Mouth of the Sigatoka River, Viti Levu, Fiji
The Sigatoka dunes
About 2600 to 1300 years ago, in phases (a three-level model that was challenged)
Sand west of a Fijian river mouth keeps giving up pots and burials, and then covering them again. The early ceramics are late Lapita, buried in a hurry about 2,600 years ago. Whether the dune’s history is three quiet phases or one long scramble is the argument.
Why this is extraordinary
West of the mouth of the Sigatoka, the second largest river on Viti Levu, a field of parabolic dunes covers 650 hectares. The tentative World Heritage listing says about half of that field is unstable, especially on the east. Pottery, stone tools, and human remains keep coming out of the sand because the wind and the river put them there and then uncover them. The National Trust records a burial ground found in the late 1940s with more than 50 people, dated to about 2,600 years ago and read as Lapita, and a later set of burials about 1,650 to 1,500 years ago. The same park has produced the largest collection of complete and nearly complete Lapita pots in the Pacific. Many of those pots are in the Fiji Museum in Suva. They are not in these photographs.
For more than thirty years the orthodox sequence was three phases of occupation, each tied to a time when the dune stood still. That sequence was the standard model of Fijian prehistory. A challenge then said the sand does not support three separate levels. Occupation, on that reading, was broken, complicated, and continuous inside a dune that never sat quietly. New optical dates on the sand and radiocarbon dates were published to answer that challenge. Optically stimulated luminescence, the optical method, is a light-clock in grains of sand. It resets when the grains last saw the sun, and it runs again once they are buried and dark. Radiocarbon, used beside it, dates once-living material. Together they keep a three-phase model, more complicated in the detail. The longest stable phase, Level 2, begins 2,500 to 2,300 calibrated years before present. Substantial dunes pile up after about 1,300 calibrated years before present.
A separate excavation of an intact patch of the early ceramic layer found the pots had been laid down and buried quickly on a back-beach flat, in a short interval around 2,500 radiocarbon years before present, about 2,600 calibrated. Few houses, few other artefacts, little bone. The reading offered for that patch is a canoe landing, not a village. The decoration on the pots is limited. One reading calls that a decline from the dense Lapita patterns of the first colonists toward plainware. Another calls it a working industry with a job to do, not a fall from grace. No azimuth of a shrine or a dune crest is published. Lapita sailors reached Fiji by wind and stars, as later Oceanic navigators did. You hold a course by keeping a star's rising place off the bow, and by the wind on the sail. This river mouth has no printed bearing, and none is published for the landing.
Why it matters
If three stable land surfaces hold, Sigatoka is the spine of Fijian prehistory. If the sand was always moving, the phases are an artefact of where the excavators stopped, and the first settlement of this valley is older or simply lost.
The photographs are the dune. The pots are in the museum. The pin is the river mouth.
Photo tour
2 photographs, unedited. On a phone, swipe the row. Choose one to look closely.
.jpg){kind=link}
- The Lapita pots and the burials are not in either photograph. An open face-sherd from Fiji was not used, because the file does not say it came from Sigatoka.
What is still unexplained
These sit beside the consensus account, with the same visual weight. None of them is marked solved.
Location and dating
- Location
- Sigatoka Sand Dunes, immediately west of the river mouth on the south coast of Viti Levu. The pin is 18.17 south, 177.489 east. The park is 650 hectares.
- Period label
- Current consensus — open to revision Late Lapita ceramics around 2,600 calibrated years before present. Level 2 stability from 2,500 to 2,300 calibrated years. Later burials about 1,650 to 1,500 years ago in the Trust’s account. Dune accumulation after about 1,300 calibrated years.
- How this dating was done
- Radiocarbon on the occupation, and optical dating of the sand grains, a light clock that resets when a grain is buried and dark. The phases are stratigraphy in a dune, which is the disputed part.
- What could change it
- A date from under Level 1, if any older human layer exists, would show whether Lapita here is the first landing or only the oldest one that survived. The valley survey looked for that older trace and treated it as unproven.
What is actually there
- The dune field is 650 hectares, about half of it unstable.
- A late-1940s burial ground held more than 50 people, put at about 2,600 years ago.
- Later burials are given as about 1,650 to 1,500 years ago.
- Level 2 begins 2,500 to 2,300 calibrated years before present.
- No star bearing is published for the river mouth.
Open questions
| Question | Status |
|---|---|
| Do three stable surfaces exist, or did the dune stay in motion? | Open. The optical dates defend three phases. The challenge said the sections do not. |
| Was the early pot scatter a landing, a camp, or a village the sand destroyed? | Open. Few houses were found. Sand removes houses. |
The current narrative
Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.
The current narrative — open to revision
Three occupations on stable dune surfaces
People used the river mouth in three phases, each while the sand was quiet, from late Lapita onward. The optical dates keep that model.
What supports this
- Optical and radiocarbon dates were gathered to test the challenge, and the three-phase reading remained the most robust.
- Level 2 has a start, 2,500 to 2,300 calibrated years before present, and dune-building picks up after about 1,300.
- The early pots form a temporally tight assemblage, which a single stable surface explains.
What pushes back
- The challenge said the sections do not show three discrete levels.
- Dunes move. A stable surface can be a pause in one stormy history, not a cultural period.
- The reply itself calls the detail more complicated than the textbook version.
What would settle it
It is the current narrative because the dates were collected against the objection. The objection is why the room stays open.
Emerging narratives
Other published readings. Each may be right. None is adopted here unless the evidence carries it.
An emerging narrative
A late Lapita canoe landing, not a village sequence
The early layer is a short-lived landing on a back beach. The later “phases” are whatever the moving sand preserved, not a planned sequence of cultures.
What supports this
- Few habitation features, few non-ceramic artefacts, and little fauna in the intact patch.
- Pots were buried rapidly, in a brief interval, which fits a visit more than a town.
- The shoreline of a river mouth is where a canoe comes in.
What pushes back
- More than 50 burials, and the largest set of restorable Lapita pots in the Pacific, are a lot of material for a landing.
- Later burials, 1,650 to 1,500 years ago, show the place was used again in a different century.
- A landing does not, by itself, dissolve the optical dates for Level 2.
What would settle it
It explains one patch of pots. It does not yet explain the whole 650 hectares.
Should the current view be reevaluated?
Open to revision
Late Lapita pottery at the river mouth, quickly buried about 2,600 years ago, holds. Three cultural phases hold only as the best current reading of a dune that another study called continuous. The landing reading and the decline-of-decoration reading are both still on the table.
What drives this
- A thirty-year three-phase model and a published challenge to it.
- Optical dates that keep the phases and complicate them.
- A pot assemblage with two meanings, decline or a working industry.
This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.
Weigh it yourself
Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.
Visitor lean so far — not evidence. On this browser only; a shared tally is not collected.
If the contested reading is right
What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.
The three phases are real
Fijian prehistory keeps a spine at this river mouth, from late Lapita to the centuries when the big dunes grew.
Why it is still worth a test
A section in which all three surfaces, and the dates on them, can be seen in one face.
The early layer is only a landing
The famous pots are cargo or a camp, and the village of the first Sigatoka people is somewhere else in the valley, or was never built.
Why it is still worth a test
House floors upriver, dated to the same centuries as the back-beach pots.
Sources
- Sigatoka Sand Dunes, tentative list UNESCO
- Sigatoka Sand Dunes, archaeology and Lapita National Trust of Fiji
- Times of sand, sedimentary history and archaeology at the Sigatoka Dunes Geoarchaeology
- Late Lapita occupation and its ceramic assemblage at the Sigatoka Sand Dune site Archaeology in Oceania
Image credits
- The dune field west of the Sigatoka River mouth. Footprints are modern. No pot or burial is in the frame. The park covers 650 hectares, and about half the dunes are unstable. . Radosław Botev, CC BY 3.0. Shown unedited.
- The same park, looking toward the sea through the vegetated dunes. The river mouth is the reason the sand is here. Pots and bones are in the unstable faces, not in this view. . Radosław Botev, CC BY 3.0. Shown unedited.