Journées d'étude "Grammaires étendues"

Jeudi 03 Novembre 2016, 09:30 to Vendredi 04 Novembre 2016, 17:00
Émilie Aussant et Aimée Lahaussois (HTL)

ODG – Salle Laplanche (576)

Jeudi 3 novembre

Thème : Asie

9:30-10:15 Émilie Aussant
Extended Grammars and Extended Sanskrit Grammar
After having presented the main lines of the Extended Grammars research programme, this paper focuses on what Jean-Luc Chevillard and I call the « Extended Sanskrit Grammar ». The aim is to draw up an overview of the various researches led on the topic so far and, from there, to start to bring out the features of Sanskrit grammatical descriptions’ spread as well as their adaptations to languages other than Sanskrit (Indo-aryan languages, Dravidian languages, South-East Asia languages, Tibetan languages, Buriat, Algonquian…). In other words, it is about reviewing what was extended, how and why.

10:15-11:00 Nalini Balbir
Les grammaires étendues : du sanskrit au moyen-indien
Les langues moyen-indiennes présentent une situation contrastée : les prakrits littéraires sont décrits par des grammairiens qui utilisent le sanskrit, modèle normatif par excellence ; en revanche, le pali (langue des écritures du bouddhisme Theravāda), est décrit au moyen du pali. On se propose d’examiner les raisons susceptibles d’expliquer cette différence, alors même que prakrits et pali présentent de nombreux points communs dans leurs évolutions phonétiques ou morphologiques, et d’explorer dans quelle mesure l’utilisation d’une autre langue, le pali, est plus qu’une différence superficielle. On verra, notamment, comment fonctionne la description du verbe et comment la notion de racine, centrale en sanskrit, est traitée dans des états de langue où les conjugaisons, tendant à prendre pour forme de base le thème du présent, développent des paradigmes réguliers.

11:00-11:15 pause café

11:15-12:00 Jean-Luc Chevillard
The organization of the Syntax chapters inside Beschi's 1738 Grammatica Latino-Tamulica and Walther's 1739 Observationes Grammaticae
Constantius Joseph Beschi [1680-1746/1747], an Italian jesuit, and Christoph Walther [1699-1741], a protestant missionary, both contributed to the “extended” description of Tamil during the early part of the 18th century, by composing respectively the Grammatica Latino-Tamulica [ubi de Vulgari Tamulicae Linguae idiomate ...] (dated 1728 in the introduction) and the Observatione grammaticae [quibus linguae tamulicae idioma vulgare ...]. Their works, which both appeared in printed form in Tranquebar, respectively in 1738 and in 1739, and seem to have sometimes circulated bundled together, to the displeasure of Beschi, have respectively 175 pages and 60 pages, and both contain FOUR main parts [I. De litteris (26 p., for Beschi, versus 12 p., for Walther); II. De nomine et pronomine (17 p. vs. 7 p.); III. De Verbo (33 p. vs. 17 p.) ; IV. De syntaxi (67 p. vs. 21 p.)], followed in Beschi’s case by a fifth section [V. De variis quotidiano usui praecipue necessariis (26 p.)], encyclopedic in nature. This presentation, which is part of an ongoing effort of translating Walther’s Observationes from Latin into English, will examine the fourth sections of both works, concentrating on the divergences between the two authors.

12:00-12:45 Victor D'Avella 
Vīracōḻiyam’s case system
My presentation will focus on the case system in the Vīracōḻiyam, an 11th century grammar of Tamil by Puttamittiraṉ together with the gloss of Peruntēvaṉar, and its likely antecedents in Sanskrit grammatical works. The description of the case system in Pāṇini's grammar, from which all later grammars borrow or draw inspiration, involves a two step process. First, six kārakas “factors” are defined. Each one is an object that has a particular role in bringing about an action, e.g. the agent, the location etc. The seven cases are then variously mapped onto the nouns so labelled under certain conditions. In later works, each kāraka is further subdivided into upakārakas “sub-factors” in a manner somewhat akin to the various types of cases in Latin grammar, e.g. ablativus instrumentalis. In my presentation I will show how Puttamittiraṉ and Peruntēvaṉar borrowed and adopted the terminology for the kārakas and upakārakas into Tamil with particular reference to the two non-Pāṇinian schools of grammar, the Cāndravyākaraṇa and Kātantra, as well as the Vararucisaṅgraha and the allied Prayogamukha, popular and widely spread works that summarize some basics of Sanskrit grammar.

13:00-14:30 déjeuner 

14:30-15:15 Deven Patel
Metalinguistic terminology in the Āndhraśabdacintāmaṇi, an early grammar of the Telugu language
One of the more remarkable yet understudied features of early India’s celebrated contribution to linguistic science is the regional-language grammar composed in Sanskrit. A unique contribution in this regard is a short text entitled the Āndhraśabdacintāmaṇi [“A treatise (literally: a wish-fulfilling gem) on Andhra speech-forms”], a grammar of the Telugu language composed in Sanskrit. Tradition (and a few modern scholars) date the work to the 11th century while most hold it to be a work of the 16th or 17th century. The grammar consists of two hundred and seventy-four sūtras and is organized into five chapters (pariccheda): saṃjñā (metalinguistic terminology), sandhi (euphonic coalescence), ajanta (nominal bases/verbal stems ending in vowels), halanta (nominal bases/verbal stems ending in consonants), and kriyā (verbal action). While the Āndhraśabdacintāmaṇi closely follows Pāṇini’s Aṣṭādhyāyi in the construction of sutras and in its use of technical terms, the structuring of topics and explanations of grammatical issues show awareness of the Kātantra grammar, Vararuci’s Prākṛtaprakāśa, Hemacandra’s Siddhahaima-śabdānuśāsana, Kaccāyana’s Pāli grammar, and Bhaṭṭoji Dīkṣita’s Siddhāntakaumudi. In addition to offering a brief overview of the entire work, this presentation focuses especially on Chapter 1 of the Āndhraśabdacintāmaṇi, which, in the spirit of this workshop, extends inherited technical, “meta-linguistic” terminologies (saṃjñā) found in Sanskrit, Prakrit, and Pali grammars by introducing wholly new terms to either amplify or refine existing grammatical features or to identify linguistic characteristics that are distinctive to the Telugu language.

15:15-16:00 Giovanni Ciotti
On the Sanskritic Condition for the Application of Sandhi and its Reception: From Nacciṉārkkiṉiyar to Max Müller
That phonological rules, or rules pertaining to the mutation of sounds, occur under specific conditions (or domains) is a linguistic idea shared by several grammatical traditions. Within the Sanskritic grammatical tradition it corresponds to the concept of saṃhitā (“combination”), a term with a rather rich history in itself within South Asia. This presentation aims at reconstructing some of the steps of the reception that such a concept has had in the history of Indian and Western linguistics. As the Sankritic grammatical tradition served as a model for other traditions, saṃhitā was part of the package of theoretical tools taken up by those who described other South Asian languages, such as Tamil. In the Tamil grammatical tradition one can in fact observe the reflections of the different interpretation of saṃhitā in Sanskrit. Much later and much far away, the same reflections can be observed as Sanskrit was discovered by European linguists and its indigenous grammatical tradition became the object of study of some prominent modern Western linguists, such as Max Müller.

16:00-16:15 pause café

16:15-17:00 Arlo Griffiths
The extension of Sanskrit poetological knowledge systems to Java: towards a new study of the Chandaḥkaraṇa alias Candrakiraṇa

Vendredi 4 novembre

Thème : Linguistique descriptive

9:30-10:15 Aimée Lahaussois
The shapes of verbal paradigms in Kiranti languages
Kiranti languages have polypersonal indexation, with two arguments encoded in verb agreement markers. In contemporary descriptions of Kiranti languages (from 1975 on), the tables presenting transitive verb paradigms are arranged in a rather similar way, in a matrix format with the different person/number combinations for the agent argument represented horizontally and the patient argument person/number combinations vertically. In essence, an areal style of presenting paradigms has come into being. In earlier grammars, however, a number of different formats for representing the combination of two arguments was used.
I shall present the different paradigm formats found in a sampling of grammars of Kiranti languages from the present day back to 1850, with a view to tracing the origins of the current layout, and, in cases where significantly different layouts are encountered, attempting to retrace the model which may have influenced the presentation of the data.

10:15-11:00 François Jacquesson
The Verb in Agglutinative languages: Analytic sequencing vs. form history (with notes on the related technical lexicon.)
Innu, like other Algonquian languages, has a complex verbal morphology, and morphemes are often not easily segmentable. Much of the terminology used in Algonquian linguistics was influenced by early Algonquianists, especially Leonard Bloomfield, and morphemes are described in structural terms. Innu verb stems consists of as least two morphemes, an initial and a final. They may also contain a medial. In addition, there is also the possibility of having morphemes in between these three, called postradical, premedial, postmedial, prefinal. Later, following the Bloomfieldian tradition, a distinction has been proposed between initial derivation for the derivation of a verb stem, and secondary derivation for morphemes that attach to a verb stem to form a new verb. In this presentation, I am going to explain what the verbal complex in Innu is and determine what these morphemes are, and show how the Bloomfieldian heritage impacts on the understanding of Algonquian languages.

11:00-11-15 pause café

11:15-12:00 Nicolas Quint 
Capeverdean grammar(ian)s: one more account of the difficulties of describing a language for its own sake
Capeverdean is an Afro-Portuguese Creole language natively spoken by cca. 1,000,000 people worldwide, including more than 95% of the total population of the Republic of Cape Verde (West Africa). Since 1887 (publication of the first Capeverdean grammar by António de Paula Brito), Capeverdean has been studied by several generations of scholars, among whom several native speakers of the language (e.g. Almada, Baptista, Cardoso, Veiga...).

12:00-12:45 Tatiana Nikitina
Descriptions of verbal morphology in early grammars of Mande languages

13:00-14:30 déjeuner 

14:30-15:15 Martine Vanhove
The verbal system of Beja: A look at the history of its terminology
The description of Beja, a North-Cushitic language spoken mainly in Sudan, has a short tradition of descriptive and grammatical studies which dates back only to the second half of the nineteenth century (Almqvist 1881). Like many other Afroasiatic languages, this tradition was, at its beginning, deeply entrenched in the western linguistic tradition of Semitic and Indo-European studies. The scholars’ analyses of the verbal system (Almqvist 1881; Reinisch 1893-94; Roper 1928; Hudson 1976; Dahl 1984; Morin 1995; Appleyard 2004) were mainly concerned with the morphological make up of TAM, derivational patterns and non-finite verb forms, as well as with the evolution of this system as compared with other Cushitic languages (Cohen 1973; Zaborski 1975; Voigt 1988). Labels to the various paradigms and patterns were often given without taking real care of the semantics of the grammatical categories underlying the terminology. This state of affairs resulted in a wide terminological discrepancy (see table 1), albeit with some commonalities regarding one of the paradigms and some derivational patterns. None of them, even Dahl’s (1984) study of the aspectual system which was based on his questionnaire and translated examples, looked at their actual use in natural discourse.
This presentation will focus on the transmission or non-transmission of the various terminologies from one scholar to another, and their implications for a correct analysis of the verbal system of Beja which lead until recently to the non-recognition of certain categories in Beja such as converbs or the middle derivation pattern for instance (Vanhove 2014 & forth.). TAM, derivational patterns and non-finite verb forms will be dealt with in turn.

15:15-16:00 Renée Lambert-Bretière
The verbal complex in Innu and the Bloomfieldian heritage

16:00-16:15 pause café

16:15-17:00 Discussion