NEWS: Very Early Maya Ceremonial Architecture at Ceibal

The most recent issue of Science includes an article on the remarkable finds recently made at Ceibal (Seibal), Guatemala. Excavations there have revealed very early evidence of Maya ceremonial buildings and civic space, dating as far back as 1000 BCE. It’s wonderful and significant work, extending the roots of Maya religious architecture back to the Early Preclassic. Congratulations go out to Takeshi Inomata (my old Vanderbilt classmate and road-trip companion), Daniela Triadan and their colleagues.

Science News article

Link to Science article (subscription required for full access)

“Early Maya Ceremonial Constructions at Ceibal, Guatemala, and the Origins of Lowland Maya Civilization”

Takeshi Inomata, Daniela Triadan, Kazuo Aoyama, Victor Castillo, and Hitoshi Yonenobu

Science, Vol. 340, no. 6131, pp. 467-471

The spread of plaza-pyramid complexes across southern Mesoamerica during the early Middle Preclassic period (1000 to 700 BCE) provides critical information regarding the origins of lowland Maya civilization and the role of the Gulf Coast Olmec. Recent excavations at the Maya site of Ceibal, Guatemala, documented the growth of a formal ceremonial space into a plaza-pyramid complex that predated comparable buildings at other lowland Maya sites and major occupations at the Olmec center of La Venta. The development of lowland Maya civilization did not result from one-directional influence from La Venta, but from interregional interactions, involving groups in the southwestern Maya lowlands, Chiapas, the Pacific Coast, and the southern Gulf Coast.

Dead Bugs and Olmec Writing

by Stephen Houston

The difficulties of working with an undeciphered writing system are many. The first hurdle is simply that of sorting out the reading order, i.e., determining the ways and directions in which signs are to be read–up-down, top-bottom, right-left, and all the possible variations in laying out a text. The absence of this understanding is fatal. If there is no known reading order, there can be no possibility of penetrating syntax and all the functional attributes implied by the arrangement of signs. At some point, writing must, as a record of language, reflect the organization of language and sequenced sound.

Figure 1. The Phaistos Disk

Yet, there are some navigational aids—all is not lost! For example, with hieroglyphic systems, even undeciphered ones, there is strong evidence that signs obey the strictures of iconography. Sign “behavior” in a text—the ways in which scribes orient signs–tends to conform to its disposition in imagery. Consider the human head or body. The unmarked position shows them much as we would see them, head aloft, feet where they should be. The marked position inverts them, blood rushing to the head, feet jerked aloft in high state of discomfort. Marked signs occur, but they are often rare. To judge from evidence in scripts as diverse as Maya, Aztec, and Egyptian, faces and bodies also orient towards the scanning or reading eye. As I have suggested elsewhere, there is almost a conceit—and perhaps a perceived reality–of social interaction between the reader and the sign. In unmarked form, a face or body addresses the approaching interlocutor. The marked form reverses them, and the face and body look away from the scanning eye. A case in point is the Phaistos Disk (Figure 1), cynosure of every traveler along the bumps and byways of decipherment. Little is known—or knowable—about this system, stamped on a clay disk some 3500 years ago. But there can be little doubt that the faces and bodies are directed to the reader. They almost certainly show the reading order of the disk, with passages that run from right-to-left, from outer edge and spiraling inwards to the center. The right-to-left orientation may have something to do with the fact that the signs were stamped into the clay, with the first typeface known to humanity.

The reason for this near-universal in hieroglyphic systems—the notable fact that such texts conform to the behavior of unmarked imagery–probably has to do with script origins. Codified imagery may well have been a necessary and sufficient condition for codified script (Houston 2004:288-293). In hieroglyphs, there may also have been a need to retain the legibility of iconic referents, well beyond the time of origins, when script coalesced from images. By definition, signs in hieroglyphic systems convey the notion of a discrete object, oriented in much the same way as that object would appear in imagery. For this reason, cursivity tends not to occur, as it would bruise the notion of a separable, distinct thing, slicing through or diffusing sign boundaries. (Of course, Maya script dealt with the need to compress text by the expedient of merging of infixing elements.)

This brings us to Dead Bugs. A recent effort to reexamine the celebrated Cascajal Block, of Olmec affiliation, and perhaps as early as 1100 BC in date, perhaps a few centuries later, proposes a radical reworking of the reading order in the block (Mora-Marín 2009). There is some accompanying PR to add brio to the claim, now available in a peer-reviewed journal ( The original publication suggested that the orientation was of a rectangular text, longer than wide (Rodríguez Martínez et al. 2006). The new, proposed reading order places it on its side, 90 degrees off, wider than it is long. By this view, the text must be read in columns, left-to-right, and with the necessity of a new numbering of the text (Figure 2; Mora-Marín 2009:404).

Figure 2. Contrasting Orientations of the Cascajal Block

Since publishing the Cascajal Block, we have received many suggestions about the text. Some can only be described as bizarre—that the block records Chinese, that it tabulates the conditions of crops, that it shows human dentition or a chart of chromosomes. Perhaps the best and most reasoned comment comes from John Wood of Australia, who, in an unpublished paper, argues that there is more columnar organization in the top of the block than we had recognized—the slight disorganization of the text, with some irregular spaces and alignments, makes the text a difficult piece to parse, even under the best of circumstances. Wood’s views may well be correct, however. They would make sense of several peculiar irregularities in the layout of the text.

But it is highly improbable that the block should be viewed on its side.

First, as noted in the original publication, many of the signs, even though undeciphered, conform to codified elements in Olmec imagery. There are the famed “knuckle-dusters,” maize cobs, probable bloodletters, even a throne. Several excellent compilations of Olmec images attest to their usual positioning (e.g., Guthrie 1995: passim). The claim for side-ways orientation would pivot these elements on their sides, in violation of all known canons of Olmec imagery. Quite simply, there would need to be a radical disconnect between representational conventions and the signs that securely descend from Olmec imagery. Curiously, when the author responsible for the new theory lays out sequences for comparison, he re-positions them from his claimed orientation, fitting them into a display that conforms precisely to the orientation he disputes (e.g., Mora-Marín 2009:Figures 10-13). One proposed example, intended to show that such counter-intuitive orientations are relatively common, is etched on the belly of a raptor, scratched in turn on an obsidian core from La Venta. What the author does not emphasize is that he has had to reverse the image to force a resemblance to the Cascajal sign (Mora-Marín 2009:Figure 14c-f). The similarity is weak in any case. Another drawing of the obsidian fails to show the same internal details (e.g., Joralemon 1971:Figure 197; cf. Mora-Marín 2009:Figure 15c; Kent Reilly’s updated rendering, done after study of the original, also departs from Mora-Marín, cf. Reilly’s Figure 11, in

Figure 3. Incised Vase, Snite Museum of Art, University of Notre Dame (Kerr Photograph 6441, copyright Kerr Associates).

Second, some of the Cascajal signs, such as a possible bottle gourd, are also found on images such as an incised vase, c. 900 BC, from the Snite Museum of Art at the University of Notre Dame (Figure 3, Karl Taube, personal communication, 2007). Taube demonstrates that these are essentially the same fetishes, albeit with a slightly leaning quality to fit into the spaces on the side of the vessel. They are not shown on their sides, as the new theory would predict, yet the orientation of the vessel is patent. There can be no question here of their correct positioning.

Figure 4. Dwarf from “area of Teotihuacan,” Covarrubias notebooks (Joralemon 1971:Figure 19).
Taube has also shown that a small carving of an Olmec dwarf, in standing position, has two signs in the Cascajal system on either side of his head—and, crucially, the glyphs are placed in an orientation that is not on its side (Figure 4, Joralemon 1971:Figure 19). A paired set of glyphs on the Cascajal Block (glyphs 21-22
Figure 5. Cascajal couplet and orientation of same elements in eyes of Monument 1, Laguna de los Cerros, Mexico (Joralemon 1971:Figures 125, 153).
in the original numbering) occurs within the eyes of a sculpture from Laguna de Cerros, again in a position that runs counter to the sideways proposal (Figure 5, e.g., Joralemon 1971:Figures 125, 153). The observations by Taube augment the corpus, further link the block to Olmec imagery, and confirm that the correct orientation of the signs on the Cascajal Block corresponds to the presentation in the initial publication. It is likely that similar signs will appear with closer study of existing collections, as in this quick sketch of what appear to be glyphs on display at the Walters Museum of Art (Figure 6); exciting finds by David Cheetham of Arizona State University are forthcoming from Cantón Corralito, Chiapas.
Figure 6. Sketch of possible signs in Cascajal-system, vase on exhibit at the Walters Art Museum, Baltimore.

Finally, a side-ways reading order would also take the insect in the Cascajal Signary (glyphs 1, 23, 50) and put it on its back, feet in the air—quite dead, unprecedented in contemporary imagery, and far different from the semblant creature found in Monument 43 at San Lorenzo, Mexico (Figure 7, Coe and Diehl 1980, I:Figures 481-482).

Figure 7. San Lorenzo, Monument 43 (Coe and Diehl 1980, I:Figures 481-482, drawing by Felipe Dávalos)

The points have to be put as plainly as possible:

– in all cases where the Cascajal signs appear in iconography, they do not correspond to the claim for a side-ways orientation of the block;
– in all cases where the Cascajal signs appear in text-formats, or in paired, couplet-like form, they do not correspond to the claim for a side-ways orientation of the block.

The new claim is thus incorrect. The perceived “repeated sequences” do not exist (e.g., Mora-Marín 2009:Figures 6, 7), and they have no bearing on any patterned syntax in the text. However, even if mistaken, the proposal has created a useful occasion to reaffirm the close ties in Mesoamerica of imagery to texts. When those ties are disregarded, confusions ensue. The block must be placed back in its correct position, in the vertical orientation it craves. The Dead Bugs live!


Coe, Michael D., and Richard Diehl. 1980. In the Land of the Olmec. 2 vols. University of Texas Press, Austin.

Guthrie, Jill. 1995. The Olmec World: Ritual and Rulership. The Art Museum, Princeton University, Princeton.

Houston, Stephen D. 2004. Writing in Early Mesoamerica. In The First Writing: Script Invention as History and Process, ed. S. D. Houston, pp. 274-309. Cambridge University Press, Cambridge.

Joralemon, Peter D. 1971. A Study of Olmec Iconography. Studies in Pre-Columbian Art and Archaeology 7. Dumbarton Oaks, Washington, D.C.

Mora-Marín, David F. 2009. Early Olmec Writing: Reading Format and Reading Order. Latin American Antiquity 20(3):395-412.

Rodríguez Martínez, María del Carmen, Ponciano Ortíz Ceballos, Michael Coe, Richard Diehl, Stephen Houston, Karl Taube, Alfredo Delgado Calderón. 2006. Oldest Writing in the New World. Science 313:1610-1614.