Further Evidence for Diffusion Between the Mayan and Isthmo-Colombian Areas:  The “Transformation” Motif

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

25/4/2026

In a presentation in 2007 at the Denver Art Museum, I suggested the possibility that an artistic theme referred to in Maya archaeology as the Charlie Chaplin figurine theme for most of the past century had diffused from the Maya region to the Chibchan area of ancient Costa Rica (Mora-Marín 2007). The idea developed further, eventually coming out as an article in Cuadernos de Antropología in 2016 (Mora-Marín 2016). Figure 1 illustrates the basic point-for-point correspondences between the Mayan-style figurines (Figures 1A and 1C) and the Chibchan greenstone pendants in question (Figure 1B). No one iconographic element or motif is obligatory in the Mayan version, but the most common ones are the sideways-pointing feet and the arms-bent-at-elbows components (Traits 2 and 4).

 

Figure 1. Comparison of Mayan Ritual Specialist Theme with Chibchan counterpart.

 

 

I proposed, based on the available dates, that the Mayan theme, itself a local variant of a pan-Mesoamerican theme that can be traced back to Olmec art in the Middle Preclassic (1000–400/300 BCE) period, with examples dating to as early as the end of the Late Preclassic (400/300 BCE-CE 200) period, served as the inspiration for the Chibchan Ritual Specialist Theme, whose earliest examples are not well dated, but may start around CE 100–500. I proposed that the two represent the same idea: a Ritual Specialist Theme.

 

Figure 2 shows how the Chibchan artists adapted the theme, either in its plain form (Figure 2A), or with elaborations based on local artistic motifs and themes (Figures 2B–C), such as the double-zoomorphic motif on top of the head (Figures 2B–C), sometimes shown as avian and other times as reptilian, as well as the occasional addition of miniaturized trophy heads (Figure 2C) and reptilian feet (Figure 2C).

 

Figure 2. Examples of Chibchan Ritual Specialist Theme with and without local elaborations.

 

I also proposed that this borrowed Ritual Specialist Theme influenced the development of new themes or subthemes in the Chibchan artisan’s repertoire. Figure 3 illustrates the possibility —raised in that same paper— that the Mayan theme (Figure 3A) may have inspired not only the Chibchan version of the Ritual Specialist Theme (Figure 3B), but also the evolution of the Chibchan Anthropomorphic Celtiform Pendant Theme (Figures 3D–E) as a result of convergence or conflation of the earlier Chibchan Avian Celtiform Pendant Theme (Figure 3C) and the Chibchan Ritual Specialist Theme (Figure 3B). Moreover, as also seen in Figure 3, I suggested that the Chibchan avian celtiform pendant theme, which had appeared already around 500 BCE in the Pacific Northwest of Costa Rica, combined with the new Ritual Specialist Theme, led to the creation of the anthropomorphic celtiform pendant theme.

 

Figure 3. Development of Anthropomorphic Celtiform Pendant from conflation of Ritual Specialist and Avian Celtiform Pendant.

 

In other words, there is now reasonable evidence that there was diffusion of information —artistic, ideological— between the two regions, the Maya region and the Chibchan region of ancient Costa Rica, more specifically. It is also evident that this artistic theme in turn influenced local artistic developments. This was not an isolated phenomenon: it was no doubt a consequence of the interregional interaction network of exchange in preciosities, involving jadeite and other greenstones, ceramics, and likely other types of items, such as Spondylus shell, among others. Such exchange was responsible for the arrival and ultimate deposition of Chibchan-style jade pendants from Costa Rica at Maya sites (in royal burials and cosmopolitan caches), and Mesoamerican-style (Olmec, Teotihuacan, Zapotec, Mayan) artifacts in Costa Rica, including several dozen whole or fragmented Mayan royal jade belt plaques (e.g. Reents-Budet and Fields 1991; Fields and Reents-Budet 1992; Mora-Marín 2002, 2008; Reents-Budet et al. 2016, 2018; Mora-Marín et al. 2018).

 

The present note raises the possibility of another example of iconographic diffusion, possibly as a simple motif, but perhaps, at times, as a more comprehensive theme. I initially thought of this new case after John Hoopes raised the possibility of the two of us co-authoring a paper on the topic of interaction between Mesoamerica and Isthmo-Colombia, following his participation in the 10th European Maya Conference in 2005, with the theme “The Maya and Their Neighbours – Internal and External Contacts Through Time.” In the end, however, many factors conspired against our collaboration, and we were not able to contribute to the proceedings volume. I then thought some more about the case, and gathered a few examples, and presented them in public on my own (Mora-Marín 2016b, 2021), and with my colleague Dorie Reents-Budet 2021 (Mora-Marín and Reents-Budet 2021).

 

This new case can be referred to as a “Transformation” motif because of its function in ancient Mayan writing, and some of its associations in the Chibchan art from Costa Rica. I should note that not all cases of its occurrence in Mayan art have been studied, and as a result, its meanings and functions in artistic contexts are not yet known; similarly, the range of contexts in Chibchan art also shows great variation, so that it is not possible to know with great certainty whether it conveyed the same meaning or had the same function in all cases. The sign in question is ZS5/T645, which depicts two interlocked bands or a single twisted band, sometimes showing a series of dots or spots, as seen in Figure 4A, and only partially, as explained next, ZAD/T539, which depicts a face partially covered on one side with jaguar skin/pelt, is seen in Figure 4B. They appear to be allograms: each one is a logogram, a sign representing a lexeme or word, and they both bear the same value, WAY for Ch’olan *wäy ‘co-essence’ (Houston and Stuart 1987), and in some cases seemingly related to the notion of ‘shape-shifter’ as well. There are other signs in Mayan writing that may also be allograms with this same value, but they are not relevant to the problem at hand. (There is another Mayan grapheme that appears to be graphically related to ZS5, but bears a different value: YS5/T676 TAL. A future note will discuss these graphic relationships and explore possible semantic and contextual associations between the different values.)

 

Figure 4. ZS5 and ZAD, allograms for WAY ‘co-essence; shape-shifter’. Exemplars from from Maya Hieroglyphic Database (Looper and Macri 1991–2026).

 

As Figure 5 shows, the two signs briefly overlap during the Early Classic (CE 200–600) period, but it appears that ZS5 was the more common of the two during that time period, with dated examples bearing a mean date of 433.6. Nevertheless, ZAD had replaced ZS5 before the end of the Early Classic period, and was the most common allogram by far with the value WAY for *wäy ‘co-essence; shape-shifter’ thereafter, showing a mean date of 727.1.

 

Figure 5. Distribution of ZS5 and ZAD allograms for WAY in dated texts based on data from Maya Hieroglyphic Database (Looper and Macri 1991–2026).

 

Figure 6 shows a few examples from actual texts. Importantly, note that ZS5 can include a line that spans at least part of the length of the interlocked bands or single twisted band (Figures 6A and 6E), in addition to dots or spots (Figures 6A–D), or just the line dividing the interlocked bands or twisted band (Figures 6D–E). Also, note that some cases show a transition from a simple ZS5 to ZAD (Figures 6D–E), showing the partial covering with jaguar skin/pelt; Figure F shows an example of ZAD, which is composed of half of ZA1/T533 on the left side and half of a jaguar skin/pelt. The apparent transition supports the notion that ZS5 and ZAD were in fact allograms, signs bearing the same value.

 

Figure 6. Gradual evolution of ZS5 into ZAD?

Inscriptions indicated with codes from Maya Hieroglyphic Database (Looper and Macri 1991–2026). a) COLK3261:A02. Drawing by the author. b) PNGPan12:M01. Drawing by Dana Moot II. c) TIKSt39:zA01. Drawing in Laporte and Fialko (1995:Fig. 37). d) COLCCVCox9172:I. Drawing by Lin Crocker in Hellmuth (1987:Fig. 434), courtesy of Nicholas Hellmuth and FLAAR, www.Maya-archaeology.org. e) TIKSt31:K01. Drawing by William R. Coe in Jones and Satterthwaite (1982:Figs. 51–52). f) PALTFC:O03. Drawing by Linda Schele, Copyright David Schele (SD-172). Photo courtesy Ancient Americas at LACMA (ancientamericas.org).

 

For the present purposes, the sign that matters is really only ZS5, the sign depicting either two interlocked bands, or a single twisted band. It is this allogram that can be connected to the Chibchan motif seen in Figure 7. Note especially the example in Figure 7C, showing the line dividing the dots or spots (cf. Figures 6A–C), as well as the line spanning much the length of the bands (cf. Figures 6A and 6C).

 

Figure 7. Motifs corresponding to ZS5 in Chibchan art from Costa Rica.

a) Motif on back of polychrome effigy vessel. Drawing by the author. b) Motif on cheek of Carrillo Polychrome anthropomorphic effigy vessel (cf. Snarskis 1981:Pl. 14, Cat. no. 82). Drawing by the author. c) Detail from abdomen of female figure on vessel from Museo del Jade (#4817). Drawing by the author based on photograph taken on site. d) Elaborate version of motif on back of Galo Polychrome effigy vessel. Drawing by the author. e) Elaborate version of motif on back of Guinea Incised effigy vessel. Drawing by the author.

 

In fact, in Mayan art and writing, it appears that ZS5 (Figure 8A) is a simplified or abbreviated version of an iconographic motif involving two interlocked bands (Figure 8B), attested on a polychrome vessel housed at LACMA, as observed by Dorie Reents-Budet (personal communication, 2021). This version is graphically identical to some versions of grapheme YS5, which bears the logographic TAL. Previously, John Hoopes (personal communication, 2005) had suggested a connection with a symbol appearing in the Codex Mendoza in association with silver smithing, while Dorie Reents-Budet (personal communication, 2021) also observed its presence on an Aztec jade head (Coyolxauhqui sculpture) where it serves as a tag on a representation of a gold bell. It is possible the that common theme among these uses of ZS5 has to do with “transformation.”

 

Figure 8. Mayan ZS5 grapheme and Motif painted on lid of polychrome vessel M.90.104a-b at Los Angeles County Museum of Art. Drawings by the author.

 

Grapheme ZS5 is attested in at least one Mayan jade pendant recovered from Costa Rica, a fragment of an Early Classic royal belt plaque in the Metropolitan Museum of Art collection, seen in Figure 9. This example shows ZS5 atop a grapheme depicting a SKULL, possibly for JOL or B’AK, itself atop a grapheme resembling the syllabogram so. Together these three graphemes likely spell a glyphic name of a ruler, likely embedded in the headdress of a portrait of said ruler, though this cannot be determined for sure. Thus, ZS5 can be shown to have been seen by Chibchan artisans. (So was YS5/T676 TAL, the grapheme mentioned above to be graphically related to ZS5. YS5/T676 TAL is also attested in an Early Classic Maya royal jade belt plaque reported from Costa Rica.)

 

Figure 9. Reworked fragment of Early Classic Maya royal jade belt plaque from Metropolitan Museum of Art. Note ZS5 in probable ruler’s name phrase. Drawing by the author.

 

Also it is worth noting a possible association with transformation, specifically with shape-shifting. John Hoopes (personal communication, 2005) noted that the Galo Polychrome “man” (Figure 10A) shows evidence of reptilian traits, perhaps crocodilian specifically (Figures 10B–C). Hoopes (2004, 2007, 2023) has discussed the evidence —ethnohistorical and archaeological— for the concept of therianthropic transformation of shamans, in particular, in the Isthmo-Colombian region, including the Chibchan peoples of Costa Rica. Hoopes and Mora-Marín (2009:309–313) have also offered some comparative remarks on therianthropic transformation and its representation in both Chibchan and Mayan art.

 

Figure 10. Galo Polychrome effigy vessel from Museo Nacional de Costa Rica.

 

On the effigy’s back (Figure 11), we see one of the interlocked band motifs already illustrated (Figure 7D). Thus, this shows an association between the motif and the notion of transformation, or at the very least, of an association with some sort of crocodilian alter ego or co-essence.

 

Figure 11. Back of Galo Polychrome effigy vessel from Museo Nacional de Costa Rica showing Chibchan version of ZS5.

 

Another example, seen in Figure 12, consists of a Galo Polychrome effigy vessel in the shape of a jaguar, though one that suggests a human pose. It also bears an elaborate version of the intertwined motif on its back.

 

Figure 12. Galo Polychrome effigy vessel (jaguar) from Museo del Jade.

 

Examples also appear in association with bat imagery; there is ample evidence for associations between ritual specialists and bats in Chibchan art across several media (e.g. jade and ceramic). Figure 13 illustrates an example.

 

Figure 13. Guinea incised effigy vessel (bat) from Museo del Jade with ZS5 as motifs on wings.

 

The motif appears also on stone metates, like the example seen in Figure 14. The metate is carved partly in the shape of a macaw, once again suggesting an association with animal transformation or co-essences. This type of metate has been estimated by Chaves Chaves and Fontana Coto (1993) to date to ca. CE 300–700.

 

Figure 14. Andesite tripod metate from Museo del Jade.

 

The motif is also found in contexts where there is no obvious connection to the concept of animal transformation or co-essences. Figure 15A shows an instance incised on the pronounced belly of a female Guinea incised effigy vessel. Figure 15B isolates the motif, already illustrated in Figure 7C. It can be compared to Mayan ZS5 in Figure 15C. Perhaps the transformation of interest here has to do with pregnancy and birthing.

 

Figure 15. Guinea incised effigy vessel from Museo del Jade.

 

Figure 16 provides similar examples on another female Guinea incised effigy vessel, bearing the example of the motif already illustrated in Figure 7E.

 

Figure 16. Guinea incised effigy vessel from Museo del Jade. 

 

Figure 17 provides an example of the motif, already illustrated in Figure 7A, on the back of a Carrillo Polychrome effigy head vessel. Though there is no obvious connection to any kind of transformation or animal co-essence, perhaps the motif marks the portrayed individual as a person who had the power to shape-shift.

 

Figure 17. Carrillo Polychrome vessel.

 

Figure 18 shows two more Carrillo Polychrome effigy head vessels, this time showing the motif in question on the head’s cheeks.

 

Figure 18. Carrillo Polychrome effigy head vessels.

 

Finally, Figure 19 shows an example of a ceramic stamp from the Museo del Jade, probably the exact same device used to apply the motif on a person’s cheek.

 

Figure 19. Stamp from Museo del Jade collection.

 

The case at hand is also interesting because it can be suggested that the direction of diffusion was the reverse of the Ritual Specialist Theme: from the Chibchan area of Costa Rica to the Maya region. This is supported, tentatively, by chronology: Table 1 shows the temporal ranges for the earliest ceramics where this motif is attested.

 

Table 1. Temporal ranges of ceramics with early attestations of “Transformation” motif.

 

It is quite possible, based on these temporal ranges, starting as early as 300 BCE and 100 BCE, that the Chibchan version of the motif preceded the Mayan version, ZS5, which is so far attested in dated texts only as early as the fourth century. However, a version of ZS5 may be attested in at least one Late Preclassic inscription, Kaminaljuyu Monument 10, the subject of a future note. But at the very least, it seems like this is another possible case of interregional diffusion between Mayan and Chibchan societies.

 

Acknowledgments. I would like to thank Sergio García Piedra at the Museo del Jade for his help in finding additional photos of several relevant artifacts. I am also especially indebted to John Hoopes and Dorie Reents-Budet for discussions of this and related topics.

 

Appendix Figure: YS5 for TAL.

One graphic variants of YS5, specifically example A, is identical to the Chibchan motif in question.

References

Chaves Chaves, Sergio, and Amalia Fontana Coto. 1993. Lítica precolombina: artefactos de piedra de la colección del Instituto Nacional de Seguros. San José de Costa Rica: Instituto Nacional de Seguros.

Hellmuth, Nicholas M. 1987. Monster und Menschen in der Maya-Kunst: eine Ikonographie der alten Religionen Mexikos und Guatemalas. Graz, Austria: Akademische Druckund Verlagsanstalt.

Hoopes, John W. 2004. Atravesando fronteras y explorando la iconografia sagrada de los antiguos chibchas en Centroamerica meridional y Colombia septentrional. Arqueologia del Area Intermedia 6:129-166.

Hoopes, John W. 2007. Sorcery and the Taking of Trophy Heads in Ancient Costa Rica. In The Taking and Displaying of Human Body Parts as Trophies by Amerindians, edited by Richard J. Chacon and David H. Dye, 444–80. New York: Springer.

Hoopes, John W. 2023. Teriantropía, hechicería y curación: el papel de los actores sobrenaturales de cambio de forma en el Área Istmocolombiana. Paper presented at the Coloquio Internacional Espacios Sagrados: Umbrales y Fronteras, on May 25, 2023. Oaxaca de Juárez, Oaxaca, México.

Hoopes, John W., and David Mora-Marín. 2009. Violent Acts of Curing: Pre-Columbian Metaphors of Birth and Sacrifice in the Diagnosis and Treatment of Illness ‘Writ Large’. Book Chapters. In Blood and Beauty: Organized Violence in the Art and Archaeology of Mesoamerica and Central America, edited by Heather Orr and Rex Koontz, pp. 291–330. Cotsen Institute of Archaeology.

Houston, Stephen, and David Stuart. 1987. The Way Glyph: Evidence for “Co-Essences” among the Classic Maya. Research Reports on Ancient Maya Writing 30.

Jones, Christopher, and Linton Satterthwaite. 1982. The Monuments and Inscriptions of Tikal: The Carved Monuments. Tikal Report 33, University Museum Monograph 44. Philadelphia: University Museum, University of Pennsylvania.

Laporte, Juan Pedro, and Vilma Fialko. 1995. Un Reencuentro con Mundo Perdido, Tikal, Guatemala. Ancient Mesoamerica 6:41–94.

Looper, Matthew G. and Martha J. Macri. 1991-2024. Maya Hieroglyphic Database. Department of Art and Art History, California State University, Chico. URL: http://www.mayadatabase.org/.

Mora-Marín, David F. 2002. An Epi-Olmec Jade Pendant Found in Costa Rica. Mexicon 24:14–19

Mora-Marín, David F. 2007. A Brief History of the Middle American Jade Exchange Network: The Costa Rica/Maya Interaction. Invited paper for conference: Costa Rica and the Pre-Columbian World: Honoring the Contributions of Frederick Mayer, Saturday, November 10, 2007. Sharp Auditorium (Hamilton Building), Denver Art Museum.

Mora-Marín, David F. 2008. Notes on Three Foreign Artifacts from the Cerro de las Mesas Jade Cache. Mexicon 30(1):20–22.

Mora-Marín, David F. 2016a. The “Charlie Chaplin” Silhouette Figural Theme: A Pan-Middle American Ritual Performer Theme. Cuadernos de Antropología 26(1):9–45.

Mora-Marín, David F. 2016b. Textos jeroglíficos mayas en Costa Rica: ¿Qué nos enseñan sobre la historia de las culturas precolombinas de Costa Rica? Presented at the Museo del Jade Fidel Tristán in San José, Costa Rica.

Mora-Marín, David F. 2021. Las inscripciones mayas halladas en Costa Rica: métodos, fechas, eventos, personajes y contextos históricos. Paper presented on Wednesday, February 24th, 2021, as part of series Ciclo de Conferencias del Museo del Jade del Institution Nacional de Seguros, hosted by the Museo del Jade (Jade Museum) in San José, Costa Rica.

Mora-Marín, David, and Dorie Reents-Budet. 2021. Intercambio a larga distancia y adaptación local de artefactos foráneos: Conexiones entre las sociedades prehispánicas de Costa Rica y Mesoamérica. Keynote presented at the Segundo Ciclo de Conferencias, Estilo y tecnología en arqueología, Museo del Templo Mayor, Mexico City, Wednesday 17, November 2021.

Mora-Marín, David, Dorie Reents-Budet, and Virginia Fields. 2017. Costa Rica and the Maya: Prestige Goods and International Relations. In Trading Spaces: The Archaeology of Migration, Interaction, and Exchange, edited by Margaret Patton, and Jessica Manion, pp. 181–188. Proceedings of the 46th Annual Chacmool Conference. Calgary: Chacmool Archaeological Association, University of Calgary.

Mora-Marín, David, Dorie Reents-Budet, and Virginia Fields. 2018. El estudio epigráfico e iconográfico de las placas de jade y discos de pizarra mayas encontrados en Costa Rica. In XXXI Simposio de Investigaciones Arqueológicas en Guatemala 17 al 21 de julio, 2017, Tomo I, editado por Bárbara Arroyo, Luis Méndez Salinas y Gloria Ajú Álvarez, pp. 143–156. Guatemala: Asocación Tikal.

Reents-Budet, Dorie, David Mora-Marín, and Virginia Fields. 2018. El intercambio de preciosidades entre los mayas de Mesoamérica y los chibchas de Costa Rica: El caso de las placas de jade y discos de pizarra mayas encontrados en Costa Rica. In XXXI Simposio de Investigaciones Arqueológicas en Guatemala 17 al 21 de julio, 2017, Tomo I, editado por Bárbara Arroyo, Luis Méndez Salinas y Gloria Ajú Álvarez, pp. 131–142. Guatemala: Asocación Tikal.

The -he-wa Distance Number Glyph as ʔew/hew ‘day’ based on the Central Mayan term *ʔeew-ii(r) ‘yesterday’

 

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

3/21/25

In this note I review the evidence for the explicit spellings of the Distance Number glyph for ‘day’ in Epigraphic Mayan, and propose a reading based on a term reconstructed to Central Mayan as *ʔeew-ii(r) ‘yesterday’ by Kaufman with Justeson (2003:1448–1449). More specifically, I propose that the spellings NUMERAL-he-wa ~ NUMERAL-he represent a root ʔeew ‘day’, implied by the etymology of *ʔeew-ii(r) ‘yesterday’, a term that must have been lost from all the Ch’olan, Tzeltalan, and Yucatecan languages, but which must have been retained into proto-Ch’olan.

 

First, I begin by pointing out the signs, seen in Figure 1, that are used in the spellings of interest here, using the exemplars from the Maya Hieroglyphic Database (MHD) by Looper and Macri (1991–2026). The AV3 sign is a “deer (Odocoileus virginianus) with crossed bones in eye” (Looper and Macri 1991–2026).

 

 

Now, as is well known, Mayan scribes used Distance Numbers to express the amount of time that had transpired between events, and also, in general, they did not spell out the period name ‘day’ in such DN sequences, providing only a numeral for the corresponding number of days. Using datasets compiled by means of the MHD, it is apparent that about 82.2% of DNs did not provide an explicit spelling for ‘day(s)’ after the corresponding numeral. In the remaining number of cases, they employed one of several spellings to indicate the notion of ‘days’, but none of these represents the more common term for ‘day’, proto-Ch’olan *k’in ‘day; sun’. Table 1 lists the spellings in question. Table 2 does the same but this time showing a breakdown by Time Period, distinguish Early Classic (ca. CE 200–600) and Late Classic (ca. CE 600–900).

 

 

Figure 2 is a Box Plot that shows the distribution of these various spellings in dated texts. Although spellings using only -wa are very few, and not always very clear, they are in general the earliest ones, followed by -he-wa spellings. After these two, it is a bit trickier to determine with certainty the progression, though it is very likely that -he spellings followed in third place, that –AV3-na spellings appeared fourth. After these four spelling types, it is not obvious what the progression was. Figure 3 simply breaks down the spelling types by Period.

 

 

Next, Table 3 and Figure 4 allow us to say a few things about the regional distribution of these spellings.

 

What is most important for our purposes here is the temporal distribution of four spelling types: -he, -he-wa, –AV3, and –AV3-wa. As seen in Figure 5, the -he-wa and -he spellings precede the similar –AV3-wa and –AV3 spellings. It is only during the middle of the sixth century when the –AV3-wa begins to appear with some frequency, and the same may be said of the –AV3 spellings.

 

 

So, the -wa, -he-wa, -he spellings, which constitute the earliest spellings, point to a form of the general shape <hew>. The -he-na spellings, clearly the latest of them all, point to a form of the general shape <hen>. The uses of AV3 in these spellings mirror those of ZR7 he, as discussed in more detail below. Whether these <hew> and <hen> forms are grammatical or lexical morphemes is an open question. This is where a review of the literature can offer some insights, with a focus, first, on what has been said about the spelling patterns and possible readings and values of the expression in question.

 

First, Lacadena and Wichmann (2005:33) suggested that these forms represent a sequence of suffixes or enclitics, with the he-wa sequence cueing a -(V)h-e’w suffix sequence, and the he-na sequence a -(V)h-e’n suffix sequence. The authors admit to two problems with their proposal. Their proposed values -e’w and -e’n for the second suffix in the sequence is derived from their proposal for the diacritic use of synharmonic and disharmonic syllabograms (Lacadena and Wichmann 2004). First, their putative -(V)h suffix in this suffix sequence would correspond, in their analysis, to proto-Ch’olan *-ij ‘in the future’, which bears /j/ instead of /h/, and yet, as Grube (2004) has shown, the scribes distinguished between /h/ and /j/ systematically since the Early Classic. And second, they admit that “[they] are not sure” what their proposed -e’w and -e’n suffixes might be. A third problem would be that the enclitic of Distance Numbers referring to periods of time above days is a sequence *-ij-iy, spelled -ji-ya, usually abbreviated orthographically to just the -ya. This sequence consists of *-ij ‘in the future’ and *-iy ‘in the past’, and together in a sequence they function as *-ij-iy ‘in the past’. For their analysis to work, the putative suffixes -e’w and -e’n would have to function in the same manner as *-iy ‘in the past’. No suffixes or enclitics with such shapes and functions are known from any Mayan language.

 

Next, Carter (2009:14) credits Marc Zender and Stephen Houston with the suggestion that AV3 in this context constitutes a logographic sign for ‘day’ representing a lexeme distinct from *k’in ‘sun; day’, and points out that AV3 is not employed in contexts where it can be analyzed as he.

 

Davletshin and Houston (2021) transcribed the spelling AV3-wa as heew, and transliterated the AV3 sign as HEEW, under the assumption that the use of wa, instead of we, was meant to indicate that that the vowel of the root was complex, in this case long, /ee/ (cf. Houston et al. 1998, 2004). Mora-Marín (2010) has presented evidence against the notion that disharmonic vowels served to indicate vowel complexity of a preceding syllable nucleus, and thus, the use of wa is not considered to be indicative of a diacritic function (i.e. vowel complexity of preceding syllable nucleus) in this note.

 

More recently, Kelly (2022:169) observed that, despite the proposal that AV3 was logographic, the spelling patterns support the idea that AV3 is merely another syllabogram with the value <he>, like ZR7 he, and thus, that AV3 was he₂. Indeed, as is clear from Table 4, the -he spelling with ZR7 is similar to the –AV3 spelling type, the spelling that makes AV3 “look” like a logogram, and in fact, ZR7 is more frequent in such spelling type than AV3, which would suggest, by such criterion, that ZR7 is also a logogram in such contexts. Kelly (2022:169) also concedes that “Of course, since [AV3] is not known in other contexts outside of providing the “day” period name in Distance Numbers, the identification of this as he is tenuous.”

 

Most recently, Carter et al. (2025:271) have presented strong evidence that he-wa spelled a lexical morpheme, not a grammatical morpheme. The upper text panel on Ixkun Stela 4, in particular, appears to bear a prepositional phrase ta-he-wa for tä hew, which they gloss as ‘on the same day’. Here it is clear that he-wa cannot be a suffix, but a nominal expression functioning as the complement of the preposition tä. However, there is nothing in the expression ta-he-wa that would call for ‘the same’ as part of a gloss. Instead, it seems that tä hew calls for a translation as ‘on a day’.[1]

 

With this in mind, the likelihood that the expression represents a lexical root or stem, Table 5 presents the data from Kaufman with Justeson (2003) for the etymon *ʔeew-ii(r) ‘yesterday’, which they reconstruct to proto-Central Mayan (i.e. Eastern Mayan plus Western Mayan). This etymon is absent from the Greater Lowland Mayan languages (Ch’olan, Tzeltalan, Yucatecan). Kaufman with Justeson (2003:1449) propose an areal innovation *wol-ej-eer ‘yesterday’, attested in Yucatecan and Tzeltalan, but not Ch’olan. The Ch’olan languages have preferred the use of forms derived from proto-Ch’olan *ʔahk’äb’  ‘night’ for expressions for ‘yesterday’, reconstructible as *ʔäk’b’-i ‘yesterday’, with -i ‘earlier (in the past)’.

 

 

The proto-Central Mayan form *ʔeew-ii(r) ‘yesterday’ seems curious, as it contains a reflex of the enclitic *-eer ‘in the past’, and since it is in the Ch’olan languages where this enclitic experienced a shift of *ee > ii > i, the reconstruction with *-ii(r) seems problematic. And yet, not only is this expression unattested in the Ch’olan languages, but it seems that most of the cognates bear evidence of a vowel /i/ or /ii/ in the enclitic or suffix. The few exceptions showing /e/ (e.g. Q’eqchi’ ewer(-aq)) could be explained as the result of perseverative vowel assimilation at a distance. However, what matters for now is this: if the form *ʔeew-ii(r) ‘yesterday’ acquires its meaning via derivation by means of the suffix or enclitic ‘in the past’, then without it, the root *ʔeew should mean ‘day’. And this form, *ʔeew, presumably ‘day’, looks surprisingly similar to the proposed form <hew>, suggested by the vastly more frequent and earlier set of spellings -wa ~ he-w(a) ~ -he for ‘day’ in the Distance Number context. It is conceivable that this noun had become grammaticalized into a numerical classifier, or even into an enclitic or a suffix deriving a numeral term, and that as such, it would be omissible from orthographic expression. Indeed, the representation of numerical classifiers was the exception, not the norm, in Mayan texts, and suffixes, whether derivational or inflectional, may also be omitted orthographically.

 

With this in mind, then, I propose the reading ʔew/hew ‘day’, grammaticalized as a numerical classifier in the Distance Number context, resulting in -hew ~ -ew, and reanalyzed as /hew/ in cases where it was used as a noun (e.g. Ixkun Stela 4), from an earlier form ʔew < *ʔeew. As Figure 6 shows, I support the possibility that AV3 might be a syllabogram he₂, rather than a logogram.

 

 

What my proposal does not answer is the obvious lingering question: what about the he-na and AV3-na spellings? There is no evidence whatsoever for a root for ‘day’ of the general form <hen>, nor any phonological relationship between /n/ and /w/. A possible answer requires an excursus at this point.

 

In conjunction with the ‘day’ expression of Distance Number contexts, the other frequent context for ZR7 he is the quotative particle, reconstructed as proto-Ch’olan *cheʔ (Kaufman and Norman 1984:139). Before reviewing the cognates and the history of this expression, I will begin by introducing its Classic period equivalents, seen in Figure 7. The two most common spellings are che-ʔe-n(a) and che-he-n(a).

 

 

Unfortunately, because this particle tends to occur in portable texts in contexts that are less likely to be accompanied by calendrical information, the majority of instances lack dates. Table 6 provides their distribution by Period. The spelling che-ʔe-n(a) is rare, with only four examples, two from the Early Classic and two from the Late Classic. Thus far, the spelling che-he-n(a) is exclusively attested in Late Classic texts. Most cases of quoted speech are dated to the Late Classic; with only a very few cases from the Early Classic, three of them from the same site (Copan) and using the che-ʔe-n(a) expression (in one case preserved fragmentarily as -ʔe-n(a), which is why it was omitted from my dataset.

 

 

As Table 7 shows, the spelling che-ʔe-n(a) is restricted to the Southern and Central regions, while the spelling che-he-n(a) is more widely distributed, appearing in all except the Southern region. Figure 8 provides a visualization of the regional distribution. I suspect this wider distribution of che-he-n(a) could indicate that such form is the more conservative form, despite the lack of Early Classic attestations thus far, which is likely due to the fact that the genres documenting quoted speech became more common in the Late Classic period.

 

 

 

Kaufman with Justeson (2003:739) reconstruct a proto-Central Mayan morpheme *kih ‘quotative particle’ based on the cognates in Table 8. To these, I have added the Ch’ol cognate cheʔen, which matches the spelling che-ʔe-n(a) quite closely. The following scenario can be proposed: proto-Central Mayan *kih experienced the proto-Ch’olan-Tzeltalan *k > ch shift, resulting in *chih, which is what is attested in Tzeltal and Tzotzil (after taking into account the *h > Ø/__# characteristic of all Tzotzil varieties and some, but not all, of the Tzeltal varieties). Pre-Ch’olan likely inherited this form *chih, but Ch’olan speakers likely reshaped this morpheme by analogy with proto-Ch’olan *cheʔ ‘thus’, resulting in *cheʔ ‘quotative particle’. Then, proto-Ch’olan speakers added *-en, possibly from ‘first person singular absolutive’, which became amalgamated —reanalyzed as part of the root— for some speakers, attesting to its use in Ch’ol and also in Classic texts. What matters for the present purposes is that this reanalysis likely took place in stages: 1) *chih ‘quotative particle’ > 2) *chih ~ (*cheh ~) *cheʔ ‘quotative particle’ (reanalysis based on *cheʔ ‘thus’ with variable uses of both forms, possibly three forms, if the reanalysis applied first to the vowel, and later to the final consonant) > 3) *cheʔ ‘quotative particle’ and *cheʔ ‘thus’ (perhaps simply ‘quotative particle; thus’, as suggested by Kaufman and Norman (1984:139)). During the second stage, the forms *cheh and *cheʔ would have coexisted, and with the addition of *-en, the forms *cheh-en and *cheʔ-en would have coexisted as well, hence the Classic-period attestations.

 

 

Consequently, what is important here, is that an alternation between sequences <ʔen> and <hen> would have become common place in the language, resulting in the che-ʔe-n(a) and che-he-n(a) alternation in the texts. Also, in this scenario, the che-he-n(a) spellings would reflect the more conservative form, where the particle still retained its original *h from proto-Central Mayan *kih.

 

Finally, the spelling of this particle as che-he-n(a), during the Late Classic period, especially during the eighth century, could have led scribes to associate the two most common uses of ZR7 he: the one in the Distance Number context as -he-wa, and the one in the quotative particle context as -he-na. It is in fact especially during the late seventh century and the entirety of the eighth century when we see the -he-na and the few –AV3-na spellings of the Distance Number ‘day’ spellings. As suggested in Figure 9, then, the -he-na and –AV3-na could be the result of orthographic analogical reanalysis, during a time when it is quite likely that the root ʔew/hew‘day’ was already obsolete, and perhaps no longer used outside of the highly specialized context of Distance Number counts on stone monuments.

 

 

To conclude, I propose the Distance Number expression for ‘day’ was ʔew/hew ‘day’, and that the he-na and AV3-na spellings may be analogical, orthographically artificial, taking place at a point in time when the root ʔew/hew may have already disappeared from the spoken registers.

 

Acknowledgments. I want to thank the folks who attended the Q&G meeting on 11/11/25, where I presented a very preliminary version of this idea; special thanks to John Justeson and Barb MacLeod for more detailed comments.

 

Endnotes

[1] Carter et al. (2025) propose an identity between AV2 and AV3. AV2, distinct from AV3 in that it contains an infixed JAWBONE in the deer’s eye, instead of the CROSSED.BONES, occurs on K1398, immediately after a Calendar Round date, and immediately before a verbal expression inflected in the first person singular, in a very different context from AV3. Here, the authors analyze AV2 as HEW and gloss it as ‘[on that] day’ despite the absence of a preposition, and with no accounting of the meaning ‘that’, which would require the presence of a deictic demonstrative, and in fact, quite possibly, a deictic demonstrative frame —a determiner before the noun, and a deictic demonstrative after the noun.

 

References

Carter, Nicholas P. 2009. Paleographic trends and linguistic processes in Classic Ch’olti’an: a spatiotemporal distributional analysis. Master’s thesis, Department of Anthropology, Brown University, Providence.

Carter, Nicholas P., Barbara MacLeod, Aliyah Anderson, and Jacob Lozano. 2025. The Enemy of My Enemy: Ixkun Stela 4 and Classic Maya International Politics on the Eve of War. Ancient Mesoamerica 36:264–283.

Davletshin, Albert, and Stephen Houston. 2021. Maya Creatures VI: A Fox Cannot Hide Its Tail. Maya Decipherment: Ideas on Ancient Maya Writing and Iconography. Posted January 8, 2021. http://decipherment.wordpress.com/2021/01/08/maya-creatures-vi-a-fox-cannot-hide-its-tail/.

Houston, Stephen, David Stuart, and John Robertson. 1998. Disharmony in Maya Hieroglyphic Writing: Linguistic Change and Continuity in Classic Society. In Anatomía de una civilización: Aproximaciones interdisciplinarias a la cultura maya, edited by A. Ciudad Ruiz, Y. Fernández Marquínez, and J. M. García Campillo, M. J. Iglesias Ponce de León, A. Lacadena García-Gallo, and L. T. Sanz Castro, pp. 275-296. Madrid: Sociedad Española de Estudios Mayas.

Houston, Stephen, David Stuart, and John Robertson. 2004. Disharmony in Maya Hieroglyphic Writing: Linguistic Change and Continuity in Classic Society. In The Language of Maya Writing, edited by Søren Wichmann, pp.  83–99. Salt Lake City: University of Utah Press.

Kaufman, Terrence, and William Norman. 1984. An outline of proto-Cholan phonology, morphology, and vocabulary. In Phoneticism in Maya Hieroglyphic Writing, edited by John S. Justeson and Lyle Campbell, pp. 77–166. Institute for Mesoamerican Studies Publication No. 9. Albany: State University of New York.

Kaufman, Terrence, with John Justeson. 2003. Preliminary Mayan Etymological Dictionary. (http://www.famsi.org/reports/01051/index.html) (Accessed January of 2017.)

Kelly, Mary Kate. 2022. Speech Carved in Stone: Language Variation Among the Ancient Lowland Maya. PhD Dissertation, Department of Anthropology, Tulane University.

Lacadena García-Gallo, Alfonso, and Søren Wichmann. 2004. On the representation of the glottal stop in Maya writing. In The Linguistics of the Maya Script, edited by Søren Wichmann, pp. 100–164. Salt Lake City: University of Utah Press.

Lacadena García-Gallo, Alfonso, and Søren Wichmann. 2005. The dynamics of language in the Western Lowland Maya region. In Andrea Waters-Rist, Christine Cluney, Calla McNamee & Larry Steinbrenner (eds.), Art for archaeology’s sake: Material culture and style across the disciplines, Proceedings of the 33rd Annual Chacmool Conference (2000), 32–48. México, D.F.: Instituto Nacional de Antropología e Historia & Universidad Autónoma de Yucatán.

Looper, Matthew G. and Martha J. Macri. 1991-2024. Maya Hieroglyphic Database. Department of Art and Art History, California State University, Chico. URL: http://www.mayadatabase.org/.

Mora-Marín, David F. 2010. Consonant Deletion, Obligatory Synharmony, Typical Suffixing: An Explanation of Spelling Practices in Mayan Writing. Written Language and Literacy 13: 118-179.

Mora-Marín, David F. 2025. Some speculation on the explicit expression for ‘day’ in the Distance Numbers context. Quarantinis and Glyphs Presentation #64, 11/11/25.

Rzymski, Christoph and Tresoldi, Tiago et al. 2019. The Database of Cross-Linguistic Colexifications, reproducible analysis of cross- linguistic polysemies. Scientific Data 7, 13:1–12. https://clics.clld.org/. https://doi.org/10.1038/s41597-019-0341-x.

Linguistic Background to Pictish and the Pictish Symbols Writing System

 

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

16/3/26

This note is about some of the sources that review the evidence for the nature and affiliation of the language referred to as Pictish, the most likely candidate for the language of the ancient Pictish Symbols writing system (e.g. Forsyth 1997a). With this in mind, what is known about Pictish, and how can we learn more?

First, Pictish became extinct about a millennium ago, and very little in the way of attestations has survived to date. As Kenneth Jackson (1955:133) explained, there are five kinds of evidence: “(1) direct or indirect statements on Pictish by medieval writers while the language was still alive; (2) northern Scottish names in classical sources; (3) the [Ogham and Roman] inscriptions of Pictland; (4) names in medieval works such as Adamnan’s Life of St Columba, Bede’s Ecclesiastical History, and the Pictish Chronicle; and (5) the modern place-names of the Pictish area.” While this sounds like plenty of material to work with, basically every type of evidence is problematic in some way or another. The most direct type of evidence, the Pictish inscriptions in Ogham script, Jackson notes, “constitute a problem so thorny that many writers who have discussed the Pictish question, however confident they may have been, have carefully avoided attempting to deal with them” (1955:138).

Based on these various types of information, some have suggested that the Picts may have spoken a non-Indo-European language, a non-Celtic Indo-European language, a Brittonic Celtic language, a Goidelic Celtic language, a Celtic language (“Pritenic”) perhaps more closely affiliated with Transalpine Celtic/Gaulish than with either Brittonic or Goidelic, or perhaps two different languages combining two of the options just described. What is clear is that there were multiple languages in contact during the time of the Pictish Symbol Stones writing system, namely Brittonic, Goidelic, Latin, and Anglo-Saxon, and also that by the ninth and tenth centuries the Picts were shifting toward the use of a Goidelic variety, a process completed not long —a century or two— thereafter.

The Pictish Ogham and Roman inscriptions do not make it easier, but in fact, more complicated: there is evidence of influence from Latin and Goidelic, which as K. Jackson (1955:141–142) has explained, should be no surprise due to the influence of Christianity, for one, and the fact that Ogham scribal practices originated with Goidelic speakers, so their spread and that of Goidelic simply went hand-in-hand (along with the likely practice of “Gaelicising” Pictish names written in Ogham script), for another. Also, such inscriptions may bear evidence of Indo-European traits that have not been preserved in the Brittonic and Goidelic languages, such as the apparent presence of a singular genitive -s, according to authors like Rodway (2020) and Mees (2023a). However, their proposed singular genitive -s is not always present in what should be names in genitive case (e.g. those following the <MAQQ>/<MEQQ> ‘son’ expression in parentage formulas), which could suggest that the -s endings have a different function, or perhaps a variety of functions.

The Pictish Oghams also bear what seems like tentative evidence of verb-second word order, which Mees (2023a) analyzes as countering the notion that Pictish would have been a verb-initial language like Brittonic and Goidelic (Insular Celtic). Nevertheless, it would seem that this is not inconsistent, for as Eska (2017:1242) notes, “In the Middle phase of the Brittonic languages, however, the unmarked configuration is verb-second, with the initial constituent generally indicating topicality.”

For now, enough evidence exists from personal and place-names pointing to Celtic etymologies to support the hypothesis of Pictish as a Celtic language to follow the seemingly grown consensus that Pictish was Celtic, and most likely Brittonic (e.g. Forsyth 1996, 1997; Rhys 2015, 2020). Given this, and the evidence that Pictish was a P-Celtic language, that is, a language that experienced the shift of proto-Indo-European *kw > p, a shift no longer believed to be indicative of shared innovations and therefore diagnostic in subgrouping, but likely the result of independent innovations, I will assume three possible models, illustrated in Figures 1A-1C. The first involves a Gallo-Brittonic subgroup as a sibling to Goidelic within Core Celtic (cf. Koch 1992). The second, perhaps the more commonly assumed model, posits that Transapline Celtic and Insular Celtic are sibling subgroups within Core Celtic. The third proposes that Transalpine Celtic is really part of a Continental Celtic continuum, and also assumes an Insular Celtic subgroup (Eska 2024).

 

 

My goal moving forward will be to apply a more generalized linguistic framework, seen in Table 1, based on morphological and syntactic typology, elaborated through the comparison of Celtic languages. Eska (2017) and Stüber (2017) will constitute the foundation for such a framework, along with Fife’s (2009[1993]) characterization of diagnostic Celtic traits, and Stifter’s (2008:9–19) re-evaluation of such traits.

 

 

Table 1 suggests that, if Pictish is in fact an Insular Celtic language, and perhaps more narrowly a Brittonic language, then several of the traits should correlate with the head-complement order of V-initial clause structure (VSO). Identifying several of these morphosyntactic traits, occurring in tandem, especially those considered particularly diagnostic by Fife (2009), could offer strong evidence for the Celtic and perhaps even Insular Celtic nature of the Pictish Oghams and Pictish Symbols. For instance, if evidence of VSO clause structures (1), Bifurcated Demonstrative (Determiner + Noun + Demonstrative) phrases (2), and Confirming/Supplementary Pronouns constructions (3), were all attested, in combination, it could go a long way to confirming the Insular Celtic nature of such texts. Moreover, these are traits that might be identifiable, at least in part, through contextual and syntax-level structural analyses before the accomplishment of an actual decipherment.

 

(1) OLD WELSH (Eska 2017:1242), VSO

prinnit                      hinnoid             .iiii.             aues

buy.3.SG.PRES    DISTAL             four           birds

‘That buys four birds’

 

(2) MIDDLE WELSH (Eska 2017:1226), Bifurcated Demonstrative

y                pryd                 hwnnw

DEF         creature        DISTAL

‘that creature (lit. the creature there)’

 

(3) MODERN WELSH (Fife 2009:17), Confirming/Supplementary Pronouns

Dyma      fy        llyfr       i

here         my      book    I

‘Here is my book’

 

Additional sources are provided below that were not mentioned in this note, but which offer important background to the problem. Clearly the most important direct source on Pictish is made up of the corpus of Pictish Oghams, so future work on their decipherment should resolve at least some of the current controversies and offer a more constrained approach to the study of the Pictish Symbols as well.

 

Acknowledgments: I would like to thank Joseph Eska for his comments,  suggestions, and corrections regarding Celtic historical linguistics over the past few weeks (January–March of 2026). Any remaining errors are entirely my own.

 

References

Ball, Martin J., and Nicole Müller, editors. 2009. The Celtic Languages. Second edition. London: Routledge.

Broderick, George. 2015. *Pixti /*Pexti, Picti? The Name ‘Picti’ Revisited. Journal for Scottish Name Studies 9:9–42.

Diack, Francis C. 1944. The Inscriptions of Pictland: An Essay on the Sculptured and Inscribed Stones of the North-East and North of Scotland: with other writings and collections. Aberdeen.

Eska, Joseph F. 2004. Continental Celtic. In The Cambridge Encyclopedia of the World’s Ancient Languages, edited by Roger D. Woodard, pp. 857–880. Cambridge University Press.

Eska, Joseph F. 2009. The emergence of the Celtic languages. In The Celtic Languages, edited by Martin J. Ball and Nicole Müller, pp. 22–27. Second edition. London: Routledge.

Eska, Joseph F. 2017. The syntax of Celtic. In Handbook of Comparative and Historical Indo-European Linguistics, Volume 2, edited Jared Klein, Brian Joseph, and Matthias Fritz, pp. 1218–1249. Berlin: De Gruyter Mouton.

Eska, Joseph F. 2024. The Continental Celtic Dialect Continuum. In The Method Works. Studies on Language Change in Honor of Don Ringe, edited by Joseph F. Eska, Olav Hackstein, Ronald I. Kim, and Jean-François Mondon, pp. 3–19. Palgrave McMillan.

Etchingham, Colmán, and Catherine Swift. 2004. English and Pictish Terms for Brooch in an 8th-Century Irish Law-Text. Medieval Archaeology 48 (1): 31–49.

Fife, James. 2009. Typological aspects of the Celtic languages. In The Celtic Languages, edited by Martin J. Ball and Nicole Müller, pp. 3–21. Routledge.

Forsyth, Katherine. 1996. The Ogham Inscriptions of Scotland: An Edited Corpus. PhD Thesis, Cambridge, Massachusetts, Harvard University.

Forsyth, Katherine. 1997a. Some Thoughts on Pictish symbols as a formal Writing System. In The Worm, the Germ, and the Thorn: Pictish Studies Presented in Honor of Isabel Henderson, edited by David Henry, pp. 85–98. Balgavies: The Pinkfoot Press.

Forsyth, Katherine. 1997b. Language in Pictland. The case against ‘non-Indo-European Pictish’ . Utrecht: De Keltische Draak.

Forsyth, Katherine. 1998. Literacy in Pictland. In Literacy in Medieval Celtic Societies, edited by Huw Pryce, pp. 39–61. Cambridge, UK: Cambridge University Press.

Forsyth, Katherine. 2006. Pictish language and documents. In Celtic culture. A historical encyclopedia, edited by John T. Koch, pp. 1444–1446. Santa Barbara: ABC Clio.

Fournet, Arnaud. 2010. A linguist’s comment on ‘Pictish symbols revealed as a written language through application of Shannon entropy’. Proceedings of the Royal Society A: 467(2121):305–308.

Fraser, James E. 2011. From Ancient Scythia to The Problem of the Picts: Thoughts on the Quest for Pictish Origins. In Pictish Progress: New Studies on Northern Britain in the Middle Ages, edited by S. T. Driscoll, J Geddes, and M. A. Hall, pp. 15–43. Leiden: Brill.

Jackson, Kenneth H. 1955. The Pictish language. In The Problem of the Picts, edited by Frederick T. Wainwright, pp. 129–166. Edinburgh.

James, Alan G. 2013. P-Celtic in Southern Scotland and Cumbria: A Review of the Place-name Evidence for Possible Pictish
Phonology. Journal of Scottish Name Studies 7: 1–50.

Koch, John T. 1983. The loss of final syllables and loss of declension in Brittonic. Bulletin of the Board of Celtic Studies 30:201–233.

Koch, John T. 1992. ‘Gallo-Brittonic’ vs. ‘Insular Celtic’: The inter-relationships of the Celtic languages re-considered. In Bretagne et pays celtique. Langues, histoire, civilisation. Mélanges offerts à la memoire de Léon Fleuriot, edited by Gwennolé Le Menn, pp. 471–495. Saint-Brieue: Presses Universitaires Rennes.

Koch, John T. 2022. The Neo-Celtic Verbal Complex and Earlier Accentual Patterns. Unpublished paper. Accessed here: https://repository.uwtsd.ac.uk/id/eprint/3342/.

Lee, Rob. 2010.  The use of information theory to determine the language character type of Pictish symbols. Scottish Archaeological Journal 32:137–176.

Lee, Rob, Philip Jonathan, and Pauline Ziman. 2010a. Pictish symbols revealed as a written language through application of Shannon entropy. Proceedings of the Royal Society A (466): 2545–2560.

Lee, Rob, Philip Jonathan, and Pauline Ziman. 2010b. Pictish Symbol Stones: religious imagery, heraldic arms or a language? Significance 7(4):159–163.

Matasović, Ranko. 2012. The substratum in Insular Celtic. Journal of Language Relationship 8:153–168.

Matasović, Ranko. Insular Celtic as a Language Area. In The Celtic Languages in Contact, Papers from the Workshop within the Framework of the XIII International Congress of Celtic Studies, Bonn, 26–27 July 2007, edited by Hildegard L.C. Tristram, pp. 93–113. Postdam: Postdam University Press.

Mees, Bernard. 2023a. The Morphology of Pictish. Celtica 35:33–55.

Mees, Bernard. 2023b. Caledonia and the language of the Picts. Scottish Language 42:1–15.

Nicolaisen, W. F. H. 1970. Gaelic Place-Names in Southern Scotland. Studia Celtica V:15–35.

Nicolaisen, W. F. H. 1972. P-Celtic Place-Names in Scotland: A Reappraisal. Studia Celtica VIII:1–11.

Nicolaisen, W. F. H. 2001. Scottish Place-Names: their study and significance. Revised Edition. Edinburgh: John Donald.

Nicolaisen, W. F. H. 1989. Place-Name Maps—How Reliable Are They? In Studia Onomastica (Festskrift till Thorsten Andersson), edited by L. Peterson and S. Strandberg, pp. 261–268. Stockholm: Almqvist & Wiksell.

Rhys, Guto. 2015. Approaching the Pictish language: historiography, early evidence and the question of Pritenic. PhD thesis. University of Glasgow.

Rhys, Guto. 2020. The Pictish language. History Scotland. URL: www.historyscotland.com. Accessed from Academia.edu: https://www.academia.edu/41294628/Pictish_Language_Guto_Rhys.

Rodway, Simon. 2020. The Ogham Inscriptions of Scotland and Brittonic Pictish. Journal of Celtic Studies 21(3): 173–234.

Russell, Paul. 2017. The evolution of Celtic. In Handbook of Comparative and Historical Indo-European Linguistics, Volume 2, edited Jared Klein, Brian Joseph, and Matthias Fritz, pp. 1274–1297. Berlin: De Gruyter Mouton.

Samson, Ross. 1992. The reinterpretation of the Pictish Symbols. Journal of the British Archaeological Association 145(1): 29–65.

Sproat, Richard. 2010. Ancient symbols, computational linguistics, and the reviewing practices of the general science journals. Computational Linguistics 36 (3):586–594.

Stifter, David. 2008. Old Celtic Languages. (Sommersemester 2008 course materials.) https://web.archive.org/web/20121002035607/http://www.univie.ac.at/indogermanistik/download/Stifter/oldcelt2008_1_general.pdf.

Stüber, Karin. 2017. The morphology of Celtic. In Handbook of Comparative and Historical Indo-European Linguistics, edited Jared Klein, Brian Joseph, and Matthias Fritz, pp. 1203–1218. Berlin: De Gruyter Mouton.

Taylor, Simon. 2011. Pictish Place-names Revisited. In Pictish Progress: New Studies on Northern Britain in the Middle Ages, edited by S. T. Driscoll, J Geddes, and M. A. Hall, pp. 67–118. Leiden: Brill.

Woolf, Alex. 2017. On the Nature of the Picts. The Scottish Historical Review 96(2): 214-217.

Mora-Marín, David F. 2024. A reappraisal of the development of the Mayan Syllabary. Written Language & Literacy 27(2):218–265.

https://www.jbe-platform.com/content/journals/10.1075/wll.00089.mor

 

Abstract

This paper revisits the problem of the derivation of phonographic signs in Mayan writing (cf. Mora-Marín 2003). The primary goal is to evaluate Lacadena’s (2005, 2010a, 2010b) proposal that the Mayan syllabary was initially derived through the wholesale adoption of a Mixe-Zoquean-based script. Lacadena argued that Mayan <CV> syllabograms with consonants shared by both Mayan and Mixe-Zoquean generally lack acrophonic explanations, while <CV> syllabograms for consonants exclusive to Mayan can be explained either by Mayan-based acrophony or graphic modifications of existing graphemes (i.e. addition of graphic elements, or coinage of digraphs). After 1) preparing a more comprehensive dataset of Mayan syllabograms using the Maya Hieroglyphic Database (Looper and Macri 1991–2024), and 2) conducting a more thorough review of the literature for possible acrophonic derivations and the nature of graphic modifications and seemingly digraphic combinations, the paper carries out a statistical test of a syllabary development model based on Lacadena’s proposal, concluding that such model is not viable, and that on present evidence, the Mayan syllabary was mostly an autochtonous invention.

Key words: acrophony, Mayan epigraphy, Mixe-Zoquean, orthographic conventions

 

Supplementary Materials:

Appendix1Comprehensive

Appendix2MajorAllograms

Appendix1References

MORA-MARÍN, D. Historical Sociolinguistics of the Classic Maya Lowlands: The Generic Preposition Variable. Cadernos de Linguística, [S. l.], v. 6, n. 1, p. e794, 2025. DOI: 10.25189/2675-4916.2025.v6.n1.id794.

 

 

Drawings of Two Incised Earflares with Comments on Embedded Glyphs

 

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

8/23/25

 

In 2003 the author visited a private collection to study an inscribed greenstone statuette, cataloged as (K3261), whose partial documentation and analysis constitute one of the topics of the report in Mora-Marín (2004). During the visit to the same collection, the author had the chance to photograph several other artifacts, including a pair of large jade earflares, of essentially the same dimensions, and carved out of the same type of jadeite—both probably fashioned from the same block of raw material, seen in Figures 1 and 2.  As noted by Houston and Tokovinine (2011), the two earflares were likely crafted and inscribed by the same artisan and scribe.

 

Figure 1

Figure 2

A few years after my visit I learned of the research by Tokovinine and Fialko (2007a, 2007b), documenting Stela 45 at the site and the iconographically- and textually-embedded glyphic mentions of some early royal names at the site of Naranjo, specifically, the one referred to by those authors as Naatz Chan Ahk and Tzik’in B’ahlam, which I began to suspect could be related to the names attested on one of the earflares, especially since the other earflare appeared to bear the Naranjo  emblem glyph main sign, as will be reviewed below. In 2011, I emailed several epigraphers with a slideshow illustrating the earflares and raising the possible connection to Naranjo’s rulers mentioned on Stela 45. At the time, Alexandre Tokovinine responded, expressing uncertainty about whether the content of the flares showed an unambiguous connection with Naranjo.

 

Two years later, Houston and Tokovinine (2013) published an online report on three artifacts, including the earflares in question, in which they not only provided line drawings of the incised designs, but also discussed their iconography and embedded glyphs (cf. Houston and Tokovinine 2013:Fig. 3).

 

Next, I present my own drawings of the earflares, highlighting the places where my drawings differ from those in Houston and Tokovine (2013). I also offer a brief review of the evidence for the embedded glyphic names.

 

The present drawings, seen in Figures 3A and 3B, are based on the photographs that I took in 2003 (Figure 1), my freehand sketches prepared during that visit, not illustrated here, as well as corrections undertaken with the help of photos generously shared by Donald Hales in June of 2021, seen in Figure 4. Better drawings based higher-resolution photographs taken with better lighting, or by means of RTI or 3D scanning technologies, should be attempted in the future, as there remain many details that need clarification.

 

Figure 3

Figure 4

Since I originally used my own photographs as a basis for tracing the imagery and glyphs, Figures 5 and 6 juxtapose those photos with my drawings.

 

Figure 5

Figure 6

 

Now, returning to the imagery and glyphs, I strongly recommend the reader consult Houston and Tokovinine (2011) first. Those authors discuss the imagery on each earflare, for one, as follows:

 

Carved by the same lapidary artist, the pair clearly forms a coherent whole. One depicts the so-called “baktun” bird, perhaps a celestial eagle. Its pectoral indicates some close but unspecified tie to the Principal Bird Deity. The other displays, not a bird in full flight, but a swimming lizard with scutes running up and down his front and back legs. The central design, a quadripartite element with four lobes, appears to represent a cavity at the center of each creature. Was this a witty reference to the central perforation or an allusion to their emergent state?

 

Next, a few notes comparing the two sets of drawings available. Figure 7 shows the areas where my drawing differs in minor or more significant details from that by Alexandre Tokovinine’s in Houston and Tokovinine (2011:Fig. 3). A few of the details are probably more accurate in Tokovinine’s drawing, but others are more clearly visible in the photos provided by Donald Hales, and thus are more complete in my drawings. Neither set of drawings is complete, though.

 

Figure 7

 

Regarding the embedded glyphs, I am in general agreement with the findings in Houston and Tokovinine (2011). My renderings of the glyphs are extracted and shown in Figures 8 and 9. Despite minor differences of graphic details, the glyphs in the first earflare are rendered largely in agreement with Tokovinine’s rendering (Figure 8).

 

Figure 8. Earflare with Reptilian/Amphibian creature.

 

In the second earflare, though, a few details are somewhat different. One small difference is seen in Figure 9A, where the embedded glyphic collocations contains a syllabogram tzi (or logogram TZIK for tzik ‘to (re)count’), which was not rendered in Tokovinine’s drawing. A second small detail is the more complete rendering of what may be the logogram SAK ‘white’ (Figure 9C).

 

Figure 9

 

As noted by Houston and Tokovinine (2011), the earflares likely came from either Naranjo or Rio Azul, with the na-tzu ʔAK name documented in at least six Early Classic texts from those two sites, including two stelae and a pottery vessel from Naranjo (Lopes 2005; Tokovinine and Fialko 2007a, 2007b; Stuart et al. 2023) and a mural and a vessel from Rio Azul (Stuart 1987; Kerr 1989:84), and of course, the likely presence of the main sign of the Naranjo Emblem Glyph, SAʔ, tilting the balance in favor of that site.

 

The next obvious step is to further document the earflares with better imaging and more accurate drawings.

 

Acknowledgments: Thanks are due to Alexandre Tokovinine for a brief email correspondence regarding the earflares in 2011, and to Donald Hales for the photos of the earflares.

 

References

Houston, Stephen, and Alexandre Tokovinine. 2013. REPORT: An Earful of Glyphs from Guatemala. Maya Decipherment: Ideas on Ancient Maya Writing and Iconography. https://mayadecipherment.com/2013/07/14/report-an-earful-of-glyphs-from-guatemala/.

Kerr, Justin. 1989. The Maya Vase Book: A Corpus of Rollout Photographs of Maya Vases 1. New York: Kerr Associates.
Lopes, Luís. 2005 A Probable Reference to Na-“Gourd” Chan Ahk on Naranjo Stela 15. Mesoweb: www.mesoweb.com/articles/lopes/ProbableReference.pdf.

Mora-Marín, David F. 2004. The Primary Standard Sequence: Database Compilation, Grammatical Analysis, and Primary Documentation. FAMSI report. URL: http://www.famsi.org/reports/02047/index.html.

Stuart, David. 1987. The Paintings of Tomb 12, Rio Azul. In Rio Azul Reports Number 3: The 1985 Season, edited by Richard E. W. Adams, pp. 161–167. San Antonio, Texas: University of Texas at San Antonio.

Stuart, David, Tomás Barrientos, Alexandre Tokovinine, and Daniel Aquino. 2023. La recuperación de la estela 43 de Naranjo: un breve acercamiento a la importancia histórica e iconográfica de un monumento maya perdido. In 35 Simposio de Investigaciones Arqueológicas en Guatemala, 2022, Tomo 1, edited by Bárbara Arroyo, Luis Méndez Salinas, and Gloria Aju Alvarez, pp. 219–232. Guatemala City: Ministerio de Cultura y Deportes, Instituto de Antropología e Historia, Asociación Tikal.

Tokovinine, Alexandre y Vilma Fialko. 2007a. La Estela 45 de Naranjo: Un análisis preliminar de su iconografía y epigrafía. In 20 Simposio de Investigaciones Arqueológicas en Guatemala, 2006, edited by Juan Pedro Laporte, Bárbara Arroyo, and Héctor Mejía, pp. 1140–1159. Museo Nacional de Arqueología y Etnología, Guatemala. (Versión digital). https://www.mesoweb.com/Simposio/pdf/20/Tokovinine_y_Fialko.2007.pdf.

—–. 2007b. Stela 45 of Naranjo and the Early Classic Lords of Sa’aal. The PARI Journal 7(4):1–14. https://www.precolumbia.org/pari/journal/archive/PARI0704.pdf.

CSV Files

Appendix 1. Accession Statements Dataset. AccessionDataset

Appendix 2. Generic Preposition Dataset. GenericPrepositions

Appendix 3. Northern Region Ch’olan Terms Dataset. NorthernRegionCholanTerms

HE6 Is ʔu Not u (Repeat as Needed with Other ʔV Syllabograms)

 

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

7/23/25, minor edits on 7/28/25

This note aims to change folks’ opinion on the notion that there are V phonograms in Mayan writing. Instead, so-called vocalic signs need to be conceptualized as canonical ʔV syllabograms.

 

The idea of vocalic signs stems from two common but erroneous ideas: 1) that glottal stops are “weak” consonants that are typically ignored in writing systems; and 2) that glottal stops are not phonemic in initial positions in Mayan languages. Both of these ideas are unsurprisingly held by linguists and scholars whose native languages typically lack phonemic glottal stops, and who are literate in writing systems that evolved in languages lacking phonemic glottal stops. These ideas are also undoubtedly influenced by the long-standing tradition of study of Colonial orthographies of indigenous languages, which were typically designed by speakers of European languages (e.g. Spanish, Portuguese, Italian, French, English) lacking phonemic glottal stops, and typically ignoring their orthographic representation.

 

Despite the lack of a grapheme for glottal stop in such European-derived orthographies, indigenous scribes who have adopted such orthographies have developed ways to indicate the presence of an initial glottal stop in contexts that matter (when the root-initial glottal stop is itself preceded by a proclitic, or by a prepound when forming part of a compound word).

 

This note therefore has a similar objective to Kaufman’s (2015) and Hopkins’s (2018): to explain and rectify the misconceptions about initial glottal stops in the Mayanist literature. But before reviewing what these authors have said, and providing additional evidence supportive of their observations and appeals, I begin with a recent argument by Campbell (2017), who supports the lack of phonemic status of the glottal stop in initial position. Indeed, in what is likely to become a highly influential overview of Mayan historical linguistics, Campbell (2017:18, footnote 3) remarks:

 

Most Mayan morphemes are monosyllabic, and PM [Proto-Mayan] had the possible syllable shapes (canonical forms): CVC, CV:C, CVC₁C₂, and CV₁ʔV₁C, where in CVC₁C₂ the C₁ of the consonant cluster was limited to h, ʔ, or a fricative s, š, or x. In each of these, the initial consonant is in fact optional. That is, traditionally Mayanists have followed Terrence Kaufman in repeating these canonical shapes with an initial C under the belief that the vowel initial ones really had a glottal stop as their onsets. However, since Proto-Mayan has a different set of possessive and ergative pronominal markers that attach to vowel-initial roots distinct from the set that attaches to consonant- initial forms, it is clear that not all Mayan morphemes should be considered consonant-initial, that PM also had vowel-initial morphemes, so that in the conventional canonical forms, the first C should be understood as optional, as (C)VC, (C)V:C, (C)VC₁C₂, and (C)V₁ʔV₁C. (Kaufman sometimes interprets CV₁ʔV₁C as equivalent to CV:ʔC.)

 

Campbell circulated his unpublished draft in 2015. This is the same year when Hopkins and Kaufman, separately, wrote and circulated papers on the topic. I begin with Hopkins, whose draft was circulated first.

 

Writing with epigraphers in mind, very few of whom recognize the importance of the glottal stop as a phoneme, much less consider it systematically when studying the orthographic principles of the ancient script, Hopkins (2018:268) observed: “Without exception, the phonological inventories of Mayan languages include the glottal stop as a consonant, and this consonant occurs in all structural positions: initial, medial, and final.”[1] In his review of the literature, Hopkins also traces the problem to the transcription practices by Spanish scribes during the Colonial period, to whom the glottal stop was an unfamiliar phonological unit which they typically did not represent. Then, more recently, scholars began to write the glottal stop by means of an apostrophe (saltillo), except in initial position, in which case words in Yucatec Maya, for example, are written as if vowel-initial, even by scholars who knew better. The result, Hopkins (2018:270) notes, was to be expected:

 

[…] the orthographic convention left the impression that the glottal stop did not occur at the beginning of words. This impression is now firmly established. It is a rare dictionary that writes initial glottal stops, and a rare grammar that does not refer to “vowel-initial” words […].

 

Hopkins also alludes to Terrence Kaufman’s convention of not writing initial glottal stops, but does not describe the context for such practice.

 

This is, in fact, where Kaufman (2015) picks up the story regarding the confusion. He begins by describing the background research carried out during the early 1970s, which led to the Proyecto Lingüístico Francisco Marroquín (PLFM), which trained native speakers of Mayan languages in linguistics. At this point, Kaufman (2015:5) notes that at that time both William Norman and himself believed that lexemes in the K’iche’an Proper languages could begin with vowels, but that that [ʔ] was inserted predictably in “absolute phrase-initial position.” Kaufman (2015:5) then explains (using <7> as a practical replacement for <ʔ>):

 

This led us to see as good the notion of not writing initial [7], because in these languages [but these languages ONLY!] initial [7] can be viewed as inserted phonetically. Unfortunately, we unwisely adopted this custom for all the languages of the PLFM, and this custom has been wrongly followed by some Mayanists working on Mexican Mayan languages.

 

Kaufman and Norman’s “unwise” lack of representation of initial glottal stops was applied in their proto-Ch’olan vocabulary and phonology (Kaufman and Norman 1984), whose influence on Mayan epigraphic research cannot be overstated. Kaufman with Justeson (2003) have corrected this, with proto-Ch’olan forms now cited with initial glottal stop, as in the case of *7@k’b’.@l [*ʔək’b’.əl] ‘darkness’ (2003:450).

 

It is perhaps in recognition of the impending influence of Campbell’s (2017) essay, as well as to once and for all reverse the earlier mistake during the PLFM regarding orthographic conventions, as well as his earlier characterization of the K’iche’an languages, that Kaufman (2015) presented a detailed discussion of initial glottal stops in Mayan.

 

One of Kaufman’s key arguments relates to the fact that in K’iche’an there exist contexts that suggest that initial glottal stop is in fact phonemic. This is important, as already noted, because Greater K’iche’an is the only subgroup where Kaufman himself, along with William Norman, once thought that initial glottal stops were purely phonetic. One such context that strongly points to the phonemic status of initial glottal stops in Greater K’iche’an relates to reduplication (Kaufman 2015:4–5):

 

In fact, the analysis we made of K’ichee7-Poqom was wrong. When morphemes of the shape /7VC/ are reduplicated, /7/ is present as part of the reduplication. Thus when the numeral root {7ox} ‘three’ is reduplicated, it is [7ox7ox] ‘three by three’, which it would be bone-headed to phonemicize as /ox7ox/: it must be /7ox=7ox/. Reduplication has to reduplicate something; therefore underlyingly the root is {7ox}. This being so, there has to be another way of accounting for the presence vs absence of [7] at the beginning of lexical roots in word initial position. It is preferable to analyze these words as starting with /7/, and postulate a rule that says ..VC # 7V.. => ..VCV..

 

It is in fact not difficult to find similar justifications, as well as new ones, to support the initial phonemic status of the glottal stop in other Mayan languages. Indeed, in almost all Mayan languages, including the Ch’olan-Tzeltalan and Yucatecan languages, glottal stops are phonemic in all positions (Kaufman 2015; Hopkins 2018), and only a very few morphemes, all of them grammatical proclitics belonging to the Set A pronominal paradigm for ergative/possessive agreement (i.e. a-, u- generally; in- in some languages), are truly vowel-initial.

 

This statement can be readily supported by observing that initial glottal stops are retained in a variety of contexts: 1) after the reflexes of proto-Mayan *ʔaj= and *ʔix=, the so-called “male” and “female” classifiers, respectively, both of which are characterized by Kaufman (2015) as prepounds (i.e. the first term in a compound term) but which may be proclitics instead; 2) after more common roots used as prepounds in compound terms; and 3) as the initial consonant of certain consonant-initial suffixes.

 

I begin by illustrating a few cases of phonemic glottal stops immediately following the “male” and “female” classifiers:

 

1. After a male or female classifier (i.e. reflexes of proto-Mayan *ʔaj= and *ʔix=, as in K’ichee’ ʔaj=ʔiitz for ‘witch’ (Kaufman 2015:6), or Ch’ol x’ec’ ‘chaya, malamujer’, based on x= ‘female proclitic’ and ʔek’ ‘chaya’, and Ch’ol aj’e’telob ‘autoridades’, based on ʔaj= ‘male proclitic’ and ʔeʔtel(-ob) ‘trabajo; autoridad(es)’ (Aulie and Aulie 2009:88).

 

Next are a few cases illustrating the phonemic status of root- and stem-initial glottal stop after a prepound in compound terms, both simplex and complex:

 

2. After the first term of a compound, as in Ch’ol tun’at for tun=ʔat ‘testicles’ based on tun ‘stone’ and ʔat ‘penis’ (Aulie and Aulie 2009:31, 94), where the glottal stop may even influence the final consonant of the preceding root causing its glottalization, as in Ch’ol pepech’ak’ ‘vine that is spread out on the ground’ (Aulie and Aulie 2009:71), analyzed as pe-pech=ʔak’, based on the prepound pech ‘flat’ > pepech ‘very flat’ (partial duplication derives intensified adjetives), and the postpound ʔak’ ‘vine’.

 

The following is an example where a -CV suffix begins with glottal stop:

 

3. With the highly productive inchoative suffix –ʔa-n ~ ʔa in Ch’ol, as in k’än-ʔan ‘to ripen; to become pale’, k’un-ʔan ‘to soften; to weaken’, sen-ʔan ‘to get stiff/numb’ (Aulie and Aulie 2009:20, 23, 83), among many others.

 

And last, the following are cases like the word-initial one adduced by Kaufman (2015:6) for K’iche’, cited above:

 

4. In reduplication processes, whether word-initially, as in K’ichee’ (ch+)ʔox=ʔox junaab’ ‘(at) a frequency of every three years’, or word-medially, as in Tzotzil derivations with -C₁un ‘affective suffix’, such as x-bik’-b’un ‘wanting to kill’, based on bik’ ‘to swallow’, and x-ʔach’-ʔun ‘dampness coming out’, based on ʔach’ ‘wet’ (Kaufman 2015:6), and as in affective derivations with -[CVC]₁-na(l) in Ch’ol and Yokot’an, as in Yokot’an ʔej-ʔej-na ‘bujando, gruñendo (to make noise, to grunt)’ (Keller and Luciano 1997:113).

 

What such examples tell us is that the deletion of initial glottal stops upon addition of a Set A prefix (e.g. y-), or proclitic-plus-prefix sequence (e.g. a-w-, u-y-), is not evidence that noun and verb roots that experience such deletion are vowel-initial. Instead, such examples constitute evidence of a morphophonological process specific to Set A proclitics and prefixes, not a strictly phonological process. The phonological process often cited to is epenthesis. However, as observed by Kaufman (2015:9–10), the epenthetic glides in Mayan languages are typically [h], [ʔ], and [y], not [r] or [w]. And historically speaking, at least, the [y] of Ch’olan (i)y- or Yucatecan (u)y- ‘third person singular prevocalic ergative/possessive’ is a reflex of proto-Mayan *r, which did not function as a glide in the Set A pronominal system of proto-Mayan, just as it does not in contemporary languages that have preserved /r/ from proto-Mayan *r.

 

Moreover, as Kaufman (2015:9) explains, both the omission of initial /ʔ/ upon addition of a Set A marker regardless of its underlying shape as either //VVC++// (e.g. proto-Mayan *aaw++ ‘A2’ and *eer++ ‘A5’) or //CV++// (e.g. proto-Mayan *qa++ ‘A4’ and *ki++ ‘A6’), as well as the contraction of vowel clusters to a single /V/ upon addition of a Set A marker of an underlying //CV++// shape, occur only with Set A proclitic markers. Such processes do not occur with other proclitics (aspect markers, absolutive markers) or with enclitics (absolutive markers), or in other contexts involving non-clitic morphemes (as already illustrated above). According to Kaufman, these two phenomena occur only with the Set A markers, which unlike the other clitics or word classes, are extrametrical; Kaufman (2015:9) notes that otherwise, /VV/ clusters will typically result in long vowels (i.e. [V:]). Thus, Kaufman (2015:9) attributes such behavior to the extrametrical proclitic status of Set A markers.

 

I will illustrate the problem now with the case of T1/HE6, the syllabogram ʔu. Some scholars in the past and present have assumed it to be “vocalic sign” u (e.g. Justeson 1989:33; Helmke et al. 2018:139; Kettunen and Helmke 2020:74), instead of a consonant-initial syllabogram ʔu. While it is true that it was used to spell /#u…/ sequences, specifically in the spelling of the proclitic u- ‘third person singular ergative/possessive’, the only /#u…/ initial morpheme in your typical Mayan language, it is also used to spell /#ʔu…/ sequences, as in the spelling of ʔuk’ ‘to drink’ (e.g. Piedras Negras Panel 3:P1, ʔu-ʔUK’-ni for ʔuk’-n-i-Ø ‘he drank’; Copan Altar U:K2, ti-ʔu-ʔUK’ for ti ʔuk’[-Vl] ‘drinking’) or ʔuht ‘to be finished; to happen’ (e.g. Tikal Stela 39:zB7, ʔu-ʔUH/ʔuh-ti for ʔu[h]t-i-Ø ‘it got finished; it happened’).

 

It is also noteworthy that native-speaker writers of Mayan languages, who typically employ alphabetic orthographies designed or inspired by Spanish orthography, and which, as a result, typically lack a grapheme for glottal stop since it lacks phonemic status in Spanish, have developed ways to indicate, at least in situations when it really matters, that a root-initial glottal stop is in fact present. Sometimes a hyphen is employed, for example, after male or female proclitics, as in the cases of Yucatec x-uunil ~ uunil ‘type of large ant that lives in the cracks and holes of dried trees’, and j-il kaab ‘abejero (male beekeeper)’ or x-il kaab ‘abejera (female beekeeper)’ (ALMY 2007:238, 278), based on ʔil ‘to see’ and kaab ‘bee’. Other times, a space is used after the first term of a compound when the following term begins with a glottal stop, as in Yucatec kok áak ‘pequeña tortuga terrestre que se usa para curar el asma y la tos persistente (small tortoise that is used for curing asthma and persistent cough)’ (ALMY 2007:98), for the compound kok=ʔáak, based on kok ‘asthma’ and ʔáak ‘turtle’. It is syllabified /kok.ʔáak/, not /ko.káak/, with the latter being the syllabification expected if the glottal stop of /ʔáak/ were not phonemic. Given this, glottal stops generally behave like any other consonant, and there is no reason to believe that ancient scribes would have treated them otherwise, with the exception being situations when a root-initial /ʔ/ is omitted when a Set A proclitic is applied.

 

The idea of V signs instead of ʔV signs was influenced by the Colonial alphabetic orthographies, such as that of Yucatec, in which vowel-doubling was used to spell final glottal stops: to the Spanish ear, a final glottal stop in some instances can sound like a repetition (rearticulation) of a final vowel. But again, the Colonial Yucatec orthography is based on Spanish orthographic practices and the absence of a grapheme for glottal stop, consequently, is due to the absence of a such a phoneme in Spanish. The idea has also been influenced, undoubtedly, by the fact that Mayanist epigraphers and linguists are themselves primarily speakers of Indo-European languages lacking a phonemic glottal stop. This parallels, as Hopkins (2018:265) has succinctly explained, the way in which the Hebrew (more generally Semitic) ‘aleph sign (which means ‘ox’), for the glottal stop, was reinterpreted by the Greeks: “Foreshadowing its later misunderstanding in Maya studies, when this letter passed from Middle Eastern writing to Greek, which lacked the glottal stop phoneme, it was taken to transcribe a vowel rather than a consonant, and became Greek alpha, representing the vowel a. Thus, A (an inverted ox head) begins the alphabet in Western writing.”

 

To conclude, when it comes to the glottal stop phoneme in Mayan languages and its representation in Mayan writing, the null hypothesis should be that ancient scribes treated glottal stops as consonants in all positions, and therefore, that a grapheme like HE6 was understood by Mayan scribes canonically as ʔu, not u, and that this was the case of all glottal stop-initial syllabograms (i.e. ʔi, ʔe, ʔa, ʔu, ʔo). The burden of proof should rest on epigraphers referring to such graphemes as “vocalic signs” (i.e. i, e, a, u, o).

 

References

Academia de la Lengua Maya de Yucatán. 2007. Diccionario Maya Popular, Maya – Español, Español – Maya. Mérida, Yucatán: ALMY, A.C.

Campbell, Lyle. 2017. Mayan history and comparison. In The Mayan Languages, edited by Judith Aissen, Nora C. England, and Roberto Zavala Maldonado, pp. 43-61. London and New York: Routledge/Taylor and Francis Group.

Helmke, Christophe, Joseph W. Ball, and Jennifer T. Taschek. 2018. A Classic Maya Carved Capstone from Buenavista del Cayo, Belize. Mexicon, 40: 134-142.

Hopkins, Nicholas. 2018. Saltillo: Not Just a Town in Northern Mexico. In Tiempo detenido, tiempo suficiente: ensayos y narraciones mesoamericanistas en homenaje a Alfonso Lacadena García-Gallo, edited by Harri Kettunen, Verónica Amellali Vázquez López, Felix Kupprat, Cristina Vidal Lorenzo, Gaspar Muñoz Cosme, María Josefa Iglesias Ponce de León, pp. 265-274. Belgium: European Association of Mayanists WAYEB.

Justeson, John S. 1989. The representational conventions of Mayan hieroglyphic writing. In Word and Image in Maya Culture: Explorations in Language, Writing, and Representation, edited by William F. Hanks and Don S. Rice, pp. 25–38. Salt Lake City: University of Utah Press.

Kaufman, Terrence. 1967. Preliminary Mocho Vocabulary. Working Paper Number 5, Laboratory for Language-Behavior Research, University of California, Berkeley.

—–. 2015. Initial glottal stop in Mayan languages. Unpublished note circulated by email May of 2015. URL: < https://www.albany.edu/ims/pdlma-web/Kaufman-2015-initial_glottal_stop_in_Mayan.pdf >.

—–. 2017. Aspects of the Lexicon of Proto-Mayan and Its Earliest Descendants. In The Mayan Languages, edited by Judith Aissen, Nora C. England, and Roberto Zavala Maldonado, pp. 62-111. London and New York: Routledge/Taylor and Francis Group.

Kaufman, Terrence, with John Justeson. 2003. Preliminary Mayan Etymological Dictionary. http://www.famsi.org/reports/01051/index.html.

Kaufman, Terrence, and William Norman. 1984. An outline of Proto-Cholan phonology, morphology, and vocabulary. In Phoneticism in Maya Hieroglyphic Writing, edited by John S. Justeson and Lyle Campbell, pp. 77-166. Institute for Mesoamerican Studies Publication No. 9. Albany: State University of New York.

Kettunen, Harri, and Christophe Helmke. 2020. Introduction to Maya Hieroglyphs. Seventeenth Revised Edition. Wayeb.

 

[1] Nick Hopkins first shared his draft in May of 2016. I am citing from the rewritten, published version from 2018.

Un nuevo dibujo de la inscripción de la “máscara” de “Phoenix” o “Wray”

 

David F. Mora-Marín
davidmm@unc.edu
University of North Carolina
Chapel Hill

27/6/25, 9/7/25

Esta nota tiene como objetivo el de introducir un nuevo dibujo de la inscripción del reverso de una de las dos “máscaras” supuestamente provenientes del sitio de Rio Azul (Petén, Guatemala).

 

El artefacto en cuestión, asignado el código (objabbr) COLJMsk en el Maya Hieroglyphic Database (MHD) por Looper y Macri (1991–2025), consiste de una “máscara” de piedra verde “de 19,7 cm de alto, 15 cm de ancho y 7 cm de grosor” (Mayer 1987:138), introducida al público en general por medio de una representación artística a color (Figura 1) en la portada de la revista National Geographic Magazine en 1986 (número 4, volumen 169). El editor de la revista en aquel entonces, Wilbur E. Garrett, explicó que la máscara seguramente habría sido una de una cantidad desconocida de artefactos saqueados de la Tumba 1 de Rio Azul.

 

Figure 1. Portada del número 4, volumen 169, de NGM.

 

Aunque hubo menciones de la máscara antes del árticulo de National Geographic (Emmerich 1984: No. 19; Berjonneau and Sonnery 1985: No. 326; Graham 1986: 45), fue hasta un año despúes de tal artículo que dos autores, Hellmuth (1987) y Mayer (1987), publicaron comentarios mas detallados sobre el mismo, e incluyeron dibujos de las superficies grabadas. Hellmuth, en su Monster und Menschen in der Maya-Kunst: eine Ikonographie der alten Religionen Mexikos und Guatemalas, escribió principalmente sobre la iconografía de la máscara y su relación con parte de la inscripción. Por su parte, Mayer, en su artículo en la revista del museo Volkermunde, menciona que la máscara parece estar hecha de fucsita y también provee un resumen de su iconografía y algunos de las expresiones glíficas. Ambos autores incluyeron los mismos dibujos (Figure 2), preparados por Melih Yerlikaya, y ambos describen al personaje representado como una versión de GI (God I), definido originalmente como un dios patrono del sitio de Palenque, pero más generalmente, un aspecto del dios Chahuk/Chahk, el dios de la lluvia. Éste lleva, a manera de tocado, una versión de la “Insignia Cuatripartita” (Quadripartite Badge), el cual representa un plato de ofrendas ceremonial.

 

Figure 2. Dibujos por Melih Yerlikaya.

 

Existen por lo menos dos dibujos más de la inscripción. Uno por Stephen Houston, pero solo un par de detalles del mismo han sido publicados (p.ej. Houston 1986:6, Fig. 7d; Houston et al. 2001:6, Fig. 4a). Otro es el de Sebastian Matteo, disponible en el archivo de dibujos de Wayeb.org (Figura 3).

 

Figura 3. Dibujo del reverso de la máscara por Sebastian Matteo.

 

La versión por el presente autor (Figura 4), basada en dos fotografías generosamente proporcionadas por Donald Hales, no debe de considerarse definitiva: sin duda existen muchos detalles que podrían definirse con mayor certeza mediante una examinación directa de la pieza. (El pigmento rojo añadido en las incisiones a veces no se implementó correctamente; incisiones sin pigmento son muy difíciles de percibir, algunos detalles con pigmento no son incisiones intencionales sino que rayones accidentales.) Pero la versión presente sí contiene detalles ausentes de por lo menos algunas de las versiones anteriores. El dibujo es un calco realizado con una tableta digital Wacom en Photoshop.

 

Figure 4. Dibujo por el autor. Se permite su uso con propósito didáctico/académico.

Por ahora, concluyo esta nota con una explicación básica pero preliminar (transliteración, transcripción, traducción) del texto, como se aprecia en la Figura 5. Pronto ofreceré más detalles al respecto y opciones más variadas para el análisis. (B4b podría contener, alternativamente a lo que se sugiere en esta ilustración, una expresión de los tres años transcurridos, lo que significa que B4b y A5 constituirían un doblete repetitivo, al menos de manera parcial.) La diferencia más importante con respecto a comentarios previous (cf. Houston et al. 2001:6, Fig. 4a) es que no considero el uso de T12/1G4 ʔAJ/ʔaj/ʔa como modo de marcar un sufijo derivacional *-aʔ (para formar sustantivos agentivos), sino que más bien como forma de marcar un sufijo derivacional incoativo *-ʔa.

 

Figura 5. Transliteración, transcripción, traducción.

 

Agradecimientos: a Nicholas Hellmuth for permitirme el uso de sus dibujos, y a Donald Hales, por proporcionarme las fotografías que fueron instrumentales para esta labor.

 

 

Referencias

Aulie, Wilbur H., and Evelyn W. de Aulie. 1999[1978]. Diccionario Ch’ol-Español, Español-Ch’ol. Mexico City: Instituto Lingüístico de Verano.

Berjonneau, Gerald, and Jean-Louis Sonnery. 1985. Rediscovered Masterpieces of Mesoamerica. Editions Arts 135.

Bricker, Victoria, Eleuterio Po’ot Yah, and Ofelia Dzul de Po’ot. 1998. A Dictionary of The Maya Language As Spoken in Hocabá, Yucatán. Salt Lake City: University of Utah Press.

Emmerich, André. 1984. Masterpieces of Pre-Columbian Art from the Collection of Mr. and Mrs. Peter G. Wray. André Emmerich Galery and Perls Galleries. Nwe York, NY.

Garrett, Wilbur E. 1986. Editorial. National Geographic 169:419. Washington, D.C.

Graham, lan. 1986. Looters Rob Our Graves and History. National Geographic 169:452–461.
Washington, D.C.

Houston, Stephen D. 1986. Problematic Emblem Glyphs: Examples from Altar de Sacrificios, El Chorro, Río Azul, and Xultun. Research Reports on Ancient Maya Writing 3. Washington, D.C.: Center for Maya Research.

Houston, Stephen D., John Robertson, and David S. Stuart. 2001. Quality and Quantity in Glyphic Nouns and Adjectives. Research Reports on Ancient Maya Writing 47. Washington D.C.: Center for Maya Research.

Looper, Matthew G., and Martha J. Macri. 1991–2025. Maya Hieroglyphic Database. Department of Art and Art History, California State University, Chico. URL: http://www.mayadatabase.org/.

Looper, Matthew G., Martha J. Macri, Yuriy Polyukhovych, and Gabrielle Vail. 2022a. MHD Reference Materials 1: Preliminary Revised Glyph Catalog. Glyph Dwellers Report 71. http://glyphdwellers.com/pdf/R71.pdf.

Looper, Matthew G., Martha J. Macri, Yuriy Polyukhovych, and Gabrielle Vail. 2022b. The Historical Development of the Maya Script: Preliminary Results. Glyph Dwellers Report 75. http://glyphdwellers.com/pdf/R75.pdf.

Matteo, Sebastian. https://wayeb.org/drawings/raz_jade_mask.png

Mayer, Karl Herbert. 1987. Drei Frühklassische Maya Miszellen-Texte. Archiv für Völkerkunde 41:137–144. Vienna.

Appendices for paper on H/J Merger and Male Proclitic Variables in Mayan writing

 

Detailed citation will be provided soon

 

Appendix1MergerStatistics

Appendix2MaleProcliticDNIGStatistics

Appendix3HJMergerDataset

Appendix4MaleProcliticDNIGDataset