- Clause structure in South Asian languages
The languages of South Asia (India, Pakistan, Nepal, Bhutan, Bangladesh, the Mal-dives, and Sri Lanka) are genetically diverse. They belong either to the Indo-Aryan group (i.e. the eastern group of Indo-European), to the Dravidian group, to part of the Austro-Asiatic group, or to part of the Sino-Tibetan group of languages. Nevertheless, they show much convergence in terms of grammatical features. With respect to syntax these include head-final order of constituents, WH-in-situ, serial verb constructions, nonnominative subjects, and clause-final complementizers. Common features like these have given rise to viewing South Asia as a specific linguistic area (see Emeneau 1956, Masica 1976, 1991). While earlier research has mainly concentrated on sociolinguistic aspects of this linguistic area, it has become a focus of interest within linguistic typology and especially in the course of a revived interest in formal comparative language re-search. In their introduction to the present volume the editors emphasize the fact that it was in the 1980s and 1990s that investigations of South Asian languages began to have a critical impact on the development of grammatical theory. Starting out from an account of agreement, scope, and word order in Hindi, both editors have made important contributions in their previous work. Theories of variable word order, theories about the syntax-semantics interface, WH-scope, binding, and so forth, and theories of argument structure, predication, and Case can hardly ignore the results that have emerged from the study of South Asian languages over the years.
The present volume offers a collection of ten articles that deal with various aspects of clause structure in South Asian languages. The languages under investigation are 'big' languages throughout. Hindi (or Urdu) as one of the most-studied languages is next to Bangla (also known as Bengali), Kashmiri, and Sinhala (which is heavily influ-enced by Dravidian) at the center of interest in Indo-Aryan. The Dravidian languages that play a role in individual chapters are Kannada and Malayalam. Austro-Asiatic or Sino-Tibetan languages play no role in the present collection.1 The contributions are loosely grouped in four blocks: functional projections (R. AMRITAVALLI on the develop-ment of functional categories in Kannada, GILLIAN C. RAMCHAND on negation in Ban-gla), argument structure (K. A. JAYASEELAN on serial verb constructions in Malayalam, JEFFREY LIDZ on causativity and reflexivity in Kannada, P. MADHAVAN on light verb raising and the role of an empty preposition in Malayalam), Case theory (MIRIAM BUTT and TRACY HOLLOWAY KING, as well as ALICE DAVISON on Case in Hindi-Urdu), and movement phenomena (PAUL HAGSTROM on the movement of scope-sensitive particles in Sinhala and Japanese, AYESHA KIDWAI on topic movement and questions of the extended projection principle (EPP) with special reference to Kashmiri and German, [End Page 597] JAMES D. MCCAWLEY on different types of relative clauses and relative clause extraposi-tion in Hindi).
Most of the contributions are committed to the principles-and-parameters (P&P) framework as developed in Noam Chomsky's government and binding theory (GB) and continued in the minimalist program. The chapter by Butt and King is couched in lexical functional grammar (LFG). The authors argue that this framework offers advan-tages in capturing Case phenomena in Hindi-Urdu. McCawley's paper remains largely neutral but makes crucial reference to syntactic versus interpretive accounts of relative clause extraposition. The chapters by Amritavalli and by Butt and King include details about diachronic aspects.
R. Amritavalli's chapter, 'Some developments in the functional architecture of the Kannada clause', takes its starting point from observations about negation in the Dravi-dian language Kannada. It concentrates on a diachronic development according to which Kannada has undergone a change in which finiteness was first expressed by agr(eement) and then turned into a projection of Mood (MoodP), thereby allowing finiteness...