The Pauwasi languages are a likely family of Papuan languages, mostly in Indonesia. The subfamilies are at best only distantly related. The best described Pauwasi language is Karkar, across the border in Papua New Guinea. They are spoken around the headwaters of the Pauwasi River in the Indonesian-PNG border region. Based on earlier work, the East and West Pauwasi languages of Indonesia were classified together in Wurm, though he did not recognize that Yuri of Papua New Guinea was also East Pauwasi. That connection was made by Usher, though anthropologists had long known of the connection. Later the South Pauwasi languages were also identified by Usher, and the West Pauwasi family tentatively expanded. Wichmann, Foley and Pawley & Hammarström, noting the sharp differences between the three groups, are agnostic about whether West Pauwasi, East Pauwasi and South Pauwasi are related.
The inclusion of Molof is especially tentative. The languages are not close: though the Eastern languages are clearly related, Yafi and Emumu are only 25% lexically similar. Pawley and Hammarström also question whether Eastern Pauwasi and Western Pauwasi are really related. They also note that Tebi and Towei are very different from each other, and may not necessarily group with each other. Karkar-Yuri, long thought to be an isolate in Papua New Guinea, is clearly related and may actually form a dialect continuum with Emumu in Indonesia. On the other hand, the Western languages are so poorly attested that it is not certain that they are part of the Pauwasi family, or if the common words are loans and they constitute a separate family or families, though a family connection appears likely. The proto-forms of the pronouns have not been reconstructed. Attested forms include: Yafi and Emumu are similar, and Dubu and Towei may share 1pl *numu, but there is not apparent connection between them. Dubu no and Yafi nam might reflect pTNG *na, and Towei ngo pTNG *ga, and the plural pTNG *nu and *ni.
Classification
classified the Western languages as a branch of the Trans–New Guineaphylum, a position which Malcolm Ross tentatively retained. Ross's TNG classification is based on personal pronouns. Since no pronouns could be reconstructed from the available data on the poorly attested Indonesian Pauwasi languages, which were all that were recognized as Pauwasi at the time, only a tentative assessment could be made, based on a few lexical items. Some of the pronouns of Dubu and Yafi look like they might be TNG. However, Ross counted Karkar, for which the pronouns were known, as an isolate because its pronouns did not pattern as TNG. At this stage its identity as a Pauwasi language was unknown. Pawley and Hammarström do not consider there to be sufficient evidence for the Pauwasi languages to be classified as part of Trans-New Guinea, though they do note the following lexical resemblances between Tebi, Yafi, and proto-Trans-New Guinea.
