Maya script

From Wikipedia, the free encyclopedia
  (Redirected from Maya hieroglyphics)
Jump to: navigation, search
Maya script
Type
Mixed Logographic and Syllabic (used both logograms and syllabic characters)
Languages Mayan languages
Time period
3rd century BCE to present
ISO 15924 Maya, 090
An inscription in Maya glyphs from the site of Naranjo, relating to the reign of king Itzamnaaj K'awil, 784-810

The Mayan script, also known as Mayan glyphs or Mayan hieroglyphs, is the writing system of the Maya civilization of Mesoamerica, presently the only Mesoamerican writing system that has been substantially deciphered. The earliest inscriptions found which are identifiably Maya date to the 3rd century BCE in San Bartolo, Guatemala.[1][2] Maya writing was in continuous use throughout Mesoamerica until shortly after the arrival of the conquistadors in the 16th century CE and into the 18th century in isolated areas, such as Tayasal.

Maya writing used logograms complemented by a set of syllabic glyphs, somewhat similar in function to modern Japanese writing. Mayan writing was called "hieroglyphics" or hieroglyphs by early European explorers of the 18th and 19th centuries who did not understand it but found its general appearance reminiscent of Egyptian hieroglyphs, to which the Mayan writing system is not at all related.

Although modern Mayan languages use the Latin alphabet as standard, Mayan writing has received official support and promotion by the Mexican government and is taught in universities and public schools in several Mayan speaking areas.[citation needed]

Languages[edit]

It is now thought that the codices and other Classic texts were written by scribes, usually members of the Maya priesthood, in a literary form of the Ch’olti’ language (known as Classic Maya).[3][4] It is possible that the Maya elite spoke this language as a lingua franca over the entire Maya-speaking area, but also that texts were written in other Mayan languages of the Petén and Yucatán, especially Yucatec. There is also some evidence that the script may have been occasionally used to write Mayan languages of the Guatemalan Highlands.[4] However, if other languages were written, they may have been written by Ch’olti’ scribes, and therefore have Ch’olti’ elements.

Structure[edit]

Mayan writing consisted of a relatively elaborate set of glyphs, which were laboriously painted on ceramics, walls or bark-paper codices, carved in wood or stone, or molded in stucco. Carved and molded glyphs were painted, but the paint has rarely survived. About 90% of Mayan writing can now be read with varying degrees of certainty, enough to give a comprehensive idea of its structure.[5]

The Mayan script was a logosyllabic system. Individual symbols ("glyphs") could represent either a word (actually a morpheme) or a syllable; indeed, the same glyph could often be used for both. For example, the calendaric glyph MANIK’ was also used to represent the syllable chi. (It is customary to write logographic readings in all capitals and phonetic readings in italics.) It is possible, but not certain, that these conflicting readings arose as the script was adapted to new languages, as also happened with Japanese kanji and with Assyro-Babylonian and Hittite cuneiform. There was ambiguity in the other direction as well: Different glyphs could be read the same way. For example, half a dozen apparently unrelated glyphs were used to write the very common third person pronoun u-.

Mayan was usually written in blocks arranged in columns two blocks wide, read as follows:

Maya inscriptions were most often written in columns two glyphs wide, with each such column read left to right, top to bottom

Within each block, glyphs were arranged top-to-bottom and left-to-right, superficially rather like Korean Hangul syllabic blocks. However, in the case of Mayan, each block tended to correspond to a noun or verb phrase such as his green headband. Also, glyphs were sometimes conflated, where an element of one glyph would replace part of a second. Conflation occurs in other scripts: For example, in medieval Spanish manuscripts the word de 'of' was sometimes written Ð (a D with the arm of an E). Another example is the ampersand (&) which is a conflation of the Latin "et". In place of the standard block configuration Mayan was also sometimes written in a single row or column, 'L', or 'T' shapes. These variations most often appeared when they would better fit the surface being inscribed.

Mayan glyphs were fundamentally logographic. Generally the glyphs used as phonetic elements were originally logograms that stood for words that were themselves single syllables, syllables that either ended in a vowel or in a weak consonant such as y, w, h, or glottal stop. For example, the logogram for 'fish fin' (Maya [kah] — found in two forms, as a fish fin and as a fish with prominent fins), came to represent the syllable ka. These syllabic glyphs performed two primary functions: they were used as phonetic complements to disambiguate logograms which had more than one reading, as also occurred in Egyptian and in modern Japanese (i.e. furigana); and they were used to write grammatical elements such as verbal inflections which did not have dedicated logograms, also as in modern Japanese (i.e. okurigana). For example, bahlam 'jaguar' could be written as a single logogram, BALAM, complemented phonetically as ba-BALAM, or BALAM-ma, or ba-BALAM-ma, or written completely phonetically as ba-la-ma.

Harmonic and disharmonic echo vowels[edit]

Phonetic glyphs stood for simple consonant-vowel or bare-vowel syllables. However, Mayan phonotactics is slightly more complicated than this: Most Mayan words end in a consonant, not a vowel, and there may be sequences of two consonants within a word as well, as in xolte’ [ʃolteʔ] 'scepter', which is CVCCVC. When these final consonants were sonorants (l, m, n) or gutturals (j, h, ’) they were sometimes ignored ("underspelled"), but more often final consonants were written, which meant that an extra vowel was written as well. This was typically an "echo" vowel that repeated the vowel of the previous syllable. That is, the word [kah] 'fish fin' would be underspelled ka or written in full as ka-ha. However, there are many cases where some other vowel was used, and the orthographic rules for this are only partially understood; this is largely due to the difficulty in ascertaining whether this vowel may be due to an underspelled suffix. Lacadena and Wichmann (2004) proposed the following conventions:

  • A CVC syllable was written CV-CV, where the two vowels (V) were the same: yo-po [yop] 'leaf'
  • A syllable with a long vowel (CVVC) was written CV-Ci, unless the long vowel was [i], in which case it was written CiCa: ba-ki [baak] 'captive', yi-tzi-na [yihtziin] 'younger brother'
  • A syllable with a glottalized vowel (CV’C or CV’VC) was written with a final a if the vowel was [e, o, u], or with a final u if the vowel was [a] or [i]: hu-na [hu’n] 'paper', ba-tz’u [ba’tz’] 'howler monkey'.
  • Preconsonantal [h] is not indicated.

That is, a simple vowel is intended if the vowels are the same (harmonic), and either two syllables are intended (likely underspelled) if the vowels are not the same (disharmonic), or else a single syllable with a long vowel (if V1 = [a e? o u] and V2 = [i], or else if V1 = [i] and V2 = [a]) or with a glottalized vowel (if V1 = [e? o u] and V2 = [a], or else if V1 = [a i] and V2 = [u]). The long-vowel reading of [Ce-Ci] is still uncertain, and there is a possibility that [Ce-Cu] represents a glottalized vowel (if it is not simply an underspelling for [CeCuC]), so it may be that the disharmonies form natural classes: [i] for long non-front vowels, otherwise [a] to keep it disharmonic; [u] for glottalized non-back vowels, otherwise [a].

A more complex spelling is ha-o-bo ko-ko-no-ma for [ha’o’b kohkno’m] 'they are the guardians'. (Vowel length and glottalization are not always indicated in common words like 'they are'.) A minimal set is,

ba-ka [bak]
ba-ki [baak]
ba-ku [ba’k] = [ba’ak]
ba-ke [baakel] (underspelled)
ba-ke-le [baakel]

Verbal inflections[edit]

Despite depending on consonants which were frequently not written, the Mayan voice system was reliably indicated. For instance, the paradigm for a transitive verb with a CVC root is as follows:

Voice Transliteration Transcription Gloss
Active u-TZUTZ-wa utzutzu’w "s/he finished it"
Passive TZUTZ-tza-ja tzuhtzaj "it was finished"
Mediopassive TZUTZ-yi tzutzuuy "it got finished"
Antipassive TZUTZ-wi tzutzuuw "s/he finished"
Participial TZUTZ-li tzutzuul "finished"

The active suffix did not participate in the harmonic/disharmonic system seen in roots, but rather was always -wa.

However, the language changed over 1500 years, and there were dialectical differences as well, which are reflected in the script, as seen next for the verb "s/he sat" (h is an infix in the root chum for the passive voice):

Period Transliteration Transcription
Late Preclassic CHUM? chuhm?
Early Classic CHUM-ja chuhm-aj
Classic (Eastern Ch'olan) CHUM-mu-la-ja chum-l-aj
Late Classic (Western Ch'olan) CHUM-mu-wa-ni chum-waan

Emblem glyphs[edit]

Tikal or "Mutal" Emblem Glyph, Stela 26 in Tikal's Litoteca Museum

An "emblem glyph" is a kind of royal title. It consists of a word ajaw—a Classic Maya term for "lord" of yet unclear etymology but well-attested in Colonial sources[6]—and a place name that precedes the word ajaw and functions as an adjective. An expression "Boston lord" would be a perfect English analogy. Sometimes, the title is introduced by an adjective k’uhul ("holy, divine" or "sacred"), just as if someone wanted to say "holy Boston lord". Of course, an "emblem glyph" is not a "glyph" at all: it can be spelled with any number of syllabic or logographic signs and several alternative spellings are attested for the words k’uhul and ajaw, which form the stable core of the title. The term "emblem glyph" simply reflects the times when mayanists could not read Classic Maya inscriptions and had to come up with some nicknames isolating certain recurrent structural components of the written narratives.

This title was identified in 1958 by Heinrich Berlin,[7] who coined the term "emblem glyph". Berlin noticed that the "emblem glyphs" consisted of a larger "main sign" and two smaller signs now read as k’uhul ajaw. Berlin also noticed that while the smaller elements remained relatively constant, the main sign changed from site to site. Berlin proposed that the main signs identified individual cities, their ruling dynasties, or the territories they controlled. Subsequently, Marcus[8] argued that the "emblem glyphs" referred to archaeological sites, broken down in a 5-tiered hierarchy of asymmetrical distribution. Marcus' research assumed that the emblem glyphs were distributed in a pattern of relative site importance depending on broadness of distribution, roughly broken down as follows: Primary regional centers (capitals) (Tikal, Calakmul, and other "superpowers") were generally first in the region to acquire a unique emblem glyph(s). Texts referring to other primary regional centers occur in the texts of these "capitals", and dependencies exist which use the primary center's glyph. Secondary centers (Altun Ha, Luubantuun, Xunantunich, and other mid-sized cities had their own glyphs but are only rarely mentioned in texts found in the primary regional center, while repeatedly mentioning the regional center in their own texts. Tertiary centers (towns) had no glyphs of their own, but have texts mentioning the primary regional centers and perhaps secondary regional centers on occasion. These were followed by the villages with no emblem glyphs and no texts mentioning the larger centers, and hamlets with little evidence of texts at all.[9] This model was largely unchallenged for over a decade until Mathews and Justeson,[10] as well as Houston[11] argued once again that the "emblem glyphs" were the titles of Maya rulers with some geographical association.

The debate on the nature of "emblem glyphs" received a new spin with the monograph by Stuart and Houston.[12] The authors convincingly demonstrated that there were lots of place names-proper, some real, some mythological, mentioned in the hieroglyphic inscriptions. Some of these place names also appeared in the "emblem glyphs", some were attested in the "titles of origin" (various expressions like "a person from Boston"), but some were not incorporated in personal titles at all. Moreover, the authors also highlighted the cases when the "titles of origin" and the "emblem glyphs" did not overlap, building upon an earlier research by Houston.[13] Houston noticed that the establishment and spread of the Tikal-originated dynasty in the Petexbatun region was accompanied by the proliferation of rulers using the Tikal "emblem glyph" placing political and dynastic ascendancy above the current seats of rulership.[14] Recent investigations also emphasize the use of emblem glyphs as an emic identifier to shape socio-political self-identity.[15]

Numerical system[edit]

List of Maya numerals from 0 to 19 with underneath two vertically oriented examples
Main article: Maya numerals

The Mayas used a positional base-twenty (vigesimal) numerical system which only included whole numbers. For simple counting operations, a bar and dot notation was used. The dot represents 1 and the bar represents 5. A shell was used to represent zero. Numbers from 6 to 19 are formed combining bars and dots. Numbers can be written horizontally or vertically.

These four examples show how the value of Maya numerals can be calculated

The value of a number depends on its position going from the bottom line upward in the configuration. The initial position (viz bottom line) has the value represented in the symbol. On the following line, the value of the symbol is multiplied by 20; on the third line from the bottom it's multiplied by 400, and each successive line is growing by powers of 20. That is to say, Mayan numerals use base 20. This positional system allows the calculation of large figures, necessary for chronology and astronomy.[16]

History[edit]

It was until recently thought that the Maya may have adopted writing from the Olmec or Epi-Olmec culture, who used the Isthmian script. However, murals excavated in 2005 have pushed back the origin of Maya writing by several centuries, and it now seems possible that the Maya were the ones who invented writing in Mesoamerica.[17] Regardless, it is generally believed that the Maya developed the only complete writing system in Mesoamerica, meaning that they were the only civilization that could write everything they could say.

Knowledge of the Maya writing system continued into the early colonial era and reportedly a few of the early Spanish priests who went to Yucatán learned it. However, as part of his campaign to eradicate pagan rites, Bishop Diego de Landa ordered the collection and destruction of written Maya works, and a sizable number of Maya codices were destroyed. Later, seeking to use their native language to convert the Maya to Christianity, he derived what he believed to be a Maya "alphabet" (the so-called de Landa alphabet). Although the Maya did not actually write alphabetically, nevertheless he recorded a glossary of Maya sounds and related symbols, which was long dismissed as nonsense but eventually became a key resource in deciphering the Maya script, though it has itself not been completely deciphered. The difficulty was that there was no simple correspondence between the two systems, and the names of the letters of the Spanish alphabet meant nothing to Landa's Maya scribe, so Landa ended up asking the equivalent of write H: a-i-tee-cee-aitch "aitch", and glossed a part of the result as "H".

Landa was also involved in creating a Latin orthography for the Yukatek Maya language, meaning that he created a system for writing Yukatek in the Latin alphabet. This was the first Latin orthography for any of the Mayan languages,[citation needed] which number around thirty.

Only four Maya codices are known to have survived the conquistadors. Most surviving texts are found on pottery recovered from Maya tombs, or from monuments and stelae erected in sites which were abandoned or buried before the arrival of the Spanish.

Knowledge of the writing system was lost, probably by the end of the 16th century. Renewed interest in it was sparked by published accounts of ruined Maya sites in the 19th century.

Decipherment[edit]

Maya glyphs in stucco at the Museo de sitio in Palenque, Mexico

The decipherment of the writing was a long and laborious process. 19th century and early 20th century investigators managed to decode the Maya numbers and portions of the texts related to astronomy and the Maya calendar, but understanding of most of the rest long eluded scholars. In the 1930s, Benjamin Whorf wrote a number of published and unpublished essays, proposing to identify phonetic elements within the writing system. Although some specifics of his decipherment claims were later shown to be incorrect, the central argument of his work, that Maya hieroglyphs were phonetic (or more specifically, syllabic), was later supported by the work of Yuri Knorozov, who played a major role in deciphering Maya writing.[18] In 1952, Knorozov published the paper "Ancient Writing of Central America" arguing that the so-called "de Landa alphabet" contained in Bishop Diego de Landa's manuscript Relación de las Cosas de Yucatán was actually made of syllabic, rather than alphabetic symbols. He further improved his decipherment technique in his 1963 monograph "The Writing of the Maya Indians"[19] and published translations of Maya manuscripts in his 1975 work "Maya Hieroglyphic Manuscripts". In the 1960s, progress revealed the dynastic records of Maya rulers. Since the early 1980s it has been demonstrated that most of the previously unknown symbols form a syllabary, and progress in reading the Maya writing has advanced rapidly since.

As Knorozov's early essays contained several older readings already published in the late 19th century by Cyrus Thomas,[20] and the Soviet editors added propagandistic claims[21] to the effect that Knorozov was using a peculiarly "Marxist-Leninist" approach to decipherment,[21] many Western Mayanists simply dismissed Knorozov's work. However, in the 1960s more came to see the syllabic approach as potentially fruitful, and possible phonetic readings for symbols whose general meaning was understood from context began to be developed. Prominent older epigrapher J. Eric S. Thompson was one of the last major opponents of Knorozov and the syllabic approach. Thompson's disagreements are sometimes said to have held back advances in decipherment.[22] For example, Coe (1992) says[23] "the major reason was that almost the entire Mayanist field was in willing thrall to one very dominant scholar, Eric Thompson".

In 1959, examining what she called "a peculiar pattern of dates" on stone monument inscriptions at the Classic Maya site of Piedras Negras, Russian-American scholar Tatiana Proskouriakoff determined that these represented events in the lifespan of an individual, rather than relating to religion, astronomy, or prophecy, as held by the "old school" exemplified by Thompson. This proved to be true of many Maya inscriptions, and revealed the Maya epigraphic record to be one relating actual histories of ruling individuals: dynastic histories similar in nature to those recorded in literate human cultures throughout the world. Suddenly, the Maya entered written history.[24]

Although it was now clear what was on many Maya inscriptions, they still could not literally be read. However, further progress was made during the 1960s and 1970s, using a multitude of approaches including pattern analysis, de Landa's "alphabet", Knorozov's breakthroughs, and others. In the story of Maya decipherment, the work of archaeologists, art historians, epigraphers, linguists, and anthropologists cannot be separated. All contributed to a process that was truly and essentially multidisciplinary. Key figures included David Kelley, Ian Graham, Gilette Griffin, and Michael Coe.

Dramatic breakthroughs occurred in the 1970s, in particular at the first Mesa Redonda de Palenque, a scholarly conference organized by Merle Greene Robertson at the Classic Maya site of Palenque held in December, 1973. A working group was led by Linda Schele, an art historian and epigrapher at the University of Texas at Austin, which included Floyd Lounsbury, a linguist from Yale, and Peter Mathews, then an undergraduate student of David Kelley's at the University of Calgary (whom Kelley sent because he could not attend). In one afternoon they managed to decipher the first dynastic list of Maya kings, the ancient kings of the city of Palenque. By identifying a sign as an important royal title (now read as the recurring name K'inich), the group was able to identify and "read" the life histories (from birth, to accession to the throne, to death) of six kings of Palenque.

From that point, progress proceeded at an exponential pace, not only in the decipherment of the Maya glyphs, but also towards the construction of a new, historically based understanding of Maya civilization. Scholars such as J. Kathryn Josserand, Nick Hopkins and others published findings that helped to construct a Mayan vocabulary.[25] In 1988, Wolfgang Gockel published a translation of the Palenque inscriptions based on a morphemic rather than syllabic interpretation of the glyphs. The "old school" continued to resist the results of the new scholarship for some time. A decisive event which helped to turn the tide in favor of the new approach occurred in 1986, at an exhibition entitled "The Blood of Kings: A New Interpretation of Maya Art". It was organized by InterCultura and the Kimbell Art Museum and curated by Schele and Yale art historian Mary Miller. This exhibition and attendant catalogue—and international publicity—revealed to a wide audience the new world which had latterly been opened up by progress in decipherment of Maya hieroglyphics. Not only could a real history of ancient America now be read and understood, but the light it shed on the material remains of the Maya showed them to be real, recognisable individuals. They stood revealed as a people with a history like that of all other human societies: full of wars, dynastic struggles, shifting political alliances, complex religious and artistic systems, expressions of personal property and ownership and the like. Moreover, the new interpretation, as the exhibition demonstrated, made sense out of many works of art whose meaning had been unclear and showed how the material culture of the Maya represented a fully integrated cultural system and world view. Gone was the old Thompson view of the Maya as peaceable astronomers without conflict or other attributes characteristic of most human societies.

However, three years later in 1989, a final counter-assault was launched by supporters who were still resisting the modern decipherment interpretation. This occurred at a conference at Dumbarton Oaks. It did not directly attack the methodology or results of decipherment, but instead contended that the ancient Maya texts had indeed been read but were "epiphenomenal". This argument was extended from a populist perspective to say that the deciphered texts tell us only about the concerns and beliefs of the society's elite, and not about the ordinary Maya. Michael Coe in opposition to this idea described "epiphenomenal" as "a ten penny word meaning that Maya writing is only of marginal application since it is secondary to those more primary institutions—economics and society—so well studied by the dirt archaeologists."[26]

Linda Schele noted following the conference that this is like saying that the inscriptions of ancient Egypt—or the writings of Greek philosophers or historians—do not reveal anything important about their cultures. Most written documents in most cultures tell us about the elite, because in most cultures in the past, they were the ones who could write (or could have things written down by scribes or inscribed on monuments).

Progress in decipherment continues at a rapid pace today, and it is generally agreed by scholars that over 90 percent of the Maya texts can now be read with reasonable accuracy.[attribution needed] As of 2008, at least one phonetic glyph was known (or had been proposed) for each of the syllables marked in this chart. Based on verbal inflection patterns, it would seem that a syllabogram for [wu] did not exist rather than simply being unattested. No syllable in [p’] is attested.[27]

Syllables for which at least one phonetic glyph has been found
(’) b ch ch’ h j k k’ l m n p p’ s t t’ tz tz’ w x y
a Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
e Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
i Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
o Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
u Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes

Notes[edit]

  1. ^ K. Kris Hirst (6 January 2006). "Maya Writing Got Early Start". Science. 
  2. ^ "Symbols on the Wall Push Maya Writing Back by Years". The New York Times. 2006-01-10. Retrieved 2010-05-12. 
  3. ^ Houston, Stephen D.; Robertson, John; Stuart, David (2000). "The Language of Classic Maya Inscriptions". Current Anthropology 41 (3): 321–356. doi:10.1086/300142. ISSN 0011-3204. PMID 10768879. 
  4. ^ a b Kettunen and Helmke (2005, p.12)
  5. ^ See here for a more substantial discussion and, from page 70 on, a partial list of glyphs and glyph blocks
  6. ^ Lacadena García-Gallo, A. and A. Ciudad Ruiz (1998). Reflexiones sobre la estructura política maya clásica. Anatomía de una Civilización: Aproximaciones Interdisciplinarias a la Cultura Maya. A. Cuidad Ruiz, M. I. Ponce de León and M. Martínez Martínez. Madrid, Sociedad Española de Estudios Mayas: 31-64
  7. ^ Berlin, H. (1958). "El Glifo Emblema en las inscripciones Maya." Journal de la Société des Américanistes de Paris 47: 111-119
  8. ^ Marcus, J. (1976). Emblem and state in the classic Maya Lowlands: an epigraphic approach to territorial organization. Washington, Dumbarton Oaks Trustees for Harvard University
  9. ^ Marcus, J. (1973) Territorial Organization of the Lowland Classic Maya. Science. 1973 June 1;180 (4089):pp. 911-916
  10. ^ See Mathews (1991)
  11. ^ Houston, S. D. (1986). Problematic emblem glyphs: examples from Altar de Sacrificios, El Chorro, Río Azul, and Xultun. Washington, D.C., Center for Maya Research
  12. ^ Stuart, D. and S. D. Houston (1994). Classic Maya place names. Washington, D.C., Dumbarton Oaks Research Library and Collection
  13. ^ Houston (1993; in particular, pp.97–101)
  14. ^ Source: A.Tokovinine(2006) People from a place: re-interpreting Classic Maya "Emblem Glyphs". Paper presented at the 11th European Maya Conference "Ecology, Power, and Religion in Maya Landscapes", Malmö University, Sweden, December 4–9, 2006
  15. ^ S. Gronemeyer (2012) Maya Political Relations and Strategies. Proceedings of the 14th European Maya Conference, Cracow, 2009 [Contributions in New World Archaeology, 4], edited by Jarosław Źrałka, Wiesław Koszkul and Beata Golińska: 13-40. Cracow: Polska Akademia Umiejętności and Uniwersytet Jagielloński
  16. ^ Information panel in the Museo Regional de Antropología in Mérida (state of Yucatán), visited on 2010-08-04
  17. ^ Saturno, William A.; David Stuart; Boris Beltrán (3 March 2006). "Early Maya Writing at San Bartolo, Guatemala". Science 311 (5765): 1281–1283. doi:10.1126/science.1121745. PMID 16400112. 
  18. ^ Yuri Knorozov at Britannica
  19. ^ (Russian)Yuri Knorozov
  20. ^ Coe, M. (1992), pp. 151
  21. ^ a b Coe, M. (1992), pp. 147
  22. ^ Coe, M. (1992), pp. 125-144
  23. ^ Coe, M. (1992), pp. 164
  24. ^ Coe, M. (1992), pp. 167-184
  25. ^ Google Docs
  26. ^ Coe, M. (1992), p. 268 (3rd ed)
  27. ^ Kettunen & Helmke (2008) pp 48–49

Bibliography[edit]

External links[edit]

Media related to Maya writing at Wikimedia Commons