The settlement of Island Southeast Asia and the Pacific remains one of the most astounding feats of human history. The large island of New Guinea, lying just north of Australia, was settled some 60,000 years ago by humans migrating from the Southeast Asian mainland. Today New Guinea remains the most linguistically diverse place on earth, home to over 1000 languages comprising more than 20 different families. One of these families, Austronesian, spread far and wide to become the most geographically dispersed language family on Earth. The Austronesians migrated from Taiwan some 6000 years ago, eventually reaching past New Guinea and as far as Hawaii and Easter Island, as well as westward to Madagascar. With two notable exceptions, all of the languages of modern day Indonesia west of New Guinea belong to this large Austronesian family. The language of the Alor and Pantar archipelago in southern Indonesia are one such exception. All but one of the some two dozen languages spoken there do not belong to the Austronesian family. They are either remnants of an earlier population which predates the arrival of the Austronesians, or they are more recent migrants from the New Guinea mainland some 1000 km to the north.

Until very recently almost nothing was known about the Alor-Pantar languages, so judgements about their linguistic relationships remained highly speculative. Through a concerted research effort the NSF project gathered information from 12 of the Alor-Pantar languages representing a distributed geographic sample of the archipelago. Working with fluent speakers of the languages project members recorded diagnostic vocabulary lists and other grammatical information which could be systematically compared to identify sound changes over time. Using the standard Comparative Method technique, the ancestral proto-Alor-Pantar language was reconstructed, providing the first scientific confirmation that the Alor-Pantar languages did indeed descend from a common ancestor. By employing advanced statistical techniques the project was also able to posit the first internal classification, showing how the individual languages are of the Alor-Pantar family are related to each other (see image).

Given that the Alor-Pantar languages are clearly unrelated to any of the Austronesian languages which surround them, the project next tackled the problem of the wider relations of the Alor-Pantar languages. Prior to our work it was widely assumed that the Alor-Pantar languages must be related to the Trans-New Guinea family of languages, but this relationship had never been demonstrated. Armed with new data on the proto-Alor-Pantar language, we found no evidence for this relationship. Instead, the most likely relationship appears to be with the languages of the Bomberai Peninsula in New Guinea. However, those languages are poorly documented, so a thorough evaluation of that proposal will require additional data. The most likely remaining scenario is that Alor-Pantar is not related to mainland New Guinea languages but instead represents a remnant population of what once was a much larger family of related languages in Eastern Indonesia. Those ancestral related languages were replaced by Austronesian languages as speakers of those languages migrated through the region 3000-6000 years ago. The Alor-Pantar family today remains a unique and distinct family of languages in a region teeming with diversity.

This research was the US portion of a larger international collaboration that was conceived under the European Science Foundation, EUROCORES Programme, EuroBABEL (Better Analyses Based on Endangered Languages). The full ESF project was a collaboration of researchers from three countries, including the US, the Netherlands, and the United Kingdom. This collaboration greatly enhanced the efficiency of data collection, allowing relevant data to be shared among project partners.

Supplementary work conducted by the project in 2016 evaluated the potential for ethnobotanical research in the region, focusing on the the Abui language. This pilot research resulted in the documentation of more than 500 Abui plant names, most never before recorded. The PI’s plan to expand this research in the future to include additional Alor-Pantar languages, working in conjunction with a local non-profit agency LENDOLA. Comparing plant terminology among the languages will provide clues as the ancestral homeland of the Alor-Pantar languages.

The results of this project have been widely circulated to a diverse range of audiences at national and international conferences and local workshops. An extended summary of results can be found in the open access volume, The Alor-Pantar Languages: History and Typology. Primary data generated by the project, including audio files and transcriptions, can be accessed via the LAISEANG repository at The Language Archive. These data provide primary documentation of the Alor-Pantar languages and thus have the potential to impact fields beyond linguistics, including archaeology, botany, and history. These data also serve as an enduring record of a unique group of highly endangered languages. 

Last Modified: 04/24/2017
Modified by: Gary Holton