Rongorongo
UnsolvedSummary
Twenty-seven wooden objects, carved with several hundred distinct glyphs, from the most isolated inhabited island on Earth. If rongorongo is writing, it is one of a handful of times in human history that writing was invented from nothing.
It is undeciphered, and it is very likely to stay that way, because of what happened to the people who could read it. In 1862 and 1863, Peruvian slave raiders took a large part of the island's population, including much of its elite. Smallpox followed. By the time anyone from outside thought to ask what the tablets said, there was almost nobody left who could answer, and the answers that were given do not hold up.
What is documented
The corpus. Fewer than thirty objects survive, generally counted as twenty-seven. Wooden tablets, a chief's staff, a chest ornament, and a few other items. None of them is on Rapa Nui; they are scattered through museums and religious institutions in Europe and elsewhere.
The script. More than four hundred distinct glyphs have been catalogued, across roughly fifteen thousand characters in total. That total is the fundamental problem: it is a very small corpus by the standards of decipherment.
How it is written. In reverse boustrophedon. You read a line, then rotate the object one hundred and eighty degrees to read the next. It is an unusual system and it is consistently applied.
The one passage anyone can read. A section of the Mamari tablet is generally accepted to be a lunar calendar, on the grounds that its structure aligns with Rapa Nui and other Polynesian lunar calendars. This is the only part of the corpus with a relatively understood meaning, and it is a structural match rather than a translation.
The collapse. In 1862 and 1863, Peruvian slave raids removed a large fraction of the island's population. The raids fell disproportionately on the elite, who are precisely the people a script like this would have belonged to. Smallpox, introduced by returning survivors, killed a large part of the remainder.
Eugène Eyraud, a French lay missionary, reported the tablets to the outside world in 1864, which is to say: one to two years after the population that could read them had been destroyed.
The attempted readings, and why they fail. In the 1870s Bishop Jaussen of Tahiti worked with a Rapa Nui man, Metoro, who chanted over the tablets. The chants were recorded. They do not correspond to one another consistently, and when Metoro read the same tablet twice he produced different chants. The prevailing assessment is that he was describing the pictures rather than reading a text, which is what a person does when they have seen a script but were never taught it.
The 2024 dating. Silvia Ferrara and colleagues at the University of Bologna radiocarbon-dated four rongorongo tablets held in Rome. Three came from trees felled in the eighteenth or nineteenth centuries. One, tablet D, known as Échancrée, came from a tree felled between roughly 1493 and 1509, around two hundred years before Europeans reached the island.
Leading explanations
It is an independent invention of writing (the headline claim, and it goes further than the evidence).
The 2024 result was reported worldwide as proof that Rapa Nui invented writing on its own. It is a genuinely important paper and it does not establish that. The registry sets out the gap carefully, because this is exactly the sort of gap the registry exists for.
Radiocarbon dates the wood, not the carving. It tells you when a tree was cut down. It does not tell you when somebody carved a glyph into the plank. The authors argue that using centuries-old wood would have been impractical, and that is an argument, not a measurement.
The wood is not local. Tablet D is made of Podocarpus latifolius, an African species. It did not grow on Rapa Nui. It arrived, and the most likely way for a plank of African wood to reach the most isolated island on Earth is as driftwood. Driftwood can float for a very long time. A piece of wood that grew around 1500 can be picked up off a beach in 1780 and carved the following week.
It is one specimen. The other three tablets in the same study date to the eighteenth or nineteenth centuries.
It has not been shown to be writing. This is the deepest objection, and it was raised in the journal's own comments. To call rongorongo an independent invention of writing, in the sense that Mesopotamia, Egypt, China and Mesoamerica were, you must first establish that it encodes language. Nobody has. Until it is read, it might be writing, and it might be a mnemonic system, or a proto-script, or something else.
The paper's actual result is real and useful: one plank used for a rongorongo tablet is older than European contact. The claim built on top of it is bigger than the plank.
Stimulus diffusion. The alternative. Rapa Nui islanders saw European writing, most plausibly the Spanish annexation ceremony of 1770 at which islanders were asked to make marks on a document, understood the idea that marks can hold speech, and invented their own system to do it. This would make rongorongo an original script but not an independent invention of the idea of writing, which is a different and much commoner event.
The 2024 date makes this harder to sustain, if the carving is as old as the wood. That conditional is doing all the work.
What the popular version gets wrong
"Radiocarbon dating proved the script predates European contact." It dated a plank. The relationship between when a tree was felled and when somebody carved it is an assumption, and in the case of a plank of African wood on a Pacific island, it is an assumption that has to survive the question of how the wood got there.
"Rongorongo is a lost writing system." It is an undeciphered glyph system, and whether it is writing in the technical sense, encoding language, is itself unresolved. Almost every account skips this.
"The islanders could still read it in the 1870s." One man produced chants over the tablets and could not reproduce them. That is not reading.
"It will be deciphered eventually." Probably not. Decipherment normally requires either a bilingual text or a very large corpus in a known language. Rongorongo has neither: no bilingual, fifteen thousand characters in total, and a language whose relationship to the script is unknown. The people who could have told us were taken in 1862.
Current status
Unsolved. Twenty-seven objects, four hundred glyphs, one probable lunar calendar, and no reading.
The registry keeps this record as its clearest case of a mystery that was not preserved but created. Rongorongo did not become unreadable through the slow attrition of time. It became unreadable inside about eighteen months, because the people who could read it were taken away in ships and the ones who came back brought smallpox.
Sources
- Ferrara, S., Tassoni, L., Kromer, B., Wacker, L., Friedrich, M., Tonini, F., Lastilla, L., Ravanelli, R. and Talamo, S. (2024). "The invention of writing on Rapa Nui (Easter Island). New radiocarbon dates on the Rongorongo script." Scientific Reports 14, 2794. doi:10.1038/s41598-024-53063-7
- Published critical commentary on the above, including on the identification of tablet D's wood as an African species and on the inference from wood date to inscription date.
- Barthel, T. (1958). Catalogue and tracing of the rongorongo corpus.
- Horley, P. (2021). Revised tracing of the corpus.
- Guy, J. and subsequent work on the Mamari lunar calendar passage.
- Jaussen, T. and the Metoro chants (1870s), and later assessments of their reliability.
- Eyraud, E. (1864). First outside report of the tablets.
Last reviewed: July 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.