The Construction of the Codex In Classic- and Postclassic-Period Maya Civilization

Thomas J. Tobin

How did the Maya make books? This simple question is difficult to answer because researchers today must rely on what are often no better than educated guesses in order to reconstruct the practices of the ancient Maya scribes. However, based on the aggregation of evidence from varied sources, it is possible to create a reproduction of a Maya codex with reasonable certainty of the methods. This essay will first explore the history of Maya papermaking and codex production, then enumerate the papermaking and bookbinding processes that can be adopted today as being reasonably close to the practices of the Maya. Although it is tempting to infer that this project is an attempt to prove exactly what the Maya did when they made paper and codices, it is impossible to recapture a historical moment; one can only create description based on evidence. Thus, the problem to be solved in this project is to set down the steps by which researchers today can create reasonable facsimiles of Maya codices. The solution to this problem lies in identifying materials, codifying recipes and processes, and testing hypotheses of practice.

A short history of the Maya codex
             600-1519: Vases, codices, and buildings
             1519: The Spanish conquest
             1940s: Von Hagen and Lenz
             1980-2000: The codex as book form
Ritual preparations
The recipe for paper
Making paper
Assembling the codex
Making brushes/pens
Making ink
Writing the text
             Grid lines
A Short History of The Maya Codex

During the period 100-700 C.E., the Romans invented and improved upon the codex, the form of the book that would endure for nearly two millennia. However, during roughly the same period of time, the Maya culture in and around what is now the Yucatan Peninsula in southern Mexico produced codices, as well, independently of Europeans. The factors leading to the development of the codex in European culture have been well documented, and the close identity between the Roman and Maya solutions to the problems of codex design, manufacture, purpose, and usage suggests that the codex form of the transmission of information is not merely an accident of geography or history. The Maya developed paper quite early in the millennium: “archaeological evidence of manufacture and use of bark paper by Maya dates from the early 5th century [C.E.]” (López). The Maya named their paper huun, and saw it as a writing surface when they appropriated their bark-cloth tunics as a possible means of transmitting information: “early in their history the Mayas produced a kind of tapa cloth from the inner bark of certain trees. . . . Bark-cloth manufacture apparently evolved into papermaking, although when this occurred is not known” (Sandstrom and Sandstrom, Traditional Papermaking 13). The Maya developed paper screen-fold codices as a direct step beyond carving information into stone buildings and stelae, unlike Western papermaking, which took a more circuitous route to reach its final form (single sheets, papyrus rolls, and then leafed codices).

Part of the difficulty in producing a history of the development of Maya papermaking has been that so few examples survive. Of the thousands of codices mentioned in the glyphic records of the Maya and the inventories of Spanish conquistadors, all but four partial examples were burned as suspected satanic manuals, thrown to the bottom of the ocean in ship raids, or moldered into dust in hot and humid storage conditions. Thus, the task of the historians of Maya papermaking is made more difficult in that they must reconstruct practices and relative dates from a melange of indirect data: ethnobotanical analysis; historical accounts at second, third, and sometimes no hand; sometimes-unreliable translations of Maya glyphs into Spanish; external evidence presented by the surviving codices themselves, including microscopic identification of fibers; archaeologically associative clues; and anthropological evidence based on the paper-making practices of Maya-language speakers today. Each of these types of evidence introduces a factor of uncertainty to the results obtained due to the barriers of time, language, scarcity, and interpretive difference.

However, the development of current theories of Maya codex-production can be delineated fairly clearly. The first accumulation of data occurred during the late Classic and early Postclassic Periods (c. 600-1519 C.E.), and the information was recorded by the Maya themselves, in the form of carved monuments, painted ceramics, and codices.  
600-1519: Vases, Codices, and Buildings

Although the focus of this project is neither epigraphy nor the study of Maya monuments and vases, a word is in order about the interrelation of epigraphic study and the information to be found on Maya architecture and pottery. Based on the evidence we have today, the Maya seem to have conducted their daily lives in accordance with their learning. During the Classic (ca. 300-900 C.E.) and Postclassic (ca. 900-1540 C.E.) periods of Maya history, dynasties, economies, and modes of learning came and went, and nearly all were recorded by Maya scribes. We can identify three main types of Maya records: historical records which delineate the deeds of noble persons, rulers, and states; scientific records which encompassed the learning of Maya scholars with regard to such diverse subjects as astronomy, agronomy, military science, statesmanship, and religious ritual; and economic records which served to mark down routes of trade, amounts given and received in tribute, and economic policies of the state. There is considerable overlap among these categories, but if we remember nothing else about Maya record keeping, it is that the Maya named nearly everything and structured their rituals, their agriculture, and their warfare around their calendar. As in European cabalistic thinking during the same time period, the Maya believed that to name a thing was to have power over it; likewise, the ability to keep records that encompassed learning beyond the span of one lifetime allowed the Maya to profit from their past.

Thus, the occupation of scribe, or ah ts’ib, which means “he of the writing,” was an important one in Maya culture. We have evidence of the output of Maya scribes in glyphs and pictographs carved into Maya architecture, painted onto Maya codices, and glazed onto Maya pottery. Maya pottery has been especially helpful in preserving two issues important to the reconstruction of a Maya codex: Maya glyphic/phonemic writing and depictions of Maya codices. Part of the glyphs on Maya pottery are always the same:

The Primary Standard Sequence or PSS is a string of about 35 glyphic signs--in effect a formula--laid out in an invariable sequence. No one ceramic text contains them all: there may be as few as 3 or 4 PSS glyphs on a vessel, or as many as 22, but they are always in the same order. When the present writer discovered them in 1972, he had thought them to be a ritual formula, perhaps a chant, recited to accompany the deceased owner of the vessel into Xibalba, the Underworld. Since that time, however, epigraphers have concluded that the PSS is one more case of name-tagging, in this case for the vessel itself, with the owner or patron being named, along with his or her titles, at the end of the PSS string. (Coe 60)

The PSS allowed twentieth-century scholars to begin to decipher the grammar and meaning of Maya language, and, more importantly for this project, some Maya vases contain scenes of scribes writing in codices, which, as we shall see in the next section, is an important clue to the physical makeup of the Maya codex. Paper itself was a valued commodity in Maya society, as it was used as an adjunct element in rituals as well as record keeping. The historian of Mesoamerican papermaking Alan Sandstrom surmises that “in addition to precious stones and metals, feathers, and decorative clothing, one of the most sought-after items in ancient Mexico was handmade paper” (Sandstrom and Sandstrom, Traditional Papermaking 7). Further, the historical records we have suggest that paper manufacture and use were widespread throughout the region by the time the Spaniards arrived in 1519. In particular, Maya texts written in Roman characters after the Spanish Conquest reveal the extent of the depth of Maya book making:

Lo más conocidos son los libros de Chilam Balam, en los que se consignaron asuntos diversos. En varios pasajes se hace referencia a los códices, por lo que es claro que que los manuscritos fueron copiados directamente de los libros jeroglíficos. [The best known of the post-conquest books are the books of the Chilam Balam, in which diverse subjects are covered. In several passages there are reference to codices, because of this it is clear that that the manuscripts were copied directly from hieroglyphic books.] (Santos 37)

This suggests that the invention of paper and of codices may have been ancient. “Most authorities are convinced that the Mayas of the Yucatan were the first to produce paper” (Sandstrom and Sandstrom 13) on the American continent. Paper was often part of the cache of tribute given to appease a conquering ruler, as we shall see later.  
1519: The Spanish Conquest

The primary reason why book historians must rely on evidence other than Maya codices in order to learn about their manufacture, structure, and use is that there are very few of them left. When the Spanish landed in central America in 1519, they established a missionary state which sought to spread Christianity among the Maya. Although a few Spanish missionaries attempted to learn the Maya language and glyphic writing system, most of the occupying religious leaders elected to burn as heretical most Maya codices, sending back only a few specimens to their superiors in Europe, of which small number even more codices were lost when ships were attacked and the books thrown overboard as valueless (von Hagen 34). Further, most traditional Maya practices, including codex-making, were quickly overtaken by the Spanish:

Pre-Columbian “books” had no influence on printed books in Mexico after the Conquest of 1519-1521. In the early colonial period the native manuscript art was useful to the Spanish for a while and aided exploration, government, and conversion of the Indians to Christianity. But by 1600 Indian painting (writing) was very much acculturated and almost completely eclipsed by the widespread use of alphabetic script. Only a few survivals appear in later centuries, primarily in rural areas, and in folk art. . . . Arriving in Mexico in 1528 [Fray Juan] Zumárraga immediately saw the need for libraries and a printing press to aid in converting the Indians. (Robertson)

Most scholars concur that the arrival of the Spanish precipitated a catastrophically rapid devolution and destruction of Maya technology, literature, and culture:

Prior to the arrival of the Spanish, native Mexican cultures enjoyed a rich technological history. These various cultures made paper, books, and pigments, painted murals and manuscripts, erected elaborate sculpture and architecture, cultivated crops and botanical gardens, and engaged in medical practice. The Mexican Early Colonial Period (ca. 1521-1600) immediately followed the Spanish invasion of Mexico. It was a time of rapid change during which the outward signs and formal apparatus of Mexican native societies disappeared before the proselytizing energy of Catholicism and the superior technology of Renaissance Europe. Christianity quickly supplanted the various indigenous Mexican religions, and within a short time most native technologies either assimilated European technologies or were supplanted by them. Native architecture and sculpture were rapidly and systematically destroyed by the Spanish in an attempt to eliminate any connection with Mexico’s “pagan past.” Only in the fragile medium of the manuscript painter did the vocabulary of form and artistic conventions of the old civilization and proscribed religion survive, changing continually under the impact of the new European ways. (Haude)

A few missionaries did attempt to create a syllabary of Spanish-language equivalents to some Maya glyphs; some also recorded in passing the appearance of Maya codices and the methods of their manufacture, but such records are scant and reveal little of substance about the Maya codex as a book form. The result of the combination of scarce recording of Maya bookmaking processes and the destruction of the vast majority of Maya codices is that very few codices escaped, none wholly intact. This loss is comparable to the burning and looting of the Alexandrian Library: the majority of the record of an entire civilization was erased:

Although pre-Colombian libraries contained thousands of these handmade books, few escaped burning by Spanish priests and soldiers. Today, only 22 original codex books remain, along with about 50 other volumes written shortly after the conquest. (Krayna 92)

Of these twenty-two volumes, only four are of Maya origin, and it is from only these four codices, named for the cities or institutions where they were originally housed (Dresden, Paris, Madrid, and Grolier), that present-day book historians can derive data about the content, manufacture, structure, and materials used in Maya codices. Because of their rarity, scientific access to these examples is extremely difficult. For example, the last microscopic analysis of any of the codices took place in 1910. Fortunately, these partial codices, of which none is a complete specimen and of which none has its original covers, are not the only kinds of evidence scholars can use in order to reconstruct the Maya codex.  
The 1940s: Von Hagen and Lenz

During the 1940s, two scholars, Victor von Hagen and Hans Lenz, conducted independently of each other ethnological and anthropological studies into the modern-day descendants of the Maya, many of whom at that time still spoke Quechua, though to be descended from the Maya language. Both von Hagen and Lenz were able to observe the Otomi Indians making handmade huun paper. From the accounts given by these two scholars, we get a description of the nature and function of paper up to the arrival of the Spanish. Von Hagen links the development of paper among the Maya to similar development around the world at nearly the same time:

As a corollary to writing, a smooth surface for writing had to be perfected. Thus, in some remote epoch of Maya civilization, bark-cloth tunics left the backs of the people and became “paper.” This “paper,” superior in texture, durability, and plasticity to Egyptian papyrus, was thus perfected anonymously and communally by the Maya. And this writing surface they called huun. . . . Paper aided the endless unraveling of the intricacies of the Maya calendar and helped develop Maya writing. . . . In time paper became one of the transmissional agents of Maya civilization and, by aiding the continuity from on epoch to another, gave substance to oral traditions. The Maya accumulated books, as man had done elsewhere--in China, in Egypt, in Rome, in Greece. These books, and there were actually books, were housed and protected down through the centuries. (von Hagen, “Paper and Civilization” 302)

Likewise, Lenz prefaces his ethnographic work with a nod to the Spanish missionary Peter Martyr, whose records of the contents of Maya paper help present-day scholars to link the practices of the Otomi Indians to those of their Maya ancestors:

Pedro Mártir de Anglería fue el primer cronista que hizo una descripción del papel americano: “En lo que ellos escriben son unas hojas de cierta delgada corteza interior de los árboles que se cría debajo de la corteza superior; creo que se llama philira confrme lo vemos, no en el sauce u olmo, sino en la de dos palmitos que se comen, que hay una tela dura que separa las hojas exteriores a modo de redes con agujeros y mallas estrechas, y las embetunan con unto fuerte. Cuando están blands, les dan la forma que quieren y las extienden a su arbitrio, se supone que con yeso o con alguna materia parecida.” [Peter Martyr was the first historian who made a description of American paper: “In that which they write there are leaves of the thin inner bark of the trees that grow underneath the outer bark; I believe it is called philira. We see it, not in the willow or elm tree, but in the one of two small edible palms, that is, a durable fabric that separates the outer leaves, as a networks with holes and narrow meshes, and they solidify the pages with strong grease. When they are whitened, the Indians give the form which they want to them and they use them as they will, I assume, coated with plaster or some similar substance."] (Lenz 71-72)

This detailed description of the process of Maya papermaking has been rather loosely updated into modern Spanish by Lenz, but the main points of papermaking remain strikingly similar to those practiced by the Otomi Indians in the 1940s. Although von Hagen and Lenz observed the papermaking processes of the Otomi and postulated about the manufacture and role of paper in Maya Postclassic period civilization, their research did not examine the Maya codex as an artifact: as a book qua book. Only recently has scholarship sought to examine the Maya codex in the same way that book historians and descriptive bibliographers currently study European handmade and early-press books.  
1980-2000: The Codex as Book Form

Only recently has scholarship focused on the surviving four Maya codices as artifacts, and the extent of such scholarship has been limited mostly to physical description. The leaves of the codices are small compared to present-day standards of page dimensions (Letter or A4 paper), and are connected in screenfold manner: one long piece of paper folded accordion-wise creates many pages on both sides of the paper. The long sheet of paper is usually formed of two or more layers of smaller paper sheets which have been glued together to create the long sheets needed to create a codex. Indeed, tribute records indicate that paper was not usually turned into codices right away, but was bundled for transport into standard-sized packets, and only glued or felted together when a codex was being made. The dimensions of the surviving codices indicate the large amount of work involved in creating a codex:

Even in their present somewhat fragmentary states, the surviving books are of impressive length when spread out flat. The Dresden is now 20.5 cm. (86 in.) high and 3.56 m. (3 ft. 8.4 in.) long; since both sides of each leaf had been prepared, the scribe, or scribal team, was faced with 74 pages totaling at least 2,268 inches of unpainted, white surface. But this is nothing compared with the Madrid: 23 cm. (9.6 in.) high, it measures no less than 6.82 m. (22 ft. 6 in.) in length, offering its scribe 112 pages covering slightly under 5,000 inches of available surface. The dimensions of the double-sided leaves--and hence the individual pages--resulting from the creasing process vary somewhat from screenfold to screenfold. (They also vary from leaf to leaf, to a greater or lesser extent, because of wear.) Although it is demonstrably extremely late pre-Conquest in date, based on its content, the Dresden Codex probably most closely resembles the books of the Classic period; its pages are only 9 cm. wide, in contrast to the 12.2 cm. of the Madrid, and the 12.5 cm of the Paris and Grolier. The small, calligraphically elegant writing of the Dresden is in perfect harmony with the comparatively small size of its pages. (Coe 171)

In this project, my aim is to create a codex of only three or four folds, both for reasons of economy of materials and of text: I have comparatively little bark from which to make paper, and I have comparatively little to say using Maya glyphs and pictographs. The original codices we have today are almost certainly copies of earlier works that had either deteriorated or were valuable enough to merit preservation reformatting. This suggests that the development of the codex may be more ancient than archaeological evidence suggests. Further, the Maya codex as a format for the transmission of information, like the papyrus scroll and the codex in Europe and the Middle East, seems to have existed relatively unchanged throughout its history because it was an efficient means of recording, preserving, and disseminating information. That the Maya thought enough of the texts that survive to have copied them from earlier documents that were in poor condition speaks to the value placed on information in Maya culture:

We should now reconsider why no economic codices survived. Only four ancient Maya books remain of the hundreds that still existed in Postclassic times. These four were devoted to religious and astronomical data, and are considered in more than one case to be copies of earlier, Classic Period originals. Why would later Maya scribes and rulers want to recopy codices containing royal genealogies, or tribute records, or kingdoms that had long since perished? (Fash 194)

This question can be answered by looking not only at the Maya codices, but by including the records found on Maya architecture and pottery, as discussed above. The data we currently have about Maya codices suggest that although very solid connections have been made between the papermaking skills of the Otomi and those of the ancient Maya, few if any researchers have attempted to recreate a Maya codex from scratch, as it were. Western scholars have tended to focus on the codices as texts, privileging the information to be found on their pages over the codices as artifacts themselves. There seems not to be a similar disjuncture with regard to Maya architecture, where the structure and materials of construction are given as much attention as the glyphs carved into buildings, nor is this disjuncture seen in the study of Maya pottery, where the form and function of vessels are studied by linking the writing on them to their structures and intended uses.

This separation of epigraphic data from physical format is puzzling, given the many breakthroughs obtained in other areas of Maya scholarship when, for example, a scholar guesses that a vase could be a ceremonial drinking vessel, and a character-set in the Primary Standard Sequence is found to translate to something like “One Jaguar, his drinking bowl, made in such and such a year in such and such a place.” Because the Maya named nearly everything, tagging bowls, cities, rulers, stars, gravesites, and tax records as having unique names, it stands to reason that there may eventually be found a similar correlation between the texts of Maya codices and their format. Thus, it seems important to be able to reproduce to the best of our ability a “real” Maya codex.

On 13 February 2001, the program “Lost Kings of the Maya” aired on PBS’s Nova. One segment showed David Stuart of Harvard University, one of the foremost epigraphers of Maya glyphs, holding what appeared to be a reproduction of a Maya codex:

We’ve had a lot of lucky breaks along the way. One of them is that we know the language in which it’s written; a lot of Mayan people still live today speaking Mayan languages. And also, one of the clues that was sort of like a Rosetta Stone for us, was the discovery of a manuscript by a Spanish priest named [Fray Diego de] Landa, who actually wrote down a little about the hieroglyphic system and the way it was structured. He didn’t really understand it, but scholars about 50 years ago began working on it and picking up clues that really told us the way the writing worked and the real structure that allowed us to crack the code. (“Tour Copán with David Stuart”)

In a telephone conversation, Stuart subsequently revealed that the codex in the NOVA episode was a facsimile of the Dresden Codex, created using a printing press and European paper stock. Dr. Stuart said that although scholars like Alan Sandstrom have documented the papermaking techniques of the Maya, relying on the huun-making techniques of modern-day Mexican Indians, few, if any, recreations of Maya codices have been produced using the methods and materials of the Maya (Stuart).

Complicating the task of reproduction is the fact that we have only four extant examples of Maya codices, and each of these examples is off-limits to invasive examination, such as microscopy and chemical analysis, as well as most physical examination, due to the rarity and fragility of the surviving codices. Even the leading scholars in the field, such as Michael Coe, William Fash, and David Stuart, have not been able to examine the objects of their research. For example, the scholarly knowledge of the content of the gesso on the surface of the codices, which I’ll examine later, is the result of speculation based on other writing surfaces from the present day or from contemporaneous but geographically disparate cultures.

A final problem in defining the Maya method of making codices is that the codices recently discovered in archaeological digs have been so badly decomposed or altered by the conditions of their burial that we cannot study them:

Among the Zoque, immediate western neighbors of the Maya in Tabasco and Chiapas and unrelated linguistically to this latter group, two ancient codices were recovered as part of the mortuary offerings accompanying two elite male burials at the site of Mirador, in the municipality of Jiquipilas in Chiapas and dating to the Early Classic period, around A.D. 450. A most disturbing fact is that these two ancient books have been so badly preserved that they cannot be opened. One was apparently made on bark paper, the inner bark of the wild fig tree called amate locally, while the other may have been painted on deerskin. Both were sized with a white lime background. The organic portions of the books have decomposed and the lime sizing has recrystallized in such a way that the individual crystals cross between the layers of fan-folded pages and literally nail them all together. In spite of numerous attempts to separate the pages and study them, none has been successful. The oldest Maya codices known also date to the pre-Hispanic period and have been found by archaeologists as mortuary offerings with burials in excavations in Uaxactún, San Agustín Acasaguastlán, and Nebaj in Guatemala; at Altun Ha in Belize; and at Copán in Honduras. The six examples of Maya books discovered in excavations date to the Early Classic (Uaxactún and Altun Ha), Late Classic (Nebaj, Copán), and Early Postclassic (San Agustin Acasaguastlán) periods and, unfortunately, all have been changed by the pressure and humidity in the ground during their many years in the ground, eliminating the organic backing and reducing all into unopenable masses or collections of very small flakes and bits of the original lime sizing and multicolor painting. The result being, unfortunately, more old books which will probably never be read. (Whiting 207-208)

Poor conditions similar to those related about the Zoque codices were found in Maya scribal tombs, as well:

An elite tomb recently was found inside Structure 10L-26 at Copán. This vaulted tomb contained sumptuous offerings of ceramic sculpture and pottery vessels, carved bone objects and jadeite jewelry, a desiccated codex and containers of pigment. Together, these identify the burial as that of an artist/scribe. (Reents-Budet, Painting the Maya Universe 57)

The fact that any Maya codices survive today in useful condition at all seems to be almost miraculous, given the legion of forces which have destroyed or rendered unusable all but four codices. I use the word “miraculous” advisedly, as my next section deals with the spiritual aspect of the creation of a Maya codex.

In order to create a codex, it may be helpful first to be familiar with some of the uses to which codices were put, and to enter as much as possible into the mindset of the Maya scribe. To do so, we must familiarize ourselves with a few deities and ritual practices.  
Ritual Preparations

In order to approximate the process by which the Maya made paper, I must briefly digress here in order to introduce the papermakers themselves. Although it is not clear who among Maya society were tasked with making paper, it is certain that only scribes were privileged to make codices. The Maya scribe was a member of the educated aristocracy and was part of the religious and social economy of Maya culture:

Among the post-Classic Maya of Yucatan, the artist/scribe (the ah ts’ib) was a member of the elite stratum of Maya society. He was highly educated in the multiplicity of subjects preserved in the pre-Colombian manuscripts, including mathematics, astronomy, astrology, cosmology, and human history. (Reents-Budet, Painting the Maya Universe 65)

Before a scribe could perform his duties, he needed to pray to and to consult the gods about the most auspicious times for every step of the process of codex creation. The masculine pronoun is not, in this case, gender inclusive. Scribes are almost always depicted as men, and their glyphs translate to “he,” “his,” and “him,” although there exists very rarely evidence of high-ranking court ladies who are described by the scribal titles “she of the writing” and “she of the inkpot” (Coe 74ff).

Three Maya deities will serve as our patrons during the process of papermaking and codex construction. I have significantly ignored an important god, the Young Maize God, who is often depicted on Maya pottery as a scribe writing in a codex, but this is due more to personal affection for our patron gods than to any reasoned scholarly decision. We will now place ourselves under the guidance of Itsamná and the Monkey-Man twins Hun Batz and Hun Chouen.

According to ethnohistoric sources, Itsamná was the inventor of writing as well as the patron of the priesthood. . . . His role as a scribe is made clear at the Terminal Classic site of Xcalumkin, where he bears the title of ah ts’ib, “he of the writing,” and there are several images of Itsamná as scribe in the Madrid codex. . . . Another aged god, Pawahtún (God N), is far more widespread than Itsamná on Maya ceramics dealing with scribes. . . . Iconographically, Pawahtún is distinguished by his aged features and by his floppy, netted headdress. (Coe 102, 104)

Figure 1 shows Itsamná or Pawahtún in the upper-right corner discoursing with a monkey-man and a scribe on the left. The monkey-man holds a codex, and the scribe behind him has a “printout” roll of paper with glyphic numerals on it issuing from underneath his arm and a conch-shell inkpot in his hand. The bottom cover of the codex held by the monkey-man seems to be detached from the sheets of paper (this will be important to remember later).

Figure 1 (K0501). © Justin Kerr.

The most striking of the lesser gods consist of a pair of beings with human bodies but simian or part-simian heads. These are the Monkey-men, . . . Hun Batz and Hun Chouen--both calendrical names signifying the date 1 Monkey in the 260-day count; they are the malevolent half-brothers of the Hero Twins, turned by them into monkeys. Scribes par excellence, they are often shown on Codex-style and other ceramics actually writing in codices, or reading from them, and are frequently pictured with the Spangled Turban headdress into which pens and other writing instruments are thrust. (Coe 106)

Figure 2 shows a rollout photograph of a vase depicting Hun Batz and Hun Chouen recording tribute from members of the animal or spirit world. Figure 3 shows Michael Coe’s drawing of the monkey-man gods in Figure 2.

Figure 2 (K3413). © Justin Kerr.

Figure 3. (Coe 106)

Another anthropomorphic figure is the Rabbit Scribe. Human scribes with animal heads are often depicted on Maya pottery: there are examples of scribes with vulture, deer, and monkey heads on various vessels. However, the Rabbit Scribe is not anthropomorphic, but is depicted simply as a rabbit holding a brush pen.

In a Classic Maya funerary vase from northern Guatemala, now in the Princeton University Art Museum, the rabbit was made into a scribal god! Among the Maya, divine scribes usually are monkeys. It is not known if there were any rabbit scribe gods among the Maya. The scene on the vase describes the ghastly rituals in the underworld. It may even be an illustrated chapter from the adventures of the Hero Twins (as recorded in the Quiché Maya holy book Popol Vuh). The rabbit sits below an important old god, perhaps diligently recording the human sacrifice happening in front of the old god in a jaguar-pelt bound codex. (Lo)

A line-drawing rendering of a detail of this vase is shown in Figure 4 below:

Figure 4. Detail of codex-style vase. © Michael Coe.

The Rabbit Scribe is also described in terms of the puzzle of how Maya codices were used:

Several surviving Maya pottery vessels bear images of scribes with their codices. In one scene, a rabbit portrayed painting in an open codex bound in jaguar skin in the palace of a Maya lord. . . . The social context in which books were used is still elusive. Scribes held an elevated position and their books were entombed in lavish crypts; yet beyond this, scant detail comes to the art historian or archaeologist about how the books were used. (Love 3)

We can, however, makes some educated guesses about the manner in which codices were used. Each of the scribal gods and patrons described above was consulted in turn by the scribes to determine the auspicious times for engaging in war, planting crops, holding rituals, and even preparing and copying books. As an aside, the recent animated film The Road to El Dorado, although it depicts an obvious mish-mash of Aztec, Maya, and Toltec societies, shows in one scene the villain of the film, a jaguar-hide-wearing scribal lord, consulting screenfold codices to determine the auspices of the arrival of the European heroes of the film, the planning of a feast, and the proper time for destroying the heroes by sending them to Xibalba (the Maya underworld) by pushing them into a giant whirlpool.

In order to avoid the possible pitfalls of undertaking my experiment at an unlucky time, I settled on reverential supplication as the surest possible means of attaining a proper attitude. Although I did not know how to consult directly with Hun Batz and Hun Chouen in order to determine the most favorable time for creating my codex, I offered up a prayer and a sacrifice (a chicken dinner with rice and new potatoes) to them, and began the process of making the paper for the codex.  
The Recipe for Paper

Our knowledge of how the Maya made paper is based on three indirect methods of observation. First, we have the records the Maya left behind, few of which survive in any detail sufficient to allow us to reconstruct their recipes or methodologies with any certainty. Although four codices survive, apart from a visual inspection, very little chemical or microscopic analysis has been performed on them, and no such access is likely to be granted in the near future. Second, we have the eyewitness accounts of the Spanish, among the papers of whom we find accounts of the daily lives of the Maya, giving us insight into their dress, government, agricultural practices, and, in a very short passage, their papermaking techniques.

It is not known when or how the Maya process of making paper began, nor is it clear how it evolved over time. We find ourselves in the position of deducing from Spanish records nearly an entire millennium of papermaking history among the Maya. Last, we have the evidence of present-day papermaking practices among the descendants of the Maya. Although, as we have seen, the ethnographic research of von Hagen and Lenz provide ample evidence of how paper is made today, we must be cautious about adducing from such practices that the Maya did or used exactly, or even nearly, the same processes and materials as present-day Indians. That said, the findings of scholars as to the composition of Maya paper are based on the best information available:

At the beginning of the twentieth century, most scholars thought that amate, the paper used by the Pre-Columbian peoples of Mesoamerica to make books (including the Dresden Codex), maps, and the like was manufactured from maguey fiber. It is now known that this is completely erroneous: all surviving examples of pre-Conquest paper were made from the inner bark of one or more species of Ficus, a wild fig of the order Moraceae. In the case of the Dresden Codex, the crucial analysis was done by Dr. Rudolph Schwede in 1910; through microscopic analysis of some of the fibers of the paper supporting the surface of the book, he found these to be identical with specimens from living fig trees. Later he had the opportunity to study samples from the Madrid and Paris codices, as well as pre-Conquest documents from central Mexico, and came to the same conclusion about them. As for the fourth Maya book, the Grolier Codex, the present writer was able to make a close examination in 1972, and there is no doubt in his mind that the primary material is also processed inner bark from a species of wild fig.

In the Maya lowlands the tree in question is Ficus cotinifolia, to which the Yucatec Maya gave the name of hu’un or hun, a name which they also applied to paper and to the books made from it (amate is a Mexican term derived from the Nahuatl amatl). Although we have no information on how the Maya themselves turned tree bark into paper, native Ficus paper from another species is still manufactured by the Otomi in central Mexico and by various Nahuatl-speaking villages in Guerrero; and there is also an account by the royal botanist Francisco Hernandez, writing in the late sixteenth century, who describes paper-making as it was practiced by the Colonial-period descendants of the Aztec. There is no reason to believe that Maya paper- manufacture was different from that of non-Maya Mesoamerica. (Coe 144)

Coe’s summary of the chain of evidence linking the Maya codices to the papermaking practices of the present day shows us that although the link between the two practices cannot be certain, there is enough evidence to suggest that we come closest to the methods and materials of the ancient Maya when we study and employ the techniques of the Otomi and other central-American Indian tribes. Coe continues by delineating the steps in making paper:

In contemporary Mexican villages, the entire process of making paper from a tree involves six steps:

  1. Fig branches over 1.5 m. (5 ft.) in length and about 25 mm. (1 in.) in diameter are harvested.
  2. The freshly cut branches are slit lengthwise, and the bark (outer and inner) is stripped off in one piece.
  3. The outer bark is peeled off from the strips, and the inner bark soaked in running water; the latex (present in all Ficus species) is allowed to coagulate, and scraped off.
  4. Nixtamalization is then carried out on the bark fibers; this is directly comparable to the way dried maize kernels are converted into a hominy-like substance, nixtamal, by alkalizing the kernels in lime mixed with water. In the case of bark paper, the inner bark fibers are first dried and then boiled in a pot containing the same water in which maize kernels had been left to soak, along with more lime or lye from wood ash.
  5. The now pliable bast-fibers are removed from the solution, rinsed in cold water to remove traces of the alkali, and placed in a large gourd for further processing.
  6. The nixtamalized fibers are cut to conform to the dimensions of a flat, wooden drying-board and laid on it in a grid formation, with the first fibers laid lengthwise, and the next laid crosswise-in other words, the constituent fiber layers are at right angles to each other, exactly as in the preparation of papyrus in Egypt. Then, with the goal of felting the loose fibers together, the whole is pounded with a striated beater; this instrument is now totally of wood, but in pre-Columbian times beaters were of stone held in wooden hafts. (Coe 143-144)

More specific than Coe is Marie Van der Meeren, who reports on the recipe used in modern-day San Pablito in the state of Chiapas. Van der Meeren gives approximate cooking times and material amounts, and compares the Maya methods of making paper to the “slight modifications” that present-day papermakers find it necessary to employ:

Cuando comparamos la elaboración de una hoja de papel amate al inicio de la Colonia con el tipo de manufactura que actualmente se lleva a cabo en San Pablito, nos percatamos de que existen algunas modificaciones en cuanto a las técnicas empleadas.

Retomando las fuentes del siglo XVI, vemos que Francisco Hernández, protomédico del rey Felipe II, fue el único que dejó alguna información acerca de la fabricación del papel amate. La observación personal de varies códices resguardados en la Biblioteca Nacional de Antropología e Historia permitió corroborar algunos de sus datos.

Según Francisco Hernández, los hacedores de papel cortaban únicamente las ramas gruesas de los árboles, dejando los renuevos. En seguida, se dejaba reblandecer las ramas en los ríos o arroyos cercanos durante una noche. Al día siguiente, se arrancaba la corteza de la rama y se separaba la corteza externa de la interna para guardar únicamente esta ú1tima.

Una vez limpia la corteza, se extendían los segmentos de fibras sobre una superficie plana y se golpeaban las fibras con un machacador de piedra estriado hasta obtener una hoja. Posteriormente, se golpeaba doe nuevo con otra piedra sin estrías para dejar liso el papel. Las necesidades definían la unión del varias hojas para obtener el tamaño deseado.

Actualmente, los fabricantes de papel de San Pablito compran sus fibras al jonotero, quien por lo general es una persona ajena a los habitantes del pueblo. El jonotero se traslada por diferentes puntos de la sierra para recolectar las cortezas. En el lugar, arranca la corteza del árbol y separa la interna de la externa, para posteriormente formar bultos; después lleva a vender su mercancía a los artesanos.

Al mornento de comprar los paquetes do jonote, éste se encuentra todavía húmedo, razón por la cual los artesanos ponen a secar las fibras sobre tendederos antes de almacenarlas para evitar que se pudra el material.

[When we compare the processing of a sheet of amate paper at the beginning of the Colony with the type of manufacture that at the moment is carried out in San Pablito, we notice some modifications as far as the techniques used.

Rereading the sources of the sixteenth century, we see that Francisco Hernandez, protomédico of King Felipe II, was the only one who recorded some information about the manufacture of amate paper. The personal observation of various codices protected in the National Library of Anthropology and History allowed me to corroborate some of his data.

According to Francisco Hernandez, the papermakers cut only the heavy branches of the trees, leaving the sprouts. Then, the branches were softened in the rivers or streams overnight. On the following day, the bark of the branch was removed, and the external surface of the internal bark was separated.

Once clean of bark, the fiber segments were extended on a flat surface and the fibers mashed with a fluted stone crusher until a leaf of paper was obtained. Later, the paper was beaten anew with another unstriated stone to smooth the paper. This necessitated joining several sheets to obtain the desired size of paper.

At the moment, the manufacturers of paper of San Pablito buy their fibers from the jonotero, or jonote-maker, generally a person known to the inhabitants of the town. The jonotero travels to different points along the mountain range to collect bark. In his shop, he takes the bark of the tree and separates the internal one from the external, later to form bundles; later he takes these bundles (“jonotes”) to sell to the craftsmen.

When they buy the jonote bundles, they are still humid, so the craftsmen put the fibers on dessicators to dry before storing them to avoid rotting.] (Van der Meeren 72-73)

The primary difference in the methods described by Coe and Van der Meeren is the inclusion or omission of the practice of steeping the fibers in a river overnight in order to raise the latex to the surface, where it may be scraped away before boiling in the nixtamalization process. The most difficult aspect of this codex-making project is not that of knowing the steps one must follow in order to make paper: it lies in finding the ingredients, since ficus cotinifolia and ficus padifolia are both endangered species, and my chance of access to a jaguar hide is, to say the least, slim. While attempting to use as many traditional practices and materials as possible, I am forced by circumstance to substitute plentiful for scarce materials; however, the questions of what kinds of substitutions to make have already been discussed by scholarship. In the next section, I will gather my materials and actually make huun paper.  
Making Paper

In order to make the paper for my codex, I needed some basic materials: fig-tree inner bark, a means of soaking the fibers overnight in running water, a means of boiling the fibers in lime water, a beater with which to felt the fibers, and a smooth board on which to lay out the sheet of paper. In order to create a sheet that would approximate the dimensions of the Maya codex, I relied on archeological descriptions of the extant codices:

Just as with modern books, paper was the most common material out of which codices were made. The Maya made paper from the inner bark of fig trees (Ficus), called kopó in Maya and today commonly known as amate paper. Although they also used deer skin, cotton cloth and maguey paper, apparently the Maya preferred kopó. The paper measured several meters long and, as in the case of the three known Mayan codices, measured about 20 centimeters wide. (Martí)

With these dimensions in mind and my materials gathered, I begin making huun/kopó/amate paper.  

Citlalli López has identified the plants traditionally used for papermaking. In late Classic and early Postclassic Maya culture, the two most common fibers used were the inner bast fibers of two species of fig: Ficus padifolia and Ficus cotinifolia. López also identifies as traditional papermaking plants Ficus pertusa, Ficus calyculata, Ficus goldmanii, and Morus celtidifolia (López). Today, because many of the traditional papermaking plants used by the Maya are threatened by overharvesting, handmade paper is often produced using substitute tree species: Trema micrantha, Ulmus mexicana, Brosimum alicastrum, Sapium pedicelatum, Urera caracasana, and Myriocarpa cordifolia (López).

In the case I had not been able to obtain even a good surrogate for the bark, I was prepared to follow the advice of European-style papermakers:

A large variety of raw fibers can be used for papermaking as well. These fall into three categories: bast or inner bark fibers (such as flax and kozo), leaf fibers (such as abaca and sisal), and grass fibers (such as bamboo and rice straw). Raw fibers often require cooking, retting or fermenting, and beating, which may be done by hand as an alternative to using a stamper or Hollander beater. (“Sources of Fiber and Pulp”)

Thankfully, I was able to obtain a sufficient quantity of bark from one of the species mentioned in López’s research. I must here thank Jeff Polonoli of the Phipps Conservatory and Botanical Gardens in Pittsburgh, PA for his advice and help in locating suitable specimens for this project. Because of the scarcity of the Ficus species used by the Maya, and because the Maya seem to have used whatever fig or rubber plants were to hand in making their paper, I obtained a number of Ficus pertusa branches from Pete Hernandez at the Dural Country Club in Miami. These branches were pruned from existing plants and averaged four feet in length and 1.5 inches in diameter. Figure 5 shows the process of removing the bark from the fig branches.

Figure 5. Pulling the bark.

Because the number of fig branches I was able to obtain did not supply a sufficient quantity of fibers for the project, I supplemented the Ficus fibers with flax fibers in a measure of four parts fig-tree bast fibers to one part flax. The flax fibers were added during the boiling, or nixtamalization, process below.  

It is clear from the descriptions given by von Hagen and Lenz that the freshly-pulled bast fibers are submerged overnight in a river in order to coagulate the natural latex in them. However, having no access to a river or stream, I had to make do with a large tub and a fresh supply of running tap water.

I scraped the bast fibers from the outer bark using a flint-sherd stone scraping tool and then gathered the fibers together and submerged them under a heavy stone in the tub. Figure 6 depicts the separation of the inner bark fibers from the hard outer bark, and Figure 7 shows the fiber-bundle folded and submerged in the running-water bath.

Figure 6. Scraping the bast fibers.

Figure 7. Immersion of the fibers.

I then introduced a hose in one end of the tub to produce the effect of running water. When I returned in twenty-four hours, I discovered a) that my water bill for that month would rise sharply, and b) that this step in the process may not even be necessary. In present-day southern Mexico,

en lugar de reblandecer la fibra en el río, los artesanos ponen a cocer el jonote en grandes cazos. Las labores empiezan con la preparación del agua de cocción, a la que agregan cierta cantidad de ceniza y cal, materiales que ayudan a reblandecer fibra y a eliminar ciertas sustancias. La proporción de ceniza y cal depende de la especie vegetal que se va a preparar. El tiempo de cocción puede ir de tres a seis horas, dependiendo del tipo de fibra. Durante la cocción, su mueven constantamente las fibras para controlar su cocimiento. Una vez cocidas, se enjuagan abundantemente y se dejan reposar en recipientes con agua.

[Instead of softening the fiber in the river, the craftsmen cook the jonote in great dippers. The work begins with cooking water, to which they add certain amounts of ash and lime, materials that help to soften fiber and to eliminate certain substances. The proportion of ash and lime depends on the vegetal species they are going to prepare. The boiling lasts from three to six hours, depending on the type of fiber. During the boiling, they stir the fibers constantly to control their binding. Once cooked, they are rinsed abundantly and are left to soak in containers with water.] (Van der Meeren 72-73)

Whether the soaking of the fibers or boiling them produce the same effect, I have included both in the process of making my codex. One serendipitous advantage of using running tap water for the fiber soaking instead of river water is that the mineral content of my local tap water closely resembles the lime-and-chalk water used in the next step of the process, nixtamalization.  

There is not consensus among scholars about the necessity of either soaking or boiling the plant fibers. Some scholars claim that the Maya simply soaked the fibers in a river, scraped them of their latex, and then began to beat the fibers with felting clubs or stones. Other researchers, like Van der Meeren above, see no reason to soak the fibers at all if they are to be boiled. One of the earliest records of the Otomi papermaking process is Victor von Hagen’s, and he reports that the Otomi both soaked and boiled their paper fibers:

[Papermakers] pull off five-foot long strips from the paper-plants. The outer bark is then removed by peeling, the freshly peeled bark allowed to soak in running water, so that the abundant latex may coagulate and be scraped off. Then. . . the fibers are dried and then boiled in a large kettle filled with nixtamal, the lime-water residue in which kernels of maize have been soaked. After hours of boiling, the softened bast-fibers are taken from the lime-water, washed again in cold water and placed in a large gourd. (von Hagen, The Aztec and Maya Papermakers 57)

I have followed von Hagen, both because it is difficult to determine the actual practices of the ancient Maya based on present-day Otomi Indian practices and because it is impossible to say beforehand with certainty which process, soaking or boiling, may be dispensed with safely. Figure 8 depicts the fibers being removed from the pot after having been boiled over a wood fire.

Figure 8. Boiled fibers.

After I blanched the fibers in lime-water, I rinsed them thoroughly to remove any residue of latex or lime. I also reserved the soot from the bottom of the boiling-pot in order to make black ink, the process of which is described below. Before I can make the ink, however, I need to felt the fibers together to create a contiguous sheet of paper.  

Two sorts of beaters have been found in Maya archaeological digs: the first, and earliest, form of beater is a squat stone with a handle carved into the top surface and striations scored into the bottom. This instrument is also used by the Otomi today in a form little changed since the days of the Maya. The second form of beater is like a war-club, with a rounded handle and a four-faced head, into which striations are carved parallel to the short side on one face. This second form has a greater mechanical advantage than the first, and is thought to be of later origin than the first example. Setting aside the form of the beater, the process by which strips of fiber were felted together is described relatively consistently by researchers:

To fashion paper, only two instruments are necessary: a flat, smooth, wooden surface like a bread-board, and a striated stone called a muinto. . . . Pieces of boiled fiber are now taken, cut to conform to the shape of the board and pounded with a muinto until the strips are felted together to form a continuous surface. Then, the paper, still sticking to the board, is sun-dried. When finished, it presents a smooth surface on the side which faced the board, and a roughened surface over which the muinto was rubbed. (von Hagen, The Aztec and Maya Papermakers 57)

Likewise, Alan Sandstrom gives a historical context to the use of the striated stone with a handle:

Paper was produced from felted mashed vegetable fibers, particularly agave leaves and the inner bark of fig (Ficus) trees. . . . Striated stone beaters were probably used in this process; archaeological examples found on the Pacific coast of Guatemala date as far back as 1000 B.C. (Sandstrom and Sandstrom, “Paper” 583)

I crafted a stone-cum-handle beater by gouging out striations in the bottom of a flat piece of stone. Figure 9 shows the process of beating the strips of fiber, which have been arranged in two layers, one running the length and one the breadth of the work board:

Figure 9. Beating the fibers.

A surprising result of the process of felting the paper is that although I did not use any European-method mold, deckle, or chains in making this sheet of paper, the finished product contains distinctive marks analogous to a watermark: the impressions of the striations of the beater. Scholars have examined many samples of Maya paper, and found striation-marks from a beating stone still evident after many hundreds of years. Thus, we may be able to use these striation marks as sorts of “fingerprints” of various Maya papermaking workshops; such evidence would certainly be visible in an X-ray examination of the extant codices. Unfortunately, scholarly access to the codices is virtually non-existent.  

The final step in making the paper is to allow it to dry in the sun. Michael Coe discusses the various products that can result:

Finally, the material is dried in the sun and peeled off. The resulting paper is smooth on the board side, and rough on the other. Hernandez tells us that polishing was carried out with smooth stones, which were probably heated; more will be said about them in the context of scribal tools. The color of the paper varies: maize water with lime produces yellow paper, while woodash lye results in whiter fibers. The paper in the two codices that I have examined personally varies from a nondescript grey in the case of the Dresden Codex--rather like cheap cardboard--to a light to brownish tan in the case of the Grolier Codex, but that may be a result of the aging process. (Coe 144)

I could find no discussion in the scholarship on the extant Maya codices as to whether the paper on which they are painted is smooth on only one or both sides. It seems a simple matter to create paper smooth on both sides by pressing the felted sheets between two smooth boards, although such a process would greatly increase the drying time, and might have an adverse effect on the shrinkage of the sheet as it dries. I opted to follow Coe’s suggestions, and laid out the board and felted sheet on a warm, sunny day.

Figure 10. Sheet of paper drying.

After eight hours of drying, the sheet lifted easily from the board, and I trimmed the sheet to conform to the dimensions closest to those of the four extant codices. My paper came out a light yellow, as Coe predicts.  

Now that the paper has been produced, the next step in preparing it for use in a codex is to bend the paper into screenfold sheets. In order to make each page the same size as its neighbors, I had little in the scholarship to guide me:

Of course, the coated paper which was to make up a Maya screenfold book or hun had to be divided into pages which would neatly fold, and that is a part of the operation that is not yet fully understood. But once this was done, the scribe--or, more likely, consortium of scribes--would lay out the network of light red gridlines to which all of the text and all the pictures were to conform. (Coe 145)

However, the extant codices show remarkable consistency in the width of their pages, suggesting that the Maya had some kind of measuring device, like a ruler, that would allow them to mark off the precise widths of the pages. However, no marks are seen on the codices near or on the fold lines. This suggests a second possibility, one which I adopted in creating uniform pages: a template. I fashioned a block of inch-thick wood that was five inches wide and fourteen inches long, making certain that the sides were square. Using this as my template, I began folding the paper by placing the block atop the paper and folding upwards. This produced a clean, crisp fold without tearing the paper. Figure 11 depicts this process:

Figure 11. Creasing the screenfold with a template.

Once the screenfold had been created, the next step was to prepare the surface of the codex for painting.  

The Maya did not write directly on the surface of their codices; I have already pointed out the extremely porous and rough nature of the paper resulting from the methods employed by the Otomi Indians. In order to create a suitable surface on which to draw and write, the Maya coated their screenfolds with a sort of gesso underlayment composed of chalk and a thickening medium such as sap or animal fat. Although I will call this substance “gesso” throughout this essay, it is really a mixture of ingredients that sometimes approaches plaster, sometimes gesso, and sometimes sizing. The chemical composition of the coating on at least one extant codex has been determined:

On the surface of all four Maya codices, on both sides, is a fine white coating of what is either plaster or gesso, or a mixture of both. On examining the Dresden, Schwede concluded that it was a form of calcium carbonate, which would make it similar to the plaster surfaces upon which Maya artists painted their murals. In contrast to the Maya bark-paper codices, the surviving pre-Conquest books of non-Maya Mesoamerica were made of leather, but were also coated with a fine white layer as a paint base; tests on the Selden Codex, a Mixtec manuscript in Oxford’s Bodleian Library, showed this substance to be a mixture of calcium sulphate (gesso), calcium carbonate, and animal glue. (Coe 144)

Beatriz Martí confirms Coe’s assessment of the necessity of limning the pages before writing on them:

The large codices were folded like screens, covered with [a] layer of starch, and then with a thin, white, calcium carbonate paste. Each page was separated by a thick, red frame and then horizontal and vertical lines were painted to further separate texts. (Martí)

In order to create my own backing material, I combined one part beef tallow to five parts calcium carbonate. The result was a suitably sticky thick material which I applied to four of my screenfold pages using a large brush of decidedly non-Maya origin. I applied two thin layers of gesso, allowing the first layer to dry overnight before applying the second layer. The resulting pages were a dull dun color, and showed the groove-marks of my brush strokes quite clearly.

Figure 12. Applying the gesso.

Coe makes an interesting point about the suitability of plain bark paper to support the writing tools of the time:

What this comes down to is that the scribes actually never wrote directly on paper at all, but on miniature mural surfaces laid over that paper: uncoated bark paper would simply have been too rough and porous for their delicate calligraphy, and for the instruments they were using. (Coe 145)

Now that the pages are coated with gesso, the next step is either to polish the gesso or to move directly to assembling the codex and its covers. I will discuss the process of assembly momentarily, after I have examined the possibility that the Maya polished their pages before writing on them.  

Michael Coe puts forward an interesting hypothesis about the necessity of the scribe to polish the surface of the gesso in order to obtain as smooth and impermeable a surface as possible before writing on it. Coe’s speculations are based on paper-polishing practices among early papermaking cultures worldwide, and on what he admits is unclear evidence in some Maya pottery:

On a Codex-style vase . . . where three lesser Monkey-men pay homage to an enthroned Monkey-man God, each holds out to him in one hand a small ball-like object; could these be polishing stones? In another court scene, the principal figure--perhaps the Young Maize God--opens a codex, while before him sit two Monkey-man Gods, one bearing two small, round objects, and the other holding up an offering of paper tied with a band. Similarly suggestive is a scene on a Nahua-style vase with two identical ah k’u hun scribes, each holding a stack of bound paper; atop each stack is what I believe to be a “stick bundle,” i.e. a bundle of quill pens decorated with quetzal feathers; and on top of one of these bundles is a further object which I feel sure is a round, stone paper-polisher (there can be no doubt that these personages are scribes, as the glyph collocation ts’ib, “writing,” appears next to them).

If these really are paper-polishers, then they are quite common in scribal situations on Classic pottery. For example, on one polychrome vase a round object with a dot in its center appears atop a cut conchshell inkpot being held by a supernatural who appears to be Hunahpu; and on another Hunahpu and a Monkey-man God hold inkpots with the same objects on them. A search of the ceramic corpus would probably turn up many similar scenes.

While I acknowledge that depictions of alleged paper-polishers actually in use are absent in the record, I feel that that the association with known scribal paraphernalia, including paper, is sufficient to establish at least a circumstantial case for their function. It is also a possibility that the instruments with curved ends that appear in scribal headdresses were really paper-polishers rather than tools for carving clay; but at present there is no way to determine their true function. (Coe 152)

In order to test Coe’s theory, I prepared only two of the four gesso-treated pages of my codex to a steady rubbing with a smooth heated stone, working with firm pressure as though I were waxing a car, making small overlapping circles with the stone from the center of the page out toward the edges. The polishing resulted in two significant advantages that could be seen and felt after only a few minutes of polishing. First, the surface of the gesso became noticeably smoother and more consistent in thickness. Further, as I continued polishing the gesso, through the stone I could feel the grains of the gesso penetrating the fibers of the paper, and on examining the reverse of the sheet, I found that some gesso had penetrated through the sheet of paper entirely. I can imagine that in applying gesso to both sides of the paper, polishing would integrate the gesso with the paper to synthesize a hard, stable page on which to write.

Figure 13. Polishing the gesso.

I also noted that with regard to the gesso-treated pages which I did not polish, after a few days the gesso tended to flake off if the page was handled roughly, whereas the pages which have been polished stood up to much more vigorous handling. Despite the fact that the cultures of the world cited by Coe as using paper-polishing stones used them in order to burnish the paper itself, my evidence leads me to suspect that Coe’s hypothesis about Maya paper polishing may indeed be correct.  
Assembling the Codex

With the paper prepared for painting, the final step in creating the codex is to attach covers to it, primarily to protect the screenfold pages from becoming soiled, as Maya scribes are usually depicted as sitting cross-legged on the floor or ground when they write in codices.

As an aside, I may point out that since my demonstration codex has very few screenfold pages in it, there was no necessity of gluing together many sheets to produce a long screenfold. Michael Coe suggests that long screenfolds were produced by gluing together two layers of paper:

The various steps in the manufacture of a Maya screenfold book have been touched upon in the previous chapter. The process began with the making of bark paper. Once each dried sheet had been peeled from the board on which it had been beaten out, it was necessary to attach it to others like it. Each sheet had to be fixed on top of another sheet to produce book leaves of the necessary thickness, either by felting or gluing (or both together). And each sheet had also to be attached horizontally to its neighbor until there were enough sheets to make a continuous length suitable for creasing into the leaves that were going to make up the accordion-like screenfold. It is not clear which of these processes came first, but the fact that Spanish paper has been incorporated into two leaves of the Madrid Codex suggests that horizontal attachment preceded the thickening operation.

Next, the leaves of the screenfold were formed by creasing, perhaps with a wooden straight-edge or something like a weaving batten. This had to be carried out before the plaster or gesso sizing or coating was applied, otherwise that surface would crack or flake off when subjected to folding. How the coating was applied, and its exact composition, will remain a mystery until modern laboratory analyses of the extant codices can be carried out. Presumably, after the entire blank codex had been completed by specialists in these matters, and suitably polished, it was ready for inscribing by the ah ts’ib--the scribe or scribes. (Coe 171)

The final phase of the process of creating the codex is to attach the covers. This presents unique problems for this experiment because we do not know how the covers were constructed, and how they were attached to the screenfold sheets, if at all.  

The next step in creating a codex is to bind the screenfold paper between covers. One of the subjects on which the collective record is silent is the form of the covers of Maya codices. Since no codices seem to have survived with their covers intact, it is a matter of conjecture as to how--or if--the covers were attached to the screenfold paper in order to create a codex. However, as a start, we can rule out a number of methods.

Because the paper sheets are not sewn together to make leaves or sewn at the edges to make what the Chinese refer to as a “butterfly book,” the notion of the codex’s “cover” perhaps needs to be rethought. Rather than an encompassing contiguous envelope of material which protects the screenfold inside from harm, the covers of Maya codices are separate, one attached to the “front” of the first sheet-fold and one cover attached to the “back” of the last sheet-fold. Indeed, many scholars count as usable pages the beginning and ending leaves of extant screenfold codices: what we might call “end papers” meant, perhaps, to be glued to the covers. This working definition of the cover is a bit problematic, for, depending on the number of folds made, it is possible to create a codex with covers on either the same or opposing sides of the paper. Figure 14 shows the two possible configurations, between which contemporary Maya depictions of codices do not conclusively differentiate.

Figure 14. Two possible codex configurations.

A further lacuna in our knowledge of Maya codex binding deals with the methods used to attach the covers to the screenfold paper. Michael Coe implies that the same sorts of glues used to attach the screenfold sheets together were used to attach the sheets to the covers, but he never follows up on how this might have been accomplished, preferring (rightly, I admit) to base his findings on the extant record:

The thickness of the paper of the Dresden Codex makes it certain that there are two layers which have been felted or glued on top of each other, and the same holds for the Grolier. Moreover, there must have been some kind of glue to fasten together numbers of sheets horizontally to form these codices: the Dresden alone is some 3.5 m. (10 ft.) long, and the Madrid almost double that in length. In central Mexico, analysis has shown that the native Mapa de Quinatzin of 1546 consists of two sheets of bark paper glued together. Francisco Hernandez saw a vegetable glue (interestingly, he calls it “papyrus glue”) in use among papermakers of the Nahua settlement of Tepoztlan to fasten sheets together; this was manufactured from the roots of amatzauhtli, a species of orchid. (Coe 144)

Jack Rau, on the other hand, in his short monograph on The Codex as a Book Form, puts forward a theory about the composition of Maya codex covers, but never mentions the source of his information:

The Maya were a unit of culture, custom and religious concepts apart from the rest of [what is now] Mexico. Their records were set down by the priests in hieroglyphic writing in the codices. These were made of strips of leather, bark paper, and (later) cotton cloth, folded like a screen, each fold forming a page with writing on both sides. Often wood covers protected the volume, and, practical note, one codex still to be seen has an inset of jade to indicate at which end to begin reading. (Rau n. pag.)

Neither Rau nor Coe surmise from the evidence the methods used by the Maya to attach the covers to the codex, so it is time to revisit the pictographic evidence to see if we can pick up any clues. Remember the rabbit-scribe writing in the jaguar-hide bound codex? On closer inspection of the rabbit scribe image (Figure 15), we find a few hints as to the possible configuration of the codex and its covers:

Figure 15. Detail of codex-style vase.

First, the covers themselves are represented as considerably thicker than the screenfold pages, suggesting that they may indeed be made of wood that is covered with precious or durable materials, in this case jaguar hide. The second clue about their construction is the odd shape of the “inside” edge of the covers. The cover represented here as the top cover has rounded edges which terminate in hook- or flap-like structures, and there are two similar hooks or flaps along the edge of the cover which makes contact with the sheet-fold. The bottom cover is even more interesting, because, in addition to exhibiting the hook/flap structures, it is clear that one page of the sheet-fold slips beneath the hooks or flaps on the side of the codex closest to the scribe.

In most depictions of codices, the artist draws the codex sideways: in the example above, the rabbit scribe is seated at the right-hand side of the page, not at the bottom. Why is this pictorial convention followed among Maya artists? One possible explanation is that by reorienting the codex, it can best be shown as a codex, depicting the screenfold structure and the binding to best advantage:

One peculiarity of these Classic images, though, is that the codex is always shown in the “wrong” position: the opened book is depicted in profile, as if it were turned 90 degrees clockwise, so that the ah ts’ib appears to face not its bottom edge but the right edge of the second page. This is, of course, a Maya artistic convention, emphasizing the folding-screen aspect of the book, and thus--along with the jaguar-skin covers--enhancing its identification in the eye of the beholder. (Coe 148)

Figure 16 shows what the open codex might look like if it were shown in its “proper” position in relation to the rabbit scribe. It would appear like a European book with its spine facing the viewer and its cover open, only the “spine” of the codex would show the edges of the screenfold paper:

Figure 16. The codex from Figure 15 seen the “proper” way.

Thus, by turning the depiction of the codex ninety degrees, the Maya artist shows us that this object is indeed an open codex, and not simply a bundle of unbound blank paper sheets, which are shown in Figure 17 as being given in tribute, represented by a drawing very similar to that in Figure 16.

Figure 17 (K1785). © Justin Kerr.

This evidence suggests that the pages of the codex were held to the covers by at least one method, and probably two. We can say with relative certainty that the rabbit scribe’s codex is held together by binding cords of the jaguar hide around the screenfold paper. Each hook or flap can be seen as the side edge of a cord that, untied, would dangle from the sides of the covers, and which would, when tied, help to keep the screenfold paper in place against the wooden cover. This is especially to be seen along the bottom cover, where the bottom page of the screenfold is plainly thrust underneath the hooks or flaps.

Coe and Rau may be correct in assuming that the covers were glued to the screenfolds, but an argument against the gluing of the covers to the paper--and a possible explanation about why to date no codices with covers have been unearthed--is that the covers were not meant to be permanent at all, but acted more like dust jackets do for modern books.

The Spanish burned literally thousands of codices; did all of them have jaguar-skin covers, or even precious stones? Although it is likely that not all codices had valuable covers (see the carved wooden covers between the Monkey-Men Gods in Figures 2 and 3 above), the overriding characteristic of the codex covers is that they seem detachable. Consider the problem of storage, which, given contemporary accounts of huge Maya libraries, necessitates an efficient use of space. Sheet-folds are depicted on Maya architecture and pottery as often without covers as with them, and it is tempting to infer that Maya codices were stored sans covers; covers would be attached temporarily when the codices were in use. This also explains why, despite the fact that the Maya seem to have labeled nearly every object with regard to ownership and use, no codex covers are shown as containing specific “title” glyphs for the codices they contain.

A second difficulty in knowing how to cover a codex is the variety of ways in which codices and their covers are depicted. Figures 18 through 23 show depictions of codices in many different possible forms, from the most common the to the most unique. Most depictions of codices in Maya pottery and architectural carving show the codex as a bundle of screenfold leaves bound top and bottom with animal-skin covers, as in Figure 18, which depicts two scribes, likely the Hero Twins, writing in such codices:

Figure 18 (K1523). © Justin Kerr.

A variation on this manner of codex construction is seen in another Maya vase. In Figure 19, we see three scribes and three codices. Each codex has only one plain bottom cover, and in this instance, it is detached from the screenfold, which is held together by a sort of wrap-around “spine” of animal skin:

Figure 19 (K5352). © Justin Kerr.

A third variation on the “wrap-around” construction is seen in Figure 20, in which the scribal deities write in codices which have traditional jaguar-hide top covers but are held together by the wrap-around skin on the bottom. Two other vessels (K5721 and K5824) not depicted here show codices with a jaguar-skin top cover only (Kerr).

Figure 20 (K0760). © Justin Kerr.

Different from the above methods of binding a codex is that seen in Figure 21, where the screenfolds of paper are used without any covers by the Young Maize God. A similar depiction can be found in vase K5814 (Kerr).

Figure 21 (K1787). © Justin Kerr.

Also of note is a depiction of an “inside-out” codex, in which the raised bumps on the codex covers with typically face in toward the screenfold, are reversed, facing outward, apparently attached by the protuberances to a block of wood or other solid cover material. Figure 22 shows two scribes with codices in this configuration, and another vessel (K2095) not shown here depicts two bearded scribes writing in “inside-out” codices, as well.

Figure 22 (K1565). © Justin Kerr.

The most intriguing depiction of codices, however, shows three codices, each one different. In Figure 23, we see a traditionally-bound codex in the hands of each of the monkey-man gods. At the feet of the monkey-man god on the left is a codex without any covers, drawn in the traditional style of zigzag vertical lines joined to flat horizontal lines to create a side view of a screenfold. Most intriguing is the codex at the feet of the monkey-man god on the right. The lines of the unbound codex all curve down toward a point at the bottom right of the codex, suggesting, perhaps, a side view of a leafed book as we know it in the traditional European sense.

Figure 23 (K1225). © Justin Kerr.

This theory is at present admittedly little more than conjecture, and further research needs to be done to determine the nature and use of covers for Maya codices. In the meantime, I will follow the detachable-covers model when constructing my own codex.  
Making Brushes/Pens

As with the covers of codices, the process by which the Maya made their writing instruments remains an inexact science to us. No specimens of writing implements survive, but based on the evidence we have from the four extant codices, we can derive the sorts of instruments used to create the thick and thin lines in them. In each of the four codices, we can see thick lines of red and black, and in some cases, the lines are not uniform in thickness and contain in places wisps of ink that trail off from the main line, strongly suggesting a pliable-hair brush of some kind:

The major tool of the Maya scribe and painter was the brush pen. None have survived, but a fairly good idea of what they looked like can be gained from numerous representations on Classic pictorial ceramics, and on the already cited incised bone from the tomb of Hasaw Cha’an K’awil at Tikal. Most of those shown have an almost identical appearance to the traditional Chinese brush pen: a tube-like handle made of reed, bamboo, or the like, into which a tip made from animal hairs has been inserted. In the majority of cases, the tip of the Maya instrument seems to have been bound and then inserted and glued inside the handle, but some pottery images show a tying up of the hairs outside the end of the handle. Clearly, the artist/scribe had different sizes at his disposal, depending on the nature of the task: for very fine work, particularly in codices (see the pen held by the Young Maize God on one painted vase), the tips would have been very slender, while for painting texts on bare, relatively rough, rock walls in tombs and caves, much coarser brushes were called for. (Coe 146-147)

There is also evidence on Maya pottery of writing instruments in use; these instrument, as mentioned above, are sometimes brushes of varying sizes and thickness, but, more often than not, the instrument is shown simply as a short, straight line, suggesting a carved or whittled pen.

Figure 24 (K5824). © Justin Kerr.

To support the theory of the existence of quill pens or other firm-tipped writing instruments, we can rely on evidence from the four surviving codices, whose pages contain fine, even lines which do not vary in thickness and which do not betray evidence of brush-hair “stray.”

Thus, the Maya seem to have had two different sorts of writing implements: brush pens and hard-tipped fine-line pens. In order to make the latter, I simply whittled a goose quill, using the sharp cutting edge of a broken-off flint to do the cutting. Despite the clumsiness of this process relative to using, say, a penknife, the nib I was able to produce was very strong and thin.

Figure 25. Hard-tipped pen.

In order to create a brush pen, I glued an assembly of hair-bristles into the hollowed-out marrow of a chicken bone. I then cut the tip of the brush at a slight angle, again using my flint-sherd knife. The resulting brush, was very sturdy, and was surprisingly comfortable in the hand.

Figure 26. Human-hair brush pen.  
Making Ink

The ink used to create Maya codices has undergone careful scrutiny, and most scholars agree that the black ink used in Maya codices is a combination of water and fire-soot scraped from the bottoms of cooking pots. This carbon-black is one of the most permanent and durable inks possible, and has stood up to environmental conditions extremely well over time. Likewise, the recipe for colored pigments are combinations of minerals and water. In their codices, the Maya used red and black heavily and other colors more sparingly, as evinced by the four surviving codices. After the uncertainty and debate about the format and function of the codex covers, I was relieved to see scholarly agreement about the inks used by the Maya, and I was even more relived to discover that Maya ink is relatively easy to manufacture. Michael Coe describes the conch-shell vessels used by scribes to hold ink:

In his 1977 study of the patron gods of Maya scribes, the present writer was able to recognize that these supernaturals sometimes held a writing implement in one hand and in the other an ink or paint container consisting of a conch shell cut in half lengthwise. The shell would have made an ideal “inkpot” since the several compartments produced by cutting could have contained pigments of different colors. In subsequent years, knowledge of this container has been extended by new archaeological and iconographic research: the statue of the Monkey-man God found in the scribal palace at Copán, for example, has just such an inkpot in its left hand, as do the scribal figures in the niches of the building exterior. And in her 1994 exhibition “Painting the Maya Universe,” Dorie Reents-Budet put on display two surviving examples identified as conch-shell paintpots, as well as a ceramic effigy cut-conch-shell container found in the Late Classic tomb of the Tikal ruler Hasaw Cha’an K’awil. (Coe 150)

Conch shells were ideal containers for ink, as their smooth and slippery inside surfaces allowed for good mixing of the ingredients, and their sharp edges facilitated wiping of one’s brush in order to maintain a proper amount of colorant on one’s brush (Reents-Budet 41-43). The Maya recipe for ink usually contains only two ingredients: water as a vehicle, and a colorant in suspension:

How would the soot used in black ink have been produced by the burning of wood? A clue may again be furnished by early Yucatec dictionaries, in which there is an entry pok, glossed as “soot, lampblack,” as well as a more complete phrase u pokil kum, “soot from a cooking-pot,” suggesting that the soot was scraped from the bottoms of kitchen ollas and mixed with water to produce ink. Such an ink, whether called abak, sabak, or pok, is the most permanent one known, since it is virtually pure carbon, and has survived remarkably well on the walls of temples, caves, and tombs, as well as on the pages of the extant Maya codices.

The Aztecs knew of the Maya lowlands as the “Land of the Black and the Red,” i.e. the land of the codices, and it is true that both black and red pigments were employed to write the hieroglyphic texts in the manuscripts. I can say from personal examination that the red used in the Dresden, Madrid, and Grolier codices is hematite. Hematite is an iron oxide pigment that is far superior to red ochre, but it was probably far more costly as it does not occur naturally in much of the Maya lowlands. (Coe 151)

I mixed together a number of batches of black ink, having reserved the soot from the bottom of my kettle during the nixtamalization process above. The ideal solution of vehicle to suspension (about XXX parts water to XXX parts soot) produced an ink that was sludgy, blacker and thicker than modern bottled ink, which did not run or bleed on the writing surface during preliminary writing experiments.

Mary Elizabeth Haude has conducted a study of colorant composition from maps made by Mesoamerican and Spanish artists during the early Spanish colonial period. Although I will not delineate her findings here, the results of her research clearly indicate that the Maya had mastered the creation of opaque permanent pigments for use in mural-painting, pottery decoration, and writing (Haude). Based on her findings and the suppositions of scholars such as Michael Coe and Dorie Reents-Budet, I also made a batch of red ink. For red ink, I found the process more demanding, mostly because of the scarcity of one of the ingredients:

Primarily, colors were obtained from natural minerals which were mixed together in cactus juice, water or other natural liquids. Colors ranged from browns, reds, oranges, greens, blues, turquoises, purples, yellows, black and white. Variation of pigments was achieved through dilution rather than the mixing of pigments. Darker hues were obtained by coating layers of the pigment on top of each other. (Copeland)

Red ink is composed of hematite (a mineral rust from iron ore) and water, and the samples of hematite I collected came from scraps of old steel-mill iron ore, which contain much more than just hematite. It is unknown whether the Maya used any means of separating hematite from other minerals with which it is sometimes found. I used a steel file (a decidedly un-Maya time-saving tool) to obtain iron-ore dust to produce a fair facsimile of “Maya red.” The vehicle-to-suspension ration for red ink (XXX parts water to XXX parts hematite) produced ink of similar consistency and qualities to my black ink. Figure 26 shows lines and ah ts’ib glyphs written with differing recipes of black and red ink.

Figure 27. Ink recipe results.  
Writing The Text

In order to begin thinking about how to write out my text, I must first consider what we might call “penmanship.” How did the Maya hold their brushes? Did they rest their hands on the page, or did they utilize a “painter’s grasp,” dangling the brush over the surface of the codex? Fortunately, the archaeological and pottery evidence is fairly clear on these points:

The Western custom of holding writing instruments at an angle, with the hand supported on the actual writing surface, tends to dictate sequences of slanting strokes, all leaning at the same angle (for right-handed writers, this means slanting in a northeast to southwest direction). A perusal of the Classic Maya texts presented in the next chapter will demonstrate the same general northeast-southwest bias in painted or incised glyphs and glyph blocks, indicating that Maya scribes also held their pens at an angle to the writing surface, rather than vertically; this is confirmed by a number of representations of scribal deities on Codex-style ceramics. (Coe 147)

Having settled the question of how to hold the instruments, I now needed to figure out where to put the codex in order to write in it. My choices seemed fairly limited, as the evidence we have already seen points to the codex being placed on the floor with the scribe seated cross-legged before it. However, Michael Coe suggests a few different possibilities:

From this point on, the scribe or scribes were in total control of the preparation of a particular book. From imagery on some pictorial Maya vases, we might suppose that the codex rested at floor level, perhaps on a protective mat, and the scribe sat cross-legged before it, writing instrument in hand. . . .

It is possible that the blank codex was propped up on some kind of lectern, so that the seated scribe would not have been obliged to view the pages on which he was working at an acute angle. In illuminated manuscripts of medieval and early Renaissance Europe, scribes are often depicted before such lecterns seated on chairs or stools; it is doubtful that the Maya scribe--like his counterpart in the world of ancient Egypt--sat anywhere but cross-legged on the floor or on a low platform when working on manuscripts. The paper or silk surfaces on which Chinese and Japanese calligraphers write are laid flat on a table, to be sure, but the artist either sits in a raised chair or stands up, giving him sufficient height over the surface for the brush to be held in the desired vertical position. But this Oriental verticality is absent in the Maya script: the northeast to southwest bias of Classic painted and incised texts shows that the scribe held the pen on a slant, and probably leaned slightly forward over the writing surface while he worked. In other words, the ah ts’ib may not have needed a lectern at all. (Coe 148)

I have adopted the writing stance seen in the majority of the depictions of scribes: seated on the floor with the codex resting on the ground before me. Once I had gathered my brushes, inks, and other materials, I began to lay out the page with red grid lines to mark where text and illustrations should go. Figure 28 shows the material laid out and ready for use. Although I am not using conch shells as inkpots, this arrangement is very much like what a Maya scribe would have used.

Figure 28. The assembled materials.

It is thought that it took several days to write each codex. Each of the figures was delineated with black ink made of a coal base. The initial drawing was done with an instrument made from the thorns of the maguey cactus or from bone splinters of small animals-mainly birds. Later, the details on the inside of the square were filled in with a thicker paintbrush made with animal hair. (Martí)

One advantage of the Maya scribes posited by many scholars is that the codices which survive today are actually copies of much older texts. If this is the case, then the scribes laid out their grids and texts using an exemplar. I, too, used an exemplar for my text, and I am indebted to Nancy McNelly for use of her fictional “practice text,” which I have reproduced here in Figure 28 in a slightly modified form. It describes the conquests and life circumstances of Flint Jaguar of Minal (literally “Zero-place,” or, more figuratively, “Nowhere”).

Figure 29. Practice text ( © Nancy McNelly.

McNelly has provided, thankfully, a narrative translation of the text:

On (August 9, 813) Flint Jaguar of Minal was born. His mother, Lady Red Mountain, was a woman of the upper aristocracy. His father, One Jaguar, the ruler of Minal, lived to be at least 59 years old. One Jaguar had taken a captive, White Heron, who [was] important enough to be mentioned in his titles. After his father’s death, which took place when he was 19, Flint Jaguar became thirteenth member of his lineage to be seated as the ruler of Minal. As part of his accession ritual he called forth a spirit (k’awil), possibly by means of bloodletting.

Approximately 17 years later, Flint Jaguar took part in ceremonies to commemorate the completion of the first k’atun of the bak’tun by scattering sacred drops and performing a ritual dance with a hasaw-chan, a ceremonial banner that decorated the ruler’s palanquin.

Approximately 2 years later, warfare took place within the boundaries of Minal. One of Flint Jaguar’s wives, Lady Yellow Snake, performed a bloodletting, presumably to insure that her husband would be victorious in battle. Flint Jaguar captured Fire Bat, an enemy who was of high enough rank to be named. Sixty-four days later, Flint Jaguar had Fire Bat beheaded and had a stela erected to commemorate the event. Approximately 15 years after this event, when he was 53, Flint Jaguar of Minal died. (

This text more properly belongs perhaps on a tribute stela or other piece of monumental architecture, but the basic elements of Maya grammar are inherent in this text. The text is read from left to right and top to bottom in two-glyph columns, beginning with the date column, which locates the writing of the text according to the Maya calendar. The composition of texts is itself a subject worthy of lengthy study, and I encourage readers to go to McNelly’s web site ( in order to learn more about the manner in which this text was created. In order to copy this text in an aesthetically pleasing way, it was imperative to sketch the figures first on a scratch pad to create an exemplar (more on this process below), and then to lay out the design of the page on the codex using grid lines.  
Grid lines

The Maya scribe seldom, it seems, worked freehand from the exemplar to the copy codex. The best examples of codex art are laid out with red grid lines to separate units of text and illustrations, and the glyphs and figures fit neatly within the space provided. The first step in creating a pleasing and skillful page layout was to create grid lines, much as a present-day newspaper compositor decides where to put columns of text and images:

The first task of the ah ts’ib was to lay out a network of gridlines in thin red or reddish-tan wash over those pages on which he was to place his text and accompanying pictures. Internal evidence indicates that never more than four pages were laid out at any one time: the thicker red horizontal lines which were brushed over the fine gridlines dividing up individual sections of each page join up with similar lines over two, three, and four pages, but never in excess of four (not even in the case of the five-page Venus Table of the Dresden [Codex]). Most commonly, two adjacent pages will have linking lines; thus, the scribe/artist usually had the codex open to just two pages at a time. Unfortunately, because of fading and other vicissitudes of time, the gridline system of the Dresden Codex is now almost invisible, and barely appears in most modern facsimiles, but some idea of what it once looked like can be gained from the excellent copy made by Agostino Aglio in the early nineteenth century, and published by Lord Kingsborough in Volume 3 of his monumental Antiquities of Mexico. From this, and from an examination of the original manuscript in Dresden, it is apparent (1) that these lines were produced with a quill or reed pen and a straight-edge, and (2) that the ah ts’ib knew exactly what he was going to write, and where he was going to place the relevant illustrations to his text. He took great care to leave a special box for almost every glyph (excepting single vertical columns of day glyphs, which were simply bounded with a pair of vertical lines) and occasionally even for the bar-and-dot coefficients, as well as one for every single picture--nothing was left to chance. For such a methodology to work, he must have had another codex at his side to guide him in this task. When we look at a codex like the Dresden, we can be confident that we are looking at a compendium based on older sources, some of them perhaps harking back to the period before the Classic Collapse. (Coe 172)

Often, faint sketch lines can be discerned underneath the heavier black lines of the finished images, and one codex exists where the scribe often disregarded his own sketch lines, leaving the codex looking sloppy and hurried:

Faint brownish-red gridlines are still visible throughout the Madrid Codex, but they are very crudely and hastily drawn by comparison with those in the Dresden; they were executed with a brush pen, rather than a quill. It is also obvious that when it came time to lay down the written text and the pictures, the artist/scribe of the Madrid often ignored his own gridline boundaries. As for the Paris Codex, because no truly reliable facsimile in color has yet been published, it is difficult to say much about the gridline system, but lines are faintly visible in some editions. . . . Even though the very fragmentary Grolier lacks an overall gridline network, its columns of day signs are bounded by light reddish-brown lines as in the other three codices; but it is unique among these in that the same finely brushed lines were used to sketch in the pictorial part of each page--in this case, a series of Venus divinities. (Coe 173)

In accordance with the methods described by Michael Coe as used by the scribes who produced the Dresden Codex, I used my paper-bending wood block as a straight edge, laying out a design first on scrap paper with a modern pen in order to create an exemplar page that contained McNelly’s text and an image of Flint Jaguar acceding to the rulership of Minal at the age of 19, witnessing a sacrifice, as described in the text. Figure 30 shows a side-by-side comparison of the exemplar and the grid lines being laid out in the codex.

Figure 30. Laying out gridlines.

Once the grid lines were laid out and allowed to dry, the process of copying the glyphs began.  

In transcribing the glyphs from McNelly’s text, I found that although the glyphs in the sample text are relatively square, the slant of my hand imparted the southwest-to-northeast “lean” described by Coe above. Figure 31 depicts the process of drawing glyphs, again in a side-by-side comparison to the exemplar.

Figure 31. Drawing glyphs.

An unlooked-for result of copying the glyphs was confirmation of Michael Coe’s idea of the hand position employed by Maya scribes. He suggests that scribes used a grip much like the modern pen-grip, with the brush held primarily between the forefinger and thumb, with the ball of the hand resting on the page. One of the results of using Coe’s suggested hand position was far less smearing of the ink, primarily because the glyphs are arranged in columns two glyphs wide (about three inches), or nearly the maximum amount of space that would fit between the brush tip and the ball of my hand. Because of this characteristic of the text layout, my hand never rested on wet ink. For drawing the graphics, I employed a similar technique of working from the northwest corner of the page in columns of figures.  

Nearly every page in the extant codices contains at least one illustration of the text on the same page. In order to create the scene of Flint Jaguar becoming the ruler of Minal, I composed my illustration based on elements of a number of Postclassic-period Maya pottery and architectural carving. The Maya scribe, too, used a controlled visual vocabulary in order to create his images. As in Maya iconography, Flint Jaguar wears a headdress that contains his name glyphs, and his full title is drawn in glyphs to his left. It is reasonable to think that the scribes who drew the glyphs were also responsible for the illustration of their texts, especially if working from an exemplar, as I was:

Es claro que, en el mundo maya, escribir implicaba a la vaz el empleo de glifos y figuras, por ello, en los diccionarios encontramos la voz ts’ib, que significa tanto “escritura” como “escribir y pintar,” de donde se infiere que la escritura requería muchas veces el complemento de las figuras, y, posiblemente, del color. Los tonos empleados en esos códices son el rojo y el negro, así como el característico azul maya. [It is clear that, in the Maya world, to write simultaneously implied the use of glyphs and figures. For this reason, in the dictionaries we find the word ts’ib, that means as much “writing” as “to write and to paint,” of which it is inferred that writing often required the complement of figures, and, possibly, of color. The tones used in those codices are red and black, as well as the characteristic Maya blue.] (Santos 39-40)

A final characteristic of Maya illustration is that important figures are almost always named with blocks of glyphs in single or double columns. In architectural carvings, the artist sometimes “signed” his work in a lower corner, but such a practice is not commonly found in the codices. Figure 32 shows the graphics for my codex in a partially-completed state.

Figure 32. Drawing graphics.


The process by which the Maya made paper is still in dispute, while their techniques for creating codices are well documented but based on evidence obtained at many cultural, linguistic, and temporal removes from original sources. However, while there may be scholarly debate about the purposes and uses of codex books in Maya ritual and social life, it is possible today to make a reasonably accurate reproduction of a Maya codex. Figure 32 depicts the “finished” codex, with screenfolds showing the base huun paper, a page with unpolished gesso coating, a page to which gesso has been applied and which has been polished, and a finished treated page with glyphs and graphics on it.

Figure 33. The finished codex.

This experiment has shown that there are a number of variables yet to be worked out in describing accurately the processes by which the Classic- and early Postclassic-period Maya produced their codices. This area of the book arts is often overshadowed by the study of European and Chinese papermaking and fine bookbinding, mostly, it seems, because few book historians have collected, tested, and theorized about the history behind our admittedly limited knowledge of the techniques of Maya codex creation. It is the hope of the author that this tentative and hypothetical essay will spur further research into the physical processes by which the Maya made their books.  

Clarkson, Persis B. “Classic Maya Pictorial Ceramics: A Survey of Content and Theme.” Papers on the Economy and Architecture of the Ancient Maya. Ed. Raymond Sidrys. Los Angeles: Institute of Archaeology, U of California, 1978. 86-141.

Coe, Michael D. The Art of the Maya Scribe. Ill. Justin Kerr. New York: Harry N. Abrams, 1997.

Copeland, Jessica Lin. “A Qualitative Analysis of Pigment Mixers used by the Maya.” Maya Art. 1998.
<>. 16 Mar. 2001.

Dienhart, John M. The Mayan Languages: A Comparative Vocabulary. Odense, Denmark: Odense UP, 1997. <>. 20 Feb. 2001.

Fash, William L. “Changing Perspectives on Maya Civilization.” Annual Review of Anthropology 23 (1994): 181-208.

Haude, Mary Elizabeth. “Identification and Classification of Colorants Used During Mexico’s Early Colonial Period.” The Book and paper Group Annual 16 (1997).
<>. 16 Mar. 2001.

Kerr, Justin. The Maya Vase Database. <>. 21 March 2001.

Krayna, Philip. “History, Unfolded.” Print 53.1 (Jan./Feb. 1999): 90-94.

Lenz, Hans. El Papel Indígena Mexicano. Mexico City: Secretaría de Educación Pública, 1948.

Lo, Lawrence. “Maya Hieroglyphs.” Ancient Scripts of the World.
<>. 1 Mar. 2001.

López, Citlalli. “Hand-made Bark Paper in Mexico: Local Production--Regional Harvest Transitions in Tree Knowledge, Harvest Strategies and Land Use Systems.” Cultivating (in) Tropical Forests: The Evolution and Sustainability of Intermediate Systems Between Extractivism and Plantations. Lofoten, Norway, 28 Jun.-1 Jul. 2000. <>. 16 Mar. 2001.

Love, Bruce. The Paris Codex: Handbook for a Maya Priest. Austin: U of Texas P, 1994.

Martí, Beatriz. “Maya Codices.” Mundo Maya. <>. 1 Mar. 2001.

The Mayan Epigraphic Database Project. <>. 1 Mar. 2001.

McNelly, Nancy. Rabbit in the Moon: Mayan Glyphs and Architecture. <>. 6 Mar 2001.

---. Mayan Hieroglyphic Practice Text. <>. 6 Apr. 2001.

Murray, David C. “Mesoamerican Web Bookmarks.”
<>. 16 Mar. 2001.

Porter, James B. “Mayan Glyphs.” <>. 1 Mar. 2001.

Rau, Jack. The Codex as a Book Form: Three Maya Codices: Dresden, Tro-Cortesianus, Per[e]sianus. New York: Pre-Columbian P, 1970.

Reddick, Greg. Maya. <>. 21 Mar. 2001.

Reents-Budet, Dorie. “Codices.” Email message. 21 Mar. 2001.

---. Painting the Maya Universe: Royal Ceramics of the Classic Period. Durham: Duke UP, 1994.

The Road to El Dorado. Dirs. Don Paul and Eric Bergeron. Dreamworks, 2000.

Robertson, Martha Barton. “Early Chronicles of the Americas in Manuscripts and Printed Books: Histories, Missionary Accounts, Indian Languages, and New World Incunabula in the Latin American Library.” 47th International Congress of Americanists. Tulane University, New Orleans, 1991.
<>. 16 March 2001.

Sandstrom, Alan R. “Papermaking.” The Oxford Encyclopedia of Mesoamerican Cultures. Ed. Davíd Carrasco. Oxford: Oxford UP, 2000.

Sandstrom, Alan R. and Pamela E. Sandstrom. “Paper.” Archaeology of Ancient Mexico and Central America: An Encyclopedia. Eds. S. T. Evans and D. L. Webster. New York: Garland, 2001. 583.

---. Traditional Papermaking and Paper Cult Figures of Mexico. Norman, OK: U of Oklahoma P, 1986.

Santos, Laura Elena Sotelo. “Los Códices Maya.” Arqueología Mexicana 4.23 (1997): 34-43.

“Sources of Fiber and Pulp.” Hand Papermaking Newsletter 17 (Jan. 1992).
<>. 16 Mar. 2001.

Stuart, David. Telephone Interview. 8 March 2001.

“Tour Copán with David Stuart.” NOVA Online: Lost King of the Maya.
<>. 2 Mar. 2001.

Van der Meeren, Marie. “El papel amate: origen y supervivencia.” Arqueología Mexicana 4.23 (1997): 70-73.

von Hagen, Victor Wolfgang. The Aztec and Maya Papermakers. New York: J. J. Augustin, 1944.

---. “Paper and Civilization.” Scientific Monthly 57.4 (Oct. 1943): 301-314.

Whiting, Thomas A. Lee. “The Maya Codices.” Maya. Eds. Peter Schmidt, Mercedes de la Garza, and Enrique Nalda. New York: Rizzoli, 1998. 206-215.

Contact Dr. Tobin at with questions & comments.
Copyright © 2001 Thomas J. Tobin.
Last update 7 May 2001.