Dresden Codex

Most of the codex consists of almanacs: the 260-day count laid out in rows, divided and subdivided, each interval attached to a deity, an offering, an omen and a task. They cover planting and rain, beekeeping, deer hunting, weaving, disease and its cures, and the ceremonies of the year-bearers. This is the working material of a Maya priest, closer to a manual than to a narrative, and it is written in the hieroglyphic script with numbers alongside.

Set among it are the astronomical tables that made the book famous. The Venus table tracks the synodic cycle of the planet, its appearances as morning star and evening star and its two disappearances, using a canonical figure of 584 days against a true value of about 583.92, and supplies correction intervals to be applied periodically so that the table does not drift out of true across centuries of use. That correction is the impressive part: the Maya astronomers knew their round number was slightly wrong and built the compensation into the instrument. The eclipse table sets out 405 lunar months in groups of five and six, marking the stations at which a solar eclipse becomes possible. It does not predict which eclipses will be visible from Yucatan and was not designed to. Multiplication tables of the kind needed to run these cycles against each other and against the 365-day year sit alongside them.

The final page shows a flood: water pouring from the body of a sky serpent and from a celestial band above, an aged goddess emptying a jar, and a dark deity below carrying weapons. It is a scene of destruction by water, of a kind attested elsewhere in Maya and central Mexican thought, and it has since been made to carry a great deal more weight than it will bear.

The codex was bought in Vienna in 1739 by Johann Christian Gotze, director of the royal library at Dresden, from a private owner, and given to the library. How it had come to Vienna is not documented. The usual assumption is that it went to Europe among the early consignments sent back from the Spanish colonies and reached the Habsburg court, but no record ties it to a particular shipment, and its history before 1739 is guesswork.

It is one of only four Maya books generally accepted as authentic. The others are the codices in Madrid and Paris, both damaged and both later, and the Maya Codex of Mexico, long called the Grolier Codex, which surfaced in the 1960s without archaeological context and was suspected of being a forgery for decades. Radiocarbon dating and materials analysis reported between 2016 and 2018 established it as genuine and in fact the oldest of the four, and it is now accepted by most specialists, though it is a fragment of a Venus almanac rather than a full book.

In February 1945 the bombing of Dresden destroyed the building in which the codex was housed. The manuscript survived, but water, from the hoses and from the flooded cellar in which it lay, soaked the pages, running pigment and staining the lime ground. The damage is permanent and plainly visible. It is the reason the photographic facsimiles made in the nineteenth century, before the war, now count as primary evidence in their own right rather than as reproductions. The codex is held by the Saxon State and University Library and has been digitised at high resolution.

The Dresden Codex, one of only four Maya books to survive the Spanish conquest, devotes one of its most carefully worked sections to a table tracking the planet Venus through its full 584-day cycle of morning-star and evening-star appearances and periods of invisibility, corrected across multiple cycles with real astronomical precision.

This was not simply a matter of intellectual curiosity. Maya inscriptions and painted ceramics record specific historical military campaigns explicitly timed to Venus's heliacal rising as morning star, glyphs modern epigraphers read as "star war" events, when a particular king's attack on a rival center is dated to coincide with Venus's reappearance. Susan Milbrath's Star Gods of the Maya documents this pattern across the corpus: Venus in its morning-star aspect was read as an omen of conquest, and warfare was scheduled to fall under its influence deliberately, the way a modern general might choose a date for symbolic reasons, except that the date itself was believed to carry real supernatural weight for the outcome.

Quetzalcoatl, in his aspect as Tlahuizcalpantecuhtli, is this same morning star among the Aztec, inheriting a Venus identification shared across Mesoamerican cultures generally rather than confined to the Maya, and his own myth of descent into the underworld and difficult return echoes the planet's own disappearance and reemergence. The connection between a documented astronomical table, a documented military practice, and a major deity's own name and myth is one of the clearest cases anywhere in the world of a society building both practical calculation and religious meaning around the same observed celestial cycle.

Ernst Wilhelm Forstemann was the royal librarian at Dresden, not a Mesoamericanist, and he had the codex in front of him. Working in the 1880s he broke the numeration, the bar-and-dot system, the shell sign for zero, the place value running in twenties with a break at the third position, and from it reconstructed the Long Count, the absolute chronology by which the Maya dated events from a fixed zero point. He identified the Venus table for what it was and worked out the architecture of the eclipse table. All of this was done half a century and more before the phonetic values of the script were established: the Maya could be read as mathematicians and astronomers long before they could be read as speakers of a language.

The same work began the codex's least useful legacy. Forstemann described the final page as a picture of the destruction of the world; Sylvanus Morley repeated and amplified that reading, and through his widely read survey of 1946 the flood page entered popular circulation as Maya apocalypse. When the thirteenth bak'tun of the Long Count ended in December 2012, the two were welded together in books, films and television, and the Dresden Codex was cited as the ancient authority.

There is nothing in it. The flood page carries no date and no reference to any bak'tun ending. The Long Count did not stop in 2012; Maya inscriptions record dates far beyond it, and a text at Palenque projects forward to a date in the year 4772. Mayanists said so consistently for years and were consistently ignored. The codex records a calendar built to keep running, not to expire.