What Diego de Landa Burned

Four Maya codices survive. There were, on any reasonable estimate, thousands. The gap between those two numbers has a name attached to it, a date, and a first-person account written by the man responsible.

Diego de Landa was a Franciscan friar in the Yucatán. In July 1562, at Maní, he presided over an auto-da-fé at which Maya religious objects and manuscripts were destroyed. In his own later account, Relación de las cosas de Yucatán, he wrote that they found a large number of books in these characters, and that as they contained nothing but superstition and lies of the devil, they burned them all, which the Maya regretted to an astonishing degree and which caused them much affliction.

That last clause is the part worth sitting with. He recorded the grief. He understood what he was destroying was valuable to the people it belonged to, wrote it down, and did not consider it a reason to stop.

What was lost is not fully knowable, which is the nature of the loss. The four surviving codices are the Dresden, the Madrid, the Paris, and the Maya Codex of Mexico, whose authenticity was disputed for decades and is now generally accepted. Between them they contain astronomical tables of remarkable precision, including a Venus table tracking the planet's synodic period with an error small enough that corrections were built into the table itself, and eclipse predictions covering centuries.

That is what survived by accident from a corpus that was destroyed on purpose. It is a reasonable inference that the destroyed material included more of the same, and no way at all to know what else.

The second half of the story is stranger and is the reason the episode resists a simple moral. Landa was recalled to Spain to answer for exceeding his authority, and while defending himself he wrote his Relación, an extensive description of Maya life, calendar, ritual and writing. In it he recorded what he believed was a Maya alphabet, obtained from an informant, matching Spanish letters to Maya glyphs.

He had misunderstood the system. Maya writing is largely syllabic rather than alphabetic, and for centuries Landa's alphabet was treated as a curiosity or a fraud. In the 1950s the Russian linguist Yuri Knorozov recognised that the correspondences were syllabic, not alphabetic, and that Landa's informant had been giving the sounds of Spanish letter-names. That reinterpretation is the foundation of the modern decipherment.

So the man who destroyed the library also wrote, in the course of justifying himself, the single document that made the surviving fragments readable. The record of a culture was burned and partially recovered by the same hand, and neither act was intended to produce the outcome it produced.

The methodological lesson is not about villains. It is that the suppression of a body of knowledge and the preservation of the key to it can be performed by the same institution, for reasons internal to that institution, and that neither the destruction nor the preservation tells you anything reliable about the value of what was lost.

Four books out of thousands. Every general claim anyone makes about Maya thought rests on those four, plus inscriptions in stone that were too heavy to burn.