Endangered Languages

Partiendo de la premisa de que los cambios lingüísticos no solo responden a estructuras internas de las lenguas, sino también a factores extralingüísticos, como el contexto sociohistórico, migratorio, cultural, tecnológico y las experiencias individuales de los hablantes, este capítulo compara el español de monolingües hispanohablantes que residen en Ecuador con el español hablado en comunidades diaspóricas de ecuatorianos que que viven en Nueva York. En los dos casos, los datos provienen de lengua natural obtenidos en contextos formales e informales, medios de comunicación y redes sociales. El capítulo concluye señalando que, aunque las variedades son relativamente estables, surgen innovaciones constantes que responden a las necesidades comunicativas de sus hablantes, principalmente monolingües.

Palabras clave: Ecuador, Nueva York, castellano andino ecuatoriano, innovaciones lingüísticas, identidades transfronterizas

Gelecek zamanda konuşma zamanından sonra ortaya çıkacak ya da çıkma ihtimali bulunan olaylar anlatılır. Gelecek zaman, zaman kategorisinde farklı semantik alt ayrımlarla ele alınır. Bu sebeple gelecekte ortaya çıkacak veya çıkma ihtimali bulunan olaylar, Türk dili alanlarında semantik olarak ayırt edilebilmekte ve farklı biçim bilgisi unsurlarıyla işaretlenebilmektedir. Söz konusu durum, Türk dili incelemelerinin terminolojisine de yansır ve gelecek zaman, "gelecek, kesin gelecek, kategorik belirli gelecek vb." ile "geniş, belirsiz, kesin olmayan, şüpheli, tahminî, bitmeyen şimdiki zaman vb." semantik alt alanlara ayrılır. Türk dili alanlarında -Ar ~ -(y)(X)r eki, genellikle belirsiz gelecek zaman ve şimdiki zaman için kullanılagelir. -Ar ~ -(y)(X)r ekinin belirsiz gelecek zaman işlevi, Türk dilinin bütün tarihî ve çağdaş alanlarında görülür. -Ar ~ -(y)(X)r eki, şimdiki zaman işleviyle ise Eski Türkçede ve bugün bazı lehçelerde/ağızlarda yer alır. Bu ek, tarihî Türk dili alanlarında belirsiz gelecek zaman ve şimdiki zaman yanında, yer yer, kesin gelecek zaman için de kullanılır. Türk dili alanlarında biçim bilgisinin yenilendiği 13. yüzyıldan sonra kesin gelecek zaman için başka ekler de gelişir. Horasan Türkçesinin Bocnurd ağzında Oğuz ya da Çağatay grubu lehçelerinde gelişen kesin gelecek zaman ekleri yer almaz. Bocnurd ağzında daha eski bir kesin gelecek zaman eki olarak -(y)Ar bulunur. -(y)Ar eki, Bocnurd ağzında aynı zamanda belirsiz gelecek zaman için de kullanılır.

【本論文の中心テーゼ】
【本論文の中心テーゼ（一行で言えば）】
人間の知性が普遍的に生み出す「螺旋構造」は、計算機の規律（DAG）と原理的に衝突する。
本論文は、正岡子規の「柿くへば鐘が鳴るなり法隆寺」を人類の知性が生む螺旋の典型として据え、
その意味を一滴も損なわないままマルコフ決定過程（MDP）により 3ms 以内で DAG へ動的変換する
「非破壊的写像エンジン」が、第 4 巻の核心である。

本論文は、スキルズ・フィルター・フレームワーク（第 1〜3 巻）の上に構築される第 4 巻であり、「動的自律修復エンジン（MDP-based Dynamic Recompiler）」を提案・実証する。

第 3 巻が確立したプレレンダリング・静的ガード（RULE-07〜12）は、入力コードが描画エンジンに到達する前に DAG 違反を検出・修正する「事前防止（シフトレフト）」の仕組みであった。しかし、静的ガードには根本的な限界がある。人間の高次元な思考や LLM が生成する洗練されたコンテキストには、原理的に巡回（サイクル）を含む。

正岡子規の「柿くへば鐘が鳴るなり法隆寺」は、「現在の身体感覚 → 空間の響き → 千年の記憶 → 再び現在へ」という螺旋構造を持つ有向巡回グラフ（Cyclic Directed Graph）そのものである。この螺旋を「消去」することなく、計算機が 100% 正確にレンダリングできる DAG 形式へと安全に「写像」する——これが Phase 4 の固有の問いである。

本論文が提案する MDP ベースの動的リコンパイラは、巡回の意味（コンテキスト）を「アノテーション」として構造内に保全したまま、最大 50 ステップ・平均 3ms 以内で完全修復を達成する。N=30 の巡回コンテキスト構造（「柿くへば」を含む）を対象とした実験において、MDP モデルは 100% の成功率・RSR ≥ 97.8% を記録した。

【平易な説明】
「3ms」とは何か——それは速度の自慢ではない。我々の写像方程式が、千年の時を駆け巡る螺旋の意味を寸分も破壊せずに DAG 空間へと正しく変換できることの、動かぬ数学的証明である。

キーワード：マルコフ決定過程（MDP）、動的リコンパイラ、有向巡回グラフ（DCG）、DAG 変換、非破壊的写像、C4 モデル、螺旋構造、俳句工学、アノテーション保全、論文道場

Within most subfields of linguistics, the term “speaker” is often used in a shorthand, nonspecific way. In referring simply to“speakers”of endangered languages, the nuances of proficiency, language use, self-identification, and local language ideologies are collapsed into a binary: speaker vs. non-speaker. Despite the central role of local language ideologies in shaping patterns of language shift and maintenance, insiders’ perceptions of speaker status are not often investigated as part of language documentation projects. This paper approaches the issue of speaker status in Iyasa, a threatened Coastal Bantu language of Cameroon and Equatorial Guinea, through the firsthand accounts of self-identified Iyasa speakers. Using a discourse-analytic approach and the framework of identity and interaction (Bucholtz & Hall 2005), this paper examines the ways Iyasa speakers construct “speakerhood” in discourse, respond to researchers’ language ideologies, and position their own and others’ prof...

Kohgiluyeh va Boyer Ahmad Province, which constitutes less than one percent of Iran’s land area and a similar proportion of its population, is located in the mountainous south-west of the country. In the literature as well as in popular discourse in Iran, this region has been regarded as uniformly Lori-speaking. In this paper, we survey the distribution of languages in the province and investigate whether any languages other than Lori are spoken there as a mother tongue. Adopting the methodology outlined in the Atlas of the Languages of Iran, this article presents static and interactive maps showing each of the province’s languages. The results of our survey indicate that in addition to seven distinct varieties of Southern Lori, Ghashghāi Turkic, Khuzestāni Arabic, and Bakhtiari are spoken as ancestral languages in the southern districts of Kohgiluyeh va Boyer Ahmad. Further, we signal the growing prevalence of Persian, the dominant prestige language across Iran, as a mother tongue.

Les obstacles qui entravent la décision de parler patois et les motivations qui la favorisent sont multiples. Non seulement ils varient selon la compétence, l’origine et le caractère des locuteurs, mais ils dépendent également de la représentation mentale de la variation dialectale ainsi que de facteurs situationnels et sociaux. Nous nous appuierons dans la présente contribution sur deux projets de recherche qui ont particulièrement mis en lumière l’importance de certains défis dans l’utilisation de la langue minoritaire. Il s’agit, d’une part, d’un sondage sur l’auto-efficacité communicative en patois qui a été effectué entre autres en Vallée d’Aoste, et, de l’autre, d’une enquête sur les choix de langue menée à Saint-Nicolas, qui inclut, outre l’échelle d’auto-efficacité, des questions ouvertes concernant diverses difficultés perçues par les locuteurs. Les résultats démontrent le type de soutien qui peut être utile pour augmenter la motivation à choisir le francoprovençal comme langue de communication chez les Valdôtains en général et plus particulièrement chez les Sen-Nicolaèn.

English phrasal verbs pose complex semantic interpretations and cause significant problems in their translation into the target language. The paper selects some frequent and highly polysemous English phrasal verbs listed with 10 or more senses in the English WordNet, collects sample English sentences from the British National Corpus and examines their Hindi translation equivalents in terms of the morpho-syntactic structure of Hindi verbs and the representation of semantic information. This study attempts to present the semantic equivalence in rendering the identical semantics in the verbal system of both languages under consideration. The study found that the most frequent and common way of realising the concepts represented by English phrasal verbs in Hindi is by incorporating the compound verb constructions in the Hindi language. Besides that, we observe multiple mapping patterns of English phrasal verbs in Hindi, including the simple Hindi verb (v), compound verb (v1-v2), and conjunct verb (noun/adjective/adverb-verb) constructions.

Being able to thoroughly assess massive multitask language understanding (MMLU) capabilities is essential for advancing the applicability of multilingual language models. However, preparing such benchmarks in high quality native language is often costly and therefore limits the representativeness of evaluation datasets. While recent efforts focused on building more inclusive MMLU benchmarks, these are conventionally built using machine translation from high-resource languages, which may introduce errors and fail to account for the linguistic and cultural intricacies of the target languages. In this paper, we address the lack of native language MMLU benchmark especially in the under-represented Turkic language family with distinct morphosyntactic and cultural characteristics. We propose two benchmarks for Turkic language MMLU: TUMLU is a comprehensive, multilingual, and natively developed language understanding benchmark specifically designed for Turkic languages. It consists of middle-and high-school level questions spanning 11 academic subjects in Azerbaijani, Crimean Tatar, Karakalpak, Kazakh, Tatar, Turkish, Uyghur, and Uzbek. We also present TUMLU-mini, a more concise, balanced, and manually verified subset of the dataset. Using this dataset, we systematically evaluate a diverse range of open and proprietary multilingual large language models (LLMs), including Claude, Gemini, GPT, and LLaMA, offering an in-depth analysis of their performance across different languages, subjects, and alphabets. To promote further research and development in multilingual language understanding, we release TUMLU-mini and all corresponding evaluation scripts 1 .

Since social media and technology permeate contemporary life, activism and environmental protection are hot topics in the digital era. Our news feeds are flooded with pictures of a damaged environment, reports about climate change and its negative repercussions are everywhere, and the necessity of protecting our natural environment, including its plants and animals, is a constant worry. Communities that have a better grasp of the effects of industrialization and rapid development are more sensitive to environmentalist views and are embracing cleaner living and ethical behaviors. As a result of this, in perspective over the last few decades, language preservation should be included under the preservation umbrella in order to reach a wider audience. The challenges surrounding language preservation advancements will be discussed in this paper.

Our nation is filled with variety of culture and traditions. There are above 570 distinctive tribal groups residing all over India. Tribal and folk traditions reveal and outline beliefs, essence, and desires of people that contribute them a unique national identity. The significance of protecting our cultural heritage and value is that it keeps us integrated as citizens of one nation. Photography as a medium of art, which not only captures the real life, moment, emotions and expressions but also, helps in recording and communicating it to the world. This paper is an initiative to review the role and impact of photography in preserving Folk and tribal culture, tradition and art in India.

Script for a keynote speech given at the 6th International Conference on Content and Language Integrated Learning (CLIL), Bukhara State University
Abstract
Content and Language Integrated Learning (CLIL) can be an effective tool in language education, including that for minority and endangered languages. However, there may be difficult challenges involved; this talk will examine these challenges and ways of overcoming them. An obvious one has to do with expertise in the minority or endangered language; it is, of course, more difficult to find qualified teachers of e.g. Crimean Tatar or an Australian Aboriginal language than those of French or German. It is even more difficult to find such people who can also teach a non-language subject such as physics. In some cases, specialized dictionaries might need to be created. In fact, new vocabulary might have to be created; this should be done in consultation with the relevant communities. Intercultural competence is also required of teachers, as some subjects, particularly social studies and history, might require delicate and sensitive treatment for some student audiences. The cooperation of governmental and educational authorities is also important; in some cases the necessary resources may not be available. Finally, there is the question of whether CLIL is even possible for severely endangered languages, i.e. what is the minimum number of younger speakers necessary to make a program viable? Again, input from the language communities is crucial in deciding such matters. In spite of all these potential problems, when possible, the implementation of CLIL programs should be pursued.

Pronoun is a type of word that functions as noun substitutes and they can have the same number, gender, and case markings as nouns. Pronouns are divided into several subsets: indefinite, reflexive, demonstrative, interrogative, and personal pronouns. In Dhundhari, there are subject and object pronouns that are categorized for three persons and two numerals. Personal pronouns are not gender sensitive in first and second person but in third person. Different inflectional markers can be suffixed to the personal pronouns. Gender is marked in the vowel at the word final position. Reflexivity is marked by the agentive pronoun kʰʊɖ 'self' and possessive reflexive pronoun apɳo 'self'. The demonstrative pronoun makes a distinction for spatial distance on two levels. Another way to differentiate them is with a two-tier number system. Dhundhari has root forms /ʤo/, /ʤi/ or /ʤa for relative pronouns. Different morphemes are suffixed to these root forms based on different uses in relative clauses, based on various factors including animacy. All the interrogative pronouns are person as well as object specific in Dhundhari. The language also has indefinite pronouns such as koi 'anyone', dusro 'someone else'etc.

The world over languages includes a variety of narratives and folklore. Embedded within these narratives are often many aspects of social, cultural, religious and political issues of Human Rights. Thus, as most of these folklores have been around for centuries, these issues too carry a sense of historical significance with them. Today, these narratives and folklore serve as important repositories to understand many perspectives of Human Rights issues. This paper is an attempt to look into this
specific aspect of folklore and it does so by highlighting through examples of endangered indigenous languages of South Asia. This paper looks at an identity of a region, being synonymous to their local and indigenous languages and dialects and the paper proceeds to explain these through
examples of South Asia, especially spanning the region of the SAARC (South Asian Association of Regional Cooperation) countries. This is an important factor as these languages preserve narratives, which give an insight into many aspects concerning human rights as well as issues of migrations, resettlements and various aspects of regional history. Thus, the languages and their narratives are important representations in the modern global world. The paper also looks into few noteworthy questions, which are often argued as significant in modern times e.g. - ‘Are indigenous languages endangered?’ or ‘The context of indigenous languages in the present global world’. A closer study of these queries provides answers to look at representation, survival and continuity of
narratives of indigenous languages across the region and within the subtexts of these narratives, lie questions of human rights and of lives and their everyday struggles, which are oft left in oblivion from the pages of mainstream books. Thus, far from the confines of rigid urban discussions, these narratives can truly reflect issues of human rights and its sublime approach can help to connect across cultures by cutting through geographical boundaries.
The paper looks at the issue through few endangered languages across the region of South Asia with few examples of narratives from each. This approach of understanding issues of Human Rights through narratives and folklore is an area which is least explored till the present. The present paper feels that it is important to study the same as many unheard voices are often lost in oblivion and these are often seen in folklore and narratives. The paper follows an extensive work of the author across the region of South Asia across 2017 till the present.
---------------------------------------------------------------X-------------------------------------

It tum s out that a word-final long vowel tends to be maintained when it forms a morpheme on its own. When the final syllable does not constitute a morpheme, the long vowel /a:/ is shortened (e.g. sutore.,to-pa.,ma → sutopa) more frequently than the long vowel /u:/ (e.g. roNgu-bu' tsu → roNbu '). In addition, three-mora abbreviations with their final vowels shortened are more likely to be medially accented than other three mora abbreviations in the Kinki dialect. M oreover, a compound word is more frequently abbreviated to a single component when the component consists of one heavy syllable followed by one light syllable than when it contains four moras or one light syllable fol lowed by one heavy syllable (e.g. fis・ shu-pe pa・ → tisshu, ・ suno・ 一 bo.,do → bo・do).

The present article studies the visibility of a nearly extinct language Tjwao (Khoe-Kwadi) in the linguistic landscape (LL) of the rural district of Tsholotshoan area historically associated with the Tjwa in Zimbabwe. The analysis of the physical, functional, and multilingual characteristics of the signage reveals the following hierarchy of visibility: English is more conspicuous in the signage than Ndebele and Tjwao, which are, in turn, more prominent than Kalanga and other African languages. Shona is absent in the signage contrary to the LLs of the cities of Harare, Bulawayo, and Masvingo previously examined in literature. The considerable visibility of Tjwao is viewed as 'sustained' stemming from statal and local interventions related to linguistic/cultural activism, education, and health. In contrast, Tjwao is absent from the signage pertaining to daily community life. Overall, the near extinction of a language-such as Tjwao-does not necessarily imply erasure from its LL.

Assamese (ISO 639-3 code: asm) and Bodo (ISO 639-3 code: brx) employ conjunct verbs (CjV) of both [Adjective + Light Verb] and [Noun + Light Verb] patterns where the light verbs (LV) have limited meanings but grammatical functions. The CjVs in these languages share some common structural, grammatical, and semantic features. The light verbs are sometimes "restricted" in that they only collocate with specific nouns or adjectives whereas some nominals collocate with only their corresponding light verbs. Thus, the restriction of collocation in CjVs may be of two types, namely restriction of collocation of light verbs and restriction of collocation of nominals. For example-the light verbs [niɹua] 'REMOVE' (Assamese) and [pʰɯ ́ɾ] 'pw' (Bodo) co-occur only with the nominals [bɔn] 'weed (n)' (Assamese) and [kʰūɡà] 'mouth' (Bodo) respectively in the conjunct verbs [bɔn niɹua] 'weed' (Assamese) and [kʰūɡà pʰɯ ́ɾ] 'introduce' (Bodo); on the other hand, the nominals [buɹ] 'pw' (Assamese) and [dìɡūn] 'value of water' (Bodo) co-occur only with the light verbs [maɹ] 'HIT' (Assamese) and [zá] 'EAT' (Bodo) respectively in the conjunct verbs [buɹ maɹ] 'immerse; dip' (Assamese) and [dìɡūn zá] 'become trench foot' (Bodo). This study investigates the suitable collocations of nominals (nouns & adjectives) and light verbs in the conjunct verbs. CjVs in Assamese and Bodo may be found in two different categories based on their meanings, namely literal CjVs and idiomatic CjVs. And, based on the transitivity, CjVs are found to be intransitive and transitive in these languages. Structurally, a considerable set of members of Assamese and Bodo CjVs are of multi-word patterns such as [N + Adj + LV], [N + N + LV], [Adj + N + LV], [N + Adv + LV], and so on. Moreover, a few instances of CjVs are observed in the baby talk of these languages.

Linguistic typology studies and classifies languages according to their structural features rather than historical genealogy. The underlying assumption came to be called universal grammar, i.e. that all languages were somehow similar. Typology involves breaking down language into analytic units and then tracking their similarities across different languages. The question is what can be learnt from this, what question does it ask which provide useful insightful answers. The paper focuses on the language situation in Central Nigeria, where a large number of languages (200+) from different phyla and Niger-Congo subgroups form an intricate mosaic of mutual influences. The paper presents examples from phonology and morphology of nouns and verbs to illustrate this. Even where neighbouring languages can be shown to be related they can often exhibit rapid typological change. If so, then the outputs from typological studies are largely circular, i.e. they demonstrate typology is extremely fluid and can evolve rapidly. This in turn is hardly a striking insight. If so, then approaches from evolutionary linguistics might be more helpful in understanding complex situations such as in Central Nigeria.

Open access, peer-reviewed electronic and print journals, multimedia, and monographs on documentation and support of endangered languages, including theory and practice of language documentation, language description, sociolinguistics, language policy, and language revitalisation.

The preservation of linguistic diversity is a critical challenge in the digital age, as thousands of languages face the risk of extinction. Modern Natural Language Processing (NLP) tools often focus on high-resource languages, leaving a significant "digital divide" for endangered dialects. This paper presents KaybolanDiller (Endangered Languages), an opensource, AI-powered translation framework specifically designed for rare and endangered languages, with a primary focus on Turkic dialects (e.g., Meskhetian, Karamanl, Gagauz) and global isolates such as Ainu. We describe a hybrid system architecture that combines Neural Machine Translation (NMT) via Transformer-based models with a phrase-level lexicon fallback for extremely low-resource pairs. The framework features a decoupled FastAPI backend and a SvelteKit-based web interface, providing an accessible platform for both linguists and native speakers. Our results demonstrate the feasibility of cross-lingual transfer learning to support dialects with minimal digital corpora.

Multidisciplinary evidence indicates that the Austroasiatic (AA) language family is the earliest known language in Mainland Southeast Asia (MSEA), dating back to the Neolithic. Yet, the genomic formation and structure of MSEA AA groups remain understudied. Here, we generate genome-wide data for seven AA-speaking and two Sino-Tibetan-speaking populations from Thailand/Laos/ Myanmar, which together with published data comprises the largest AA genome-wide dataset to date. We find substantial genetic heterogeneity across both geographic regions and linguistic branches, with the greatest observed in Northern Mon-Khmer highland groups. Analyses with ancient DNA data indicate that northern AA groups exhibit higher East Asian ancestry linking to Iron Age northern Thailand/Cambodia, whereas southern AA groups display additional South Asian ancestry and affinities with Neolithic Laos/Vietnam. Notably, the South Asian-related ancestry is detectable in Neolithic MSEA. Overall, both isolation and contact have together shaped the pronounced genetic heterogeneity observed across linguistic branches of MSEA AA groups.

The article examines the efforts to revitalize Juhuri, the spoken language of Mountain Jews, in Moscow. The data for the analysis were collected in 2024-25. The main actor of the Juhuri revitalization is STMEGI Charitable Foundation established by emigrants from the village of Krasnaya Sloboda (Azerbaijan). Here, the social conditions have hindered language shift, allowing Juhuri to remain the primary means of communication until recently. STMEGI foundation publishes textbooks, organizes language courses, hosts literary events and a Juhuri festival, supports writers, funds a master's program in studying the Juhuri language and culture of Mountain Jews. Among Mountain Jews both in Moscow and abroad, particularly in Israel, there are many enthusiasts who initiate their own projects for language revival: they translate and dub cartoons and film episodes, compose and perform songs, translate books, collect folklore, host chats and communities focused on Juhuri. The foundation’s language policy mirrors state-level policies: it operates institutionally, delegates tasks, demands accountability, and strives for efficiency, whereas grassroots language initiatives reflect individual preferences and interests without aiming for effectiveness. Both STMEGI foundation's work and independent activists contribute to expanding access to Juhuri but have almost no impact on its actual use in communication. However, increased communicative load of Juhuri may be an unintended consequence of other foundation projects such as creating Mountain Jewish communal centers and spaces where Juhuri can serve as the primary language of interaction.

Berlin. 2) to appear 'Imperfective stems, moods and tenses in Budugh', to appear in: Linguistic Typology at the Crossroads, Special Issue, Edited by Murad Suleymanov and Steven Kaye. 3) to appear 'Anticausative alternation in Budugh simple verbs' to appear in Diachronic and typological perspectives on anticausativization' Edited by Guglielmo Inglese et al. Typological Studies in Language, Benjamins. 4) to appear 'Heterogeneous flagging in Budugh' to appear in Gilles Authier, & Kirill Kozhanov, The Diachrony of Flagging. Language Science Press 5) 2026 'Generalization of bipartite stems in Budugh' to appear in "Historical morphology and morphological theory", Edited by Francesco Gardani and Franz Rainer, special issue of Word Structure, volume 19, issues 1-2. 6) (avec A. Haciyev) 7) 2016. « Budugh « in: Word-Formation. An International Handbook of the Languages of Europe Edited by P. O. Müller, I. Ohnheiser, S. Olsen and F. Rainer. Mouton De Gruyter: Berlin. 8) 2013. « The origin of two genitive cases and inalienability split in Budugh (East Caucasian) «, Faits de Langues 41. 9) 2010. « Inessif du verbe, subordination et modalité « In: Essais de typologie et de linguistique générale. Mélanges offerts à Denis Creissels, éd. F. Floricic Presses de l'ENS: Lyon. 10) 2010. « Finite and non-finite: prosodic distinctions on Budugh verb stems « In: Clause-hierarchy and clause-linking: the syntax and pragmatics interface, ed. I. Bril. John Benjamins: Amsterdam. 11) 2010. « From adlocative case to debitive mood in Lezgic languages « In: Tense, Mood, Aspect and Finiteness in East-Caucasian, Authier, G. & Maisak T., eds. Diversitas Linguarum, Brockmeier: Bochum. 12) 2008. « Aperçu de l'histoire des phonèmes du Budugh » Mélanges Kodzasov.