Why did the Polynesians Stop Sailing For Over 1000 Years?

Ancient rainfall records suggest climate helped set the timing of one of history’s great migrations.

Around 3,500 BCE, people began voyaging out across the western Pacific Ocean. They hopped from one island to the next, settling, building villages, establishing unique cultures, and moving on to the next island on the distant horizon. They used the stars, currents, and winds to guide them. Eventually, their descendants, a seafaring culture known as the Lapita, reached the Western Polynesian islands of Samoa and Tonga, but then, after hundreds of years of expansion and exploration, they stopped.

This sudden period when long-distance sea voyages eastward ceased is called the ‘Long Pause,’ and it’s a bit of a mystery why it began and why it ended. But a group of researchers led by Dr. David Sear may have an answer, pointing to climate records of past rainfall to explain why people stayed, and why they chose to sail the open seas again after more than a thousand years.

The Voyager Period

The settlement of the tropical South Pacific was one of humanity’s most remarkable feats.

For over 4,500 years, from 3,500 BCE to 1250 CE, people crossed ever longer stretches of ocean to reach ever smaller islands, eventually culminating in the colonization of some of the furthest-flung islands, including Rapa Nui and Hawai’i.

The traditional archaeological tale goes as follows: The Lapita culture had been expanding west-to-east until they reached the Polynesian islands of Samoa and Tonga around 1000–800 BCE. After a period of inactivity, their descendants fanned out to settle the remaining islands of the eastern Pacific, including Hawai’i and New Zealand, in a burst of expeditions that lasted less than 300 years.

But over the past decade, that neat story has been somewhat complicated by genetic and linguistic studies showing that Micronesia (northwest of Polynesia) was settled west to east, and that eastern Polynesia may also have been settled partly via Micronesia. What emerges is not a single neat chain of migrations but a complex web of repeated voyages going back and forth, west, east, south, and north, though Samoa and Tonga still seem to have been the major departure points for voyages into eastern Polynesia.

But why people set out, and when, rather than sooner or later, has also been a much-debated question. Globally, island colonization has myriad reasons, from resource and population pressures to technological developments, challenges posed by geographical distance, and climate change.

The Long Pause

One of the puzzles in this story is the Long Pause. After reaching Samoa and Tonga, roughly 1,700 years passed before people appeared on the islands further east. Some explanations have been proposed, including a lack of suitable technology to successfully make this journey farther eastward, the time it took for the islands to become overcrowded and resource-scarce, and environmental change that made living on the current islands unsustainable.

In terms of environmental change, most previous studies relied on climate records from the northern hemisphere, suggesting a link between the Medieval Climate Anomaly to the Little Ice Age as drivers of the sudden restart of voyages. But climates are complex, and drivers and causes in the Northern hemisphere are often felt very differently in the Pacific, a fact that has quietly been largely ignored. As a result, climate has been quietly dropped from recent overviews of Pacific archaeology.

But climate disasters, especially droughts, have been known to cause devastation on island communities, and thus, Dr. Sear and his colleagues propose that a new look at what happened during the Long Pause is needed. The problem… climate records for this period in the Pacific are often patchy and coarse. So, the researchers decided to go looking for their own records.

Sailing Again

Using climate records from lake and swamp sediments across the study region, chemical signatures in ancient plant and algae remains, and thousands of simulations to account for dating and measurement uncertainty, the researchers were able to reconstruct past rainfall regimes.

What they noticed was something almost too perfect. During the centuries in which people began pushing east, the tropical Pacific shifted into a persistently El Niño-like state, and the southwest Pacific, including Samoa, Tonga, and Vanuatu, dried out, losing as much as a third to a half of its normal annual rainfall in the worst drying event of the past 1,500 years, while the islands further east grew wetter and more stable. 

Exactly what caused this shift is unclear, but it did not just affect rainfall; it also affected the strength of the trade winds, making the long journeys east and northeast less difficult (though whether this shift really made a big difference is uncertain since recent simulations suggest the canoes of the day may not have needed the help. The debate is unlikely to be settled until a voyaging canoe from the era is found, and the effects of different wind strengths and directions can be tested.)

Climate alone cannot explain the sudden restart of voyaging; after all, people deal with droughts all the time, and don’t always resort to packing up and leaving as a result. Drought only becomes a problem needing relocation when coupled with water and crop shortages, population crises, and time.

It is, more or less, the setup of Moana in miniature: an island whose crops and fishing are failing, and someone deciding to look beyond the reef to solve it.

The researchers’ models showed short dry spells were survivable; they were part of the ordinary rhythm of life, but long, intense droughts were problematic. The more crowded an island was, the worse the effects, and beyond roughly a hundred people per square kilometer, severe drought began to outstrip what an island could provide. 

Island density, though, is misleading; you can have a massive island with only a few thousand people, suggesting low population density, but people do not inhabit islands equally; they cluster where there is fresh water and fertile land. When that reliable land on lagoon edges, in island valleys, and along coastlines dries up, population pressures increase.

So the Long Pause may have ended not because people suddenly could sail east, but because they suddenly had reason to. For over 1000 years, the islands they had settled were sufficiently stable, and climate change and population pressures were short-term and manageable. It may have been only when Samoa, Tonga, and other western Polynesian islands became too dry, for too long, with too many people to feed and care for, that they voyaged east.

New research suggests that the Long Pause in Polynesian sea voyaging ended when drought helped push island inhabitants to sail eastward in search of wetter, more stable islands. Coincidence in time is not proof of cause, and the researchers are careful to say so. But the drought was real, the crowding was real, and together they offer a compelling reason why, after over 1000 years of settled life, people once more took to the seas.

You may wonder how they knew these islands would be wetter, more stable? Well, I do not actually know, but the researchers point out that the islands of Atiu and Mangaia in the southern Cooks had first been sighted as early as 350–600 CE, but were only inhabited some 250 to 350 years later (850–950 CE). It is possible that the people of Polynesia knew of these islands further east, having seen or traveled there in generations past, but having no compelling reason to settle there.

Do you think the timing of El Niño and the end of the Long Pause was a coincidence? And what other reasons may have contributed to its end?

Let me know your thoughts, and if you’d like to support the countless hours I pour into researching, writing, and bringing these hidden corners of history to life, why not Buy Me A Coffee

Reference

David Sear et al, Did changing climate in the tropical South Pacific contribute to the eastward migration and settlement of Polynesia?, Journal of Pacific Archaeology (2026). DOI: 10.70460/jpa.v16i2.399