KEYWORDS: Animal introductions, extinction, faunal analysis, Polynesia, Mangareva
Despite images of paradise, eastern Polynesian islands were devoid of most economically-useful animals and cultigens when colonised by humans about 1500 years ago; other views on chronology are more conservative. We analyse the faunal material from the 1959 excavations by Green in Mangareva, southeast Polynesia where 13,598 bones (NISP) were identified to nearest taxon. We discuss the significance of the purposely introduced chicken or Pacific Jungle Fowl (Gallus gallus), dog (Canis familiaris) and pig (Sus scrofa) not known from the historic records of that island group as well as the Pacific rat (Rattus exulans) and human dental and skeletal material. The dog remains now represent the most eastern limit of this species known prehistorically for Oceania. Many of the pig remains were associated with a marae (religious edifice) confirming the importance of this animal in its ceremonies. The majority of fractured human bones and teeth were recovered from midden contexts, thus alluding to the possibility of cannibalism as reported in late prehistoric oral traditions. We suggest that rats, and not human predation, were responsible for the early local extinction of the chicken in the prehistoric sequence for Mangareva.
Chicken, pigs, dogs, rats and, of course, people constitute the five main terrestrial animals transported in prehistory across Remote Oceania. On many islands, all five were present from the earliest times but, in some cases, one or more of these taxa went extinct during late prehistory. Andrew Sharp (1964:94-97) used the presence or absence of the pig, dog and chicken to bolster some of his arguments for accidental voyaging, but he drew mostly on the European contact ethnographic evidence with only limited data from the archaeological record. More recently, archaeology has considerably altered the record of humanly-introduced animals for most of the island groups which he discussed. Moreover, it can now be shown that the transport of various animals to numerous islands of Near Oceania (islands west of the eastern Solomons) goes back to the Pleistocene with different species of rats being transported even more widely (Flannery 1995; Roberts 1991; Spriggs 1997:53-55; White et al. 2000). The Remote Oceanic pattern of introduction and subsequent extinction is a continuation of an ancient process that began in Near Oceania. In this paper we describe the Mangarevan example of the early transport and then local extinction of the chicken, pig and dog by the time of first contact with Europeans in the early 19th century as deduced from an analysis of the faunal material. We also show how this evidence may be interpreted within the eastern Polynesian context.
The fauna we discuss come principally from three sites (GK-1, -2 and -3) on the small islet of Kamaka at the southern margin of the large Mangarevan lagoon excavated by Roger Green in 1959. The details of their excavation, stratigraphy, contexts and dating of the sequence are discussed elsewhere (Green and Weisler 2000, 2002). In our view the sites represent the mid to late end of an 800 year sequence extending from 1200 AD to the 1840s. Dating the beginning of human habitation of the Mangarevan group has been undertaken recently and will be presented elsewhere. From extensive research on islands further east (Weisler 1995, 1997, 1998a) we estimate that 200 to 400 years is missing from the earliest portion of the Mangareva sequence.
The faunal sample is a well-preserved collection of 13,598 bones and midden. The molluscs were processed and analysed in the field. Numerous boxes of bone went missing after 1960 from the American Museum of Natural History, but were later noticed in the basement of the Washington State Burke Memorial Museum around 1995 where they were identified by Lisa Nagaoka as coming from excavations in Mangareva by Roger Green. …