CONTENTS:

FINDING SPACE FOR ORALITY A FIGHT AGAINST LINGUISTIC APARTHEID

Anvita Abbi
 
Abstract
Discrimination between various languages, especially those which are not written down and those which are, between those which are used as the medium of education and those which are not, between those which are dominant as the languages of administrators and those which are dominated, leads to a situation of linguistic apartheid. This is a recent phenomenon. It has risen in the new rule of imposing the so-called modern education of the elite on the ones who were educated but illiterate in written languages. Modern education that does not take cognizance of our multilingual nature or does not incorporate indigenous knowledge imbibed in heritage languages with their oral tradition in the school system has resulted in an immitigable divide in societies. Single language domination, however, primarily through education, judiciary, and political system, pushes the society to violence, intolerance, and subjugation. It encourages language shift and ultimately, language endangerment.

LINGUISTIC DIVERSITY AND SHARED LINGUISTIC FEATURES ACROSS LANGUAGE FAMILIES IN INDIA
      
Warsi M.J.

Abstract
The exceptional diversity of Indian linguistic landscape positions it as one of the most linguistically diverse countries worldwide. The 2011 census data shows that there are 121 main languages and 1599 additional languages in the country, indicating a complex distribution of languages. Despite belonging to different language families, languages of India have converged to share many features which make India a linguistic area. The Indian language region is characterized by unique features such as retroflex sounds, a Subject-Object-Verb (SOV) word order, morphological reduplication, echo formations, and diverse syntactic structures as explicator compound verbs and oblique marked subjects. This paper offers a glance into the vast linguistic landscape of India, showcasing its intricate and unified nature by highlighting common linguistic features found across different language families.
[Keywords: Linguistic Diversity, Language Family, Linguistic Features, Linguistic Area]
Keynote Address presented at the 51st AICDL, Thanjavur.

TELUGU MODIFIERS AND THE STRUCTURE OF THE NOMINAL PHRASE      
Aida Talić
 
Abstract
Regarding the structure of the nominal domain, many have suggested or adopted the Universal DP Hypothesis, where it is argued that every nominal domain universally projects a DP (Bowers 1991, Longobardi 1994). However, robust differences between languages with articles (e.g., the) and those without articles have led to a parametric approach to the nominal structure, where only languages with articles project a DP (Bošković 2008). Some later cross-linguistic works further show that some languages exhibit mixed behavior and are not easily classified within the two-way cut between NP and DP languages, arguing for a three-way typology, where languages with affixal articles represent a middle case between NP and DP languages (Talić 2015, 2017). In this paper, I investigate several properties of nominal modifiers in Telugu, comparing their behavior to that of adjectives and adverbs in English, Bosnian, and Dholuo, to shed light on the syntactic structure of the nominal domain in Telugu. I explore left-branch extraction out of nominal phrases, adverb extraction out of adjectival phrases and a type of pronominalization with the particle -di/-vi, which resembles English one-pronominalization and Bosnian adjective-stranding ellipsis. 
[Key words: Telugu modifiers, nominal phrases, pronominalization, ellipsis, crosslinguistic variation]

LOCALITY VIOLATION IN LOWERING:  THE CURIOUS CASE OF TYPE 5 CLITICS IN MUNDARI  
Gurujegan Murugesan
Abstract
Embick and Noyer (2001) model post-syntactic head movement as a type of syntactic head movement that follows locality by not skipping intervening heads is discussed in this article. The data cited here are for a  special pattern of clitics called Klavans' (1985) type 5 clitics, which may not strictly fit in with the idea of locality. In this article, it is demonstrated that only a post-syntactic head movement that violates locality can account for the characteristics of the type 5 clitic in Mundari, an Austro-Asiatic language.
[Keywords:  Locality, Head movement, Clitics, PF branch, Lowering, Vocabulary insertion, Local dislocation, Head movement constraint.
Abbreviations
ACC=accusative, DAT=dative, DL=dual, ERG= ergative, INGR = ingressive, IND=indicative, OM=object marker, PL=plural, PPRT=past participle SBJ=subjunctive, SM=subject marker, TR=transitive, 1=first person, 2=second person, 3=3rd person]

ECONOMETRICS TECHNIQUES FOR QUANTITATIVE ANALYSIS IN LINGUISTICS RESEARCH  
Nagaraj V. & Palanirajan G.
Abstract
This work explores the application of econometric techniques for quantitative analysis in linguistic research. For the empirical evidence, it used the primary data comprising 297 samples surveyed from Kasaragod District, Kerala. The data aims to study linguistic competence in Malayalam through 10 linguistic competence indicators among native and non-native students at different levels of school education. Based on the mother tongue of the students, they are grouped into i) Dravidian native (Malayalam), ii) Dravidian non-native (Tamil. Tulu, Kannada, and Kudiya), and iii) non-Dravidian non-native (Hindi, Marathi). To quantify their linguistic competence in Malayalam, the Multi-variate Linear Discriminant Function is applied for the analysis. The estimated values show that most of the linguistic competence indicators positively differentiate the Dravidian native (Malayalam) student from other groups. This study confirms the potential of econometric techniques to quantify linguistic phenomena through appropriate techniques.
[Keywords: Econometrics, Linguistic Competence, Natives and Non-natives, Discriminant Function]

EFFECTIVE COMBINATIONS OF ACOUSTIC PARAMETERS FOR SPEAKER IDENTIFICATION USING MACHINE LEARNING METHODS      
shagi G.U., Aji S. & Prema S.
 
Abstract
The speaker identification system uses feature extraction techniques to extract features from speech signals. The spectral feature consists of MFCC, LPC, or hybrid forms of these features. Other speech features like pitch and formant frequencies can concatenated with spectral features for the state-of-art problem. The main contribution of this study is the creation of feature sets from acoustic parameters such as pitch, first four formants, and intensity. The paper also explores using more than seven hundred features derived from the base feature sets to identify the speakers effectively. Commonly used scaling methods are used to normalize the feature set. The evaluation techniques include machine learning classifiers such as Support Vector Machines (SVM), Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Logistic Regression (LR). The model evaluation uses the metric parameter Accuracy, and the study employs speech corpora SLR 63 and SDUoK of the Malayalam language.
[Keywords: Speaker identification, Acoustic parameters, Statistical measures, Machine learning]

PATTERN OF BICONSONANTAL CLUSTERS IN OLD TAMIL TEXTS     
Ramprashanth Venkatakrishnan, Kumarasamy R. & BalaSundaraRaman Lakshmanan 
 
Abstract
In this paper, the frequency of biconsonantal clusters in the Sangam texts is computed using the programming language Python and its library NLTK, specifically made for natural language processing. The Sangam corpora taken for study are the oldest extant literature of Old Tamil. It consists of text corpora like the Eṭṭutokai (‘Eight Anthologies’) and Pattuppāṭṭu (‘Ten Idylls’), which are collections of Old Tamil poems of Akam and Puram genres. For the present study, both the texts of Eṭṭutokai and Pattuppāṭṭu are taken. Eṭṭutokai is an anthology of eight poetic texts, while Pattuppāṭṭu is a compilation of ten texts. Our main aim in this paper is to see what the pattern in the frequencies of biconsonantal clusters is, and to explain the pattern (why is a given cluster pattern more frequent compared to the others?) emerging from Old Tamil grammatical rules. Our paper is one of the first attempts at finding a pattern of frequencies of biconsonantal clusters across all extant Old Tamil texts and then explaining it.

CONTACT AND CONVERGENCE BETWEEN TELUGU-URDU: A CASE STUDY OF URDU SPOKEN IN PRAKASAM DISTRICT      
Devune Datturam & Anusree Sreenivasan
 
Abstract
The present paper aims to examine and illustrate some of the aspects of the Urdu language spoken in the Prakasam district in the state of Andhra Pradesh in India. Urdu is having long and stable contact with Telugu in this district which has resulted in the convergence of these languages. Extensive contact between Urdu and Telugu has led to large-scale lexical borrowings into Urdu spoken in this area. Apart from direct lexical borrowings from Telugu, Urdu which is spoken in this area also shows certain similarities with Telugu in case marking as well as number systems. This paper is a preliminary attempt to look into some of these aspects.
[Keywords: contact, convergence, lexical borrowings, Telugu, Urdu]

FOSTERING LINGUISTIC DIVERSITY IN EDUCATIONAL SETTINGS: STRATEGIES FOR SUPPORTING MINORITY LANGUAGE LEARNERS  
Sruthi S.
Abstract
This paper investigates the importance of fostering linguistic diversity within educational settings, with a focus on supporting minority language learners. The study explores strategies for promoting language retention and proficiency among linguistic minority students. It alsoemphasizes the need for collaborative efforts among educators, policymakers, and community stakeholders to ensure the successful implementation of strategies for supporting minority language learners.
[Keywords: Linguistic diversity, Educational settings, Minority languages, Language retention, Inclusive learning environments]