Kashmiri Language

Presentation material for the World Sanskrit Conference, 2025, Kathmandu, Nepal. Special Panel on Vernacular Tantra. The paper presents constructions in Śaiva Tantric Sanskrit that have an active verb with a subject in the instrumental. It argues that these constructions are not to be seen as bad Sanskrit, but as demonstrating the influence of period vernacular Apabhraṃśa.

This paper conducts a comparative analysis of the genitive case system in three Eastern-Indo-Aryan languages: Goalparia, Assamese, and Bangla. The primary objective is to investigate the morpho-syntactic properties of genitive case markers, exploring potential differences or similarities among these languages that share close genetic affiliations and exhibit readily visible features. While Assamese adopts a split-ergative case system, Bangla and Goalparia employ a nominative-accusative case system. Our examination focuses on declarative sentences, utilizing the Generative Approach to unveil the inherent structure and distribution of genitive case markers in these languages. The paper presents a syntactic tree illustrating the relationship between the head, D, and the specifiers, DP spec, along with the complement, demonstrating the representation of the genitive case marker across the three languages.

Using the framework of Processability Theory (PT), this chapter examines the development of the Hindi case system in the interlanguage of Dutch-speaking foreign language learners. A number of studies have already adopted the PT framework to describe, predict and explain the development of case marking systems in different languages (e.g., German, Russian and Serbian). These studies demonstrated that PT was a suitable framework for predicting and explaining case development. Because German, Russian and Serbian are accusative languages, the question arises whether PT as a conceptual framework is also applicable to the foreign language acquisition of a language with ergative features such as Hindi. The present chapter therefore addresses case development in L2 Hindi, thereby testing the claim that PT possesses universal applicability. Spontaneous oral production data were collected from 11 foreign language learners of Hindi. In accordance with PT, the study results indicate that three stages of development occur; in particular, learners begin with no feature unification, transition through direct mapping, and eventually attain feature unification.

This is an interesting study exploring in a considerable detail the rise of the lowlevel configurationality in New Indo-Aryan from the non-configurationality in early Indo-Aryan. The data were extracted from ca. 60 primary sources (Vedic, Pali, Apabhraṃśa and Old Awadhi). The author concurs with the general assumption that Vedic Sanskrit lacks any phrasal structures. As explicated in Chapter Two ("Grammaticalization and configurationality"), in addition to the free word order (basic SOV, and OSV and VSO making the VP discontinuous) there are no obligatory function words (articles, adpositions) in Vedic Sanskrit which impose phrasal constraints on nominal expressions. At the end of a development that lasted 2000 years, Hindi-while still allowing for free constituent order-has developed low-level configurationality in terms of postpositional phrases, accompanied by a rise of obligatory function words. Thus unlike Sanskrit, Hindi does not allow for discontinuity within the NP(Adj N) or free permutation with respect to the arguments denoting the beneficiary and direct object as in example (40), Raješ= ne (choṭe bacc)e=ko kitāb bhejī 'Rajesh sent the book to the little child' (p. 100). The author shows that in Hindi the "obligatoriness constraint" of the postposition requiring a dependent has no equivalent in Vedic *bacce choṭe ko, *choṭe ko bacce, *choṭe … bacce ko). Interestingly, within the NP adnominal elements (adjectives) do not show distinctive forms for direct and oblique cases in the plural (examples 44 and 46): (choṭ-e kamr-e) =mê 'in the small room' but (choṭ-e kamr-õ) =mê 'in (the) small rooms' (p. 101). In diachronic terms the postpositional phrases are traced back to "asymmetrical groups" consisting of a nominal (or verbal) head and a dependent; thus the above examples would start in Old Indo-Aryan with obligatoriness of AN and GN as (alp-asya veśman-as) madhye (small-GEN room-GEN) middle-LOC 'in the small room' and (alp-ānām veśman-ām) madhy-e (small-GEN.PL room-GEN.PL) middle-LOC 'in the small rooms'. The diverse origins of the Hindi simple postpositions are surveyed in Chapter Three ("The diverse origins of the Hindi simple postpositions") with

I like reading prefaces of dissertations. They always seem spontaneous, even lyrical, and often totally different from the dry and serious chapters following them. No wonder I was looking forward to writing one myself. However, now the time has come, at the end of almost four long, demanding, fascinating, depressing and euphoric years, I seem to have lost any lyrical inspiration. Hence I fear that this preface will be dry and concise, though no less sincere. I would first like to thank my supervisor, Klaas Willems. The best supervisor one could wish for, he has incessantly been providing me with feedback and advice, while simultaneously questioning and challenging my opinions. In the process of writing this dissertation, he has been enormously supportive, and I am extremely grateful for that. The support of the members of my doctoral advisory committee, Peter Hook and Eva De Clercq, was crucial for this dissertation in so many ways. I thank Peter for endlessly answering my mails, and Eva for the long and thorough conversations in Japan and here. I want to thank Eva and the other members of my defense committee, Annie Montaut, Omkar Koul and Seppo Kittilä, for agreeing to read the manuscript and to share their expertise. My acknowledgement is also due to Jyotiprakash Tamuli, Hans Hock, Liudmilla Khokhlova, Miriam Butt, Andrea Drocco, John Peterson and so many other people whom I bothered with questions, in real life and through email, and whose comments have benefited me greatly. My friends and native speakers, Rudhira, Jayeeta, Alpha, Sarit, Pratik, Rohit and Jeanne have been of tremendous help. They were there when I needed them, for language consultation as well as for simple mental support. I am in debt to Meenaxi Barkataki-Ruscheweyh, who saved me from the traps of Asamiya and let me use her careful, skilled translations. And to Omkar N. Koul, who introduced me to one of the most beautiful (but complex!) languages I‖ve ever learnt, Kashmiri. I also thank my vi Nepali informants, Vishma Raj Thapaiya, Chand Mati Pokharel, Saraswati Chaudhary and Satish. I am working in a man‖s place: Kristof, Maarten en Ludovic, I owe you my thanks for the discussions, the lunches, the occasional four-o‖-clock drinks and for simply being perfect colleagues. At Blandijn, thanks are due to my partner in Indology, Stefanie, and Gitte, thanks for the ICT support. I wish to thank my parents for their undemanding encouragement. For the necessary distraction in the form of weekend dinners, lunches, (birthday) parties, soccer tragedies and cycle races, thanks to my family and friends: Pol, Koen, my sister/favorite housemate Anneke, Lieselotte, Maya, Anne-Sophie and Annelies. And Felix, I‖m sorry for all the times I was at work while I should have been playing hide-and-seek with you. And most importantly, thank you, Bram, for so many things, but particularly for taking me to the most pristine beaches on earth, occasionally.

This is an interesting study exploring in a considerable detail the rise of the lowlevel configurationality in New Indo-Aryan from the non-configurationality in early Indo-Aryan. The data were extracted from ca. 60 primary sources (Vedic, Pali, Apabhraṃśa and Old Awadhi). The author concurs with the general assumption that Vedic Sanskrit lacks any phrasal structures. As explicated in Chapter Two ("Grammaticalization and configurationality"), in addition to the free word order (basic SOV, and OSV and VSO making the VP discontinuous) there are no obligatory function words (articles, adpositions) in Vedic Sanskrit which impose phrasal constraints on nominal expressions. At the end of a development that lasted 2000 years, Hindi-while still allowing for free constituent order-has developed low-level configurationality in terms of postpositional phrases, accompanied by a rise of obligatory function words. Thus unlike Sanskrit, Hindi does not allow for discontinuity within the NP(Adj N) or free permutation with respect to the arguments denoting the beneficiary and direct object as in example (40), Raješ= ne (choṭe bacc)e=ko kitāb bhejī 'Rajesh sent the book to the little child' (p. 100). The author shows that in Hindi the "obligatoriness constraint" of the postposition requiring a dependent has no equivalent in Vedic *bacce choṭe ko, *choṭe ko bacce, *choṭe … bacce ko). Interestingly, within the NP adnominal elements (adjectives) do not show distinctive forms for direct and oblique cases in the plural (examples 44 and 46): (choṭ-e kamr-e) =mê 'in the small room' but (choṭ-e kamr-õ) =mê 'in (the) small rooms' (p. 101). In diachronic terms the postpositional phrases are traced back to "asymmetrical groups" consisting of a nominal (or verbal) head and a dependent; thus the above examples would start in Old Indo-Aryan with obligatoriness of AN and GN as (alp-asya veśman-as) madhye (small-GEN room-GEN) middle-LOC 'in the small room' and (alp-ānām veśman-ām) madhy-e (small-GEN.PL room-GEN.PL) middle-LOC 'in the small rooms'. The diverse origins of the Hindi simple postpositions are surveyed in Chapter Three ("The diverse origins of the Hindi simple postpositions") with

Kashmiri language 'koshur' is the language of people living in the valley of Kashmir and its adjoining valleys in the south and east. According to census the magnitude of Kashmiri speakers was 55276698 during 2001. There is a general consensus among historical linguistics that Kashmiri belongs to the Dardic branch and is closely akin to Shina. Shina is also a subgroup of Dard language. Kashmiri language is also said to be a prakrit of pure and original Sanskrit. There is a controversy among scholars and philologists over the origin of the language. The present paper attempts to address the raison d'etre, which made the language recede into oblivion. This endeavor aims to highlight the reasons that are tapering the scope and virginity of the language among its own speakers.

फ्रान्स्वा त्रूफो दिग्दर्शित 'फोर हंड्रेड ब्लोज' (१९५९), 'ज्यूल अँड जिम' (१९६१) यांसारख्या फ्रेंच न्यू वेव्हच्या जाणिवा घडवणाऱ्या चित्रपटांना यश मिळालं. समीक्षकांनीही ते नावाजले. पण १९६४मध्ये प्रदर्शित झालेल्या 'सॉफ्ट स्किन' या त्रूफोच्या चित्रपटानं त्याच्या न्यू वेव्हमधल्या मित्रांना आणि चाहत्यांना नाराज केलं. खरं तर त्यात एक प्रौढ, सखोल मांडणी आहे आणि एका प्रदीर्घ तसंच वजनदार कारकिर्दीची चुणूकही आहे.

The influence of the Indo-Aryan languages on the indigenous Indian languages, especially in terms of word-borrowing, has long been recognized. Less well recognized is the influence of the indigenous languages on Indo-Aryan. This paper explores the influence of proto-Dravidian (PD) on Indo-Aryan (specifically Vedic and Pāli), by examining shared similarities and innovations between the two language groups which the Indo-Aryan languages did not inherit from Indo-Iranian, Old Indic's closest relative. The paper argues that they were adopted from the proto-Dravidian language.

Kashmiri is a resource poor language with very less computational and language resources available for its text processing. As the main contribution of this paper, we present an initial version of the Kashmiri Dependency Treebank. The treebank consists of 1,000 sentences (17,462 tokens), annotated with part-of-speech (POS), chunk and dependency information. The treebank has been manually annotated using the Pān. inian Computational Grammar (PCG) formalism (Begum et al., 2008; Bharati et al., 2009). This version of Kashmiri treebank is an extension of its earlier version of 500 sentences (Bhat, 2012), a pilot experiment aimed at defining the annotation guidelines on a small subset of Kashmiri corpora. In this paper, we have refined the guidelines with some significant changes and have carried out inter-annotator agreement studies to ascertain its quality. We also present a dependency parsing pipeline, consisting of a tokenizer, a stemmer, a POS tagger, a chunker and an inter-chunk dependency parser. It, therefore, constitutes the first freely available, open source dependency parser of Kashmiri, setting the initial baseline for Kashmiri dependency parsing.

Kashmiri is a resource poor language with very less computational and language resources available for its text processing. As the main contribution of this paper, we present an initial version of the Kashmiri Dependency Treebank. The treebank consists of 1,000 sentences (17,462 tokens), annotated with part-of-speech (POS), chunk and dependency information. The treebank has been manually annotated using the Pān. inian Computational Grammar (PCG) formalism (Begum et al., 2008; Bharati et al., 2009). This version of Kashmiri treebank is an extension of its earlier version of 500 sentences (Bhat, 2012), a pilot experiment aimed at defining the annotation guidelines on a small subset of Kashmiri corpora. In this paper, we have refined the guidelines with some significant changes and have carried out inter-annotator agreement studies to ascertain its quality. We also present a dependency parsing pipeline, consisting of a tokenizer, a stemmer, a POS tagger, a chunker and an inter-chunk dependency parser. It, therefore, constitutes the first freely available, open source dependency parser of Kashmiri, setting the initial baseline for Kashmiri dependency parsing.

Целью исследования было охарактеризовать шугнанские начальные взрывные с точки зрения категорий VOT, а также определить влияние открытости гласного, места образования согласного и ударности слога на количество придыхания.

The purpose of this research was to describe the VOT of the initial plosives in Shughni and then, identify if there was any influence of vowel height, consonant place of articulation and stress.

This paper describes how bootstrapping was used to extend the development of the Urdu Noisy Text dependency treebank. To overcome the bottleneck of manually annotating corpus for a new domain of user-generated text, MaltParser, an opensource, data-driven dependency parser, is used to bootstrap the treebank in semi-automatic manner for corpus annotation after being trained on 500 tweet Urdu Noisy Text Dependency Treebank. Total four bootstrapping iterations were performed. At the end of each iteration, 300 Urdu tweets were automatically tagged, and the performance of parser model was evaluated against the development set. 75 automatically tagged tweets were randomly selected out of pre-tagged 300 tweets for manual correction, which were then added in the training set for parser retraining. Finally, at the end of last iteration, parser performance was evaluated against test set. The final supervised bootstrapping model obtains a LA of 72.1%, UAS of 75.7% and LAS of 64.9%, which is a s...

The present paper proposes a non-parametric approach to ergativity. It is argued that ergativity does not depend on a parametric choice, but on morphological properties of certain constructions that are distributed throughout a wide and varied array of languages. The main hypotheses propounded are two: (i) that the notion of ergativity should include not only canonical ergative patterns, but also, as a part of the very same phenomenon, oblique subject constructions (i.e. dative experiencers and so-called quirky subjects), and (ii) that ergativity can be understood as the consequence of a defective accusative case assignment in transitive constructions. In this sense, we propose to consider ergativity in general as transitive unaccusativity.

Данная публикация представляет собой некоторые соображения о статье А. И. Когана «Genealogical classification of New Indo-Aryan languages and lexicostatistics» в № 14 «Вопросов языкового родства». Автор высказывает ряд предложений по исправлению списков Сводеша, представленных А. И. Коганом, на основании данных исторической фонетики и семантики индоарийских языков в целом, а также анализирует списки языков хинди, бенгали, ория и куллуи, сопоставляя их со своими полевыми данными и сведениями из словарей и корпусов. После внесения предложенных исправлений автор сопоставляет деревья, построенные программой Starling на основании первоначальных и исправленных списков. Изменения не очень значительны, однако повышают достоверность результатов, а кроме того, показывают, что дальнейшее исправление списков может сделать дерево значительно более точным.

Отражение древнеиндийского среднего рода в цыганском* В статье показано, что древнеиндийский средний род не исчез бесследно в цыганском языке, а сохранился в формах на á им.мн. существительных (синхронно мужского рода), тогда как старый мужской род дал формы им. мн. без окончания (прежде не имевшие объяснения). Ключевые слова: цыганский язык, древнеиндийский язык, лингвистическая реконструкция, историческая морфология. * Выражаю глубокую благодарность Анастасии Крыловой (Москва) за ценные поправки. 1 Кэлдэрарский диалект относится к влашской группе диалектов цыганского языка и сформировался на румыноязычной территории. Сейчас кэлдэрары живут во многих странах мира, в частности на террито

Kashmiri is a resource poor language with very less computational and language resources available for its text processing. As the main contribution of this paper, we present an initial version of the Kashmiri Dependency Treebank. The treebank consists of 1,000 sentences (17,462 tokens), annotated with part-of-speech (POS), chunk and dependency information. The treebank has been manually annotated using the Pān . inian Computational Grammar (PCG) formalism . This version of Kashmiri treebank is an extension of its earlier version of 500 sentences (Bhat, 2012), a pilot experiment aimed at defining the annotation guidelines on a small subset of Kashmiri corpora. In this paper, we have refined the guidelines with some significant changes and have carried out inter-annotator agreement studies to ascertain its quality. We also present a dependency parsing pipeline, consisting of a tokenizer, a stemmer, a POS tagger, a chunker and an inter-chunk dependency parser. It, therefore, constitutes the first freely available, open source dependency parser of Kashmiri, setting the initial baseline for Kashmiri dependency parsing.

Koshur, also known as Kashmiri, is a Dardic language spoken primarily in the Kashmir Valley of South Asia, the center of the international Kashmir conflict, in India-administered Kashmir. It has about 7 million speakers globally, and at least 14 dialects (Eberhard, Simons, & Fennig, 2019) though sociolinguistic research into the dialect distinctions is nearly nonexistent. This is due to a variety of reasons, many of them politically charged and fraught with diplomatic faux-pas. In this thesis, I will explore Kashmir’s and Koshur’s linguistic history and examples of linguistic persecution and conflict in modern South Asia to compare them to Koshur’s current standing. I will also discuss the lack of dialect-sensitive, objective research in Koshur’s documentation and stress the urgency with which this research is needed. Additionally, I will explore the treatment of the largest and starkest dialect split, which is between the co-dialects of Koshur spoken by the Kashmiri Hindu and Kashmiri Muslim communities. While this split encompasses many smaller sub-dialects and varieties, this split is the most obvious and simultaneously least acknowledged.

На основе сравнительного метода сопоставления хурритской лексемы aɣаri (названия ладана) с этимологией чеченского значения jagar («сжигать, гореть»), представляющего также семантику значения, связанного с доисламской обрядовой практикой k'ur toxar («окуривание дымом»), автор предлагает новую этимологическую реконструкцию для хурритской лексемы.

This article addresses the variable alignment properties of experiencer constructions in Indo-Aryan (IA) languages in the light of the available historical data from Vedic Sanskrit onwards. The first aimof the article is to shed light on the possible historical sources, emergence and expansion of constructions with non-canonically marked arguments in Old IA in general. The second aim is to gain a better understanding of the variation in case marking and agreement patterns that can be attested in New IA experiencer constructions, given that the interplay among morphological cases, semantic roles and additional semantic motivations poses many unsolved questions.

А.И.Коган. Некоторые проблемы кашмирской диалектологии. В статье исследуются генетические отношения литературного кашмири и четырех диалектов, распространенных за пределами Кашмирской долины. В прошлом эти диалекты нередко классифицировались как кашмирские, однако аргументы в пользу такой классификации зачастую были неубедительны. Автор статьи приходит к выводу, что 2 диалектасираджи и рамбанине являются не только кашмирскими, но и дардскими, а относятся к индоарийской группе. В то же время 2 других диалекта погули и каштаваридействительно обнаруживают большую генетическую близость к стандартному кашмири.

The present paper proposes a non-parametric approach to ergativity. It is argued that ergativity does not depend on a parametric choice, but on morphological properties of certain constructions that are distributed throughout a wide and varied array of languages. The main hypotheses propounded are two: (i) that the notion of ergativity should include not only canonical ergative patterns, but also, as a part of the very same phenomenon, oblique subject constructions (i.e. dative experiencers and so-called quirky subjects), and (ii) that ergativity can be understood as the consequence of a defective accusative case assignment in transitive constructions. In this sense, I propose to consider ergativity in general as transitive unaccusativity.

२८ नोव्हेंबर २०१० रोजी जगभरातल्या अमेरिकी दूतावासांच्या अडीच लाख केबल्समधली गोपनीय माहिती प्रसारमाध्यमांतर्फे जगासमोर उघड होऊ लागली. अनेकांना `विकीलीक्स'च्या अस्तित्वाचा तेव्हा शोध लागला, पण विकीलीक्सचा सर्वेसर्वा असणाऱ्या ज्यूलिअन असांज याने २००६पासून विकीलीक्समार्फत विविध प्रकारची गोपनीय माहिती उघड केलेली आहे. पाश्चिमात्य देशांनी, विशेषत: अमेरिकेने त्याची वेळोवेळी दखलही घेतलेली आहे. आताच्या परिस्थितीत पूर्वीपेक्षा एक फरक मात्र दिसतो आहे : असांजला गौप्यस्फोट करता येऊ नये यासाठी अनेकांनी आता कंबर कसलेली आहे. असांजला आणि त्याच्या विकीलीक्सला जगभरात मिळालेली प्रसिद्धी आणि त्यांना संपवण्याचे होणारे प्रयत्न, यांमुळे या सर्व प्रकाराविषयी विविध मते आणि विचार मांडण्याची अहमहमिका लागलेली आज सर्वत्र दिसते आहे. त्या गदारोळात न सापडता थोडे गंभीर मूल्यमापन करण्याचा प्रयत्न या लेखात केलेला आहे.

पूर्वप्रकाशन: 'आजचा सुधारक' मार्च २०११