Poking through some of the innermost rubble of an ancient pyramid known as Las Pinturas in San Bartolo, Guatemala, graduate student Boris Beltrn uncovered a boulder-size chunk of plaster. Mayan builders had created the boulder when constructing the third version of Las Pinturas, following their practice of supporting subsequent structures with the ruined remnants of the preceding pyramid. This particular fragment happened to be part of an ancient mural and thick black hieroglyphics ran down its side, following a faint pinkish-orange guideline.

"When he showed it to me, I asked him: 'Do you know what you have there?'" recalls team leader William Saturno of the University of New Hampshire. "'That is likely the earliest text in the Maya era. It is likely among the earliest texts in the New World as a whole.'"

In a paper published in the current Science Express the researchers describe the 10-glyph fragment found on the boulder. "The glyph ajaw, meaning ruler, is clearly visible," Saturno notes. "It's not rudimentary. The style is distinct. It's clearly not the first one ever painted."

But according to radiocarbon dating of burnt wood bits found in the plaster and from surrounding strata, it is by far the oldest known Mayan writing--dating from between 300 and 200 B.C., which is roughly concurrent with the earliest writings of other Mesoamerican cultures. Previous examples of Mayan script could only be confidently dated to around A.D. 250, leading to speculation that the Mayans may have inherited their writing from other, older cultures, such as the Zapotec, despite stylistic differences.

The find seems to upend that theory, proving that the ancient Maya were as literate as their descendants. Unfortunately, the plaster piece only provides what appears to be the tail end of a longer text and, without the accompanying pictures of the mural it accompanied, it defies easy translation. "We're talking about 400 to 600 years before [modern researchers are] literate. It's pretty arcane," Saturno explains. "Most likely it's part of a title referring to some person, mythic or real, we don't know."

Archaeologists will have to wait for more ancient writings--hopefully attached to murals for easy translation, Saturno says--to offer any hope of piecing together the script. "The more of it we find," he observes, "the more likely we are to recognize the patterns and make connections with later texts to be able to trace things back and forth."