards Native Americans involved forced cultural assimilation, with schools that punished speaking the native languages. The history of White-Indian interactions is replete with devastation from diseases brought from Europe, wars, massacres, and genocide, economic and religious pressures, starvation, etc. Economic factors have played a large role in motivating language shift in North America, with forced resettlement, dispersal, and migration of populations, Indian wars, gold rushes, resource depletion, forced changes in subsistence patterns, lack of economic opportunities, ongoing industrialization, destruction of habitat, globalization, etc. Political and social factors are also important factors: issues of the status of languages, discrimination, repression (as in the case of the boarding schools for North American Indians where, as mentioned, use of native languages was forbidden and their use punished), official language policies, etc. Subjective attitudes about the languages have been important causal factors. These include the symbolism attached to the dominant language (e.g. symbol of nation; socioeconomic symbol of civilization, progress and the future vs. of the past, etc.); stigmatization (low prestige of the endangered language); absence of institutional support (the minority language not represented in schools, academies, texts, the media); etc.

164

The world’s endangered languages

The current status of endangered languages in North America While the world’s 3,138 still-spoken endangered languages amount to 45.6% of all its 6,879 living languages,21 100% of North America’s 147 living indigenous languages are endangered. This difference shows the striking magnitude of language endangerment in North America. The Catalogue of Endangered Languages includes 204 languages of North America.22 Of these, only 156 actually still have speakers; 48 have no known speakers, as 22 are listed as “Dormant,” and 26 as “Awakening.” Two of the surviving endangered languages involve Hawai‘i (Hawaiian and Hawaiian Sign Language). Five of ELCat’s endangered languages of North America are sign languages, three of which are indigenous. There are two pidgin and creole languages, one of them indigenous (Chinook Wawa, a.k.a Chinook Jargon), and one mixed language, Michif (Copper Island Aleut [a.k.a. Mednuy Aleut] no longer has any known native speakers). Though 156 endangered languages are still spoken in North America, this figure is in a real sense misleading, since many of these are extremely endangered, and unless revitalization efforts are successful, will very soon have no remaining native speakers, just like the 48 that have recently lost their last remaining mother-tongue speakers. That is, almost certainly very soon the number of 156 surviving languages will shrink dramatically. ELCat finds that 24 languages have fewer than 10 speakers; 68 have fewer than 100. Of the 156 remaining languages, 53 are “Critically Endangered,” 36 “Severely Endangered,” and together 99 are “Critically” or “Severely Endangered.” Only one, Navajo, is listed as “At Risk,” ELCat’s level of least endangerment, and nine others as “Vulnerable,” the next level of increasing endangerment. There is intergenerational transmission in fewer than 20 of these languages, and each day fewer children are learning even these. Below is an alphabetical listing of the languages of North America included in the Catalogue of Endangered Languages.

Endangered languages of North America Achumawi (Achumawi, Achomawi, Pitt River) [acv]: Critically endangered, fewer than ten speakers; Palaihnihan; spoken in the United States of America. Acoma-Laguna (Western Keres Pueblo, Western Keresan, Acoma) [kjq]: Threatened, approximately 4,000 speakers; Keresan; spoken in the United States of America. Ahtna (Ahtena, Nabesna, Tanana) [aht]: Critically endangered, 30 speakers; Athabaskan-EyakTlingit; spoken in the United States of America. Alabama (Alibamu) [akz]: Endangered, 250–300 speakers; Muskogean; spoken in the United States of America. Aleut (Unangan, Unangam Tunuu, алеутский язык) [ale]: Endangered, 70–100 speakers; Eskimo-Aleut; spoken in Russia and the United States of America. Arapaho (Arrapahoe, Northern Arapaho, Arapaho-Atsina) [arp]: Severely endangered, approximately 1,000 speakers; Algic; spoken in the United States of America. Arikara (Ree, Ricara, Arikari) [ari]: Critically endangered, fewer than ten speakers; Caddoan; spoken in the United States of America. Assiniboine (Assiniboin, Stoney, Hohe) [asb]: Endangered, fewer than 150 speakers; Siouan; spoken in the United States of America and Canada. Atsugewi (Atsugewi, Hat Creek) [atw]: Dormant; Palaihnihan; formerly spoken in the United States of America. Attikamek (Cree, Atikamekw, Tête de Boule) [atj]: Vulnerable, approximately 3,000 speakers; Algic; spoken in Canada. 165

Anna Belew et al.

Barbareño (Šmuwič, Chumash) [boi]: Awakening; Chumashan; spoken in the United States of America. Bearlake (Bearlake Slavey, North Slavey) [no ISO 639-3 code]: Endangered, 580 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Blackfoot (Blackfeet, Pikanii, Siksika) [bla]: Endangered, 3,300 speakers; Algic; spoken in the United States of America and Canada. Caddo (Kadohadacho, Petit Caddo, Upper Nasoni) [cad]: Critically endangered, fewer than 25 speakers; Caddoan; spoken in the United States of America. Cahuilla [chl]: Critically endangered, six speakers; Uto-Aztecan; spoken in the United States of America. Catawba (Catawaba) [chc]: Awakening; Siouan; spoken in the United States of America. Cayuga [cay]: Severely endangered, approximately 100 speakers; Iroquoian; spoken in the United States of America and Canada. Central Alaskan Yup’ik (Yugcestun, Yugtun, Yupik) [esu]: Threatened, 10,400 speakers; Eskimo-Aleut; spoken in the United States of America. Central Pomo (Oat Valley, Ballo-Kai-Pomo) [poo]: Dormant; Pomoan; formerly spoken in the United States of America. Central Siberian Yupik (St. Lawrence Island Yupik, Bering Strait Yupik, чаплинский язык) [ess]: Threatened, 1,300 speakers; Eskimo-Aleut; spoken in the United States of America and Russia. Cherokee (ᏣᎳᎩ, Tsalagi, Tslagi) [chr]: Vulnerable, approximately 16,000 speakers; Iroquoian; spoken in the United States of America. Cheyenne [chy]: Endangered, 2,100 speakers; Algic; spoken in the United States of America. Chickasaw (Chikashshanompa’, Chicasa) [cic]: Endangered, approximately 600 speakers; Muskogean; spoken in the United States of America. Chinook Wawa (Chinook Jargon, Chinook Pidgin, Chinuk Wawa) [chn]: Critically endangered, a few speakers; pidgin or creole; spoken in the United States of America and Canada. Chipewyan (Dene, Montagnais, Dene Suline) [chp]: Threatened, 2,235 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Chitimacha [ctm]: Awakening; isolate; spoken in the United States of America. Chiwere (Oto, Iowa-Oto) [iow]: Awakening; Siouan; spoken in the United States of America. Choctaw [cho]: Vulnerable, 9,000–11,000 speakers; Muskogean; spoken in the United States of America. Coast Miwok [csi]: Dormant; Utian; formerly spoken in the United States of America. Cocopah (Cocopá, Cucupá) [coc]: Endangered, 300–600 speakers; Cochimi-Yuman; spoken in the United States of America and Mexico. Coeur d’Alene (Snchitsu’umshtsn, snčícuʔumšcn, Skitswish) [crd]: Critically endangered, two speakers; Salishan; spoken in the United States of America. Columbian (Nxaʔamxcín, Columbia-Wenatchi, Wenatchi-Columbia) [col]: Critically endangered, 25 speakers; Salishan; spoken in the United States of America. Comanche [com]: Severely endangered, fewer than 100 speakers; Uto-Aztecan; spoken in the United States of America. Crow (Apsaalooke, Apsaroka, Apsaloka) [cro]: Threatened, 3,000–4,000 speakers; Siouan; spoken in the United States of America. Cupeño (Agua Caliente) [cup]: Awakening; Uto-Aztecan; spoken in the United States of America. Dakelh (ᑕᗸᒡ, Carrier, les Porteurs) [crx, caf]: Severely endangered, 680 speakers; AthabaskanEyak-Tlingit; spoken in Canada. 166

The world’s endangered languages

Dane-Zaa (ᑕᓀ ᖚ, Beaver, Dunne-za) [bea]: Severely endangered, 156 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Danezāgé’ (Kaska, Kaska Dena, Nahani) [kkz]: Severely endangered, 250 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Dena’ina (Tanaina, K’naia-khotana, Kinayskiy) [tfn]: Severely endangered, approximately 75 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Dene K’e (Slavey, Dene Tha’) [xsl]: Threatened, approximately 3,260 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Diitiidʔaatx̣ (Nitinat, Ditidaht, Nitinaht) [dtd]: Critically endangered, seven speakers; Wakashan; spoken in Canada. Dogrib (Thlingchadine, Tłı̨chǫ) [dgr]: Endangered, 2,470 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. East Cree [crl]: Vulnerable, 12,000 speakers; Algic; spoken in Canada. Eastern Ojibwe (Eastern Ojibwa, Ojibway) [ojg]: Threatened, 1,500 speakers; Algic; spoken in Canada. Eastern Pomo (Clear Lake Pomo) [peb]: Dormant; Pomoan; formerly spoken in the United States of America. Eel River Athabaskan (Saiaz, Eel River Athapaskan) [no ISO 639-3 code]: Dormant; Athabaskan-Eyak-Tlingit; formerly spoken in the United States of America. Éy7á7juuthem (Comox, Mainland Comox, Sliammon) [coo]: Critically endangered, 36 speakers; Salishan; spoken in Canada. Eyak [eya]: Awakening; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Gitsenimx̱ (Gitksan, Gitsanimx̱, Gitksanimx) [git]: Severely endangered, 350 speakers; Tsimshian; spoken in Canada. Gros Ventre (Atsina, Gros Ventres, White Clay People) [ats]: Awakening; Algic; spoken in the United States of America. Gwich’in (Dinjii Zhuh K’yaa, Kutchin, Kuchin) [gwi]: Severely endangered, 550 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America and Canada. Hailhzaqvla (Heiltsuk, Bella Bella, Heiltsuk-Oweek’ala) [hei]: Critically endangered, 60 speakers; Wakashan; spoken in Canada. Halq’eméylem (Halkomelem, Hul’q’umi’num’, Halq’eméylem) [hur]: Endangered, 265 speakers; Salishan; spoken in the United States of America and Canada. Han (Han-Kutchin, Moosehide, Dawson) [haa]: Critically endangered, nine speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America and Canada. Hanis (Coos, Anasitch) [csz]: Dormant; Coosan; formerly spoken in the United States of America. Hare (North Slavey) [no ISO 639-3 code]: Severely endangered, 75 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Hawai‘i Sign Language (Hawai‘i Pidgin Sign Language, Pidgin Sign Language) [hps]: Critically endangered, 30 speakers; sign language; spoken in the United States of America. Hawaiian (Ōlelo Hawai‘i) [haw]: Severely endangered, approximately 300 speakers; Austronesian; spoken in the United States of America. Hidatsa (Gros Ventre, Minitari, Duan) [hid]: Critically endangered, approximately 100 speakers; Siouan; spoken in the United States of America. Holikachuk (Upper Innoko, Innoko) [hoi]: Dormant; Athabaskan-Eyak-Tlingit; formerly spoken in the United States of America. Hopi (Tusayan, Moki, Moqui) [hop]: Threatened, at least 5,000 speakers; Uto-Aztecan; spoken in the United States of America. 167

Anna Belew et al.

Hupa (Hoopa-Chilula, Hupa-Chilula-Whilkut, Hoopa) [hup]: Critically endangered, approximately one speaker; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Ingalik (Kaiyuh-khotana, Deg Hit’an, Deg Xinag) [ing]: Severely endangered, 14 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Inuit Sign Language (Inuit Uukturausingit, IUR) [iks]: Severely endangered, fewer than 40 speakers; sign language; spoken in Canada. Inuktitut (ᐃᓄᒃᑎᑐᑦ, Eastern Canadian Inuktitut) [ike]: Vulnerable, approximately 30,000 speakers; Eskimo-Aleut; spoken in Canada. Inupiaq (Inupiaqtun, Iñupiatun, Alaskan Inuit) [ipk]: Severely endangered, 26,644 speakers; Eskimo-Aleut; spoken in the United States of America and Canada. Jemez (Towa) [tow]: Threatened, approximately 3,000 speakers; Kiowa-Tanoan; spoken in the United States of America. Jicarilla Apache (Jicarilla, Hikariya) [apj]: Endangered, 300 speakers; Athabaskan-EyakTlingit; spoken in the United States of America. Kalaallisut (West Greenlandic) [kal]: Vulnerable, 53,000 speakers; Eskimo-Aleut; spoken in Greenland (Denmark). Kalispel-Spokane-Pend d’Oreille-Salish (Nqlispélišcn) [fla, spo]: Severely endangered, 70 speakers; Salishan; spoken in the United States of America. Kanza (Kansa, Kansas, Kaw) [ksk]: Awakening; Siouan; spoken in the United States of America. Karuk (Karok, Karuk, Quoratem) [kyh]: Critically endangered, fewer than 12 speakers; isolate; spoken in the United States of America. Kashaya (Southwestern Pomo, Kashayn) [kju]: Critically endangered, several dozen speakers; Pomoan; spoken in the United States of America. Kawaiisu [xaw]: Critically endangered, three speakers; Uto-Aztecan; spoken in the United States of America. Kickapoo (Kikapoo, Kikapú) [kic]: Threatened, approximately 1,100 speakers; Algic; spoken in the United States of America and Mexico. Kiowa (Kiowan) [kio]: Critically endangered, fewer than 100 speakers; Kiowa-Tanoan; spoken in the United States of America. Kiowa Apache (Oklahoma Apache, Plains Apache) [apk]: Dormant; Athabaskan-Eyak-Tlingit; formerly spoken in the United States of America. Klallam (Clallam, S’klallam, Na’klallam) [clm]: Awakening; Salishan; spoken in the United States of America. Klamath-Modoc (Lutuami, Modoc, Klamath) [kla]: Awakening; Plateau; spoken in the United States of America. Koasati (Coushatta) [cku]: Severely endangered, 200 speakers; Muskogean; spoken in the United States of America. Kodiak Russian Creole [no ISO 639-3 code]: Critically endangered, five speakers; pidgin or creole; spoken in Russia and the United States of America. Konkow (Maidu, Northwest Maidu, Meidoo) [mjd]: Critically endangered, three speakers; Maiduan; spoken in the United States of America. Koyukon (Denaakk’e, Ten’a) [koy]: Critically endangered, 150 speakers; Athabaskan-EyakTlingit; spoken in the United States of America. Ktunaxa (Kootenai, Kutenai, Kootenay) [kut]: Severely endangered, 26 speakers; isolate; spoken in the United States of America and Canada. 168

The world’s endangered languages

Kwak̓wala (Kwakiutl, Kwakwaka’wakw, Kwak’wala) [kwk]: Endangered, 195 speakers; Wakashan; spoken in Canada. Ladino (Judeo-Spanish, Sephardic, Hakitia) [lad]: At risk, 400,000 speakers; Indo-European; spoken in Israel, Greece, Turkey, and other countries. Lake Miwok [lmw]: Dormant; Utian; formerly spoken in the United States of America. Lakota (Lakhota, Lakȟótiyapi, Teton) [lkt]: Severely endangered, 2,000 speakers; Siouan; spoken in the United States of America and Canada. Lipan (Eastern Apache) [apl]: Awakening; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Louisiana Creole (Kréyol, Kreyol lwizyàn, Kouri-vini) [lou]: Severely endangered, fewer than 7,000 speakers; pidgin or creole; spoken in the United States of America. (Lower) Tanana [taa]: Critically endangered, 25 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Luiseño (Luiseño-Juaneño) [lui]: Awakening; Uto-Aztecan; spoken in the United States of America. Lushootseed (xʷəlšucid, Puget Sound Salish, Skagit-Nisqually) [lut, sno, ska]: Critically endangered, five speakers; Salishan; spoken in the United States of America. Maidu (Northeastern Maidu, Mountain Maidu) [nmu]: Dormant; Maiduan; formerly spoken in the United States of America. Makah (Kwe-Nee-Chee-Aht, Kweedishchaaht, Macaw) [myh]: Awakening; Wakashan; spoken in the United States of America. Maliseet-Passamaquoddy (Passamaquoddy-Maliseet, Malecite-Passamaquoddy, Pennacook) [pqm]: Endangered, approximately 500 speakers; Algic; spoken in the United States of America and Canada. Mandan [mhq]: Dormant; Siouan; formerly spoken in the United States of America. Maricopa (Piipaash, Pee Posh, Cocomaricopa) [mrc]: Severely endangered, approximately 100 speakers; Cochimi-Yuman; spoken in the United States of America. Maritime Sign Language (Nova Scotian Sign Language) [nsr]: Critically endangered, fewer than 100 speakers; sign language; spoken in Canada. Menominee (Menomini) [mez]: Severely endangered, 35 speakers; Algic; spoken in the United States of America. Mescalero-Chiricahua [apm]: Threatened, at least 1,500 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Mi’kmaq (Micmac, Mi’gmaq, Mi’gmaw) [mic]: Threatened, 8,145 speakers; Algic; spoken in the United States of America and Canada. Miami-Illinois (Myaamia, Miami, Illinois) [mia]: Awakening; Algic; spoken in the United States of America. Michif (French Cree, Mitchif, Métif) [crg]: Critically endangered, approximately 200 speakers; mixed language; spoken in the United States of America and Canada. Mikasuki (Hitchiti, Mikasuki Seminole, Miccosukee) [mik]: Threatened, approximately 500 speakers; Muskogean; spoken in the United States of America. Mohawk (Kanien’kehaka) [moh]: Endangered, 3,850 speakers; Iroquoian; spoken in the United States of America and Canada. Mojave (Mohave, Amaquaqua, Jamajab) [mov]: Critically endangered, fewer than 100 speakers; Cochimi-Yuman; spoken in the United States of America. Mono (Monachi, Saidyuka) [mnr]: Critically endangered, fewer than 40 speakers; Uto-Aztecan; spoken in the United States of America. 169

Anna Belew et al.

Munsee (Delaware, Ontario Delaware, Munsee Delaware) [umu]: Critically endangered, few speakers; Algic; spoken in the United States of America and Canada. Muskogee (Creek, Northern Muskogean, Creek-Seminole) [mus]: Severely endangered, 4,200 speakers; Muskogean; spoken in the United States of America. Natchez [ncz]: Awakening; unclassified (isolate?); spoken in the United States of America. Navajo (Diné, Navaho) [nav]: At risk, 120,000 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Nēhiyawēwin (ᓀᐦᐃᔭᐍᐏᐣ, Plains Cree) [no ISO 639-3 code]: Severely endangered, 68 speakers; Algic; spoken in Canada. Nez Perce (Niimi’ipuutímt, Chopunnish, Nimipu) [nez]: Critically endangered, fewer than 50 speakers; Sahaptian; spoken in the United States of America. Nisenan (South Maidu, Neeshenam, Pujuni) [nsz]: Dormant; Maiduan; formerly spoken in the United States of America. Nisga’a (Nishga, Niska’, Nisk’a’) [ncg]: Severely endangered, 857 speakers; Tsimshian; spoken in Canada. Nłeʔkepmxcín (Thompson, Nlha7kçpmx, Thompson River Salish) [thp]: Severely endangered, 127 speakers; Salishan; spoken in Canada. Northern Paiute (Paviotso) [pao]: Critically endangered, approximately 700 speakers; UtoAztecan; spoken in the United States of America. Northern Pomo (Redwood Cañon) [pej]: Dormant; Pomoan; formerly spoken in the United States of America. Northern Straits Salish [str]: Critically endangered, fewer than 20 speakers; Salishan; spoken in the United States of America and Canada. Northern Tutchone [ttm]: Severely endangered, approximately 200 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Nuuchahnulth (Nootka, Nuu-chah-nulth) [nuk]: Critically endangered, 134 speakers; Wakashan; spoken in Canada. Nuxalk (Bella Coola) [blc]: Critically endangered, 17 speakers; Salishan; spoken in Canada. O’odham (Pima-Papago) [ood]: Vulnerable, at least 15,000 speakers; Uto-Aztecan; spoken in the United States of America and Mexico. Oji-Cree (Severn Ojibwe, Severn Ojibwa, Ojicree) [ojs]: Vulnerable, 6,110 speakers; Algic; spoken in Canada. Okanagan-Colville (Nsyilxcən) [oka]: Critically endangered, 350–400 speakers; Salishan; spoken in the United States of America and Canada. Omaha-Ponca (Uman, Mahairi) [oma]: Severely endangered, fewer than 85 speakers; Siouan; spoken in the United States of America. Oneida [one]: Endangered, approximately 212 speakers; Iroquoian; spoken in the United States of America and Canada. Onondaga (Onandaga, Onondoga) [ono]: Critically endangered, approximately 52 speakers; Iroquoian; spoken in the United States of America and Canada. Osage (Wazhazhe) [osa]: Awakening; Siouan; spoken in the United States of America. Ottawa (Odawa, Odaawaa, Eastern Ojibwa) [otw]: Severely endangered, approximately 2,000 speakers; Algic; spoken in the United States of America and Canada. Pacific Yupik (Pacific Gulf Yupik, Alutiiq, Sugpiak) [ems]: Severely endangered, 200 speakers; Eskimo-Aleut; spoken in the United States of America. Panamint (Panamint Shoshone, Tümpisa Shoshone, Timbisha) [par]: Critically endangered, fewer than 20 speakers; Uto-Aztecan; spoken in the United States of America. 170

The world’s endangered languages

Patwin [pwi]: Critically endangered, approximately one speaker; Wintuan; spoken in the United States of America. Pawnee (Grand Pawnee, Republican Pawnee) [paw]: Dormant; Caddoan; formerly spoken in the United States of America. Pennsylvania German (Pennsylvania Deitsh, Pennsylvanish, Pennsylvania Dutch) [pdc]: Vulnerable, 98,700 speakers; Indo-European; spoken in the United States of America and Canada. Picuris (Taos-Picuris) [no ISO 639-3 code]: Threatened, fewer than 225 speakers; KiowaTanoan; spoken in the United States of America. Plains Indian Sign Language (Plains Sign Language) [psd]: Critically endangered; sign language; spoken in the United States of America and Canada. Plains Miwok (Valley Miwok) [pmw]: Dormant; Utian; formerly spoken in the United States of America. Potawatomi (Pottawotomi, Bodéwadmi, Bodewadmi) [pot]: Critically endangered, fewer than ten speakers; Algic; spoken in the United States of America and Canada. Quapaw (Arkansas, Arkans, Alkansea) [qua]: Awakening; Siouan; spoken in the United States of America. Quebec Sign Language (Langue Signe Quebecars, Langue des Signes Québécoise, LSQ) [fcs]: Threatened, 730–1,365 speakers; sign language; spoken in Canada. Quechan (Yuma, Kechan, Quecl) [yum]: Critically endangered, 150–200 speakers; CochimiYuman; spoken in the United States of America. Quileute (Quillayute, Quile-ute) [qui]: Dormant; Chimakuan; formerly spoken in the United States of America. Rio Grande Keresan (Eastern Keresan, Eastern Keres Pueblo) [kee]: Endangered, approximately 6,125 speakers; Keresan; spoken in the United States of America. Sahaptin [waa, tqn, uma, yak]: Critically endangered, approximately 125 speakers; Sahaptian; spoken in the United States of America. San Francisco Bay Costanoan (Čočenyo, Northern Costanoan, Northern Ohlone) [cst]: Awakening; Utian; spoken in the United States of America. Sauk-Fox (Fox, Mesquakie, Meskwakie) [sac]: Severely endangered, approximately 200 speakers; Algic; spoken in the United States of America. Saulteaux (Nakawēmowin, ᓇᐦᑲᐌᒧᐎᓐ, Western Ojibwa) [ojw]: Threatened, fewer than 10,000 speakers; Algic; spoken in Canada. Secwepemctsin (Shuswap, Secwepemc) [shs]: Endangered, 210 speakers; Salishan; spoken in Canada. Seneca (Tsonnontouan, Taroko) [see]: Critically endangered, fewer than 50 speakers; Iroquoian; spoken in the United States of America and Canada. Serrano (Serran, Kitanemuk) [ser]: Awakening; Uto-Aztecan; spoken in the United States of America. Shawnee [sjw]: Severely endangered, fewer than 100 speakers; Algic; spoken in the United States of America. She shashishalhem (Sechelt, sháshishálem, Siciatl) [sec]: Critically endangered, four speakers; Salishan; spoken in Canada. Shoshone (Shoshoni, Shoshoni-Goshiute, Shonshoni) [shh]: Endangered, approximately 1,000 speakers; Uto-Aztecan; spoken in the United States of America. Sierra Miwok [csm, nsq, skd]: Critically endangered, one speaker; Utian; spoken in the United States of America. 171

Anna Belew et al.

Sioux (Lakota) [dak, lkt, sto]: Endangered, 25,000 speakers; Siouan; spoken in the United States of America and Canada. Siuslaw (Umpqua, Ku-itc, Kuitsh) [sis]: Dormant; isolate; formerly spoken in the United States of America. Ski:xs (Sgüüxs, Southern Tsimshian, Old Klemtu) [no ISO 639-3 code]: Dormant; Tsimshian; formerly spoken in the United States of America and Canada. Sḵwx̱wú7mesh sníchim (Squamish, Sqwxwumish, Skwxwu7mesh) [squ]: Critically endangered, seven speakers; Salishan; spoken in Canada. Sm̓algya̱x (Coast Tsimshian) [tsi]: Critically endangered, fewer than 430 speakers; Tsimshian; spoken in the United States of America and Canada. Southeastern Pomo (Elem Pomo, Lower Lake Pomo) [pom]: Awakening; Pomoan; spoken in the United States of America. Southern Pomo (Russian River, Gallinoméro) [peq]: Critically endangered, one speaker; Pomoan; spoken in the United States of America. Southern Tiwa (Isleta-Sandia) [tix]: Endangered, approximately 1,600 speakers; KiowaTanoan; spoken in the United States of America. Southern Tutchone (Tuchone) [tce]: Severely endangered, approximately 200 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Southwestern Ojibwa (Ojibwe, Ojibway) [ciw]: Threatened, 5,000 speakers; Algic; spoken in the United States of America. St̓át̓imcets (Lillooet) [lil]: Critically endangered, 137 speakers; Salishan; spoken in Canada. Stoney (Nakota, Assiniboin) [sto]: Endangered, 1,000–1,500 speakers; Siouan; spoken in Canada. Tagish [tgx]: Critically endangered, approximately one speaker; Athabaskan-Eyak-Tlingit; spoken in Canada. Tāłtān (Tahltan, Nahanni, Tahl-tan) [tht]: Critically endangered, 45 speakers; Athabaskan-EyakTlingit; spoken in Canada. Tanacross (Nee’anděg’, Transitional Tanana) [tcb]: Critically endangered, 50 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Taos (Northern Tiwa) [twf]: Endangered, 800 speakers; Kiowa-Tanoan; spoken in the United States of America. Tewa [tew]: Endangered, 1,500 speakers; Kiowa-Tanoan; spoken in the United States of America. Tipai (Tiipai, Tipay, Mexican Diegueño) [no ISO 639-3 code]: Severely endangered, approximately 100 speakers; Cochimi-Yuman; spoken in the United States of America and Mexico. Tlingit (Łingít, Tlinkit, Thlinget) [tli]: Critically endangered, 200 speakers; Athabaskan-EyakTlingit; spoken in the United States of America and Canada. Tolowa (Smith River, Chetco-Tolowa) [tol]: Awakening; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Tse’khene (Sekani, Tsek’ene, Tsek’ehne) [sek]: Critically endangered, 30 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Tsilhqot’in (Chilcotin, Tinneh, Tsilhqot’in) [clc]: Severely endangered, 864 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Tsuut’ina (Sarcee, Sarsi, Tsutina) [srs]: Critically endangered, 47 speakers; AthabaskanEyak-Tlingit; spoken in Canada. Tubatulabal (Tübatulabal, Kern River, Kern) [tub]: Critically endangered, one speaker; UtoAztecan; spoken in the United States of America. 172

The world’s endangered languages

Tunica (Tonika, Tunixka) [tun]: Awakening; isolate; spoken in the United States of America. Tuscarora (Skarohreh) [tus]: Critically endangered, two to three speakers; Iroquoian; spoken in the United States of America and Canada. Unami (Oklahoma Delaware, Lenni-Lenape, Lenape) [unm]: Dormant; Algic; formerly spoken in the United States of America. Upland Yuman (Havasupai-Walapai-Yavapai, Upper Colorado River Yuman, Northern Pai) [yuf]: Threatened, 1,600–1,650 speakers; Cochimi-Yuman; spoken in the United States of America. Upper Kuskokwim (Kolchan) [kuu]: Critically endangered, 25 speakers; Athabaskan-EyakTlingit; spoken in the United States of America. Upper Tanana (Nee’aanegn, Nabesna) [tau]: Critically endangered, 55 speakers; AthabaskanEyak-Tlingit; spoken in the United States of America and Canada. Ute (Ute-Southern Paiute, Pai Ute, Paviotso) [ute]: Endangered, approximately 3,500 speakers; Uto-Aztecan; spoken in the United States of America. Wampanoag (Wôpanâak, Massachusett-Narragansett) [wam]: Awakening; Algic; spoken in the United States of America. Wappo (Ashochimi, Napa) [wao]: Awakening; Yukian; spoken in the United States of America. Wasco-Wishram (Kiksht, Wasco, Wishram) [wac]: Dormant; Chinookan; formerly spoken in the United States of America. Washo (Washoe, Washoa) [was]: Severely endangered, approximately 20 speakers; isolate; spoken in the United States of America. Western Abenaki (Western Abnaki) [abe]: Critically endangered, fewer than ten speakers; Algic; spoken in the United States of America and Canada. Western Apache (San Carlos-Southern Tonto, San Carlos Apache, White River Apache) [apw]: Threatened, fewer than 14,000 speakers; Athabaskan-Eyak-Tlingit; spoken in the United States of America. Wichita (Witchita) [wic]: Dormant; Caddoan; formerly spoken in the United States of America. Winnebago (Ho-Chunk, Hocak Wazijaci, Hocák) [win]: Endangered, at least 250 speakers; Siouan; spoken in the United States of America. Wintu (Colouse, Wintun, North Wintun) [wnw]: Dormant; Wintuan; formerly spoken in the United States of America. Witsuwit’en (Witsuwit’en-Nedut’en, Witsuwit’en-Babine) [bcr]: Severely endangered, 434 speakers; Athabaskan-Eyak-Tlingit; spoken in Canada. Wiyot (Wishosk, Kowilth, Wiyoshk) [wiy]: Awakening; Algic; spoken in the United States of America. Xaad Kil (Haida, K’iis Xaat’aay) [hdn, hax]: Critically endangered, 50–100 speakers; isolate; spoken in the United States of America and Canada. X̄enaksialak̓ala (Haisla, X̄a’’islak̓ala, Xiʔslak’ala) [has]: Critically endangered, 80 speakers; Wakashan; spoken in Canada. Yokuts [yok]: Critically endangered, fewer than ten speakers; Yokutsan; spoken in the United States of America. Yuchi (Euchee) [yuc]: Critically endangered, five to seven speakers; isolate; spoken in the United States of America. Yurok (Chillula, Mita, Pekwan) [yur]: Awakening; Algic; spoken in the United States of America. Zuni (Zuñi, Shiwi’ma) [zun]: Threatened, approximately 9,000 speakers; isolate; spoken in the United States of America. 173

Anna Belew et al.

The Pacific Russell Barlow Linguistic diversity in the Pacific The Pacific is one of the regions of the globe with the greatest linguistic diversity. Ethnologue includes 1,108 languages of the Pacific (1,059 in Melanesia, 27 in Micronesia, and 22 in Polynesia), representing over 15% of its total figure of 7,099 “living languages”23 (Simons & Fennig 2017). While the number of languages Ethnologue lists as spoken in Africa is 2,144 and in Asia 2,294, these regions are also home to much larger populations. In contrast, the languages of the Pacific are spoken by less than 0.1% of the world’s population. The average language of the Pacific has fewer than 10,000 speakers.24 Of perhaps even greater interest than the total number of languages in the Pacific is the large number of language families represented. The indigenous (pre-colonial, non-creole) languages of the region have traditionally been divided into two categories: Austronesian and Papuan. This is a false dichotomy, however, which obscures the facts of genetic diversity, since a number of different language families (and isolates) are grouped under the “Papuan” label: over 30 (Ross 2005:30) even by those most inclined to tolerate proposals of distant relatedness among languages – over 100 families and isolates by scholars less inclined to accept proposals of remote kinship without compelling evidence (Hammarström 2012:469). While there are more languages in the Pacific, South America has a larger number of distinct language families (including language isolates). In ELCat these Pacific languages are represented by endangered languages belonging to 28 families and 18 isolates, in addition to five unclassified languages. However, it should be noted that some of these “Papuan” languages are spoken in eastern Indonesia and East Timor, which are both considered (politically) to be part of island southeast Asia. The largest of these Papuan families – both in number of speakers and in number of languages – is the Trans-New Guinea (TNG) family. In earlier scholarship (e.g., Wurm et al. 1975) this term has been applied to a postulated massive “phylum” now considered to comprise several families that are not demonstrably related (Pawley 2005). There is, nevertheless, a “core” of TNG languages that are certainly related, mostly spoken throughout the highlands of Papua New Guinea. The Austronesian family, on the other hand, is a clearly established single family of languages that have all descended from a common ancestral language. Nevertheless, there is much linguistic diversity even within this single family, which (with over 1,200 living members) is the world’s second-largest language family (behind Niger-Congo). Since the proto-language is estimated to have last been spoken around 5,500–6,000 years ago, there has been ample time for the subgroups to diverge markedly from one another (Blust 2013). While some of these languages are spoken in continental southeast Asia (and even as far west as Madagascar), the vast majority are spoken in the Pacific, either in island southeast Asia or Oceania. A staggering number of languages in the Pacific, both Austronesian and Papuan, are endangered, many critically so. Following is an overview of language endangerment in the Pacific region.

174

The world’s endangered languages

Causes of language endangerment in the Pacific Historically, the single greatest contributor to language loss in the Pacific has been colonization. As Europeans spread throughout the Pacific in the eighteenth and nineteenth centuries, in search of trade, plantation agriculture, whaling, and other economic gains (and while proselytizing), they brought their languages with them. While Dutch, English, French, German, Portuguese, and Spanish were all spoken at various times in various colonies in the Pacific, English has certainly had the greatest impact. This is especially the case in Micronesia and Polynesia: where there is language shift in these regions, the target language is almost always English. The colonial histories of Polynesian nations such as New Zealand and Tuvalu reflect the shifts occurring to English (from Māori and Tuvaluan, respectively). Micronesian languages such as Nauruan and Mokilese are suffering similar fates. The picture in Melanesia, however, is different. European colonization never penetrated Melanesia as deeply as in Micronesia or Polynesia. Although every Melanesian nation and territory has at least one European language as an official language (English or French, and both in the case of Vanuatu), these are not commonly spoken by the people at large, nor is there much language shift towards English or French. Rather, it is another legacy of European colonialism towards which many Melanesian languages are shifting: English-based creoles. Unlike in Micronesia and Polynesia, where economic pressures have been the primary factor in language shift, the linguistic landscape in Melanesia is changing rapidly, in part as a result of its immense linguistic diversity. The adoption of an English-based creole (e.g., Tok Pisin in Papua New Guinea, Bislama in Vanuatu, or Pijin in the Solomon Islands) has enabled speakers of hundreds of languages to communicate across long-standing language barriers. As more and more children acquire these creoles as their first language, indigenous languages face increasingly dire threats.

The current status of endangered languages in the Pacific ELCat identifies 537 endangered languages (which includes seven dormant and one awakening languages, 529 still spoken) in the 20 countries and territories of the Pacific. Of this total of 537 languages, 220 are Austronesian languages, 272 belong to various “Papuan” groups, and three are pidgin/creole languages. Not included in this number, though, are the Austronesian and Papuan languages that are located in what is politically island southeast Asia. Of the 537 languages of the Pacific listed in ELCat, 247 are “Endangered,” 71 are “Severely Endangered,” and 35 are “Critically Endangered.” There are 101 languages that have at most 100 speakers. When attempting to obtain reliable information on the endangerment status of languages in the Pacific, one is immediately struck by the dearth of information available for many languages, especially of Melanesia. There are undoubtedly more endangered languages of the Pacific than those included in ELCat, but about which there is essentially no information, either by which to gauge their vitality or even to classify them. For example, the speaker numbers for 14 of the languages listed in ELCat are just unknown. Due in part to the lack of information about linguistic vitality in some parts of the Pacific, lesser-known languages are often reported to be much more vital than they really are. Especially in Melanesia, websites such as Ethnologue often rely on linguistic surveys conducted decades ago; sadly, however, the present sociolinguistic context for the languages of the Pacific is drastically different from that of the 1970s and 1980s. One asset of ELCat, in contrast, is the Catalogue’s access to up-to-date personal communications from specialists working in the field for many of the languages. For example, Aisi (also known as Musak [mmq]), a Madang language 175

Anna Belew et al.

in Papua New Guinea, is described in Ethnologue as being “Vigorous” (Simons & Fennig 2017), and cited for this assessment are population figures from 1981 (themselves actually from even earlier survey work). The Catalogue of Endangered Languages, however, classifies Aisi as “Endangered” based on more recent field reports (Don Daniels, personal communication, 2014) – more and more children are speaking Tok Pisin instead of their heritage language there. Another Madang language, Manat (also known as Paynamar [pmr]), is in even more dire condition, as there are only about 50 speakers, none of whom are children (Don Daniels, personal communication, 2014). Accordingly, ELCat classifies Manat as “Severely Endangered,” a much more urgent degree of endangerment than Ethnologue’s assessment of “Threatened,” which is based in part on population figures published in 1975. Similarly, Papapana [ppn], an Austronesian language spoken in Bougainville, is considered “Severely Endangered” in ELCat, based on careful sociolinguistic research conducted by Ellen Smith (personal communication 2014); Ethnologue, however, treats Papapana as merely “Threatened.” By using information from linguists currently working in the field, ELCat is able to offer a more accurate picture of just how endangered many Pacific languages are. Indeed, the greatest challenge to assessing the endangerment of the languages of the Pacific is the absence of reliable current information. As more linguists conduct research in the region, it is hoped that the picture of linguistic endangerment in the Pacific will become clearer. Below is an alphabetical listing of all languages of the Pacific currently included in the Catalogue of Endangered Languages.

Endangered languages of the Pacific Abaga [abg]: Critically endangered, five speakers; Trans-New Guinea; spoken in Papua New Guinea. Abau (Green River, Djarok) [aau]: Threatened, at least 7,000 speakers; Sepik; spoken in Papua New Guinea. Abom [aob]: Critically endangered, 15 speakers; Trans-New Guinea; spoken in Papua New Guinea. Abu (Adjora, Adjoria, Azao) [ado]: Threatened, 2,347 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Abu’ (Abu’ Arapesh, Ua) [aah]: Threatened, 2,560 speakers; Torricelli; spoken in Papua New Guinea. Adzera (Azera, Acira, Atsera) [adz]: Vulnerable, 28,900 speakers; Austronesian; spoken in Papua New Guinea. Agi (Metru) [aif]: Endangered, 670 speakers; Torricelli; spoken in Papua New Guinea. Aiku (Yangum Mon, Malek, Menandon) [ymo]: Endangered, 819 speakers; Torricelli; spoken in Papua New Guinea. Aimele (Eibela, Kware) [ail]: Endangered, 140 speakers; Trans-New Guinea; spoken in Papua New Guinea. Ainbai [aic]: Endangered, 110 speakers; Border; spoken in Papua New Guinea. Aiome (Ayom) [aki]: Endangered, 620 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Aisi (Musak, Green Miao, Blue Miao) [mmq]: Endangered, approximately 250 speakers; TransNew Guinea; spoken in Papua New Guinea. Äiwoo (Reef, Ayiwo, Naaude) [nfl]: Vulnerable, 7,926 speakers; Austronesian; spoken in Solomon Islands. Ak [akq]: Severely endangered, fewer than 80 speakers; Sepik; spoken in Papua New Guinea. 176

The world’s endangered languages

Akoye (Mahigi, Akoinkake, Lohiki) [miw]: Endangered, 800 speakers; Trans-New Guinea; spoken in Papua New Guinea. Alamblak (Arambak) [amp]: Threatened, 1,200 speakers; Sepik; spoken in Papua New Guinea. Alekano (Gahuku, Gafuku, Gahuku-Gama) [gah]: Vulnerable, 25,000 speakers; Trans-New Guinea; spoken in Papua New Guinea. Alfendio (Arafundi, Nanubae, Aunda) [afk, afp]: Endangered, 633 speakers; Arafundi; spoken in Papua New Guinea. Alu (Dia, Galu, Metru) [dia]: Threatened, 1,880 speakers; Torricelli; spoken in Papua New Guinea. Ama (Sawiyanu, Waniabu, Neinbonei) [amm]: Vulnerable, 450 speakers; Left May; spoken in Papua New Guinea. Amal (Amel, Alai) [aad]: Endangered, 388 speakers; Sepik; spoken in Papua New Guinea. Amanab [amn]: Threatened, 3,500 speakers; border; spoken in Papua New Guinea and Indonesia. Amara (Longa, Bibling) [aie]: Endangered, fewer than 231 speakers; Austronesian; spoken in Papua New Guinea. Ambakich (Aion, Porapora) [aew]: Endangered, 770 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Ambulas (Abelam, Abulas, Ambelas) [abt, wos]: Vulnerable, 33,000 speakers; Sepik; spoken in Papua New Guinea. Amto (Ki, Siwai, Siawi) [amt]: Endangered, 230 speakers; Amto-Musan; spoken in Papua New Guinea. Anam (Pondoma) [pda]: Threatened, 1,070 speakers; Trans-New Guinea; spoken in Papua New Guinea. Andarum [aod]: Endangered, 830 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Ande (Morouas) [no ISO 639-3 code]: Endangered, 500 speakers; Austronesian; spoken in Vanuatu. Anejom̃ (Aneityum) [no ISO 639-3 code]: Endangered, 900 speakers; Austronesian; spoken in Vanuatu. Anem (Anêm) [anz]: Endangered, 843 speakers; isolate; spoken in Papua New Guinea. Angor (Anggor, Senagi, Watapor) [agg]: Threatened, 1,250 speakers; Senagi; spoken in Papua New Guinea. Angoram (Olem, Pondo) [aog]: Threatened, 6,514 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Anor [anj]: Endangered, 450 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Anuki (Gabobora) [aui]: Threatened, 574 speakers; Austronesian; spoken in Papua New Guinea. Anuta [aud]: Endangered, fewer than 340 speakers; Austronesian; spoken in Solomon Islands. Aore [aor]: Dormant; Austronesian; formerly spoken in Vanuatu. Ap Ma (Ap Ma Botin, Botin, Kambot) [kbx]: Threatened, 6,500 speakers; isolate; spoken in Papua New Guinea. Araki (Banks-Inseln) [akr]: Critically endangered, 15 speakers; Austronesian; spoken in Vanuatu. Arawum [awm]: Severely endangered, 60 speakers; Trans-New Guinea; spoken in Papua New Guinea. Arhâ (Ara, Neukaledonien) [aqr]: Severely endangered, 35 speakers; Austronesian; spoken in New Caledonia. 177

Anna Belew et al.

Arhö (Aro, Neukaledonien) [aok]: Severely endangered, 62 speakers; Austronesian; spoken in New Caledonia. Ari [aac]: Severely endangered, 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Arifama-Miniafia (Miniafia-Arifama) [aai]: Threatened, 3,470 speakers; Austronesian; spoken in Papua New Guinea. Aru (Alang, Alatil, Amol) [alx]: Endangered, 270 speakers; Torricelli; spoken in Papua New Guinea. Aruamu (Ariawia, Ariawiai, Gumasi) [msy]: Threatened, 5,350 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Aruek (Djang, Jang) [aur]: Endangered, 614 speakers; Torricelli; spoken in Papua New Guinea. Aruop (Alatil, Lau’u, Lauisaranga) [lsr]: Endangered, 700 speakers; Torricelli; spoken in Papua New Guinea. Asaba (Duranmin, Akiapmin, Suarmin) [seo]: Threatened, 145 speakers; isolate; spoken in Papua New Guinea. Asumboa (Asumbua, Asumuo) [aua]: Severely endangered, ten speakers; Austronesian; spoken in Solomon Islands. Ati (Polonombauk, Meris) [no ISO 639-3 code]: Severely endangered, 85 speakers; Austronesian, Malayo-Polynesian, Oceanic, Northern Vanuatu; spoken in Vanuatu. Atin [no ISO 639-3 code]: Endangered, 120 speakers; Austronesian; spoken in Vanuatu. Au [avt]: Threatened, 4,200 speakers; Torricelli; spoken in Papua New Guinea. Aulua [aul]: Endangered, 750 speakers; Austronesian; spoken in Vanuatu. Austral (Tubuai-Rurutu) [aut]: Threatened, 8,000 speakers; Austronesian; spoken in French Polynesia. Auwe (Simog) [smf]: Endangered, 270 speakers; Border; spoken in Papua New Guinea. Avava (Katbol, Navava, Bangsa’) [tmb]: Vulnerable, 700 speakers; Austronesian; spoken in Vanuatu. Aveteian (Dixon Reef) [dix]: Severely endangered, 50 speakers; Austronesian; spoken in Vanuatu. Avok [no ISO 639-3 code]: Endangered, 500 speakers; Austronesian; spoken in Vanuatu. Awar [aya]: Severely endangered, fewer than 1,153 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Awiakay (Arafundi) [no ISO 639-3 code]: Threatened, 300 speakers; Arafundi; spoken in Papua New Guinea. Awtuw (Autu, Kamnum) [kmn]: Endangered, 400 speakers; Sepik; spoken in Papua New Guinea. Awun (Awon, Auwan) [aww]: Endangered, 384 speakers; Sepik; spoken in Papua New Guinea. Axamb (Ahamb) [ahb]: Endangered, 750 speakers; Austronesian; spoken in Vanuatu. Ayi [ayq]: Endangered, 430 speakers; Sepik; spoken in Papua New Guinea. Bagupi [bpi]: Endangered, 58 speakers; Trans-New Guinea; spoken in Papua New Guinea. Bahinemo (Bahenemo, Gahom, Inaru) [bjh]: Vulnerable, 700 speakers; Sepik; spoken in Papua New Guinea. Baibai [bbf]: Endangered, 271 speakers; Baibai-Fas; spoken in Papua New Guinea. Baikeno (Baikenu, Vaikenu, Vaikino) [bkx]: Vulnerable, 20,000 speakers; Austronesian; spoken in East Timor. Baki (Burumba, Paki) [bki]: Endangered, 150 speakers; Austronesian; spoken in Vanuatu. 178

The world’s endangered languages

Banaro (Banar, Banara, Waran) [byz]: Threatened, 2,569 speakers; isolate; spoken in Papua New Guinea. Bebeli [bek]: Severely endangered, 780 speakers; Austronesian; spoken in Papua New Guinea. Beli (Akuwagel, Mukili, Makarim) [bey]: Vulnerable, fewer than 1,453 speakers; Torricelli; spoken in Papua New Guinea. Bepour [bie]: Severely endangered, 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Berinomo (Bitara, Apowasi, Perbi) [bit]: Endangered, 178 speakers; Sepik; spoken in Papua New Guinea. Bierebo (Bonkovia-Yevali) [bnk]: Threatened, 900–1,000 speakers; Austronesian; spoken in Vanuatu. Bieria (Bieri, Vovo, Wowo) [brj]: Severely endangered, 25 speakers; Austronesian; spoken in Vanuatu. Bilakura (Banks-Inseln) [bql]: Severely endangered, 30 speakers; Trans-New Guinea; spoken in Papua New Guinea. Biliru (Tambotalo) [tls]: Severely endangered, fewer than 50 speakers; Austronesian; spoken in Vanuatu. Bilua [blb]: Threatened, 8,740 speakers; isolate; spoken in Papua New Guinea. Bimin [bhl]: Vulnerable, 2,000 speakers; Trans-New Guinea; spoken in Papua New Guinea. Bisis (Yambiyambi) [bnw]: Endangered, 395 speakers; Sepik; spoken in Papua New Guinea. Bisorio (Bi, Gne, Sidi) [bir, bic]: Endangered, approximately 200 speakers; Trans-New Guinea; spoken in Papua New Guinea. Biwat (Munduguma, Mundugumor, Mundokuma) [bwm]: Threatened, 1,642 speakers; Yuat; spoken in Papua New Guinea. Blablanga (Gema, Goi) [blp]: Threatened, approximately 2,000 speakers; Austronesian; spoken in Solomon Islands. Bo (Po, Sorimi) [bpw]: Severely endangered, 76 speakers; Left May; spoken in Papua New Guinea. Bogaya (Bogaia, Pogaya) [boq]: Endangered, 300 speakers; isolate; spoken in Papua New Guinea. Boiken (Boikin, Nucum, Yangoru) [bzf]: Vulnerable, 35,204 speakers; Sepik; spoken in Papua New Guinea. Bola (Bakovi, Bola-Bakovi) [bnp]: Vulnerable, 13,700 speakers; Austronesian; spoken in Papua New Guinea. Bonkiman [bop]: Endangered, 180 speakers; Trans-New Guinea; spoken in Papua New Guinea. Borei (Borei, Gamai, Gamay) [gai]: Threatened, 1,202 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Boselewa (Bosilewa, Bosalewa, Mwani’u) [bwf]: Endangered, 250 speakers; Austronesian; spoken in Papua New Guinea. Bosmun (Bosman, Bosngun) [bqs]: Endangered, fewer than 1,768 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Botovro (Mpotovoro, Mallicolo) [mvt]: Endangered, 430 speakers; Austronesian; spoken in Vanuatu. Bragat (Alang, Alauagat, Braget) [aof]: Endangered, 355 speakers; Torricelli; spoken in Papua New Guinea. Breri (Kuanga) [brq]: Endangered, 720 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. 179

Anna Belew et al.

Budibud (Nada) [btp]: Endangered, 120 speakers; Austronesian; spoken in Papua New Guinea. Bugawac (Bukawa, Bukaua, Bukawac) [buk]: Threatened, 9,690 speakers; Austronesian; spoken in Papua New Guinea. Bukiyip (Mountain Arapesh, Kavu, Kawu) [ape]: Endangered, 12,000 speakers; Torricelli; spoken in Papua New Guinea. Bulgebi [bmp]: Severely endangered, 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Bulu [bjl]: Endangered, 910 speakers; Austronesian; spoken in Papua New Guinea. Bumbita (Bumbita Arapesh, Bambita Arapesh, But Arapesh) [aon]: Endangered, 2,353 speakers; Torricelli; spoken in Papua New Guinea. Bun [buv]: Endangered, 194 speakers; Yuat; spoken in Papua New Guinea. Buna [bvn]: Threatened, 1,259 speakers; Torricelli; spoken in Papua New Guinea. Bunak (Buna, Buna’, Bunake) [bfn]: Vulnerable, 100,000 speakers; Timor-Alor-Pantar; spoken in Indonesia and East Timor. Bungain [but]: Threatened, 2,451 speakers; Torricelli; spoken in Papua New Guinea. Bura [no ISO 639-3 code]: Endangered, 300 speakers; Austronesian; spoken in Vanuatu. Burmbar (Banam Bay, Vartavo) [vrt]: Endangered, 900 speakers; Austronesian; spoken in Vanuatu. Burui (Sawos) [bry]: Endangered, approximately 150 speakers; Sepik; spoken in Papua New Guinea. Busa (Odiai, Busan, Uriai) [bhf]: Endangered, 240 speakers; isolate; spoken in Papua New Guinea. Butmas-Tur (Ati, Farafi) [bnr]: Endangered, 520 speakers; Austronesian, Malayo-Polynesian, Oceanic, Northern Vanuatu; spoken in Vanuatu. Bwatoo (Neukaledonien) [bwa]: Endangered, 300 speakers; Austronesian; spoken in New Caledonia. Caac (Moenebeng, Neukaledonien, Cawac) [msq]: Endangered, 890 speakers; Austronesian; spoken in New Caledonia. Carolinian (Saipan Carolinian, Southern Carolinian) [cal]: Threatened, 3,100 speakers; Austronesian; spoken in Northern Mariana Islands. Chambri (Tchambuli, Tshamberi) [can]: Threatened, 1,050 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Chamorro (Tjamoro) [cha]: Vulnerable, 92,700 speakers; Austronesian; spoken in Northern Mariana Islands and Guam. Changriwa [cga]: Endangered, 498 speakers; Yuat; spoken in Papua New Guinea. Chenapian (Tsenap, Zenap, Chenap) [cjn]: Endangered, 180 speakers; Sepik; spoken in Papua New Guinea. Chini (Akrukay, Akruray) [afi]: Severely endangered, 75 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Daonda [dnd]: Endangered, 135 speakers; Border; spoken in Papua New Guinea. Daruru [no ISO 639-3 code]: Severely endangered, 100 speakers; Austronesian; spoken in Vanuatu. Dengalu [no ISO 639-3 code]: Endangered, 140 speakers; Austronesian; spoken in Papua New Guinea. Dibiyaso (Bainapi, Dibiasu, Pikiwa) [dby]: Threatened, 1,950 speakers; isolate; spoken in Papua New Guinea. Doga (Magabara) [dgg]: Endangered, 200 speakers; Austronesian; spoken in Papua New Guinea. 180

The world’s endangered languages

Dorig (Wetamut, Banks-Inseln) [wwo]: Endangered, 300 speakers; Austronesian; spoken in Vanuatu. Dumun (Bai) [dui]: Severely endangered, 40 speakers; Trans-New Guinea; spoken in Papua New Guinea. Duwet (Guwet, Guwot, Waing) [gve]: Endangered, 300 speakers; Austronesian; spoken in Papua New Guinea. Eitiep (Endangen, Indiang) [eit]: Endangered, 394 speakers; Torricelli; spoken in Papua New Guinea. Elepi (Elapi, Samap, Sumup) [ele]: Endangered, 149 speakers; Torricelli; spoken in Papua New Guinea. Elkei (Olkoi) [elk]: Threatened, 1,640 speakers; Torricelli; spoken in Papua New Guinea. Elu (Admiralitäts-inseln) [elu]: Endangered, 220 speakers; Austronesian; spoken in Papua New Guinea. Emae (Makatea, Emwae, Mae) [mmw]: Endangered, 400 speakers; Austronesian; spoken in Vanuatu. Eton [etn]: Endangered, 500 speakers; Austronesian; spoken in Vanuatu. Fagani (Faghani, Südliche Salomons-Inseln,) [faf]: Endangered, 900 speakers; Austronesian; spoken in Solomon Islands. Faiwol (Angkiyakmin, Faiwolmin, Fegolmin) [fai]: Threatened, 4,500 speakers; Trans-New Guinea; spoken in Papua New Guinea. Farnanto [no ISO 639-3 code]: Severely endangered, 100 speakers; Austronesian, MalayoPolynesian, Oceanic, Northern Vanuatu; spoken in Vanuatu. Farsaf (Narango, Nambel) [nrg]: Endangered, 160 speakers; Austronesian; spoken in Vanuatu. Fas (Bembi, Momu) [fqs]: Threatened, 5,000 speakers; Baibai-Fas; spoken in Papua New Guinea. Fasu [faa]: Threatened, 1,200 speakers; isolate; spoken in Papua New Guinea. Fembe (Sinale, Agala) [agl]: Endangered, 350 speakers; Trans-New Guinea; spoken in Papua New Guinea. Gaikundi (Gaikunti, Gaikwundi, Sawos) [gbf]: Endangered, approximately 650 speakers; Sepik; spoken in Papua New Guinea. Galoli (Galole, Galolen) [gal]: Vulnerable, 13,000 speakers; Austronesian; spoken in East Timor. Galu (Metru, Sinagen) [siu]: Endangered, 208 speakers; Torricelli; spoken in Papua New Guinea. Gao (Nggao) [gga]: Threatened, approximately 900 speakers; Austronesian; spoken in Solomon Islands. Gizrra (Gizra, Toga) [tof]: Threatened, 1,050 speakers; Eastern Trans-Fly; spoken in Papua New Guinea. Gnau [gnu]: Endangered, 980 speakers; Torricelli; spoken in Papua New Guinea. Gorovu (Yerani, Gorova) [grq]: Critically endangered, fewer than 20 speakers; Lower SepikRamu; spoken in Papua New Guinea. Grass Koiari [kbk]: Threatened, 1,700 speakers; Trans-New Guinea; spoken in Papua New Guinea. Guramalum [grz]: Dormant; Austronesian; formerly spoken in Papua New Guinea. Guriaso (Menóu) [grx]: Endangered, fewer than 421 speakers; isolate; spoken in Papua New Guinea. Guya (Guiarak, Guyarak) [gka]: Endangered, 130 speakers; Trans-New Guinea; spoken in Papua New Guinea. Gweda (Garuwahi) [grw]: Severely endangered, 26 speakers; Austronesian; spoken in Papua New Guinea. 181

Anna Belew et al.

Habu (Kairui) [hbu]: Threatened, 2,700 speakers; Austronesian; spoken in East Timor. Haeke (Aeke, ‘Aeke, Haeake) [aek]: Severely endangered, 100 speakers; Austronesian; spoken in New Caledonia. Hahon (Hanon) [hah]: Vulnerable, approximately 3,000 speakers; Austronesian; spoken in Papua New Guinea. Haruai (Harway, Taman, Waibuk) [tmd]: Endangered, 100 speakers; Piawi; spoken in Papua New Guinea. Haveke (Aveke, ‘Aveke, Neukaledonien) [hvk]: Endangered, 300 speakers; Austronesian; spoken in New Caledonia. Hewa (Sisimin, Umairof, Yoliapi) [ham]: Threatened, 2,000 speakers; Sepik; spoken in Papua New Guinea. Heyo (Arima, Arinua, Arinwa) [auk]: Threatened, fewer than 1,824 speakers; Torricelli; spoken in Papua New Guinea. Hiw (Hiu, Torres, Torres Island) [hiw]: Endangered, 120 speakers; Austronesian; spoken in Vanuatu. Hmwaveke (Moaveke, Ceta, Faa Ceta) [mrk]: Endangered, 300 speakers; Austronesian; spoken in New Caledonia. Hoava (Hoava-Kusaghe) [hoa]: Threatened, approximately 2,360 speakers; Austronesian; spoken in Solomon Islands. I’saka (Isaka, Krisa) [ksi]: Severely endangered, 347 speakers; Sko; spoken in Papua New Guinea. Iatmul (Big Sepik, Middle Sepik) [ian]: Threatened, 46,000 speakers; Sepik; spoken in Papua New Guinea. Igana [igg]: Endangered, 114 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Imonda [imn]: Endangered, 274 speakers; Border; spoken in Papua New Guinea. Inapang (Itutang, Midsivindi) [mzu]: Threatened, 1,290 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Ipiko (Ipikoi, Higa, Epai) [ipo]: Endangered, 200 speakers; Trans-New Guinea; spoken in Papua New Guinea. Iteri (Alowiemino, Iyo, Laro) [itr]: Endangered, 447 speakers; Left May; spoken in Papua New Guinea. Kaian (Kayan) [kct]: Threatened, fewer than 792 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Kaiep (Samap, Sumup, Terebu) [kbw, trb]: Endangered, 300 speakers; Austronesian; spoken in Papua New Guinea. Kairiru [kxa]: Threatened, 3,507 speakers; Austronesian; spoken in Papua New Guinea. Kairui-Midiki (Cairui, Midiki) [krd]: Threatened, 15,000 speakers; Austronesian; spoken in East Timor. Kaki Ae (Tate, Raepa Tati, Tati) [tbd]: Endangered, 510 speakers; isolate; spoken in Papua New Guinea. Kalou (Yawa) [ywa]: Endangered, 820 speakers; Sepik; spoken in Papua New Guinea. Kamasa [klp]: Critically endangered, seven speakers; Trans-New Guinea; spoken in Papua New Guinea. Kamasau (Komasau, Komassau, Sagi) [kms]: Endangered, 700 speakers; Torricelli; spoken in Papua New Guinea. Kambaira [kyy]: Endangered, 140 speakers; Trans-New Guinea; spoken in Papua New Guinea. Kamula (Wawoi) [xla]: Endangered, 800 speakers; isolate; spoken in Papua New Guinea. 182

The world’s endangered languages

Kandas (King) [kqw]: Endangered, approximately 300 speakers; Austronesian; spoken in Papua New Guinea. Kanggape (Igom) [igm]: Endangered, 930 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Kaningra (Kaningara) [knr]: Threatened, 327 speakers; Sepik; spoken in Papua New Guinea. Kaninuwa (Kaokao, Wataluma) [wat]: Endangered, 360 speakers; Austronesian; spoken in Papua New Guinea. Kap (Ali, Yakamul) [ykm]: Threatened, 2,079 speakers; Austronesian; spoken in Papua New Guinea. Kapingamarangi (Kirinit) [kpg]: Threatened, 3,000 speakers; Austronesian; spoken in Micronesia. Kapriman (Sare, Mugumute, Wasare) [dju]: Vulnerable, 1,339 speakers; Sepik; spoken in Papua New Guinea. Karawa (Bulawa) [xrw]: Severely endangered, 44 speakers; Sepik; spoken in Papua New Guinea. Karawari (Tabriak) [tzx]: Threatened, 1,300 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Karkar-Yuri (Karkar, Yuri) [yuj]: Threatened, at least 1,000 speakers; Pauwasi; spoken in Papua New Guinea. Karore [xkx]: Endangered, 550 speakers; Austronesian; spoken in Papua New Guinea. Kawacha (Kawatsa) [kcb]: Severely endangered, 12 speakers; Trans-New Guinea; spoken in Papua New Guinea. Kayik (Wanap, Menandon, Minendon) [wnp]: Endangered, 769 speakers; Torricelli; spoken in Papua New Guinea. Kemak (Ema) [kem]: Vulnerable, 72,000 speakers; Austronesian; spoken in East Timor and Indonesia. Kene [no ISO 639-3 code]: Endangered, 300 speakers; Austronesian; spoken in Vanuatu. Kiai (Fortsenal) [frt]: Endangered, 450 speakers; Austronesian; spoken in Vanuatu. Kibiri-Porome (Gibiri, Polome) [prm]: Threatened, 1,180 speakers; isolate; spoken in Papua New Guinea. Kilmeri (Kilmera) [kih]: Threatened, 2,200 speakers; Border; spoken in Papua New Guinea. Kinalakna [kco]: Endangered, 220 speakers; Trans-New Guinea; spoken in Papua New Guinea. Kire (Gire, Giri, Kire-Puire) [geb]: Threatened, 1,900 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Kis [kis]: Endangered, 216 speakers; Austronesian; spoken in Papua New Guinea. Koitabu (Koita) [kqi]: Threatened, 2,700 speakers; Trans-New Guinea; spoken in Papua New Guinea. Koiwat (Sawos) [kxt]: Endangered, approximately 450 speakers; Sepik; spoken in Papua New Guinea. Kokota [kkk]: Threatened, 1,200 speakers; Austronesian; spoken in Solomon Islands. Kol [kol]: Threatened, approximately 4,000 speakers; isolate; spoken in Vanuatu. Kombio (Anamagi, Endangen) [xbi]: Endangered, 2,970 speakers; Torricelli; spoken in Papua New Guinea. Kominimung [xoi]: Endangered, 328 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Komnzo [no ISO 639-3 code]: Endangered, approximately 200 speakers; Morehead-Upper Maro; spoken in Papua New Guinea. 183

Anna Belew et al.

Konomala [koa]: Endangered, 800 speakers; Austronesian; spoken in Papua New Guinea. Kopar [xop]: Endangered, 229 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Koro [krf]: Endangered, 250 speakers; Austronesian; spoken in Vanuatu. Kosraean (Kusaie, Kosrae) [kos]: Threatened, 8,570 speakers; Austronesian; spoken in Nauru and Micronesia. Kove (Kaliai-Kove) [kvc]: Threatened, 6,750 speakers; Austronesian; spoken in Papua New Guinea. Kowaki [xow]: Severely endangered, approximately 30 speakers; Trans-New Guinea; spoken in Papua New Guinea. Kuot (Panaras, Kuat, Neu-Pommern) [kto]: Threatened, 2,400 speakers; isolate; spoken in Papua New Guinea. Kursav (Faita) [faj]: Critically endangered, ten speakers; Trans-New Guinea; spoken in Papua New Guinea. Kwanga (Gawanga, Kawanga, Mende) [kwj]: Threatened, 7,520 speakers; Sepik; spoken in Papua New Guinea. Kwaruwi Kwundi (Sos Kundi, Agbane, Agbanekundi) [sdk, keh]: Threatened, approximately 3,300 speakers; Sepik; spoken in Papua New Guinea. Kwoma (Kuome, Washkuk) [kmo]: Threatened, 2,925 speakers; Sepik; spoken in Papua New Guinea. Kwomtari [kwo]: Threatened, fewer than 800 speakers; Kwomtari-Nai; spoken in Papua New Guinea. Label [lbb]: Endangered, 140 speakers; Austronesian; spoken in Papua New Guinea. Labu (Labu’, Labo, Hapa) [lbu]: Threatened, 1,600 speakers; Austronesian; spoken in Papua New Guinea. Laeko-Libuat (Laeko, Laeko-Limbuat, Watalu) [lkl]: Threatened, fewer than 538 speakers; Torricelli; spoken in Papua New Guinea. Laghu (Lagu, Katova) [lgb]: Dormant; Austronesian; formerly spoken in Solomon Islands. Lakon (Lakona, Gaua, Gog) [lkn]: Endangered, 800 speakers; Austronesian; spoken in Vanuatu. Lamen (Lamenu, Varmali) [lmu]: Endangered, 850 speakers; Austronesian; spoken in Vanuatu. Langam [lnm]: Endangered, 254 speakers; Mongol-Langam-Ulwa; spoken in Papua New Guinea. Larëvat (Larevat, Laravat, Mallicolo) [lrv]: Endangered, 750 speakers; Austronesian; spoken in Vanuatu. Laua (Labu, Lauwa) [luf]: Critically endangered, one speaker; Trans-New Guinea; spoken in Papua New Guinea. Lavukaleve (Laumbe, Laube, Russell Island) [lvk]: Threatened, approximately 1,700 speakers; isolate; spoken in Solomon Islands. Lehali (Teqel, Tekel) [tql]: Endangered, 250 speakers; Austronesian; spoken in Vanuatu. Lelepa (Havannah Harbour) [lpa]: Endangered, 400 speakers; Austronesian; spoken in Vanuatu. Lemerig (Sasar, Banks-Inseln) [lrz]: Critically endangered, two speakers; Austronesian; spoken in Vanuatu. Lendamboi (Small Nambas, Letemboi) [nms]: Endangered, 800 speakers; Austronesian; spoken in Vanuatu. Lenkau [ler]: Endangered, 250 speakers; Austronesian; spoken in Papua New Guinea. Likum [lib]: Severely endangered, 80 speakers; Austronesian; spoken in Papua New Guinea. Lilau (Ngaimbom) [lll]: Endangered, 410 speakers; Torricelli; spoken in Papua New Guinea. 184

The world’s endangered languages

Litzlitz (Naman, Litzlitz-Visele, Mallicolo) [lzl]: Critically endangered, 15–20 speakers; Austronesian; spoken in Vanuatu. Lo-Toga (Loh, Toga) [lht]: Endangered, 580 speakers; Austronesian; spoken in Vanuatu. Longgu (Logu) [lgu]: Threatened, 1,500 speakers; Austronesian; spoken in Solomon Islands. Lorediakarkar (Shark Bay) [lnn]: Severely endangered; Austronesian; spoken in Vanuatu. Lou (Anamagi, Torricelli) [tei]: Endangered, 953 speakers; Torricelli; spoken in Papua New Guinea. Löyöp (Lehalurup, Ureparapara, East Ureparapara) [urr]: Endangered, 200 speakers; Austronesian; spoken in Vanuatu. Madi (Gira, Girara) [grg]: Endangered, 380 speakers; Trans-New Guinea; spoken in Papua New Guinea. Magori [zgr]: Endangered, fewer than 200 speakers; Austronesian; spoken in Papua New Guinea. Makolkol (Neu-Pommern) [zmh]: Dormant; East New Britain; formerly spoken in Papua New Guinea. Maku’a (Lovaea, Lovaia) [lva]: Critically endangered, 50 speakers; Austronesian; spoken in East Timor. Malalamai (Bonga) [mmt]: Endangered, approximately 150 speakers; Austronesian; spoken in Papua New Guinea. Malua Bay (Middle Nambas) [mll]: Endangered, 500 speakers; Austronesian; spoken in Vanuatu. Mamaa (Mama) [mhf]: Endangered, 200 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mambae (Mambai, Manbae) [mgm]: At risk, 131,000 speakers; Austronesian; spoken in East Timor. Manam (Manum) [mva]: Threatened, 6,000 speakers; Austronesian; spoken in Papua New Guinea. Manambu [mle]: Threatened, approximately 2,800 speakers; Sepik; spoken in Papua New Guinea. Manat (Paynamar) [pmr]: Severely endangered, approximately 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mand (Atemble, Atemple-Apris, Atemple) [ate]: Critically endangered, eight speakers; TransNew Guinea; spoken in Papua New Guinea. Manem (Jeti, Yeti, Skofro) [jet]: Endangered, 400 speakers; Border; spoken in Papua New Guinea and Indonesia. Mangareva (Mangarevan) [mrv]: Endangered, fewer than 900 speakers; Austronesian; spoken in French Polynesia. Māori (Maori) [mri]: Endangered, 127,000 speakers; Austronesian; spoken in New Zealand. Mape [mlh]: Threatened, 5,120 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mari [mbx]: Endangered, 80 speakers; Sepik; spoken in Papua New Guinea. Maring (Mareng, Yoadabe-Watoare) [mbw]: Vulnerable, 11,000 speakers; Trans-New Guinea; spoken in Papua New Guinea. Marino (Naone, North Maewo, Sunwadia) [mrb]: Endangered, approximately 500 speakers; Austronesian; spoken in Vanuatu. Mavea (Mafea, Mav̋ea) [mkv]: Critically endangered, 32 speakers; Austronesian; spoken in Vanuatu. Mawak [mjj]: Severely endangered, 25 speakers; Trans-New Guinea; spoken in Papua New Guinea. 185

Anna Belew et al.

Mawan [mcz]: Endangered, 470 speakers; Trans-New Guinea; spoken in Papua New Guinea. May River Iwam [iwm]: Threatened, 3,000 speakers; Sepik; spoken in Papua New Guinea. Mbwenelang [no ISO 639-3 code]: Critically endangered, fewer than ten speakers; Austronesian; spoken in Vanuatu. Mea (Ha Mea, Hameha, Neukaledonien) [meg]: Endangered, 300 speakers; Austronesian; spoken in New Caledonia. Meakambut (Andai, Arafundi, Pundungum) [afd]: Endangered, 100 speakers; Arafundi; spoken in Papua New Guinea. Mehek (Driafleisuma, Indinogosima, Me’ek) [nux]: Threatened, fewer than 4,027 speakers; Sepik; spoken in Papua New Guinea. Mekmek [mvk]: Threatened, 1,036 speakers; Yuat; spoken in Papua New Guinea. Mende (Seim, Nihamber, Sambu) [sim]: Threatened, fewer than 5,057 speakers; Sepik; spoken in Papua New Guinea. Merei (Tiale, Lametin, Malmariv) [lmb]: Endangered, 400 speakers; Austronesian; spoken in Vanuatu. Mian (Mianmin, Miyanmin) [mpt]: Threatened, 1,750 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mindiri [mpn]: Endangered, approximately 90 speakers; Austronesian; spoken in Papua New Guinea. Miyak (Kyenele, Bulang, Keñele) [kql]: Endangered, 548 speakers; Yuat; spoken in Papua New Guinea. Mkir (Maii, Mae-Morae, Mafilau) [mmm]: Endangered, 180 speakers; Austronesian; spoken in Vanuatu. Moere [mvq]: Severely endangered, 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Moiso [no ISO 639-3 code]: Endangered, 100 speakers; Austronesian; spoken in Vanuatu. Mokerang (Mokareng, Mokoreng) [mft]: Endangered, 200 speakers; Austronesian; spoken in Papua New Guinea. Mokilese (Mokil, Mwoakilese, Mwoakiloa) [mkj]: Threatened, 1,230 speakers; Austronesian; spoken in Micronesia. Momare (Momale, Momole, Mumare) [msz]: Endangered, 374 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mongol-Kaimba (Mongol) [mgt]: Endangered, 338 speakers; Mongol-Langam-Ulwa; spoken in Papua New Guinea. Mono (Alu, Mono-Alu) [mte]: Threatened, 3,340 speakers; Austronesian; spoken in Solomon Islands. Monumbo (Mambuwan) [mxk]: Endangered, 450 speakers; Torricelli; spoken in Papua New Guinea. Mores (Ko, Morouas, Moruas) [mrp]: Endangered, 150 speakers; Austronesian; spoken in Vanuatu. Moresada (Murisapa, Murusapa-Sarewa, Murusapa) [msx]: Endangered, 200 speakers; TransNew Guinea; spoken in Papua New Guinea. Mortlockese (Mortlock, Nomoi) [mrl]: Threatened, 5,900 speakers; Austronesian; spoken in Micronesia. Mosimo [mqv]: Severely endangered, 50 speakers; Trans-New Guinea; spoken in Papua New Guinea. Mota [mtt]: Endangered, 700 speakers; Austronesian; spoken in Vanuatu. 186

The world’s endangered languages

Motlav (Motalava, Mwotlap, Mwotlav) [mlv]: Vulnerable, 2,100 speakers; Austronesian; spoken in Vanuatu. Mufian (Mufwian, Muhian, Muhiang) [aoj]: Vulnerable, 10,646 speakers; Torricelli; spoken in Papua New Guinea. Muniwara (Juwal, Mambe, Muniwara) [mwb]: Endangered, 826 speakers; Torricelli; spoken in Papua New Guinea. Munkip (Mungkip) [mpv]: Severely endangered, 12 speakers; Trans-New Guinea; spoken in Papua New Guinea. Murik (Nor, Nor-Murik Lakes) [mtf]: Threatened, fewer than 1,256 speakers; Lower SepikRamu; spoken in Papua New Guinea. Musom (Misatik) [msu]: Endangered, 219 speakers; Austronesian; spoken in Papua New Guinea. Mussau-Emira (Emira-Mussau, Musau-Emira, Mussau) [emi]: Threatened, 3,651 speakers; Austronesian; spoken in Papua New Guinea. Mwatebu [mwa]: Endangered, 120 speakers; Austronesian; spoken in Papua New Guinea. Mwesen (Mosina) [msn]: Critically endangered, ten speakers; Austronesian; spoken in Vanuatu. Nafi (Sirak) [srf]: Endangered, 160 speakers; Austronesian; spoken in Papua New Guinea. Nagu (Engdewu, Nanggu) [ngr]: Severely endangered, 206 speakers; Austronesian; spoken in Solomon Islands. Naha’ai (Malfaxal, Malvaxal-Toman Island, Taman) [mlx]: Endangered, 600 speakers; Austronesian; spoken in Vanuatu. Nahavaq (South West Bay, Sinesip, Seniang) [sns]: Endangered, 700 speakers; Austronesian; spoken in Vanuatu. Nai (Biaka, Nai-Biaka) [bio]: Endangered, 454 speakers; Kwomtari-Nai; spoken in Papua New Guinea. Najit [no ISO 639-3 code]: Critically endangered, fewer than five speakers; Austronesian; spoken in Vanuatu. Nake (Ale) [nbk]: Endangered, 170 speakers; Trans-New Guinea; spoken in Papua New Guinea. Nakwi [nax]: Endangered, 216 speakers; Left May; spoken in Papua New Guinea. Namat [nkm]: Endangered, 180 speakers; Morehead-Upper Maro; spoken in Papua New Guinea. Nambi (Menandon, Metan, Minendon) [mty]: Endangered, 484 speakers; Torricelli; spoken in Papua New Guinea. Namia (Edawapi, Lujere, Namie) [nnm]: Vulnerable, 3,500 speakers; Sepik; spoken in Papua New Guinea. Namo (Dorro, Mari, Tais) [mxw]: Endangered, 380 speakers; Morehead-Upper Maro; spoken in Papua New Guinea. Namonuito (Namon Weite) [nmt]: Endangered, 940 speakers; Austronesian; spoken in Micronesia. Narmoris (Polonombauk) [plb]: Endangered, approximately 120 speakers; Austronesian; spoken in Vanuatu. Nasarian (Mallicolo) [nvh]: Critically endangered, fewer than 20 speakers; Austronesian; spoken in Vanuatu. Nasvang [no ISO 639-3 code]: Endangered, 275 speakers; Austronesian; spoken in Vanuatu. Nāti (Naati) [no ISO 639-3 code]: Severely endangered, 25 speakers; Austronesian; spoken in Vanuatu. Nauete (Nauhete, Naueti, Naóti) [nxa]: Endangered, 1,000 speakers; Austronesian; spoken in East Timor. 187

Anna Belew et al.

Nauna (Naune) [ncn]: Severely endangered, 100 speakers; Austronesian; spoken in Papua New Guinea. Nauru Pacific Pidgin (NPP, Melanesian-Chinese Mixed Pidgin English in Nauru) [cpi]: Threatened; pidgin or creole; spoken in Nauru. Nauruan (Nauru) [nau]: Threatened, 6,000 speakers; Austronesian; spoken in Nauru. Navwien [no ISO 639-3 code]: Critically endangered, five speakers; Austronesian; spoken in Vanuatu. Nawaru (Sirio) [nwr]: Endangered, 190 speakers; Trans-New Guinea; spoken in Papua New Guinea. Neku [nek]: Endangered, 220 speakers; Austronesian; spoken in New Caledonia. Nemi [nem]: Endangered, 320 speakers; Austronesian; spoken in New Caledonia. Nen [nqn]: Endangered, 250 speakers; Morehead-Upper Maro; spoken in Papua New Guinea. Nese (Matanvat) [no ISO 639-3 code]: Critically endangered, 15 speakers; Austronesian; spoken in Vanuatu. Ngala (Gala, Kara, Nggala) [nud]: Endangered, 136 speakers; Sepik; spoken in Papua New Guinea. Ngalum (Sibil) [szb]: Vulnerable, 18,000 speakers; Trans-New Guinea; spoken in Papua New Guinea and Indonesia. Ngatik Men’s Creole (Ngatik Men’s Language, Micronesian Pidgin) [ngm]: Severely endangered, 15–30 speakers; pidgin or creole; spoken in Micronesia. Ngatikese [no ISO 639-3 code]: Vulnerable, approximately 700 speakers; Austronesian; spoken in Micronesia. Ngen (Shark Bay) [ssv]: Endangered, 800 speakers; Austronesian; spoken in Vanuatu. Niksek (Meiyari, Sumwari, Setiali) [gbe]: Endangered, 300 speakers; Sepik; spoken in Papua New Guinea. Nila [nil]: Threatened, 1,800 speakers; Austronesian; spoken in East Timor. Nimo (Nimo-Wasawai, Nimo-Wasuai, Birua) [niw]: Endangered, 178 speakers; Left May; spoken in Papua New Guinea. Ninde (Labo, Mewun, Meaun) [mwi]: Threatened, 1,100 speakers; Austronesian; spoken in Vanuatu. Ningera (Nagira, Negira, Ninggera) [nby]: Endangered, 213 speakers; Border; spoken in Papua New Guinea. Ninggerum (Kasiwa, Kativa, Muyu) [nxr]: Threatened, at least 3,500 speakers; Trans-New Guinea; spoken in Papua New Guinea and Indonesia. Ningil (Gul, Ral) [niz]: Endangered, 500 speakers; Torricelli; spoken in Papua New Guinea. Niolean (Repanbitip) [rpn]: Severely endangered, fewer than 90 speakers; Austronesian; spoken in Vanuatu. Nisvai (Vetbon) [no ISO 639-3 code]: Endangered, 200 speakers; Austronesian; spoken in Vanuatu. Nitita [no ISO 639-3 code]: Critically endangered, fewer than five speakers; Austronesian; spoken in Vanuatu. Niue (Niuean) [niu]: Threatened, 7,990 speakers; Austronesian; spoken in New Zealand, Cook Islands, Tonga, and Niue. Nivat [no ISO 639-3 code]: Critically endangered, fewer than ten speakers; Austronesian; spoken in Vanuatu. Njav [no ISO 639-3 code]: Critically endangered, ten speakers; Austronesian; spoken in Vanuatu. Nokuku (Nogugu) [nkk]: Endangered, 250 speakers; Austronesian; spoken in Vanuatu. 188

The world’s endangered languages

North Ambrym (Ambrym, North) [mmg]: Threatened, 5,250 speakers; Austronesian; spoken in Vanuatu. Nukna (Komutu) [klt]: Vulnerable, 1,000 speakers; Trans-New Guinea, Finisterre-Huon; spoken in Papua New Guinea. Nukuoro (Nukoro, Nuguor) [nkr]: Endangered, 900 speakers; Austronesian; spoken in Micronesia. Numbami (Siboma, Sipoma) [sij]: Endangered, 270 speakers; Austronesian; spoken in Papua New Guinea. Nume (Tasarag) [tgs]: Endangered, 700 speakers; Austronesian; spoken in Vanuatu. Numèè (Nââ Numèè, Kapone, Touaouru) [kdk]: Threatened, 1,810 speakers; Austronesian; spoken in New Caledonia. Odoodee (Kalamo, Nomad, Tomu) [kkc]: Endangered, 490 speakers; Trans-New Guinea; spoken in Papua New Guinea. Oksapmin (Nalu) [opm]: Vulnerable, 8,000 speakers; Trans-New Guinea; spoken in Papua New Guinea. Olo (Orlei, Wape, Wapi) [ong]: Vulnerable, 13,000 speakers; Torricelli; spoken in Papua New Guinea. Olrat [olr]: Critically endangered, three speakers; Austronesian; spoken in Vanuatu. One (Aunalei, Onele, Oni) [aun, oin, okk, onk, onr, osu]: Threatened, 4,450 speakers; Torricelli; spoken in Papua New Guinea. Onjob (Onjab) [onj]: Endangered, 150 speakers; Trans-New Guinea; spoken in Papua New Guinea. Orkon (Fanbak) [no ISO 639-3 code]: Severely endangered, 30 speakers; Austronesian; spoken in Vanuatu. Ôrôê (Orowe, Boewe) [bpk]: Endangered, 590 speakers; Austronesian; spoken in New Caledonia. Oroha (Mara Ma-Siki, Oraha) [ora]: Severely endangered, 38 speakers; Austronesian; spoken in Solomon Islands. Ouma [oum]: Critically endangered, four speakers; Austronesian; spoken in Papua New Guinea. Owiniga (Bero, Samo, Taina) [owi]: Endangered, 219 speakers; Left May; spoken in Papua New Guinea. Pááfang [pfa]: Threatened, 1,320 speakers; Austronesian; spoken in Micronesia. Pagi (Bembi, Pagei) [pgi]: Threatened, 1,100 speakers; Border; spoken in Papua New Guinea. Pahi (Lugitama, Riahoma, Wansum) [lgt]: Endangered, fewer than 598 speakers; Sepik; spoken in Papua New Guinea. Papapana [ppn]: Severely endangered, 120 speakers; Austronesian; spoken in Papua New Guinea. Papi (Paupe, Baiyamo) [ppe]: Severely endangered, 70 speakers; Unclassified; spoken in Papua New Guinea. Papitalai [pat]: Endangered, 520 speakers; Austronesian; spoken in Papua New Guinea. Pasi (Besi, Warasai, Yau) [psq]: Endangered, fewer than 154 speakers; Sepik, Tama; spoken in Papua New Guinea. Pawaia (Aurama, Pavaia, Pawaian) [pwa]: Threatened, 4,000 speakers; isolate; spoken in Papua New Guinea. Pei (Pai) [ppq]: Severely endangered, 80 speakers; Unclassified; spoken in Papua New Guinea. Pele-Ata [ata]: Threatened, approximately 2,000 speakers; isolate; spoken in Papua New Guinea. 189

Anna Belew et al.

Penrhyn (Tongareva, Mangarongaro, Penrhynese) [pnh]: Severely endangered, approximately 500 speakers; Austronesian; spoken in Cook Islands. Piamatsina [ptr]: Endangered, 250 speakers; Austronesian; spoken in Vanuatu. Piame (Biami) [pin]: Severely endangered, fewer than 100 speakers; Sepik; spoken in Papua New Guinea. Pije (Pinje, Pindje, Neukaledonien) [piz]: Endangered, 160 speakers; Austronesian; spoken in New Caledonia. Pinai-Hagahai (Aramaue, Aramo, Erem) [pnn]: Vulnerable, 600 speakers; Piawi; spoken in Papua New Guinea. Pingelapese (Pingelap, Pingilapese) [pif]: Threatened, 3,000 speakers; Austronesian; spoken in Guam and Micronesia. Pitcairn-Norfolk (Pitcairnese, Pitcairn English, Pitkern) [pih]: Endangered, fewer than 630 speakers; pidgin or creole; spoken in Norfolk Island and Pitcairn. Piu (Sanbiau, Lanzog, Kuruko) [pix]: Endangered, 100 speakers; Austronesian; spoken in Papua New Guinea. Port Sandwich (Lamap) [psw]: Threatened, 1,200 speakers; Austronesian; spoken in Vanuatu. Pouye (Bouye, Buiye) [bye]: Threatened, fewer than 817 speakers; Sepik; spoken in Papua New Guinea. Puare (Puari, Bkwala) [pux]: Severely endangered, 10–99 speakers; Sko; spoken in Papua New Guinea. Pukapuka (Bukabukan, Pukapukan) [pkp]: Endangered, 2,030 speakers; Austronesian; spoken in New Zealand and Cook Islands. Puluwatese (Puluwat) [puw]: Threatened, 1,360 speakers; Austronesian; spoken in Micronesia. Purari (Evorra, Iai, Koriki) [iar]: Threatened, 7,000 speakers; isolate; spoken in Papua New Guinea. Pwaamei (Poamei) [pme]: Endangered, 220 speakers; Austronesian; spoken in New Caledonia. Pwapwâ (Poapoa) [pop]: Severely endangered, 16 speakers; Austronesian; spoken in New Caledonia. Pyu [pby]: Endangered, approximately 100 speakers; isolate; spoken in Papua New Guinea and Indonesia. Qaqet (Baining, Maqaqet, Kakat) [byx]: Threatened, 6,350 speakers; East New Britain; spoken in Papua New Guinea. Rakahanga-Manihiki (Manihiki-Rakahanga) [rkh]: Endangered, 320 speakers; Austronesian; spoken in New Zealand and Cook Islands. Raljago (Lonwolwol, Craig Cove, Fali) [crc]: Threatened, 1,200 speakers; Austronesian; spoken in Vanuatu. Rao (Annaberg, Rao Breri) [rao]: Threatened, 4,000 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Rapa (Rapan) [ray]: Critically endangered, fewer than 50 speakers; Austronesian; spoken in French Polynesia. Rarotongan (Cook Island, Cook Islands Maori, Maori) [rar]: Vulnerable, 33,220 speakers; Austronesian; spoken in French Polynesia, New Zealand, and Cook Islands. Ratsua [no ISO 639-3 code]: Vulnerable, approximately 200 speakers; Austronesian; spoken in Papua New Guinea. Rawo [rwa]: Endangered, 506 speakers; Sko; spoken in Papua New Guinea. Rema (Bothar) [bow]: Severely endangered; Morehead-Upper Maro; spoken in Papua New Guinea. 190

The world’s endangered languages

Rennell Island Sign Language (Rennellese Sign Language) [rsi]: Critically endangered, one speaker; sign language; spoken in Solomon Islands. Rerep (Pangkumu, Tisman) [pgk]: Endangered, 380 speakers; Austronesian; spoken in Vanuatu. Retlatur [no ISO 639-3 code]: Severely endangered, 100 speakers; Austronesian; spoken in Vanuatu. Ririo [rri]: Severely endangered, 79 speakers; Austronesian; spoken in Solomon Islands. Romkun (Romkuin) [rmk]: Endangered, 350 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Saliba [sbe]: Threatened, 2,500 speakers; Austronesian; spoken in Papua New Guinea. Samosa [swm]: Severely endangered, 90 speakers; Trans-New Guinea; spoken in Papua New Guinea. Saniyo-Hiyewe (Sanio, Sanio-Hiowe) [sny]: Endangered, 644 speakers; Sepik; spoken in Papua New Guinea. Saruga [sra]: Endangered, 130 speakers; Trans-New Guinea; spoken in Papua New Guinea. Satawalese [stw]: Endangered, 460 speakers; Austronesian; spoken in Micronesia. Savosavo (Savo, Savo Island) [svs]: Threatened, 2,500 speakers; isolate; spoken in Solomon Islands. Se (Fanafo) [no ISO 639-3 code]: Severely endangered, 20 speakers; Austronesian, MalayoPolynesian, Oceanic, Northern Vanuatu; spoken in Vanuatu. Seke (Ske) [ske]: Threatened, 600 speakers; Austronesian; spoken in Vanuatu. Sene [sej]: Severely endangered, ten speakers; Trans-New Guinea; spoken in Papua New Guinea. Sengo (Manambu) [spk]: Endangered, approximately 300 speakers; Sepik; spoken in Papua New Guinea. Sengseng (Asengseng) [ssz]: Threatened, 1,750 speakers; Austronesian; spoken in Papua New Guinea. Sepa [spe]: Endangered, 200 speakers; Austronesian; spoken in Papua New Guinea. Sepen (Akukem) [spm]: Endangered, 409 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Sepik Iwam (Yawenian) [iws]: Threatened, 2,500 speakers; Sepik; spoken in Papua New Guinea. Sera (Ssia, Serra) [sry]: Endangered, 432 speakers; Austronesian; spoken in Papua New Guinea. Seta (Aunalei) [stf]: Endangered, 155 speakers; Torricelli; spoken in Papua New Guinea. Setaman (Setamin) [stm]: Threatened, 381 speakers; Trans-New Guinea; spoken in Papua New Guinea. Seti (Aunalei) [sbi]: Endangered, 113 speakers; Torricelli; spoken in Papua New Guinea. Siawi (Musa, Musan, Musian) [mmp]: Severely endangered, approximately 75 speakers; Amto-Musan; spoken in Papua New Guinea. Sibe (Nagovisi, Sibbe, Sibe-Nagovisi) [nco]: Threatened, 5,000 speakers; South Bougainville; spoken in Papua New Guinea. Siliput (Mai, Maimai, Seleput) [mkc]: Endangered, fewer than 263 speakers; Torricelli; spoken in Papua New Guinea. Silopi [xsp]: Endangered, 180 speakers; Trans-New Guinea; spoken in Papua New Guinea. Sinia (Navut) [nsw]: Endangered, 520 speakers; Austronesian, Malayo-Polynesian, Oceanic, Northern Vanuatu; spoken in Vanuatu. Sirva (Sileibi) [sbq]: Endangered, approximately 200 speakers; Trans-New Guinea; spoken in Papua New Guinea. 191

Anna Belew et al.

Sîshëë (Zire, Zira, Sirhe) [sih]: Awakening; Austronesian; spoken in New Caledonia. Sissano (Sisano, Sinano, Sinama) [sso]: Endangered, 300 speakers; Austronesian; spoken in Papua New Guinea. Som (Somm) [smc]: Severely endangered, 80 speakers; Trans-New Guinea; spoken in Papua New Guinea. Sonsorol (Sonsorolese, Sonsorol-Tobi) [sov]: Endangered, at least 600 speakers; Austronesian; spoken in Northern Mariana Islands and Palau. Sörsörian (Sosorian) [no ISO 639-3 code]: Critically endangered, four speakers; Austronesian; spoken in Vanuatu. South Marquesan [mqm]: Threatened, 1,000–2,000 speakers; Austronesian; spoken in French Polynesia. Sowa [sww]: Dormant; Austronesian; formerly spoken in Vanuatu. Sowanda (Punda-Umeda, Pundina, Waina) [sow]: Threatened, 1,100 speakers; Border; spoken in Papua New Guinea and Indonesia. Suena (Yema, Yarawe, Yarawi) [sue]: Threatened, 3,000 speakers; Trans-New Guinea; spoken in Papua New Guinea. Suganga (Blimo, North Mianmin, Wagarabai) [sug]: Endangered, 680 speakers; Trans-New Guinea; spoken in Papua New Guinea. Sulka [sua]: Threatened, 2,500 speakers; isolate; spoken in Papua New Guinea. Sumariup (Sogoba, Latoma) [siv]: Severely endangered, 80 speakers; Sepik; spoken in Papua New Guinea. Susuami [ssu]: Severely endangered, ten speakers; Trans-New Guinea; spoken in Papua New Guinea. Tabo (Hibaradai, Hiwi, Waia) [knv]: Threatened, 3,000 speakers; isolate; spoken in Papua New Guinea. Taiap (East Parana, Guana, Chuala) [gpn]: Severely endangered, 75 speakers; isolate; spoken in Papua New Guinea. Tamambo (Malo, Tamabo) [mla]: Vulnerable, 3,000–4,000 speakers; Austronesian; spoken in Vanuatu. Tanapag (Northern Carolinian, Tallabwog) [tpv]: Threatened, 4,400 speakers; Austronesian; spoken in Northern Mariana Islands. Tanema (Tanima) [tnx]: Critically endangered, four speakers; Austronesian; spoken in Solomon Islands. Tangoa (Movono) [tgp]: Endangered, 370 speakers; Austronesian; spoken in Vanuatu. Tangu (Tanggu, Tanggum) [tgu]: Threatened, 2,330 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Tanguat [tbs]: Endangered, 510 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Tanimbili (Nyisunggu) [tbe]: Severely endangered, 50 speakers; Austronesian; spoken in Solomon Islands. Tape (Maragus, Maragaus, Mallicolo) [mrs]: Critically endangered, 10–15 speakers; Austronesian; spoken in Vanuatu. Tasmate (Meri, Oa) [tmt]: Endangered, 300 speakers; Austronesian; spoken in Vanuatu. Taulil (Butam, Neu-Pommern) [tuh]: Endangered, 800 speakers; East New Britain; spoken in Papua New Guinea. Teanu (Buma, Puma) [tkw]: Severely endangered; Austronesian; spoken in Solomon Islands. Telefol (Eripataman, Teleefool, Telefool) [tlf]: Threatened, 4,500 speakers; Telefol; spoken in Papua New Guinea. 192

The world’s endangered languages

Tenis (Tench) [tns]: Severely endangered, 30 speakers; Austronesian; spoken in Papua New Guinea. Teop [tio]: Threatened, 5,000 speakers; Austronesian; spoken in Papua New Guinea. Terebu (Terepu, Turupu, Turubu) [trb]: Endangered, 130 speakers; Austronesian; spoken in Papua New Guinea. Tholp (Nethalp) [no ISO 639-3 code]: Dormant; Austronesian; formerly spoken in Vanuatu. Tifal (Tifalmin) [tif]: Threatened, 2,800 speakers; Trans-New Guinea; spoken in Papua New Guinea. Tirax (Mae, Dirak) [mme]: Endangered, 1,000 speakers; Austronesian; spoken in Vanuatu. Tobian (Tobi, Hatohobei) [tox]: Severely endangered, 22 speakers; Austronesian; spoken in Palau. Tokelauan (Tokelau, Fakaafo) [tkl]: Threatened, fewer than 8,000 speakers; Austronesian; spoken in New Zealand and Tokelau. Toksiki (Soisoru, Roria, Mores) [rga]: Severely endangered, 75 speakers; Austronesian; spoken in Vanuatu. Tolomako (Tolomako-Jereviu, Big Bay, Marina) [tlm]: Endangered, fewer than 450 speakers; Austronesian; spoken in Vanuatu. Touo [tqu]: Threatened, 1,870 speakers; isolate; spoken in Papua New Guinea. Tuamotuan (Pa’umotu) [pmt]: Vulnerable, 14,400 speakers; Austronesian; spoken in French Polynesia. Tumleo [tmq]: Endangered, 610 speakers; Austronesian; spoken in Papua New Guinea. Turaka [trh]: Severely endangered, 25 speakers; Trans-New Guinea; spoken in Papua New Guinea. Turumsa [tqm]: Critically endangered, five speakers; Unclassified; spoken in Papua New Guinea. Tutuba [tmi]: Threatened, 120 speakers; Austronesian; spoken in Vanuatu. Tuvaluan (Ellice, Ellicean, Tuvalu) [tvl]: Vulnerable, approximately 9,000 speakers; Austronesian; spoken in Nauru, New Zealand, Fiji, Kiribati, and Tuvalu. Tuwari (Akiapmin) [tww]: Severely endangered, 90 speakers; Unclassified; spoken in Papua New Guinea. Uisai [uis]: Threatened, 2,500 speakers; South Bougainville; spoken in Papua New Guinea. Ukuriguma [ukg]: Endangered, 170 speakers; Trans-New Guinea; spoken in Papua New Guinea. Ulau-Suain (Suain) [svb]: Threatened, 1,271 speakers; Austronesian; spoken in Papua New Guinea. Ulithian [uli]: Threatened, 3,075 speakers; Austronesian; spoken in Micronesia. Ulwa (Yaul) [yla]: Severely endangered, 600 speakers; Mongol-Langam-Ulwa; spoken in Papua New Guinea. Unua (Onua) [onu]: Vulnerable; Austronesian; spoken in Vanuatu. Ura (Uramät, Uramit, Uramet) [uro]: Threatened, 1,900 speakers; East New Britain; spoken in Papua New Guinea. Ura (Aryau, Arau) [uur]: Critically endangered, approximately six speakers; Austronesian; spoken in Vanuatu. Urapmin [urm]: Endangered, 370 speakers; Trans-New Guinea; spoken in Papua New Guinea. Urat (Urakin, Wasepnau) [urt]: Threatened, 6,280 speakers; Torricelli; spoken in Papua New Guinea. Urim (Kalp, Kurom, Tayon) [urim]: Threatened, 2,538 speakers; Torricelli; spoken in Papua New Guinea. 193

Anna Belew et al.

Urimo (Yaugiba) [urx]: Endangered, 835 speakers; Torricelli; spoken in Papua New Guinea. Uruava [urv]: Dormant; Austronesian; formerly spoken in Papua New Guinea. Uya (Usu) [usu]: Endangered, 270 speakers; Trans-New Guinea; spoken in Papua New Guinea. Vaeakau-Taumako (Pileni, Pilheni, Taumako) [piv]: Vulnerable, 1,662 speakers; Austronesian; spoken in Solomon Islands. Valpei [vlp]: Endangered, 300 speakers; Austronesian; spoken in Vanuatu. Vamale (Moaeke, Hmwaeke, Pamale) [mkt]: Endangered, 150 speakers; Austronesian; spoken in New Caledonia. Vanimo (Dumo, Duso, Manimo) [vam]: Vulnerable, 1,660 speakers; Sko; spoken in Papua New Guinea. Vano (Vanikoro, Vanikolo, Lovono) [vnk]: Critically endangered, five speakers; Austronesian; spoken in Solomon Islands. Varavara (Amblong, Aje) [alm]: Endangered, at least 100 speakers; Austronesian; spoken in Vanuatu. Vehes (Buasi, Vehees) [val]: Severely endangered, 70 speakers; Austronesian; spoken in Papua New Guinea. Vera’a [vra]: Endangered, 275 speakers; Austronesian; spoken in Vanuatu. Vinmavis (Vinmavis, Lambumbu, Mallicolo) [vnm]: Endangered, 500 speakers; Austronesian; spoken in Vanuatu. Vivti [no ISO 639-3 code]: Critically endangered, fewer than five speakers; Austronesian; spoken in Vanuatu. Volow (Valuwa) [mlv]: Critically endangered, one speaker; Austronesian; spoken in Vanuatu. Vovo [no ISO 639-3 code]: Endangered, 475 speakers; Austronesian; spoken in Vanuatu. Vunapu [vnp]: Endangered, 380 speakers; Austronesian; spoken in Vanuatu. Vurës (Mosina, Mosin) [msn]: Threatened, 1,050 speakers; Austronesian; spoken in Vanuatu. Wab (Som) [wab]: Endangered, 120 speakers; Austronesian; spoken in Papua New Guinea. Wailapa (Ale) [wlr]: Endangered, 100 speakers; Austronesian; spoken in Vanuatu. Waima’a (Uai Ma’a, Waimaha, Waimoa) [wmh]: Threatened, 3,000 speakers; Austronesian; spoken in East Timor. Walio [wla]: Endangered, 200 speakers; isolate; spoken in Papua New Guinea. Walman (Valman, Koroko) [van]: Severely endangered, 1,740 speakers; Torricelli; spoken in Papua New Guinea. Wamas [wmc]: Endangered, 220 speakers; Trans-New Guinea; spoken in Papua New Guinea. Wantoat (Yagawak, Bam) [wnc]: Threatened, 8,200 speakers; Trans-New Guinea; spoken in Papua New Guinea. Warapu (Barupu, Varofu, Waropu) [wra]: Endangered, a few hundred speakers; Sko; spoken in Papua New Guinea. Waris (Walsa) [wrs]: Threatened, fewer than 4,400 speakers; Border; spoken in Papua New Guinea and Indonesia. Watakataui (Waxe, Waxei) [wtk]: Endangered, 160 speakers; Sepik; spoken in Papua New Guinea. Watam (Marangis) [wax]: Threatened, fewer than 1,086 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Weliki (Weleki, Karangi) [klh]: Endangered, 200 speakers; Trans-New Guinea; spoken in Papua New Guinea. Wiaki (Menandon, Miniden, Minidien) [wii]: Critically endangered, 100 speakers; Torricelli; spoken in Papua New Guinea. 194

The world’s endangered languages

Wiarumus (Mandi, Imandi) [tua]: Severely endangered, 460 speakers; Torricelli; spoken in Papua New Guinea. Wiru (Witu) [wiu]: Vulnerable, 15,300 speakers; isolate; spoken in Papua New Guinea. Wogamusin (Wongamusin) [wog]: Endangered, 368 speakers; Sepik; spoken in Papua New Guinea. Wogeo (Wokeo, Uageo) [woc]: Threatened, approximately 1,500 speakers; Austronesian; spoken in Papua New Guinea. Woleaian [woe]: Threatened, 1,630 speakers; Austronesian; spoken in Micronesia. Wom (Wam) [wmo]: Threatened, 2,500 speakers; Torricelli; spoken in Papua New Guinea. Wusi (Wusi-Kerepua, Kula) [wsi]: Endangered, 350 speakers; Austronesian; spoken in Vanuatu. Wutung (Udung, Sangke) [wut]: Endangered, 610 speakers; Sko; spoken in Papua New Guinea. Yahang (Rurihi’ip, Ruruhip, Ruruhi’ip) [rhp]: Endangered, fewer than 1,182 speakers; Torricelli; spoken in Papua New Guinea. Yakaikeke [ykk]: Endangered, 100 speakers; Austronesian; spoken in Papua New Guinea. Yalë (Yale, Yade, Yadë) [nce]: Endangered, 580 speakers; isolate; spoken in Papua New Guinea. Yambes [ymb]: Endangered, 860 speakers; Torricelli; spoken in Papua New Guinea. Yarawata [yrw]: Endangered, 130 speakers; Trans-New Guinea; spoken in Papua New Guinea. Yau [yyu]: Endangered, 140 speakers; Torricelli; spoken in Papua New Guinea. Yawiyo (Yabio) [ybx]: Severely endangered, 60–100 speakers; Unclassified; spoken in Papua New Guinea. Yele (Yélî Dnye) [yle]: Threatened, 6,000 speakers; isolate; spoken in Papua New Guinea. Yelogu (Yalaku, Buiamanambu, Kaunga) [ylg]: Endangered, approximately 700 speakers; Sepik; spoken in Papua New Guinea. Yerakai (Yerekai, Garamambu) [yra]: Endangered, 390 speakers; isolate; spoken in Papua New Guinea. Yeri (Yapunda, Reiwo, Hiro) [yev]: Severely endangered, 60 speakers; Torricelli; spoken in Papua New Guinea. Yessan-Mayo (Maio-Yesan, Mayo-Yesan, Yasyin) [yss]: Endangered, 900 speakers; Sepik; spoken in Papua New Guinea. Yil [yll]: Threatened, 2,134 speakers; Torricelli; spoken in Papua New Guinea. Yimas [yee]: Endangered, 250 speakers; Lower Sepik-Ramu; spoken in Papua New Guinea. Yis [yis]: Endangered, 489 speakers; Torricelli; spoken in Papua New Guinea. Yonggom (Yongom, Yongkom) [yon]: Threatened, 4,000 speakers; Trans-New Guinea; spoken in Papua New Guinea and Indonesia. Zazao (Jajao, Kilokaka) [jaj]: Critically endangered; Austronesian; spoken in Solomon Islands.

South America Lyle Campbell Linguistic diversity in South America25 If the extensive linguistic diversity of the Americas is famous, with over 40% the language families (including isolates) of the world,26 that of South America is infamous, with 108 independent language families (including language isolates) – that comes to 62% of the language families in the Americas and to 26.5% of those of all the world. South America’s 108 195

Anna Belew et al.

language families include 53 families that have at least two languages as members that unquestionably are genetically related to one another, and 55 language isolates – languages with no known relatives, in effect language families that have only a single member. Of the 53 families, 43 are small, with six or fewer members, while ten are relatively large, both in terms of number of languages and their geographical expanse (Campbell 2012). There are 308 languages listed in ELCat for South America; one is awakening and 43 are dormant, with 264 endangered languages still spoken. It is uncertain how many languages there were in South America at the time of first European contact, though the number was certainly large. The number often cited is that of Loukotka’s (1968) 1,492 “languages,” i.e. names of languages listed for South America, though many of these are known to be errors, for example, alternate names for groups known by other names, or merely names of rivers, clans, etc. A more likely but still vague estimate is c.800 different languages at the time of the arrival of the first Europeans. The number of endangered languages in ELCat for each South American country is: Argentina 17 (two non-indigenous, one of them a sign language; in addition, there are three dormant languages); Bolivia 39 (one non-indigenous, a sign language; eight are dormant); Brazil 179 (one is awakening, 20 are dormant); Chile seven (one is a non-indigenous sign language; one, Rapa Nui, is Polynesian); Colombia 76 (three are dormant; two are sign languages, one non-indigenous; two are creole languages, one indigenous); Ecuador 15 (one is a non-indigenous sign language); French Guiana seven; Guyana ten (one is a dormant non-indigenous creole); Paraguay 16; Peru 72 (six dormant; one non-indigenous sign language); Suriname 11 (two dormant languages, one of which is a non-indigenous pidgin); Uruguay two (one is a nonindigenous sign language); and Venezuela 39 (four are dormant; one is a non-indigenous sign language). As in North America, a number of languages are spoken in more than one country.

Causes of language endangerment in South America The causes of language endangerment in South America vary from region to region, and have varied over time, although there are broad, recurring factors that are clear enough. Diseases introduced from Europe devastated many indigenous populations. Estimates of pre-European contact populations sizes vary widely; nevertheless, a figure of 90% decline in the indigenous populations soon after 1520 is often repeated (see Denevan 1992). Even today, more isolated groups are especially vulnerable to these introduced diseases. It would not be amiss to cite sheer greed as a major cause of language endangerment in South America. The rubber boom (1879–1912) had horrible consequences for the indigenous population across the Amazon basin, where Indians were pressed into forced labor and large numbers died. Greedy resource exploitation has continued up to the present day, with a long record of murder and massacres of indigenous groups by unscrupulous loggers, miners, ranchers, and exploiters of oil, gas, and other mineral resources, resulting in displacement and genocide of whole groups. Drug trafficking and armed conflicts have also had an impact on several groups, leading also to displacement in several regions. Other causes include political, social, and other other economic factors. The history of violence and brutal attacks against indigenous groups extends from the conquistadores to the present. Cases of deliberate attempts to introduce infections, to poison water supplies, etc. are well-known. Colonial policies fostered a shift to Spanish and Portuguese. Bilingualism between local languages and more influential indigenous languages – such as Quechua, Tukano, and Nheengatu – has led to language shift to these larger indigenous languages. The same issues of discrimination, repression, stigmatization, negative status of the indigenous languages, official 196

The world’s endangered languages

language policies, and lack of institutional support are in force here, too. Intermarriage with Europeans has contributed to loss of indigenous languages and assimilation of these populations. In some instances, intermarriage among tribes has also contributed to endangerment of some languages, as well.

The current status of endangered languages in South America As elsewhere in the Americas, all South America’s indigenous languages are considered endangered by ELCat’s criteria, some obviously much more endangered than others. If we contrast the figure that 100% of South American languages are endangered with the figure that 46% of languages are endangered worldwide,27 we see the seriousness of language endangerment in South America in high relief. The Catalogue of Endangered Languages (ELCat) lists 374 endangered languages in South America (43 of them are dormant; one, Umutina, is awakening). Most of them are indigenous; most of the languages of the many immigrant groups from Europe and Asia in South America are not listed here, rather are treated with their countries of origin. Ten of ELCat’s endangered languages in South America are sign languages (one is indigenous, Urubú-Kaabor Sign Language). There are four endangered pidgin and creole languages, two of them indigenous (Ndyuka-Trio Pidgin and Nheengatu), and there are two mixed languages (Kallawaya and Media Lengua). The vitality of many of the 330 endangered languages that still have speakers is far from robust: 54 are “Critically Endangered” and 25 others are “Severely Endangered”; together, 79 are “Critically” or “Severely Endangered.” Of these, 36 have 10 or fewer speakers; 71 have 100 or fewer speakers. Below is an alphabetical listing of all languages of South America currently included in the Catalogue of Endangered Languages.

Endangered languages of South America Achagua (Achagua, Xagua, Ajagua) [aca]: Endangered, 283 speakers; Arawakan; spoken in Colombia. Aché (Axe, Guaiaqui, Guoyagui) [guq]: Severely endangered, 200 speakers; Tupian; spoken in Paraguay. Achuar (Achual, Achuar-Shiwiar, Achuale) [acu]: Threatened, 5,443 speakers; Jivaroan; spoken in Peru and Ecuador. Agavotaguerra (Agavotoküeng, Agavotoqüeng, Agavo-queng) [avo]: Dormant; Arawakan; formerly spoken in Brazil. Aguaruna (Aguajun, Ahuajun, Awaruna) [agr]: Vulnerable, approximately 55,000 speakers; Jivaroan; spoken in Peru. Aikanã (Aikaná, Huarí, Warí) [tba]: Vulnerable, 175–200 speakers; isolate; spoken in Brazil. Akuntsú (Akunsu, Akunts’u) [aqz]: Critically endangered, four speakers; Tupian; spoken in Brazil. Akuriyo (Tiriyometesem, Triometesen, Akurio) [ako]: Dormant; Cariban; formerly spoken in Suriname. Alto Marañón Quechua (Alto Pativilca, Margos-Yarowilca-Lauricocha Quechua) [qvm]: Vulnerable, 83,400 speakers; Quechuan; spoken in Peru. Alto Pativilca (Alto Marañón, Ambo-Pasco Quechua, San Rafael-Huariaca Quechua) [qva, qxh]: At risk, 140,000 speakers; Quechuan; spoken in Peru. 197

Anna Belew et al.

Amahuaca (Amawaka, Ameuhaque, Sayaco) [amc]: Severely endangered, approximately 100 speakers; Pano-Tacanan; spoken in Brazil and Peru. Amanayé (Amanajé, Manaze, Amanage) [ama]: Dormant; Tupian; formerly spoken in Brazil. Amuesha (Amuexa, Yanesha’, Amoesha) [ame]: Threatened, 8,000 speakers; Arawakan; spoken in Peru. Anambé of Cairarí [aan]: Dormant; Tupian; formerly spoken in Brazil. Andoa (Andoa, Gae, Siaviri) [anb]: Dormant; Zaparoan; formerly spoken in Peru. Andoque (Andoke, Cho’oje, Patsiaehé) [ano]: Endangered, 597 speakers; isolate; spoken in Colombia. Angaité (Angaite, Angate, Enenlhet) [aqt]: Severely endangered, 780 speakers; Mascoyan; spoken in Paraguay. Apalaí (Apalai, Aparai, Apalay) [apy]: Endangered, 317 speakers; Cariban; spoken in Brazil. Apiaká (Apiacá, Apiake) [api]: Dormant; Tupian; formerly spoken in Brazil. Apinayé (Apinage, Apinajé) [apn]: Threatened, 1,500 speakers; Jean; spoken in Brazil. Apurí Quechua (Central Yungay Quechua) [no ISO 639-3 code]: Endangered, fewer than 6,500 speakers; Quechuan; spoken in Peru. Apuriná (Ipuriná, Kangite, Apurinã) [apu]: Threatened, 2,000 speakers; Arawakan; spoken in Brazil. Arabela (Chiripuno, Chiripunu, Arabela-Andoa) [arl]: Critically endangered, 30 speakers; Zaparoan; spoken in Peru. Araona (Cavina, Carina) [aro]: Endangered, 111 speakers; Pano-Tacanan; spoken in Bolivia. Arapaso (Arapaáo, Araspaso, Koneá) [arj]: Dormant; Tucanoan; formerly spoken in Brazil. Arára of Mato Grosso (Arara do Beiradão, Arara do Aripuanã) [axg]: Dormant; isolate; formerly spoken in Brazil. Araweté (Bïde, Asuriní, Asuriní do Xingú) [awt]: Endangered, 339 speakers; Tupian; spoken in Brazil. Argentine Sign Language (Lengua de Señas Argentina, LSA) [aed]: No estimate available; sign language; spoken in Argentina. Arikapú (Arikapu, Aricapú, Aricapu) [ark]: Critically endangered, one speaker; Jabutian; spoken in Brazil. Aruá (Aruaxi, Aruashí, Cinta Larga) [arx]: Critically endangered, five speakers; Tupian; spoken in Brazil. Asháninka (Asháninga, Axininca Campa, Asháninca) [cni]: Vulnerable, 89,516 speakers; Arawakan; spoken in Brazil and Peru. Ashéninka (Ashéninga, Ashéninka Perené) [cpc, cjo, prq, cpu, cpy, cpb]: Vulnerable, 20,000 speakers; Arawakan; spoken in Brazil and Peru. Asuriní do Tocantins (Assuriní) [asu]: Endangered, 384 speakers; Tupian; spoken in Brazil. Asuriní of Xingú (Asuriní de Coatinema, Asuriní do Koatinema) [asn]: Endangered, 125 speakers; Tupian; spoken in Brazil. Auré (Aurê, Aurá) [aux]: Dormant; Tupian; formerly spoken in Brazil. Avá (Avá-Canoeiro, Canoeiros, Canoe) [avv]: Severely endangered, 16 speakers; Tupian; spoken in Brazil. Awa Pit (Coaiquer, Cuaiquer, Quaiquer) [kwi]: Vulnerable, fewer than 17,464 speakers; Barbacoan; spoken in Colombia and Ecuador. Awaké (Auaqué, Arutani, Auaké) [atx]: Dormant; isolate; formerly spoken in Brazil and Venezuela. Awetí (Awety, Awetö, Aueto) [awe]: Endangered, 170 speakers; Tupian; spoken in Brazil. 198

The world’s endangered languages

Ayacucho Quechua (Ayacucho, Runasimi, Chanka) [quy]: At risk, 850,000 speakers; Quechuan; spoken in Peru. Aymara (Aymará, Southern Aymara) [ayc]: At risk, fewer than 1,471,256 speakers; Aymaran; spoken in Peru, Argentina, Bolivia, and Chile. Ayoreo (Ayoré, Zamuco, Morotoco) [ayo]: Threatened, 3,154 speakers; Zamucoan; spoken in Bolivia and Paraguay. Bakairí (Bakairi, Kura, Bacairí) [bkq]: Endangered, 900 speakers; Cariban; spoken in Brazil. Banawá (Banawá Yafí, Baniwá-Jafí) [no ISO 639-3 code]: Endangered, 100 speakers; Arawan; spoken in Brazil. Baniwa (Baniwa, Baniva, Baniva do Içana) [bwi]: Threatened, at least 9,554 speakers; Arawakan; spoken in Brazil, Colombia, and Venezuela. Bará (Waimaha, Waimaja, Bará) [bao]: Endangered, 130 speakers; Tucanoan; spoken in Brazil and Colombia. Barasano (Barasana, Southern Barasano, Paneroa) [bsn]: Threatened, 1,924 speakers; Tucanoan; spoken in Brazil and Colombia. Baré (Bare, Mitua, Barauna) [bae]: Critically endangered, two speakers; Arawakan; spoken in Brazil and Venezuela. Barí (Motilón, Motilon, Motilone) [mot]: Vulnerable, 4,000 speakers; Chibchan; spoken in Colombia and Venezuela. Baure (Bauré, Chiquimiti) [brg]: Critically endangered, 20 speakers; Arawakan; spoken in Bolivia. Berbice Creole Dutch (Berbice Dutch) [brc]: Dormant; pidgin or creole; formerly spoken in Guyana. Bolivian Quechua (Boliviano, South Bolivian Quechua, Central Bolivian Quechua) [quh]: At risk, 2,300,000 speakers; Quechuan; spoken in Argentina and Bolivia. Bolivian Sign Language [bvl]: Threatened; sign language; spoken in Bolivia. Bora (Imihita, Miranya, Boro) [boa]: Severely endangered, 2,000–3,000 speakers; Boran; spoken in Brazil, Peru, and Colombia. Bororo (Boe, Borôro, Bororo proper) [bor]: Endangered, approximately 700 speakers; Bororoan; spoken in Brazil. Cabiyarí (Cauyari, Caryari, Cabiuarí) [cbb]: Endangered, 311 speakers; Arawakan; spoken in Colombia. Cahuarano [cah]: Dormant; Zaparoan; formerly spoken in Peru. Cajamarca Quechua (Northern Quechua) [qvc]: Vulnerable, approximately 30,000 speakers; Quechuan; spoken in Peru. Camsá (Coche, Kamsá, Sibundoy) [kbh]: Threatened, 4,773 speakers; isolate; spoken in Colombia. Candoshi (Candoxi, Candoši, Kandoshi) [cbu]: Threatened, 1,586 speakers; isolate; spoken in Peru. Canela (Canela, Canela-Krahô, Timbira) [ram]: Threatened, 2,502 speakers; Jean; spoken in Brazil. Canichana (Kanichana, Canesi, Canisiana) [caz]: Dormant; isolate; formerly spoken in Bolivia. Caquinte (Kakinte, Campa, Atiri) [cot]: Endangered, 500 speakers; Arawakan; spoken in Peru. Carabayo [cby]: No estimate available; Tikuna-Yurí; spoken in Colombia. Carapana (Carapano, Mochda, Moxdoa) [cbc]: Endangered, 527 speakers; Tucanoan; spoken in Brazil and Colombia. 199

Anna Belew et al.

Carijona (Hianácoto, Hianacoto, Jianácoto) [cbd]: Critically endangered, fewer than ten speakers; Cariban; spoken in Colombia. Cashibo (Cashibo-Cacataibo, Catcataibo, Kashibo) [cbr]: Threatened, 2,191 speakers; PanoTacanan; spoken in Peru. Cashinahua (Kashinawa, Kaxinawa, Tuxinawa) [cbs]: Threatened, 5,253 speakers; PanoTacanan; spoken in Brazil and Peru. Cavineña (Cavinena, Kavinenya) [cav]: Endangered, 600–1,200 speakers; Pano-Tacanan; spoken in Bolivia. Cayuvava (Cayubaba, Cayuwaba, Kayuvava) [cyb]: Dormant; isolate; formerly spoken in Bolivia. Cha’palaa (Cha’palaachi, Chachi, Cayapa) [cbi]: Threatened, 9,393 speakers; Barbacoan; spoken in Ecuador. Chachapoyas Quechua [quk]: Threatened, 7,000 speakers; Quechuan; spoken in Peru. Chácobo (Chákobo, Chacobo, Shinabo) [cao]: Vulnerable, approximately 1000 speakers; Pano-Tacanan; spoken in Bolivia. Chaima (Chayma, Sayma, Kumaná) [ciy, cuo]: Endangered, 112 speakers; Cariban; spoken in Brazil, Venezuela, Guyana, Suriname, and French Guiana. Chamacoco (Ebidoso, Tumraha, Chamakoko) [ceg]: Threatened, approximately 1,300 speakers; Zamucoan; spoken in Brazil and Paraguay. Chamicuro (Chamikuro, Chamicura, Chamicolo) [ccc]: Critically endangered, eight speakers; Arawakan; spoken in Peru. Chayahuita (Chayhuita, Chaui, Chawi) [cbt]: Vulnerable, 14,000 speakers; Cahuapanan; spoken in Peru. Chilean Sign Language (Lengua de Señas Chilena, Lenguaje Chileno de Signos) [csg]: Vulnerable, 21,000 speakers; sign language; spoken in Chile. Chimila (San Jorge, Chamila, Caca Weranos) [cbg]: Endangered, 900 speakers; Chibchan; spoken in Colombia. Chipaya [cap]: Threatened, 1,625 speakers; Chipaya-Uru; spoken in Bolivia. Chiquitano (Chiquito, Tarapecosi, Tao) [cax]: Threatened, 4,665 speakers; isolate; spoken in Brazil and Bolivia. Chocos Quechua (Central Yungay Quechua) [no ISO 639-3 code]: Vulnerable; Quechuan; spoken in Peru. Chorote (Choroti, Chorotí, Manjuy) [crq, crt]: Threatened, 2,057 speakers; Matacoan; spoken in Argentina and Paraguay. Cinta Larga [cin]: Endangered, 645 speakers; Tupian; spoken in Brazil. Cocama-Cocamilla (Cocama, Kokama, Kukama-Kukamiria) [cod]: Critically endangered, approximately 1,000 speakers; Tupian; spoken in Brazil, Peru, and Colombia. Cofán (Kofane, Cofan, Kofán) [con]: Threatened, 1,017 speakers; isolate; spoken in Colombia and Ecuador. Cogui (Kogui, Coghui, Kogi) [kog]: Threatened, 9,911 speakers; Chibchan; spoken in Colombia. Colombian Sign Language [csn]: Vulnerable; sign language; spoken in Colombia. Conchucos Quechua (Corongo Ancash Quechua, Sihuas Ancash Quechua, Northern Conchucos Ancash Quechua) [qwa, qws, qxn, qxo]: At risk, 515,000 speakers; Quechuan; spoken in Peru. Coreguaje (Corehuaje, Correguaje-Tama) [coe]: Threatened, 2,212 speakers; Tucanoan; spoken in Colombia. 200

The world’s endangered languages

Cuiva (Cuiba, Cuiba-Wámonae, Kuiva) [cui]: Threatened, 2,885 speakers; Guajiboan; spoken in Colombia and Venezuela. Culina (Culino, Culiña, Culinha) [cul]: Threatened, 2,854 speakers; Arawan; spoken in Brazil, Peru. Curripaco (Kurripako, Ipeka-Kurripako, Curipaco) [kpc]: Vulnerable, 12,075 speakers; Arawakan; spoken in Brazil, Colombia, and Venezuela. Cuzco Quechua (Cuzco-Collao, Cuzc, Colla) [quz]: At risk, 1,500,000 speakers; Quechuan; spoken in Peru. Dâw (Kamã, Kamã Makú, Kam) [kwa]: Endangered, 120 speakers; Nadahup; spoken in Brazil. Demushbo (Dëmushbo) [no ISO 639-3 code]: Dormant; Pano-Tacanan; formerly spoken in Bolivia. Dení (Deni, Den, Dani) [dny]: Threatened, approximately 1,200 speakers; Arawan; spoken in Brazil. Desano (Desana, Desano, Desâna) [des]: Endangered, approximately 450 speakers; Tucanoan; spoken in Brazil and Colombia. Diahói (Diahui, Jahoi, Diarrui) [no ISO 639-3 code]: Critically endangered, four speakers; Tupian; spoken in Brazil. Dyapá (Katukína-Kanamarí, Bendiapa, Txunhuã-Djapá) [knm]: Threatened, approximately 1,650 speakers; Harakmbut-Katukinan; spoken in Brazil. Ecuadorian Sign Language [ecs]: At risk, 231,000 speakers; sign language; spoken in Ecuador. Emberá-Catío (Emperã, Eberã Bed’ea, Eperã Pedea) [cto]: Vulnerable, 88,747 speakers; Chocoan; spoken in Colombia, Panama. Emerillon (Emerilon, Melejo, Mereo) [eme]: Endangered, 400 speakers; Tupian; spoken in French Guiana. Enawené-Nawé (Eneuene-Mare, Salumã) [unk]: Endangered, 445 speakers; Arawakan; spoken in Brazil. Enenlhet (Maskoy, Toba-Maskoy) [tmf]: Threatened, 1,275 speakers; Mascoyan; spoken in Paraguay. Enlhet (Lengua, Enxet, Vowak) [enl]: Threatened, 6,439 speakers; Mascoyan; spoken in Paraguay. Enxet (Lengua Sur) [enx]: Threatened, 3,842 speakers; Mascoyan; spoken in Paraguay. Epena (Eperara, Southern Embera, Emberá-Saija) [sja]: Vulnerable, 88,800 speakers; Chocoan; spoken in Colombia, Ecuador, and Panama. Ese’jja (Huarayo, Tiatinagua, Chama) [ese]: Endangered, 518 speakers; Pano-Tacanan; spoken in Peru and Bolivia. Gavião (Gavião do Jiparaná, Gavião do Rondônia, Digüt) [gvo]: Endangered, 523 speakers; Tupian; spoken in Brazil. Gavião de Parã (Pukobyé, Piokob, Bocobu) [gvp]: Endangered, 476 speakers; Jean; spoken in Brazil. Guajá (Awá, Avá, Guaxare) [gvj]: Endangered, 283 speakers; Tupian; spoken in Brazil. Guajajára (Guajajára, Guazazzara, Tenetehara) [gub]: Vulnerable, 14,000 speakers; Tupian; spoken in Brazil. Guajibo (Guahibo, Guaybo, Sikuani) [guh]: Vulnerable, 35,006 speakers; Guajiboan; spoken in Colombia and Venezuela. Guajiro (Goajiro, Wayuu, Wahiro) [guc]: Vulnerable, approximately 312,730 speakers; Arawakan; spoken in Colombia and Venezuela. 201

Anna Belew et al.

Guamaca (Malayo, Marocasero, Maracasero) [mbp]: Threatened, 1,922 speakers; Chibchan; spoken in Colombia. Guambiano (Guambiano, Guambia, Moguex) [gum]: Vulnerable, 23,462 speakers; Barbacoan; spoken in Colombia. Guana (Etelena, Chané, Kinikinao) [gqn]: Dormant; Arawakan; formerly spoken in Brazil. Guana (Paraguay) (Guaná, Kaskiha, Kaskihá) [gva]: Critically endangered, 19 speakers; Mascoyan; spoken in Paraguay. Guarayu (Guarayú, Chapacura, Guarayo) [gyr]: Threatened, 8,433 speakers; Tupian; spoken in Bolivia. Guarequena (Warekena, Arequena, Guarekena) [gae]: Endangered, 180 speakers; Arawakan; spoken in Brazil and Venezuela. Guató (Guato) [gta]: Critically endangered, five speakers; isolate; spoken in Brazil. Guayabero (Cunimía, Jiw, Mítus) [guo]: Threatened, 1,118 speakers; Guajiboan; spoken in Colombia. Gününa-Küne (Gennaken, Northern Tehuelche) [pue]: Dormant; isolate; formerly spoken in Argentina. Harakmbut (Harákmbut, Harakmbet, Hate) [hug, amr]: Threatened, 1,206 speakers; Harakmbut-Katukinan; spoken in Brazil and Peru. Himarimã (Hi-Merimã, Hi-Marimã, Maimã) [hir]: No estimate available; unclassified (possibly Arawan?); spoken in Brazil. Hixkaryana (Hishkariana, Hixkaryána, Hishcariana) [hix]: Endangered, 600 speakers; Cariban; spoken in Brazil. Huallaga Quechua (Huánuco Alto Huallaga, Huallaga Huánuco Quechua) [qub]: Vulnerable, 40,000 speakers; Quechuan; spoken in Peru. Huambisa (Jíbaro, Xívaro, Huambiza) [hub]: Threatened, 8,000 speakers; Jivaroan; spoken in Peru. Huaylas Quechua (Huaylas Ancash Quechua, Huaraz Quechua) [qwh]: Vulnerable, 336,000 speakers; Quechuan; spoken in Peru. Huilliche (Williche, Veliche, Huiliche) [huh]: Severely endangered, a few speakers; Araucanian; spoken in Argentina and Chile. Hupda (Hup, Hupdë, Hupdá Makú) [jup]: Threatened, 1,500 speakers; Nadahup; spoken in Brazil and Colombia. Ica (Arhuaco, Aruaco, Bintuk) [arh]: Vulnerable, 14,799 speakers; Chibchan; spoken in Colombia. Ignaciano (Moxo, Mojo, Mojeño) [ign]: Threatened, 1,080 speakers; Arawakan; spoken in Bolivia. Ikpéng (Txikāo, Ikpeng, Txicão) [txi]: Endangered, 342 speakers; Cariban; spoken in Brazil. Iñapari (Inapori, Inapari, Inamarí) [inp]: Critically endangered, four speakers; Arawakan; spoken in Peru. Inga (Highland Inga, Colombia Quechua, Ingo) [inb, inj]: Threatened, 8,000 speakers; Quechuan; spoken in Colombia. Iquito (Iquita, Ikito, Amacacore) [iqu]: Critically endangered, 12–15 speakers; Zaparoan; spoken in Peru. Irantxe (Iranshe, Iranxe, Irntxe) [irn]: Severely endangered, approximately 90 speakers; isolate; spoken in Brazil. Isconahua (Iskonawa, Iscobakebo, Iscobaquebu) [isc]: Critically endangered, five speakers; Pano-Tacanan; spoken in Peru. 202

The world’s endangered languages

Isolado do Tanaru (Isolado do Buraco) [no ISO 639-3 code]: Critically endangered, one speaker; unclassified; spoken in Brazil. Itene (Iteneo, Iténez, Moré) [ite]: Severely endangered, 44 speakers; Chapacuran; spoken in Bolivia. Itonama (Saramo, Machoto) [ito]: Critically endangered, one speaker; isolate; spoken in Bolivia. Jabutí (Yabutí, Jabotí, Djeoromitxi) [jbt]: Severely endangered, approximately 50 speakers; Jabutian; spoken in Brazil. Jamamadí (Yamamadi, Jamamadi, Yamadi) [jaa]: Endangered, 980 speakers; Arawan; spoken in Brazil. Japréria (Japrería, Japreria, Yucpa-Yaprería) [jru]: Severely endangered, 95 speakers; Cariban; spoken in Venezuela. Jaqaru (Haqearu, Haqaru, Haq’aru) [jqr]: Endangered, 725 speakers; Aymaran; spoken in Peru. Jauja-Huanca Quechua (Huaylla Wanca Quechua, Southern Huancayo Quechua, Huanca Huaylla Quechua) [qvw]: At risk, 250,000 speakers; Quechuan; spoken in Peru. Jotí (Joti, Hodï, Waruwaru) [yau]: Endangered, 767 speakers; isolate; spoken in Venezuela. Júma (Arara-Kawahib, Kagwahiv, Kavahiva) [jua]: Critically endangered, five speakers; Tupian; spoken in Brazil. Juruna (Jurúna, Yurúna, Iuruna) [jur]: Endangered, approximately 300 speakers; Tupian; spoken in Brazil. Kaapor (Urubú-Kaapor, Urubú, Kaaporté) [urb]: Endangered, approximately 800 speakers; Tupian; spoken in Brazil. Kadiwéu (Mbaya-Guaikuru, Caduveo, Ediu-Adig) [kbc]: Threatened, 1,600 speakers; Guaicuruan; spoken in Brazil. Kaingang (Kaingan, Bugre, Coroado) [kgp]: Vulnerable, 18,500 speakers; Jean; spoken in Brazil. Kaiwá (Kaiwá, Caiwa, Caingua) [kgk, pta]: Vulnerable, approximately 26,364 speakers; Tupian; spoken in Brazil, Argentina. Kaixana (Caixana) [no ISO 639-3 code]: Dormant; Arawakan; formerly spoken in Brazil. Kakua (Cacua, Macu de Cubeo, Macu de Guanano) [cbv]: Endangered, 220 speakers; KakuaNukak; spoken in Colombia. Kallawaya (Callahuaya, Machaj-Juyay, Pohena) [caw]: Severely endangered; mixed language; spoken in Bolivia. Kamayurá (Kamaiurá, Camaiura, Kamayirá) [kay]: Endangered, 400 speakers; Tupian; spoken in Brazil. Kapixaná (Capixana, Kanoé, Kapishaná) [kxo]: Critically endangered, three speakers; isolate; spoken in Brazil. Kapong (Kapon, Ka’pong) [ake, pbc]: Vulnerable, fewer than 11,550 speakers; Cariban; spoken in Brazil, Venezuela, and Guyana. Karajá (Karajá-Xambioá, Karaho, Caraja) [kpj]: Threatened, 3,060 speakers; isolate; spoken in Brazil. Kari’nja (Kari’nya, Carib, Galibí) [car]: Threatened, approximately 8,600 speakers; Cariban; spoken in Brazil, Venezuela, Guyana, Suriname, and French Guiana. Karipuna (Karipúna do Guaporé, Caripuna, Jau-Navo) [kuq]: Severely endangered, ten speakers; Tupian; spoken in Brazil. Karitiana (Karitiâna, Caritiana, Karitiána) [ktn]: Endangered, 320 speakers; Tupian; spoken in Brazil. Karo (Arara, Urukú, Káro) [arr]: Endangered, 200 speakers; Tupian; spoken in Brazil. 203

Anna Belew et al.

Katawixi (Catauisi, Catauichí, Catawixi) [xat]: Dormant; Harakmbut-Katukinan; formerly spoken in Brazil. Katukina (Catuquina, Pidádiapá, Catuquino) [kav]: Vulnerable, 550 speakers; HarakmbutKatukinan; spoken in Brazil. Katukina do Acre (Panoan Katukína, Panoan Catuquina, Waninnawa) [knt]: Endangered, 318 speakers; Pano-Tacanan; spoken in Brazil. Katxúyana (Kashuyana-Warikyana, Kaxuiana, Kaxuiâna) [kbb]: Severely endangered, fewer than 150 speakers; Cariban; spoken in Brazil. Kawésqar (Kawesqar, Kaweskar, Qawasqar) [alc]: Critically endangered, seven speakers; Kawesqaran; spoken in Chile. Kaxararí (Cacharari, Kashararí, Kaxariri) [ktx]: Threatened, approximately 300 speakers; PanoTacanan; spoken in Brazil. Kayabí (Kajabí, Caiabi, Parua) [kyz]: Threatened, 1,000 speakers; Tupian; spoken in Brazil. Korubo (Caceteiros, Chankuëshbo) [xor]: Severely endangered, 26 speakers; Pano-Tacanan; spoken in Brazil. Krahô [xra]: Threatened, 2,184 speakers; Jean; spoken in Brazil. Kreen-Akarore (Kren Akarore, Panará, Ipewí) [kre]: Endangered, approximately 250 speakers; Jean; spoken in Brazil. Krenak (Botocudo, Aimoré, Nakrehé) [kqq]: Severely endangered, approximately ten speakers; Krenakan; spoken in Brazil. Kreye (Crengez, Krenjé, Crenge) [xre]: Endangered, 496 speakers; Jean; spoken in Brazil. Krikatí (Krinkati, Krikati-Timbira, Canela) [xri]: Endangered, 682 speakers; Jean; spoken in Brazil. Kubeo (Cubeo, Hahänana, Hehenawa) [cub]: Threatened, 7,900 speakers; Tucanoan; spoken in Brazil and Colombia. Kuikuro (Língua Karib do Alto Xingu, Kuikúro-Kalapálo, Kuikuru) [kui, mzo]: Threatened, 1,106 speakers; Cariban; spoken in Brazil. Kulina (Culina, Culino, Kulino) [xpk]: Severely endangered, 30 speakers; Pano-Tacanan; spoken in Brazil. Kuruaya (Kuruáya, Caravare, Curuaia) [kyr]: Dormant; Tupian; formerly spoken in Brazil. Kwaza (Koiyá, Koaiá, Quaiá) [xwa]: Severely endangered, 25 speakers; isolate; spoken in Brazil. Lambayeque Quechua (Cañaris-Incahuasi, Lambayeque Quechua, Ferreñafe) [quf]: Vulnerable, 20,000 speakers; Quechuan; spoken in Peru. Lapachu (Apolista, Lapacho, Aguachile) [no ISO 639-3 code]: Dormant; Arawakan; formerly spoken in Bolivia. Laraos Quechua [no ISO 639-3 code]: No estimate available; Quechuan; spoken in Peru. Leco (Leko, Lapalapa, Chuncho Aleniano) [lec]: Dormant; isolate; formerly spoken in Bolivia. Lincha Quechua (Lincha, Central Yungay Quechua) [no ISO 639-3 code]: No estimate available; Quechuan; spoken in Peru. Loja Quichua (Saraguro, Ecuadorian Quichua, Loja Highland Quichua) [qvj]: Severely endangered, 30,500 speakers; Quechuan; spoken in Ecuador. Lokono (Arawak, Arawák, Aruak) [arw]: Severely endangered, fewer than 2,000 speakers; Arawakan; spoken in Venezuela, Guyana, Suriname, and French Guiana. Lowland Quichua (Quijo, Napo, Napo Quichua) [qvo]: Vulnerable, approximately 17,855? speakers; Quechuan; spoken in Peru, Colombia, and Ecuador. Macaguán (Macaguane, Agualinda Guahibo, Hitnü) [mbn]: Endangered, 441 speakers; Guajiboan; spoken in Colombia. 204

The world’s endangered languages

Machiguenga (Machaguenga, Amachengue, Macheyenga) [mcb]: Threatened, 5,000 speakers; Arawakan; spoken in Peru. Macuna (Erulia, Yeba, Buhagana) [myy]: Threatened, 1,041 speakers; Tucanoan; spoken in Brazil and Colombia. Maká (Macca, Towothli, Enimaga) [mca]: Threatened, 1,042 speakers; Matacoan; spoken in Paraguay. Mako (Maku, Macu, Maco) [wpc]: Vulnerable, approximately 1,078 speakers; Saliban; spoken in Venezuela. Makurap (Makuráp, Makurápi, Macuráp) [mpu]: Severely endangered, 45 speakers; Tupian; spoken in Brazil. Makushi (Makushí, Macuxi, Teweya) [mbc]: Threatened, fewer than 7,830 speakers; Cariban; spoken in Brazil, Venezuela, and Guyana. Mamaindé (Northern Nambikwara, Mamainde, Northern Nambikuára) [wmd]: Endangered, 250 speakers; Nambikwaran; spoken in Brazil. Mandahuaca (Mitua, Mandauaca, Mandawaka) [mht]: Dormant; Arawakan; formerly spoken in Brazil and Venezuela. Mapoyo (Mapoyo-Yavarana, Mapoyo, Mapayo) [mcg]: Critically endangered, approximately three speakers; Cariban; spoken in Venezuela. Mapudungu (Mapudungun, Araucano, Maputongo) [arn]: Threatened, 258,413 speakers; Araucanian; spoken in Argentina and Chile. Marubo (Marobo, Marúbo, Maruba) [mzr]: Threatened, 1,043 speakers; Pano-Tacanan; spoken in Brazil. Mashco Piro (Mashco, Mashco Piro) [cuj]: Endangered, 200–600 speakers; Arawakan; spoken in Peru. Matís (Matis, Mayoruna-Matsés) [mpq]: Endangered, at least 280 speakers; Pano-Tacanan; spoken in Brazil. Matsés (Mayoruna, Matse, Matís) [mcf, pig]: Threatened, 2,200 speakers; Pano-Tacanan; spoken in Brazil and Peru. Mawayana (Mahuayana, Madipian, Maiopitian) [mzx, mpw]: Critically endangered, two speakers; Arawakan; spoken in Brazil, Guyana, and Suriname. Mawé (Sateré-Mawé, Maué, Mabue) [mav]: Threatened, 6,219 speakers; Tupian; spoken in Brazil. Maxakalí (Tikmũ’ũn Yĩy’ax) [mbl]: Threatened, 1,271 speakers; Maxakalian; spoken in Brazil. Maxineri (Manitenere, Mashineri, Machinere) [mpd]: Endangered, 950 speakers; Arawakan; spoken in Brazil and Bolivia. Mbyá (Mbua, Mbiá, Bugre) [gun]: Vulnerable, 19,924 speakers; Tupian; spoken in Brazil, Argentina, Paraguay, and Uruguay. Mebengokre (Kayapó, Cayapo, Kokraimoro) [txu]: Threatened, 7,266 speakers; Jean; spoken in Brazil. Mehináku (Mehinacú, Minaco, Mehinacu) [mmh]: Endangered, 227 speakers; Arawakan; spoken in Brazil. Mekéns (Sakirabiát, Sakirabiá, Sakirabiáp) [skf]: Critically endangered, 22 speakers; Tupian; spoken in Brazil. Minica Huitoto (Minica, Mɨnɨca, Mɨnɨka) [hto]: Threatened, 7,343 speakers; Witotoan; spoken in Peru and Colombia. Mirití (Miriti-Tapuyo, Neenoá) [mmv]: Dormant; Tucanoan; formerly spoken in Brazil. 205

Anna Belew et al.

Mocoví (Mocovi, Mocobí, Mbocobí) [moc]: Threatened, 2,780 speakers; Guaicuruan; spoken in Argentina. Mondé (Sanamaiká, Sanamaykã, Sanamaica) [mnd]: Dormant; Tupian; formerly spoken in Brazil. Mosetén (Chimané, Chimane, Chiman) [cas]: Vulnerable, at least 4,800 speakers; isolate; spoken in Bolivia. Movima (Mobima, Móvima, Móbima) [mzp]: Threatened, 1,173 speakers; isolate; spoken in Bolivia. Muinane (Muinana, Muinani, Muename) [bmr]: Severely endangered, approximately 78 speakers; Boran; spoken in Colombia. Mundurukú (Mundurucu, Monjoroku, Weidyenye) [myu]: Threatened, 8,000 speakers; Tupian; spoken in Brazil. Munichi (Muniche, Paranapura, Otonabe) [myr]: Dormant; isolate; formerly spoken in Peru. Murui Huitoto (Murui, Murui Witoto, Huitoto Murui) [huu]: Threatened, approximately 3,000 speakers; Witotoan; spoken in Peru and Colombia. Nadëb (Nadobo, Nadeb Macu) [mbj]: Endangered, 370 speakers; Nadehup; spoken in Brazil. Nambikwara (Nambiquara, Southern Nambikwara) [nab]: Endangered, 721 speakers; Nambikwaran; spoken in Brazil. Ñandeva (Chiripá, Tsiripá, Txiripá) [nhd]: Endangered, 19,308 speakers; Tupian; spoken in Brazil, Argentina, and Paraguay. Nanti (Cogapacori, Kogapakori) [cox]: Vulnerable, 450 speakers; Arawakan; spoken in Peru. Ndyuka-Trio Pidgin [njt]: Dormant; pidgin or creole; formerly spoken in Suriname. Nheengatú (ñeengatú, Nhengatu, Yeral) [yrl]: Endangered, 19,060 speakers; Tupian; spoken in Brazil, Colombia, and Venezuela. Ninam (Yanam, Nimam, Xiriâna) [shb]: Vulnerable, 560 speakers; Yanomaman; spoken in Brazil and Venezuela. Nipode (Huitoto Muinane, Nïpode, Nüpode Huitoto) [hux]: Severely endangered, 50 speakers; Witotoan; spoken in Brazil, Peru, and Colombia. Nivaclé (Chulupi, Churupi, Nivaklé) [cag]: Threatened, 8,800 speakers; Matacoan; spoken in Argentina and Paraguay. Nomatsiguenga (Machinguenga, Nomatsigenga) [not]: Threatened, 5,531 speakers; Arawakan; spoken in Peru. Nonuya (Nononota) [noj]: Dormant; Witotoan; formerly spoken in Peru and Colombia. Northern Bolivian Quechua (Boliviano, North la Paz Quechua) [qul]: At risk, 100,000 speakers; Quechuan; spoken in Bolivia. Northern Emberá (Emperã, Eberã Bed’ea, Eperã Pedea) [emp]: Vulnerable, approximately 88,799 speakers; Chocoan; spoken in Colombia and Panama. Northern Pastaza Quichua (Canelo, Loreto, Puyo Pongo) [qvz]: Threatened, 10,000 speakers; Quechuan; spoken in Peru and Ecuador. Nukak (Nukak Makú, Maczsa, Guaviare) [mbr]: Endangered, 390 speakers; Kakua-Nukak; spoken in Colombia. Nukini (Nuquini, Nucuini, Inukuini) [nuc]: Dormant; Pano-Tacanan; formerly spoken in Brazil. Ocaina (Okaina) [oca]: Critically endangered, approximately 50 speakers; Witotoan; spoken in Peru and Colombia. Ofayé (Guachi, Opaie, Opaié-Shavante) [opy]: Severely endangered, 12 speakers; isolate; spoken in Brazil. Omagua (Omagua-Yete, Ariana, Pariana) [omg]: Critically endangered, fewer than ten speakers; Tupian; spoken in Brazil and Peru. 206

The world’s endangered languages

Ona (Selknam, Selk’nam, Shelknam) [ona]: Dormant; Chonan; formerly spoken in Argentina. Orejón (Maijuna, Maihuna, Coto) [ore]: Critically endangered, 90–100 speakers; Tucanoan; spoken in Peru. Orowari (Oro Win, Pakaás-novos, Pacasnovas) [orw]: Critically endangered, five speakers; Chapacuran; spoken in Brazil. Pacaraos Quechua (Quechua, Pacaraos) [qvp]: Severely endangered, approximately 35 speakers; Quechuan; spoken in Peru. Paez (Páez, Nasa Yuwe, Pez) [pbb]: Vulnerable, 60,000 speakers; isolate; spoken in Colombia. Paiter (Suruí do Jiparaná, Suruí de Rondônia, Paiter) [sru]: Endangered, 920 speakers; Tupian; spoken in Brazil. Palenquero (Lengua, Palenque) [pln]: Endangered, 500 speakers; pidgin or creole; spoken in Colombia. Palikur (Palicur, Palikúr, Palikr) [plu]: Threatened, approximately 2,000 speakers; Arawakan; spoken in Brazil and French Guiana. Panare (Eñepa, E’ñepa, Panari) [pbh]: Vulnerable, 3,100 speakers; Cariban; spoken in Venezuela. Parakanã (Parakanân, Parocana, Awaeté) [pak]: Threatened, 1,286 speakers; Tupian; spoken in Brazil. Paraujano (Parauhano, Parahujano, Parujano) [pbg]: Critically endangered, one speaker; Arawakan; spoken in Venezuela. Paresí (Parecís, Paressí, Ariti) [pab]: Threatened, 2,000 speakers; Arawakan; spoken in Brazil. Paumarí (Paumari, Pammarí, Curucuru) [pad]: Endangered, 200 speakers; Arawan; spoken in Brazil. Paunaka (Paunaca, Pauna-Paikone, Pauna-Paicone) [pnk]: Critically endangered; Arawakan; spoken in Bolivia. Pauserna (Guarasu, Paucerne, Guarayu-Ta) [psm]: Critically endangered, three to four speakers; Tupian; spoken in Bolivia. Pemón (Pemong, Pemon, Taulipang) [aoc]: Vulnerable, 23,983–24,083 speakers; Cariban; spoken in Brazil, Venezuela, and Guyana. Peruvian Sign Language (Lenguaje de Señas Peruanas, Lenguaje de Señas Peruano, Lengua de Signos Peruana) [prl]: Threatened; sign language; spoken in Peru. Piapoco (Dzaze, Piapoko, Tsae) [pio]: Threatened, 6,041 speakers; Arawakan; spoken in Colombia and Venezuela. Piaroa (Kuakua, Guagua, Quaqua) [pid]: Vulnerable, 13,773 speakers; Saliban; spoken in Colombia and Venezuela. Pilagá (Pilaga, Pilaca) [plg]: Threatened, 6,000 speakers; Guaicuruan; spoken in Argentina. Pirahã (Pirahán, Piraha, Múra-Pirahã) [myp]: Vulnerable, 389 speakers; isolate; spoken in Brazil. Piratapuyo (Uaíana, Waíana, Uaicana) [pir]: Threatened, 1,330 speakers; Tucanoan; spoken in Brazil and Colombia. Piro (Yine, Chontaquiro, Simirinch) [pib]: Threatened, 3,490 speakers; Arawakan; spoken in Peru. Playero (Rio Arauca Guahibo, Yamarero, Pepojivi) [gob]: Endangered, fewer than 400 speakers; Guajiboan; spoken in Colombia and Venezuela. Poyanawa (Poyanáwa, Poianáua, Puinahua) [pyn]: Dormant; Pano-Tacanan; formerly spoken in Brazil. Providencia Sign Language (Providence Island Sign Language) [prz]: Threatened, approximately 1,500 speakers; sign language; spoken in Colombia. 207

Anna Belew et al.

Puinave (Guipuinave, Guaipunavos, Puinabe) [pui]: Threatened, 7,154 speakers; isolate; spoken in Colombia and Venezuela. Puruborá (Puruba, Aurã, Pumbora) [pur]: Critically endangered, two speakers; Tupian; spoken in Brazil. Rapa Nui (Easter Island, Pascuense, Rapanui) [rap]: Threatened, approximately 2,000 speakers; Austronesian; spoken in Chile. Resígaro (Resigaro, Resígero) [rgr]: Critically endangered, one speaker; Arawakan; spoken in Peru. Reyesano (Maropa, San Borjano) [rey]: Critically endangered, 10–15 speakers; Pano-Tacanan; spoken in Bolivia. Rikbaktsá (Rikbaktsa, Erikbatsa, Aripaktsá) [rkb]: Threatened, 1,085 speakers; isolate; spoken in Brazil. Sabané (Sabane, Sabanês, Sabones) [sae]: Critically endangered, three speakers; Nambikwaran; spoken in Brazil. Sáliva (Saliba, Sáliba (in Colombia)) [slc]: Critically endangered, approximately 45 speakers; Saliban; spoken in Colombia and Venezuela. San Martín Quechua (Canelo, Lamaño, Lamanyo) [qvs]: Threatened, 15,000 speakers; Quechuan; spoken in Peru. Sanapaná (Quilyacmoc, Kelya’mok, Saapa’ang) [spn]: Endangered, 984 speakers; Mascoyan; spoken in Paraguay. Santiago del Estero Quechua (Argentino, Tucumano) [qus]: Vulnerable, 60,000–80,000 speakers; Quechuan; spoken in Argentina. Sanumá (Sanuma, Sanum, Sanɨma) [xsu]: Threatened, 3,497 speakers; Yanomaman; spoken in Brazil and Venezuela. Sapé (Caliana, Kaliana, Kariana) [spc]: Dormant; isolate; formerly spoken in Venezuela. Saraveca (Sarave, Saraveka, Zarabe) [sar]: Dormant; Arawakan; formerly spoken in Bolivia. Shanenawa (Katukina Shanenawa) [swo]: Critically endangered, 23 speakers; Pano-Tacanan; spoken in Brazil. Shikuyana (Sikiana, Chikena, Chiquena) [sik]: Endangered, fewer than 162 speakers; Cariban; spoken in Suriname. Shipibo (Shipibo-Conibo, Shipibo-Konibo) [shp]: Vulnerable, 22,517 speakers; Pano-Tacanan; spoken in Peru. Shiwilu (Jebero, Chébero, Jébero) [jeb]: Critically endangered, fewer than 30 speakers; Cahuapanan; spoken in Peru. Shuar [jiv]: Vulnerable, 42,840 speakers; Jivaroan; spoken in Ecuador. Simba Guaraní (Simba Guaraní, Western Bolivian Guaraní, Chiriguano) [gnw]: Threatened; Tupian; spoken in Argentina and Bolivia. Siona-Secoya (Piohe, Pioje) [snn, sey]: Threatened, 1,174 speakers; Tucanoan; spoken in Colombia and Ecuador. Siriano (Sirianó, Siriana, Desano-Siriano) [sri]: Endangered, 820 speakers; Tucanoan; spoken in Brazil and Colombia. Sirionó (Mbia Chee, Mbya, Siriono) [srq]: Critically endangered, fewer than 100 speakers; Tupian; spoken in Bolivia. Southern Emberá (Emberá-Chamí) [cmi]: Vulnerable, approximately 88,799 speakers; Chocoan; spoken in Colombia and Panama. Suruí do Pará (Suruí, Suruí do Tocantins, Akewere) [mdz]: Endangered, 264 speakers; Tupian; spoken in Brazil. Suyá (Suya, Kisêdjê, Suiá) [suy]: Endangered, 409 speakers; Jean; spoken in Brazil. 208

The world’s endangered languages

Tacana (Takana, Tupamasa) [tna]: Threatened, 1,153 speakers; Pano-Tacanan; spoken in Bolivia. Taiwano (Taiwano, Taibano, Taiwaeno) [no ISO 639-3 code]: Severely endangered, 22 speakers; Tucanoan; spoken in Colombia. Tanimuca-Retuarã (Uairã, Tanimuca-Letuama, Retuama) [tnc]: Threatened, fewer than 1,952 speakers; Tucanoan; spoken in Colombia. Tapieté (Chawuncu, Ava, Chané) [tpj]: Endangered, 207 speakers; Tupian; spoken in Argentina, Bolivia, and Paraguay. Tapirapé (Tapi’irape) [taf]: Endangered, 564 speakers; Tupian; spoken in Brazil. Tariana (Tariano, Tarîna, Taliáseri) [tae]: Critically endangered, 90 speakers; Arawakan; spoken in Brazil and Colombia. Taruma (Aroaqui, Taruamá) [tdm]: Critically endangered, a few speakers; isolate; spoken in Brazil and Guyana. Tatuyo (Oa, Tatuyo, Pamoa) [tav]: Endangered, 331 speakers; Tucanoan; spoken in Colombia. Taushiro (Pinche, Pinchi) [trr]: Critically endangered, one speaker; isolate; spoken in Peru. Tehuelche (Tehuelche, Aoniken, Aonek’enk) [teh]: Critically endangered, one to two speakers; Chonan; spoken in Argentina. Tembé (Tenetehára, Tenetehára, Timbé) [tqb]: Severely endangered, 60 speakers; Tupian; spoken in Brazil. Tenharín (Tenharim, Tenharem, Tenharin) [pah]: Endangered, 350 speakers; Tupian; spoken in Brazil. Tequiraca (Tekiraka, Aushiri, Auishiri) [ash]: Dormant; isolate; formerly spoken in Peru. Terena (Tereno, Etelena, Terêna) [ter]: Vulnerable, 8,439 speakers; Arawakan; spoken in Brazil. Teteté (Eteteguaje) [teb]: Dormant; Tucanoan; formerly spoken in Ecuador. Tikuna (Ticuna, Tucuna, Tukuna) [tca]: Vulnerable, 50,102 speakers; Tikuna-Yurí; spoken in Brazil, Peru, and Colombia. Tinigua (Tiniguan, Tiniwa, Tiniguas) [tit]: Critically endangered, one speaker; Tiniguan; spoken in Colombia. Toba (Qom, Toba Qom, Toba Sur) [tob]: Vulnerable, 31,593 speakers; Guaicuruan; spoken in Argentina and Paraguay. Torá (Tora, Toraz) [trz]: Dormant; Chapacuran; formerly spoken in Brazil. Totoró (Totoro, Polindara) [ttk]: Dormant; Barbacoan; formerly spoken in Colombia. Trinitario (Trimitarios, Moxo, Moxos) [trn]: Threatened, 3,140 speakers; Arawakan; spoken in Bolivia. Trió (Tiríyo, Tiriyó, Trio) [tri]: Threatened, 2,456 speakers; Cariban; spoken in Brazil and Suriname. Trumai (Trumaí, Tramalhy) [tpy]: Critically endangered, 50 speakers; isolate; spoken in Brazil. Tsafiki (Colorado, Tsachila, Colima) [cof]: Threatened, 1,872 speakers; Barbacoan; spoken in Ecuador. Tukano (Tucano, Tukána, Tukana) [tuo]: Vulnerable, 13,996 speakers; Tucanoan; spoken in Brazil and Colombia. Tunayana (Katuena) [no ISO 639-3 code]: Critically endangered, fewer than ten speakers; Cariban; spoken in Guyana and Suriname. Tunebo (Uwa, U’wa, U’w Cuwa) [tbn, tnb, tnd, tuf]: Threatened, 7,231 speakers; Chibchan; spoken in Colombia. Tuparí (Tupari) [tpr]: Endangered, 276 speakers; Tupian; spoken in Brazil. Tuyuka (Tuyuca, Dochkafuara, Tejuca) [tue]: Threatened, approximately 1,442 speakers; Tucanoan; spoken in Brazil and Colombia. 209

Anna Belew et al.

Umutina (Umotína, Umutina, Barbados) [umo]: Awakening; Bororoan; spoken in Brazil. Urarina (Simacu, Shimacu, Itukale) [ura]: Threatened, 2,000 speakers; isolate; spoken in Peru. Uru (Uro, Morato, Uru-Murato) [ure]: Dormant; Chipaya-Uru; formerly spoken in Bolivia. Urubú-Kaapor Sign Language (Urubú Sign Language) [uks]: Critically endangered, seven speakers; sign language; spoken in Brazil. Uruewawau (Uru-Eu-Wau-Wau, Uru-Eu-Uau-Uau, Eru-Eu-Wau-Wau) [urz]: Endangered, 183 speakers; Tupian; spoken in Brazil. Uruguayan Sign Language (Lengua de Señas Uruguaya, LSU) [ugy]: Threatened, 7,000 speakers; sign language; spoken in Uruguay. Venezuelan Sign Language (Lengua de Señas Venezolana) [vsl]: No estimate available; sign language; spoken in Venezuela. Vilela [vil]: Dormant; Lule-Vilela; formerly spoken in Argentina. Waimirí-Atroarí (Yawaperí, Atruahí, Atroarí) [atr]: Threatened, 1,120 speakers; Cariban; spoken in Brazil. Waiwai (Uaiuai, Wai Wai, Katawiana) [waw]: Threatened, 3,154 speakers; Cariban; spoken in Brazil and Guyana. Wanano (Uanana, Ananas, Wanana) [gvc]: Threatened, approximately 2,000 speakers; Tucanoan; spoken in Brazil and Colombia. Waorani (Auishiri, Auca, Sabela) [auc]: Threatened, 1,616 speakers; isolate; spoken in Ecuador. Wapixana (Wapishana, Aruma, Upichana) [wap]: Vulnerable, fewer than 10,900 speakers; Arawakan; spoken in Brazil and Guyana. Warao (Guarao, Guasay, Warrau) [wba]: Vulnerable, approximately 32,400 speakers; isolate; spoken in Venezuela, Guyana, and Suriname. Wari’ (Yaru, Jaru, Ituarupa) [pav]: Threatened, 2,381 speakers; Chapacuran; spoken in Brazil. Waurá (Waura, Uaurá, Aura) [wau]: Endangered, 410 speakers; Arawakan; spoken in Brazil. Wayampí (Wayampi, Oiampí, Oyapí) [oym]: Threatened, 1,655 speakers; Tupian; spoken in Brazil and French Guiana. Wayana (Northern Wayana, Oyana, Urukuyana) [way]: Threatened, 1,738 speakers; Cariban; spoken in Brazil, Suriname, and French Guiana. Wayoró (Ayurú, Wayurú, Ajurú) [wyr]: Critically endangered, five speakers; Tupian; spoken in Brazil. Wichí (Mataco, Wichí Lhamtés Güisnay, Mataco Güisnay) [mzh]: Vulnerable, 30,560 speakers; Matacoan; spoken in Argentina and Bolivia. Wounaan (Waunana, Noanamá, Huaunana, Woun Meu) [noa]: Threatened, approximately 13,000 speakers; Chocoan; spoken in Colombia and Panama. Xavante (Chavante, Xavánte, A’uwe Uptabi) [xav]: Vulnerable, 13,303 speakers; Jean; spoken in Brazil. Xerente (Xerénte, Sherenté, Xerenti) [xer]: Threatened, 2,569 speakers; Jean; spoken in Brazil. Xetá (Aré, Sheta, Cheta) [xet]: Critically endangered, two speakers; Tupian; spoken in Brazil. Xipaya (Shipaja, Xipaia, Shipaya) [xiy]: Dormant; Tupian; formerly spoken in Brazil. Xokleng (Aweikoma, Botocudo, Bugre) [xok]: Endangered, 790 speakers; Jean; spoken in Brazil. Yagan (Yahgan, Yaghan, Yamana) [yag]: Critically endangered, one speaker; isolate; spoken in Chile. Yagua (Nijyamïï Nikyejaada, Yahua, Llagua) [yad]: Threatened, 4,297 speakers; Yaguan; spoken in Peru and Colombia. 210

The world’s endangered languages

Yahuna (Yaúna, Yayuna, Yauna) [ynu]: Dormant; Tucanoan; formerly spoken in Colombia. Yaminawa (Yaminahua, Yawanawa, Morunahua) [yaa]: Dormant; Pano-Tacanan; formerly spoken in Brazil, Peru, and Bolivia. Yanomam (Yanomami, Waiká, Yanoam) [wca]: Threatened, at least 4,000 speakers; Yanomaman; spoken in Brazil. Yanomamö (Guaicá, Shamateri, Guaharibo) [guu]: Vulnerable, 18,234 speakers; Yanomaman; spoken in Brazil and Venezuela. Yaru Quechua (Tarma, Junín) [qvn]: Vulnerable, 60,000 speakers; Quechuan; spoken in Peru. Yaruro (Jaruro, Pumé, Llaruro) [yae]: Threatened, 7,400 speakers; isolate; spoken in Venezuela. Yaté (Fulnio, Fulino, Fulnió) [fun]: Threatened, 1,998 speakers; isolate; spoken in Brazil. Yauyos Quechua [qux]: Critically endangered, 6,500 speakers; Quechuan; spoken in Peru. Yavitero (Paraene, Yavitano) [yvt]: Dormant; Arawakan; formerly spoken in Venezuela. Yawalapití (Jaulapiti, Yaulapiti, Yawalpití) [yaw]: Critically endangered, eight speakers; Arawakan; spoken in Brazil. Yawarana (Yabarana, Yavarana, Yauarana) [yar]: Critically endangered, approximately 25 speakers; Cariban; spoken in Venezuela. Yekuana (Ye’kuana, Yecuana, Ye’cuana’ Maquiritare) [mch]: Threatened, 6,630 speakers; Cariban; spoken in Brazil and Venezuela. Yucuna (Chucuna, Matapi, Yukuna) [ycn]: Endangered, 550 speakers; Arawakan; spoken in Colombia. Yuhup (Makú-Yahup, Yëhup, Yahup) [yab]: Vulnerable, 400-550 speakers; Nadahup; spoken in Brazil and Colombia. Yuki (Yuqui, Bia, Yúki) [yuq]: Endangered, 140 speakers; Tupian; spoken in Bolivia. Yukpa (Yucpa-Yaprería, Japrería, Macoíta) [yup]: Vulnerable, fewer than 11,173 speakers; Cariban; spoken in Colombia and Venezuela. Yuracaré (Yurakaré, Yurujure) [yuz]: Threatened, 1,809 speakers; isolate; spoken in Bolivia. Yurutí (Yuruti, Juruti, Jurutí) [yui]: Endangered, 687 speakers; Tucanoan; spoken in Brazil and Colombia. Záparo (Záparo, Zápara, Kayapwe) [zro]: Critically endangered, three speakers; Zaparoan; spoken in Peru and Ecuador. Zo’é (Zoé, Jo’é, Buré) [pto]: Endangered, 177 speakers; Tupian; spoken in Brazil. Zoró (Zoró, Panginey, Cabeça Seca) [no ISO 639-3 code]: Endangered, 599 speakers; Tupian; spoken in Brazil. Zuruahã (Suruahá, Sorowahá, Suruwahá) [swx]: Endangered, 136 speakers; Arawan; spoken in Brazil.

South Asia Yen-ling Chen Linguistic diversity in South Asia South Asia in the Catalogue of Endangered Languages consists of six countries: Bangladesh, Bhutan, India, Nepal, Pakistan, and Sri Lanka.28 The Catalogue has identified 273 endangered languages in this region, among them 24 languages that have no ISO 639-3 code (and are not included in Ethnologue). Ethnologue (Simons & Fennig 2017) lists a total of 727 languages in these countries. Adding the new languages (those as yet lacking an ISO 639-3 code in Ethnologue), we can conclude that c.751 languages are spoken in South Asia.29 211

Anna Belew et al.

A majority of these languages belong to the Indo-European (specifically to the Indo-Aryan branch), Sino-Tibetan (specifically to Tibeto-Burman), Tai-Kadai, Dravidian, Austroasiatic, Great Andamanese, and South Andamanese families, although one creole language, three isolates, and two sign languages, as well as several unclassified languages, are also found here. In short, South Asia is linguistically very diverse.

Causes of language endangerment in South Asia In South Asia’s high level of linguistic diversity, multilingualism is a norm. The role that religion has played in self-identity and choice of language in this region is highly significant, where Hinduism (associated especially with Hindi speakers), Islam (to which Urdu is connected), and Sikhism (often linked with Punjabi) are particularly influential (Khokhlova 2014). Riots involving language disputes have been common in India’s history since partition. Some of the major languages in South Asia are Baluchi, Bengali, Dzongkha, Hindi and Urdu,30 Malayalam, Nepali, Pashto, Punjabi, Tamil, Telugu, and Sinhalese. These and other major languages influence the smaller languages there. Several major political movements centered around language have taken place in Bangladesh, India, and Pakistan, including the Bengali Language Movement (1948–1952), which eventually led to the independence of Bangladesh (formerly East Pakistan) from Pakistan (formerly West Pakistan) in 1971 (Alam 1991; Rahman 2002). In recent years, economic pressures have increasingly affected language choice, because ability to speak regional, national or official languages often enhances one’s socio-economic opportunities. With infrastructure improvements, more remote regions are now accessible to outsiders, and more people from rural regions have left their home areas to seek a better economic opportunities in the cities, where dominant languages are favored. Language maintenance has been correlated also with the size of a linguistic group; languages with small speaker populations have less strength to resist pressures to shift towards dominant languages. In addition, natural disasters such as earthquakes, floods, mudslides, tsunamis, and pandemics have at times presented threats to smaller languages in the region. In India, Hindi and English are designated national languages, while Indian states also designate their official languages. English has never lost its importance and prestige mostly because English has become a language “whose use prevents any single indigenous language from gaining prestige and privileging its speakers at the expense of other language and their speakers” as well as being a global lingua franca (Thomason 2001:43).31 In Pakistan, Urdu was selected as the official national language because (1) its speakers are politically dominant, and (2) none of the indigenous languages is favored above others (leaving English favored, as in India), although only 7.57% of the entire population speaks Urdu (Khokhlova 2014). In the modern era, language shift in these countries often begins with speakers of the minority languages extending their use of religiously, politically and/or economically dominant languages into more domains. As for minority languages spoken on the Nicobar islands and Andaman islands of India, their populations are small, and shrank further as the result of the 2004 Indian Ocean earthquake and tsunami. Survival of these languages has been challenged by immigration from the mainland in the twentieth century (Abbi 2013). The only exception is Sentinelese [std], since no outsiders are allowed to go to Sentinel Island. In Sri Lanka, both Tamil and Sinhala serve as the official languages, with English being influential in education, science, and economy. The status of English in Sri Lanka mirrors its importance and prestige in India, Pakistan, and Bangladesh. 212

The world’s endangered languages

Before Nepal’s transition to democracy, the policy “one nation, one culture, and one language” equated Nepalese identity with the Nepali language. The use of minority languages was disdained (Noonan 1996; Turin 2005). As a result, bilingualism has become the norm among people whose traditional language is not Nepali. Since becoming a republic in 2008, mother tongue education has been developed for certain languages as supported by Nepal’s National Language Policy Recommendations Commission. However, “some linguistic activists in Nepal see the development of a script for their language as primarily important for the status that this will accord their community on the national stage. . . rather than for any resulting mother tongue or bilingual education programme that may ensue” (Turin 2006:68). Noonan (2006:166) also pointed out that “what the census numbers reveal is an increase in ethnic consciousness, while at the same time hiding the true condition of the languages.” That is, persons’ self-identification is often not reflected in their ability to use the language they identify themselves with. In addition, economic pressures as well as education favor the use of Nepalese and English, especially among younger generations. In sum, the improved language rights in Nepal do not ensure the maintenance of minority languages. All the languages spoken in Bhutan belong to the Tibeto-Burman branch of Sino-Tibetan, except for Nepali (Indo-Aryan) spoken by recent immigrants. Several Tibeto-Burman “isolates,” that is, Tibeto-Burman languages without other Tibeto-Burman relatives to which they might be shown to have a closer relationship, have been discovered, such as Lepcha [lep], Lhokpu [lhp], Olekha [ole], and Gonduk [goe] (van Driem 2007a; Hyslop 2016). According to van Driem (2007a; 2007b), language shift to the official languages, i.e. English and Dzongkha, and to Sharchop/Tshangla [ tsj], the lingua franca of eastern Bhutan, is gradually taking place. The Royal Government of Bhutan values and encourages documentation and conservation of indigenous languages. However, the influx of Nepali-speaking immigrants in southern Bhutan has threatened the local varieties, causing language loss at an accelerated rate.

The current status of endangered languages in South Asia The Catalogue of Endangered Languages has identified 275 endangered languages in this region (which includes three dormant languages: Aka-Bo [abm], Aka-Kora [ack], and Veddah [ved]). This number includes three language isolates (Burushaski [bsk], Kusunda [kgg], and Nihali [nll]), one creole (Sri Lanka Malay [sci]), two sign languages (Alipur Sign Language and Sri Lankan Sign Language [sqs]), and five unclassified languages. There are no “Awakening” languages in South Asia (that is, no known revival efforts for formerly dormant languages). Out of the 273 languages, ten are “Critically Endangered” and ten “Severely Endangered”; 40 languages are listed in the “Endangered” category, while 106 are classified as “Threatened,” 83 as “Vulnerable,” and 21 as “At Risk.” There are a few other potentially endangered languages identified by van Driem (2007a), the vitality of which remains unknown. They are Bhalesi (Indo-European) and Chorei and Puiron (Sino-Tibetan), all spoken in India. Of the 727 living languages listed in South Asia by Ethnologue, 14 are listed as extinct (Simons & Fennig 2017). Of the 713 languages listed as still spoken, 273, c.38.2%, are endangered and thus are included in the Catalogue. Generally speaking, a high rate of language endangerment is attested in Arunachal Pradesh, Manipur, and Nagaland States of India, as well as the Eastern Development Region of Nepal, which are culturally highly diverse and mostly mountainous. Although a large number of these endangered languages belong to the Tibeto-Burman branch of Sino-Tibetan, the most endangered language families in this region are the Great Andamanese and Ongan (a.k.a. South Andamanese family, or Jarawa-Onge) families, both of the Andaman Islands, which belong 213

Anna Belew et al.

to India. Önge [oon] (one of the two Ongan languages) is listed as “Threatened” in the Catalogue of Endangered Languages, whereas Mixed Great Andamanese [gac] and Aka-Bo [abm] of the Great Andamanese family are considered “Critically Endangered” and “Dormant,” respectively. Below is an alphabetical listing of all languages of South Asia included in the Catalogue of Endangered Languages.

Endangered languages of South Asia Adi (崩尼-博嘎爾语, Lho-Pa, Abhor) [adi]: At risk, 100,190 speakers; Sino-Tibetan; spoken in India and China. Aer [aeq]: Endangered, 200 speakers; Indo-European; spoken in Pakistan. Aimol (Langrong, Aimal) [aim]: Endangered, 2,640 speakers; Sino-Tibetan; spoken in India. Aiton (Aitonia, Sham Doaniya) [aio]: Threatened, at least 1,000 speakers; Tai-Kadai; spoken in India. Aka-Bo (Bo, Ba) [abm]: Dormant; Great Andamanese; formerly spoken in India. Aka-Kora (Kora, Khora, Aka-Khora) [ack]: Dormant; Great Andamanese; formerly spoken in India. Alipur Sign Language [no ISO 639-3 code]: Threatened; sign language; spoken in India. Allar (Alan, Alanmar, Alar) [all]: Endangered, 350 speakers; Dravidian; spoken in India. Alu Kurumba (Pal Kurumba, Hal Kurumba) [xua]: Threatened, approximately 1,000 speakers; Dravidian; spoken in India. Anal (Namfau) [anm]: Endangered; Sino-Tibetan; spoken in India and Myanmar. Apatani (Apa, Patan, Apa Tani) [apt]: Vulnerable, approximately 14,000 speakers; Sino-Tibetan; spoken in India. Aranadan (Aranatan, Eranadans) [aaf]: Endangered, 200 speakers; Dravidian; spoken in India. Ashing [no ISO 639-3 code]: Endangered, fewer than 1,000 speakers; Sino-Tibetan; spoken in India. Asuri (Ashree, Asura, Assur) [asr]: Threatened, approximately 7,000 speakers; Austro-Asiatic; spoken in India. Athpariya (Athpare, Rai, Athapre) [aph]: Threatened, fewer than 3,658 speakers; SinoTibetan; spoken in Nepal. Atong (A’tong) [aot]: Threatened; Sino-Tibetan; spoken in India. Baghati [no ISO 639-3 code]: Threatened, fewer than 4,000 speakers; Indo-European; spoken in India. Bahing (Bhojpuri, Khaling, Rai) [bhj]: Critically endangered; Sino-Tibetan; spoken in Nepal. Balti (Sbalti, Baltistani, Bhotia of Baltistan) [bft]: At risk, 337,000 speakers; Sino-Tibetan; spoken in Pakistan and India. Bangani [no ISO 639-3 code]: Vulnerable, 12,000 speakers; Indo-European; spoken in India. Bantawa (Kiranti, Rai, Bantawa Rai) [bap]: Threatened, 371,056 speakers; Sino-Tibetan; spoken in Nepal. Baram (Barhamu, Brahmu, Bhramu) [brd]: Critically endangered, 53 speakers; Sino-Tibetan; spoken in Nepal. Bateri (Bateri Kohistani, Batera Kohistani, Baterawal) [btv]: Vulnerable, 29,051 speakers; IndoEuropean; spoken in Pakistan and India. Bawm Chin (Bawm, Banjogi, Bawng) [bgr]: Vulnerable; Sino-Tibetan; spoken in India, Myanmar, and Bangladesh. 214

The world’s endangered languages

Belhariya (Belhare, Athpariya, Athpahariya) [byw]: Severely endangered, few speakers; SinoTibetan; spoken in Nepal. Bellari [brw]: Endangered, 1,000 speakers; Dravidian; spoken in India. Betta Kurumba (Kadu Kurumba, Urali Kurumba) [xub]: Threatened, 2,000–6,000 speakers; Dravidian; spoken in India. Bhadrawahi (Bradrawah, Baderwali, Badrohi) [bhd]: Vulnerable, fewer than 69,000 speakers; Indo-European; spoken in Pakistan and India. Bhalesi [no ISO 639-3 code]: No estimate available; Indo-European; spoken in India. Bhaya [bhe]: Endangered, 700 speakers; Indo-European; spoken in Pakistan. Bhumij (Sadar Bhumij, Bhumij Munda, Bhumij Thar) [no ISO 639-3 code]: At risk, at least 100,000 speakers; Austro-Asiatic; spoken in India. Biete (Bete, Baite, Biate) [biu]: Vulnerable, 19,000 speakers; Sino-Tibetan; spoken in India. Birhor (Birhar, Bīrhaṛ, Bihor) [biy]: Threatened, fewer than 2,000 speakers; Austro-Asiatic; spoken in India. Bishnupriya (Bishnupuriya Manupuri, Bisna Puriya, Bishnupria Manipuri) [bpy]: Vulnerable; Indo-European; spoken in India, Myanmar, and Bangladesh. Bokar (Boga’er Luoba, 崩尼-博嘎爾语) [no ISO 639-3 code]: Threatened, 3,500 speakers; Sino-Tibetan; spoken in India and China. Bori [no ISO 639-3 code]: Threatened, 2,000 speakers; Sino-Tibetan; spoken in India. Bote-Majhi (Kushar) [bmj]: Vulnerable, 24,664 speakers; Indo-European; spoken in Nepal. Brokkat (Brokskad, Jokay) [bro]: Endangered, 300 speakers; Sino-Tibetan; spoken in Bhutan. Brokpake (Mira Sagtengpa, Dakpa, Brokpa) [sgt]: Threatened, fewer than 5,000 speakers; SinoTibetan; spoken in Bhutan. Brokskat (Brokpa, Brokpa of Dah-Hanu, Dokskat) [bkk]: Threatened, 3,000 speakers; IndoEuropean; spoken in India. Bugun (Khowa, Kho, Khoa) [bgg]: Threatened, 1,700 speakers; Sino-Tibetan; spoken in India. Burushaski (Mishaski, Brushaski, Burushaki) [bsk]: Threatened, 100,300 speakers; isolate; spoken in Pakistan and India. Byangsi (Byangkho Lwo, Byasi, Byanshi) [bee]: Threatened, approximately 1,500 speakers; Sino-Tibetan; spoken in India and Nepal. Camling (Rai, Rodong, Chamling) [rab]: Vulnerable, 30,000 speakers; Sino-Tibetan; spoken in Nepal. Car Nicobarese (Pu, Car) [caq]: Vulnerable, 37,000 speakers; Austro-Asiatic; spoken in India. Central Nicobarese (Muöt, Nicobarese, Nancowry) [ncb]: Vulnerable, 10,100 speakers; Austro-Asiatic; spoken in India. Chalikha (Chali, Tshali, Chalipkha) [tgf]: Threatened, 1,000 speakers; Sino-Tibetan; spoken in Bhutan. Chambeali (Chameali, Chamaya, Chambiali) [cdh]: At risk, 129,654 speakers; Indo-European; spoken in India. Chantyal (Chentel, Chantel, Chhantel) [chx]: Threatened, at least 2,000 speakers; SinoTibetan; spoken in Nepal. Chaudangsi (Bangba Lo, Bangbani, Chanpa Lo) [cdn]: Endangered, fewer than 3,500 speakers; Sino-Tibetan; spoken in India and Nepal. Chaura (Chowra, Tutet, Tətet) [crv]: Threatened, 2,020 speakers; Austro-Asiatic; spoken in India. Chepang (Tsepang, Chēpāng, Tśepang) [cdm]: Vulnerable, 25,097 speakers; Sino-Tibetan; spoken in Nepal. 215

Anna Belew et al.

Chhintange (Chhintange, Teli, Chintang Rûng) [ctn]: Threatened, 3,500 speakers; SinoTibetan; spoken in Nepal. Chhulung (Chilling, Chɨlɨng, Chulung) [cur]: Endangered, 1,314 speakers; Sino-Tibetan; spoken in Nepal. Chilisso (Chiliss, Galos, Dardu) [clh]: Threatened, 2,000–3,000 speakers; Indo-European; spoken in Pakistan. Chiru (Chhori, Tśiru) [cdf]: Threatened, 7,000 speakers; Sino-Tibetan; spoken in India. Chocangacakha (Maphekha, Rtsamangpa’ikha, Tsagkaglingpa’ikha) [cgk]: Vulnerable, 20,000 speakers; Sino-Tibetan; spoken in Bhutan. Chokri Naga (Chokri, Eastern Angami, Chakrima Naga) [nri]: Vulnerable, fewer than 20,000 speakers; Sino-Tibetan; spoken in India. Chorei [no ISO 639-3 code]: Severely endangered; Sino-Tibetan; spoken in India. Chukwa (Cukwa Ring, Pohing, Pohing Kha) [cuw]: Critically endangered, five speakers; SinoTibetan; spoken in Nepal. Chungli (Ao-Chungli, Chongli, Tśungli) [no ISO 639-3 code]: Vulnerable, 65,275 speakers; Sino-Tibetan; spoken in India. Churahi (Churahi Pahari, Chaurahi, Churai Pahari) [cdj]: At risk, 110,552 speakers; IndoEuropean; spoken in India. Dakpa (Dakpakha) [dka]: Vulnerable, 31,000 speakers; Sino-Tibetan; spoken in India and Bhutan. Dameli (Damēlī, Damel, Damedi) [dml]: Vulnerable, 5,000 speakers; Indo-European; spoken in Pakistan. Darai [dry]: Endangered, approximately 3,060 speakers; Indo-European; spoken in Nepal. Darma (Darmiya, Darimiya, Sauka) [drd]: Threatened, fewer than 2,600 speakers; Sino-Tibetan; spoken in India and Nepal. Deori (Chutia, Deuri, Chutiya) [der]: Threatened, 28,000 speakers; Sino-Tibetan; spoken in India. Dhanwar (Dhanvar, Danuwar Rai, Danuwar) [dhw]: Threatened, 23,721 speakers; IndoEuropean; spoken in Nepal. Dhimal (Dhīmāl) [dhi]: Endangered, approximately 20,000 speakers; Sino-Tibetan; spoken in India and Nepal. Digaro-Mishmi (Darang Deng, 達讓語) [mhu]: Threatened, approximately 6,000 speakers; Sino-Tibetan; spoken in India and China. Dimasa (Hill Kachari, Dimasa Kachari) [dis]: At risk, 112,000 speakers; Sino-Tibetan; spoken in India. Dolpo (Phoke Dolpa, Dolpa Tibetan, Dolpike) [dre]: Threatened, approximately 5,000 speakers; Sino-Tibetan; spoken in Nepal. Domaaki (Dumaki, Dumākī, Doma) [dmk]: Severely endangered, 350 speakers; IndoEuropean; spoken in Pakistan. Dumi (Rai, Dumi Bo’o, Dumi Bro) [dus]: Critically endangered, fewer than eight speakers; Sino-Tibetan; spoken in Nepal. Dungmali (Dungmali Pûk, Dungmali-Bantawa, Arthare) [raa]: Threatened, 4,609 speakers; Sino-Tibetan; spoken in Nepal. Duruwa (Parji, Dhurwa) [pci]: Vulnerable, 51,200 speakers; Dravidian; spoken in India. Eastern Gurung (Eastern Gurnung) [ggn]: Threatened, 325,000 speakers; Sino-Tibetan; spoken in Nepal. Eastern Meohang (Rai, Newang, Newahang) [emg]: Severely endangered, 904 speakers; SinoTibetan; spoken in Nepal. 216

The world’s endangered languages

Gaddi (Bharmauri Bhadi, Pahari Bharmauri, Panchi Brahmauri Rajput) [gbk]: At risk, 120,000 speakers; Indo-European; spoken in India. Gahri (Ghara, Lahuli of Bunan, Boonan) [bfu]: Threatened, 3,500–4,000 speakers; Sino-Tibetan; spoken in India. Galo (Gallong) [adl]: Vulnerable, 30,000–40,000 speakers; Sino-Tibetan; spoken in India. Gamale Kham (Kham, Proto-Kham, Khamkura) [kgj]: Vulnerable, 30,000 speakers; SinoTibetan; spoken in Nepal. Gangte (Gante) [gnb]: Vulnerable, 15,100 speakers; Sino-Tibetan; spoken in India and Myanmar. Garhwali (Gadhavali, Gadhawala, Gadwahi) [gbm]: Vulnerable, 2,920,000 speakers; IndoEuropean; spoken in India. Garo (Garrow, Mande) [grt]: Vulnerable, 700,000 speakers; Sino-Tibetan; spoken in India and Bangladesh. Gongduk (Gongdubikha) [goe]: Threatened, approximately 1,000 speakers; Sino-Tibetan; spoken in Bhutan. Gorum (Gorum, Pareng, Parengi) [pcj]: Endangered, 750–6,000 speakers; Austro-Asiatic; spoken in India. Gowro (Gabaro, Gabar Khel, Dardu) [gwf]: Threatened, 1,000 speakers; Indo-European; spoken in Pakistan. Gta’ (Gataʔ, Gata’, Gataq) [gaq]: Threatened, approximately 3,000 speakers; Austro-Asiatic; spoken in India. Gutob (Gutob, Gadaba, Gadba) [gbj]: Endangered, 8,000 speakers; Austro-Asiatic; spoken in India. Gyalsumdo [gyo]: Threatened, 300-400 speakers; Sino-Tibetan; spoken in Nepal. Hakha Chin ( , Lai, Hakha Lai) [cnh]: Threatened; Sino-Tibetan; spoken in India and Myanmar. Helambu Sherpa (Yolmo, Hyolmo) [scp]: Threatened, 9,842 speakers; Sino-Tibetan; spoken in Nepal. Hill Miri (Miri, Sarak) [no ISO 639-3 code]: Threatened, approximately 10,000 speakers; SinoTibetan; spoken in India. Hinduri (Handuri) [hii]: Endangered, 138 speakers; Indo-European; spoken in India. Hmar (Hamar, Mhar, Hmari) [hmr]: Vulnerable, 83,404 speakers; Sino-Tibetan; spoken in India. Hruso (Aka, Hrusso, Angka) [hru]: Threatened, fewer than 3,000 speakers; isolate; spoken in India. Idu-Mishmi (Idu, Midu, Lho-Pa) [clk]: Vulnerable, 11,080 speakers; Sino-Tibetan; spoken in India and China. Irula (Iruḷa, Eravallan, Erukala) [iru]: Threatened, 5,000 speakers; Dravidian; spoken in India. Jad (Bhotia, Dzad) [jda]: Endangered, approximately 2,000 speakers; Sino-Tibetan; spoken in Pakistan and India. Jangshung (Jangrami, Zangram, Zhang-Zhung) [jna]: Threatened, 1,990 speakers; SinoTibetan; spoken in India. Jarawa [anq]: Vulnerable, 340 speakers; South Andamanese; spoken in India. Jaunsari (Janusari, Jaunsauri, Jansauri) [jns]: Vulnerable, 96,995 speakers; Indo-European; spoken in India. Jerung (Jero, Jerum, Jerunge) [jee]: Endangered, 271 speakers; Sino-Tibetan; spoken in Nepal. Jirel (Ziral, Jiri, Jirial) [jul]: Threatened, 4,919 speakers; Sino-Tibetan; spoken in Nepal. 217

Anna Belew et al.

Juang (Patua, Patra-Saara, Puttooas) [jun]: Vulnerable, 30,875 speakers; Austro-Asiatic; spoken in India. Kachari (Plains Kachari, Cachari) [xac]: Endangered, 59,000 speakers; Sino-Tibetan; spoken in India. Kagate (Shuba, Shyuba, Syuba) [syw]: Threatened, 100 speakers; Sino-Tibetan; spoken in Nepal. Kaike (Tarali Kham) [kzq]: Vulnerable, 794 speakers; Sino-Tibetan; spoken in Nepal. Kalami (Baškarīk, Gāwrī, Gārwī) [gwc]: Threatened, 1,500 speakers; Indo-European; spoken in Pakistan. Kalasha (Kalasa, Kalashamon, Kalash) [kls]: Threatened, 2,900–5,700 speakers; IndoEuropean; spoken in Pakistan. Kanashi (Kanasi, Kanāshī, Kanaśi) [xns]: Threatened, 1,400 speakers; Sino-Tibetan; spoken in India. Karbi (Mikir, Arleng, Manchati) [mjw]: At risk, at least 150,000 speakers; Sino-Tibetan; spoken in India. Karko [no ISO 639-3 code]: Threatened, 2,000 speakers; Sino-Tibetan; spoken in India. Khaling (Rai, Kaling, Khalinge Rai) [klr]: Vulnerable, 15,000 speakers; Sino-Tibetan; spoken in India and Nepal. Khamba (Khamba Khaadi) [kbg]: Threatened, 1,330 speakers; Sino-Tibetan; spoken in India. Khamti (Hkamti, Khampti, Khamti Shan) [kht]: Vulnerable, 13,000 speakers; Tai-Kadai; spoken in India and Myanmar. Khamyang (Khamjang, Khamiyang, Shyam) [ksu]: Critically endangered, 50 speakers; TaiKadai; spoken in India. Kharia (Haria, Kharvi, Khatria) [khr]: Vulnerable; Austro-Asiatic; spoken in India and Nepal. Khengkha (Khenkha, Khen, Keng) [xkf]: Vulnerable, approximately 40,000 speakers; SinoTibetan; spoken in Bhutan. Khezha Naga (Kheza, Kezami, Khezhama) [nkh]: Vulnerable, fewer than 23,000 speakers; Sino-Tibetan; spoken in India. Khiamniungan Naga (Khiamngan, Aoshedd, Khiemnungan) [kix]: Vulnerable, 47,800 speakers; Sino-Tibetan; spoken in India. Khowar (Khowār, Citālī, Čitarī) [khw]: At risk, 300,000 speakers; Indo-European; spoken in Pakistan and India. Kinnauri (Kinnaura Yanuskad, Kanoreunu Skad, Kanorug Skadd) [kfk]: Threatened, fewer than 84,000 speakers; Sino-Tibetan; spoken in India. Koch (Koc, Kocch, Koce) [kdq]: Vulnerable; Sino-Tibetan; spoken in India and Bangladesh. Koda (Kora, Kōḍā, Kaora) [cdz]: Vulnerable, 25,000 speakers; Austro-Asiatic; spoken in India and Bangladesh. Kodagu (Kodava, Coorge, Kadagi) [kfa]: Threatened, 97,011 speakers; Dravidian; spoken in India. Koi (Koyu, Kohi, Koyi) [kkt]: Threatened, 2,000–3,000 speakers; Sino-Tibetan; spoken in Nepal. Koireng (Kwoireng, Koirng, Kolren) [nkd]: Vulnerable, 1,410 speakers; Sino-Tibetan; spoken in India. Kom (Kom Rem, Kolhreng) [kmm]: Vulnerable, 15,467 speakers; Sino-Tibetan; spoken in India. Konda-Dora (Porja, Koṇḍa, Kūbi) [kfc]: Vulnerable, 15,000 speakers; Dravidian; spoken in India. Konyak Naga (Konyak, Tableng, Kanyak) [nbe]: Vulnerable, approximately 70,000 speakers; Sino-Tibetan; spoken in India. 218

The world’s endangered languages

Korku (Kurku, Muwasi, Muasi) [kfq]: At risk, 200,000 speakers; Austro-Asiatic; spoken in India. Koro (Aka, Hrusso, Angka) [jkr]: Endangered, 800–1,200 speakers; unclassified; spoken in India. Korra Koraga (Koragar, Koragara, Korangi) [kfd]: Vulnerable, 14,000 speakers; Dravidian; spoken in India. Korwa (Ernga, Singli, Erngga) [kfp]: Vulnerable, at least 35,000 speakers; Austro-Asiatic; spoken in India. Kota (Kotta, Kowe-Adiwasi, Kother-Tamil) [kfe]: Endangered, 900 speakers; Dravidian; spoken in India. Kui (Kui, Kandh, Khondi) [kxu]: At risk, fewer than 500,000 speakers; Dravidian; spoken in India. Kulung (Rai, Khulunge Rai, Kulu Ring) [kle]: Threatened, 33,200 speakers; Sino-Tibetan; spoken in India and Nepal. Kumarbhag Paharia (Malto, Malti, Maltu) [kmj]: Vulnerable, 20,200 speakers; Dravidian; spoken in India. Kumhali (Kumhale, Kumbale, Kumkale) [kra]: Endangered, 12,200 speakers; Indo-European; spoken in Nepal. Kurichiya (Kurichia, Kurichchia, Kowohans) [kfh]: Endangered, 29,400 speakers; Dravidian; spoken in India. Kurtokha (Gurtü, Kurtopakha, Kürthöpka) [xkz]: Vulnerable, 15,000 speakers; Sino-Tibetan; spoken in Bhutan. Kusunda (Kusanda, Kusūndu) [kgg]: Critically endangered, two speakers; isolate; spoken in Nepal. Kuvi (Kuwi, Kuvinga, Kuvi Kond) [kxv]: At risk, 246,513 speakers; Dravidian; spoken in India. Ladakhi (Ladaphi, Ladhakhi, Ladak) [lbj]: Threatened, 150,000 speakers; Sino-Tibetan; spoken in India and China. Lakha (Tshangkha) [lkh]: Threatened, 8,000 speakers; Sino-Tibetan; spoken in Bhutan. Lamkang [lmk]: Endangered, 5,894 speakers; Sino-Tibetan; spoken in India. Lepcha (Lapche, Lapcha, Rong) [lep]: Severely endangered, a few thousand speakers; SinoTibetan; spoken in India, Nepal, and Bhutan. Levai (Bangru) [no ISO 639-3 code]: Threatened, approximately 1,000 speakers; unclassified; spoken in India. Lhokpu (Lhobikha, Taba-Damey-Bikha) [lhp]: Threatened, 2,500 speakers; Sino-Tibetan; spoken in Bhutan. Lhomi (Lhoket, Shing Saapa, Kath Bhote) [lhm]: Vulnerable, 3,130 speakers; Sino-Tibetan; spoken in India, Nepal, and China. Liangmai Naga (Liangmai, Kwoireng, Kacha) [njn]: Vulnerable, 34,232 speakers; SinoTibetan; spoken in India. Limbu (Yakthung Pan, Limbo, Lumbu) [lif]: Threatened, 300,000 speakers; Sino-Tibetan; spoken in India and Nepal. Lotha Naga (Lotha, Lhota, Chizima) [njh]: At risk, 170,001 speakers; Sino-Tibetan; spoken in India. Mahasu Pahari (Baghati, Western Pahari, Mahasui) [bfz]: At risk, approximately 500,000 speakers; Indo-European; spoken in India. Majhi (Manjhi) [mjz]: Endangered, 24,422 speakers; Indo-European; spoken in India and Nepal. 219

Anna Belew et al.

Mal Paharia (Mal Pahoria, Malto, Malti) [mkb]: Vulnerable, 51,000 speakers; Indo-European; spoken in India. Manange (Manang, Manangba, Ngyeshang) [nmm]: Threatened, approximately 3,000 speakers; Sino-Tibetan; spoken in Nepal. Manda (Manḍa) [mha]: Threatened, 4,040 speakers; Dravidian; spoken in India. Mao Naga (Mao, Sopvoma, Spowama) [nbi]: Vulnerable, 77,810 speakers; Sino-Tibetan; spoken in India. Mara Chin (Mara, Lakher, Chin) [mrh]: Vulnerable, 42,000 speakers; Sino-Tibetan; spoken in India and Myanmar. Maram Naga (Maram, Naga, Maram) [nma]: Vulnerable, 37,340 speakers; Sino-Tibetan; spoken in India. Maring Naga (Maring, Naga, Maring) [nng]: Threatened, 22,326 speakers; Sino-Tibetan; spoken in India. Miju-Mishmi (Geman Deng, 格曼語) [mxj]: Threatened, approximately 9,000 speakers; Sino-Tibetan; spoken in India and China. Milang (Holon) [no ISO 639-3 code]: Vulnerable, 2,150 speakers; unclassified; spoken in India. Minyong (Padam-Minyong Adi) [no ISO 639-3 code]: Vulnerable, approximately 20,000 speakers; Sino-Tibetan; spoken in India. Mising (Mishing, Takam, Plains Miri) [mrg]: At risk, 551,224 speakers; Sino-Tibetan; spoken in India. Mixed Great Andamanese (Present Great Andamanese, Jeru, Andamanese) [gac]: Critically endangered, seven speakers; Great Andamanese; spoken in India. Mru (Murung, Mro, Mrung) [mro]: Vulnerable, 50,000 speakers; Sino-Tibetan; spoken in Bangladesh, Myanmar, and India. Mudhili Gadaba (Kondekor Gadaba) [gau]: Threatened, 8,000 speakers; Dravidian; spoken in India. Mudu Koraga [vmd]: Threatened, approximately 1,000 speakers; Dravidian; spoken in India. Mzieme Naga (Mzieme, Northern Zeme, Zeme) [nme]: Vulnerable, approximately 29,000 speakers; Sino-Tibetan; spoken in India. Na [nbt]: Threatened, 1,500 speakers; Sino-Tibetan; spoken in India. Nachering (Nachereng, Nacering Ra, Nachering Tûm) [ncd]: Vulnerable, 10,041 speakers; Sino-Tibetan; spoken in Nepal. Nahari (Nahali) [nhh]: Vulnerable, 20,400 speakers; Indo-European; spoken in India. Nar Phu (Nar-Phu, Narpa) [npa]: Endangered, 500 speakers; Sino-Tibetan; spoken in Nepal. Nepali Kurux (Kurukh, Oraon, Kurux) [kxl]: Vulnerable, 28,615 speakers; Dravidian; spoken in Nepal. Nihali (Nihal, Nahali, Kalto) [nll]: Threatened, 1,000–2,000 speakers; isolate; spoken in India. Nocte Naga (Nocte, Namsangia, Naga) [njb]: Threatened, 35,000 speakers; Sino-Tibetan; spoken in India. Northern Lorung (Lohorong, Lohrung, Lohrung Khanawa) [lbr]: Threatened, 1,207 speakers; Sino-Tibetan; spoken in Nepal. Northern Rengma Naga (Rengma North, Ntenyi, Ntenyi Naga) [nnl]: Threatened, fewer than 6,600 speakers; Sino-Tibetan; spoken in India. Nubri (Kutang Bhotia, Larkye) [kte]: Threatened, 2,000 speakers; Sino-Tibetan; spoken in Nepal. Nupbikha [npb]: Vulnerable, 30,000 speakers; Sino-Tibetan; spoken in Bhutan. Nyenkha (Henkha, Lap, Mangsdekha) [neh]: Vulnerable, 8,692 speakers; Sino-Tibetan; spoken in Bhutan. 220

The world’s endangered languages

Olekha (Monpa, Ole Mönpa, ‘Olekha) [ole]: Critically endangered, three speakers; Sino-Tibetan; spoken in Bhutan. Önge (Ong, Onge) [oon]: Threatened, 94 speakers; South Andamanese; spoken in India. Padam (Standard Adi, Padam Adi, Padam-Minyong Adi) [no ISO 639-3 code]: Vulnerable, approximately 40,000 speakers; Sino-Tibetan; spoken in India. Pailibo (Palibo) [no ISO 639-3 code]: Endangered, approximately 1,000 speakers; SinoTibetan; spoken in India. Paite Chin (Paite, Oarte, Paithe) [pck]: Vulnerable, 64,100 speakers; Sino-Tibetan; spoken in India. Pangwali (Satlaj, Pahari, Pangi) [pgg]: Threatened, 17,000 speakers; Indo-European; spoken in India. Pani Koch (Banai, Wanang) [no ISO 639-3 code]: Endangered, 300 speakers; Sino-Tibetan; spoken in India and Bangladesh. Pankhu (Pankho, Pankhua, Pang Khua) [pkh]: Vulnerable; Sino-Tibetan; spoken in India and Bangladesh. Pattani (Manchati, Manchad, Patni) [lae]: Vulnerable, 11,000 speakers; Sino-Tibetan; spoken in India. Phake (Phakial, Phakey, Faake) [phk]: Threatened, 2,000 speakers; Tai-Kadai; spoken in India. Phalura (Palūla, Palola, Dangarīk) [phl]: Vulnerable, 9,841 speakers; Indo-European; spoken in Pakistan. Phom Naga (Chingmengnu, Tamlu) [nph]: Vulnerable, approximately 19,000 speakers; SinoTibetan; spoken in India. Pochuri Naga (Southern Sangtam, Eastern Rengma) [npo]: Threatened, fewer than 13,000 speakers; Sino-Tibetan; spoken in India. Pottangi Ollar Gadaba (Ollari, Ollaro) [gdb]: Vulnerable, 15,000 speakers; Dravidian; spoken in India. Puiron (Puiron) [no ISO 639-3 code]: No estimate available; Sino-Tibetan; spoken in India. Puma (Puma Pima, Puma La, Puma Kala) [pum]: Threatened, 4,000 speakers; Sino-Tibetan; spoken in Nepal. Purik (Purigskad, Burig, Purig) [prx]: At risk, 132,000 speakers; Sino-Tibetan; spoken in Pakistan and India. Puroik (Sulung, 蘇龍語 [suv]: Threatened, 5,000–10,000 speakers; Sino-Tibetan; spoken in India and China. Rabha (Rava) [rah]: At risk, 139,365 speakers; Sino-Tibetan; spoken in India. Raji (Rajibar) [rji]: Endangered, 472 speakers; Sino-Tibetan; spoken in India and Nepal. Ramo [no ISO 639-3 code]: Endangered, fewer than 800 speakers; Sino-Tibetan; spoken in India. Raute (Rautye, Harka Gurung, Khamchi) [rau]: Vulnerable, 700 speakers; Sino-Tibetan; spoken in India and Nepal. Remo (Remsan, Bonda, Bondo) [bfw]: Threatened, 9,000 speakers; Austro-Asiatic; spoken in India. Riang (Reang, Kau Bru, Tipra) [ria]: Threatened, 5,900 speakers; Sino-Tibetan; spoken in India and Bangladesh. Rongmei Naga (Nruanghmei, Rongmei, Kabui) [nbu]: Vulnerable, 59,000 speakers; SinoTibetan; spoken in India. Rongpo (Rangpo) [rnp]: Threatened, approximately 12,000 speakers; Sino-Tibetan; spoken in India. Ruga [ruh]: Critically endangered, a few speakers; Sino-Tibetan; spoken in India. 221

Anna Belew et al.

Saam (Rai, Saam Rai, Samakha) [raq]: Severely endangered, 23 speakers; Sino-Tibetan; spoken in Nepal. Sajalong (Miji, Dammai, Dhammai) [sjl]: Threatened, 4,000 speakers; unclassified; spoken in India. Sampang (Rai, Sampange Rai, Sangpang) [rav]: Severely endangered, approximately 1,000 speakers; Sino-Tibetan; spoken in Nepal. Sauria Paharia (Malto, Malti, Maltu) [mjt]: Vulnerable, 117,000 speakers; Dravidian; spoken in India and Bangladesh. Seke [skj]: Threatened, approximately 7,000 speakers; Sino-Tibetan; spoken in Nepal. Sentinel (Sentinelese) [std]: Vulnerable; unclassified; spoken in India. Sherdukpen (Ngnok) [sdp]: Threatened, fewer than 2,000 speakers; Sino-Tibetan; spoken in India. Sherpa (Sharpa, Sharpa Bhotia, Xiaerba) [xsr]: Threatened, 15,000–70,000 speakers; SinoTibetan; spoken in India, Nepal, and China. Shimong [no ISO 639-3 code]: Threatened, approximately 2,000 speakers; Sino-Tibetan; spoken in India. Sholaga (Kadu Sholigar, Sholiga, Sholigar) [sle]: Vulnerable, 24,000 speakers; Dravidian; spoken in India. Shom Peng (Shom Pen, Shompeng, Shompen) [sii]: Threatened, 300–400 speakers; AustroAsiatic; spoken in India. Simte [smt]: Threatened, 7,150 speakers; Sino-Tibetan; spoken in India. Singpho (Sing-Fo, Kachin, Jingphaw) [sgp]: Threatened, approximately 2,500 speakers; SinoTibetan; spoken in India. Sirmauri (Sirmouri, Sirmuri) [srx]: At risk, 400,000 speakers; Indo-European; spoken in India. Sonha (Sonaha, Sonahaa, Sunah) [soi]: Endangered, 579 speakers; Indo-European; spoken in Nepal. Sora (Saora, Savara, Sabara) [srb]: At risk, 150,000–300,000 speakers; Austro-Asiatic; spoken in India. Southeastern Kolami (Kōlāmì, Kolami, Naiki) [nit]: Threatened, 1,500 speakers; Dravidian; spoken in India. Southern Ghale (Gurung, Galle Gurung, Ghale) [ghe]: Vulnerable, 1,649 speakers; SinoTibetan; spoken in Nepal. Southern Nicobarese (Condul, Great Nicobar) [nik]: Threatened, approximately 7,500 speakers; Austro-Asiatic; spoken in India. Southern Rengma Naga (Rengma, Rengma Proper, Rengma Naga) [nre]: Vulnerable; SinoTibetan; spoken in India. Southern Yamphu (Lohorong, Lohrung, Lohrung Khap) [lrr]: Threatened, 3,000–5,000 speakers; Sino-Tibetan; spoken in Nepal. Spiti Bhoti (Spiti, Spitian, Piti Bhoti) [spt]: Threatened, 10,000 speakers; Sino-Tibetan; spoken in Pakistan and India. Sri Lanka Malay (Sri Lankan Creole Malay, Melayu Bahasa, Java Jati) [sci]: Endangered; pidgin or creole; spoken in Sri Lanka. Sri Lankan Sign Language [sqs]: Vulnerable, 12,800 speakers; Sign Language; spoken in Sri Lanka. Sumi Naga (Simi, Sema, Sumi) [nsm]: At risk, 100,000 speakers; Sino-Tibetan; spoken in India. Sunwar (Sunuwar, Sunbar, Sunwari) [suz]: Threatened, 26,611 speakers; Sino-Tibetan; spoken in Nepal. 222

The world’s endangered languages

Surel [no ISO 639-3 code]: Endangered, a few hundred speakers; Sino-Tibetan; spoken in Nepal. Tagin [tgj]: Vulnerable, approximately 25,000 speakers; Sino-Tibetan; spoken in India. Takpa (Dwags, Dakpa, Northern Monpa) [twm]: Threatened, a few thousand speakers; SinoTibetan; spoken in India, China, and Bhutan. Tangam [no ISO 639-3 code]: Severely endangered, fewer than 200 speakers; Sino-Tibetan; spoken in India. Tangkhul Naga (Tangkhul, Tagkhul, Thangkhulm) [nmf]: Endangered, 58,167 speakers; Sino-Tibetan; spoken in India. Tarao Naga (Tarao, Tarau, Taraotrong) [tro]: Endangered, 870 speakers; Sino-Tibetan; spoken in India. Teressa (Taih-Long, Teressa-Bompoka, Təih-lɔng) [tef]: Threatened, 2,080 speakers; AustroAsiatic; spoken in India. Thado Chin (Thado, Kuki-Thado, ) [tcz]: Severely endangered; Sino-Tibetan; spoken in India and Myanmar. Thakali (Tapaang, Thaksya, Panchgaunle) [ths]: Threatened, 7,113 speakers; Sino-Tibetan; spoken in Nepal. Thangal Naga (Khoirao, Khoirao Naga, Koirao) [nki]: Vulnerable, fewer than 23,600 speakers; Sino-Tibetan; spoken in India. Thangmi (Thami, Dolakha, Thāmī) [thf]: Threatened, 18991 speakers; Sino-Tibetan; spoken in India and Nepal. Thulung (Rai, Thulunge Rai, Thulu Luwa) [tdh]: Severely endangered, approximately 1,000 speakers; Sino-Tibetan; spoken in India and Nepal. Tiddim Chin (Chin, Tedim, ) [ctd]: At risk, at least 100,000 speakers; Sino-Tibetan; spoken in India and Myanmar. Tilung (Tiling, Tilling, Tilung Blama) [tij]: Critically endangered, seven speakers; SinoTibetan; spoken in Nepal. Tinani (Gondhla, Gondla, Lahauli) [lbf]: Threatened; Sino-Tibetan; spoken in India and China. Tiwa (Lalung, Dowyan) [lax]: Vulnerable, 27,072 speakers; Sino-Tibetan; spoken in India. Toda (Todi, Tuda) [tcx]: Endangered, approximately 1,000 speakers; Dravidian; spoken in India. Torwali (Torwālī, Turvali, Dardu) [trw]: Vulnerable, 80,000 speakers; Indo-European; spoken in Pakistan. Toto (Tōtō) [txo]: Threatened, 1,400 speakers; Sino-Tibetan; spoken in India. Tsum (Tsumge) [ttz]: Threatened, 4,790 speakers; Sino-Tibetan; spoken in Nepal. Turi [trd]: Threatened, 2,000 speakers; Austro-Asiatic; spoken in India. Ushojo (Ushuji, Dardu) [ush]: Threatened, 1,000 speakers; Indo-European; spoken in Pakistan. Vaiphei (Bhaipei, Vaipei, Veiphei) [vap]: Vulnerable, fewer than 39,673 speakers; Sino-Tibetan; spoken in India. Veddah (Veda, Vedda, Veddha) [ved]: Dormant; Indo-European; formerly spoken in Sri Lanka. Vishavan (Malankudi, Malarkuti) [vis]: Endangered, approximately 150 speakers; Dravidian; spoken in India. Walungge (Olangchung Gola, Walungchung Gola, Walung) [ola]: Threatened, approximately 1,000 speakers; Sino-Tibetan; spoken in India and Nepal. Wambule (Umbule, Chaurasya, Tsaurasya) [wme]: Vulnerable, 15,000 speakers; Sino-Tibetan; spoken in Nepal. Wancho Naga (Wancho, Banpara, Banpara Naga) [nnp]: Vulnerable, approximately 30,000 speakers; Sino-Tibetan; spoken in India. 223

Anna Belew et al.

Wayu (Hayu, Vayu, Wayo) [vay]: Endangered, 1,520 speakers; Sino-Tibetan; spoken in Nepal. Western Gurung (Gurung, Tamu Kyi, Gurnung) [gvr]: Threatened, 325,000 speakers; SinoTibetan; spoken in India and Nepal. Western Mewahang (Rai, Newang, Newahang) [raf]: Threatened, 4,650 speakers; SinoTibetan; spoken in Nepal. Yakha (Yakkha, Yakkhaba, Yakkhaba Cea) [ybh]: Endangered, 14,000 speakers; Sino-Tibetan; spoken in India and Nepal. Yamphu (Rai, Yamphu Rai, Yamphu Kha) [ybi]: Threatened, 1,722 speakers; Sino-Tibetan; spoken in Nepal. Yidgha (Yudgha, Yudga, Yidga) [ydg]: Threatened, fewer than 6,000 speakers; Indo-European; spoken in Pakistan. Yimchungru Naga (Yimchungrü, Yachimi, Yanchunger) [yim]: Vulnerable, 92,100 speakers; Sino-Tibetan; spoken in India. Zakhring (Charumba, Zaiwa, Meyor) [zkr]: Endangered, fewer than 1,000 speakers; SinoTibetan; spoken in India and China. Zangskari (Zanskari, Zaskari, Zangs-dkar) [zau]: Threatened, 5,000 speakers; Sino-Tibetan; spoken in Pakistan and India. Zeme Naga (Zeme, Empeo, Kachcha) [nzm]: Vulnerable, 34,100 speakers; Sino-Tibetan; spoken in India.

Southeast Asia Yen-ling Chen Linguistic diversity in Southeast Asia The Southeast Asia region in the Catalogue of Endangered Languages includes ten countries: Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand, and Vietnam. This region can be further divided into Mainland Southeast Asia – with Cambodia, Laos, Malaya of Malaysia, Myanmar, Thailand, and Vietnam – and Island Southeast Asia – with Brunei, Indonesia, Sabah and Sarawak of Malaysia, Singapore, and the Philippines. Indian culture had influence on many of the languages and cultures of the region. Centuries ago groups adapted and developed various Brahmi-based scripts as Indian influence expanded. The Catalogue of Endangered Languages lists 417 endangered languages identified in this region (6 of which are dormant, 411 still spoken), 23 of which have no ISO 639-3 codes (and are not included in Ethnologue). Ethnologue (Simons & Fennig 2017) lists 1488 languages for the region, 20 of which are extinct, so 1,468 languages still spoken, both endangered and nonendangered, in Southeast Asia. A majority of the endangered languages in this region belong to Tibeto-Burman (a large branch of the Sino-Tibetan family),32 Austronesian, Austroasiatic, and Tai-Kadai/Kra-Dai families, and others are “Papuan” languages.33 Several pidgins and creoles, language isolates, sign languages and unclassified languages are also included.

Causes of language endangerment in Southeast Asia The history of modern Southeast Asia is closely associated with colonialism. Some new national boundaries and independent states were created after World War II, and nationalism has played an important role in policies towards languages in these countries. These postcolonial 224

The world’s endangered languages

governments tend to be monarchies or to be ruled by the military, who tend to favor one ethnic group over the others. Minority groups that have different cultural or religious orientations are typically expected to accommodate. Instability in these newly established countries has in some cases resulted in political unrest or violent conflict. For unity, an official language34 was typically selected and designated to be a lingua franca to be shared by citizens. National language policy has favored one single language, such as Thai/Laotian in Thailand and Laos, Malay in Malaysia, Indonesia, and Brunei, Filipino (Tagalog) in the Philippines, Vietnamese in Vietnam, Khmer in Cambodia, and Burmese in Myanmar. The selection of one language over another has not necessarily been correlated with numbers of speakers. Preference often correlates with: (1) the political status of the language’s speakers; and (2) the social status of a language. For example, Malay/Bahasa Indonesia was chosen because it had been an important trade language in the region, although it was not widely spoken in today’s Indonesia before that country was founded. The bilingualism or multilingualism involving the official language(s) developed through the imposition of the official language in formal domains and in education. TV or radio programs also have promoted the official languages. While English, a global lingua franca, is employed as either an official language or a medium of instruction in higher education, the colonial languages French, Dutch, and Spanish, are no longer in use (Portuguese remains one of the official languages of East Timor, beside Tetum). Major regional languages, such as Javanese and Sundanese in Indonesia, Cebuano and Ilocano in the Philippines, and Jingpho and Shan in Myanmar, are prestigious in their home provinces and have managed to thrive, mostly in spoken forms. In contrast, other minority languages have gradually lost domains of use, with speakers shifting to official and regionally dominant languages. Other factors that have had a role in triggering language shift and/or endangerment include socio-economic opportunities, political instability (with riots, wars, land mines, etc.), and natural disasters. Transition from traditional lifestyles (chiefdoms, non-sedentary subsistence, etc.) to urbanization has often led to relocations and reform of tribal societies. Often this has weakened ties among speakers of certain languages, at the cost of ethnic-linguistic identities. As countries have improved their infrastructure, they have often forced relocations in order to build dams or roads or for resource extraction. Influxes of members of dominant groups into new or remote areas have had profound impacts on the indigenous languages and peoples. The Constitution of the Republic of the Philippines names two official languages – Filipino35 and English. Speakers of minority languages are usually bilingual/multilingual in strong regional languages, such as Bikol, Cebuano, Ilonggo/Hiligaynon, Ilocano, Kapampangan, and Pangasinan, with at least a working knowledge of Filipino/Tagalog (or English) through education. Before the arrival of speakers of Austronesian languages c.4,500 years ago, the Philippines were inhabited by “Negrito” people, physically different from the Austronesian (AN) groups. However, now all speak Austronesian languages with a non-Austronesian linguistic substratum (L. Reid 1987, 1994a, 1994b). The names of languages spoken by Negrito groups are often similar to “Agta,” “Arta,” “Ata,” etc. Although the Negrito groups have been linguistically assimilated to Austonesian, several still maintain a hunter-gatherer life-style, living in remote areas, whereas the non-Negrito speakers of Austronesian languages are typically agriculturalists and occupy the lowlands (Blust 2013). Given their small populations and their lifestyle, it is not surprising that their languages are challenged by urbanization and socio-economic-political pressures, with speakers shifting to dominant languages of the country. In recent years, the government of the Philippines has changed its English-and-Tagalog-only language policy, and has now decreed mother tongue-based multilingual education (MTB-MLE) in elementary 225

Anna Belew et al.

schools. However, the languages registered in the Catalogue of the Endangered Languages may very well not survive until any potential effects from the MTB-MLE program can be observed. Today’s Malaysia, Singapore, and Brunei were all under the rule of the British East India Company during the colonial era. Prior to that, the Portuguese and Dutch had both controlled the region. However, it should be acknowledged that several influential empires, such the Brunei Empire, the Srivijaya Empire, and the Majapahit Empire, existed before the arrival of the European colonial powers. During the Brunei Empire, Islam was brought to this region. Trading and cultural exchange between the South China Sea and the Indian Ocean also introduced Hinduism and Buddhism with associated material culture. Two creole or trade languages, Malaccan Creole Portuguese [mcm] and Baba Malay [mbf], developed out of contact around the Strait of Malacca. Most indigenous languages of this region are Austronesian; however, Malaya/Peninsular Malaysia also hosts some Austroasiatic languages, 12 of which are identified in the Catalogue as endangered. The majority of the population in Brunei and Malaysia speaks Malay. Given its speaker numbers and its function for centuries as a lingua franca for trade, it is of no surprise that Malay acquired official status after the establishment of Malaysia (in 1957), Singapore (in 1965), and Brunei (in 1984). Malay is the only official language in Malaysia and Brunei, and must be learned by every citizen. In contrast, Malay (i.e. Indonesian), together with English, Mandarin Chinese, and Tamil, are listed as official languages in the Constitution of Singapore. After the fall of the British East India Company, the economy of Malaysia, Singapore, and Brunei continued to thrive. The ability to speak an official language is the key to socio-economic opportunities in these countries. The creole/trade languages, Malaccan Creole Portuguese [mcm] and Baba Malay [mbf] (Lee 2014), in Peninsular Malaysia and Singapore have gradually lost their roles and have mostly been replaced by the official languages. Speakers of other languages, including Chinese, are bilingual in official languages through education. Speakers of certain languages, such as Belait [beg] in Brunei, and Lanoh [lnh] and Minriq [mnq] in Malaysia, began to shift to Malay varieties via bilingualism some decades ago. In addition to Malay, in Malaysian Borneo, Iban [iba] of Sarawak has also put pressure on minority languages, including Kanowit-Tanjong [kxn] and Remun [lkj]. The Orang Asli “original people” department, which administers indigenous peoples’ affairs in Malaysia does not make efforts to preserve languages and cultures of the indigenous groups (Bradley 2007: 388). Today both Brunei and Malaysia are Islamic states, where homogeneity is campaigned for through a shared religion and language. In recent years, infrastructure development, the construction of a dam in the Baram River basin, and log and palm oil exploitations in Malaysian Borneo have further threatened the indigenous peoples there, including the Dayak, Kenyah, Kayan, Kelabit, and Penan groups, as well as local flora and fauna like Orangutans, leading to both language endangerment and species extinction – the same holds true for Indonesian Borneo (see below). The islands of Java and Sumatra have served as important international ports of trade, the location of which also fostered the spread of Hindu, Buddhist, and Muslim religions and cultures, and the developments of Brahmi-based writing systems. Consequently, languages of Java and Sumatra have large numbers of speakers and enjoy higher social status and longer written traditions. However, the official language of the new sovereign state of Indonesia, founded in 1949, is Bahasa Indonesia,36 a dialect of Malay that had long been an important trading language and lingua franca of island Southeast Asia. In order to create a shared sense of belonging of the country, the Indonesian government strongly promoted the use of Indonesian in all domains. 226

The world’s endangered languages

Although Indonesian was not the language with the largest number of speakers when the country was established, it is now considered the dominant language of younger generations, with widespread bilingualism. Languages, ethnic groups, and cultures in West Papua (formerly Iryan Jaya, on the island of New Guinea) are much closer to those of Papua New Guinea than to the rest of Austronesiandominant and Islamic-dominant Indonesia. Nevertheless, after West Papua became part of Indonesia (annexed in 1969), immigrants from Java and Sumatra poured in; the forced use of Bahasa Indonesian and the negative attitude of immigrants toward the indigenous peoples have resulted in local languages becoming endangered. Instead of Standard Indonesian, the de facto lingua franca as well as dominant language of West Papua is Papuan Malay [pmy],37 which is widely spoken among the young Papuans as their L1, especially those living in the coastal regions (Kluge 2014:5–7; Scott et al. 2008:11). In addition to the socio-economic pressures, the government has changed political boundaries and cut regions up, which mixed people that once were isolated and increased the use of Papuan Malay. The level of linguistic vitality, in general, remains unclear in West Papua due to its political instability and lack of information. Vietnam has been heavily influenced by the Chinese languages, writing system, and culture for centuries. The landscape of western Vietnam, which is mountainous, blocked the spread of the Han Chinese impact on regions such as present day Cambodia, Thailand, Laos, and Malaysia that do not border China, as opposed to northern Vietnam where the Red River Delta provided easy access for the Chinese to enter. The Austroasiatic-speaking Kinh, who originated in today’s Northern Vietnam and southern China, conquered the Austronesianspeaking Champa kingdom, i.e. Central and Southern Vietnam, in the tenth century. The regime of Vietnam’s earlier inhabitants was replaced by the ancestors of present-day Vietnamesespeaking Kinh, even though small Cham polities survived until the nineteenth century when they were completely annexed by the Vietnamese. Speakers of the Chamic languages Bih, Chru [cje], Eastern Cham [cjm], Northern Roglai [rog], and Southern Roglai [rgs] are all bilingual in Vietnamese, with shrinking domains of use for their heritage languages. In addition to the Chamic languages, most of the endangered languages of Vietnam belong to the Austroasiatic or Tai-Kadai language families; most are Austroasiatic, cases such as Arem [aem], Hung [hnu], Iduh [tyh], and Maleng [pkt], spoken in hilly regions of Vietnam bordering Laos; Tai-Kadai languages there include Green Gelao [giq], Laha (Vietnam) [lha], and Laqua [laq] on the Vietnam plain close to the Vietnam-China border. The Constitution of Vietnam grants minorities the right to preserve their languages, cultures, and identities, and to develop their own Latin-based orthographies; however, minorities in Vietnam do not seem to be interested in language maintenance and mother-tongue education (Bradley 2007:389). The only exception may be the Phan Rang Cham/Eastern Cham [cjm] communities, who have developed teaching materials based on the traditional Cham scripts (Akhar Thrah scripts) and have maintained good intergenerational transmission of their language, although its grammar reflects Vietnamese influence due to widespread bilingualism in Vietnamese (Brunelle 2008). Cambodia declared its independence from France in 1954. The new sovereign state promoted the use of the Khmer language (spoken by the majority of its people) and the Brahmic-based writing system as the official orthography. However, Cambodia has suffered catastrophic and devastating consequences of civil disorder. How many languages and ethnic groups became extinct as a result of the Khmer Rouge is unknown. Somray [smu] is reported to as having been nearly completely extirpated during this period of time, with only two speakers left in the 1980s (Bradley 2007:416). Today, all the endangered languages of Cambodia listed in the 227

Anna Belew et al.

Catalogue belong to the Austroasiatic language family, and a majority of the minority groups are bilingual in Khmer. The government provides no support for maintenance of the languages of the minority groups (Bradley 2007:387). Even though some minority groups, such as Chung [scq] and Kaco’ [xkk], are able to maintain balanced bilingualism in their heritage languages and Khmer, languages such as Chong [cog], Pear [pcb], Samre [sxm], and Suoy [syo] are unlikely to survive for more generations. Before the arrival of the Tai from southern China (Guangxi and Yunnan) around the eighth to tenth centuries (Pittayaporn 2014), present-day Thailand, Laos, and their neighboring regions were inhabited by the Cham, Khmer, and Mon. Thailand was the only Southeast Asian sovereign state that was not colonized during the European incursions into Southeast Asia. Central Thai, spoken in Bangkok, was selected as the official language of Thailand, and was imposed via education. Minority groups, especially hunter-gatherers and sea nomads, were brought to Thai-speaking towns for education (Bradley 2007:387). As a result, bilingualism in Thai among the minorities is high. Some people later settled down in towns for economic opportunities. Minority groups, such as Mlabri [mra] and Lua’ [prb], may have managed to maintain linguistic equilibrium, regardless of their small populations. By contrast, Ugong [ugo], Thavung [thm], and Mpi [mpz] are no longer passed on to children. Documentation of minority languages by Thai scholars is in progress. It should be acknowledged that the drastic change of lifestyle (nomadic to sedentary) of Mlabri [mra], Urak Lawoi’ [urk], and Moklen [mkm] threatens the maintenance of their languages. The Constitution of Laos authorized the official status of Lao and its script. However, Lao is spoken by only half of the population as a first language, and more speakers of Lao dwell in northeastern Thailand (the Isan region) than in Laos.38 Even though the constitution granted language and culture rights, the Lao government does not support ethnic language maintenance for fear of creating a separate national identity (Bradley 2007:386). The majority of (potentially) endangered languages of Laos belong to the Austroasiatic language family, and are spoken along the borders with Vietnam, Thailand, and China. These linguistic communities are usually small in size, and speakers of these languages tend to shift to dominant languages such as Vietnamese, Khmer, Akha, and Lahu (China), and Lao via bilingualism, as exemplified by Phana’ [phq], Chepya [ycp], Bit [bgk], Sach [scb], and Tum. Myanmar (formerly Burma) remained isolated due to the rule of military regimes until into the 2010s. Although the Bamar are the dominant group in the country, they were not the earliest inhabitants. The Tibeto-Burman-speaking Pyu and the Austroasiatic-speaking Mon had arrived before the Bamar, and had been politically dominant before the ascendant Bamar. The Burmese language and culture has gradually gained in importance in Myanmar since the Bamar Kingdom of Pagan of the twelfth to thirteenth centuries. Although Pyu become extinct, the Mon remained influential in southern Myanmar from time to time. In the nineteenth century the Mon started to assimilate and become acculturated due to the influx of the Bamar into lower Irrawaddy. After gaining its independence in 1948, Burmese and its writing system were promoted as the official language and orthography in Myanmar. However, the termination of British colonial rule was followed by decades of militias and military regimes. Major ethnic groups such as the Karen, the Kachin, and the Shan of the north have been fighting for autonomy for centuries. No support is given for maintenance of the languages of minority groups, which “has already resulted in the Burmanisation of much of the population of the central plain and lower eastern hills, including many Mon, Shan and others.” (Bradley 2007:388). Armed conflicts have lasted for decades. In 2008, Cyclone Nargis killed more than 10,000 people in Myanmar. Although there are only 28 endangered languages listed in the Catalogue for Myanmar, most Tibeto-Burman, this might not reflect the actual state of language 228

The world’s endangered languages

endangerment in Myanmar because little documentation work was done while the country remained politically isolated. A majority of these endangered languages are also spoken in nearby countries, such as China, Thailand, and India, from which for the most part the status of each language is determined.

The current status of endangered languages in Southeast Asia As mentioned, the Catalogue of Endangered Languages has identified 419 endangered languages in Southeast Asia, with 22 language isolates, 13 sign languages, five pidgins and creoles (Baba Malay [mbf], Iha-Based Pidgin [ihb], Javindo [jvd], Malaccan Creole Portuguese [mcm], and Petjo [pey]), and 379 other endangered languages. Of these 419 languages listed in ELCat, six are dormant, 34 are “Critically Endangered,” and 52 “Severely Endangered”; 160 languages fall into the category labeled “Endangered,” 101 “Threatened,” and 60 “Vulnerable.” There are also a few languages the vitality of which remains unknown; these linguistic groups tend to have a small population or lower social status, for example Original Laos Sign Language [lso] and Mok [mqt]. Moreover, Iha-Based Pidgin [ihb] of Indonesian, which is only used in trade settings, is the only language that has no native speakers. Little recent documentation work has been done in politically unstable or war-ridden zones, specifically West Papua, southern Philippines, southern Thailand, Myanmar, and Cambodia. As our understanding of this part of the world progresses, more endangered languages can be expected to be uncovered beyond the ones currently identified in ELCat. Below is an alphabetical listing of all languages of Southeast Asia currently included in the Catalogue of Endangered Languages.

Endangered languages of Southeast Asia Abai Sungai [abf]: Endangered, fewer than 500 speakers; Austronesian; spoken in Malaysia. Abellen Ayta (Abenlen, Ayta Abenlen Sambal, Aburlin Negrito) [abp]: Threatened, 3,500 speakers; Austronesian; spoken in the Philippines. Abinomn (Avinomen, Foya) [bsa]: Endangered, fewer than 50 speakers; isolate; spoken in Indonesia. Abui (Barue, Namatalaki) [abz]: Threatened, 16,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. Abun (Karon Pantai, Madik, Yimbun) [kgr]: Threatened, 3,000 speakers; isolate; spoken in Indonesia. Adang (Alor) [adn]: Endangered, 7,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. Adasen Itneg (Adasen, Addasen Tinguian, Itneg) [tiu]: Threatened, 4,000 speakers; Austronesian; spoken in the Philippines. Afra (Usku) [ulf]: Endangered, 115 speakers; isolate; spoken in Indonesia. Alabat Island Agta (Alabat Island Dumagat) [dul]: Critically endangered, 50 speakers; Austronesian; spoken in the Philippines. Alune (Sapalewa, Patasiwa Alfoeren) [alp]: Vulnerable, 17,200 speakers; Austronesian; spoken in Indonesia. Amahai (Amahei) [amq]: Critically endangered, 50 speakers; Austronesian; spoken in Indonesia. Amarasi [aaz]: Vulnerable, 70,000 speakers; Austronesian; spoken in Indonesia. Ampibabo [no ISO 639-3 code]: Threatened, 6,000 speakers; Austronesian; spoken in Indonesia. 229

Anna Belew et al.

Anakalangu (Anakalang) [akg]: Vulnerable, at least 25,000 speakers; Austronesian; spoken in Indonesia. Andio (Masama, Andio’o, Imbao’o) [bzb]: Threatened, 1,700 speakers; Austronesian; spoken in Indonesia. Anong (Fuche Naw, 阿儂語, ) [nun]: Critically endangered, approximately 4,040 speakers; Sino-Tibetan; spoken in China and Myanmar. Anus (Koroernoes, Korur) [auq]: Endangered, 320 speakers; Austronesian; spoken in Indonesia. Aputai (Ilputih, Opotai, Tutunohan) [apx]: Endangered, 150 speakers; Austronesian; spoken in Indonesia. Arem (A-Rem, Chomrau, Chombrau) [aem]: Critically endangered, fewer than 100 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Arguni (Argoeni) [agf]: Endangered, 150 speakers; Austronesian; spoken in Indonesia. Arta [atz]: Critically endangered, 11 speakers; Austronesian; spoken in the Philippines. As [asz]: Endangered, 230 speakers; Austronesian; spoken in Indonesia. Ata [atm]: Critically endangered, four speakers; Austronesian; spoken in the Philippines. Ati (Inati, Bisaya) [atk]: Threatened, 1,500 speakers; Austronesian; spoken in the Philippines. Auye (Auwje) [auu]: Endangered, 350 speakers; Trans-New Guinea; spoken in Indonesia. Awbono (Kvolyab) [awh]: Vulnerable, approximately 100 speakers; Bayono-Awbono; spoken in Indonesia. Awera [awr]: Endangered, 100 speakers; Lakes Plain; spoken in Indonesia. Awyi (Auyi, Awye, Awje) [auw]: Endangered, 400 speakers; Border; spoken in Indonesia. Baba Malay (Chinese Malay, Baba, Straits Malay) [mbf]: Critically endangered, 2,000 speakers; pidgin or creole; spoken in Malaysia and Singapore. Bädi Kanum (Kanum, Enkelembu, Knwne) [khd]: Severely endangered, ten speakers; Morehead-Upper Maro; spoken in Indonesia. Bahonsuai [bsu]: Endangered, 200 speakers; Austronesian; spoken in Indonesia. Balaesang (Balaesan, Balaisang, Pajo) [bls]: Threatened, 3,200 speakers; Austronesian; spoken in Indonesia. Ban Khor Sign Language [bfk]: Endangered, at least 400 speakers; sign language; spoken in Thailand. Bangka (Lom, Belom, Mapor) [mfb]: Critically endangered, 50 speakers; Austronesian; spoken in Indonesia. Baras (Ende) [brs]: Endangered, 250 speakers; Austronesian; spoken in Indonesia. Bataan Ayta (Mariveles Ayta, Bataan Sambal) [ayt]: Endangered, 527 speakers; Austronesian; spoken in the Philippines. Batak (Babuyan, Tinitianes, Palawan Batak) [bya]: Endangered, approximately 200 speakers; Austronesian; spoken in the Philippines. Batek (Bateq, Bateg, Batok) [btq]: Endangered, approximately 960 speakers; Austro-Asiatic; spoken in Malaysia. Bati (Gah) [bvt]: Threatened, 3,500 speakers; Austronesian; spoken in Indonesia. Bayono [byl]: Vulnerable, approximately 100 speakers; Bayono-Awbono; spoken in Indonesia. Bedoanas [bed]: Endangered, 180 speakers; Austronesian; spoken in Indonesia. Belait (Balait Jati, Lemeting, Meting) [beg]: Severely endangered, 700 speakers; Austronesian; spoken in Malaysia and Brunei. Benggoi (Bengoi, Kobi-Benggoi, Uhei-Kaclakin) [bgy]: Endangered, 350 speakers; Austronesian; spoken in Indonesia. Besoa (Behoa) [bep]: Threatened, 8,000 speakers; Austronesian; spoken in Indonesia. Betaf (Tebi) [bfe]: Endangered, 600 speakers; Orya-Tor; spoken in Indonesia. 230

The world’s endangered languages

Biak (Biak-Numfor, Noefoor) [bhw]: Severely endangered, 70,000 speakers; Austronesian; spoken in Indonesia. Biga [bhc]: Endangered, 300 speakers; Austronesian; spoken in Indonesia. Bih [no ISO 639-3 code]: Critically endangered; Austronesian; spoken in Vietnam. Bintulu [bny]: Vulnerable; Austronesian; spoken in Malaysia. Biritai (Ati, Aliki, Biri) [bqq]: Endangered, 250 speakers; Lakes Plain; spoken in Indonesia. Bisu (Misu, 畢蘇語, 米蘇語) [bzi]: Threatened, fewer than 2,740 speakers; Sino-Tibetan; spoken in China, Myanmar, Thailand, and Laos. Bit (布興話, Kha Bit, Khabit) [bgk]: Severely endangered, fewer than 1,000 speakers; AustroAsiatic; spoken in China and Laos. Blagar (Belagar, Balagur, Kaera) [beu]: Threatened, fewer than 10,000 speakers; Timor-AlorPantar; spoken in Indonesia. Boano (Maluku, Indonesia) (Buano) [bzn]: Threatened, 3,240 speakers; Austronesian; spoken in Indonesia. Boano (Sulawesi, Indonesia) (Bolano, Djidja) [bzl]: Threatened, 2,350 speakers; Austronesian; spoken in Indonesia. Bobongko [bgb]: Threatened, 1,500 speakers; Austronesian; spoken in Indonesia. Bonerif (Beneraf) [bnv]: Endangered, 200 speakers; Orya-Tor; spoken in Indonesia. Bonggo (Armopa, Bgu, Bogu) [bpg]: Severely endangered, 432 speakers; Austronesian; spoken in Indonesia. Budong-Budong (Tangkou, Tongkou) [bdx]: Severely endangered, 70 speakers; Austronesian; spoken in Indonesia. Bulo Stieng (Stieng, Xtieng, Xa-Dieng) [sti]: Threatened; Austro-Asiatic; spoken in Vietnam and Cambodia. Burate [bti]: Endangered, 100 speakers; East Geelvink Bay; spoken in Indonesia. Burmeso (Taurap, Borumeso, Boromeso) [bzu]: Endangered, 250 speakers; isolate; spoken in Indonesia. Burumakok [aip]: Severely endangered, 40 speakers; Trans-New Guinea; spoken in Indonesia. Busoa (Bosoa) [bup]: Threatened, 2,300 speakers; Austronesian; spoken in Indonesia. Camarines Norte Agta (Manide, Agiyan) [abd]: Vulnerable, approximately 4,000 speakers; Austronesian; spoken in the Philippines. Cambodian Sign Language (CSL, CBDSL, Khmer Sign Language) [no ISO 639-3 code]: Endangered, 1500 speakers; sign language; spoken in Cambodia. Casiguran Dumagat Agta (Dumagat (Casiguran), Agta (Casiguran Dumagat), Casiguran Dumagat) [dgc]: Threatened, approximately 600 speakers; Austronesian; spoken in the Philippines. Central Cagayan Agta [agt]: Endangered, 780 speakers; Austronesian; spoken in the Philippines. Central Mnong (Pnong, Phong) [cmo]: Vulnerable, 52,500 speakers; Austro-Asiatic; spoken in Vietnam and Cambodia. Central Tagbanwa [tgt]: Endangered, fewer than 1,000 speakers; Austronesian; spoken in the Philippines. Chatong [no ISO 639-3 code]: Endangered, fewer than 580 speakers; Austro-Asiatic; spoken in Vietnam. Chepya [ycp]: Threatened, 2,000 speakers; Sino-Tibetan; spoken in Laos. Chewong (Cheq Wong, Che’wong, Siwang) [cwg]: Severely endangered, 400 speakers; Austro-Asiatic; spoken in Malaysia. 231

Anna Belew et al.

Chong (Chawng, Shong, Xong) [cog]: Severely endangered, approximately 500 speakers; Austro-Asiatic; spoken in Thailand and Cambodia. Chrau (Jro, Ro, Tamun) [crw]: Vulnerable, fewer than 15,000 speakers; Austro-Asiatic; spoken in Vietnam. Chru (Churu, Choru, Chu Ru) [cje]: Threatened, fewer than 11,000 speakers; Austronesian; spoken in Vietnam. Chu-ng (Sauch, Saotch, Saoch) [scq]: Threatened, 17 speakers; Austro-Asiatic; spoken in Thailand and Cambodia. Côông (Công) [cnc]: Severely endangered, 450 speakers; Sino-Tibetan; spoken in Vietnam. Cosao (Caosao, 搓梭語, tsho55 sɔ55) [no ISO 639-3 code]: Endangered; Sino-Tibetan; spoken in Laos and China. Dabe (Mangambilis) [dbe]: Endangered, 440 speakers; Orya-Tor; spoken in Indonesia. Dakka (Nordost-Celebes) [dkk]: Threatened, 1,500 speakers; Austronesian; spoken in Indonesia. Dakkang (Daak Kang) [no ISO 639-3 code]: Threatened, fewer than 1,198 speakers; AustroAsiatic; spoken in Laos. Damal (Uhunduni, Amung, Amung Kal) [uhn]: Threatened, 4,000–5,000 speakers; isolate; spoken in Indonesia. Dampelas (Dian, Dampal, Dampelasa) [dms]: Vulnerable, 10,300 speakers; Austronesian; spoken in Indonesia. Danau (Danaw, Kano’, Kano’ Chon) [dnu]: Endangered; Austro-Asiatic; spoken in Myanmar. Dao (Maniwo) [daz]: Endangered, 250 speakers; Trans-New Guinea; spoken in Indonesia. Dela-Oenale (Rote, Roti, Rotinese) [row]: Vulnerable, 7,000 speakers; Austronesian; spoken in Indonesia. Dem (Lem, Ndem) [dem]: Threatened, 1,000 speakers; isolate; spoken in Indonesia. Demisa [dei]: Endangered, 400 speakers; East Geelvink Bay; spoken in Indonesia. Dera (Dla, Dra, Kamberataro) [kbv]: Threatened, 1,200 speakers; Senagi; spoken in Indonesia and Papua New Guinea. Dhao (Ndao, Dao, Ndaonese) [nfa]: Threatened, approximately 3,000 speakers; Austronesian; spoken in Indonesia. Diuwe [diy]: Endangered, 100 speakers; Trans-New Guinea; spoken in Indonesia. Dondo [dok]: Threatened, approximately 13,000 speakers; Austronesian; spoken in Indonesia. Doutai (Taori-So, Taori, Tolitai) [tds]: Severely endangered, 70 speakers; Lakes Plain; spoken in Indonesia. Duano’ (Duano, Orang Kuala, Desin Dola’) [dup]: Endangered; Austronesian; spoken in Malaysia. Dubu (Tebi) [dmu]: Endangered, 130 speakers; Pauwasi; spoken in Indonesia. Dumpas (Doompas) [dmv]: Endangered, fewer than 1,100 speakers; Austronesian; spoken in Malaysia. Dupaninan Agta (Eastern Cagayan Agta, Dupaningan Agta) [duo]: Threatened, approximately 1,400 speakers; Austronesian; spoken in the Philippines. Duriankere (Esaro, Sailen, Duriankari) [dbn]: Severely endangered, 30 speakers; Trans-New Guinea; spoken in Indonesia. Dusner (Dusnir) [dsn]: Critically endangered, six speakers; Austronesian; spoken in Indonesia. Eastern Bru (Bruu, Brou) [bru]: Vulnerable, 82,300 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Eastern Cham (Phan Rang Cham) [cjm]: Vulnerable, 100,000 speakers; Austronesian; spoken in Vietnam. 232

The world’s endangered languages

Eastern Penan (Punan) [pez]: Threatened, fewer than 10,000 speakers; Austronesian; spoken in Indonesia, Malaysia, and Brunei. Elseng (Djanggu, Janggu, Malluo) [mrf]: Endangered, 300 speakers; isolate; spoken in Indonesia. Emplawas [emw]: Endangered, 250 speakers; Austronesian; spoken in Indonesia. Emumu (Emem, Imimkal, Kiamerop) [enr]: Threatened, approximately 1,100 speakers; Pauwasi; spoken in Indonesia. En (Nung Ven, 恩語, 儂環語) [enc]: Endangered, 200 speakers; Tai-Kadai; spoken in China and Vietnam. Enggano (Engganese) [eno]: Threatened, 1,000 speakers; Austronesian; spoken in Indonesia. Erokwanas [erw]: Endangered, 250 speakers; Austronesian; spoken in Indonesia. Faire Atta (Southern Atta) [azt]: Endangered, 400–550 speakers; Austronesian; spoken in the Philippines. Falam Chin (Hallam Chin, ) [cfm]: Vulnerable, 107,300 speakers; Sino-Tibetan; spoken in Myanmar and India. Fayu (Sehudate) [fau]: Threatened, 400 speakers; Lakes Plain; spoken in Indonesia. Foau (Doa) [flh]: Endangered, 230 speakers; Lakes Plain; spoken in Indonesia. Gana (Ganaq, Gana’, Minansut) [gnq]: Endangered, 1,000 speakers; Austronesian; spoken in Malaysia. Gebe (Gebi) [gei]: Threatened, 2,650 speakers; Austronesian; spoken in Indonesia. Gorap [goq]: Endangered, 1,000 speakers; Austronesian; spoken in Indonesia. Hai Phong Sign Language (ngôn ngữ ký hiệu Hải Phòng, HPSL) [haf]: Endangered, approximately 1,800 speakers; sign language; spoken in Vietnam. Hanoi Sign Language (ngôn ngữ ký hiệu Hà Nội, Ha Noi Sign Language) [hab]: Endangered, 39,000 speakers; sign language; spoken in Vietnam. Hatam (Hattam, Borai, Mansim) [had]: Vulnerable, 16,000 speakers; isolate; spoken in Indonesia. Hawu (Sabu, Havunese, Savu) [hvn]: At risk, 100,000 speakers; Austronesian; spoken in Indonesia. Helong (Helon, Semau, Kupang) [heg]: Vulnerable, 14,000 speakers; Austronesian; spoken in Indonesia. Ho Chi Minh City Sign Language (ngôn ngữ ký hiệu thành phố Hồ Chí Minh, Southern Vietnamese Sign Language) [hos]: Endangered, approximately 45,000 speakers; sign language; spoken in Vietnam. Hoti [hti]: Critically endangered, ten speakers; Austronesian; spoken in Indonesia. Hpon (Hpön, Phun, Phön) [hpo]: Dormant; Sino-Tibetan; formerly spoken in Myanmar. Hukumina (Bambaa) [huw]: Critically endangered, one speaker; Austronesian; spoken in Indonesia. Hulung [huk]: Critically endangered, ten speakers; Austronesian; spoken in Indonesia. Hung (Cuói, K’katiam-Pong-Houk, Pheng) [hnu]: Endangered, fewer than 500 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Ibu [ibu]: Severely endangered, 35 speakers; North Halmahera; spoken in Indonesia. Iduh (O Du, O’du, ‘Iduh) [tyh]: Severely endangered, approximately 30 speakers; AustroAsiatic; spoken in Vietnam and Laos. Iha Based Pidgin [ihb]: No estimate available; pidgin or creole; spoken in Indonesia. Iraya [iry]: Vulnerable, 10,000 speakers; Austronesian; spoken in the Philippines. Iresim (Iroquois, Beduba, Yerisiam) [ire]: Severely endangered, 70 speakers; Austronesian; spoken in Indonesia. 233

Anna Belew et al.

Isarog Agta (Bikol) [agk]: Critically endangered, six speakers; Austronesian; spoken in the Philippines. Isinay (Isinai, Inmeas, Malaates) [inn]: Critically endangered; Austronesian; spoken in the Philippines. Isirawa (Saberi, Saweri, Okwasar) [srl]: Threatened, 1,800 speakers; unclassified; spoken in Indonesia. Itik (Ittik, Betef, Ittik-Tor) [itx]: Severely endangered, 80 speakers; Orya-Tor; spoken in Indonesia. Iwur (Dintere, Iwoer, Iwur) [iwo]: Threatened, 1,000 speakers; Trans-New Guinea; spoken in Indonesia. Jah Hut (Jah Het, Eastern Sakai, Ost-Sakai) [jah]: Threatened; Austro-Asiatic; spoken in Malaysia. Jahai (Jehai, Pangan, Jehehr) [jhi]: Threatened; Austro-Asiatic; spoken in Malaysia. Jakarta Sign Language [no ISO 639-3 code]: Threatened, 4,800–9,600 speakers; sign language; spoken in Indonesia. Jakun (Jaku’d, Jakud’n, Jakoon) [jak]: Vulnerable, 10,000 speakers; Austronesian; spoken in Malaysia. Javindo (Krontjong) [jvd]: Critically endangered; pidgin or creole; spoken in Indonesia. Jejara (Para, Para Naga) [pzn]: Vulnerable, 1,460 speakers; Sino-Tibetan; spoken in Myanmar. Juk (Suai) [no ISO 639-3 code]: Threatened, approximately 1,500 speakers; Austro-Asiatic; spoken in Laos. Kaco’ (Kachah’, Kacho’, Kachok) [xkk]: Vulnerable, 3,365 speakers; Austro-Asiatic; spoken in Cambodia. Kadai [kzd]: Endangered, 350 speakers; Austronesian; spoken in Indonesia. Kadu (Kado, ) [zkd]: Vulnerable, 30,000 speakers; Sino-Tibetan; spoken in Myanmar. Kaera (Dorit) [jka]: Vulnerable, approximately 5,500 speakers; Timor-Alor-Pantar; spoken in Indonesia. Kaibobo (Kaibubu) [kzb]: Endangered, 500 speakers; Austronesian; spoken in Indonesia. Kaiy (Taori-Kei, Taori-Kaiy, Kai) [tcq]: Endangered, 220 speakers; Lakes Plain; spoken in Indonesia. Kalao (Kalaotoa) [kly]: Endangered, 500 speakers; Austronesian; spoken in Indonesia. Kamang (Woisika, Waisika, Kamana-Kamang) [woi]: Threatened, approximately 6,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. Kamarian (Kamariang, Seruawan) [kzx]: Critically endangered, ten speakers; Austronesian; spoken in Indonesia. Kanan ( , Kado) [zkn]: Threatened, approximately 9,000 speakers; Sino-Tibetan; spoken in Myanmar. Kanowit-Tanjong (Kanowit) [kxn]: Endangered, fewer than 100 speakers; Austronesian; spoken in Malaysia. Kao (Kau, Ka’u) [kax]: Endangered, 400 speakers; North Halmahera; spoken in Indonesia. Kapori (Kapauri) [khp]: Vulnerable, approximately 200 speakers; isolate; spoken in Indonesia. Karas [kgv]: Endangered, 200 speakers; Trans-New Guinea; spoken in Indonesia. Kasong (Song, Chong of Trat) [no ISO 639-3 code]: Severely endangered, 40 speakers; AustroAsiatic; spoken in Thailand. Kata Kolok (Benkala Sign Language, Bengkala Sign Language, Balinese Sign Language) [bqy]: Endangered, approximately 175 speakers; sign language; spoken in Indonesia. Kaure (Kaureh, Kaurne) [bpp]: Endangered, 450 speakers; Kaure; spoken in Indonesia. 234

The world’s endangered languages

Kayeli (Kajeli, Cajeli, Caeli) [kzl]: Critically endangered, three speakers; Austronesian; spoken in Indonesia. Kayupulau (Kajupulau) [kzu]: Severely endangered, 573 speakers; Austronesian; spoken in Indonesia. Keder [kdy]: Endangered, 370 speakers; Orya-Tor; spoken in Indonesia. Kehu [khh]: Severely endangered, 25 speakers; isolate; spoken in Indonesia. Kelon (Kelong, Kalong, Klon) [kyo]: Threatened, 5,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. Kembra [xkw]: Severely endangered, 20 speakers; unclassified; spoken in Indonesia. Keningau Murut (Central Murut) [kxi]: Threatened, 7,000 speakers; Austronesian; spoken in Malaysia. Kensiu (Mos, Mengo, Tiong) [kns]: Endangered, 500 speakers; Austro-Asiatic; spoken in Thailand and Malaysia. Kepo’ (Kepoq, Kepo) [kuk]: Threatened, 6,000 speakers; Austronesian; spoken in Indonesia. Ketum [ktt]: Endangered, 900 speakers; Trans-New Guinea; spoken in Indonesia. Khamti (Hkamti, Khampti, Khamti Shan) [kht]: Vulnerable, 13,000 speakers; Tai-Kadai; spoken in India and Myanmar. Kháng (Khaang, Tayhay, Tay Hay) [kjm]: Threatened, 3,921 speakers; Austro-Asiatic; spoken in Vietnam. Kimki (Aipki, Kimgi, Sukubatom) [sbt]: Threatened, 500 speakers; isolate; spoken in Indonesia. Kintaq (Kenta, Kintak, Kintk) [knq]: Endangered; Austro-Asiatic; spoken in Thailand and Malaysia. Kiput [kyi]: Endangered, 450 speakers; Austronesian; spoken in Malaysia. Kirikiri (Kirira) [kiy]: Endangered, 250 speakers; Lakes Plain; spoken in Indonesia. Kodeoha (Kondeha) [vko]: Threatened, 1,500 speakers; Austronesian; spoken in Indonesia. Kofei [kpi]: Severely endangered, 100 speakers; East Geelvink Bay; spoken in Indonesia. Koneraw (Konorau) [kdw]: Threatened, 1,200 speakers; Trans-New Guinea; spoken in Indonesia. Kopkaka (Kopka) [opk]: Endangered, 400 speakers; Trans-New Guinea; spoken in Indonesia. Koroni [xkq]: Endangered, 500 speakers; Austronesian; spoken in Indonesia. Korowai (Kolufaup) [khe]: Threatened, 3,500 speakers; Trans-New Guinea; spoken in Indonesia. Kosadle (Kosare) [kiq]: Endangered, 250 speakers; isolate; spoken in Indonesia. Kowiai (Koiwai, Kaiwai, Kuiwai) [kwh]: Endangered, 600 speakers; Austronesian; spoken in Indonesia. Kraol (Southern-Central Mnong, Chrau, Crau) [rka]: Vulnerable, 1,960 speakers; Austro-Asiatic; spoken in Cambodia. Kravet (Kowet, Khvek, Kavet) [krv]: Threatened, 2,380 speakers; Austro-Asiatic; spoken in Cambodia. Kri [no ISO 639-3 code]: Severely endangered, approximately 250 speakers; Austro-Asiatic; spoken in Laos. Kru’ng 2 (Krueng, Kru’ng, Krung) [krr]: Vulnerable, 18,400 speakers; Austro-Asiatic; spoken in Cambodia. Kui (Lerabain, Masin-Lak) [kvd]: Threatened, 4,240 speakers; Timor-Alor-Pantar; spoken in Indonesia. Kula (Lantoka, Lamtoka, Kola) [tpg]: Vulnerable, 5,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. 235

Anna Belew et al.

Kumbewaha (Umbewaha) [xks]: Threatened, 3,400 speakers; Austronesian; spoken in Indonesia. Kwer [kwr]: Severely endangered, 100 speakers; Trans-New Guinea; spoken in Indonesia. Kwerisa (Taogwe) [kkb]: Severely endangered, 32 speakers; Lakes Plain; spoken in Indonesia. Laboya (Lamboya) [lmy]: Threatened, at least 23,000 speakers; Austronesian; spoken in Indonesia. Lachi (Lipulio, Y To, Y Pí) [lbt]: Threatened, 5,000 speakers; Tai-Kadai; spoken in Vietnam. Laghuu (Yi, Laopa, Xá Phó) [lgh]: Endangered, fewer than 1,000 speakers; Sino-Tibetan; spoken in Vietnam. Laha (Central Ambon) [lhh]: Threatened, 3,890 speakers; Austronesian; spoken in Indonesia. Laha (Xá Khao, Khlá Phlao, Klá Dong) [lha]: Severely endangered; Tai-Kadai; spoken in Vietnam. Laiyolo (Da’ang, Barang-Barang) [lji]: Endangered, 800 speakers; Austronesian; spoken in Indonesia. Lanoh (Lanoh Jengjeng, Jengjeng) [lnh]: Endangered, approximately 350 speakers; AustroAsiatic; spoken in Malaysia. Laqua (Qabiao, 普標語) [laq]: Severely endangered, 432 speakers; Tai-Kadai; spoken in China and Vietnam. Lauje (Laudje, Tinombo, Ampibabo-Lauje) [law]: Threatened, 38,000 speakers; Austronesian; spoken in Indonesia. Lave (Brao, Braou, Brau) [brb]: Vulnerable, 32,230 speakers; Austro-Asiatic; spoken in Vietnam, Laos, and Cambodia. Lawa (Wa, Wa Proper, Lava) [lcp]: Vulnerable; Austro-Asiatic; spoken in Thailand. Lebo’ Vo’ Kenyah (Kenyah, Long San Kenyah) [no ISO 639-3 code]: Endangered; Austronesian; spoken in Malaysia. Legenyem (Laganyan) [lcc]: Endangered, 250 speakers; Austronesian; spoken in Indonesia. Lelak [llk]: Endangered, 220 speakers; Austronesian; spoken in Malaysia. Lemolang (Baebunta) [ley]: Threatened, 2,000 speakers; Austronesian; spoken in Indonesia. Lengilu [lgi]: Critically endangered, four speakers; Austronesian; spoken in Indonesia. Lepki [lpe]: Endangered, 530 speakers; unclassified; spoken in Indonesia. Liabuku (Liabuka) [lix]: Severely endangered, 75 speakers; Austronesian; spoken in Indonesia. Liki (Moar) [lio]: Severely endangered, 11 speakers; Austronesian; spoken in Indonesia. Lisela (Buru, North Buru, Li Enyorot) [lcl]: Vulnerable, 11,900 speakers; Austronesian; spoken in Indonesia. Lola (Aru) [lcd]: Endangered, 830 speakers; Austronesian; spoken in Indonesia. Lolak [llq]: Severely endangered; Austronesian; spoken in Indonesia. Lole (Rote, Roti, Rotinese) [llg]: Vulnerable, 20,000 speakers; Austronesian; spoken in Indonesia. Loun [lox]: Dormant; Austronesian; formerly spoken in Indonesia. Lua’ (East Pua Pray, Pray) [prb]: Threatened, 6,281 speakers; Austro-Asiatic; spoken in Thailand. Luang (Letri Lgona, Literi Lagona, Lgona) [lex]: Vulnerable, 18,000 speakers; Austronesian; spoken in Indonesia. Mah Meri (Besisi, Cellate, Southern Sakai) [mhe]: Threatened, 2,896 speakers; AustroAsiatic; spoken in Malaysia. Maklew (Makleu) [mgf]: Endangered, 120 speakers; Bulaka River; spoken in Indonesia. Malaccan Creole Portuguese (Malaysian Creole Portuguese, Malaccan, Papia Kristang) [mcm]: Endangered, approximately 2,150 speakers; pidgin or creole; spoken in Malaysia. 236

The world’s endangered languages

Maleng (Malieng, Malang, Kri) [pkt]: Endangered, approximately 500 speakers; AustroAsiatic; spoken in Vietnam and Laos. Mamboru (Mamboro, Memboro) [mvd]: Endangered, at least 13,000 speakers; Austronesian; spoken in Indonesia. Mander [mqr]: Severely endangered, 20 speakers; Orya-Tor; spoken in Indonesia. Mang (Mang U, 莽语, Xá Ó ) [zng]: Severely endangered, fewer than 1,750 speakers; AustroAsiatic; spoken in Vietnam and China. Maniq (Mos, Tonga, Tonga’) [tnz]: Endangered, approximately 300 speakers; Austro-Asiatic; spoken in Thailand and Malaysia. Mansaka (Mandaya Mansaka) [msk]: Vulnerable, 57,800 speakers; Austronesian; spoken in the Philippines. Mapia (Mapian) [mpy]: Dormant; Austronesian; formerly spoken in Indonesia. Maremgi (Marengge) [mrx]: Severely endangered, 55 speakers; Orya-Tor; spoken in Indonesia. Marori (Morori, Moraori, Moaraeri) [mok]: Endangered, fewer than 150 speakers; isolate; spoken in Indonesia. Masep (Massep, Wotaf, Potafa) [mvs]: Severely endangered, 25 speakers; isolate; spoken in Indonesia. Masimasi [ism]: Severely endangered, ten speakers; Austronesian; spoken in Indonesia. Mawes [mgk]: Endangered, 693 speakers; isolate; spoken in Indonesia. May [no ISO 639-3 code]: Endangered, approximately 500 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Meoswar (War) [mvx]: Endangered, 250 speakers; Austronesian; spoken in Indonesia. Mer (Muri, Miere) [mnu]: Severely endangered, 85 speakers; Trans-New Guinea; spoken in Indonesia. Minriq (Menriq, Menrik, Mendriq) [mnq]: Endangered, approximately 145 speakers; AustroAsiatic; spoken in Malaysia. Mintil (Mitil) [mzt]: Severely endangered, 40 speakers; Austro-Asiatic; spoken in Malaysia. Mlabri (Mla, Mla Bri) [mra]: Threatened, fewer than 400 speakers; Austro-Asiatic; spoken in Thailand and Laos. Mlap (Kwansu, Kwansu-Bonggrang, Kuangsu-Bonggrang) [kja]: Endangered, 300 speakers; Nimboran; spoken in Indonesia. Moi (West Makian, Makian Luar) [mqs]: Vulnerable, approximately 12,000 speakers; North Halmahera; spoken in Indonesia. Mok (Amok, Hsen-Hsum, Yā-āng Lawa) [mqt]: No estimate available; Austro-Asiatic; spoken in Myanmar, China, Thailand, and Laos. Moken (Selung, Mawken, Basing) [mwt]: Endangered; Austronesian; spoken in Myanmar and Thailand. Moklen (Chau Pok) [mkm]: Endangered; Austronesian; spoken in Thailand. Mombum (Komelom, Kemelom, Kemelomsch) [mso]: Endangered, 250 speakers; Trans-New Guinea; spoken in Indonesia. Momina [mmb]: Endangered, 200 speakers; Trans-New Guinea; spoken in Indonesia. Momuna (Somahai, Somage, Sumohai) [mqf]: Threatened, 2,000 speakers; Trans-New Guinea; spoken in Indonesia. Mor (Bomberai Peninsula, Indonesia) (Mor2, Mor of Bomberai) [moq]: Severely endangered, 60 speakers; isolate; spoken in Indonesia. Mor (Mor Islands, Indonesia) (Austronesian Mor) [mhz]: Endangered, 700 speakers; Austronesian; spoken in Indonesia. 237

Anna Belew et al.

Mount Iraya Agta (Mt. Iraya Agta, Bikol, Inagta of Mt. Iraya) [atl]: Critically endangered, few, if any speakers; Austronesian; spoken in the Philippines. Mpi (Mpi-Mi) [mpz]: Endangered, 1,200 speakers; Sino-Tibetan; spoken in Thailand. Mpur [akc]: Threatened, 7,000 speakers; isolate; spoken in Indonesia. Naka’ela [nae]: Critically endangered, five speakers; Austronesian; spoken in Indonesia. Nakai (Na’ai, Nagai, Na’i) [nkj]: Endangered, a few hundred speakers; Trans-New Guinea; spoken in Indonesia. Namla [naa]: Severely endangered, 36 speakers; Namla-Tofanma; spoken in Indonesia. Napu (Pekurehua) [npy]: Threatened, 6,000 speakers; Austronesian; spoken in Indonesia. Narau [nxu]: Severely endangered, 85 speakers; Kaure-Narau; spoken in Indonesia. Nasal [nsy]: Vulnerable, approximately 3,000 speakers; isolate; spoken in Indonesia. Nedebang (Klamu) [nec]: Critically endangered, approximately 200 speakers; Timor-AlorPantar; spoken in Indonesia. Ngkâlmpw Kanum (Kanum, Enkelembu, Knwne) [kcd]: Endangered, 150 speakers; MoreheadUpper Maro; spoken in Indonesia. Nguôn (Ngouan) [nuo]: Endangered, approximately 1,000 speakers; Austro-Asiatic; spoken in Vietnam and Laos. North Muyu (Kataut, Kati Ninanti, Kati-Ninanti) [kti]: Threatened, 8,000 speakers; TransNew Guinea; spoken in Indonesia. Northern Alta (Edimala, Baler Negrito, Ditaylin Alta) [aqn]: Endangered, 240 speakers; Austronesian; spoken in the Philippines. Northern Roglai (Radlai, Adlai) [rog]: Vulnerable, 25,000 speakers; Austronesian; spoken in Vietnam. Nusa Laut (Nusalaut) [nul]: Severely endangered, ten speakers; Austronesian; spoken in Indonesia. Nyahkur (Tha Pong, Nyah Kur, Nyakur) [cbn]: Endangered, approximately 10,000 speakers; Austro-Asiatic; spoken in Thailand. Nyeu (Yeu, Yoe) [nyl]: Endangered, 200 speakers; Austro-Asiatic; spoken in Thailand. Obokuitai (Obogwitai, Ati, Aliki) [afz]: Endangered, 120 speakers; Lakes Plain; spoken in Indonesia. Oirata (Maaro) [oia]: Threatened, 1,220 speakers; Timor-Alor-Pantar; spoken in Indonesia. Onin (Onim, Sepa) [oni]: Endangered, 600 speakers; Austronesian; spoken in Indonesia. Orang Kanaq [orn]: Severely endangered, approximately 87 speakers; Austronesian; spoken in Malaysia. Orang Seletar [ors]: Endangered, 880 speakers; Austronesian; spoken in Malaysia and Singapore. Original Bangkok Sign Language (Old Bangkok Sign Language) [no ISO 639-3 code]: Critically endangered, approximately 495 speakers; sign language; spoken in Thailand. Original Chiangmai Sign Language (Old Chiangmai Sign Language) [csd]: Critically endangered, approximately 19 speakers; sign language; spoken in Thailand. Original Laos Sign Language [lso]: No estimate available; sign language; spoken in Laos. Ormu [orz]: Endangered, 600 speakers; Austronesian; spoken in Indonesia. Pa Di (Padi) [pdi]: Threatened; Tai-Kadai; spoken in China and Vietnam. Padoe (South Mori, Padoé, Alalao) [pdo]: Threatened, 6,000 speakers; Austronesian; spoken in Indonesia. Palu’e (Palue, Lu’a, Paluqe) [ple]: Threatened, 10,000 speakers; Austronesian; spoken in Indonesia. 238

The world’s endangered languages

Panasuan (To Panasean, To Pamosean) [psn]: Endangered, 800 speakers; Austronesian; spoken in Indonesia. Papuma [ppm]: Endangered, 600 speakers; Austronesian; spoken in Indonesia. Paulohi [plh]: Severely endangered, 50 speakers; Austronesian; spoken in Indonesia. Pear (Por, Kompong Thom) [pcb]: Severely endangered; Austro-Asiatic; spoken in Cambodia. Penang Sign Language [psg]: Endangered, fewer than 1,000 speakers; sign language; spoken in Malaysia. Pendau (Ndau, Ndaoe, Umalasa) [ums]: Threatened, 3,200 speakers; Austronesian; spoken in Indonesia. Petjo (Petjoh, Pecok) [pey]: Critically endangered, a few speakers; pidgin or creole; spoken in Indonesia. Phalok (Mae Rim Lawa, Phalo) [no ISO 639-3 code]: Dormant; Austro-Asiatic; formerly spoken in Thailand. Phong (Vietic Phong) [no ISO 639-3 code]: Endangered; Austro-Asiatic; spoken in Laos. Phong-Kniang (Pong 3, Khaniang, Kenieng) [pnx]: Threatened, 1,000 speakers; AustroAsiatic; spoken in Laos. Phonsung [no ISO 639-3 code]: Endangered, approximately 250 speakers; Austro-Asiatic; spoken in Laos. Phula (Phù Lá, Phu Khla, Fu Khla) [phh]: Vulnerable, 10,000 speakers; Sino-Tibetan; spoken in Vietnam and China. Phunoi (Phounoy, Phu Noi, Punoi) [pho]: Vulnerable, fewer than 23,618 speakers; Sino-Tibetan; spoken in Laos. Piru (Pirupiru) [ppr]: Severely endangered, ten speakers; Austronesian; spoken in Indonesia. Pong (Phong, Poong, Pong 1) [no ISO 639-3 code]: Endangered, approximately 1,500 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Powle-Ma (Molof, Ampas) [msl]: Endangered, 200 speakers; isolate; spoken in Indonesia. Punan Batu [pnm]: Severely endangered, 30 speakers; Austronesian; spoken in Malaysia. Punan Merah [puf]: Endangered, 140 speakers; Austronesian; spoken in Indonesia. Punan Merap [puc]: Endangered, 200 speakers; Austronesian; spoken in Indonesia. Rahambuu (Wiau, Wiaoe) [raz]: Threatened, 5,000 speakers; Austronesian; spoken in Indonesia. Rajong (Razong) [rjg]: Threatened, 6,000 speakers; Austronesian; spoken in Indonesia. Rasawa [rac]: Endangered, 200 speakers; Lakes Plain; spoken in Indonesia. Ratagnon (Datagnon, Latagnun, Latan) [btn]: Critically endangered; Austronesian; spoken in the Philippines. Ratahan (Toratán, Bentenan, Pasan) [rth]: Endangered; Austronesian; spoken in Indonesia. Rembong [reb]: Threatened, 5,000 speakers; Austronesian; spoken in Indonesia. Remun (Milikin, Millikin) [lkj]: Vulnerable, 3,542 speakers; Austronesian; spoken in Malaysia. Riang (Black Karen, Riang-Lang, Black Yang) [ril]: Vulnerable, 15,500 speakers; AustroAsiatic; spoken in China and Myanmar. Rikou (Ringgou, Rote, Roti) [rgu]: Vulnerable, 12,000 speakers; Austronesian; spoken in Indonesia. Romam (Lamam) [rmx]: Endangered, 250 speakers; Austro-Asiatic; spoken in Vietnam. Rongga [ror]: Threatened, 4,000 speakers; Austronesian; spoken in Indonesia. Ruc (Kha Mu Gia, Tac Cui) [no ISO 639-3 code]: Endangered, approximately 250 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Sabüm [sbo]: Dormant; Austro-Asiatic; formerly spoken in Malaysia. 239

Anna Belew et al.

Sach (Chut, Salang, Ruc) [scb]: Endangered, approximately 500 speakers; Austro-Asiatic; spoken in Vietnam and Laos. Saek (Sek, Tai Sek, Set) [skb]: Threatened, 10,000 speakers; Tai-Kadai; spoken in Thailand and Laos. Salas (Liambata, Lenkaitahe, Salas Gunung) [sgu]: Severely endangered, 50 speakers; Austronesian; spoken in Indonesia. Saluan (Loinang) [loe]: Vulnerable, 74,000 speakers; Austronesian; spoken in Indonesia. Samre (East Pear) [sxm]: Critically endangered, 20–30 speakers; Austro-Asiatic; spoken in Thailand and Cambodia. Saparua [spr]: Vulnerable, 10,200 speakers; Austronesian; spoken in Indonesia. Saponi [spi]: Critically endangered, four speakers; Lakes Plain; spoken in Indonesia. Sara (Riok, Sara Bakati’) [sre]: Threatened, 4,000 speakers; Austronesian; spoken in Indonesia. Sauri [srt]: Endangered, 300 speakers; East Geelvink Bay; spoken in Indonesia. Sause (Seuce) [sao]: Endangered, 300 speakers; isolate; spoken in Indonesia. Sawila (Manata, Wullili, Sawali) [swt]: Threatened, approximately 3,000 speakers; Timor-AlorPantar; spoken in Indonesia. Selangor Sign Language (KLSL, Kuala Lumpur Sign Language) [kgi]: Endangered, fewer than 500 speakers; sign language; spoken in Malaysia. Semai (Senoi, Sengoi, Sakai) [sea]: Vulnerable, fewer than 29,000 speakers; Austro-Asiatic; spoken in Malaysia. Semaq Beri (Semaq Bri, Semoq Beri, Ulu Tembeling) [szc]: Threatened, fewer than 2,488 speakers; Austro-Asiatic; spoken in Malaysia. Semelai (Bera, Semelai Bera, Serting) [sza]: Vulnerable, 4,103 speakers; Austro-Asiatic; spoken in Malaysia. Senggi (Sengi, Viid) [snu]: Endangered, 120 speakers; Border; spoken in Indonesia. Sepa (Indonesia) [spb]: Threatened, 2,600 speakers; Austronesian; spoken in Indonesia. Serili [sve]: Endangered, 330 speakers; Austronesian; spoken in Indonesia. Sian (Sihan) [spg]: Severely endangered, 50 speakers; Austronesian; spoken in Malaysia. Sila (Sida, Asong, Kha Pai) [slt]: Threatened, 2,118 speakers; Sino-Tibetan; spoken in Vietnam and Laos. Skou (Sekol, Sekou, Səkou) [skv]: Threatened, fewer than 700 speakers; Sko; spoken in Indonesia. Smärky Kanum [kxq]: Severely endangered, 80 speakers; Morehead-Upper Maro; spoken in Indonesia. So’a (Soa) [ssq]: Vulnerable, 10,000 speakers; Austronesian; spoken in Indonesia. Sobei (Biga, Imasi, Liki) [sob]: Endangered, 1,850 speakers; Austronesian; spoken in Indonesia. Somray (Samray, Samrai, Samre) [smu]: Critically endangered, two speakers; Austro-Asiatic; spoken in Cambodia. Sorsogon Ayta [ays]: Severely endangered, 18 speakers; Austronesian; spoken in the Philippines. Sota Kanum (Kanum, Enkelembu, Knwne) [krz]: Severely endangered, 100 speakers; Morehead-Upper Maro; spoken in Indonesia. South Muyu (Digoel, Digoeleesch, Digul) [kts]: Threatened, 4,000 speakers; Trans-New Guinea; spoken in Indonesia. Southern Alta (Baluga, Ita, Kabulowan) [agy]: Endangered, 1,000 speakers; Austronesian; spoken in the Philippines. Southern Roglai (Rai, Ríoglai) [rgs]: Vulnerable, 20,000 speakers; Austronesian; spoken in Vietnam. 240

The world’s endangered languages

Suabo (Suabau, Inanwatan, Mirabo) [szp]: Critically endangered, 800 speakers; Trans-New Guinea; spoken in Indonesia. Suoy (Su’ung, Pear) [syo]: Endangered, fewer than 500 speakers; Austro-Asiatic; spoken in Cambodia. Swoeng (Lavi) [no ISO 639-3 code]: Endangered, approximately 492 speakers; Austro-Asiatic; spoken in Laos. Taba (East Makian, Makian Dalam) [mky]: Endangered, 30,000–40,000 speakers; Austronesian; spoken in Indonesia. Tai Daeng (Red Tai, Red Thai, Tai Rouge) [tyr]: Threatened, 80,000 speakers; Tai-Kadai; spoken in Vietnam and Laos. Tai Loi (Loi, Muak Sa-aak, Wakut) [tlq]: Vulnerable, 4,460 speakers; Austro-Asiatic; spoken in Myanmar and Laos. Tai Neua [tdd]: Threatened, 70,000 speakers; Tai-Kadai; spoken in Laos. Taikat (Tajkat, Arso, Abrap) [aos]: Endangered, fewer than 600 speakers; Border; spoken in Indonesia. Taje (Petapa) [pee]: Critically endangered, 1–10 speakers; Austronesian; spoken in Indonesia. Tajio (Kasimbar, Tadjio, Ta’adjio) [tdj]: Threatened, 12,000 speakers; Austronesian; spoken in Indonesia. Taloki (Taluki) [tlk]: Endangered, 500 speakers; Austronesian; spoken in Indonesia. Talondo’ (Nordost-Celebes) [tln]: Endangered, 400 speakers; Austronesian; spoken in Indonesia. Tanahmerah (Sumeri) [tcm]: Endangered, 500 speakers; isolate; spoken in Indonesia. Tandia [tni]: Critically endangered, two speakers; Austronesian; spoken in Indonesia. Tangko [tkx]: Endangered; Trans-New Guinea; spoken in Indonesia. Tarpia (Tarfia, Sufrai) [tpf]: Endangered, 564 speakers; Austronesian; spoken in Indonesia. Tause (Doa, Darha) [tad]: Endangered, 300 speakers; Lakes Plain; spoken in Indonesia. Taworta (Dabra, Taworta-Aero, Taria) [tbp]: Endangered, 140 speakers; Lakes Plain; spoken in Indonesia. Tay Khang [tnu]: Endangered, 200 speakers; Tai-Kadai; spoken in Laos. Te’un [tve]: Threatened, 1,200 speakers; Austronesian; spoken in Indonesia. Tefaro (Demba) [tfo]: Endangered, 100 speakers; East Geelvink Bay; spoken in Indonesia. Teiwa (Tewa) [twe]: Endangered, 4,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. Temiar (Temer, Northern Sakai, Seroq) [tea]: Vulnerable; Austro-Asiatic; spoken in Malaysia. Temoq (Ulu Indau) [tmo]: Endangered, approximately 300 speakers; Austro-Asiatic; spoken in Malaysia. Temuan (Benua, Niap, Beduanda) [tmw]: Threatened, 8,000 speakers; Austronesian; spoken in Malaysia. Ternate [tft]: Threatened, 42,000 speakers; North Halmaheran; spoken in Indonesia. Thai Song (Song, Thai Song Dam, Lao Song) [soa]: Vulnerable, 32,300 speakers; Tai-Kadai; spoken in Thailand. Thavung (Aheu, Phon Soung, So) [thm]: Severely endangered, approximately 450 speakers; Austro-Asiatic; spoken in Thailand and Laos. Thu Lao [tyl]: Endangered, 200 speakers; Tai-Kadai; spoken in Vietnam. Tii (Rote, Roti, Rotinese) [txq]: Vulnerable, 20,000 speakers; Austronesian; spoken in Indonesia. Tobati (Jotafa, Yotafa, Yautefa) [tti]: Severely endangered, fewer than 100 speakers; Austronesian; spoken in Indonesia. 241

Anna Belew et al.

Tofanma (Tofamna) [tlg]: Vulnerable, 251 speakers; Namla-Tofanma; spoken in Indonesia. Tomadino [tdi]: Endangered, 600 speakers; Austronesian; spoken in Indonesia. Tombelala [ttp]: Threatened, 1,100 speakers; Austronesian; spoken in Indonesia. Tomini (Tiadje, Tialo, Mouton) [txm]: Threatened, approximately 30,000 speakers; Austronesian; spoken in Indonesia. Tonsawang (Tombatu) [tnw]: Vulnerable, 20,000 speakers; Austronesian; spoken in Indonesia and Solomon Islands. Totoli (Tontoli, Tolitoli, Gage) [txe]: Endangered, approximately 7,500 speakers; Austronesian; spoken in Indonesia. Towei (Towe) [ttn]: Endangered, approximately 115 speakers; Pauwasi; spoken in Indonesia. Triw (Triw) [no ISO 639-3 code]: Threatened, fewer than 1,328 speakers; Austro-Asiatic; spoken in Laos. Tum (Toum) [no ISO 639-3 code]: Endangered, approximately 1,000 speakers; AustroAsiatic; spoken in Laos. Uab Meto (Atoni, Meto, Uab Atoni Pah Meto) [aoz]: At risk, 700,000 speakers; Austronesian; spoken in Indonesia. Ugong (Lawa, ‘Ugong, Gong) [ugo]: Critically endangered, 50 speakers; Sino-Tibetan; spoken in Thailand. Ujir (Udjir, Aru) [udj]: Endangered, 980 speakers; Austronesian; spoken in Indonesia. Ukit [umi]: Endangered, 120 speakers; Austronesian; spoken in Malaysia. Uma’ Lung (Uma’ Lung Kenyah, Oma Longh, Òma Lóngh) [ulu]: Threatened, 3,000 speakers; Austronesian; spoken in Indonesia. Urak Lawoi’ (Orak Lawoi’, Lawta, Chaw Talay) [urk]: Endangered; Austronesian; spoken in Thailand and Malaysia. Uruangnirin (Faur, Tubiruasa) [urn]: Endangered, 400 speakers; Austronesian; spoken in Indonesia. Wae Rana (Waerana) [wrx]: Threatened, 3,000 speakers; Austronesian; spoken in Indonesia. Wahau Kenya (Kenyah, Wahau) [whk]: Threatened, 8,000 speakers; Austronesian; spoken in Indonesia. Wano (Waano) [wno]: Threatened, 7,000 speakers; Trans-New Guinea; spoken in Indonesia. Wanukaka (Wanokaka) [wnk]: Endangered, 14,000 speakers; Austronesian; spoken in Indonesia. Warembori (Warenbori, Waremboivoro) [wsa]: Endangered, 600–700 speakers; Lower Mamberamo; spoken in Indonesia. Wares [wai]: Endangered, 200 speakers; unclassified; spoken in Indonesia. Waritai (Weretai, Wari) [wbe]: Endangered, 150 speakers; Lakes Plain; spoken in Indonesia. Warkay-Bipim (Bipim As-So, Bipim, Warkay) [bgv]: Endangered; Trans-New Guinea; spoken in Indonesia. Waru (Mopute, Mapute) [wru]: Endangered, 350 speakers; Austronesian; spoken in Indonesia. Wersing (Kolana, Kolana-Wersin, Wersin) [kvw]: Vulnerable, approximately 4,000 speakers; Timor-Alor-Pantar; spoken in Indonesia. West Tarangan (West Trangan, Tarangan Barat) [txn]: Vulnerable, 8,000-9,000 speakers; Austronesian; spoken in Indonesia. Western Bru (Bruu) [brv]: Vulnerable, 20,000 speakers; Austro-Asiatic; spoken in Thailand. Western Bukidnon Manobo [mbb]: Vulnerable, 19,000 speakers; Austronesian; spoken in the Philippines. Western Pantar (Lamma, Mauta, Tubbe) [lev]: Vulnerable, 10,804 speakers; Timor-AlorPantar; spoken in Indonesia. 242

The world’s endangered languages

Weyewa (Wewewa, Wejewa, Waijewa) [wew]: At risk, at least 240,000 speakers; Austronesian; spoken in Indonesia. White Lachi (Lipupõ, 白拉基) [lwh]: Dormant; Tai-Kadai; formerly spoken in Vietnam. Woi [wbw]: Threatened, 1,800 speakers; Austronesian; spoken in Indonesia. Woria [wor]: Critically endangered, six speakers; East Geelvink Bay; spoken in Indonesia. Wotu [wtw]: Threatened, 5,000 speakers; Austronesian; spoken in Indonesia. Yafi (Jafi, Jafi Wagarindem, Wagarindem) [wfg]: Endangered, approximately 170 speakers; Pauwasi; spoken in Indonesia. Yamna [ymn]: Endangered, 560 speakers; Austronesian; spoken in Indonesia. Yarsun [yrs]: Endangered, 200 speakers; Austronesian; spoken in Indonesia. Yaw ( ) [no ISO 639-3 code]: Endangered, fewer than 10,000 speakers; Sino-Tibetan; spoken in Myanmar. Yetfa-Biksi (Yetfa, Biaksi, Biksi) [yet]: Threatened, 1,000 speakers; isolate; spoken in Indonesia. Yogyakarta Sign Language [no ISO 639-3 code]: Threatened, 1,600–3,300 speakers; sign language; spoken in Indonesia. Yoke (Yoki, Yauke, Jauke) [yki]: Endangered, fewer than 200 speakers; Lower Mamberamo; spoken in Indonesia. Zome (Zorni, ) [zom]: Vulnerable, 17,000 speakers; Sino-Tibetan; spoken in Myanmar and India.

Notes 1 Note that for their classification, pidgin and creole languages are identified as “pidgin or creole.” This does not necessarily indicate uncertainty as to whether a given language is a “pidgin” or “creole”; rather, it reflects ELCat’s policy of including both pidgins and creoles in this single category for the sake of consistency within the database, as well as allowing for cases where pidgin vs. creole status is disputed. However, given that they have no native speakers by definition, there are few pidgins in the ELCat database. 2 Though it should be acknowledged that “urbanisation has also generated variation and the creation of languages” (Lüpke 2015:71), and so urbanization should not be regarded as an unadulterated threat to linguistic diversity. 3 Burundi, Cape Verde, Djibouti, Mauritius, Rwanda, São Tomé and Príncipe, Seychelles, and Swaziland are not included in the Catalogue, as no known endangered languages are spoken in these countries. 4 Not all researchers divide languages and dialects by the same criteria as Ethnologue does, so these numbers should be understand as “plus or minus” a few languages. For instance, the number of South Caucasian languages in Ethnologue is, at 5, higher than most counts, because Ethnologue recognizes the language Judeo-Georgian in addition to the four commonly recognized South Caucasian languages (Georgia, Laz, Mingrelian, and Svan). 5 It is unlikely that these two sign languages are the only ones used within the Caucasus. In Georgia, it appears that Georgian Sign Language is used and taught in Deaf schools, although this is not recognized by Ethnologue. The subject of language use among Deaf communities within the Caucasus is currently under-researched. 6 North Korea, a.k.a. the Democratic People’s Republic of Korea, is not included here. In any case, it appears to have no endangered languages, but we do not have reliable information due to its inaccessibility. The Hamkyeong dialect of North Korea is considered by some to be mutually unintelligible with Korean, though not relevant to ELCat, since it is not endangered. 7 “Altaic” is not a valid genetic unit (language family) and therefore no language is identified as “Altaic” in this book (see Janhunen 1992, Georg 2004, 2005, and Vovin 2005, 2010). 8 Although not at an accelerated rate, language shift to Standard Mandarin has been observed among younger generations in China and Taiwan, whose heritage languages are not mutually intelligible with Mandarin. However, despite the fact that speaker numbers of some Chinese vernaculars are gradually decreasing, the level of endangerment is not as severe as that of the minority languages.

243

Anna Belew et al. 9 Archaeological evidence shows that the Austronesian language family dates back to 6000 BP (Chang 1995; Chang and Goodenough 1996; Bellwood 1991, 1997, 2000; Tsang 1992, 2005, 2012a, 2012b; among others). 10 Yeh (2002:36–37) mentioned that several emperors, including Shunzhi (順治) and Kangxi (康熙), promoted Confucianism and the imperial examination, employing Han elites for imperial civic service, as a way to pacify the Han Chinese and to gain support from the Han elites. Abe (1981: 5–11) also divided the Qing Dynasty into several stages based on the emperors’ policies of Sinicization. 11 Specifically, this section’s treatment of Russia excludes ten of the nation’s federal subjects: Adygea, Chechnya, Dagestan, Ingushetia, Kabardino-Balkaria, Kalmykia, Karachay-Cherkessia, Krasnodar, North Ossetia-Alania, and Stavropol. 12 Data from the Russian Census of 2010, as reported online: www.gks.ru/free_doc/new_site/perepis2010/ croc/perepis_itogi1612.htm. Accessed 2-19-2017. 13 www.eldia-project.org/ 14 Unless otherwise indicated, the figures, percentages, and information reported here are all based on the Catalogue of Endangered Languages as of April 4, 2017. 15 The Americas are home to c.175 language families (including language isolates); c.407 are known in all the world – that is, 43% of all the linguistics diversity of the world, calculated in terms of independent language families, is found in the Americas (cf. Campbell 1997). 16 These include Judeo-Isfahani, Judeo-Kashani, Judeo-Shirazi, and Judeo-Yazdi. They are currently listed as “no estimate available” in the Catalogue, pending more specific vitality information. 17 These include Eshtehardi, Gozarkhani, Kabatei, Kajali, Maraghei, Razajerdi, Rudbari, Savi, Shahmirzadi, Upper Taromi, Judeo-Isfahani, Judeo-Kashani, Judeo-Shirazi, and Judeo-Yazdi. 18 These include Bartang (Vulnerable), Ishkashmi (Threatened), Jowshaqani (Severely Endangered), Kesha’i (Endangered), Khufi (Vulnerable), Kuhpayi (Severely Endangered), Neo-Mandaic (Severely Endangered), Oroshor (Vulnerable), Semnani (Vulnerable), Soqotri (Endangered), Yaghnobi (Vulnerable), and Yazghulami (Vulnerable). 19 Unless otherwise indicated, the figures, percentages, and information reported here are all based on the Catalogue of Endangered Languages as of April 4, 2017. 20 The Americas are home to c.175 language families (including language isolates); c.407 are known in all the world – that is, 43% of all the linguistic diversity of the world, calculated in terms of independent language families, is found in the Americas (cf. Campbell 1997). 21 Ethnologue (Simons & Fennig 2017) lists 7,099 languages, but 220 of them have no known native speakers; ELCat lists 3,346, of which 231 have no known native speakers. 22 The countries of North America for the purposes of ELCat are Canada, the US, and Greenland – Greenland is linguistically, though not geographically, affiliated with North America. Mexico, though also part of North America, is treated together with Central American languages in ELCat. Several languages lap over borders, spoken both in Canada and the US, and in both Mexico and the US. 23 Among Ethnologue’s total of 7,099 “living languages” are 220 languages with 0 speakers, as well as an additional 199 with an unknown number of speakers (Simons & Fennig 2017). 24 Ethnologue offers 5,235 as the mean number of speakers for languages of the Pacific, although this figure includes Australia (Simons & Fennig 2017). 25 Unless otherwise indicated, the figures, percentages, and information reported here are all based on the Catalogue of Endangered Languages as of April 4, 2017. 26 The Americas are home to c.175 language families (including language isolates); c.407 are known in all the world – that is, 43% of all the linguistic diversity of the world, calculated in terms of independent language families, is found in the Americas (cf. Campbell 1997, 2012). 27 Ethnologue (Simons & Fennig 2017) lists 7,099 languages, but 219 of them have no known speakers, and one is a constructed language; thus Ethnologue has 6,879 living languages. ELCat lists 3,394, of which 254 have no known speakers, so 3,140 living endangered languages in the world. 28 Geographically, Maldives is included in South Asia. Ethnologue lists two known languages, English and Maldivian in Maldives, both having strong vitality. Hence, Maldives has no endangered languages and is not included in the Catalogue of Endangered Languages. 29 Note that languages identified in the Catalogue of Endangered Languages do not necessarily have one-to-one correspondences to those listed in Ethnologue. 30 Linguistically, Hindi and Urdu are varieties of the same language, although they employ different orthographies – Hindi written in Devanāgarī and Urdu in Arabic script. Generally speaking, Hindi is

244

The world’s endangered languages

31

32

33

34

35

36

37 38

spoken by Hindus and Urdu is used by Muslims, and the speakers have separate identities related to their religions and countries of citizenship. Indigenous groups of southern India who speak Dravidian languages especially resent attempts to impose Hindi as the only official language. These people are also culturally and ethnically different from speakers of Hindi and most other Indo-Aryan languages. It should be noted that the traditional binary branching view of the Sino-Tibetan language family into Sinitic and Tibeto-Burman has recently been much debated (see Jacques 2017; Handel 2008; van Driem 2008, 2011, 2014, 2015). The so-called Papuan languages do not represent a family of languages that are genetically related to one another, but rather “Papuan” is a cover term for the non-Austronesian languages of the area. “Official language” here refers to the languages designated as the official languages in the Constitution(s). “National language” or “regional language,” on the other hand, refers to dominant languages that have a large speaker population, but are not named as official languages. We use the term “Filipino” following the Constitution of the Philippines. However, linguistically speaking, Filipino is a Tagalog based variety. The socio-linguistic connotations involving the terms Filipino and Tagalog are not discussed here. Bahasa Indonesia/Indonesian [ind] and Malaysian Malay [zlm] do not share a single ISO 639-3 code, although they are mutually intelligible. This is because the ISO code system/Ethnologue is not entirely based on linguistic criteria, primarily mutual intelligibility, unlike the standard employed in the Catalogue of Endangered Languages. Kluge (2014) classifies Papuan Malay as an Austronesian language, where Ethnologue (Simons & Fennig 2017) lists Papuan Malay as a Malay-based creole. The 2005 Census of Laos shows that 54.6% of the entire population is ethnic Lao.

References Abe, Takeo (安部健夫). 1981. A study of the history of Qing Dynasty (清代史の研究). Tokyo: Sōbunsha (創文社). Abbi, Anvita. 2013. A grammar of the Great Andamanese language: An ethnolinguistic study. Leiden: Brill. Abu-Manga, Al-Amin. 2010. Challenges to management of linguistic diversity in the Sudan. (Paper presented at International Seminar on India and Africa: Partnership for Capacity Building and Human Resource Development Jawaharlal Nehru University, New Delhi, February 9–10, 2010.) (http:// khartoumspace.uofk.edu/handle/123456789/5080) Alam, Shamsul. 1991. Language as political articulation: East Bengal in 1952. Journal of Contemporary Asia 21(4). 469–487. Bashir, Elena. 2003. Dardic. In Cardona, George & Jain, Dhanesh (eds.), The Indo-Aryan languages, 818–894. London: Routledge. Batibo, Herman. 2005. Language decline and death in Africa: Causes, consequences, and challenges. Clevedon: Multilingual Matters. Bellwood, Peter. 1991. The Austronesian dispersal and the origins of languages. Scientific American 265(1). 88–93. Bellwood, Peter. 1997. Taiwan and the prehistory of the Austronesian-speaking peoples. Review of Archaeology 18(2). 39–48. Bellwood, Peter. 2000. The time depth of major language families: An archaeologist’s perspective. In Renfrew, Colin & McMahon, April & Trask, Larry (eds.), Time depth in historical linguistics, 109–40. Cambridge: McDonald Institute for Archaeological Research. Blench, Roger. 2018. African language isolates. In Campbell, Lyle (ed.), Language isolates, 162–192. Abingdon and New York: Routledge. Blust, Robert. 2013. The Austronesian languages (revised edition). Canberra: The Australian National University. Borjian, Habib. 2009. Median succumbs to Persian after three millennia of coexistence: Language shift in the Central Iranian Plateau. Journal of Persianate Studies (2). 62–87. Borjian, Habib. 2015. Judeo-Iranian languages. In Rubin, Aaron D. & Kahn, Lily (eds.), Handbook of Jewish languages, 234–296. Leiden: Brill.

245

Anna Belew et al. Bradley, David. 2005. Language policy and language endangerment in China. International Journal of the Sociology of Language 173. 1–21. Bradley, David. 2007. East and Southeast Asia. In Moseley, Christopher (ed.), Encyclopedia of the world’s endangered languages, 349–424. London & New York: Routledge. Brenzinger, Matthias (ed.) 1992. Language death: Factual and theoretical explorations with special reference to East Africa. Berlin: De Gruyter. Brenzinger, Matthias (ed.). 2008. Language diversity endangered. Berlin: De Gruyter. Brunelle, Marc. 2008. Diglossia, bilingualism and the revitalization of written Eastern Cham. Language Documentation & Conservation 2(1). 28–46. Buddruss, Georg. 1988. Bartangī. Encyclopædia Iranica, Vol. III, Fasc. 8. 827–830. Online: www. iranicaonline.org/articles/bartangi-a-member-of-the-sugni-q Campbell, Lyle. 1997. American Indian languages: The historical linguistics of Native America. Oxford: Oxford University Press. Campbell, Lyle. 2012. The classification of South American indigenous languages. In Campbell, Lyle & Grondona, Verónica (eds.), The indigenous languages of South America: A comprehensive guide, 59–166. Berlin: De Gruyter. Campbell, Lyle & Kaufman, Terrence & Smith-Stark, Thomas. 1986. Mesoamerica as a linguistic area. Language 62. 530–570. Chang, K. C. 1995. Taiwan Strait archaeology and proto-Austronesian. In Li, Paul Jenn-Kuei & Ho, Dahan & Huang, Ying-Kuei & Tsang, Cheng-hwa & Tseng, Chiu-yu (eds.), Austronesian studies relating to Taiwan, 161–84. Taipei, Taiwan: Academia Sinica. Chang, K. C. & Goodenough, Ward. 1996. Archaeology of southern China and its bearing on the Austronesian homeland. In Goodenough, Ward (ed.), Prehistoric settlement of the Pacific, 36–56. Philadelphia: American Philosophical Society. Chou, Wan-yao (周婉窈). 2009. A New Illustrated History of Taiwan (臺灣歷史圖說增訂本). Taipei: Linking Publishing (聯經出版公司). Comrie, Bernard. 2008. Linguistic diversity in the Caucasus. Annual Review of Anthropology 37. 131–143. Connell, Bruce. 2015. The role of colonial languages in language endangerment in Africa. In McLaughlin, Fiona & Henderson, Brent & Essegbey, James (eds.), Language documentation and endangerment in Africa, 107–130. Amsterdam: John Benjamins. Denevan, William M. 1992. The native population of the Americas. Madison, WI: University of Wisconsin Press. Dimmendaal, Gerrit J. 2015. Different cultures, different attitudes: But how different is “the African situation” really? In McLaughlin, Fiona & Henderson, Brent & Essegbey, James (eds.), Language documentation and endangerment in Africa, 37–58. Amsterdam: John Benjamins. Essegbey, James & Henderson, Brent & McLaughlin, Fiona (eds.) 2015. Language documentation and endangerment in Africa. Amsterdam: John Benjamins. Friedman, Victor A. 2010. Sociolinguistics in the Caucasus. In Ball, Martin J. (ed.), The Routledge handbook of sociolinguistics around the world, 127–138. New York: Routledge. Georg, Stefan. 2004. Review of Etymological dictionary of the Altaic languages. Diachronica 21(2). 445–450. Georg, Stefan. 2005. Reply [to Starostin 2005]. Diachronica 22(2). 455–457. Gippert, Jost. 2008. Endangered Caucasian languages in Georgia: Linguistic parameters of language endangerment. In Harrison, K. David & Rood, David S. & Dwyer, Arienne (eds.), Lessons from documented endangered languages, 159–194. Amsterdam: John Benjamins. Golla, Victor & Goddard, Ives & Campbell, Lyle & Mithun, Marianne & Mixco, Mauricio. 2008. North America. In Moseley, Christophe & Asher, Ron (eds.), Atlas of the world’s languages, 7–41. London: Routledge. Greenberg, Joseph. 1963. The languages of Africa. Bloomington: Indiana University Research Center in Anthropology, Folklore, and Linguistics, Publication no. 25a. Grenoble, Lenore. 2003. The Caucasus. In Grenoble, Lenore, Language policy in the Soviet Union, 111–136. Norwell, MA: Kluwer Academic Publishers. Häberl, Charles. Forthcoming. (Endangered languages of) the Middle East and North Africa. (www. academia.edu/27190965/_Endangered_Languages_of_The_Middle_East_and_North_Africa). Hammarström, Harald. 2012. Pronouns and the (preliminary) classification of Papuan languages. In Hammarström, Harald & van den Heuvel, Wilco (eds.), History, contact and classification of Papuan languages, 428–539. Port Moresby: Linguistic Society of Papua New Guinea.

246

The world’s endangered languages Handel, Zev. 2008. What is Sino-Tibetan? Snapshot of a field and a language family in flux. Language and Linguistics Compass 2(3). 422–441. Hobson, John & Lowe, Kevin & Poetsch, Susan & Walsh, Michael (eds.). 2011. Re-awakening languages: Theory and practice in the revitalisation of Australia’s indigenous languages. Sydney: Sydney University Press. Hyslop, Gwendolyn. 2016. Worlds of knowledge in Central Bhutan: Documentation of ‘Olekha. Language Documentation & Conservation 10. 77–106. Jacques, Guillaume. 2017. The genetic position of Chinese. In Sybesma, Rint & Behr, Wolfgang & Gu, Yuegu & Handel, Zev & Huang, C.-T. James (eds.), Encyclopedia of Chinese language and linguistics, 297–306. Leiden: Brill. Janhunen, Juha. 1992. Das Japanische in vergleichender Sicht. Journal de la Société Finno-Ougrienne 84. 145–161. Janhunen, Juha. 1997. The language of Manchuria in today’s China. In Shoji, Hiroshi & Janhunen, Juha (eds.), Northern minority languages: Problems of survival, 123–46. Osaka: National Museum of Ethnology. Janhunen, Juha. 2005. Tungusic: An endangered language family in northeast Asia. International Journal of the Sociology of Language 173. 37–54. Juaristi, Patxi & Reagan, Timothy & Tonkin, Humphrey. 2008. Language diversity in the European Union. In Arzoz, Xavier (ed.), Respecting linguistic diversity in the European Union, 47–72. Amsterdam: John Benjamins. Kandybowicz, Jason & Torrence, Harold (eds.). 2018. Africa’s endangered languages: Documentary and theoretical approaches. Oxford: Oxford University Press. Kazakevich, Olga & Kibrik, Aleksandr. 2007. Language endangerment in the CIS. In Brenzinger, Matthias (ed.), Language diversity endangered, 233–262. New York: Mouton de Gruyter. Khokhlova, Liudmila V. 2014. Majority language death. In Cardoso, Hugo C. (ed.), Language endangerment and preservation in South Asia, 19–45. Honolulu: University of Hawai‘i Press. Kluge, Angela J. H. 2014. A grammar of Papuan Malay. Leiden: Leiden University. (Doctoral dissertation.) Lee, Nala H. 2014. A grammar of Baba Malay with sociophonetic considerations. Honolulu: University of Hawai‘i at Mānoa. (Doctoral dissertation.) Loukotka, Čestmir. 1968. Classification of South American Indian languages. Los Angeles: Latin American Studies Center, UCLA. Lüpke, Friederike. 2015. Ideologies and typologies of language endangerment in Africa. In McLaughlin, Fiona & Henderson, Brent & Essegbey, James (eds.), Language documentation and endangerment in Africa, 59–106. Amsterdam: John Benjamins. McQuown, Norman A. 1955. The indigenous languages of Latin America. American Anthropologist 57. 501–70. Maddieson, Ian & Spajić, Sinisa & Sands, Bonny & Ladefoged, Peter. 1993. Phonetic structures of Dahalo. UCLA Working Papers in Phonetics 84. 25–66. Mous, Maarten. 2003. Loss of linguistic diversity in Africa. In Janse, Mark & Tol, Sijmen (eds.), Language death and language maintenance: Theoretical, practical and descriptive approaches, 157–170. Amsterdam: John Benjamins. Nichols, Johanna. 1992. Linguistic diversity in space and time. Chicago: University of Chicago Press. Noonan, Michael. 1996. The fall and rise and fall of the Chantyal language. Southwest Journal of Linguistics 15(1–2). 121–36. Noonan, Michael. 2006. The rise of ethnic consciousness and the politicization of language in west-central Nepal. In Saxena, Anju & Borin, Lars (eds.), Lesser-known languages of South Asia: Status and policies, case studies and applications of information technology, 161–174. Berlin: Mouton de Gruyter. Owens, Jonathan. 2007. Endangered languages of the Middle East. In Brenzinger, Matthias (ed.), Language diversity endangered, 263–277. Berlin: Mouton de Gruyter. Pawley, Andrew. 2005. The chequered career of the Trans New Guinea hypothesis: Recent research and its implications. In Pawley, Andrew & Attenborough, Robert & Golson, Jack & Hide, Robin (eds.), Papuan pasts: Cultural, linguistic and biological histories of Papuan-speaking peoples, 67–107. Canberra: Research School of Pacific and Asian Studies, Australian National University. Pittayaporn, Pittayawat. 2014. Layers of Chinese loanwords in Proto-Southwestern Tai as evidence for the dating of the spread of Southwestern Tai. MANUSYA: Journal of Humanities 20. 47–64. Purdie, Nola & Frigo, Tracey & Ozolins, Clare & Noblett, Geoff & Thieberger, Nick & Sharp, Janet. 2008. Indigenous languages programs in Australian schools—a way forward. A report for the

247

Anna Belew et al. Australian Government Department of Education, Employment and Workplace Relations, prepared as part of a team led by the Australian Council for Educational Research (ACER). Online: http:// research.acer.edu.au/indigenous_education/18 Rahman, Tariq. 2002. Language, power and ideology. Economic and Political Weekly 37(44). 4556– 4560. Reid, Lawrence A. 1987. The early switch hypothesis: Linguistic evidence for contact between Negritos and Austronesians. Man and Culture in Oceania 3. 41–59. Reid, Lawrence A. 1994a. Unravelling the linguistic histories of Philippine Negritos. In Dutton, T. E. & Tryon, D. T. (eds.), Language contact and change in the Austronesian world, 443–475. Berlin: Mouton de Gruyter. Reid, Lawrence A. 1994b. Possible non-Austronesian lexical elements in Philippine Negrito languages. Oceanic Linguistics 33(1). 37–72. Ross, Malcolm. 2005. Pronouns as a preliminary diagnostic for grouping Papuan languages. In Pawley, Andrew & Attenborough, Robert & Golson, Jack & Hide, Robin (eds.), Papuan pasts: Cultural, linguistic and biological histories of Papuan-speaking peoples, 15–66. Canberra: Research School of Pacific and Asian Studies, Australian National University. Salminen, Tapani. 2007a. Endangered languages in Europe. In Brenzinger, Matthias (ed.), Language diversity endangered, 205–232. New York: Mouton de Gruyter. Scott, Graham & Kim, R. Hyun & Rumaropen, Benny E. W. & Scott, Eleonora L. & Nussy, Christian G. & Yumbi, Anita C. M. & Cochran, Robert C. 2008. Tong pu bahasa: A preliminary report on some linguistic and sociolinguistic features of Papuan Malay. Sentani: SIL International Indonesia. Shibatani, Masayoshi. 1990. The languages of Japan. Cambridge: Cambridge University Press. Simons, Gary F. & Fennig, Charles D. (eds.). 2017. Ethnologue: Languages of the world, twentieth edition. Dallas, Texas: SIL International. (www.ethnologue.com) Stevenson, Mark. 2007. Mexicans debate cause of early Indian die-offs. Los Angeles Times (January 21, 2007). Thieberger, Nick. 2013. Endangered languages in Australia. In Chapelle, Carol A. (ed.), The encyclopedia of applied linguistics. Hoboken, NJ: John Wiley & Sons. Thomason, Sarah G. 2001. Language contact: An introduction. Edinburgh: Edinburgh University Press. Tsang, Cheng-hwa. 1992. Archaeology of the P’eng-Hu islands. Taipei: Academia Sinica. Tsang, Cheng-hwa. 2005. Recent discoveries at the Tapenkeng culture sites in Taiwan: Implications for the problem of Austronesian origins. In Sagart, Laurent & Blench, Roger & Sanchez-Mazas, Alicia (eds.), The peopling of East Asia, 63–73. London: Routledge. Tsang, Cheng-hwa. 2012a. 再論南島語族的起源問題 [Once again on the Austronesian origin and dispersal]. 南島研究學報 [Journal of Austronesian Studies] 3(1). 87–119. Tsang, Cheng-hwa. 2012b. 南科考古出土的稻米與小米兼論相關問題 [Issues relating to the ancient rice and millet grains unearthed from the archaeological sites in Tainan science park]. 中國飲食文 化 [Journal of Chinese dietary culture] 8(2). 1–24. Turin, Mark. 2005. Language endangerment and linguistic rights in the Himalayas: A case study from Nepal. Mountain Research and Development 25(1). 4–9. Turin, Mark. 2006. Minority language policies and politics in Nepal. In Saxena, Anju & Borin, Lars (eds.), Lesser-known languages of South Asia: Status and policies, case studies and applications of information technology, 61–72. Berlin: Mouton de Gruyter. van Driem, George. 2007a. South Asia and the Middle East. In Moseley, Christopher (ed.), Encyclopedia of the world’s endangered languages, 283–348. London: Routledge. van Driem, George. 2007b. Endangered languages of South Asia. In Brenzinger, Matthias (ed.), Language diversity endangered, 303–341. Berlin: De Gruyter. van Driem, George. 2008. To which language family does Chinese belong, or what’s in a name? In SanchezMazas, Alicia & Blench, Roger & Ross, Malcolm D. & Peiros, Ilia & Lin, Marie (eds.), Past human migrations in East Asia: Matching archaeology, linguistics and genetics, 219–253. London & New York: Routledge. van Driem, George. 2011. Tibeto-Burman subgroups and historical grammar. Himalayan Linguistics 10(1). 31–39. van Driem, George. 2014. Trans-Himalayan. In Hill, Nathan & Owen-Smith, Thomas (eds.), TransHimalayan linguistics, 11–40. Berlin: Mouton de Gruyter. van Driem, George. 2015. Tibeto-Burman. In Wang, William S.-Y. & Sun, Chaofen (eds.), Oxford handbook of Chinese linguistics, 135–148. Oxford: Oxford University Press.

248

The world’s endangered languages Vovin, Alexander. 2005. The end of the Altaic controversy (review of Starostin et al. 2003). Central Asiatic Journal 49(1). 71–132. Vovin, Alexander. 2010. Koreo-Japonica: A re-evaluation of a common genetic origin. Honolulu: University of Hawai‘i Press. Wurm, Stephen A. & Voorhoeve, C. L. & McElhanon, Kenneth A. 1975. The Trans New Guinea phylum in general. In Wurm, Stephen A. (ed.), New Guinea area languages, volume 1: Papuan languages and the New Guinea linguistic scene, 299–322. Canberra: Pacific Linguistics. Yeh, Kao-shu (葉高樹). 2002. The cultural policy in early Qing (1644–1795) (清朝前期的文化政策). Taipei: Daw Shiang Publishing (稻鄉出版社).

249

8 HOW THE CATALOGUE OF ENDANGERED LANGUAGES SERVES COMMUNITIES WHOSE LANGUAGES ARE AT RISK Raina Heaton and Sean Simpson

This chapter explores the real-world applications of the Catalogue of Endangered Languages and the Endangered Languages Project. The project provides resources not only for academics but also for teachers, students, and endangered language communities. We discuss the features of the website designed specifically with community needs in mind, including discussion forums for networking and a collection of approximately 6,500 audio-visual and text resources which complements the core content of the Catalogue. We also outline several cases where the Catalogue has aided in community projects, and we discuss applications of the Catalogue for funding, awareness, and advocacy for endangered languages.

Introduction As detailed elsewhere in this volume, the Catalogue of Endangered Languages (ELCat) was designed to be the most comprehensive and up-to-date resource for information on the world’s endangered languages available today. As such, it clearly has relevance to researchers and academics engaged in working with endangered and minority languages. However, in addition to its role as a scholarly resource, the Catalogue and the Endangered Languages Project website (endangeredlanguages.com), which houses the Catalogue, were developed with an eye towards providing a broad range of services for those outside of the academic study of language as well. This includes speakers of endangered languages, language teachers, students, funding agencies, policy makers, communities engaged in language documentation and revitalization, and lay people who are generally interested in learning more about languages, language pedagogy, or language endangerment. The Catalogue is widely available online, with interfaces in seven languages, and can be accessed free of charge by anyone who has a phone or computer with an internet connection. 250

How ELCat serves language communities

While the Catalogue’s design was intended to serve the needs of a wide range of users, it was especially crucial for the database and its surrounding infrastructure to address and serve the needs of the members of those communities whose languages are at risk. In this chapter we provide a brief overview of how the Catalogue and the Endangered Languages Project website are intended to serve those most affected by language endangerment. There are two points we would like to clarify with respect to the nature of the Catalogue and endangered language communities. First, the Catalogue has just recently reached its current state of completion, so cases of benefits to communities up to this point are less visible than those we anticipate from this point forward. We are confident that as the Catalogue continues to be used, such instances will dramatically increase. Second, while the information housed in the Catalogue can be used to aid revitalization or documentation efforts, it is not in itself designed directly to achieve either of these ends, which are the prerogative of individual communities and researchers. That said, the Catalogue was designed with an eye toward increasing the decision-making role of endangered language community members in how their languages are represented on language information sites, fostering communication between geographically disparate endangered language communities across the globe, and changing the nature of how users generally engage, interact with, and think about language information websites. We see the primary benefits of the Catalogue and the Endangered Languages Project website to endangered language community members as two-fold. First, the presentation of the most current and reliable information on a given language in a digestible manner provides educational benefits, as well as an official account of the status of the language that can be used to garner support for revitalization activities. “The Catalogue and communities” is devoted to an in-depth discussion of how such information dissemination has already benefitted endangered language stakeholders and community members to date, and to some ways in which we hope the information contained within the Catalogue will be used to benefit this audience in the future. Second, communities can find immediate use for user features which allow for the exchange of current knowledge and methodology pertaining to revitalization, language research, and language awareness. “The Endangered Languages Project and communities” details several of the features of the ELP website (www.endangeredlanguages.com, which houses the Catalogue) which were specifically designed with endangered language community members in mind, and which set this project apart from other language information compendia. While internet access and use are globally on the rise, many endangered language community members live in areas where internet access is limited or non-existent. As the Catalogue is fundamentally an online resource, this means its access for such communities is limited. Despite this limitation, however, we believe that the information housed in the Catalogue can nonetheless be of use to those who do have internet access – as evidenced by some of the anecdotes presented in “The Catalogue and communities” below – and believe that the features outlined in the section “The Endangered Languages Project and communities” position the Catalogue as a resource with useful applications to all engaged parties, rather than just to those within the realm of academia.

The Catalogue and communities There are many databases that collect information on different facets of language. Some major examples are Ethnologue (Simons & Fennig 2017), which provides general information on all known languages that they have identified; Glottolog (Hammarström et al. 2016), which gives a list of bibliographic resources for each known language; and the World Atlas of Linguistic Structures (WALS) (Dryer & Haspelmath 2013), which tracks specific typological features in 251

Raina Heaton and Sean Simpson

a wide variety of languages. The Catalogue differs from these others and excels in its mission by focusing on endangered languages. It is designed to be a comprehensive database about the endangered languages of the world, and also to have the best possible estimate of how endangered each language is. Both of these features are unique with respect to the other databases mentioned above. As such, the Catalogue is now the most comprehensive and up-to-date source of information on the world’s endangered languages. The Catalogue is a reliable source of vitality information on endangered languages. This includes information on the current number of speakers (and whether they are fluent speakers, semi-speakers, or learners), approximate age of youngest speakers, percentage of the ethnic population that speaks the language, whether the language is being passed on to children, the situations in which the language is used, if speaker numbers are increasing, decreasing, or are constant, if the language has official status or other institutional support, what other languages are in use by speakers of the particular language in question, what value speakers place on their language, how many speakers are bilingual/multilingual, and the date when all this information was last gathered. There are two primary ways in which this type of information might benefit endangered language communities: first, with respect to internal language awareness and management, and second, with respect to external perceptions of the language.

Internal awareness and management Internal awareness and language management refers to how information about a language’s vitality impacts the speaker community. First, this applies to the educational sphere, where official information about the endangerment status of a language can impact how speakers relate to and feel about their language. Frequently communities do not realize that their language is endangered, or how endangered it is, until it is too late (i.e. it is at the point where parents and children are no longer speaking the language as part of their everyday lives). As soon as only older speakers remain, the language becomes significantly harder to revitalize (cf. Fishman 1991, 2001, for ways to combat language loss even in such situations). Educating speakers and the wider community about this often sneaky transition from “safe” to “threatened” or “endangered” helps create awareness before the language is moribund. It also can motivate people to become advocates for their languages. When speakers realize how serious the situation is, the hope is that they will be motivated to do something about it, such as engaging in acts of revitalization on a personal level, and potentially also on a community level. We are aware of a few instances where speakers have found their language in the Catalogue, learned that it is endangered, and then looked for ways to get involved. For example, an Effutu speaker who is studying linguistics at Drexel University learned that Effutu (of Ghana) is endangered from the information in the Catalogue, and contacted the Catalogue team to ask how he could help preserve the language. The advocacy goals of the Catalogue also extend beyond the speakers of endangered languages themselves. The goal is that by making accessible as much information on endangered languages as possible and spreading the word about the language endangerment crisis, more people, including linguists and endangered language community members, will become aware of the problem. Currently, there is very low recognition within the general public that language endangerment and loss is a serious issue, or even something that is undesirable. Even within an endangered language community, where there is general recognition that the language is endangered, people may be unaware of the scope of the endangerment crisis. This was true of young Kaqchikel speakers and semi-speakers studying to be Kaqchikel teachers, who were surprised to learn from Heaton (co-author of this chapter) that close to half of all living languages on the planet are endangered or at risk (see Campbell and Okura, Chapter 6). 252

How ELCat serves language communities

Endangered language advocacy and education will eventually end up helping communities by increasing the profile of their languages and by increasing the number of people and organizations invested in endangered language research and preservation (e.g. the Alliance for Linguistic Diversity). Also, on a smaller and more immediate level, the availability of accurate language vitality and location information for endangered languages allows linguists to prioritize language documentation efforts (see Hauk and Heaton, Chapter 9). This brings linguists and community members together, where linguists and speakers can work together to create grammars, dictionaries, and other materials for science and for the community which ensure a lasting record of the language.

External awareness and perceptions The other primary way in which the Catalogue facilitates community endeavors through promoting external awareness of endangered languages pertains to funding. Many granting agencies that fund endangered language research and revitalization projects are more likely to give money to groups that are seen as being more in need. If a language is on the brink of becoming dormant, then the need is indeed very great, and communities can point to the data in the Catalogue to demonstrate this to foster revitalization efforts and to seek funding. We are aware of at least two cases where this has indeed happened. The first example involves a recent court case, Clarabal vs. the Department of Education of the state of Hawai‘i, pertaining to the right to Hawaiian language immersion education on the island of Lāna‘i. Hawaiian is listed in the Catalogue as “Severely Endangered” with 100% certainty. There are an estimated 300 native speakers remaining, mainly in the community on Ni‘ihau (NeSmith 2016). Hawaiian is being transmitted to younger generations via the school system, primarily on Hawai‘i and O‘ahu, which are home to some of the premiere endangered language immersion programs currently in existence. However, immersion education in Hawaiian is not available everywhere in Hawai‘i, including not on Lāna‘i. Hawaiian and English are both official languages of Hawai‘i, which, per article X, section 4 of the Hawai‘i state constitution, requires that “The State shall provide for a Hawaiian education program consisting of language, culture and history in the public schools.” The Clarabal vs. the Department of Education case is currently being appealed. The second instance where funds for a community language goal hinged on Catalogue information comes from the case of an Australian Aboriginal language community. Community members wrote Catalogue staff to inform us that the last native speaker of their language had passed away, and to request that we change the vitality assessment for their language from “Critically Endangered” to “Dormant.” This official change in the status of their language was particularly necessary, they said, because they were attempting to apply for funding for language revitalization projects, and the appropriate “Dormant” level of vitality would improve their application’s chances of success.

The Endangered Languages Project and communities Up to this point we have discussed various ways in which the information in the Catalogue database can benefit endangered language community members and stakeholders. It is necessary to recognize, however, that while the information in and of itself is crucial, the way in which community members and stakeholders are able to access and interact with that information is no less important. The Endangered Languages Project (ELP) was born with this in mind. The ELP website (www.endangeredlanguages.com) is an online community designed to complement and facilitate the information housed in the Catalogue. Through the ELP website, 253

Raina Heaton and Sean Simpson

not only can one gain access to all of the language-specific information housed in the Catalogue of Endangered Languages, one can also find big-picture statistics about language endangerment via the search functions. The database of endangered language information can be searched by region, endangerment status, number of speakers, etc.; one can also explore the languages included in the Catalogue geographically via the interactive, scalable language map on the site. It should also be stressed here that the ELP is a dynamic interface: in addition to being a portal through which users may access the language information in the Catalogue, the ELP was also designed to be a digital space in which those with an interest in endangered languages may interact with one another in discussion forums, upload resources and documents pertaining to the languages in which they are interested, access general video, audio and textual resources pertaining to various facets of language documentation and revitalization, and importantly, provide the Catalogue with feedback on the information currently available, with suggestions of additions, improvements, or corrections. In short, while the information contained in the Catalogue and the endangered language advocacy that the project generates are unquestionably useful for members of endangered language communities, the ELP provides a wide variety of resources that not only aim specifically to benefit community members, but provide a means for users directly to contribute content and to network with members of other communities. Below, we provide more detail on a few of the major features of the ELP that were developed primarily for the benefit of endangered language community members.

Resources Each language represented in the Catalogue has a dedicated language page on the ELP site. Here users can view all of the primary source material included in the Catalogue regarding speaker numbers, usage patterns, alternate names, classification, etc. This also includes access to the vitality assessment mentioned as particularly useful above, as well as a map of the area in which the language is spoken. One of the more unique features of the ELP site, however, which was designed specifically with language community members in mind, is the ability for users to upload resources about the language to any given language page. These resources can be in a variety of formats and can cover a wide range of topics. User-submitted resources constitute their own mini-database, and are separately accessible and searchable from the ELP homepage. The large amount of infrastructure surrounding resources on the site was included to maximize the utility of uploaded language materials to all users, recognizing the fact that, for instance, a novel revitalization strategy pursued by speakers of Okanagan-Colville in Canada may well be of interest to those developing revitalization projects for, say, Lule Saami in Sweden. In effect, the resources function of the site is akin to a megaphone held up to the lips of those uploading language material, broadening the impact of those materials beyond the confines of their own particular language communities. These resources also raise the profile of endangered languages by providing them with a tangible, public, digital profile that is shared with the rest of the world. While basic language information is at the heart of the Catalogue of Endangered Languages project, the resources give life to the data, adding the faces, voices, and knowledge of their speakers to the language information. There are currently about six and a half thousand resources that have been uploaded to the ELP website, most of which are likely to be useful to community members seeking to support their languages. Resources include information on a wide variety of topics, including anthropology, sociology, education, environmental sciences, and public policy in addition to language and linguistics, and all users, researchers, community members, and lay people alike, are encouraged to upload any relevant 254

How ELCat serves language communities

articles, papers, presentations, links, audio recordings, videos, photographs, etc. that they might have. We should note briefly that neither the ELP site nor the Catalogue is itself an archive. The resources and documents uploaded to the website are supported by Google Docs and YouTube, and are not stored within the site itself. Also, users wishing to upload content must first certify that they either own the rights to the content that they are uploading, or that it is otherwise already publicly accessible. Additionally, user uploads to the ELP site are not vetted by a team of experts, as is the case for all data in the Catalogue itself. User-uploaded resources that are inappropriate in any way can be flagged for review by users, and once they are flagged they are immediately taken down from the site until they can be reviewed by the ELP team. Resources can be accessed in four ways: first, via the resources tab on individual language pages, as mentioned above; second, via the overall website search function, where search results can be filtered either by language or resource; third, via the homepage, below the interactive map, where resources may be browsed by category; or fourth, via the resource-specific search function, accessible by clicking the “resources” label at the top of the homepage. The resource search function allows users to filter the thousands of resources on the site by category, tags, and format, and to sort these entries either by newest additions or by most viewed items. Resources are divided into the following eight categories based on their content: 1.

2.

3.

4.

5.

6.

7.

8.

Language research and linguistics (2,760 resources at time of writing), including publications by professional researchers and other documentarians, word lists and other types of documentation materials recorded by linguists, as well as linguistically informed information about individual languages, e.g. pronunciation guides. Language revitalization (530 resources), mostly composed of lessons designed to help people learn a particular language, but also other information about specific revitalization programs and methodologies. Language materials (3,249 resources), including, among other things, formal instructional texts, audio/video recordings instructing viewers on a particular aspects of the language, examples of language games to play with children, and links to YouTube pedagogical video series. Language education (642 resources), including a wide variety of resource types dealing with language pedagogy, including community links like talking dictionaries, instructional sites, and recorded stories. Language advocacy and awareness (698 resources), broadly targeting resources that discuss the influence of language on the lives of their speakers, as well as the Universal Declaration of Human Rights in a number of languages. Language, culture and art (2,528 resources), including interviews with elders, culturally meaningful songs/stories, demonstrations of cultural knowledge such as basket weaving or the making of traditional clothing, etc. Language and technology (1,269 resources), including examples of a wide variety of ways language can be shared electronically (e.g. recordings set to video, instructional videos on how to create digital resources, language learning programs, and links to language tools). Media (2,505 resources), a broad category mostly composed of videos showing different aspects of language use and pedagogy, as well as news stories regarding endangered languages.

While these categories are necessarily not mutually exclusive, the divisions were created to be the most useful way to search based on what users might be looking for. For example, 255

Raina Heaton and Sean Simpson

a member of an endangered language community might be interested in developing educational resources and language games to engage their community’s youth and get them interested in learning their heritage language. Rather than start from scratch, such a person could head to the resources section of the ELP, filter the uploaded material for resources tagged as “Language Revitalization” and “Language Education,” and see the sorts of things others are doing in their own communities from which inspiration can be drawn. Alternatively, a community member might be interested in learning more about how members of other communities are integrating new technology into their revitalization and/or documentation efforts. The resources tab makes it easy to access all uploaded materials tagged as “Language and Technology,” which displays examples of how others around the world are using different types of technology to similar ends. By far, the majority of uploads thus far are instructional in nature, either focused specifically on language teaching or on providing stories or other documentary-type information which convey language information to younger generations. As the Catalogue and the ELP specifically focus on threatened and endangered languages, the fact that users are employing the language material upload function to disseminate instructional and language revitalization materials digitally is particularly encouraging, and serves to benefit not only that language group, but others trying to create similar programs. Resources may also be filtered by tags, which are labels meant to help further refine search results, or to help users who are interested in a more specific resource type. Multiple tags may be added to a single search. For example, a search for the tag “language immersion” produces a list of resources tagged as pertaining to some type of immersive language instruction. These items can further be refined by adding other tags, for example the addition of “Hawaiian” will narrow the list to resources involving Hawaiian language immersion. Resources may also be filtered by their format (image, video, document, audio, or link). Currently the majority of the resources on the ELP are audio/visual, either videos or audio accompanied by displayed text. However, there are also a large number of links to websites, some of which have extensive information on particular languages, access to a collection of language lessons, an online dictionary, etc. A final aspect of the infrastructure surrounding resources on the ELP site that bears mention is that individual resources can be given a “community rating.” This feature is specifically designed to give endangered-language communities additional autonomy over the resources that are uploaded for their language. This is a way in which members/speakers can publicly endorse (or down-vote) certain resources as good or poor representations of their language, which serves as an indication of quality to other users who may be less well informed.

Discussion forums A primary function of the ELP site outside of the information associated with the Catalogue is to give community members and stakeholders a voice on the site. One way in which this is accomplished is through the discussion section at the bottom of each individual language page. In these discussion forums, users can talk with each other about things going on with a particular language in real-time, share news announcements, announce upcoming community events, and so on. The discussion forums also provide a venue to share things like anecdotes, stories and personal experiences that might not fit neatly into the language information side of the site or the language material uploads. To keep up to date with the discussion forums, users may “subscribe” to a particular language page, in order to receive a notification when anything 256

How ELCat serves language communities

is posted to the forum. Since the discussion forums are open to any ELP user, not only community members, this has also been a place where users can pose questions to the Catalogue researchers or people who are knowledgeable about the language. As such, the ELP provides more than a central location for the sharing of language material.

User feedback One of the most important features of the ELP is the ability for users to provide feedback on the information and resources provided on the site and in the Catalogue. All language pages have a tab labeled “suggest a change,” which links users to an electronic form where they can identify an issue in the current information, recommend their desired change, and provide any bibliographic references for said change. Once submitted, suggested changes are reviewed by a team of researchers at the Catalogue who, based on the recommendation of experts in that language or region, may decide to implement the change, partially implement it, or decline to implement it. If the change is declined, an email is sent to the suggesting user with an explanation of why their suggestion was not implemented. Suggested changes are typically addressed within a week or two, depending on the amount of additional research required by the research team to vet each suggestion. By providing a direct pipeline between users and Catalogue researchers, the “suggest a change” feature allows the ELP and the Catalogue to be a nimbler and more responsive resource than other compendia of language information. Importantly, it also allows the scholarly information to reflect the often more up-to-date knowledge of community members about the state of their language. To date, there have been thousands of user suggestions processed that have provided a wide variety of feedback and refinements, for example alerting us to improperly tagged language material uploads or to offensive language names, suggesting additional resources for inclusion, giving corrections or additions to geographical data, and providing information on language revitalization efforts that were previously not listed in the Catalogue database. In addition to allowing the Catalogue to be more agile and to take advantage of the wealth of information that can be accumulated by allowing users to upload to the site, this information exchange tool allows those invested in a particular language to attain an additional degree of control over the information and resources presented for that language on the site. Users in general, and endangered language community members more specifically, play an important role in making the Catalogue and the ELP a comprehensive, up-to-date, and dynamic resource on endangered languages which serves the needs of a wide variety of audiences.

Conclusions Often with an endeavor such as the Catalogue it is all too easy to start viewing languages solely in terms of numbers and statistics, and forget that these are living systems of cultural knowledge and communication, which reflect – and indeed, in part constitute – the identities of those who speak them. The ability of users to view and interact with accurate language data and examples of a language actually being used serves at once to put the facts and figures that the Catalogue provides into human context, and to give those who speak the language or whose heritage language it is a voice on the site. Both of these ends are crucially important to the goals of the ELCat project, and make it a valuable resource for all who are interested in language endangerment.

257

Raina Heaton and Sean Simpson

References Dryer, Matthew S. & Haspelmath, Martin (eds.). 2013. The world atlas of language structures. Leipzig: Max Planck Institute for Evolutionary Anthropology. (http://wals.info) Fishman, Joshua. 1991. Reversing language shift: Theoretical and empirical foundations of assistance to threatened languages. Clevedon: Multilingual Matters. Fishman, Joshua. 2001. Can threatened languages be saved? Reversing language shift, revisited: A 21st century perspective. Clevedon: Multilingual Matters. Hammarström, Harald & Forkel, Robert & Haspelmath, Martin & Bank, Sebastian. 2016. Glottolog 2.7. Jena: Max Planck Institute for the Science of Human History. (http://glottolog.org) NeSmith, Keao. 2016. Proposal to establish policies and guidelines for the county of Kaua’i regarding the use of Hawaiian language. County of Kaua‘i. Simons, Gary F. & Fennig, Charles D. (eds.). 2017. Ethnologue: Languages of the world, twentieth edition. Dallas: SIL International. (www.ethnologue.com)

258

9 TRIAGE Setting priorities for endangered language research Bryn Hauk and Raina Heaton

As a comprehensive resource on the world’s endangered languages, the Catalogue of Endangered Languages (ELCat) allows us to make data-driven recommendations, establish priorities, and target efforts to document endangered languages. This chapter identifies endangered languages of especially high priority for research efforts based on criteria measurable in ELCat. Prioritization is based on language vitality (especially intergenerational transmission) and on the availability of existing descriptive material. The list of high-priority languages is further narrowed by spotlighting language isolates and members of poorly documented language families, as well as by assigning lower priority to comparatively welldocumented languages and dormant languages.

Introduction As a comprehensive resource on the endangered languages of the world, the Catalogue of Endangered Languages (ELCat) allows us to make data-driven recommendations, establish priorities, and target efforts to document endangered languages. The goal of this chapter is to identify endangered languages of especially high priority for research efforts based on those criteria that are measurable in ELCat. Specifically, language research will be prioritized based on language vitality (especially on intergenerational transmission) and on the availability of existing descriptive material. The list of high-priority languages will be further narrowed by spotlighting language isolates and members of poorly documented language families, as well as by assigning lower priority to particularly well-documented languages, and languages that have already lost their last speakers.1

Language “triage” It should be uncontroversial that linguistic diversity is precious and important, and therefore documenting endangered languages is of high priority. A much less savory task is to perform triage upon the world’s 3,138 still-spoken endangered languages with the goal of sorting out which imperiled, neglected, and under-resourced languages deserve attention before others. 259

Bryn Hauk and Raina Heaton

“Triage” in medical usage is the assignment of degrees of urgency to wounds or illnesses in order to decide the order of treatment of patients or casualties, and it is precisely in this sense that we apply the term here to endangered languages. Triage implies, given limited resources, that some endangered languages must be left longer in the waiting room than languages of higher priority, pending the availability of qualified researchers and funding. Certainly no one wants to suggest that some endangered languages are not valuable or do not deserve treatment. Yet when resources are scarce, qualified researchers are few, and the number of languages in need of attention is high, prioritization is important for crucial decisions that must be made about how to achieve the maximum positive effect with what is available. A number of volumes on language fieldwork or endangered languages contain some guidelines for prioritizing language research. Austin and Sallabank (2011) recommend giving top priority to languages that are nearly extinct, languages where revitalization is still possible, language isolates, and languages that have not been studied by outsiders. Anderson (2011) identifies roughly 20 language “Hotspots” – regions of the world with especially high linguistic diversity and endangerment levels, analogous to hotspots of biodiversity. He suggests these regions should be given high priority for language documentation efforts and further identifies especially high priority languages within each region, by the same criteria as Austin and Sallabank (2011): language endangerment, taxonomic classification, and amount of previous documentation. In a chapter dedicated to choosing a fieldwork language, Chelliah and de Reuse (2011) discuss three possible scenarios: (1) an outsider chooses the language for the fieldworker (for example, a faculty advisor for a student); (2) the language community chooses the fieldworker; or (3) the fieldworker chooses the language. Clearly, many would hold the second scenario as highly desirable, wherein a language community decides among themselves what kind of documentation project they want and then partners with a linguist or linguists (or other appropriately trained professionals) to accomplish their goals. However, the third scenario, in which the fieldworker chooses the language, is probably the most common, and it is here, as well as in scenario (1), that a prioritized list of endangered languages such as in this chapter can prove particularly useful. Chelliah and de Reuse (2011), building on Dwyer (2006), detail several factors that will impact how a linguist would select a language project: linguistic diversity and/or conservativeness, political expediency, logistical expediency, interpersonal expediency, and personal expediency. The first factor, linguistic diversity and conservativeness, echoes criteria mentioned above, giving top priority to undescribed languages, language isolates, or languages with unique or unusual typological characteristics. The authors explicitly state that the linguist’s own interest and background will impact his or her choice: the researcher may be interested in a particular feature or phenomenon, or may be drawn towards a language related to or similar to some other language with which he or she has previous experience. Such a choice would be expedient for being able to build off previous experience to document the new language more efficiently. The other factors that Chelliah and de Reuse (2011) discuss – political expediency, logistical expediency, interpersonal expediency, and personal expediency – are external to the languages themselves. Politically, a linguist might find it difficult to gain access to a certain location or to get permission to do research depending on the government of the host country and his or her own citizenship. Logistically, a field site could be inaccessible due to climate, terrain, or location, or funding for certain field sites or languages could be difficult to obtain. In terms of interpersonal expediency, another linguist might already be working on descriptive materials for a particular language; in such a situation a linguist in search of a project might feel he or she has to ask for “permission” to work on that same language. Of course, no one 260

Triage: Setting research priorities

can claim ownership of a language, although it is polite (and most likely beneficial) to consult with others who are engaged in research with the particular language in question whenever feasible. Finally, under personal expediency, Chelliah and de Reuse remark that a linguist’s choice in a fieldwork project will be impacted by “his/her religion, national or racial background, political views, gender, handicap, health, sexual orientation, his/her ability to deal with unusual climates, diets, long-distance and uncomfortable travel, or psychologically stressful situations, and his/her interests or tastes” (2011:84). Together, sources such as these recommend the same general factors for assigning highest priority to languages for documentation: languages whose communities are interested in a documentation project or where revitalization is possible, linguistic isolates, highly endangered languages, undocumented or under-documented languages, and unusual languages. In theory, then, it should be straightforward to apply these criteria and produce a list of languages in order of priority from a resource such as ELCat, which tracks some of these criteria for the world’s endangered languages. However, a number of external factors ultimately may shuffle the list for any given researcher, meaning that recommendations based on ELCat can only be part of the equation. When attempting to prioritize these languages, we must ask, “prioritize for whom?” The recommendations made will necessarily vary depending on whose priorities are considered. For a person who senses that his mother tongue is losing ground in the community – who notices that the youth prefer a different language, or who fears that the elders’ verbal artistry will soon be lost – his own language will likely receive top priority, regardless of whether the language is critically endangered or is at some lower level of risk. For funding agencies, the highest priority would most likely be how to deploy their limited resources for maximum benefit and greatest impact. From the linguist’s perspective, priority is determined by one’s research direction, skills, training, and prior experience. For example, village (or indigenous) sign languages, which arise spontaneously in communities with a higher-than-average incidence of deafness, should be of high priority by the criteria listed above: they are, by nature, linguistic isolates (usually undocumented), and since their lifespan tends to reach only a few generations, they are highly endangered (Nonaka 2009). However, it would be a mistake to recommend that village sign languages be the top documentation priority for all linguists. Linguists who are not trained in sign languages could perhaps do more harm than good; they should not take on a sign language documentation project for which they are unprepared simply because a particular sign language is a top priority. Generally, a linguist should aim for a documentation project for which he or she is prepared, which in some cases may mean selecting a language not of the very highest priority; for example, a language in a language family with which the researcher is familiar, or a language where speakers are bilingual in a contact language (a language of wider communication) that the researcher knows and can use in the documentation project, or a spoken language instead of a sign language (or vice versa). Obviously, a database such as ELCat is not equipped to weigh these sorts of personal factors that a linguist must weigh in choosing a documentation project. Therefore, in making our recommendations for ranking the priority of endangered languages, we assume a fictitious, well-trained fieldworker aiming to undertake the most urgent documentation work needed at this time regardless of personal preference. We consider only the metrics that ELCat can provide information on: namely, endangerment status, intergenerational transmission, classification (isolate vs. member of larger family of languages), amount and recency of data or documentation on the language, and the location of the language (whether it is in a safe and accessible location). In doing this, we hope to draw attention to some of the world’s least known and 261

Bryn Hauk and Raina Heaton

least studied languages, that might otherwise be passed over as candidates for a documentation project. Readers of this chapter will have to re-evaluate these recommendations according to their own criteria, needs, and interests.

Setting priorities for research on endangered languages All 3,138 still-spoken endangered languages in the ELCat database deserve and need more linguistic investigation. However, if one goal is to gather as much information as we can about languages before they disappear, those languages that are on the verge of becoming dormant are the most in need of immediate attention. This is the first and most obvious way of approaching the triage problem: setting language research priorities according to the vitality of the languages in question. A second way involves targeting languages according to the amount of existing documentation on them. If we do not have enough information on a language even to make a reasonable assessment of how endangered it is, then the immediate necessity is to find out more about this language, then to factor its vitality into the determination of how endangered it is and what level of priority might be assigned to it. The sections on “Priority by endangerment level” and “Priority by available information” use the ELCat database to discuss information and endangerment, and how language research can be prioritized based on these criteria.

Priority by endangerment level Lee and Van Way (Chapter 5) discuss the Language Endangerment Index (LEI), the algorithm that produces vitality ratings for the languages in ELCat. It is designed to account for multiple factors and produce the best estimate of vitality based on the information available in a way that is consistent and comparable across languages. It has seven levels of vitality2: “Safe,” “At Risk,” “Vulnerable,” “Threatened,” “Endangered” (moderately), “Severely Endangered,” and “Critically Endangered.” With respect to our goal to focus on those languages most imminently in danger of disappearing, those languages in most need of documentation are those which are ranked “Severely Endangered” and “Critically Endangered.” ELCat lists 371 languages as “Severely Endangered” and 427 as “Critical,” for a total of 798 languages that will likely cease to be spoken with the death of the current generation of speakers. Some of these languages may already have disappeared, but have not yet been reported as dormant (see “Projected vitality”). The list of these 798 languages is given in Table 9.1, and can also be found on the www.endangeredlanguages.com website by searching by vitality rating and deselecting (turning off) all ranks except “Severely Endangered” and “Critically Endangered” from the menu. Note that all data discussed in this chapter reflect the ELCat database as of April 4, 2017. The 798 “Critically” and “Severely Endangered” languages are organized in Table 9.1 by name and region.

262

Triage: Setting research priorities Table 9.1 Languages in ELCat listed as “Severely” or “Critically Endangered” Code

Language

Vitality

Code

AFRICA huc ayu

bdn bsr bng bxv byf byb bwh boy bmf bnl bvf bbx bvh bso bju cet dmm dam afn doy dbo ega elh fah fgr gev goy juh idr jek jmi kbs ccj kkm kqz krm kov gnh luw

ǂHoan Ayu Baka (Far North Region, Cameroon) Baldemu Bassa-Kontagora Benga Beni Iznassen Berakou Bete Bikya Bishuo Bodo Bom Boon Boor Bubia Bure Buso Busuu Buu (Nigeria) Centúúm Dama Damakawa Defaka Dompo Dulbu Dyarim Ega El Hugeirat Fali of Baissa Fongoro Ganjulé Geviya Goundo Hõne Hwarasa Indri Jeri Kuo Jimi (Nigeria) Kande Kasanga Kiong Korana Krim Kudu-Camo Lere Luo

Language

Vitality

NORTH AMERICA Severe Severe Severe

acv aht arp ari cad chl cay chn crd col com crx, caf bea kkz tfn dtd coo git gwi hei haa

Achumawi Ahtna Arapaho Arikara Caddo Cahuilla Cayuga Chinook Wawa Coeur d'Alene Columbian Comanche Dakelh (Carrier) Dane-Zaa (Beaver) Danezāgé' (Kaska) Dena'ina Diitiidʔaatx̣ (Nitinat) Éy7á7juuthem (Comox) Gitsenimx̱ Gwich'in Hailhzaqvla (Heiltsuk) Han Hare hps Hawai‘i Sign Language haw Hawaiian hid Hidatsa hup Hupa ing Ingalik iks Inuit Sign Language ipk Inupiaq fla, spo Kalispel-Spokane-Pend d’Oreille-Salish kyh Karuk kju Kashaya xaw Kawaiisu kio Kiowa cku Koasati Kodiak Russian Creole mjd Konkow koy Koyukon kut Ktunaxa lkt Lakota lou Louisiana Creole taa (Lower) Tanana lut, sno, Lushootseed ska mrc Maricopa nsr Maritime Sign Language mez Menominee

Critical Critical Severe Severe Critical Severe Critical Critical Severe Critical Severe Severe Severe Critical Severe Critical Severe Critical Critical Critical Severe Severe Severe Severe Severe Severe Critical Critical Severe Severe Critical Severe Critical Severe Severe Severe Severe Critical Severe Critical Critical Severe Critical Critical

Critical Critical Severe Critical Critical Critical Severe Critical Critical Critical Severe Severe Severe Severe Severe Critical Critical Severe Severe Critical Critical Severe Critical Severe Critical Critical Severe Severe Severe Severe Critical Critical Critical Critical Severe Critical Critical Critical Severe Severe Severe Critical Critical Severe Critical Severe continued

263

Bryn Hauk and Raina Heaton Table 9.1 continued Code

Language

Vitality

Code

ldd mfz muj buy mdg mru kqx ngh noz gke nbd nnx nin njr noy nyp

Luri Mabaan Mabire Mani Massalat Mono (Cameroon) Mser N||ng Nayi Ndai Ngbinda Ngong Ninzo Njerep Noy Nyang'i Nyanjang Odut Okiek Omotik Ongota Pam Plapo Krumen Putai Sambe Shabo Sheni Somyev Soo Tchumbuli Tetserret Twendi Yaaku Yangkam Yeyi Zaramo Zumaya

Critical Severe Critical Severe Severe Severe Severe Critical Severe Critical Critical Critical Critical Critical Severe Critical Critical Severe Severe Severe Critical Severe Severe Severe Critical Severe Critical Severe Severe Severe Severe Severe Severe Critical Severe Critical Critical

crg mov mnr umu mus

oda oki omt bxe pmn ktj mfl xab sbf scv kgt teu bqa twn muu bsx yey zaj zuy

Michif Mojave Mono (United States) Munsee Muskogee Nēhiyawēwin nez Nez Perce ncg Nisga'a thp Nłeʔkepmxcín (Thompson) pao Northern Paiute str Northern Straits Salish ttm Northern Tutchone nuk Nuuchahnulth (Nootka) blc Nuxalk oka Okanagan-Colville oma Omaha-Ponca ono Onondaga otw Ottawa ems Pacific Yupik par Panamint pwi Patwin psd Plains Indian Sign Language pot Potawatomi yum Quechan waa, tqn, Sahaptin uma, yak sac Sauk-Fox see Seneca sjw Shawnee sec She shashishalhem (Sechelt) csm, nsq, Sierra Miwok skd squ Sḵwx̱wú7mesh sníchim (Squamish) tsi Sm̓algya̱x (Coast Tsimshian) peq Southern Pomo tce Southern Tutchone lil St̓át̓imcets (Lillooet) tgx Tagish tht Tāłtān (Tahltan) tcb Tanacross Tipai tli Tlingit sek Tse'khene (Sekani) clc Tsilhqot'in (Chilcotin) srs Tsuut'ina tub Tubatulabal tus Tuscarora kuu Upper Kuskokwim

AUSTRALIA wmi alh aid amg ajn ant adg aea amz ayd

Agwamin Alawa Alngith Amurdak Andajin Antakirinya Antekerrepenh Ariba Atampaya Ayapathu

Language

Critical Critical Critical Critical Critical Severe Critical Critical Critical Critical

264

Vitality Critical Critical Critical Critical Severe Severe Critical Severe Severe Critical Critical Severe Critical Critical Critical Severe Critical Severe Severe Critical Critical Critical Critical Critical Critical Severe Critical Severe Critical Critical Critical Critical Critical Severe Critical Critical Critical Critical Severe Critical Critical Severe Critical Critical Critical Critical

Triage: Setting research priorities Table 9.1 continued Code jbi bdy rbp bcj bpt wdj bym

Language

Badjiri Bandjalang Barababaraba Bardi Barrow Point Batjamalh Bidjara Bilinarra Binbinka bck Bunuba dda Dadi Dadi ngk Dalabon tbh Dharawal dax Dhay'yi dif, dit Diyari dyy Djabugay tjw Djabwurrung djb Djinba Djuwarliny dbl Dyirbal zmr, amy Emmi err Erre fln Flinders Island gdh Gajirrabeng gma Gamberre kld Gamilaraay gcd Ganggalida Garig wrk Garrwa Girramay giy Giyug lja Golpa gni Gooniyandi gdc Gugu-Badhun kgs Gumbaynggir gyf Gungabula kgl Gunggari gww Gunin gyy Gunya gnr Gureng Gureng gge Gurr-goni kky Guugu-Yimidhirr gwu Guwamu ilg Ilgar djd Jaminjung ddj Jaru djw Jawi djn Jawoyn jig Jingulu

Vitality

Code

Language

Vitality

Critical Severe Critical Critical Critical Severe Severe Severe Critical Critical Severe Critical Critical Critical Severe Critical Critical Severe Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Severe Severe Critical Severe Severe Critical Severe Critical Severe Severe Critical Critical Critical Severe Critical Critical Severe Severe Critical Critical Severe

tau was abe bcr hdn, hax has

Upper Tanana Washo Western Abenaki Witsuwit'en Xaad Kil (Haida) X̄enaksialak̓ala / X̄a''islak̓ala (Haisla) Yokuts Yuchi

Critical Severe Critical Severe Critical Critical

yok yuc

Critical Critical

PACIFIC abg aob akq akr awm aqr aok aac aua bnr dix aya bek bie brj bql tls bpw bmp afi dui grq grw aek ksi klp xrw kcb xow faj luf lrz lib lzl lnn lva

Abaga Abom Ak Araki Arawum Arhâ Arhö Ari Asumboa Ati (Vanuatu) Aveteian Awar Bebeli Bepour Bieria Bilakura Biliru Bo Bulgebi Chini Daruru Dumun Farnanto Gorovu Gweda Haeke I'saka Kamasa Karawa Kawacha Kowaki Kursav Laua Lemerig Likum Litzlitz Lorediakarkar Maku'a

Critical Critical Severe Critical Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Critical Severe Severe Severe Critical Severe Severe Severe Critical Critical Critical Severe Critical Severe Critical continued

265

Bryn Hauk and Raina Heaton Table 9.1 continued Code

Language

Vitality

Code

Language

Vitality

kbe xmu gbd vka

Kaanju Kamu Karajarri Kariyarra Kartujarra Kayardild Kija Kokatha Koko-Bera Kugu-Nganhcara Kuku-Mangk Kuku-Mu'inh Kuku-Muminh Kuku-Ugbanh Kuku-Uwanh Kunbarlang Kunggara Kunjen Kurrama Kurtjar Kuthant Kuuku-Ya'u Kuwema Ladji Ladji Lamalama Lardil Larrakiya Limilngan Liyagalawumirr Liyagawumirr Malak Malak Manda (Australia) Mandandanyi Mandjalpingu Mangala Mangarrayi Mangerr Margany Maridan Maridjabin Marimanindji Mariyedi Marra Marrgu Marriammu Marringarr Marrithiyel Martuthunira Mati Ke

Critical Critical Critical Critical Critical Critical Severe Critical Severe Severe Critical Critical Critical Critical Severe Severe Critical Critical Critical Critical Critical Severe Severe Severe Critical Critical Critical Critical Severe Severe Critical Critical Critical Severe Severe Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical

pmr ate mkv mjj

Manat Mand Mavea Mawak Mbwenelang Moere Mosimo Munkip Mwesen Nagu Najit Nasarian Nāti Nauna Navwien Nese Ngatik Men’s Creole Niolean Nitita Nivat Njav Olrat Orkon Oroha Ouma Papapana Papi Pei Penrhyn Piame Puare Pwapwâ Rapa Rema Rennell Island Sign Language Retlatur Ririo Samosa Se Sene Siawi Som Sörsörian Sumariup Susuami Taiap Tanema Tanimbili Tape

Severe Critical Critical Severe Critical Severe Severe Severe Critical Severe Critical Critical Severe Severe Critical Critical Severe Severe Critical Critical Critical Critical Severe Severe Critical Severe Severe Severe Severe Severe Severe Severe Critical Severe Critical Severe Severe Severe Severe Severe Severe Severe Critical Severe Severe Severe Critical Severe Critical

gyd gia ktd kkp wua xmq xmp xmh ugb uwa wlg kvs kjn vku gdj xut kuy woa llj lby lbz lrg lmc

mpb zma zmk mem mpc zme zmc zmd zmj zmm zmy mec mhg xru zmt mfr vma zmg

mvq mqv mpv msn ngr nvh ncn

ngm rpn

olr ora oum ppn ppe ppq pnh pin pux pop ray bow rsi rri swm sej mmp smc siv ssu gpn tnx tbe mrs

266

Triage: Setting research priorities

1111 2 3 4111 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8111 9 20111 1 2 3 4 5 6 7 8 9 30111 1 2 3 4 35 6 7 8 9 40111 1 2 3 4 5 6 7 8EEE

Table 9.1 continued Code

Language

Vitality

Code

Language

Vitality

zml xmy zmv mep

Matngele Mayi-Kutuna Mbariman-Gudhinma Miriwoong Mirniny Mudburra Muluridyi Muruwari Nakara Narungga Ngaatjatjara Ngamini Ngandi Ngardi Ngarinyin Ngarla Ngarluma Ngarnka Ngawun Ngiyambaa Ngomburr Nhanta Nhuwala Nimanburru Nungali Nyamal Nyawaygi Nyikina Nyiyaparli Nyulnyul Olkol Paakantyi Pakanh Palyku Panyjima Payungu Pini Putijarra Rembarrnga Tagalaka Thalanyji Tharrkari Tjungundji Tjupany Umbuygamu Umpila Umpithamu Uradhi Urningangga

Critical Critical Critical Critical Critical Critical Critical Critical Severe Severe Severe Critical Critical Severe Critical Critical Severe Critical Critical Critical Critical Critical Severe Critical Critical Critical Critical Severe Severe Critical Critical Critical Severe Critical Severe Critical Severe Critical Severe Critical Critical Critical Severe Severe Critical Critical Critical Critical Severe

tkw tns tox rga trh tqm tww yla uur vnk val

Teanu Tenis Tobian Toksiki Turaka Turumsa Tuwari Ulwa Ura (Vanuatu) Vano Vehes Vivti Volow Walman Wiaki Wiarumus Yawiyo Yeri Zazao

Severe Severe Severe Severe Severe Critical Severe Severe Critical Critical Severe Critical Critical Severe Critical Severe Severe Severe Critical

dmw vmu zmu nck nnr nmv nid rxd ung nlr nrl nji nxn wyb nrx nha nhf nmp nug nly nyt nyh xny nyv olk drl pkn nad pnw bxj pii rmb tgz dhl dhr

umg ump umd urf urc

mlv van wii tua ybx yev jaj

SOUTH AMERICA guq aqz amc aqt arl ark arx avv bae brg boa cbd ccc cod gva gta huh inp iqu irn isc ite ito jbt jru jua

Aché Akuntsú Amahuaca Angaité Arabela Arikapú Aruá Avá Baré Baure Bora Carijona Chamicuro Cocama-Cocamilla Diahói Guana (Paraguay) Guató Huilliche Iñapari Iquito Irantxe Isconahua Isolado do Tanaru Itene Itonama Jabutí Japréria Júma

Severe Critical Severe Severe Critical Critical Critical Severe Critical Critical Severe Critical Critical Critical Critical Critical Critical Severe Critical Critical Severe Critical Critical Severe Critical Severe Severe Critical continued

267

Bryn Hauk and Raina Heaton Table 9.1 continued Code

Language

Vitality

Code

Language

Vitality

wny wdu waq wbv wkw, wur wga wmb wnm wgg

Waanyi Wadjigu Wagiman Wajarri Waka-Waka

Critical Critical Severe Severe Critical

Wakaya Wambaya Wanggamala Wangganguru Wangkajunga Wardaman Warlmanpa Warluwarra Warnman Warray Warrgamay Warrwa Warumungu Wergaia Wik-Epa Wik-Iiyanh Wik-Keyangan Wik-Me'anha Wik-Ngatharr Wilawila Wiradjuri Worla Wudjari Wulna Wunambal Yandruwandha Yangkaal Yangman Yankunytjatjara Yanyuwa Yawarrawarrka Yawuru

Severe Critical Critical Critical Severe Severe Severe Critical Severe Critical Critical Critical Severe Critical Critical Severe Critical Severe Severe Critical Critical Critical Critical Critical Severe Critical Critical Critical Critical Severe Critical Critical

Yidiny Yiiji Yindjibarndi Yindjilandji Yinhawangka Yir-Yoront Yugambal Yulparija Yuwaalaraay

Severe Severe Severe Critical Critical Severe Critical Severe Critical

caw Kallawaya kxo Kapixaná kuq Karipuna kbb Katxúyana alc Kawésqar xor Korubo kqq Krenak xpk Kulina xwa Kwaza qvj Loja Quichua arw Lokono mpu Makurap mcg Mapoyo mzx, mpw Mawayana skf Mekéns bmr Muinane hux Nipode oca Ocaina opy Ofayé omg Omagua ore Orejón orw Orowari qvp Pacaraos Quechua pbg Paraujano pnk Paunaka psm Pauserna pur Puruborá rgr Resígaro rey Reyesano sae Sabané slc Sáliva swo Shanenawa jeb Shiwilu srq Sirionó Taiwano tae Tariana tdm Taruma trr Taushiro eh Tehuelche tqb Tembé tit Tinigua tpy Trumai Tunayana uks Urubú-Kaapor Sign Language wyr Wayoró xet Xetá yag Yagan yaw Yawalapití

Severe Critical Severe Severe Critical Severe Severe Severe Severe Severe Severe Severe Critical Critical Critical Severe Severe Critical Severe Critical Critical Critical Severe Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Critical Severe Critical Critical Criticalt Critical Severe Critical Critical Critical Critical

wrr wrl wrb wbt wrz wgy wwr wrm weg wie wij wif wih wik wil wrh

wux wub ynd nny jng kdd jao yww ywr, dyd yii yij yil ywg yiy yub

268

Critical Critical Critical Critical

Triage: Setting research priorities Table 9.1 continued Code

Language

Vitality

CAUCASUS bbl Batsbi bhn, hrt Bohtan Neo-Aramaic ttt Muslim Tat

Severe Severe Severe

gqu giq ili xnb kxs ckv

yln laq mnc mvi giu gld tji orh mzs uun pyu gir sxr xsy ysd ysy trv shx ryu tjs tao ssf

Ainu (Japan) Anong Ayizi Baha Buyang Bolyu Duoxu Gao Gelao Green Gelao Ili Turki Kanakanabu Kangjia Kavalan Khamnigan Mongol Kilen Laji Langnian Buyang Laqua Lopnor Uighur Manchu Manchurian Kirghiz Miyako Mulao Nadouhua Nanai Northern Tujia Ongkor Solon Oroqen Patuá Pazeh-Kaxabu Puyuma Red Gelao Saaroa Saisiyat Samatao Samatu Sanie Seediq She South-Central Okinawan Southern Tujia Tao Thao

Language

Vitality

yar zro

Yawarana Záparo

Critical Critical

SOUTH ASIA bhj brd byw

EAST ASIA ain nun yyz yha ply

Code

Critical Critical Critical Severe Severe Critical Severe Severe Critical Critical Severe Critical Severe Severe Critical Severe Severe Critical Critical Critical Severe Critical Severe Severe Severe Critical Severe Critical Critical Severe Critical Critical Severe Critical Critical Severe Critical Severe Severe Severe Severe Critical

cuw dmk dus emg ksu kgg lep gac ole ruh raq rav tdh tij

Bahing Baram Belhariya Chorei Chukwa Domaaki Dumi Eastern Meohang Khamyang Kusunda Lepcha Mixed Great Andamanese Olekha Ruga Saam Sampang Tangam Thulung Tilung

Critical Critical Severe Severe Critical Severe Critical Severe Critical Critical Severe Critical Critical Critical Severe Severe Severe Severe Critical

SOUTHEAST ASIA dul amq aem atz atm mbf khd mfb beg bhw bgk bpg bdx aip cwg cog cnc tds dbn dsn hti huw

Alabat Island Agta Amahai Arem Arta Ata Baba Malay Bädi Kanum Bangka Belait Biak Bih Bit Bonggo Budong-Budong Burumakok Chewong Chong Côông Doutai Duriankere Dusner Hoti Hukumina

Critical Critical Critical Critical Critical Critical Severe Critical Severe Severe Critical Severe Severe Severe Severe Severe Severe Severe Severe Severe Critical Critical Critical continued

269

Bryn Hauk and Raina Heaton Table 9.1 continued Code

Language

Vitality

Code

Language

Vitality

giw yoi jiu

Tuha White Gelao Yonaguni Youle Jinuo

Critical Severe Severe Severe

huk ibu tyh ire agk inn itx jvd kzx

Hulung Ibu Iduh Iresim Isarog Agta Isinay Itik Javindo Kamarian Kasong Kayeli Kayupulau Kehu Kembra Kofei Kri Kwer Kwerisa Laha (Vietnam) Lengilu Liabuku Liki Lolak Mander Mang Maremgi Masep Masimasi Mer (Indonesia) Mintil Mor Mount Iraya Agta Naka'ela Namla Narau Nedebang Nusa Laut Orang Kanaq Original Bangkok Sign Language Original Chiangmai Sign Language Paulohi Pear Petjo Piru Punan Batu 1 Ratagnon Salas

Critical Severe Severe Severe Critical Critical Severe Critical Critical Severe Critical Severe Severe Severe Severe Severe Severe Severe Severe Critical Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Severe Critical Critical Severe Severe Critical Severe Severe Critical

EUROPE sia alr

clw

eve enf yux izh ist itl itk kdr krk sjd jct fkv lud

neg nio oac oaa sje niv cjs sms sjt kim tsd enh ykg ude ulc sju

Akkala Saami Alutor Amur Nivkh Baraba Tatar Central Selkup Chulym Eastern Khanty Eastern Mansi Even Forest Enets Forest Yukagir Gardiol Ingrian Istriot Itelmen Judeo-Italian Karaim Kerek Kildin Saami Krymchak Kven Finnish Ludian Mardin Sign Language Middle Chulym Negidal Nganasan Norman Oroch Orok Pite Saami Poitevin Sakhalin Nivkh Shor Skolt Saami Southern Selkup Ter Saami Tofa Tsakonian Tundra Enets Tundra Yukagir Udege Ulch Ume Saami

Critical Severe Severe Critical Critical Severe Critical Severe Severe Severe Critical Severe Severe Severe Severe Severe Critical Critical Severe Critical Severe Critical Severe Severe Severe Severe Severe Severe Critical Critical Critical Severe Severe Critical Critical Severe Critical Severe Critical Critical Critical Critical Critical

kzl kzu khh xkw kpi kwr kkb lha lgi lix lio llq mqr zng mrx mvs ism mnu mzt moq atl nae naa nxu nec nul orn

csd plh pcb pey ppr pnm btn sgu

270

Critical Severe Severe Critical Severe Severe Critical Severe

Triage: Setting research priorities Table 9.1 continued Code

Language

Vitality

Code

Language

Vitality

vep vot

Veps Votic Yazva Yug

Severe Critical Critical Critical

sxm spi spg kxq smu ays krz szp pee tni tcz thm tti ugo wor

Samre Saponi Sian Smärky Kanum Somray Sorsogon Ayta Sota Kanum Suabo Taje Tandia Thado Chin Thavung Tobati Ugong Woria

Critical Critical Severe Severe Critical Severe Severe Critical Critical Critical Severe Severe Severe Critical Critical

yuu

MEXICO, CENTRAL AMERICA, CARRIBEAN zoo zoq itz ixc klb dih clo mhc plo

ppi ppl rma ztm scf poq

Asunción Mixtepec Zapotec Ayapanec Zoque Itza' Ixcatec Kiliwa Kumeyaay Lowland Chontal Mocho' Oluta Popoluca Original Costa Rican Sign Language Paipai Pipil Rama San Agustín Mixtepec Zapotec San Miguel Creole French Texistepec

Severe Critical Severe Critical Critical Critical Severe Critical Severe Critical Severe Critical Critical Severe Critical Critical

NEAR EAST avd bhm

aij

xkc mid sts

Alviri-Vidari Baṭḥari Central Jewish NeoAramaic Garma'i Inter-Zab Jewish NeoAramaic Jowshaqani Kho'ini Kuhpayi Neo-Mandaic Shumashti

Severe Severe Critical Severe Severe Severe Severe Severe Severe Severe

271

Bryn Hauk and Raina Heaton

In addition to the languages above which have an overall vitality rating of “Severely” or “Critically Endangered,” there are other languages that bear consideration here as well. It is also important to take into account the primary factor determining language vitality: intergenerational transmission, i.e. if the language is being passed on to children. The Language Endangerment Index (LEI) divides intergenerational transmission into six levels (0–5): 0 – All members of the community/ethnic group speak the language. 1 – Most adults and some children are speakers. 2 – Most adults of child-bearing age speak the language. 3 – Some of child-bearing age know the language, but do not speak it to the children. 4 – Many of the grandparent generation speak the language. 5 – Few speakers, all elderly. The languages with intergenerational transmission levels falling in the 3–5 range constitute cases where children are no longer learning the language, which means that the language could potentially be lost very quickly. All of the languages in Table 9.1 that are severely or critically endangered have intergenerational transmission ratings of 3 or below. However, there are a number of languages that are rated as “Endangered” (and three languages that are “Threatened”) that have intergenerational transmission ratings of 3 or 4, presumably because the language may have a relatively large number of speakers, or it is used in a variety of domains, but is not being transmitted to the next generation. If the language is not being passed on to the children, then we expect the current generation of speakers to be the final one (unless revitalization is undertaken successfully), just as with severely and critically endangered languages, and therefore could disappear just as quickly, despite having a larger speaker base. These additional 62 languages that are considered here to be at an increased risk of disappearing due to a lack of intergenerational transmission are given in Table 9.2, accompanied by their code, endangerment level, region, and intergenerational transmission (IGT) rating. Languages listed here as “Moderate,” as in “Moderately Endangered,” match those in the ECLat database labeled as simply “Endangered.” The 798 “Critically” and “Severely Endangered” languages, plus the 62 languages with 3 or 4 IGT ratings, total 860 languages, which are identified here as highly endangered and therefore merit linguistic “triage.” The highest concentrations of highly endangered languages are along the Pacific coast of North America, the Amazon and Andean regions of South America, northern sub-Saharan Africa, mainland and island southeast Asia (including New Guinea), and Australia. This distribution is of course no accident; it is the result of geopolitical forces including colonialism, conflict, environmental factors, and globalization. How this global distribution of severe language endangerment relates to documentary efforts is the subject of “Lack of accessibility.”

Priority by available information In addition to the 860 highly endangered languages identified above, it is necessary to take into account also those languages for which ELCat has incomplete vitality information. Any lack of basic information about a language indicates a need for documentation. If we are not able to make an informed and accurate evaluation of how safe or endangered a language currently is, then we cannot ensure that it gets adequately documented in time. The acquisition of the missing information for the languages discussed in this section is therefore a priority. 272

Triage: Setting research priorities Table 9.2 “Threatened” and “Endangered” languages with IGT ratings of 3 or 4 Code

Region

Language

Vitality

IGT

mjg-huz frp 08e bdp swh-mwi sir ghl zen mwp nuy gin huz udi kzg, xug, ryn, okn, ams, tkn, yox dta

East Asia Europe Europe Africa Africa Africa Africa Africa Australia Australia Caucasus Caucasus Caucasus East Asia

Huzhu Francoprovençal Vivaro-Alpine Bende Mwini Siri Uncunwee Zenaga Kalaw Kawaw Ya Wubuy Hinukh Hunzib Udi Amami-North Okinawan

Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate

4 4 4 3 3 3 3 3 3 3 3 3 3 3

East Asia East Asia East Asia East Asia East Asia Europe Europe Europe Europe Europe Europe Europe Europe Near East Near East Near East North America North America North America North America North America North America North America Pacific Pacific Pacific Pacific Pacific Pacific South America South Asia South Asia South Asia South Asia South Asia Southeast Asia Southeast Asia Southeast Asia

Dagur Hachijo Jejueo Yaeyama Zauzou Forest Nenets Kumandin Northern Mansi Western Buryat Aragonese Cypriot Spoken Arabic Inari Saami Istro Romanian Senaya Soqoṭri Zoroastrian Dari Blackfoot Cheyenne Mohawk Shoshone Southern Tiwa Stoney Ute Boselewa Bukiyip Bumbita Māori Narmoris Sirva Paumarí Darai Gorum Kachari Majhi Yakha Awera Danau Erokwanas

Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate Moderate

3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

jje rys zal yrk-for 1nt bxr arg acy smn ruo syn sqt gbz bla chy moh shh tix sto ute bwf ape aon mri plb sbq pad dry pcj xac mjz ybh awr dnu erw

continued

Bryn Hauk and Raina Heaton Table 9.2 continued Code

Region

Language

Vitality

IGT

bqy mcm mvd mpz wnk wsa mya-yaw bgx fra-gal lif

Southeast Asia Southeast Asia Southeast Asia Southeast Asia Southeast Asia Southeast Asia Southeast Asia Europe Europe South Asia

Kata Kolok Malaccan Creole Portuguese Mamboru Mpi Wanukaka Warembori Yaw Balkan Gagauz Turkish Gallo Limbu

Moderate Moderate Moderate Moderate Moderate Moderate Moderate Threatened Threatened Threatened

3 3 3 3 3 3 3 3 3 3

At the time of writing, there are 37 languages in the database that have no vitality information available at all. In some of these cases, vitality information is not available because the classification used here does not map onto the numbers given in other sources such as Ethnologue (Simons & Fennig 2017), which identifies more or fewer language-level groupings (e.g. the Quechua complex of languages and dialects; see Barlow and Campbell, Chapter 3, regarding classification). Alternatively, in some cases (particularly for the sign languages), it is difficult to get any accurate estimate of vitality with respect to numbers of speakers (or of signers in this case), domains of use, etc. However, for many of these languages, information on vitality is simply not available, which leaves open the possibility, and even likelihood, that these languages are severely or critically endangered. The languages in ELCat that have no vitality information are listed in Table 9.3, along with their codes. The spoken languages belong

Table 9.3 Languages in ELCat with no vitality information Code

Language

Code

Language

aed bxo bhd-bha cby esh gib goz hsl hir ihb

Argentine Sign Language Barikanchi (Pidgin) Bhalesi Carabayo Eshtehardi Gibanawa Gozarkhani Hausa Sign Language Himarimã Iha Based Pidgin Judeo-Egyptian Arabic Judeo-Isfahani Judeo-Kashani Judeo-Shirazi Judeo-Yazdi Kabatei Kajali Laraos Quechua Lincha Quechua

vmh mqt lso 1qa 05k rat rdb sdg sta srz sqh 0ii zab, ztt, ztj, zaw, zaq, zpf tov vsl zad, zav, zpu, zpq, ztc, zat

Maraghei Mok Original Laos Sign Language Penange Puiron Razajerdi Rudbari Savi Settla Shahmirzadi Shau Shirawa Tlacolula Upper Taromi Venezuelan Sign Language Villalta Western Baja Mixtec Zimatlán

xkp xkj qux-lar qux-lin

zph, zpp, zpl

274

Triage: Setting research priorities

to eight language families, plus four sign languages and one unclassified language (Himarimã). Judeo-Egyptian Arabic is not always treated as distinct and also lacks a code. Likewise, the four Judeo-Iranian languages are often lumped together under Dzhidi (a.k.a. Judeo-Persian, ISO 639-3 [jpr]) and therefore lack standard codes. Although we do not have specific vitality information on these four Judeo-Median languages, they have been described as “on the verge of extinction” (Borjian 2015). Additionally, there are four pidgins that meet these criteria: Barikanchi (ISO 639-3 bxo), Gibanawa (gib), Iha Based Pidgin (ihb), and Settla (sta). Since pidgins by definition have no native speakers, it is not surprising that they appear with a blank for number of native speakers, and therefore their vitality inherently differs from other languages. However, they have been included here because we currently have no other vitality metrics for these languages. While there are comparatively few languages in ELCat with no vitality information, there are many languages for which some key elements are missing. As noted in Lee and Van Way (Chapter 5) on the Language Endangerment Index (LEI), intergenerational transmission is given double the weight of other vitality indicators in the calculation of vitality. While intergenerational transmission is undeniably a crucial factor in determining overall vitality, information on it was only available for 851 of the 3,138 living languages in ELCat. We urge researchers in their publications to be explicit about how (or whether) the language is being transmitted to children to help address this deficit. Relatedly, the LEI also gives a certainty rating for its vitality estimates that reflects the vitality indicators for which there is information. For example, languages for which we have information on speaker numbers but not on domains of use, intergenerational transmission, or trends in speaker number will receive a certainty rating of 20%, while a language that has information available about all of these factors receives a certainty rating of 100%. Therefore, vitality ratings with low certainty ratings are questionable because they are based on only one vitality indicator, and may not accurately reflect the status of the language. Of the 3,138 languages that have speakers in the ELCat database at present that have vitality ratings and are neither dormant nor awakening, 2,049 have only a 20% certainty rating based on information in their primary source, calculated only from speaker numbers, domains of use, or speaker number trends. Only about 633 languages have certainty ratings of 80% or 100%. If we embrace the possibility that languages that have both low certainty ratings and no information on intergenerational transmission are potentially either more or less endangered than reported, then there is an additional set of languages that needs to be considered when determining which languages are sufficiently endangered to necessitate prompt documentation efforts. Those most likely to be at risk that were not included in the list of highly endangered languages in Tables 9.1 and 9.2 are languages classified as “(Moderately) Endangered” that lack intergenerational transmission data and have a certainty rating of 20%. There are 557 languages that meet these criteria, and are given in Table 9.4, ordered alphabetically by region. Although there is certainly the possibility that many of these languages will not be more endangered, or perhaps will be even less endangered when the data are gathered on intergenerational transmission and other vitality indicators, a lack of available information suggests a lack of documentation. With the information that ELCat currently possesses, the following languages may be more endangered than the vitality index indicates. If we consider all the severely and critically endangered languages, those where children are no longer learning the language, languages where we were unable to assign a vitality rating at all, and those that perhaps would be considered severely endangered if more vitality information were available, then the total number of languages that are potentially in higher, more urgent need of documentary efforts comes to 1,454 languages, or about 46% of the living 275

Bryn Hauk and Raina Heaton Table 9.4 Endangered languages with 20% certainty and no intergenerational transmission information Code

Language

hnh ahn aja muc soh aku myo anf bob kme btc bkf bnz bxq eot bvq bqu zaz-bot

||Ani Àhàn Aja Ajumbu Aka Akum Anfillo Animere Aweer Bakole Bati (Cameroon) Beeke Beezen Bele Beti Birri Boguru Boto Buli Burunge Buu (Cameroon) Chaari Dir Doko-Uyanga Eman Fam Geme Gyem Hasha Hijuk Isubu Iyive Ju Kapya Kari Kelo Kobiana Koke Kolbila Kujarge Kung Kwegu Laal Lafofa Langas Mahongwe Mangas Mangayat Mbara (Chad) Mbre

Code

AFRICA

bds 8uu

uya emn fam geq gye ybj hij szv uiv juu klo kbj xel kcj kou klc vkj kfl xwg gdm laf mhb zns myj mpk mka

Language PACIFIC (CONT’D)

buv vrt bry bhf bnr msq cga cjn dnd dgg wwo eit ele elu mmw etn faf nrg agl gbf siu gnu grx gka tmd hvk mrk igg imn ipo itr tbd ywa kms kyy xla igm wat xkx wnp frt kco kis kxt xoi koa xop

Bun Bura Burmbar Burui Busa Butmas-Tur Caac Changriwa Chenapian Daonda Doga Dorig Eitiep Elepi Elu Emae Eton Fagani Farsaf Fembe Gaikundi Galu Gnau Guriaso Guya Haruai Haveke Hmwaveke Igana Imonda Ipiko Iteri Kaki Ae Kalou Kamasau Kambaira Kamula Kanggape Kaninuwa Karore Kayik Kene Kiai Kinalakna Kis Koiwat Kominimung Komnzo Konomala Kopar

276

Code

Language

SOUTH AMERICA (CONT’D) cbb cot cbc cav ciy, cuo cbg cin kwa des eme unk ese gvo gvp gvj gae hix txi jaa jqr yau jur urb cbv kay ktn arr knt kre xre xri mbn wmd cuj mpq mpd mmh mbj nab nhd mbr sru pln gob spn sik sri mdz suy tpj

Cabiyarí Caquinte Carapana Cavineña Chaima Chimila Cinta Larga Dâw Desano Emerillon Enawené-Nawé Ese'jja Gavião Gavião de Parã Guajá Guarequena Hixkaryana Ikpéng Jamamadí Jaqaru Jotí Juruna Kaapor Kakua Kamayurá Karitiana Karo (Brazil) Katukina do Acre Kreen-Akarore Kreye Krikatí Macaguán Mamaindé Mashco Piro Matís Maxineri Mehináku Nadëb Nambikwara Ñandeva Nukak Paiter Palenquero Playero Sanapaná Shikuyana Siriano Suruí do Pará Suyá Tapiete

Triage: Setting research priorities Table 9.4 continued Code

Language

Code

Language

Code

Language

mlj gbn zmo boe nuh nsg njl baf lgn otr gua sxe tax tsy ttf ukq vum zaz-zak zah ziz

Miltu Mo'da Molo Mundabli-Mufu Ndunda Ngasa Njalgulgule Nubaca Opuuo Otoro Shiki Sighu Tamki Tebul Sign Language Tuotomb Ukwa Vumbu Zakshi Zangwal Zizilivakan

krf lbb lkn lkn lmu lnm lrv tql lpa nms ler lll lht tei urr grg mll mhf jet mcz meg afd lmb kql mmm

Koro (Vanuatu) Label Lakon Lakon Lamen Langam Larëvat Lehali Lelepa Lendamboi Lenkau Lilau Lo-Toga Lou Löyöp Madi Malua Bay Mamaa Manem Mawan Mea Meakambut Merei Miyak Mkir Moiso Mokerang Momare Mongol-Kaimba Monumbo Mores Moresada Mota Muniwara Mwatebu Nafi Naha'ai Nahavaq Nai Nake Nakwi Namat Nambi Namo Namonuito Nasvang Nauete Nawaru Neku Nemi Nen Ngala

taf tav pah tpr urz wau xok ycn yuq yui pto gvo-zor swx

Tapirapé Tatuyo Tenharín Tuparí Uruewawau Waurá Xokleng Yucuna Yuki Yurutí Zo'é Zoró Zuruahã

AUSTRALIA adt bvr dhg guf, gnn dji gue ibd gbb gvn thd 1j4 mpj ulk djj djd-nga nam-nan nam nbj nna pti rit wmt wig jay

Adnyamathanha Burarra Dhangu-Djangu Dhuwala Djinang Gurindji Iwaidja Kaytetye Kuku-Yalanji Kuuk Thaayorre Manjiljarra Martu Wangka Meriam Ndjebbana Ngaliwurru Ngan'gikurunggurr Ngan'gityemerri Ngarinyman Nyangumarta Pintiini Ritharrngu Walmajarri Wik-Ngathana Yan-nhangu EAST ASIA

jiy yzg llh phq

Buyuan Jinuo E'ma Buyang Lamu Phana'

mft msz mgt mxk mrp msx mtt mwb mwa srf mlx sns bio nbk nax nkm mty mxw nmt nxa nwr nek nem nqn nud

SOUTH ASIA aeq all aaf adi-ash brw bhe bro hii jee npa adi-pai kdq-ban adi-ram soi suz-sur tro tcx vis zkr

Aer Allar Aranadan Ashing Bellari Bhaya Brokkat Hinduri Jerung Nar Phu Pailibo Pani Koch Ramo Sonha Surel Tarao Naga Toda Vishavan Zakhring

SOUTHEAST ASIA abf bsa ulf auq apx agf asz auu bsu bfk brs bya btq bed bgy

Abai Sungai Abinomn Afra Anus Aputai Arguni As Auye Bahonsuai Ban Khor Sign Language Baras Batak Batek Bedoanas Benggoi continued

277

Bryn Hauk and Raina Heaton Table 9.4 continued Code

Language

Code

Language

Code

Language

1ho yrn

Tazhi Yerong

ssv gbe niw nby niz

Ngen Niksek Nimo Ningera Ningil Nisvai Nokuku Nukuoro Numbami Nume Odoodee Onjob Ôrôê Owiniga Pahi Papitalai Pasi Piamatsina Pije Pitcairn-Norfolk Piu Pukapuka Pwaamei RakahangaManihiki Rawo Rerep Romkun Saniyo-Hiyewe Saruga Satawalese Sengo Sepen Sera Seta Seti Silopi Sinia Sissano Sonsorol Suganga Tangoa Tanguat Tasmate Taulil Terebu Tirax Tumleo Ukuriguma Urapmin Urimo Uya

bfe bhc bqq bnv bti bzu 1iu dbe daz dei diy dmv mrf enc azt flh gnq kzd kzb tcq kly kax bpp kdy kns ktt knq kyi kiy opk xkq kiq kwh lgh lji lnh lcc llk lpe lcd mgf tnz mgk mvx mnq kja mso mmb mhz

Betaf Biga Biritai Bonerif Burate Burmeso Chatong Dabe Dao Demisa Diuwe Dumpas Elseng En Faire Atta Foau Gana Kadai Kaibobo Kaiy Kalao Kao Kaure Keder Kensiu Ketum Kintaq Kiput Kirikiri Kopkaka Koroni Kosadle Kowiai Laghuu Laiyolo Lanoh Legenyem Lelak Lepki Lola Maklew Maniq Mawes Meoswar Minriq Mlap Mombum Momina Mor (Mor Islands, Indonesia) Ngkâlmpw Kanum Nyeu

EUROPE cpg fss

Cappadocian Greek Finnish-Swedish Sign Language fra-dge Guernésiais Kili ynk Naukan Yupik sma South Saami MEXICO, CENTRAL AMERICA, CARIBBEAN zaf Ayoquesco 1ku Bribri Sign Language 1kv Brunca Sign Language gut Guatuso otz Ixtenco Otomí lac Lacandon tlc Misantla Totonac ocu Ocuiltec pay Pech pia Pima Bajo ztt Tejalapan Zapotec otl Tilapa Otomí zpk Tlacolulita Zapotec jic Tol zpb Yautepec Zapotec msd Yucatec Maya Sign Language ztx Zaachila Zapotec NEAR EAST 5qc syy

Aftari Al-Sayyid Bedouin Sign Language 08a Aširat Northeastern Neo-Aramaic Farvi hss Ḥarsusi hoh Hobyot Judeo-Hamadani ktl Koroshi auz Uzbeki Spoken Arabic amw Western Neo-Aramaic nli-zem Zemiaki NORTH AMERICA akz ale

Alabama Aleut

nkk nkr sij tgs kkc onj bpk owi lgt pat psq ptr piz pih pix pkp pme rkh rwa pgk rmk sny sra stw spk spm sry stf sbi xsp nsw sso sov sug tgp tbs tmt tuh trb mme tmq ukg urm urx usu

278

kcd nyl

Triage: Setting research priorities Table 9.4 continued Code

Language

Code

Language

Code

Language

asb cic coc hur

Assiniboine Chickasaw Cocopa Halq'eméylem (Halkomelem) Jicarilla Apache Kwak̓wala MaliseetPassamaquoddy Oneida Secwepemctsin (Shuswap) Taos Winnebago

vlp mkt vra

Valpei Vamale Vera'a Vovo Vunapu Wab Wailapa Walio Wamas Warapu Watakataui Weliki Wogamusin Wusi Wutung Yakaikeke Yalë Yambes Yarawata Yau Yelogu Yerakai Yessan-Mayo Yis

afz ors psn ppm psg

Obokuitai Orang Seletar Panasuan Papuma Penang Sign Language Phong Powle-Ma Punan Merah Punan Merap Rasawa Ratahan Romam Sach Selangor Sign Language Serili Southern Alta Suoy Swoeng Taloki Talondo' Tanahmerah Tause Taworta Tay Khang Tefaro Temoq Thu Lao Tomadino Towei Ujir Ukit Uruangnirin Wares Waritai Warkay-Bipim Waru Yafi Yamna Yarsun

apj kwk pqm one shs twf win

PACIFIC aif ymo ail aki miw afk, afp aad amt aod

Agi Aiku Aimele Aiome Akoye Alfendio

Amal Amto Andarum Ande Anejom̃ anz Anem anj Anor aud Anuta aur Aruek Atin aul Aulua smf Auwe Avok kmn Awtuw aww Awun ahb Axamb ayq Ayi bbf Baibai bki Baki bit Berinomo bnw Bisis bir, bic Bisorio boq Bogaya bop Bonkiman mvt Botovro aof Bragat brq Breri bjl Bulu

vnp wab wlr wla wmc wra wtk klh wog wsi wut ykk nce ymb yrw yyu ylg yra yss yis

SOUTH AMERICA aca ano apy quxapu aro awt asu asn awe bkq jaa-kit bao bor

Achagua Andoque Apalaí Apurí Quechua Araona Araweté Asuriní do Tocantins Asuriní of Xingú Awetí Bakairí Banawá Bará Bororo

279

1hq msl puf puc rac rth rmx scb kgi sve agy syo 1eu tlk tln tcm tad tbp tnu tfo tmo tyl tdi ttn udj umi urn wai wbe bgv wru wfg ymn yrs

Bryn Hauk and Raina Heaton

Figure 9.1 Geographic distribution of target languages

languages in ELCat. The geographic distribution3 of these languages, henceforth referred to as “target” languages, is given in Figure 9.1. In Figure 9.1, white triangles = “Critically Endangered,” grey markers = “Severely Endangered,” black Xs = “Endangered/Threatened with 3 or 4 IGT ratings,” white markers = “No information available,” and black triangles = “Endangered, 20% certainty.”

Narrowing the field There are of course many factors that need to be taken into account when making recommendations for endangered languages research in addition to the vitality of a given language (see, for instance, Chelliah & de Reuse 2011). This section narrows down the list of 1,454 languages that merit more urgent attention, taking into account practical criteria as well as more linguistic desiderata. While the scope of such an undertaking is potentially almost impossibly broad, we focus here only on the criteria which can practically be investigated based on the data in ELCat. As such, “Lack of accessibility” attempts to deal with practical concerns relating to location; “Projected vitality” looks at projected vitality; “Genetic classification interest” identifies particularly endangered language families and isolates; and “Narrowing by existence of documentation” discusses under-documented languages.

Lack of accessibility One parameter to look at with respect to prioritizing language documentation involves practical concerns. While there are numerous valid concerns and limitations that have been mentioned in other publications (e.g. how difficult is it to travel to the location of the language’s speakers, is the target linguistic population receptive to documentation efforts, what diseases or illnesses is one likely to encounter), these questions are either not answerable based solely on the data in ELCat, or primarily depend on the circumstances of the individual researcher. Similarly, there are restrictions involving access, which affect whether it is possible to document particular 280

Triage: Setting research priorities

languages. For example, there are a number of uncontacted tribes in the Amazonian region of South America, especially in Brazil, and also reportedly in West Papua, who are currently not accessible to researchers (see Barlow and Campbell, Chapter 3), which may or may not appear in the Catalogue, depending on how much we know about them. While some of these groups are truly uncontactable (e.g. isolados “isolated people” in Brazil, namely the Isolado de Tanaru), periodically new contacts are made and sometimes specific researchers are granted access to study the language of a particular recently contacted group (as was the case, for example, with Carolina Aragon’s work with Akuntsú; see Aragon 2014). Similarly, there are sometimes travel restrictions imposed by governments on areas where endangered languages are spoken. The most notable of these are: areas overseen by FUNAI in Brazil; all of Venezuela, which does not allow foreign researchers access to indigenous populations; Tibet and the surrounding areas in China; some areas of India; and the Solomon Islands, which oft times grants no visas for foreign researchers. It depends to some extent on who the researcher is (and on timing, as in the case of Tibet and the Solomon Islands) whether access will be granted. Because these types of restrictions are difficult to delineate and may depend on permissions, they are acknowledged here but not discussed further. At present, there are a number of indigenous languages that are spoken in areas of armed conflict and are therefore inaccessible, or at the very least, researchers would potentially be putting themselves in undue danger by trying to do fieldwork in these areas. This will of course change as the global geopolitical situation and/or local circumstances change; the list in Table 9.5 reflects the state of affairs at the time of writing. The data on the safety of various locations was taken from the maps and information provided by the UK Foreign & Commonwealth Office (FCO) travel advice website (www.gov.uk/foreign-travel-advice), which provides relatively fine-grained travel warnings and advisories, with maps delineating dangerous areas below the country level. The FCO rates locations at three levels: “See our travel advice before traveling,” “Advise against all but essential travel,” and “Advise against all travel.” As collecting information on disappearing languages could very well be considered “essential travel,” only those languages spoken in areas where the FCO advises against all travel were singled out here. This of course does not imply that other areas are necessarily safe; those languages in areas delineated as “advise against all but essential travel” are generally still in conflict areas. Many areas are also unsafe due to high crime levels, drug trafficking activities, land disputes, or political unrest. These areas are included here only to the extent that the FCO deems these areas too unsafe for any travel. Languages that are inaccessible or less accessible continue to be a high priority in terms of documentation efforts, and may be even more endangered than suggested by the vitality rating by virtue of being spoken in dangerous areas. It also may be possible to document these languages under different circumstances, i.e. with speakers somewhere else safer than in the home community. However, we have identified these languages which may be currently impossible to document, and researchers are advised to take precautions if they are considering working on these languages. Based on the above criteria, 56 of the 1,454 languages identified as at high risk of extinction or as lacking sufficient information to make that evaluation are spoken in those areas that are currently experiencing the most intense conflict (“advise against all travel”) on the planet. In some cases this includes all the highly endangered languages in entire countries, e.g. Somalia, South Sudan, the Central African Republic, Libya, Niger, Somalia, Afghanistan, Yemen, and Syria, and also in the Gaza and West Bank regions. The majority of these languages are spoken in Africa (44 of 56), although there are six in the Near East; the rest are scattered among several regions. Table 9.5 gives the names of these 56 languages, as well as their ISO codes, and their vitality rating in ELCat. 281

Bryn Hauk and Raina Heaton Table 9.5 Languages likely undocumentable due to conflict Code

Country/region

Language

Vitality

aja soh 5h1 kme bdn bkf bxv bvq bqu bnl cet dmm elh fgr geq juh idr kbj xel klc vkj laf mfz myj mpk gbn zmo mru kqx swh-mwi gke nbd nnx njl lgn otr pmn mfl tsy 1nl ghl ziz zuy hoh syn sts sqt amw nli-zem dmk

South Sudan Sudan Cameroon Cameroon Cameroon Democratic Republic of the Congo Chad Central African Republic South Sudan Somalia Nigeria Cameroon Sudan Chad Central African Republic Nigeria South Sudan Democratic Republic of the Congo South Sudan Cameroon Chad Sudan South Sudan South Sudan Chad South Sudan Sudan Cameroon Cameroon; Chad Somalia Cameroon Democratic Republic of the Congo Cameroon South Sudan Ethiopia Sudan Cameroon Nigeria Mali Niger Sudan Cameroon Cameroon Yemen; Oman Iran Afghanistan Yemen Syria Afghanistan Kashmir

Aja Aka Baka Bakole Baldemu Beeke Berakou Birri Boguru Boon Centúúm Dama El Hugeirat Fongoro Geme Hõne Indri Kari Kelo Kolbila Kujarge Lafofa Mabaan Mangayat Mbara Mo’da Molo Mono Mser Mwini Ndai Ngbinda Ngong Njalgulgule Opuuo Otoro Pam Putai Tebul Sign Language Tetserret Uncunwee Zizilvakan Zumaya Hobyot Senaya Shumashti Soqotri Western Neo-Aramaic Zemiaki Domaaki

Moderate, Info Moderate, Info Severe Moderate, Info Critical Moderate, Info Critical Moderate, Info Moderate, Info Severe Critical Critical Severe Critical Moderate, Info Severe Severe Moderate, Info Moderate, Info Moderate, Info Moderate, Info Moderate, Info Severe Moderate, Info Moderate, Info Moderate, Info Moderate, Info Severe Severe Moderate, IGT Critical Critical Critical Moderate, Info Moderate, Info Moderate, Info Severe Severe Moderate, Info Severe Moderate, IGT Moderate, Info Critical Moderate, Info Moderate, IGT Severe Moderate, IGT Moderate, Info Moderate, Info Severe continued

Triage: Setting research priorities Table 9.5 continued Code

Country/region

Language

Vitality

gin huz jct jru pbg gob

Russia Russia Crimea Venezuela Venezuela Venezuela; Colombia

Hinukh Hunzib Krymchak Japréria Paraujano Playero

Moderate, IGT Moderate, IGT Critical Severe Critical Moderate, Info

Figure 9.2 Map of languages that are currently likely to be undocumentable

The languages in Table 9.5 constitute those target languages that we consider likely to be currently undocumentable in situ due to lack of access and of security due to armed conflict. Of the 56 languages above, 11 are “Critically Endangered,” 14 are “Severely Endangered,” 6 are “Endangered” with a 3–4 intergenerational transmission rating, and 25 are “Endangered” with 20% certainty. There are many other threatened and vulnerable languages in the same areas that are not listed above, but could potentially become more endangered as the result of being in a dangerous zone. Hopefully these areas will become more accessible before their languages become dormant. The global distribution of the languages in Table 9.5 is represented on the map in Figure 9.2. Once again white triangles indicate “Critically Endangered” languages, grey markers indicate “Severely Endangered” languages, black Xs indicate “Endangered” languages with 3 or 4 IGT ratings, and black triangles indicate those “Endangered” languages with a vitality confidence level of 20%.

Projected vitality The issue of exactly when a language can safely be called “dormant” or “extinct” is often a tricky one (see Belew and Simpson, Chapter 4). ELCat has handled this issue by erring on the side of caution, by including languages that have relatively recently become dormant or have 283

Bryn Hauk and Raina Heaton

been listed as having “no known speakers,” given the possibility that unknown speakers may yet be encountered in some cases, as has happened in the past in a few cases. The design of ELCat, like most databases, is susceptible to the possibility of reporting potentially inflated speaker numbers. As the majority of the data in ELCat comes from published sources, the speaker numbers in these sources can be from a year to several decades ago. Additionally, there may be some cases where, although there is no reliable information about whether a language still has any fluent speakers, there is a heritage community who identify with the language and may be working to revitalize it. As a result of these factors, some of the 1,454 target languages in the Catalogue in reality may no longer be spoken, or may no longer be spoken by the time this book is published and documentation work on one of these languages can be undertaken. Especially worthy of attention are the languages reported to have no native speakers or languages where there is a very strong chance that, because of the dated nature of the information available, they may already no longer have native speakers. We should look at those critically endangered languages reported to have ten speakers or fewer where that speaker number estimate is out of date. For our purposes here, “out of date” means the most recent primary information appears to be from ten or more years ago (i.e., from 2007 or before). The speaker number and year reported in Table 9.6 are the most recent figures and publication dates which appear to be based on primary fieldwork or first-hand knowledge of the languages. Post-2007 publications were included when they suggested that the language may already be dormant or have no remaining native speakers. While these boundaries for delimiting speaker numbers and years are somewhat arbitrary, logically if a language had only a very few very elderly speakers ten years ago, the likelihood that the number of first language speakers has decreased, possibly to zero, is high. Of course, if these languages were found to still have speakers, they would be of very high priority for documentation. Also, since the information is out of date, work should be undertaken to obtain updated vitality information for these languages, to determine whether speakers exist and the language is therefore of the highest priority for documentation or whether it truly has lost its last speaker and is therefore is of lowest priority. Of the 427 critically endangered languages currently in the ELCat database, 191 of them are possibly now dormant by the criteria outlined above. These languages are given in Table 9.6, including their ISO code, reported speaker number, and are sorted by the last year that primary speaker number data appears to have been collected. Note that we do not claim that the languages in Table 9.6 are necessarily dormant. Rather, we are simply pointing out that based on the information available in ELCat, there is a reasonable possibility that these languages may already be dormant (and at very least are likely very soon to be dormant). There is also the seemingly perpetual issue of what it means to be a speaker of a language. In this matter, the Catalogue of Endangered Languages is at the mercy of the speaker numbers reported in the sources of information on the languages, and different sources may decide who to count as a speaker based on somewhat different criteria. That said, to the extent possible, the numbers in Table 9.6 reflect something like first-language or fully competent speakers, as opposed to semi-speakers, rememberers, and language learners. This particular set of decisions involving representation does not mean that those languages with semi-speakers or second language learners are not also worthy of study, or that the language is not being preserved by those speakers. We simply mean that if one’s goal is to document languages before the last native speaker(s) pass away, it is possible that the languages above are no longer viable candidates for that. Again, if these languages do in fact still have native speakers, they would be of very high priority for documentation. 284

Triage: Setting research priorities Table 9.6 Critically endangered languages with 10 or fewer reported speakers pre-2007 Code

Language

Speaker number

Source date

xmu rsi dsn err xmq zmk zme zmc vmu nxn nrx wdu wnm wux yil aea giy gdc nmv nmp urf wil ynd gdh xut Lmc zmv nyt wkw wie wif wrh amz wrz ayd ugb xmp smu jng nae byb bwh bju luf zuy hti kzx wmi zma

Kamu Rennell Island Sign Language Dusner Erre Kuku-Mangk Mandandanyi Mangerr Margany Muluridyi Ngawun Ngomburr Wadjigu Wanggamala Wulna Yindjilandji Ariba Giyug Gugu-Badhun Ngamini Nimanburru Uradhi Wilawila Yandruwandha Gajirrabeng Kuthant Limilngan Mbariman-Gudhinma Nyawaygi Waka-Waka Wik-Epa Wik-Keyangan Wiradjuri Atampaya Warray Ayapathu Kuku-Ugbanh Kuku-Mu’inh Somray Yangman Naka’ela Bikya Bishuo Busuu Laua Zumaya Hoti Kamarian Agwamin Manda (Australia)

2 1 6 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2