Recent years have witnessed increasing awareness of the needs (i) to preserve and record endangered languages (cf. Miyaoka 2002, Tsunoda 2005), and (ii) to accumulate and assemble various typological datasets in order to arrive at a better understanding of typological and areal distribution of grammatical phenomena (cf. Yamamoto 2003). Typologists at Max Planck Institute for Evolutionary Anthropology at Leipzig, Germany have initiated various projects relating to the objective (i). They recently accomplished a challenging task directly relating to the objective (ii), through the cooperation and assistance of an international panel of typologists and field linguists. The World Atlas of Language Structures (WALS) is the final product of this collective endeavor.
This workshop is intended to bring this important typological research tool to the attention of the Japanese linguistic community, particularly those linguists who pay close attention to typological universals and areal variation.
In this presentation, I introduce the World Atlas of Language Structures (WALS) as a research tool for linguists interested in comparing language structures. WALS shows the world-wide geographical distribution of 142 structural features (consonant inventories, stress position, word order, argument coding, voice alternations, complex sentences, color terms, and so on). Each world map shows on average 400 languages, and each feature distinguishes 2-9 different values. A large proportion of the maps show interesting geographical patterns that cannot be solely due to inheritance, showing that structural diffusion must be more widespread and important in shaping languages than has previously been thought. The electronic version of the atlas (Interactive Reference Tool) also allows users to ask questions about correlations of structural features. The classical Greenbergian correlations, as well as many other correlations proposed by linguists of various theoretical orientations, can thus readily be tested.
Many of the languages of Asia have families of expressions in which a verb meaning 'eat' takes as its subject a noun phrase bearing a patient role. Although EAT-expressions are encountered across Asia in nearly every language family, their types, numbers, and semantic range vary. Regarding their distribution as an instance of areal diffusion our paper lists and categorizes EAT-expressions in selected languages of West Asia, South Asia, Southeast Asia, East Asia and Central Asia. Preliminary results from listing and mapping EAT-expressions in these languages indicate that they have a Central Asian origin and that they have spread into Southwest, South, and North-East Asia in a pattern similar to the one found by Masica (1976) in his pioneering study defining the Indo-Turanian linguistic area. WALS serves as an effective tool in the visualization of such areal phenomenon.
In this study, I will take up one feature, namely "number of cases" described under feature no. 49 in "The World Atlas of Language Structures (WALS) 2005" and investigate as to why some languages have rich morphological case system. With respect to number of cases, WALS shows that there are 24 languages that have more than 10 cases. This study discusses two points regarding rich morphological case system: (1) which kinds of cases the 24 languages have, and (2) reasons why they have so many cases. Finally, I claim that the languages with rich case system are in fact rich in a particular case, namely the locative case and poor in terms of the diversity of cases.