Logo IMG


Rethinking the Fall of Easter Island

New evidence points to an alternative explanation for a civilization's collapse

Terry Hunt

Early Investigations

Figure 2. Often referred to as the most isolated...Click to Enlarge Image

More than 3,000 kilometers of ocean separate Rapa Nui from South America, the nearest continent. The closest habitable island is Pitcairn (settled by the infamous Bounty mutineers in the 18th century), which lies more than 2,000 kilometers to the west. Rapa Nui is small, only about 171 square kilometers, and it lies just south of the tropics, so its climate is somewhat less inviting than many tropical Pacific islands. Strong winds bearing salt spray and wide fluctuations in rainfall can make agriculture difficult.

The flora and fauna of Rapa Nui are limited. Other than chickens and rats, there are few land vertebrates. Many of the species of birds that once inhabited the island are now locally extinct. Large palm trees from the genus Jubaea long covered much of the island, but they, too, eventually disappeared. A recent survey of the island found only 48 different kinds of native plants, including 14 introduced by the Rapanui.

Figure 3. Although it is a small island...Click to Enlarge Image

Accounts by European visitors to Rapa Nui have been used to argue that by the time of European discovery in 1722 the Rapanui were in a state of decline, but the reports are sometimes contradictory. In his log, Dutch explorer Jacob Roggeveen, who led the first Europeans to set foot on Rapa Nui, portrayed the island as impoverished and treeless. After they left, however, Roggeveen and the commanders of his three ships described it as "exceedingly fruitful, producing bananas, potatoes, sugar-cane of remarkable thickness, and many other kinds of the fruits of the earth…. This land, as far as its rich soil and good climate are concerned is such that it might be made into an earthly Paradise, if it were properly worked and cultivated." In his own account of the voyage, one of Roggeveen's commanders later wrote that he had spotted "whole tracts of woodland" in the distance.

A 19th-century European visitor, J. L. Palmer, stated in the Journal of the Royal Geographic Society that he had seen "boles of large trees, Edwardsia, coco palm, and hibiscus." Coconut trees are a recent introduction to the island, so Palmer might have seen the now-extinct Jubaea palm.

Clearly the historical record leaves many gaps to be filled. Scientists have long tried to provide more definitive answers about Easter Island's prehistory, but at times have instead contributed to the confusion.

Norwegian explorer and anthropologist Thor Heyerdahl, for example, visited Rapa Nui in the 1950s and sparked widespread interest in the moai and the large stone foundations, or ahu, on which they were often placed. But he also helped to spread some misleading conclusions. Heyerdahl believed that the Polynesian islands, including Rapa Nui, were settled by voyagers from South America rather than from the western Pacific. In 1947, he launched his famous Kon-Tiki expedition, directing a small craft made of wood and other basic materials from Peru to the Tuamotu Islands to prove that the journey would have been possible for prehistoric peoples.

Figure 4. Thor Heyerdahl...Click to Enlarge Image

In 1955, Heyerdahl led an archaeological expedition on Rapa Nui. He argued that the island had been settled from the east, and he pointed to similarities between the island's statues and South American stonework. Linguistic and genetic evidence have firmly established the Polynesian origin of the Rapanui, but Heyerdahl's conclusions still cloud the archaeological record.

A charcoal sample uncovered on Poike Peninsula—marking, presumably, the site of an ancient hearth—was dated to about 400 A.D. Combined with the then-prevailing idea that the Rapanui language showed many centuries of isolation from other Polynesian groups, the radiocarbon date from this charcoal sample led scholars to conclude that human settlement here began about 400 A.D.

More recently, however, archaeologists have rejected the Poike Peninsula date. Likewise, others have questioned whether the linguistic evidence reflects Rapanui's isolation instead of early settlement. This later phase of research began to point to 800-900 A.D. as the earliest likely date of human colonization.

Although archaeologists have indeed focused a great deal of effort on establishing just when the island was settled, much of their work has been dedicated to studying the changes that these colonists brought about, especially deforestation. Heyerdahl's team took pollen samples that showed that palm trees had once been abundant on the island. In the course of their excavations, members of the expedition also found telltale features where roots had once grown, indicating more widespread vegetation in the past and pointing to the possibility that humans had caused the loss of forest cover.

Flenley has provided much of the more recent detailed evidence in this area. In the late 1970s and 1980s, he collected and analyzed sediment core deposits from three areas: Rano Aroi, a crater near the island's center; Rano Raraku, a crater adjacent to the quarry where many of the statues were carved; and Rano Kau, a crater in the southwest corner of the island. Each of these depressions hosts a shallow lake, which collects wind-blown sediment from elsewhere on the island.

The best evidence came from a 10.5-meter core from Rano Kau, which showed that the island had been forested for tens of thousands of years before the trees disappeared, a process that took place between 800 and 1500 A.D. But more recently Flenley and other scientists have raised doubts about the validity of these dates, which were derived from measurements of the radiocarbon age of lake sediment samples. In 2004, Kevin Butler of Massey University, Christine A. Prior of Rafter Radiocarbon Laboratory and Flenley showed that bulk sediment samples from such locales often contain some carbon that is considerably older than the time of deposition, suggesting that the chronology Flenley first proposed may be hundreds of years too old in dating human-induced forest clearance.

Other recent archaeological and paleo-environmental work has also challenged long-held assumptions about Rapa Nui's prehistory. Catherine Orliac of the Centre national de la recherche scientifique in France has conducted a remarkable study of 32,960 specimens of woods, seeds, fibers and roots. In addition to identifying 14 taxa not previously observed on the island, she showed that the primary source of fuel for the Rapanui changed in a dramatic fashion. Between 1300 and 1650 A.D., inhabitants burned wood from trees, but they used grass, ferns and other similar plants for fuel after that point. Orliac also argued, however, that at least 10 taxa of forest vegetation may have persisted until Europeans began visiting the island.

In another study, Orliac examined the remains of the hard shells surrounding the seeds of the Jubaea palm. Those samples that were carbonized, gnawed by rats or found in association with human materials provided evidence of human habitation on the island. She dated a number of such remains and found that they all fell after 1250 A.D.

Ecologists Andreas Mieth and Hans-Rudolf Bork of Christian-Albrechts-University Kiel in Germany have studied the process of deforestation on Rapa Nui. Using a variety of evidence, primarily from Poike Peninsula, they concluded that Jubaea palms once covered most of the island. Around 1280 A.D., they argued, deforestation began. The Rapanui largely abandoned the peninsula within 200 years, but then they resettled it in some areas from about 1500 A.D. to 1675 A.D.

In 2003, geologist Dan Mann and several colleagues obtained radiocarbon dates not from bulk samples, but from bits of charcoal found in soils from a number of locations around the island. They also documented ancient episodes of severe erosion, which according to their radiocarbon measurement began soon after 1200 A.D. Their study, like that of Mieth and Bork, pointed to deforestation taking place between 1200 and 1650 A.D., with no sign of human impact prior to that period.

Both Mann's team and Mieth and Bork reconciled their findings with earlier work by arguing that the population during the centuries prior to 1200 A.D. must have been small or transient. It was only once the number of permanent inhabitants grew larger that indications of human presence would have become clear in the paleoenvironmental record.

But this scenario makes several questionable assumptions. It requires a small founding population with a slow rate of growth and little ecological impact. After conducting our own research on Rapa Nui, we began to wonder whether the lack of evidence of human presence prior to about 1200 A.D. should be taken at face value—maybe the island hadn't actually been settled as early as most people assumed.

» Post Comment



Of Possible Interest

Spotlight: Briefings

Letters to the Editors: Only Human

Marginalia: Why Is Human Childbirth So Painful?


Related Internet Resources

Terry L. Hunt's homepage

Article by Hunt in Science magazine

Easter Island Foundation

NOVA special on Easter Island


Foreign-Language PDFs


Subscribe to American Scientist