PANEL 31: Linguistics of lesser-known languages in South Asia

Panel Organizers:

Dr. Anju Saxena - Uppsala University; Sweden
Dr. Ruth Laila Schmidt - University of Oslo, Norway

Abstract

South Asia's major national and regional languages have received a tremendous boost from the increasing availability of electronic media, which also makes it cheaper and faster to publish printed books and periodicals. Many are flourishing as new technologies and new media break the dominance of English or even of official languages (like Hindi).

Less is known about the "minor" languages: those which do not have official status as national, state or provincial languages, or in some cases, may not yet have a standard form or a widely-accepted writing system. This panel asks: what is the status of lesser-known languages in South Asia today? Are there trends which can be observed throughout the region? The panel embraces a range of topics:

Linguistic studies of lesser-known languages;

Sociolinguistics of lesser-known languages

Language shift and maintenance

language loss

Lesser-known languages and literacy issues

Lesser-known languages and educational issues

Language policies and lesser-known languages

Documenting oral traditions in lesser-known languages in South Asia;

Language contact between lesser-known languages and major languages

The impact of technology on lesser-known languages

Impact of globalization on lesser-known languages.

Anju Saxena, Uppsala University, Department of Linguistics Uppsala, Sweden

Ergative as a Discourse Marker in Kinnauri Narratives

A good narrative (whether spoken or written) is experienced by its audience as a live performance, where a series of events unfold right in front of its audience. Linguistic (and other) cues help listeners follow the narrative as well as experience it as a live performance. The ergative marker in Kinnauri occurs with transitive verbs, but its occurrence is not obligatory. It seems to function as a linguistic tool to describe a change in perspective in Kinnauri narratives. The almost obligatory occurrence of the ergative marker with an explicit subject in the he said,construction can be seen as a deictic marker which draws the listener's attention to the change in the mode of narration - from the descriptive to the expressive mode. The ergative marker in non,he said,constructions occurs regularly in situations where the clause describes something which is against expected behavior (including social norms), or it describes the magnitude of surprise, or it describes urgency. The ergative marker in such situations functions as a discourse marker, the aim of which is to direct the listeners' attention away from the default expectation mode.

Bettina Zeisler, University of Tuebingen, Tübingen, Germany

Work in progress: introduction to a "Valency Dictionary of Ladakhi Verbs"

Current linguistic classifications of Tibetan case marking represent only four or five sentence patterns. A closer look reveals as much as eleven basic sentence patterns for all Tibetan languages and quite a few additional mar,ginal patterns. In Ladakhi, where case marking is much more semantically motivated than in Old and Classical Tibetan, I have found as much as 50 different sentence patterns. As fieldwork in 2005 has shown, not only the lexicon, but also the grammar and thus the sentence patterns vary considera,bly between the two main Ladakhi dialect groups: Shamskat (Lower Ladakh, Nubra, and Purik) and Kenhat (Upper Ladakh, including Zanskar).

In 2002 I started a systematic survey of the argument structure of Ladakhi verbs focussing also on pattern variation and, increasingly, on fixed colloca,tions and idiomatic expressions. During fieldwork in the years 2002 to 2004 we were able to sample a nearly complete set of 782 verbs or main entries (plus 345 additional readings) for the Domkhar dialect of the Shamskat group. In 2005 we continued to sample data from the Gya dialect of the Kenhat group.

The data is collected in an XML annotated database, which does not only enhance the encoding and retrieval of various information, but also allows a conveniently searchable online publication in the future. At present, the da,tabase contains almost 900 main entries with more than 650 additional read,ings (Lesarten) of particular verbs and more than 7000 example sentences. In contrast to the available dictionaries, the necessary examples for each en,try or reading show complete sentences with the full set of arguments, but there might be also additional, and partly ‘incomplete', examples for par,ticular grammatical features.

I will discuss some ‘technical' problems of representing pattern variation, but also some special issues of case marking in West Tibetan.

Khawaja A. Rehman, Peshawar, Pakistan

A Brief Survey of the Languages of the Neelam Valley

The Neelam valley in Pakistan,administered Kashmir remains a relatively uncharted territory on the linguistic map. The Linguistic Survey of India (edited by Grierson in the early 20th century) does not contain substantive information on the languages of the valley, and neither does one find much information about the area in the more recent literature. The fact that the Line of Control runs right through the valley is one obvious reason as to why the area is hardly accessible to researchers.

In my paper I will present a brief overview of the language varieties spoken in the valley, based on recent research as well as my experience as a resident of the area. These varieties include forms of languages that are spoken widely elsewhere, such as Hindko, Gojri, Shina (Gurezi and Chilasi), Kashmiri, and even Pashto, but also the rather distinct language of the village of Kundal Shahi, located near the Neelam district headquarters, Authmuqam (cf. Rehman & Baart 2005).

My research shows that the predominant language of the area (which is called "Paarmi" by the local population, and sometimes also "Hindko" or "Pahari") is a form of Northern Hindko, very similar to the variety spoken in the district of Mansehra. Virtually all members of the other language communities are bilingual in Hindko. A process of language shift to Hindko is going on in many of these communities. In some of these communities this process started relatively recently, while in others it has been going on much longer.

Reference:

Rehman, Khawaja A., & Joan L.G. Baart (2005). A first look at the language of Kundal Shahi in Azad Kashmir. SIL Electronic Working Papers 2005,008.

Jan Heegård, University of Copenhagen, Copenhagen S, Denmark

Topographic and Referential Parameters in Ablative Case,Marking in Kalasha

In Kalasha, an Indo,Aryan language spoken in the Hindu Kush in Northwest Pakistan, three ablative case endings, ",ey", ",ani", and ",au", are identified. Only ",ani" and ",au" can occur on common nouns: (1) mizok gAng,ani par,iu ‘the mouse runs away from the hole' (mouse hole,abl runs).(2) uk ash,au pariu ‘water runs out of the mouth' water mouth,abl runs).

An important function of the ablative case suffixes is to provide information about how the Ground is to be conceptualized: ",ani" denotes a line,like motion from a Ground,source; and ",au" indicates that the Ground is hollow or three,dimensional in nature. However, other data suggest that a referential parameters also are involved in ablative case,marking: ",au" is associated with either plurality or an unbounded or unidentifiable Ground from which separation takes place. In contrast, ",ani" may be associated with singularity and a bounded or identifiable source Ground. As such, the ablative case markers expose striking similarities with the distribution of the locative case endings.

Judged from a survey of ablative case markers in other Hindu Kush languages (Edelman 1983, Bashir 2003) the relevance of referential parameters seems not to be widespread among languages of the area. Topographical parameters, however, are often mentioned as distinguishing ablative markers, including ablative postpositions.

My paper will give an account of the distributional pattern of the three ablative case endings and I shall show how they are to be distinguished semantically from ablative postpositions. I shall illustrate the relevance of the topographical and referential parameters, respectively, and I shall give a suggestion as to how the referential parameters may be derived from the topographic parameters.

Ram Lohani, Tribhuvan University, Kirtipur, Nepal

Lesser known languages of Nepal and ICT localization

NeLRaLEC (Nepali Language Resources and Localization for Education and Communication) project is an attempt to ice,break this gap in local languages. To localize ICT in Nepalese languages it prepares corpora and compiles corpus,based dictionary of Nepali, develops software for speech technology and open type fonts for Devanagari Unicode and prepares and trains manpower for sustainable development of ICT in Nepal. Beginning work in Nepali is basically required because this language, so far, is the means of education and media and communication. The project has its aims to further these activities in other languages as well. Technology and method of designing written, spoken and parallel corpora and principle of annotation developed for Nepali can be the starting point while working with other languages. Collection of spoken corpora will basically be the model for documenting languages that have no written tradition. Devanagari fonts (with or without additional diacritics) can be used to facilitate written tradition. Thus prepared corpora can be the basic resources for dictionary (bi,/multilingual) and grammatical analysis.

Mark Turin, University of Cambridge, Amsterdam, The Netherlands

A Lingustic Survey of Sikkim: A Report from the Field

In the last months of 2005, with support from the Department of Human Resource Development in Sikkim (formerly Education) and the Namgyal Institute of Tibetology in Gangtok, I conducted the first phase of the first modern linguistic survey of the Indian State of Sikkim. At EASAS 2006, I propose to present the preliminary findings of this survey.

A sovereign nation until 1975, Sikkim has a geopolitical importance out of proportion to its size. It shares an eastern border with Bhutan, a northern border with the Tibetan Autonomous Region of China and a western border with Nepal. To the south lies the adjacent state of West Bengal. Sikkim is home to scores of languages, including Lepcha, a Tibeto,Burman tongue indigenous to the region, and Bhutia , or Denjonke , spoken by an ethnic community of the same name who populated Sikkim's fertile valleys from eastern Tibet from the 13th century onwards.

Since Risley's excellent Gazetteer of Sikhim at the end of the 19th century, which included a section on the languages spoken by the inhabitants of the then Kingdom, the decadal Indian Census has continued to collect limited information on languages spoken across the Indian Union. From a linguistic perspective, however, the data are lacking as they say little about language use, multilingualism, mother tongue instruction in schools and the use of officially recognised state languages in administration. The present linguistic survey of Sikkim, which will continue in 2006 and through 2007, is comprehensively addressing issues of language use, and is comprised of three discrete but intersecting sections: language and population, language and education, and language and administration. At EASAS 2006, I will present for the first time an analysis of 1,200 survey questionnaires completed by school children in all four of Sikkim's districts, as well as findings based on limited travels to remoter locations in the State, such as the protected Lepcha reservation of Dzongu in North Sikkim.

Boris Zakharin, Moscow State University, Moscow, Russian Federation

Forms of address in Jaina Ardha,Ma:gadhi:( in comparison with Hindi: )

The New Indo,Aryan is generally characterized by a triple opposition of 2,nd person pronominal forms , e.g., in Hindi: tu: ‘thou' (to a very inferior or very intimate) , a:p ‘You' (to superiors in social status, age, sex etc.; formal) , tum ‘you' (to equals; informal), and by the correlating triple opposition of the forms of the imperative: kah ‘tell!' (intimate or impolite) , kahiye ‘tell, please!' (polite) , kaho ‘tell!' (neutral). Systems based on the triple oppositions of Hindi type did not exist in Ancient Indo,Aryan. They first appear in Pra:krits, and their pattern will be analysed for Ardha,Ma:gadhi: Jaina texts of the 1,st millenium a.d. , such as ‘Viva:ga sutta' and ‘Na:ya:dhammakaha:'.
The opposition ‘sg. , pl.' of the category ‘Number', relevant for the 2,nd person pronouns, for the imperative forms and for some nominals [such as ‘bho , bhaMte' (neutral or impolite) "o thou, the noble one(s)" , (polite) "o You, the noble one", or for ‘deva:Nuppiya,' "beloved of the gods", different Gender/Number forms of which are bound to the same opposition frame], serves as the main means of expressing grades of politeness in address forms of Ardha,Ma:gadhi:.
The analysis undertaken is provided for a certain set of the situation,types, among which the most important seem to be acts of communication between unequals , such as, e.g., ‘bhikkhu talking to Maha:vi:ra' (and vice versa), ‘ra:ja: , to his queen' (and vice versa), ‘father or mother , to his/her son' (and vice versa), ‘ra:ja: or queen , to a servant' (or vice versa) etc. It is shown that the triple opposition system existing in Ardha,Ma:gadhi: is of approximately the same type that we find in Hindi, with one important exception: as Ardha,Ma:gadhi: does not possess either the pronouns of the type ‘a:p' "You" or the corresponding polite imperative forms (of the type ‘V,(j)iye'), the (honorific) forms of the 3,d person plural imperative are used instead to express utmost respect. , For example, when ra:ja: and his queen address Maha:vi:ra asking him to accept their son as a disciple of the monastic Order they say: ‘paDiccha,Mtu (imp. 3 pl.) Na,M deva:Nuppiy,a:' "Please, receive (lit. ‘let them receive') him, o beloved of the gods (voc. pl.,hon.)".

The detailed analysis of the cases of use of the form ‘deva:Nuppiya,' in different contexts of Ardha,Ma:gadhi: texts is to be provided with Sanskrit and Pa:li equivalents and parallels from some other languages (compare, e.g., Russian ‘l'ubimets bogov') supplied.

Umberto Ansaldo & Lisa Lim, University of Amsterdam, Amsterdam, The Netherlands

Endangerment, Empowerment and Documentation: Dilemmas of Sri Lanka Malay

This paper addresses the issue of language shift among minority groups in Sri Lanka and the role of the researcher involved in language documentation.

The diasporic Malay communities of Sri Lanka (Javinnsu in Sri Lankan discourse) were brought to Sri Lanka through various waves of deportation from Indonesia by the Dutch and British colonial powers. Though lacking official identity, being grouped together with other minority groups under the label 'Muslims', the Sri Lanka Malays are characterized by a unique language of trilingual base (often referred to as Sri Lanka Malay creole) in which Sinhala and Tamil grammar and colloquial Malay lexicon merge (Ansaldo 2006). At present different degrees of loss of this vernacular can be noticed, in favor of the country's dominant languages (Sinhala, Tamil and English). Moreover, a new tendency can be detected in some communities: the desire to acquire standard Malay, the national language of Malaysia, in order to (a) gain a useful economic tool and (b) preserve and represent their subject position through an assumed global Malay identity , in the sense of Pavlenko and Blackledge (200!
4) , recognizable through an empowered linguistic identity, and recognized by the Malay 'homeland'.

The trend of shift towards a recognized standard in which cultural and economic functions of language come together can be interpreted as a consequence of globalization as a resource , rather than a threat , to enable a minority group to gain access to better education and enhanced political self,representation. Within an ethnography of empowerment (Delgado,Gaitan 1994, 1996), the researcher is urged to intervene in advocating and creating conditions for the recognition of minority voices; in this sense, the field linguist may be required to shift his orientation from advocating linguistic human rights to promotion of mobility and standard in a community.

Liudmila Khokhlova, Moscow University, Institute of Asian and African Studies

Moscow, Russian Federation

Clarifying the typological status of the Marwari language

The aim of this paper is to discuss some syntactic properties of one of the lesser known Western Indo,Aryan languages of Rajasthani group, i.e. of Marwari. The main objectives of the present study are: 1) to reconsider the definition of syntactic features specific for languages of South Asia and to determine the nature of their interaction; 2) to compare the hierarchy of ergative features that underwent attrition in course of Marwari's history with the hierarchically ordered sets of ergative features in modern South Asian languages. These sets described by M. Klaiman are structured in such a way that a language that lacks ‘A' will not display ‘B', a language with ‘A' but without ‘B' will lack ‘C', and so on. Out of four morphological devices described by M. Klaiman we shall consider here three that are relevant for Marwari: nominal case marking, nominal agreement on main verbs, nominal agreement on auxiliary verbs.
According to nominal case marking hierarchy suggested by M. Klaiman, Marwari is the sole exception for the rule that ergative case marking occurs somewhere in the system. This assumption is based on D.Magier's description of ergativity in Marwari. However, both my field work and the texts published in Marwari show that Magier's description is true only for singular nominal paradigms, while plural nouns have ergative type of nominal case marking in perfective domain. That means that instead of being put at the top of Klaiman's nominal case marking hierarchy, Marwari should occupy the bottom of it together with other languages demonstrating different types of nominal hierarchy split (Punjabi, Marathi). The details of nominal hierarchy split in Marwari will be described in my paper.

Another misconception concerning Marwari that has been repeated after Magier in many publications on South Asian typology, is the statement about the ‘accusative type' of present tense auxiliary verb agreement in ergative domain. I have also come across a few examples of this type in Marwari texts. However, neither my own field work, nor my analysis of literary texts prove the spreading of this pattern. The verbal concord split remains peculiar for only some of the idiolects, and it is restricted by 1,st person pronouns, the highest on animacy hierarchy scale. Lakhan Gusain , being a native speaker of Rajasthani, gives in his book samples demonstrating only ergative/default auxiliaries' agreement in Marwari. All these facts allow us to treat Marwari as occupying much less "exceptional" place on M. Klaiman's typological scale.

M. Klaiman also suggests a set of implicationally related ergative,accusative behaviour,patterns: if a given South Asian language displays any ergativity at all, then the farther it is placed on the scale, the more ergative it is. Thus, if the system has tense,aspect split and lacks the identified object marking (IOM) in ergative domain, it is a prerequisite for lacking IOM in non,ergative domain and for having verbal concord with direct object (O) for person. It is shown in the paper that the same implications might be also stated for Marwari's history.

At the most "ergative" stage of its development the language had no IOM and had object,bound verbal concord for person, later it has developed IOM in non,ergative domain, and still later - also in ergative domain, the object,bound verbal concord for person has completely stopped to exist.

Claus Peter Zoller, Institut for kulturstudier og orientalske språk (IKOS),

Oslo, Norway

Further evidence for an influence of Dardic languages on Romani

Further evidence for an influence of Dardic languages on Romani By Claus Peter Zoller

R. L. Turner has shown in his seminal paper from 1926 "The position of Romani in Indo,Aryan" that Romani is with regard to its linguistic phylogeny a Central Indo,Aryan language. According to the same author, the Northwest Indo,Aryan and Dardic elements in Romani, which he believed to have identified, entered Romani when its speakers were on their way out of India. Turner's first hypothesis has found wide acceptance, but not his second one, which has been criticized by Yaron Matras as "inconclusive".

In this paper I will show that Turner's second hypothesis is in fact correct, and that there are several archaic features and several phonological and morphological innovations plus additional vocabulary which are exclusively shared by Romani and Dardic languages, and partly by Northwest Indo,Aryan languages. I will first put Turner's arguments from phonology and lexicon on a more solid database. Then I will present new arguments from phonology and morphology

Metathesis of aspiration (comparable to the well,known Northwest Indo,Aryan r,metathesis) and "spontaneous" aspiration. Metathesis for instance in the Dardic language Indus Kohistani: phitsili ‘slippery' from Old Indo,Aryan picchila,; and in Romani: phuèel 'to ask' from Old Indo,Aryan pºcchati.
Devoicing of voiced aspirated stops and affricates. For instance, both Dardic Kalasha and Romani have the word khas ‘grass' from Old Indo,Aryan ghâsa,.

• Specific agent noun and conjunct verb constructions

The new data will not only corroborate Turner's theses, they also suggest that Dardic has left decidedly more distinct traces in Romani than previously assumed. One might tentatively conclude that the contact between the speakers of proto,Romani and of Proto,Dardic was longer and more intensive than previously conjectured.

Ruth Laila Schmidt, University of Oslo, Oslo, Norway

Lexical and Grammatical Properties of Accent in Kohistani Shina

Linguists of Shina classify Shina as a pitch accent language. This analysis states that accent is the occurrence of a high pitch on a specific syllable or part of a syllable (mora); and that the placement of the accent predicts tone. In Kohistani Shina there are four possibilities: (a) an accented short vowel (high pitch), (b) a long vowel accented on the first mora (falling), (c) a long vowel accented on the second mora (rising), and (d) an unaccented vowel (level). Minimal pairs exist for the following syllable types:

short accented : short unaccented

short accented : long accented on first mora : long accented on second mora Accent is not predictable, so is a lexical feature.

In nouns, the accent may remain on the stem or shift to the suffix. A short accented vowel may also be lengthened in inflection to a first mora accent or a second mora accent.

dón ‘tooth' > dódi ‘teeth' (stem: dód,)

dón ‘herd' > doní ‘herds' (stem: don,)

mushá ‘man' > musháaR ‘to the man'

batshó ‘calf' > batshoóR ‘to the calf''

In verbs, while the type of accent is a lexical feature, accent shift has grammatical (derivational) function. In most intransitive verb classes, it remains on the root; whereas in most transitive classes, it falls on the suffix. Historically, accent shift is a marker of valency increase.

car! ‘graze!'

cáram ‘I may graze'

cará! ‘take for grazing!'

carám ‘I may take for grazing'

byé! ‘sit!'

bayée! ‘put!'

This paper looks at Kohistani Shina grammar and asks: To what extent is it analytically useful to distinguish between strong (accented) and weak (unaccented) stems and roots? Do phonemic short accented vowels actually represent a neutralization of the contrast between first,mora accents (falling) and second,mora accents (rising); thus pointing to an analysis in terms of tone?

Further evidence for an influence of Dardic languages on Romani By Claus Peter Zoller

The aim of this paper is to discuss some syntactic properties of one of the lesser known Western Indo,Aryan languages of Rajasthani group, i.e. of Marwari. The main objectives of the present study are: 1) to reconsider the definition of syntactic features specific for languages of South Asia and to determine the nature of their interaction; 2) to compare the hierarchy of ergative features that underwent attrition in course of Marwari's history with the hierarchically ordered sets of ergative features in modern South Asian languages. These sets described by M. Klaiman are structured in such a way that a language that lacks ‘A' will not display ‘B', a language with ‘A' but without ‘B' will lack ‘C', and so on. Out of four morphological devices described by M. Klaiman we shall consider here three that are relevant for Marwari: nominal case marking, nominal agreement on main verbs, nominal agreement on auxiliary verbs.According to nominal case marking hierarchy suggested by M. Klaiman, Marwari is the sole exception for the rule that ergative case marking occurs somewhere in the system. This assumption is based on D.Magier's description of ergativity in Marwari. However, both my field work and the texts published in Marwari show that Magier's description is true only for singular nominal paradigms, while plural nouns have ergative type of nominal case marking in perfective domain. That means that instead of being put at the top of Klaiman's nominal case marking hierarchy, Marwari should occupy the bottom of it together with other languages demonstrating different types of nominal hierarchy split (Punjabi, Marathi). The details of nominal hierarchy split in Marwari will be described in my paper.

This paper attempts to discuss the status of documenting oral traditions in lesser known languages of Nepal with special reference to South Asia. With common linguistic heritage of numerous languages of four linguistic families such as Indo,European, Dravidian, Tibeto,Burman and Austro,Asiatic, the major and minor languages of the area contain rich folkloric elements. The lesser known languages of the region have continued to help identify the speakers of these languages and also to share the changes and dominance of the major linguistic communities. The themes, characters and messages of the written traditions of the major languages have come down to the oral traditions of various minor linguistic groups, and their own versions have become the markers of their identities. The great traditions of Ramayana and Mahabharata are some examples. The rich oral traditions of South Asian nations are changing rapidly because the languages which carry these traditions are also changing. Gradual increase in literacy in the dominant language of the country, mass media, internal migration and urbanization, loss of language are some of the reasons for putting various minority languages and their oral tradition at the verge of extinction. Though some attempts have been made and are being made to preserve the oral traditions of various speech communities the process is quite slow in comparison to the process of extinction. Now, valuable knowledge and wisdom as well as the verbal art of the lesser known speech communities need special attention. As preserving and promoting oral traditions are urgently needed tasks of the linguists and folklorists, doing systematic fieldworks for documenting oral traditions in South Asian region is a new challenge. The aim of this paper is to discuss some syntactic properties of one of the lesser known Western Indo,Aryan languages of Rajasthani group, i.e. of Marwari. The main objectives of the present study are: 1) to reconsider the definition of syntactic features specific for languages of South Asia and to determine the nature of their interaction; 2) to compare the hierarchy of ergative features that underwent attrition in course of Marwari's history with the hierarchically ordered sets of ergative features in modern South Asian languages. These sets described by M. Klaiman are structured in such a way that a language that lacks ‘A' will not display ‘B', a language with ‘A' but without ‘B' will lack ‘C', and so on. Out of four morphological devices described by M. Klaiman we shall consider here three that are relevant for Marwari: nominal case marking, nominal agreement on main verbs, nominal agreement on auxiliary verbs.According to nominal case marking hierarchy suggested by M. Klaiman, Marwari is the sole exception for the rule that ergative case marking occurs somewhere in the system. This assumption is based on D.Magier's description of ergativity in Marwari. However, both my field work and the texts published in Marwari show that Magier's description is true only for singular nominal paradigms, while plural nouns have ergative type of nominal case marking in perfective domain. That means that instead of being put at the top of Klaiman's nominal case marking hierarchy, Marwari should occupy the bottom of it together with other languages demonstrating different types of nominal hierarchy split (Punjabi, Marathi). The details of nominal hierarchy split in Marwari will be described in my paper.

This paper addresses the issue of language shift among minority groups in Sri Lanka and the role of the researcher involved in language documentation.

The New Indo,Aryan is generally characterized by a triple opposition of 2,nd person pronominal forms , e.g., in Hindi: tu: ‘thou' (to a very inferior or very intimate) , a:p ‘You' (to superiors in social status, age, sex etc.; formal) , tum ‘you' (to equals; informal), and by the correlating triple opposition of the forms of the imperative: kah ‘tell!' (intimate or impolite) , kahiye ‘tell, please!' (polite) , kaho ‘tell!' (neutral). Systems based on the triple oppositions of Hindi type did not exist in Ancient Indo,Aryan. They first appear in Pra:krits, and their pattern will be analysed for Ardha,Ma:gadhi: Jaina texts of the 1,st millenium a.d. , such as ‘Viva:ga sutta' and ‘Na:ya:dhammakaha:'.The opposition ‘sg. , pl.' of the category ‘Number', relevant for the 2,nd person pronouns, for the imperative forms and for some nominals [such as ‘bho , bhaMte' (neutral or impolite) "o thou, the noble one(s)" , (polite) "o You, the noble one", or for ‘deva:Nuppiya,' "beloved of the gods", different Gender/Number forms of which are bound to the same opposition frame], serves as the main means of expressing grades of politeness in address forms of Ardha,Ma:gadhi:.The analysis undertaken is provided for a certain set of the situation,types, among which the most important seem to be acts of communication between unequals , such as, e.g., ‘bhikkhu talking to Maha:vi:ra' (and vice versa), ‘ra:ja: , to his queen' (and vice versa), ‘father or mother , to his/her son' (and vice versa), ‘ra:ja: or queen , to a servant' (or vice versa) etc. It is shown that the triple opposition system existing in Ardha,Ma:gadhi: is of approximately the same type that we find in Hindi, with one important exception: as Ardha,Ma:gadhi: does not possess either the pronouns of the type ‘a:p' "You" or the corresponding polite imperative forms (of the type ‘V,(j)iye'), the (honorific) forms of the 3,d person plural imperative are used instead to express utmost respect. , For example, when ra:ja: and his queen address Maha:vi:ra asking him to accept their son as a disciple of the monastic Order they say: ‘paDiccha,Mtu (imp. 3 pl.) Na,M deva:Nuppiy,a:' "Please, receive (lit. ‘let them receive') him, o beloved of the gods (voc. pl.,hon.)".

The inflectional properties of Pashto nouns are fully investigated. The main focus is on the classification of the Pashto nouns in such a way that further computational work on them becomes easy. In the proposed classification, the class membership of each noun is unambiguous and the morphological rules for these classes are uniform. The work of three leading Pashto grammarians, Penzl (Penzl, 1955), Reshteen (Reshteen, 1994) and Zyar (Zyar, 2003), is reviewed that provides the basis for the proposed classification. There are different conjugations for the nouns ending in different phonemes. So, the ending phonemes of the nouns are of great importance in the classification. The characteristics of nouns, having a particular type of ending phoneme, are closely observed from a large collection of examples. Consequently, the Pashto nouns are classified based on these observations. This classification is such that the nouns, showing the same characteristics, are packaged into a single class.

Despite its small size, Nepal homes about one hundred languages (Ethnologue (2005) enumerates 123 and Government's report (2001) says 92 plus languages). The constitution of Nepal (1990) has declared Nepali as the language of the nation and the official language. Written in Devanagari script, Nepali is compulsory in school education. It is the lingua franca for most of the non,native speakers. All other languages are identified as National languages. Some of the Tibeto,Burman languages have their own written tradition (Tibetan, Limbu), some have adopted Devanagari (with or without additional diacritics) for printing (Newar, Gurung, Tamang), and some are developing their own script (Magar). Rest of the languages have only oral tradition. Indo,Aryan languages use Devanagari if they have written tradition, otherwise they exist only in oral communication. Literacy rate is very low among the speakers of minor languages. Access to ICT, so far, is not possible even in Nepali. Inability to access information available in the web and inability to handle technology have seriously impaired their knowledge level, rights and living style. The inflectional properties of Pashto nouns are fully investigated. The main focus is on the classification of the Pashto nouns in such a way that further computational work on them becomes easy. In the proposed classification, the class membership of each noun is unambiguous and the morphological rules for these classes are uniform. The work of three leading Pashto grammarians, Penzl (Penzl, 1955), Reshteen (Reshteen, 1994) and Zyar (Zyar, 2003), is reviewed that provides the basis for the proposed classification. There are different conjugations for the nouns ending in different phonemes. So, the ending phonemes of the nouns are of great importance in the classification. The characteristics of nouns, having a particular type of ending phoneme, are closely observed from a large collection of examples. Consequently, the Pashto nouns are classified based on these observations. This classification is such that the nouns, showing the same characteristics, are packaged into a single class.

Mohammad Abid Khan, University of Peshawar, Department of Computer Science, Pakistan
The Computational Morphology of Pashto Noun -- The Classification

The proposed classification consists of seven classes for masculine nouns and seven classes for feminine nouns. The theoretical investigation is done in such a way that it will make the computational application e.g. in a corpus-based Pashto dictionary easy. This dictionary is supposed to have root words, prefixes, suffixes, infixes and rules of morphology of Pashto language. The dictionary, in turn, will pave the way for other applications such as part-of-speech tagging and syntactic analysis of input Pashto text. Consequently, larger applications such as question-answering in Pashto by computer and machine translation from Pashto into other languages will become less difficult.

links inline

Home

Newsletters

European Conferences on South Asian Studies

PANEL 31: Linguistics of lesser-known languages in South Asia