Ahu Tongariki · Easter Island · -27.12584, -109.27664 · PYR record in the living atlas
Fifteen moai stand on Ahu Tongariki with their backs to the Pacific, on a platform that runs close to 100 metres and, counting its lateral wings, around 200 metres end to end. The largest of them stands about nine metres. The heaviest is around 86 tonnes, which makes it the heaviest statue anyone on Rapa Nui is known to have moved and set upright on an ahu, and the tallest and the heaviest are not necessarily the same figure. Everything you are looking at was picked up off the ground within living memory.
They walked, and how much weight that sentence carries
Rapa Nui oral tradition always held that the moai walked from the quarry. Outsiders spent most of a century proposing alternatives: log rollers, wooden sledges, horizontal frames, and at the sillier end of the market, visitors from elsewhere. In 1986 the Czech engineer Pavel Pavel showed on the island that a standing statue could be swivelled forward with ropes. The full case took another four decades to assemble.
It arrived in the Journal of Archaeological Science in 2025, in a paper by Carl Lipo and Terry Hunt titled The Walking Moai Hypothesis: Archaeological Evidence, Experimental Validation, and Response to Critics. The argument runs through the statues themselves. Lipo and Hunt worked from a survey of 962 moai and singled out the 62 left along the island's roads, still in transit when carving stopped. Those road statues are not miniature versions of the finished ones. They have wide D-shaped bases, they lean forward, and their eye sockets are uncut. A forward-leaning figure on a curved base is unstable standing still and hopeless lying down on a sledge, which is why the shape had never made sense. Standing, it behaves like an inverted pendulum: rock it sideways and the lean converts the rocking into a step.
The team then built a replica to road-moai proportions weighing 4.35 tonnes and walked it. Eighteen people moved it 100 metres in 40 minutes, pulling in alternating tugs from either side. Lipo's description of the experience is that once the statue is in motion it is barely hard work, with people hauling one-handed. The roads back him up. They are about 4.5 metres wide and cut with a concave cross-section, a shallow trough rather than a flat track, which is exactly the surface that keeps a rocking statue from wandering off line. Nobody builds a dished road for a sledge.
That is a strong result, and it is worth being precise about what it is. The experiment shows that the rocking method described in Rapa Nui oral tradition is physically workable. It does not show that this is how the statues were in fact moved, and it does not show how the method scaled. The replica weighed 4.35 tonnes and travelled 100 metres. The heaviest moai on this platform weighs about 86 tonnes and came roughly 1.4 kilometres from the quarry: twenty times the mass over fourteen times the distance. Transport on Rapa Nui is still an open question, several competing models have been proposed and tested, and the walking hypothesis is now the best supported of them rather than the closed answer.
What can be said without hedging is that an account the islanders gave and outsiders filed as folklore has turned out to describe workable engineering. That is worth more than it usually gets.
The quarry that stopped mid-sentence
About 1.4 kilometres inland from Tongariki, by the coordinates in our own catalogue, is Rano Raraku, the volcanic tuff quarry that supplied the stone for roughly 95 per cent of the island's known moai. Some 887 statues remain there, at every stage from a face still attached to the bedrock to a finished figure buried to the shoulders in slope wash. The largest of them was never separated from the rock: it is about 21.6 metres long and estimated at around 270 tonnes, nearly twice the length of any moai ever completed. It is worth holding those two numbers together. The islanders successfully moved 86 tonnes across country and set it on a platform. They also, at some point, started something they could not finish.
That the Tongariki statues came from Rano Raraku is not an assumption anyone has been asked to take on trust. Petrographic thin-section analysis by Jo Anne Van Tilburg and colleagues, published in 2008, matched the stone of statues from this platform directly to the quarry's tuff.
What the platform is made of, and what is inside it
The ahu is engineering as much as the statues are. It stands on a deep pavement of tightly packed vesicular basalt boulders laid over a levelled surface of volcanic clay, with the facade faced in dressed basalt slabs and the interior filled with rubble.
Some of that rubble used to be statues. The excavations during the 1990s restoration found the remains of at least seventeen earlier moai reused as constructive fill inside the older core of the platform, a practice documented at other ahu on the island. Tongariki is not a single structure. It was built, enlarged and remade repeatedly across centuries, and estimates suggest something like 30 moai stood on it over its different phases. The fifteen you photograph stand on the broken remains of their predecessors.
Radiocarbon dates on associated organic material, including human skeletal remains and corrected for the marine reservoir effect, put construction and use at roughly AD 1000 to 1500. That is the platform itself, and it starts earlier than the island-wide moai era of the thirteenth to sixteenth centuries that gets quoted for Rapa Nui in general. The boundaries between individual building phases are not established, and may never be. The 1960 tsunami destroyed the stratigraphy that would have settled them, and the excavation's chronological data has not been published in full.
The setting is busier than it looks. Hanga Nui, the bay around the ahu, holds among the highest densities of prehistoric sites anywhere on Rapa Nui, surveyed along a transect from 1979 onward under Claudio Cristino and Patricia Vargas Casanova. Our own catalogue holds almost none of them, which is a fact about our coverage and not about the bay.
Put back by crane, and by memory
The fifteen did not fall in an earthquake. Island conflict had already toppled the statues face down long before Europeans arrived. Katherine Routledge recorded all fifteen lying fallen in 1914 and 1915, and the platform was left largely undisturbed between roughly 1870 and 1960, degrading further under twentieth-century ranching. Thor Heyerdahl's Norwegian Expedition of 1955 and 1956 mapped and photographed the ahu in that ruined but still pre-tsunami condition, and that record became the baseline for everything that followed.
Then on 22 May 1960 the magnitude 9.5 Valdivia earthquake off Chile, the largest ever instrumentally recorded, sent a tsunami across the Pacific. It destroyed the platform down to its base and threw the fallen moai more than 150 metres inland, chipping and abrading their faces on the way.
The restoration ran from 1992 to 1996 under the Chilean archaeologist Claudio Cristino, after a 1988 televised appeal by the island's governor, Sergio Rapu, for heavy equipment. The Japanese crane manufacturer Tadano donated a crane and the technical staff to run it, and the Japanese government contributed funding. The work was not simply a lift. Deciding which statue belonged where, and which fragments went with which figure, drew on the community's own knowledge of the platform. Only one pukao, the red scoria topknot, was in good enough condition to be replaced; all fifteen are thought to have worn one, and the scoria for them came from Puna Pau, a separate quarry on the western side of the island. One moai never made it back into the line. It had travelled to Japan for an exhibition in Osaka and now stands separately near the entrance, known as the travelling moai, Moai Ha'ere Ki Haho.
What you are looking at is original stonework and original carving in modern positions. How closely those positions reproduce the platform's actual final arrangement before it collapsed, rather than the excavators' best reading of Heyerdahl's photographs and the scattered fragments, is a question the lead archaeologists raised themselves and did not close. It is a good deal better than guesswork. It is still an interpretation.
Two claims that circulate without measurements
The first is astronomical. It is widely repeated that the moai of Tongariki face the June solstice sunset, or that the platform's axis captures the December solstice sunrise, and at least one recent secondary source describes the orientation as exact and the astronomical intent as highly precise. No peer-reviewed measurement establishing any of that has been traced. An earlier version of this article stated that the central platform is oriented on the summer solstice sunrise. We can no longer support that sentence and we have removed it. The question is open, not answered, and we are not going to repeat a bearing nobody has published.
The second is geological. Joseph Davidovits has proposed that the moai were not carved from natural stone at all but cast from a synthetic geopolymer, and a 2025 paper in that framework applies the idea specifically to Ahu Tongariki. The thin sections say otherwise: the statues from this platform are natural Rano Raraku tuff, worked by carving. The geopolymer hypothesis has no standing in peer-reviewed Rapanui scholarship. We name it only because it is currently circulating with this site's name attached to it.
The collapse that the genomes do not show
Rapa Nui carries a second story, told as a parable: that the islanders cut down their forest to move statues, wrecked their own ecology and crashed from a large population to a remnant. Jared Diamond gave the argument its widest audience in Collapse in 2005. It is the version most visitors arrive with.
In 2024 a team published ancient genomes from 15 Rapanui individuals in Nature, working from remains held in French museum collections and radiocarbon dated between 1670 and 1950. The reconstructed effective population size does not fall off a cliff in the seventeenth century. It declines to a minimum well before the sampled individuals were born and then rises steadily until the 1860s, when the Peruvian slave raids and introduced disease did cause a documented catastrophe. When the authors simulated a severe 1600s bottleneck of the kind the collapse narrative requires, the simulated data were incompatible with what they actually observed. The same genomes carry Native American ancestry, and the authors' modelling places that contact somewhere between the mid-1200s and the early 1400s, centuries before any European ship arrived in 1722.
The deforestation itself is not in dispute. Rapa Nui did lose its palm forest. What the genetic evidence removes is the population crash that was supposed to follow, and with it the neatness of the parable.
The contact result is worth holding next to Thor Heyerdahl, whose expedition photographs made this platform's reconstruction possible and who spent decades arguing for a South American component in Rapa Nui's prehistory. Archaeology rejected his cultural model and still does: the Rapanui are Polynesian in origin by every material and genetic measure available. He was wrong about the culture. The narrower question of whether anybody ever crossed that water before Europeans did is not as closed as it looked, and the direction and the mechanics of that single admixture event are still under study.
Why the platforms sit where they sit
One more thing the tour does not usually mention. A 2019 paper in PLOS ONE by Robert DiNapoli, Carl Lipo and colleagues ran spatially explicit point-process models on 93 image-bearing ahu in the eastern part of the island, testing their positions against gardens, marine resources and fresh water. The best explanation was distance to fresh water, and specifically to coastal seeps, the brackish groundwater that emerges at the shoreline. On an island with no permanent streams, the statues appear to stand where the water is. That is a claim about position rather than about proportion, which is the same kind of question our own engine asks, and at Tongariki our engine cannot answer it.
What our engines found
Nothing usable, and the reason is ours rather than the site's. Our alignment engine measures great-circle distance ratios between catalogued sites. Inside a 200 km radius of Ahu Tongariki our atlas holds four catalogued sites, and after the 12 km separation rule only two of them are usable. Two points give a single distance, far too few to tell signal from chance, so we report the gap instead of a number.
A sparse catalogue is a hole in our data. It is never evidence of absence in prehistory, and anyone who tells you otherwise is reading a limitation as a finding. Expanding coverage in this part of the world is on our roadmap, and when it lands this section will carry a real result you can reproduce yourself.
One limit worth stating plainly, because it is the one most often glossed over. This test compares positions between monuments. It says nothing about any building's own proportions, its height, base, slope or perimeter. That is a separate question, and at several sites a genuinely open one.
Explore it yourself
Open the Ahu Tongariki record in the living atlas to see the platform in 3D, place it among the island's other catalogued sites and run the alignment and ley line engines on the real coordinates. The Ahu Tongariki Site-Emblem T-Shirt carries the row of fifteen, the site geometry and the exact coordinates.
Sources worth reading
Cristino and Vargas Casanova in the Rapa Nui Journal (1999) and the Journal of the Polynesian Society (2002), on the excavation and reconstruction of Ahu Tongariki. Van Tilburg and colleagues (2008), the petrographic analysis of the Tongariki statues. Van Tilburg (1994). Katherine Routledge, The Mystery of Easter Island (1919), which records the platform as it lay. Hunt and Lipo, The Statues That Walked (2011), and their chapter in Questioning Collapse. Lipo and Hunt in the Journal of Archaeological Science (2025) on the walking hypothesis. DiNapoli, Lipo and colleagues in PLOS ONE (2019) on why the platforms sit where they do. The 2024 Nature paper on the ancient genomes. The Easter Island Statue Project, the island-wide moai survey and catalogue.