III · THE WORK · ART HISTORY
The Florentine Codex
▶ Listen · narrated
When indigenous scholars wrote their own history under colonial supervision, they wrote it twice: once in Nahuatl, once in Spanish. The two versions diverge in instructive ways.
At a glance
- Date
- 1575
- Medium
- Ink and pigment on paper
- Collection
- Biblioteca Medicea Laurenziana, Florence
- Genre
- Chronicles of the Indies
Look closer
Parallel Nahuatl and Spanish columns
Each page is divided into two text columns, Nahuatl on the right and Spanish on the left. The arrangement suggests equivalence, but the columns do not always agree. The Nahuatl text sometimes includes detail the Spanish omits, or frames an event differently. The layout presents them as translations of one another, but they function more as parallel accounts written for different audiences with different assumptions about what needs explaining.
European shading applied to Nahua conventions
The illustrations combine two visual systems. Nahua artists used conventions for depicting figures, architecture and ritual objects that were legible within their own pictorial tradition. European techniques of shading, perspective and volumetric modelling were then applied to these forms. The result is not quite either system. A temple might be drawn in the flat, emblematic manner of pre-contact codices, but with shadows cast across its steps in the European manner.
Indigenous glyphs surviving inside the images
Scattered through the illustrations are glyphs from the Nahua writing system: name signs, place markers, calendrical symbols. They sit alongside European-style representational drawing, performing a different kind of work. Where the drawing shows what something looked like, the glyph states what it was called or when it occurred. The two systems of inscription occupy the same page without merging, each doing what the other cannot.
The story
The manuscript now known as the Florentine Codex was compiled in 1575, thirty years after the Franciscan friar Bernardino de Sahagún began interviewing Nahua elders about the culture and history of central Mexico. Sahagún worked with former students from the Colegio de Santa Cruz de Tlatelolco, a school established to train indigenous men in Latin, Spanish and European scholarly methods. These men, whose names are not recorded in the manuscript itself, conducted interviews, organised the material, wrote the Nahuatl text and created the illustrations. The project took more than four decades.
The manuscript is organised into twelve books covering cosmology, ritual practice, natural history, social structure and the Spanish conquest. Each section is presented in two columns: Nahuatl on the right, Spanish on the left. The Nahuatl text was written first; the Spanish was added later, and it is not a direct translation. The Nahuatl is more detailed, more technical in its vocabulary, and occasionally frames events in ways the Spanish does not replicate. In Book Twelve, which describes the conquest, the Nahuatl account includes speeches and internal perspectives that the Spanish summary reduces or omits entirely.
More than two thousand illustrations accompany the text, drawn by Nahua artists trained in both indigenous pictorial traditions and European techniques. A single image might show a deity rendered in the flat, frontal manner of pre-contact manuscripts, but with European shading applied to suggest volume. Glyphs from the Nahua writing system appear alongside representational drawing, marking names, places and dates in a script the Spanish text does not reproduce.
The manuscript's title page calls it "Historia General de las Cosas de Nueva España" — The General History of the Things of New Spain. It is now called the Florentine Codex because the best-preserved copy is held in the Laurentian Library in Florence. How it arrived there is not documented. Sahagún's project was not officially supported by the Spanish crown; at one point his papers were confiscated. The manuscript that survives is a fair copy made in 1575, not Sahagún's working drafts.
The first complete English translation, by Charles Dibble and Arthur Anderson, took thirty years and was published between 1950 and 1982. In 2012 the manuscript was digitised and made freely available. Scholars continue to compare the Nahuatl and Spanish columns, and the comparison remains productive. The disagreements between them are not errors. They are evidence of negotiation: what could be said, to whom, and in which language.
Why it mattered then
The Florentine Codex was compiled at a moment when indigenous knowledge systems were being simultaneously recorded and suppressed. Sahagún's project began in 1545, a generation after the conquest, when many Nahua elders still remembered pre-contact life. The Franciscans believed that understanding indigenous religion was necessary to replace it with Christianity. Sahagún argued that one could not combat idolatry without knowing what it consisted of. The Nahua scholars who worked on the manuscript had been educated in a colonial institution designed to produce indigenous men fluent in European languages and learning. They used that training to document their own culture in their own language, under the supervision of a friar whose stated purpose was evangelical. The resulting manuscript is shaped by that tension. It is encyclopaedic in ambition, recording everything from the properties of medicinal plants to the speeches made at royal funerals. It is also a work of cultural preservation created under conditions that made preservation both urgent and suspect. The two-column format made the manuscript legible to Spanish authorities while allowing the Nahuatl text to preserve detail and nuance the Spanish did not replicate. Whether this was deliberate strategy or the natural result of translation is not settled, but the effect is clear: the Nahuatl column contains knowledge the Spanish one does not fully convey. In a period when indigenous manuscripts were being destroyed as idolatrous, this one survived partly because it appeared to serve Spanish purposes.
Why it matters now
The Florentine Codex is the most extensive documentation of Nahua culture compiled in the sixteenth century, and for many aspects of that culture it is the only detailed source that survives. It has been used by historians, linguists, anthropologists, art historians and scholars of religion. It is also a text whose authorship is complicated. Sahagún's name is on it, but the manuscript was written and illustrated by Nahua scholars whose names were not recorded. The question of who authored the codex, and what authorship means in a collaborative project shaped by colonial power, has not been resolved. The digitisation of the manuscript in 2012 and the release of a high-resolution digital edition in 2023 have made it widely accessible for the first time. Scholars can now compare the Nahuatl and Spanish texts line by line, and study the illustrations in detail. The availability of the Nahuatl text has been particularly significant. For a long time the codex was read primarily through its Spanish column or through English translations based on the Spanish. Reading the Nahuatl directly reveals a different, more complex account. The codex is now studied not only for the information it contains but as an example of how knowledge is produced under colonial conditions. It is a text written by indigenous scholars for European readers, but also for themselves and for posterity. The fact that the two columns do not always agree is part of what makes the manuscript valuable. The disagreements show us the limits of translation and the persistence of perspective.
The surprising detail
The manuscript was inscribed into UNESCO's Memory of the World register in 2015, but it has been lent out precisely once. It travelled from the Laurentian Library in Florence to an exhibition in Mexico in 2009, under extraordinary security. The loan required years of negotiation, and the Mexican government provided a reciprocal loan of comparable value. The codex returned to Florence after the exhibition, and the library's position is that it is too fragile to travel again. The most extensive documentation of Nahua culture, created in Mexico by Nahua scholars, is held in Italy and cannot go home.
What is disputed
The names of the Nahua scholars who wrote the Nahuatl text and created the illustrations are not recorded in the manuscript. Sahagún refers to them collectively as former students but does not name individuals. How much editorial control Sahagún exercised over the final text, and how much autonomy the Nahua authors had, cannot be determined from the manuscript itself. Scholars disagree about whether the differences between the Nahuatl and Spanish columns represent deliberate strategy, practical constraints of translation, or something else. The question of authorship — whether this is Sahagún's work, a collaborative project, or a Nahua text compiled under Spanish supervision — remains open and has implications for how the codex is read and cited.
Remember this
The Florentine Codex is two books in one binding, written in two languages that do not always say the same thing. The divergences are the point.
Test yourself
Why would Nahua scholars working on the codex have reason to make the Nahuatl text more detailed than the Spanish, even though both were subject to Spanish oversight?
Because the Nahuatl text could preserve technical vocabulary, ritual detail and cultural knowledge that would be legible to Nahua readers but might pass unnoticed by Spanish censors who could not read Nahuatl fluently. The Spanish column provided enough information to make the project appear useful for evangelical purposes, while the Nahuatl column could function as a more complete record for a different audience. This was not necessarily a deliberate act of resistance — it may simply reflect the fact that translation always involves selection and loss — but the effect was to create a document whose full contents were not equally accessible to all its readers. The two-column format allowed the manuscript to serve multiple purposes at once, which may be one reason it survived when other indigenous texts did not.