Academic journal article Archaeology in Oceania

Radiocarbon Dates from the Ulong Site in Palau and Implications for Western Micronesian Prehistory

Academic journal article Archaeology in Oceania

Radiocarbon Dates from the Ulong Site in Palau and Implications for Western Micronesian Prehistory

Article excerpt


Recent excavations on Ulong Island in Palau recovered an early pottery assemblage that has not been recorded previously in the archipelago. Four AMS results on charcoal fragments and carbonised residues adhering to pot sherds ranged from 3150 [+ or -] 90 BP to 4330 [+ or -] 90 BE and were consistent with palaeoenvironmental work which suggests that human arrival in western Micronesia might have occurred as early as 4500 BE Redating of the Ulong samples that gave the earliest results put occupation instead around 3000-2650 BE and an error during laboratory processing of the oldest AMS samples is suspected. Revision of the Ulong dates invites reconsideration of other archaeological radiocarbon ages from western Micronesia where convincing evidence for occupation deposits older than c. 3300 cal BP has yet to be obtained.


Western Micronesia consists of a thinly distributed arc of islands (Figure 1) that is often considered to have been occupied during an early phase of Austronesian expansion from island Southeast Asia into the Pacific (Spoehr 1957: 18; Shutler 1971: 19).


Permanent occupation of Palau and other parts of western Micronesia was hypothesised by Osborne (1966: 464) to have occurred at 3750-3250 BE with the possibility that inadvertent voyages from Halmahera or New Guinea could have brought people to Palau as early as 5000-4500 BE Masse (1990a: 223-225) critically reviewed radiocarbon dates from archaeological sites to argue that Palau was colonised much later, at 2000 BP, from populations in the Indo-Malaysian Archipelago, well after human settlement of Yap and the Marianas. Similarly, Intoh (1997: Fig. 6) has suggested the Marianas were occupied first, at 3600 BP, while Yap and Palau were not colonised until 2000 BE from the Philippines or Indonesia. Recent dates on human burials and deposits containing a thin blackware in Palau provide reasonable archaeological evidence for occupation around 3000-2400 BP and possibly as early as 3400 BP (Fitzpatrick 2002, in press; Reith and Liston 2001; Wickler 2001: 187).

Wickler (2001: 189) has cautiously used palaeoenvironmental data suggesting colonisation of the Marianas at 4800 BP and Palau at 4500 BP to posit a "rapid, long distance "star burst" dispersal of Austronesia horticulturalists into western Micronesia from island Southeast Asia ...', a pattern described as similar to the dispersal of the Lapita cultural complex. However, the independent colonisation of Palau and the Marianas suggested by Wickler, and the absence of any further expansion (on current knowledge), particularly to volcanic islands such as Chuuk, Pohnpei and Kosrae lying to the east, appears unlike the Lapita dispersal in pattern, scale or tempo (e.g. Anderson 2001; Ishimura 2002). The main phase of Austronesian expansion in island Southeast Asia is now placed by some researchers at 4000-3000 BP (Bellwood 2000: 354; Spriggs 1999) rather than 5000-4000 BP as was earlier thought (e.g. Bellwood 1992: 105; Kirch 1997), suggesting that if western Micronesia was colonised in the third millennium BC it was from a discreet and minor pulse of movement from the Philippines or Taiwan, with characteristics different from later Austronesian expansion through island Southeast Asia and the west Pacific.

The current position, then, is that archaeological evidence for human colonisation throughout western Micronesia substantially post-dates the earliest plausible palaeoenvironmental indicators of anthropogenic activity and the main phase of Austronesian expansion. Palaeoenvironmental indicators of environmental change that might indicate early human occupation include microscopic charcoal particles, grass pollen, Pandanus pollen and four types of fern spores (Lycopodium cernuum, Gleichenia linearis, Psilate monolete, Psilate trilete). In four cores the frequency of charcoal and some of the disturbance pollen types increases significantly around 4200 cal BP, while in two cores from Ngerekebesang Island disturbance indicators occur at 5350 cal BP and 5750 cal BP, based on linear interpolation of calibrated radiocarbon dates (Athens and Ward 2001). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.