It is no longer reasonable to argue that evidence of earlier settlements is “missing” or archaeologically invisible through sampling or taphonomic problems [Discussion in (4)], or that particular radiocarbon dates upon specifically unidentified samples, or samples with weak stratigraphic connections to cultural remains make a case for earlier ages of colonization (9, 10).
During the last prehistoric expansion of modern humans, Polynesians from the Samoa-Tonga area dispersed through more than 500 remote, subtropical to subantarctic islands of East Polynesia (a cultural region encompassing the islands of New Zealand, Chathams, Auckland, Norfolk, Kermadecs, Societies, Cooks, Australs, Gambier, Tuamotu, Marquesas, Line, Rapa Nui, and Hawaii), an oceanic region the size of North America (Fig. The timing and sequence of this expansion, debated vigorously since Europeans rediscovered the islands of East Polynesia (1, 2) and most intensively with the advent of radiocarbon dating (3, 4), remains unresolved. 600–950 in the central, northern, and eastern archipelagos, and no earlier than A. Subsequent studies using precise AMS dating of short-lived materials alone have generally supported short chronologies (4, 6–8).
On many islands, irreconcilable long and short settlement chronologies coexist that vary by more than 400–1,000 y (4). The last systematic analysis of radiocarbon dates from archaeological and paleoecological sites throughout East Polynesia, published 17 y ago, was based on 147 radiocarbon dates (5). However, these chronologies continue to be dismissed by some scholars (9, 10) on hypothetical grounds of missing evidence or archaeological invisibility, and in favor of radiocarbon dates on materials (typically unidentified charcoal with high inbuilt age potential) incapable of providing a precise age for the event being dated.
Here accuracy is defined based on those samples that can provide a date that is the “true” age of the sample within the statistical limits of the date.
Precision is controlled by small laboratory measurement and calibration errors.
More radiocarbon dating of short-lived materials from islands lacking enough Class 1 dates for robust chronologies (Gambier, Tuamotus, Australs, Northern Cooks, Kermadec, Norfolk, and Chathams; Fig. In addition, closer scrutiny of dates at the older end of the Class 1 age ranges may also increase the precision of estimates for initial colonization.
For example, some of the oldest dates for the Auckland Islands are based on small-diameter (2-cm) wood from long-lived trees (), which, despite the size of twigs, may still contain inbuilt age and create an artificial tail to the probability distributions (19).That is, to accept only those dates that () are capable of providing a calibration that is close to the “true” age of the actual target event (i.e., human activity).One approach is to evaluate dates within their individual and comparative stratigraphic levels according to criteria of “chronometric hygiene” (11, 12) and build from those results toward a regional overview; but this method can be subjective, and it is impractical when dealing with very large databases, as is the case here.These conflicting chronologies preclude establishment of a regional pattern of settlement and hinder our understanding of cultural change and ecological impacts on these island ecosystems. Islands of East Polynesia, summarizing the two phases of migration out of West Polynesia (blue shading): first to the Society Islands (and possibly as far as Gambier) between A. ∼10 (orange shading), and second to the remote islands between A. It used a “chronometric hygiene” protocol to exclude dates with high uncertainty and to provide a chronology that proposed initial settlement A. Conflicting estimates for initial colonization in East Polynesia create great uncertainty about the historical framework within which human mobility and colonization, variations in human biology and demography, and the rates and types of human-induced ecological impacts to island ecosystems must be explained.Islands of East Polynesia, summarizing the two phases of migration out of West Polynesia (blue shading): first to the Society Islands (and possibly as far as Gambier) between A. ∼10 (orange shading), and second to the remote islands between A. As the number of radiocarbon dates from East Polynesia has increased 10-fold over those available in 1993 (5), an attempt to resolve the frustrating problem of colonization chronology for the region is now opportune.We first categorized all radiocarbon-dated materials into one of six sample material types: short-lived plant, long-lived plant, unidentified charcoal, terrestrial bird eggshell, bone, and marine shell (Fig. Dates on these materials were then sorted into reliability classes, according to whether there was potential for any disparity between the age of the radiocarbon event (i.e., Fig. Calibration probabilities were then calculated for the subset of reliable dates to derive the most precise (within radiocarbon calibration error) estimate for the age of initial colonization on all East Polynesian island groups (Chronometric range (68% probability) of calibrated radiocarbon dates for East Polynesian islands, for reliability Classes 1–3 as defined in Materials and Methods.