Preserved by Their Enemies

In July 1562, at Maní in Yucatán, the Franciscan Diego de Landa held an auto-da-fé. Maya images were smashed and Maya books were burned (Britannica). A few years later, writing his account of the Maya in Spain, he described it in one sentence:

“We found a great number of books of these letters of theirs, and because they contained nothing that did not have superstition and falsehoods of the devil, we burned them all”

The Spanish goes on: lo qual a maravilla sentian, y les dava pena, which they felt to an extraordinary degree, and it grieved them. The next sentence on the same page begins: “Of their letters, I will place here an a, b, c”. Four pages later, under the heading Siguese su a b c, Landa’s book prints twenty-seven Maya signs, each beside the Spanish letter it was supposed to stand for.

The man who burned the books also kept the only key to them that survived the century. Brasseur, the French priest who found and published the manuscript, saw that at once. Landa, he wrote, “rendered an immense service to historical sciences by compiling the precious information that follows, and by preserving for us the characters of the Maya alphabet”; the book “is the key to American inscriptions” (Brasseur’s introduction). It was not an alphabet. The key turned only in 1952, when the Soviet linguist Yuri Knorozov argued that Landa’s informant had written syllables: asked for the letter b, pronounced be, he wrote the sign for the syllable be (University of North Carolina Libraries). That reading opened the Maya script.

This is a pattern, and it is older than Yucatán. A tradition is attacked in writing; the attacker, to refute it, has to quote it; the tradition’s own books are lost; and the refutation survives because it was orthodox, and so was copied. The enemy becomes the archive. The pattern is clear in the early Church. In New Spain it is stronger, and it is also where the comparison stops working.

“Sealed up, hidden, concealed”

Around 225, Hippolytus of Rome wrote a Refutation of All Heresies. Most of it was itself lost until the 1840s, when a Greek scholar sent by the French government to the monasteries of Mount Athos came back with a manuscript of seven of its ten books (Legge’s introduction). Its value, wrote its English translator F. Legge in 1921, is the Gnostic writing it quotes, “of which well-nigh the whole has been lost to us” (p. 3).

One of those quotations is a hymn of the sect Hippolytus calls the Naassenes. He introduces it with contempt: “This psalm has been concocted by them, whereby they seem to hymn all the mysteries of their error” (p. 144). Then he copies it out. It ends with Jesus asking the Father to send him down to the soul lost in chaos:

“Holding seals I will go down,
Through entire æons I will pass,
All mysteries I will disclose;
The forms of the gods I will display;
The secrets of the holy way
Called Gnosis, I will hand down.”

In the next book Hippolytus turns to Simon Magus, accuses him of “stupidly and clumsily garbling the Law of Moses”, and quotes the opening of a book attributed to him, the Great Announcement:

“This is the writing of the Announcement of Voice and Name from the Thought of the great power of the Boundless One. Wherefore it will be sealed up, hidden, concealed and will be in the dwelling-place where the root of the universals is founded.”

A book that announces its own concealment survives in the pages of the man who wrote to expose it. (The English here is Legge’s printed translation, not ours.)

Hippolytus was not the first. Irenaeus of Lyon, about 180, explains in his preface why a bishop should spend a book on doctrines he despises: “I have deemed it necessary, after reading the commentaries of the disciples of Valentinus, as they themselves call them, and having even met with some of them and grasped their way of thinking, to reveal to you, most beloved, the monstrous and profound mysteries” (Harvey’s edition, 1857). His first chapter then sets out the Valentinian system from its first principle:

“They say that there is a certain perfect Æon, who existed before, in invisible and ineffable heights. Him they call Proarche, and Propator, and Bythos”

The chain goes one link further. Irenaeus’s own Greek is lost; his work survives whole only in an old Latin translation. For this chapter, Harvey notes at the foot of the same page, “Epiphanius in his work upon the heresies has preserved to us the Greek text”. The refuter of the Valentinians is himself preserved by a later refuter.

That later refuter, Epiphanius of Salamis, writing in the 370s, did the most generous thing a heresiologist ever did. Before attacking Ptolemy, a disciple of Valentinus, he copied out one of Ptolemy’s letters entire, and said why:

“For a more abundant refutation of the deceiver, I will set down in order and present the persuasive and poisonous words written by Ptolemy himself to a certain woman named Flora. This is so no one thinks we refute the deceiver based only on hearsay, without having first encountered his distorted teaching.”

The letter follows, beginning “My good sister Flora, since not many have understood the law established through Moses”. No other copy of it is known. It exists because a bishop wanted no one to say he had argued against a rumour.

The physician must know the disease

In New Spain the same logic ran under a new metaphor. Bernardino de Sahagún opens the twelve books of what is now called the Florentine Codex with it:

“A physician cannot properly apply medicines to a patient without first knowing from what humor or cause the illness proceeds. … Preachers and confessors are physicians of the souls, tasked with curing spiritual diseases.”

The disease is idolatry, and Sahagún says it is not over: “The sins of idolatry, rites, superstitious idolatrous omens, abuses, and idolatrous ceremonies have not yet been entirely lost.” Worse, the clergy cannot see it: “Because of a lack of this knowledge, many idolatrous things happen in my presence without the ministers understanding them.” So the friar sets out to record everything, in Nahuatl and Spanish side by side.

The fifth book, on omens, carries the metaphor through. The omens are there “so that when they are wounded by this sore, they may go to seek medicine, and the physician may easily understand them” (prologue to Book 5). Then the book lets the diviner speak. A man has heard a wild beast howling at night (the painting at the head of this essay opens the chapter) and goes to the tonalpouhqui, the reader of the day-count, who tells him what it means and on what authority: “And this is not because I say it: but because my elders and ancestors left it said and written.” In the Nahuatl column the diviner points to his source: “Listen to the word, for it is marked in the tonalamatl”, the book of days (Book 5, ch. 1). The books of the kind Landa burned are cited by name, inside the book written to root them out.

That book nearly went the same way. A royal decree of Philip II, dated 22 April 1577, ordered Sahagún’s work collected because it contained the rites, ceremonies and idolatries of the Indians, and forbade anyone to write about their superstitions and way of life (Alfredo López Austin, Augurios y abusiones, UNAM). The manuscript in our scans went to Spain and then to the Medici, and it is now in the Biblioteca Laurenziana in Florence (Historia y Espacio).

The sermon books show the same disease being charted in print. Fray Juan Bautista’s Nahuatl Sermonario of 1606 preaches on the first Sunday of Advent, “let us put on the armor of light”. He imagines his listeners objecting that “idolatry has already left, has already perished, and demon-worship scarcely shows its footprints anymore”, and answers: “Behold, listen, my children: there are still some who lead people astray, who still live in doubt” (p. 17). A list follows, and the printer set a Spanish gloss after each Nahuatl word, so that the preacher reading it would know exactly whom he was naming:

The list runs to seven kinds of specialist across the two pages: day-count diviners, those who claim to draw stones out of the body, those who revive the frightened and fainting, those who give medicines to barren women, star-readers, water-gazers, night shape-changers. A sermon meant to make them disappear is, page for page, a record of who they were.

A century and a half later they had not disappeared. Ignacio de Paredes’s Promptuario manual mexicano (1759), a book of ready-made sermons for parish priests, still warns that God is offended by “whoever goes about consulting, goes about seeking out, goes about testing” the sorcerers and “the shapeshifters Nanahualtin”; the Spanish in the margin of the print says it plainly, Hechizeros, Nahuales (read the page). And it tells its hearers what to do if someone shows them a god: “And do not be afraid of it in any way, do not fear it” (read the page). More than two centuries after the conquest, a preacher still had to tell people not to be afraid of the old gods.

Some of these records are less a sermon than a census. Hernando Ruiz de Alarcón’s Treatise on the Heathen Superstitions (1629) recorded the incantations Nahua healers and farmers used, in Nahuatl with a Spanish translation, to help the Church abolish them (University of Oklahoma Press). We do not yet hold it; it stayed in manuscript until 1892.

The Franciscans in Yucatán made the same kind of record without the sermon. The great Maya–Spanish dictionary compiled at Motul, probably by Antonio de Ciudad Real in the late sixteenth century, defines the vocabulary of Maya religion the way it defines the vocabulary of farming, flatly. Under ah kin: “a priest”. A few entries later, ah kinyah: “a sorcerer who casts lots with superstition” (vol. 1, f. 20). Ah cunal can, ceh, balam: “a charmer or sorcerer who charms snakes, deer, jaguars, and so on” (vol. 1). Ah ohel tukinam xiuob, “an herbalist”, one who knows the power of plants (vol. 1). In the Spanish–Maya volume, under hechizar and hechizero, to bewitch and sorcerer, the Maya is pulyah, and “Bewitched: pulaan yaah yokol” (vol. 2). Those words will come back.

Where the parallel breaks

Hippolytus and Sahagún did the same thing for the same reason. But the Mesoamerican record differs from the Greek one in four ways, and each matters for how you read it.

It was not written by the enemy alone. Hippolytus worked from books and, Legge suspected, from what a convert told him (Legge’s note). Sahagún worked with a team, and named it. In the prologue to Book 2 he lists the elders chosen by the town governments and the college students who checked and rewrote everything with him; among them “the principal and most wise was Antonio Valeriano, a resident of Azcapotzalco”, with Alonso Vegerano, Martín Jacobita and Pedro de San Buenaventura, “all experts in three languages”, Latin, Spanish and Nahuatl (Book 2, prologue). The Nahuatl column of the Florentine Codex is, in large part, Nahua writing.

It keeps the original language. Irenaeus summarised the Valentinians in his own Greek; Epiphanius copied one letter. The friars needed the words themselves, because they had to preach and hear confession in them. So the Sermonario prints the Nahuatl names, Paredes prints Nanahualtin, and the Motul dictionary records the exact Maya term for each kind of specialist. A reader today can go behind the friar’s gloss to the word he was glossing.

The categories are the friars’. Every one of these words reaches us sorted under hechizero, sorcerer, the category of a Spanish confessor. Sometimes the sorting shows. The Motul dictionary defines ah koh keuel as “a sorcerer or wizard who dressed in the skins of jaguars and frightened people in order to rob them” (vol. 1). A man in a jaguar skin may have been a robber, or a dancer, or a priest in regalia whose purpose the lexicographer did not know or did not record. The word is recorded exactly; the explanation attached to it is the friar’s. The Spanish–Maya volume shows the filing from the other side: the one Spanish word hechicero is answered with three Maya ones, “Sorcerer: tutuz nac: ahcunyah: ahpulyaah” (vol. 2, f. 127). Three Maya words, one Spanish category.

There is a check. For the Gnostics, the check arrived late. Legge, in 1921, could write that apart from patristic quotations “the only extant Gnostic works are the Coptic collection in the British Museum and the Bodleian at Oxford, known as the Pistis Sophia and the Bruce Papyrus respectively” (p. 3, note 1). Twenty-four years later, in 1945, farmers near Nag Hammadi in Upper Egypt dug up a jar of papyrus codices holding more than fifty treatises, most of them Gnostic, and the heretics could at last be read in their own words (Biblical Archaeology Society).

For the Maya the check was never buried. After the conquest, Maya scribes learned the Latin alphabet from the friars and used it to write their own books, the books of Chilam Balam, copied and recopied in towns across Yucatán into the nineteenth century. We hold two copies of the book from Kaua. They are Maya writing for Maya readers, and they keep what the dictionaries filed under sorcery.

The Kaua book carries a year-count that runs Christian years beside Maya ones. The line for 1810, read off the page, gives 7 Ahau, the day 9 Ik, the 2nd of the month Kankin (read the page). It prescribes remedies by plant, and gives them a lineage the friars never would: a bath of gumbo-limbo (chacah), the scribe writes, “has been used since ancient times to stop the pain and the aching of the body” (p. 5). In the copy Teobert Maler photographed, one remedy is headed “For the cure of the lower-body sickness sent by sorcery”: in the Maya, pulbil yah (read the page). It is built on the word the Motul dictionary gave for bewitching. In the dictionary it names a sin to be confessed. In the Kaua book it names a sickness that has a cure.

Reading the enemy’s archive

None of this redeems the burning. Landa’s a b c is one page against an unknown number of books, and the Florentine Codex exists because its author was racing a policy that wanted the old ways erased. But the readers who come to these books now come for what their authors tried to destroy, and the books reward them, on two conditions. Read the original words, not the friar’s category for them. And find the other witness: Nag Hammadi for the Gnostics, the Chilam Balam books for the Maya, the Nahuatl column beside the Spanish one.

Every passage quoted here links to its page, with the scan beside the transcription and the translation, so you can do both.

About the English. Except for Legge’s Hippolytus and Harvey’s notes, which are printed English, the English quoted here is machine translation from the original pages, and no scholar has reviewed it. For the two Nahuatl sermon books, an AI judge compared the translation with the page image on 20 pages (pp. 60–69 of each, 13 October 2026) and scored it 4.1 of 5 for fidelity. It flagged about one mistranslated phrase per page and one omitted phrase every two pages; that judge comes from the same model family as the translator (experiment record). The Motul dictionary was translated by Claude Opus; single pages were checked against the scans, but no error rate has been measured. The other books have no measured rate yet. For this essay, every passage the argument rests on was read against its page image, and the Spanish, French and Greek of the prints was read off the scans. If you find an error on any page, the feedback button on that page reaches us.

Related: Indigenous traditions, Maya, Aztec and Nahua, Mesoamerican