Until very recently the 15-20 languages of the Alor-Pantar (AP) archipelago in southeastern Indonesia were among the least well-documented languages of Indonesia, but a surge in field work efforts over the past decade has resulted in a wealth of new language data. This research project focuses on the extended documentation and investigation of these non-Austronesian (‘Papuan’) languages.
The project hosted by Leiden University constitutes the Dutch part of a larger international collaboration conceived under the European Science Foundation, EUROCORES Programme, EUROBABEL. In our project 7 researchers from the US, the Netherlands, and the United Kingdom collaborate, with a budget of about 1 million E.
In this project, we collect additional documentation on two domains that have turned out to be particularly interesting in the AP languages, and have not yet been studied in depth. First, we investigate the domain of reference in space: how do speakers structure the spatial domain? How are locative expressions constructed? How do they use landscape terms? The AP languages employ a particularly rich array of spatial parameters, resulting in relatively complex deictic systems.
Second, we will extend our documentation of numeral expressions in the AP languages. As far as numeral systems are concerned, the world is decimal: about 80% of the world’s languages are decimal or combine a decimal with a vigesimal (base-20) system. While bases other than 10 are extremely rare , many AP languages show traces of a quinary (base-5) system in the lower cardinals (e.g. Teiwa yes haraq ‘seven’ = yes ‘five’ + haraq ‘two’), combined with a decimal system. Numeral systems are particularly susceptible to the kinds of sociolinguistic changes that arise through language contact; numeral systems of dominant languages often replace the numeral systems of other languages, starting with the higher numerals. In this sense, numeral systems are even more endangered than languages themselves. This process is also at work in AP languages. Alongside the structure of number words, we also investigate how complex number expressions are construed, and the use of ordinals (which appear to be not frequently found in AP languages). Numerical expressions in AP languages may also contain numeral classifiers. An often assumed (and much debated) viewpoint is that languages have numeral classifiers when their nouns are not compulsory marked for number (Greenberg 1972). Numeral classifiers are frequently found throughout the Austronesian languages of Indonesia, but are largely absent in the Papuan languages on the central and eastern parts of New Guinea. Their distribution and function in AP languages varies. Research questions to be addressed include: How are numeral classifiers distributed across the AP languages? How do they function? Which dimensions or concepts do they encode? What is their historical source? Is it possible to reconstruct proto-froms of one or more numeral classifiers? If not, then they are innovative. Could they be the result of contact with Austronesian languages? Other issues pertain to the grammatical properties of numeral classifiers: How can they be distinguished from nouns? What are their distributional properties: do they form a constituent with the numeral, or with the noun?
This project will bring new data in these particular linguistic domains. It will also bear on the question of the linguistic origins and ultimate genetic relationships of the Alor-Pantar languages, the topic studied in the individual project led by Gary Holton at Fairbanks University Alaska. The implications for typology and theory will be investigated by the project led by Prof. Grev Corbett of the Surrey Morphology Group. Project time line: 2009-2012.