A Hawaiian linguistics professor believes eastern Polynesian ancestors, including Maori, began their colonisation of the Pacific from remote atolls near the Solomon Islands, not Samoa as has long been believed.
It is from these coral outcroppings, which barely break the Pacific Ocean and sustain tiny populations, that the original homeland of Pacific peoples, Hawaiki, may be located.
Professor William Wilson has been a key figure in the revitalisation of the Hawaiian language movement. Bubbling away in the background of that work over the past 30 years has been a desire to look more closely at the languages of atolls such as Takuu and Luangiua - known as Polynesian outliers because they are inside Melanesia and outside the Polynesian triangle which is bounded by New Zealand, Hawaii and Rapa Nui/Easter Island.
In a paper published in December's edition of Oceanic Linguistics, Professor Wilson argues that while anthropologists and linguists have assumed East Polynesia, including New Zealand, was settled from Central Western Polynesia, most likely Samoa, his study suggests otherwise.
The paper details 73 lexical and grammatical structures that are shared by the outlier and eastern Polynesian languages but not by Samoan or any other western Polynesian languages.
He believes outlier populations, sophisticated navigators, voyaged to Samoa and back to the outlier atolls. After a time - it is not clear how long - the language evolved and it was from the atolls that the ancestors of Maori and others eventually set out.
Speaking from Hawaii yesterday he said other evidence for the theory, such as sophisticated fishing techniques for deepsea expeditions, was shared across the outlier and eastern cultures.
Similarly, carving traditions here, in Hawaii and throughout the east do not exist in the west but do in the outlier atolls.
Hawaiki might well have been located in the outliers, Professor Wilson said, which would mean those seagoing people travelled vast distances to Aotearoa.
Otago University archaeologist Professor Richard Walter said it sounded plausible and should be tested archaeologically.