6
Differentiation among Mayan Speakers
Evidence from Comparative Linguistics and Hieroglyphic Texts
MARTHA J. MACRI
Leonard Bloomfield (1933:318) cautioned that the comparative method of linguistic reconstruction can never claim to accurately describe the historical process. Winifred Lehmann (1962:84) likewise noted that it can never yield anything other than a dialect-free corpus. Fortunately, Mayan languages are recorded in a phonetic script from at least as early as 250 ce, so for some Mayan languages, hypothetical reconstructions can be checked against written records (e.g., Lacadena 2011). A careful comparison of hieroglyphic texts reveals the presence of differentiated speech communities at several periods for which linguistic reconstruction would predict only a small number of protolanguages. The data examined here demonstrate how languages found in written records can supersede hypothetical reconstructions of historical/comparative linguistics and provide insight into various forms of social organization.
Self-identification, of both individuals and society, is an ongoing and multilayered process. Ethnogenesis examines the beginnings of this process. Keeping in mind that language variation always reflects processes of social interaction, this chapter highlights linguistic variations reflected in the Maya texts of the Classic period—points at which certain portions of the lowland Maya community seem to have differentiated themselves from other communities. Evidence of these processes can be gleaned from a chronological and geographic comparison of Classic period hieroglyphic texts.
Earlier forms of languages can be discovered by comparing known related languages and hypothesizing features that were present in their common ancestor and by subgrouping language varieties according to shared phonological, morphological, syntactic, and lexical characteristics. In linguistics, cladistic classification expresses hypothesized evolutionary relationships based on the number of shared changes. Most readers are familiar with tree diagrams of the proposed development of Mayan languages from a hypothesized proto-Mayan, through various intermediary steps, to the thirty or so Mayan languages known today. Representing language change by a tree diagram, however, introduces a distortion of historical realities, since cladistic classifications based on the comparative method fail to detect variations that once existed but were lost. Two distinguished linguists have offered caution in this regard:
The comparative method, then . . . would work accurately for absolutely uniform speech-communities and sudden, sharp cleavages. Since these presuppositions are never fully realized, the comparative method cannot claim to picture the historical process. (Bloomfield 1933:318)
We lose information also in the complexity of the language we reconstruct. In normal use of the comparative method, we proceed backward by triangulation and eventually posit for each subgroup a dialect-free corpus . . . the method itself is not designed to yield anything other than a dialect-free corpus. (Lehmann 1962:84)
Figure 6.1 shows the hypothesized relationships between the language subgroups most relevant to this discussion of the Maya hieroglyphic script. Scholars agree that the majority of the Classic texts record a Ch’olan language (see discussion in Law 2014:16–18), though there remains a question of what role Yukatekan speakers may have played, not only in the early development of the script but continuing throughout its history. Certainly, when the Spanish arrived, the script was used in northern Yucatán by Yukatekan speakers.
Figure 6.1. Mayan languages of the Yukatekan and Greater Tzeltalan subfamilies
This chapter offers evidence from several linguistic features for developments of regionally distinct social (ethnic) groups within Classic Maya civilization. The first relies on a comparison of person markers in contemporary and Colonial Mayan languages to suggest at least two distinct phases of contact between Ch’olan and Yukatekan speakers. Another has to do with phonological contrasts between Ch’olan and Yukatekan as evidenced in Classic hieroglyphic texts. A third describes the temporal and geographic distribution of two political titles. A comparison of prepositions used in hieroglyphic texts suggests a minor dialect variation both among and within Classic Maya sites. Finally, a possible spelling of nun ‘foreign-speaking person’ as part of epithets and personal names suggests that prestige was associated with certain non-local speech patterns.
Person Markers
Mayan languages differ among themselves most obviously in patterned sound change. They also differ in some cases in having completely different lexical items, that is, non-related words used to name the same item. The most complex features of comparison, however, are those related to morphology and syntax, that is, variation in word and sentence formation. A significant subset of grammatical morphemes in Mayan languages is the person markers. Generally, Mayan languages do not have obligatory freestanding pronouns occurring with verbs, as English does. Instead, they have two sets of person markers that are prefixed or suffixed to verb stems to indicate the subject of a verb. One set of person markers is used with ergative constructions (usually verbal constructions with both a subject and a direct object), and another set of person markers is used with absolutive constructions (usually verbal constructions with only a subject). For the person markers that occur as prefixes, there are forms that precede verbs beginning with a consonant—preconsonantal forms—and other forms (usually very similar) that modify verbs beginning with a vowel: prevocalic forms.
Table 6.1 shows all the person markers for the Greater Tzeltalan (also called Greater Ch’olan) and Yukatekan subgroups. They are arranged more or less geographically, beginning with Tzotzil and Tzeltal of the Chiapas highlands, followed by Ch’ol, Chontal, Acalan Chontal, Ch’orti’, Ch’olti’, then Mopan, Itzaj, Lakantun, and Yukatek (Maya). The premise of this arrangement is that the Tzeltalan and Yukatekan languages represent two historically distinct groups and that the Ch’olan languages represent language varieties of Tzeltalan that were formed as a result of contact between a group of Tzeltalan and Yukatekan speakers. In addition to the person markers, this relationship is confirmed by other morpho-syntactic features, such as split ergative systems of subject marking, and in a significant subset of vocabulary shared between the Ch’olan and Yukatekan groups (e.g., Justeson et al. 1985:7–28).
Table 6.1. Person markers of Greater Tzeltalan and Yukatekan languages (does not include Set C person markers in Ch’orti’)
Tzotzil | Tzeltal | Ch’ol | Chontal | Acalan | Ch’orti’ | Ch’olti’ | Mopan | Itzaj | Lakantun | Yukatek | |
Ergative (preconsonantal) | |||||||||||
1stSG | j- | j- | k- | kä- | ka- | in- | in- | in- | in- | in- | in- |
2ndSG | a- | a- | a- | a- | a- | a- | a- | a- | a- | a- | a- |
3rdS | s- | s- | i- | u- | u- | u- | u- | u- | u- | u- | u- |
1stPLin | j..tik | j..tik | la(k)- | kä..la | ka- | ka- | ti..e'ex | ki..e'ex | k..eex | k- | |
1stPLex | j..(ti)kotik | j..kotik | k..lojon | kä..t’okob’ | — | — | ti- | ki- | k- | — | |
2ndPL | a..ik | a..ik | la'- | a..la | la- | i- | i- | a..e'ex | a..e'ex | a..eex | a..é'ex |
3rdPL | s..ik | s..ik | i..ob’ | u..ob’ | u..ob' | u-ob' | u- | u..oo' | u..oo' | u..o' | u..ó'ob' |
Ergative (prevocalic) | |||||||||||
1stSG | k- | k- | k- | k- | k- | inw- | inw- | inw- | inw- | inw- | inw- |
2ndSG | av- | aw- | aw- | aw- | -aw | aw- | aw- | aw- | aw- | aw- | aw- |
3rdSG | y- | y- | i- | y- (j-) | y- | uy- | uy- | uy- | uy- | uy- | uy- |
1stPLin | k..tik | k..tik | la(k) | k- | — | kaw- | kaw- | tiw..e'ex | kiw..e'ex | k..eex | k- |
1stPLex | k..(ti)kotik | k..kotik | k..lojon | k- | — | — | — | tiw- | kiw- | (dual k- ) | — |
2ndPL | aw..ik | aw..ik | la'w | aw..la | la- | iw- | iw- | aw..e'ex | aw..e'ex | aw..eex | aw..é'ex |
3rdPL | y..ik | y..ik | y..ob' | y..ob’ | y..ob' | uy..ob’ | uy- | uy..oo' | uy..oo' | y..o' | uy..ó'ob' |
Absolutive | |||||||||||
1stSG | -on (-i-) | -on | -on | -on | -on | -e’n | -in | -(e)en | -(e)en | -en | -en |
2ndSG | -ot (-a) | -at | -et | -et | -et | -e’t | -et | -(e)ech | -(e)ech | -ech | -ech |
3rdSG | Ø | Ø | Ø | -Ø, -i | Ø | Ø | Ø | (-ij), Ø | (-ij), Ø | (-i'), Ø | Ø |
1stPLin | -otik | -otik | -onla | -la | — | -o’n | -on | -o'on..e'ex | -o'on-e'ex | -ooneex | ó'on |
1stPLex | -otikotik | -otikotik | -on lojon | -t’okob’ | — | — | — | -o'on | -o'on | -eno' (dual -oon) | |
2ndPL | -oxuk | -ex | etla | -la | — | -o’x | -ox | -e'ex | -e'ex | -eex | -é'ex |
3rdPL | -ik | -ik/lah- | -ob’ | -(j)ob’ | Ø | -ob' | Ø | -oo' | -oo' | -iho', Ø | -ó'ob' |
Abbreviations: 1st = first person (I, we); 2nd = second person (you); 3rd = third person (he/she/it/they); A = absolutive (subject of intransitive verb); C = preconsonantal (before a consonant); E = ergative (subject of transitive verb); ex = exclusive (we, excluding you); in = inclusive (we, including you); PL = plural; SG = singular; V = prevocalic (before a vowel).
Sources: Tzotzil (Aissen 1987:49); Tzeltal (Kaufman 1971:103) ; Ch’ol (Aulie and Aulie 1998:234–36); Chontal (Knowles 1984:77); Chontal (Keller and Plácido (1997:444); Acalan Chontal (Smailus 1975:188); Ch’orti’ (Fought 1984:47; K’ulb’il Yol Twitz Paxil 2004:61); Mopan (Hofling 2011:10); Itzaj (Hofling and Tesucún 1997:9); Lakantun (Bruce 1968:48); Yukatek (ricker et al. 199:329).
With a simple exercise in visual inspection, it is possible to observe in the person markers several stages of language change in the development of Ch’olan languages. First, boxes are drawn around identical or nearly similar forms. Then, those person markers that have similar patterns (e.g., those that have all Ch’olan forms agreeing with Yukatekan) are grouped together. These two steps in the comparison are not illustrated here. The result of this exercise is shown in table 6.2, in which all of the person markers with shared patterns, presumably representing several proposed stages of contact, have been grouped together.
Table 6.2. Person markers grouped according to possible source patterns
Tzotzil | Tzeltal | Ch’ol | Chontal | Acalan | Ch’orti’ | Ch’olti’ | Mopan | Itzaj | Lakantun | Yukatek | |
---|---|---|---|---|---|---|---|---|---|---|---|
Group 1. All forms are the same. | |||||||||||
EC2ndSG | a- | a- | a- | a- | a- | a- | a- | a- | a- | a- | a- |
EV2ndSG | av- | aw- | aw- | aw- | aw- | aw- | aw- | aw- | aw- | aw- | aw- |
A3rdSG | Ø | Ø | Ø | -Ø, -i | Ø | Ø | Ø | (-ij), Ø | (-ij), Ø | (-i’), Ø | Ø |
Group 2. All Ch’olan forms follow Yukatekan. | |||||||||||
EC3rdS | s- | s- | i- | u- | u- | u- | u- | u- | u- | u- | u- |
EC3rdPL | s..ik | s..ik | i..ob’ | u..ob’ | u..ob' | u- | u- | u..oo' | u..oo' | u..o' | u..ó'ob' |
EV3rdPL | y..ik | y..ik | y..ob' | y..ob’ | y..ob' | uy..ob’ | uy- | uy..oo' | uy..oo' | y..o' | uy..ó'ob' |
A3rdPL | -ik | -ik/lah- | -ob’ | -(j)ob’ | Ø | -ob' | Ø | -oo' | -oo' | -iho', Ø | -ó'ob' |
A1stPLin | -otik | -otik | -onla | -la | - | -o’n | -on | -o'on..e'ex | -o'on-e'ex | -ooneex | ó'on |
Group 3. Eastern Ch’olan forms follow Yukatekan. | |||||||||||
EV1stSG | k- | k- | k- | k- | k- | inw- | inw- | inw- | inw- | inw- | inw- |
EC1stSG | j- | j- | k- | kä- | ka- | in- | in- | in- | in- | in- | in- |
EV3rdSG | y- | y- | i- | y- (j-) | y- | uy- | uy- | uy- | uy- | uy- | uy- |
A1stSG | -on (-i-) | -on | -on | -on | -on | -e’n | -in | -een | -(e)en | -en | -en |
Group 4. Ch’olan forms are unique. | |||||||||||
EC2ndPL | a..ik | a..ik | la'- | a..la | la- | i- | i- | a..e'ex | a..e'ex | a..eex | a..é'ex |
EV2ndPL | aw..ik | aw..ik | la'w | aw..la | la- | iw- | iw- | aw..e'ex | aw..e'ex | aw..eex | aw..é'ex |
EC1stPLin | j..tik | j..tik | la(k)- | kä..la | ka- | ka- | ti..e'ex | ki..e'ex | k..eex | ||
EV1stPLin | k..tik | k..tik | la(k) | k- | — | kaw- | kaw- | tiw..e'ex | kiw..e'ex | k..eex | k- |
EC1stPLex | j..(ti)kotik | j..kotik | k..lojon | kä..t’okob’ | — | — | — | ki- | k- | — | |
EV1stPLex | k..(ti)kotik | k..kotik | k..lojon | k- | — | — | — | — | kiw- | (dual k-) | — |
A1stPLex | -otikotik | -otikotik | -on lojon | -t’okob’ | — | — | — | --o'on | -o'on | -eno' (dual -oon) | |
A2ndSG | -ot (-a) | -at | -et | -et | -et | -e’t | -et | -eech | -(e)ech | -ech | -ech |
A2ndPL | -oxuk | -ex | etla | -la | — | -o’x | -ox | -e'ex | -e'ex | -eex | -é'ex |
In the first group, the second-person ergative singular preconsonantal, a-, and prevocalic, aw-, and the third-person absolutive, Ø, are the same across all of the languages. These forms are very close to what has been reconstructed for proto-Mayan, the language of origin common to all Mayan languages: *aa- *aaw- *Ø, respectively (Kaufman and Norman 1984:91). Because these forms are identical (or nearly so), they do not provide any information about language prehistory other than that all of these languages share a common origin in the distant past.
In the second group, all of the Ch’olan forms agree with Yukatekan. Only the Tzelatalan forms are distinctive. Five forms—the ergative third-person preconsonantal and the ergative third-person plural preconsonantal and prevocalic, the absolutive third-person plural and the absolutive first-person inclusive—pattern in exactly the same way. A question arises as to whether the s- prefix is the original one for proto–Greater Tzeltalan and was subsequently changed in Ch’olan as a result of Yukatekan influence or whether the s- was introduced only into Tzeltal and Tzotzil (and Tojolab’al, Chuh, and Popti’) after Ch’olan languages had separated. One argument in favor of s- as the original form is that two absolutive person markers also follow the same pattern, that is, for the third-person absolutive plural and the first-person absolutive inclusive, all Ch’olan forms follow Yukatekan, not Tzeltalan. Because the first group of changes happened in all Ch’olan languages, these changes appear to date from the earliest period of contact between Greater Tzeltalan and Yukatekan, well before the Ch’olan languages had begun to differentiate.
Group 3 shows those forms in which only the Eastern Ch’olan languages follow Yukatekan. The changes that happened at this stage reflect additional Yukatekan influence only on Eastern Ch’olan languages. These changes must have occurred subsequent to the separation of Eastern and Western Ch’olan.
In Group 4 the patterning is much less regular, with the Ch’olan forms reflecting independent developments within Ch’olan languages. These changes do not appear to have occurred directly as a result of Yukatekan contact; since the patterns differ for each of the languages, they must have occurred after Eastern and Western Ch’olan groups had differentiated into the subsequent language varieties we are familiar with today.
Several conclusions can be drawn from these data. First, they support the idea of Tzeltalan and Yukatekan as two distinct groups, with Ch’olan as a Greater Tzeltalan language heavily influenced by Yukatekan. Second, these data suggest two distinct periods of influence: an initial phase of influence prior to the separation of Eastern and Western Ch’olan and a later phase of influence only on Eastern Ch’olan languages. The final grouping illustrates changes that took place during a time of no or minimal influence from Yukatekan languages. It does not appear that the person markers in Yukatekan languages, or in Tzeltal and Tzotzil, were changed as a result of contact. Ch’olan languages, in contrast, underwent at least two major periods of contact with Yukatekan that resulted in significant changes to their system of person markers.
Mayan linguists and epigraphers differ in assigning dates to the formation of proto-Ch’olan and the differentiation of Ch’olan into eastern and western forms. An early estimate would place proto-Ch’olan in the Late Preclassic period, perhaps as early as 150 bce, and the differentiation of Ch’olan languages in the Early Classic period, perhaps around 300 ce. Some estimates place these changes several hundred years later. This discussion of language change illustrates that Ch’olan communities underwent repeated phases of language contact and differentiation. The example of the variation in person markers reflects changes in self-identification for major populations. This is not, however, a deliberate behavior in the same way the use of “Maya” suggests an emerging identity (Restall and Gabbert, this volume). These unconscious variations reflect the existence of distinct speech communities that have resulted from changes in social interaction between large segments of the lowland populations. Most of the following examples reflect comparatively minor changes that affected much smaller groups.
Yukatekan and Ch’olan Spellings
Maya epigraphers agree that at least the bulk, if not all, of the Classic texts represent Ch’olan languages. Since the Maya script is a mixture of both logographic (word) signs and syllabic signs, it is sometimes possible to find phonological evidence for a Ch’olan form of a particular word that contrasts with the Yukatekan form of that word. For example, the word for ‘house’ in Yukatekan is yotoch, but in Ch’olan languages it is yotot. In Classic texts the word is most often represented by a combination of syllabic and logographic signs (the graphemes are identified below by the three-digit codes developed for the Maya Hieroglyphic Database and published as the New Catalog of Maya Hieroglyphs [Macri and Looper 2003a; Macri and Vail 2009]): the first sign is usually syllabic yo (1SA or MZC) followed by a logograph of a house (ZY5), frequently followed by the syllabic sign for ti (3M2) (figure 6.2a). Since the Yukatekan word ends in –ch, clearly the Ch’olan pronunciation is intended. In nineteen examples, most often at Chichén Itzá, the word is spelled completely with syllabic signs yo-to-ti (MZC 33A 2M1). This spelling confirms a Ch’olan pronunciation for the word for house. Evidence from the Maya codices is less decisive. Although the codices contain evidence of both Yukatekan and Ch’olan forms, the sign for ti never follows the house glyph.
Figure 6.2. a. Four spellings of yotoch ‘house’; b. glyphs for ‘fire’
One example of a variant spelling that occurs at Chichén Itzá is the word for ‘fire’, k’áak’ in Yukatekan but k’ahk in Ch’olan languages. That is, in all Ch’olan languages except Chontal, the final k has lost glottalization. In nearly all of the Classic texts the word for fire is represented by some variant of the logograph for fire k’ahk (2S6) with no doubling of the fire sign (figure 6.2b). However, at the site of Chichén Itzá, the word for ‘fire’ is spelled with repeated syllabic signs, k’a-k’a (MZ3 MZ3). Whether this spelling represents a Yukatekan pronunciation k’áak’ or simply a Chontal or other dialectal variant is not clear, but it does contrast with the representation of ‘fire’ in other Classic texts where the grapheme is never duplicated.
Another seemingly incongruent example is the word ka-yo-ma kayom ‘fisherman’ (AA1 MZC 32A) (figure 6.3). It occurs twice on the Creation Tablet from the site of Palenque in a context relating to the Maya day and night paddler deities, in which it is clear that the word ‘fisherman’ is intended. It comes from the word kay ‘fish’ and a suffix –oma, which means to do something regularly or customarily (see Kaufman 1971:58 for a discussion of a related suffix in Tzeltal). What is unexpected is that the word for ‘fish’ in Ch’olan languages in chay; kay is the Yukatekan pronunciation. The presence of this Yukatekan pronunciation in an otherwise Ch’olan text remains unexplained, though there are other anomalous spellings at Palenque (see the discussion of Kan B’ahlum below).
Figure 6.3. Creation Tablet from Palenque. Merle Greene Robertson; used with permission.
Additional evidence for Ch’olan pronunciations of words can be seen in substitution patterns. One such substitution occurs among graphemes representing ‘sky’, ‘snake’, and the number ‘four’. As can be seen from the data set below, the Ch’olan words are nearly homophonous, while each of the Yukatekan forms is distinctive in vowel quality, and the forms would thus be less likely to substitute one for another.
ká’an | chan | ‘sky; high; tall’ |
kàan | chan | ‘snake’ |
kan | chan ~chän | ‘four’ |
Ch’olan speakers were in close contact with Yukatekan speakers, with numerous examples of borrowing between the two groups. The Classic period Ch’olan scribes indicated a “foreign” pronunciation of ‘snake’ as kàan rather than the expected chan by prefixing the comb-like ka sign to the head of a snake in the site name for Calakmul (figure 6.4a) and in several personal names, including that of Kan B’ahlam at Palenque (figure 6.4b).
Figure 6.4. a. Glyphs for Calakmul with AA1 ka; b. Kan B’ahlam’s name with AA1 ka
The contrasts noted here reflect simple sound correspondences between two subgroups of Mayan languages. Syllabic spellings and substitution patterns show that the Classic Maya texts of the southern lowlands, with rare exceptions, were written and read by a population that understood itself to be distinct from speakers of Yukatekan languages. The evidence for this level of identification spans a rather large area and reflects an identity that probably grew over several centuries. Comparative linguistic data suggest this phenomenon was caused by the intrusion of a Tzeltalan-speaking population into the greater Petén region. The resulting mixed population, although very much hybridized, subsequently began to identify itself as distinct from non-hybridized Yukatekan speakers.
Limited Distributions of Lexical Items
Yet another sort of differentiation can be seen in the distribution of specific lexical items, not necessarily pointing to different languages but minimally to different traditions of language use. One example is the title b’akab’. It is spelled with syllabic signs b’a-ka-b’a; it follows personal names and other titles, nearly always the final sign in a statement—in a few examples it follows the emblem glyph (figure 6.5a). The translation is uncertain, but only one b’akab’ is named at a site at any given time. In several cases the title is preceded by the profile of a woman, indication a female b’akab’. Of 180 examples coded in the Maya Hieroglyphic Database (Macri, Looper, and Vail 2001–12), none occur before 640 ce, and most are from the Usumacinta and Pasión regions. Few occur at Palenque, few if any occur at the site of Calakmul (data for that site are incomplete), and none are known from any monuments at Tikal.
Figure 6.5. a. Glyphs ba’-ka-b’a spelling the title b’akab’; b. glyphs te-ku-yu spelling the title tekuy(u)
What conclusions can be drawn from the uneven distribution of the title? The full significance is not known, but its presence/absence would seem to be deliberate. One possibility is that by 640 ce the title b’akab’ became important to record as part of a ruler’s name phrase. It may be that the title did not exist before that time or that it was not previously considered important enough to record. The title first appears in western and southern sites in the Usumacinta River drainage; it then spreads throughout the Maya region. Its absence (or near absence) from Tikal and Calakmul inscriptions may provide evidence of a significant political difference between rulers at those sites and rulers elsewhere by 640 ce. It may be one of the few explicit confirmations from hieroglyphic texts that by the mid-seventh century, rulers from Tikal and Calakmul constituted a distinct superior category.
Another example of a limited title also originates in the Usumacinta area. Examples of the title te-ku-yu (2G1/XGC ZC1 32D) currently total twelve (figure .5b). The earliest example is dated to 9.15.5 (736 ce) in the Maya long count at Yaxchilan on the hieroglyphic stairway of Structure 44. The latest example is from 9.17.10 (780 ce) at the site of Naranjo. Nine of the occurrences are a part of expanded name phrases for the same person, the Yaxchilan ruler Yaxun B’ahlam. Since the date of the event for the earliest example is a century before the others, it is not part of Yaxun B’ahlam’s name but is better defined as a title. The latest known occurrence is at Naranjo within a parentage statement that could conceivably refer to a descendant of Yaxun B’ahlam, though that is not certain. Two additional possible examples occur much earlier, 9.0.0.0.0 about 445 ce, on Stela 31 at Tikal.
There is no Mayan word tekuy(u), but a nearly identical word does exist in Nahuatl, -te:kuiyo: ‘lordship’. It is the possessed form of the word te:uk-tli ‘lord, member of the high nobility’ (Karttunen 1983:218, 237). It appears in an early vocabulary as part of two greetings (Arenas 1611:1):
Dios sea en esta casa—Ma to Tecuiyo Dios nican amochantzinco moyetztic
[May Our Lord God be in this house.]
Dios sea con todos—Ma to Tecuiyo Dios amotlan myoetztic
[May Our Lord God be with everyone.]
The appearance of this title in the early eighth century is consistent with the seventh- and eighth-century dates for several other Nahua words spelled syllabically in Maya texts (Macri and Looper 2003b). These words suggest that certain Maya regions experienced possibly several episodes of influence from Nahua speakers. Tekuy(u) provides yet another example of how differences in the histories of individual Maya regions or communities are reflected in differences in the written texts.
“Ti” and “Ta” as Evidence for Dialect Variation
In a logosyllabic script, phonetic contrasts can be difficult to detect. Nevertheless, slight syntactic differences can provide important clues to language variation. Differences in the use of signs for ti or signs for ta as prepositions and complementizers (introducers of dependent clauses) provide one example of such variation. A preliminary discussion of ti and ta appeared in Macri (1991). The subsequent development of the Maya Hieroglyphic Database (Macri, Looper, and Vail 2001–12) has provided many additional examples along with their associated dates and locations. This compilation has allowed a much more complete picture to emerge than was available at that time. Figure 6.6 shows the most common graphemes in the substitution sets that represent ti and ta. Whatever may have accounted for the origin of the graphemes 3M2 and BV3 (most likely originally ta or ta’ from täh ‘torch; pine’ and ta’ ‘excrement’, respectively [Kaufman and Norman 1984:131]), by the height of the Classic period they occur in spellings with the syllabic value ti, suggesting that by that time, at most sites, the locative preposition was, as it is in Yukatekan and some Ch’olan languages today, ti. However, in texts from at least three sites, we find clear evidence of a contrast between ti and ta. In inscriptions at those sites, variations of the syllabic sign ta occur in both prepositional contexts and syllabic spellings, while ti occurs only in syllabic spellings or as an introduction to a subordinate clause but not as a locative preposition.
Figure 6.6. a–c. Graphemes for ti: 3M2.1, 3M2.2, BV3; d–j. graphemes for ta: 3M3, 1B1.2, 1B1.1, 1B1.3, XQB, YM2, ZS1. Drawings by Matthew G. Looper (Macri and Looper 2003a)
At the site of Palenque, variants of ta occur in the following phrases:
ta ajaw-le | ta ajawlel | ‘as ajaw (in ajawship)’ |
u-na-ta-la | unatal | ‘the first time’ |
The only times ti occurs, it appears to function to spell out words with t but never as a preposition. The most frequent example is:
u-ti | uht | ‘it happened’ |
But a variety of other words spelled syllabically include ti:
3-lu-ti-ch’uh | oxlut? ch’ul | ‘divine triad’? |
mu-ti | mut | ‘bird’ |
ti-sa-ku | tisak | ‘Tisak’ (personal name) |
u-pa-ti | upätil | ‘his work’? |
u-chi-li-ti-ni | uchitinil | ‘their sweatbath’ |
The clearest examples of the contrast between functions of ti and ta at Palenque and Chichén Itzá occur in the phrase ‘in his house,’ which begins with the preposition ta and ends with the syllabic sign for the spelling of yotot (see figure 6.2a for an illustration of the example from Chichén Itzá).
ta yo-otot-ti | ta yotot | ‘in his house’ |
Phrases that include tu are ambiguous in two respects. First, tu represents a contraction of either ti or ta with the third-person marker u. Second, by far the most frequent example is tu-b’a(-hi) tub’äh, the exact translation of which is somewhat uncertain. It may be a benefactive reflexive ‘for himself’ or an instrumental reflexive ‘by himself’.
A similar contrast between the functions of ti and ta occurs at the site of Naranjo. Here, graphemes for ta occur as the preposition up to the Maya date 9.13.10.0.0 (702 ce), at which time the texts change to reflect the more common usage of ti for the preposition. Examples of the contrast between ta and ti from the earlier texts include:
HS1 Step 6 | ta 3-te-tun-ni | ta oxte tun | ‘at the 3 stone place’ |
Altar 1 B11 | ta 13 Ix | ta oxlajun ix | ‘on 13 Ix’ |
Altar 1 D7 | u-ti | uti | ‘it happened’ |
Stele 24 A2 | 1 Sotz’ ta k’in-ni | jun sotz’ ta k’in | ‘on the day 1 Sotz’ |
From 593 to 702 ce, the only exception to ta used as a locative/temporal preposition is that ti precedes the half-period sign on Stelae 22 and 24 for the date 9.13.10 (702 ce). On Stela 24, however, ti occurs as a complementizer, that is, it introduces a dependent clause:
ti xa-k’uh | ti xak’ | ‘by standing over’ |
(see Martínez Hernández 1929:915 454r)
By 702 ce, only ti occurs as a preposition:
ti pet | ti pet | ‘on the island’? |
ti ajaw-le | ti ajawlel | ‘as ajaw (in ajawship)’ |
ti yotot | ti yotot | ‘in his house’ |
ti 5 Ak’b’al | ti ho Ak’b’al | ‘on 5 Ak’b’al’ |
and ta only occurs in syllabic spellings:
yi-ta-hi | yitah | ‘his companion’ |
ya-ta-na | yatan | ‘his wife’ |
The differing use of ti and ta is not diagnostic of either the Ch’olan or Yukatekan language family, but it does reflect a difference among varieties of Ch’olan dialects. The substitution of the set of ti graphemes for the ta set would not have seriously impaired one’s ability to read the text, but it does reflect the usage of the author of a text. Literate people would have no difficulty reading texts written in a variety of dialects. In the case of Palenque, the use of ta as a preposition seems to suggest the presence of speakers of a dialect somewhat different from that of other Classic Maya sites. The change from the preposition ta to ti at the site of Naranjo might reflect language change within a constant population or the influx of a new population (or a new scribal tradition) whose dialect is in agreement with the majority of Maya sites. Again, from our vantage point we recognize the distinctions but are not certain of their causes or their social significance. Did the people at Palenque speak in a way that sounded odd to their neighbors along the Usumacinta River and the central Petén? Did the people of Naranjo speak (or write) significantly differently after 702 ce, and was this in some way tied to the presence or the demise of Lady Six Sky? What we can infer from the differences in the representation of prepositions at Palenque, Naranjo, and Chichén Itzá is that distinctive speech communities did exist among the lowland Maya during the Classic period.
Geographic and Linguistic Identification of Lady Six Sky at Naranjo
The hieroglyphic text on Stela 24 from the site of Naranjo provides a glimpse into the life of Lady Six Sky up to the ritual celebration of the period ending 9.13.10.0.0 (702 ce), at which time her son was about nine years old (figure 6.7). The text on the sides of the monument begins with her arrival at Naranjo, then continues with the birth of her son and the celebration of the period ending on 9.13.10, and ends with a parentage statement that gives the names of her mother and her father, the ruler of the site of Dos Pilas.
Figure 6.7. Drawing of Naranjo, Stele 24, front, by Ian Graham © President and Fellows of Harvard College, Peabody Museum of Archaeology and Ethnology number 2004.15.6.2.45 (digital file #99100038)
In a phrase modifying her name in the parentage expression, there is the phrase u-b’a-hi-li aj-nu-na-ja ta-li-chan ub’ahil ah nun (n)ah tali chan (figure 6.8). The first root b’ah or b’äh is part of the general phrase translated variously as ‘her image’, ‘she does it’, ‘herself’, and the like.
Figure 6.8. Drawing of Naranjo, Stela 24, right side, D4–D7, by Ian Graham © President and Fellows of Harvard College, Peabody Museum of Archaeology and Ethnology, PM# 2004.15.6.2.47 (digital file # 99320005)
The second part of the phrase is unique to this text. It begins with the agentive aj ‘he/she who (is) . . .’ followed by nu-na or nunah. Published attestation of the root nun itself is limited to Yukatek, that is, I have not found it in dictionaries for other Mayan languages. On the contrary, Kaufman (2002:727) reconstructs *meem ‘mudo [dumb]’ for Central (Eastern + Western) Mayan languages. The Motul Dictionary provides several relevant entries:
nun, ah nun boçal, que no sabe la lengua de la tierra o que es balbuciente o tartamudo; y el rudo que no aprovecha enseñarle [someone who does not know the language of the land or speaks poorly; a coarse person who doesn’t learn]. (Martínez Hernández 1929:695, 337r)
nunum vagabundo y perdido que no quiere asenta casa [a vagabond; an incorrigible person who won’t settle down]. (ibid.)
The Cordemex offers additional examples including ‘mudo [mute]’ from a number of sources (Barrera Vasquez et al. 1980:588). At least one contemporary Maya dialect has the word nùum ‘ignorant, stupid, lazy, retarded’ (Bricker, Po’ot Yah, and Dzul de Po’ot 1998:202)—for a discussion of final n > m variation in Yukatek, see Blaha Pfeiler (1992).
These words appear to be related to the Nahuatl word no:n-tli ‘someone mute’ (Karttunen 1983:174), which may, in fact, be the source. Molina (1944:73v) lists nontli ‘mudo [mute]’ and several related forms such as nonti ‘hazerse mudo [to become mute]’ and nontilia ‘hazer mudo a otro [to act mute to another]’.
In other words, Lady Six Sky is described by a phrase suggesting she speaks the language of Naranjo poorly, as if she were a foreigner. At one time this suggested to me that there was a language or dialect difference between Naranjo and Dos Pilas. Possibly, Naranjo spoke a Ch’olan dialect closer to Yukatek and Dos Pilas a dialect with fewer similarities to Yukatek.
However, the following phrase may illuminate this epithet: ta-li-chan. This phrase begins with the locative ta instead of the more common ti, a sure sign of a dialect distinction (this also occurs on the front of Stela 24 in the phrase ta k’in). The word chan, in addition to meaning ‘sky’, can also mean ‘tall’ or ‘high’. It sometimes occurs with the meaning ‘highland’. The phrase occurs in Yukatek as tali ka’analil ‘from the highlands [para arriba]’ and tali ka’anal k’uchuk ti kab’ ‘from high to low [de alto abajo]’ (Martínez Hernández 1929:832, 409v). Although Dos Pilas is indeed south of Naranjo toward the Guatemalan highlands, it is only slightly higher in elevation than the central Petén, and the environment is not appreciably distinct, so the phrase remained puzzling.
Recent archaeological evidence suggests that some Pasión region rulers appear to have taken elite highland women as wives. So rather than the tali chan referring to Dos Pilas as a highland location, it may refer to the fact that Lady Six Sky’s mother originated from an Eastern Mayan– (K’ichean or Mamean) speaking community in the mountains of Guatemala. If this reading of Lady Six Sky as a “person who speaks poorly, “as a “foreign-speaking woman,” is correct, it would be yet another feature from the hieroglyphic texts that shows a self-awareness of the Classic Maya of Naranjo as an identifiable group separate from other Mayan-speaking peoples. Even more important, since this phrase occurs on a stela celebrating Lady Six Sky, the designation as ah nun ‘foreign speaker’ would have had to have been a mark of status. Whether her “accent” in Naranjo derives from her association with a different lowland dialect spoken by elite families at Dos Pilas or even ultimately from Tikal or whether it is a result of her mother having been a native speaker of a highland language, ah nun is a quality that contributes to her importance—it carried prestige.
The Naranjo text is not the only one on which the word nun is spelled with syllabic signs. It also occurs at Chichén Itzá on Lintel 2 of the Las Monjas structure. After the name of K’ak’upakal K’awil is the phrase u-nu-na-li ??-la b’a-te ajaw-wa-li ununal ?? b’ate ajwal ‘foreign-speaking ?? ballplayer/warrior ajaw’ (figure 6.9). In this case the context offers nothing to support a reading of ‘foreign’, but it does occur in association with a name, thus it is a phrase referring to a person.
Figure 6.9. Chichén Itzá, Las Monjas, Lintel 2aA C1. Drawing by Ian Graham (Bolles 1977:269)
A discussion of nun as represented by the logograph 3M9 and associated with several early rulers, most famously Yax Nun Ayin of Tikal, the son of Sihyaj K’ahk’, a military captain associated with Teotihuacan (Houston and Inomata 2009:10; Stuart 2000), is beyond the scope of this chapter, but it does invite some intriguing speculation about his non-local origins and about the effect a foreign intrusion would have had on the political and social boundaries of the Classic Maya. The word nun “someone who does not know the language of the land or speaks poorly” appears later in name phrases and titles from several other sites (including Chichén Itzá, Copán, Dos Pilas, Pusilha, and Yaxchilan). The reappearance of nun in later texts may have been the result of movement among populations or of elite persons speaking different Maya language varieties (languages or dialect), and it may also have resulted from repeated intrusions of Mexican groups into various parts of the Classic Maya region known from the Early Classic, the seventh century, and the Postclassic period.
Nun in the name of the early Tikal ruler represents an unusually early date for a possible loan from a Nahua language into the vocabulary of the Classic Maya, adding to a growing body of data that suggests an early form of Nahua was present in central Mexico and the Gulf region from as early as the Late Preclassic and was at least one of the languages used at Teotihuacan (Dakin and Wichmann 2000; Macri and Looper 2003b; Macri 2005). Perhaps more important for a discussion of ethnogenesis, the use of the term nun “someone who does not know the language of the land” in names of prominent individuals suggests an acknowledgment of the presence of those who speak differently from the local population and the prestige associated with those persons.
Conclusion
In summary, rigorous examination of both comparative linguistics and the written hieroglyphic record provides insight into multiple levels of social, political, and intellectual differentiation of the Classic Maya. Some of these traits are broad, such as changes in the system of person markers; others are quite limited, either spatially or temporally, such as the titles b’akab’ and tekuy(u). Identifying clusters of such traits or matching them with parallel developments or changes in ceramics, architecture, burial customs, or population density can provide useful information about how various groups of Maya people understood themselves in relation to those around them and what sorts of interactions may have transpired both among Maya groups and between Maya groups and speakers of languages from other language families.
We can no longer labor under the illusion that the Classic Maya were a monolithic or homogeneous group or even that their texts represent a single “prestige” language variety (Houston, Robertson, and Stuart 2000). Social variation, as evidenced by language change and written texts, offers evidence that the Classic Maya understood themselves through multiple layers of identification. From the written record, it is evident that these multiple layers were not static but changed gradually—even in some cases dramatically—over time. There is a tendency to under-differentiate the people and events of the past, to try to understand them more simply than they were. The more fine-grained our data about the past become and the more layers of their various identifications we see, the more closely we approach an accurate view of who they were.
References Cited
Aissen, Judith L. 1987. Tzotzil Clause Structure. Dordrecht, Germany: D. Reidel. http://dx.doi.org/10.1007/978-94-009-3741-3.
Arenas, Pedro de. 1611. Vocabulario Manual de Las Lenguas Castellana y Mexicana. Mexico: Enrico Martinez.
Aulie, H. Wilbur, and Evelyn W. de Aulie. 1998. Diccionario Ch’ol de Tumbalá, Chiapas, con variaciones dialectales de Tila y Sabanilla, ed. Emily Scharfe de Stairs. Serie de vocabularios y diccionarios indígenas, Mariano Silva y Aceves 121. Mexico City: Instituto Lingüístico de Verano.
Barrera Vasquez, Alfredo, Juan Ramón Bastarrachea Manzano, William Brito Sansores, Refugio Vermont Salas, David Dzul Góngora, and Domingo Dzul Poot. 1980. Diccionario Maya Cordemex: Maya-Español, Español-Maya. Mérida, Mexico: Ediciones Cordemex.
Blaha Pfeiler, Barbara. 1992. “Así som, los de Yucatam: El proceso fonológico Vn—> m/__ (#,C) en dos lenguas en contacto.” In Memorias del Primer Congreso Internacional de Mayistas, ed. Barbara Blaha Pfeiler, 110–22. Mexico City: UNAM.
Bloomfield, Leonard. 1933. Language. New York: Henry Holt.
Bolles, John S. 1977. Las Monjas: A Major Pre-Mexican Architectural Complex at Chichén Itzá. Norman: University of Oklahoma Press.
Bricker, Victoria R., 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.
Bruce, Robert D. 1968. Gramática del Lacandón. Mexico City: Instituto Nacional de Antropología e Historia.
Dakin, Karen, and Søren Wichmann. 2000. “Cacao and Chocolate: A Uto-Aztecan Perspective.” Ancient Mesoamerica 11 (1): 55–75. http://dx.doi.org/10.1017/S0956536100111058.
Fought, John. 1984. “Cholti Maya: A Sketch.” In Handbook of Middle American Indians: Linguistics, Supplement 2, ed. Victoria Reifler-Bricker and Munro S. Edmonson, 43–55. Austin: University of Texas Press.
Graham, Ian, and Eric von Euw. 1975. Corpus of Maya Hieroglyphic Inscriptions: Vol. 2, Part 1: Naranjo. Peabody Museum of Archaeology and Ethnology. Cambridge, MA: Harvard University.
Hofling, Charles Andrew. 2011. Mopan Maya–Spanish-English Dictionary. Salt Lake City: University of Utah Press.
Hofling, Charles Andrew, with Félix Fernando Tesucún. 1997. Itzaj Maya–Spanish-English Dictionary. Salt Lake City: University of Utah Press.
Houston, Stephen D., and Takeshi Inomata. 2009. The Classic Maya. Cambridge: Cambridge University Press.
Houston, Stephen D., John Robertson, and David Stuart. 2000. “The Language of Classic Maya Inscriptions.” Current Anthropology 41 (3): 321–56. http://dx.doi.org/10.1086/300142.
Justeson, John S., William M. Norman, Lyle Campbell, and Terrence Kaufman. 1985. The Foreign Impact on Lowland Mayan Language and Script. Middle American Research Institute 53. New Orleans: Tulane University Press.
Karttunen, Frances. 1983. An Analytical Dictionary of Nahuatl. Norman: University of Oklahoma Press.
Kaufman, Terrence S. 1971. Tzeltal Phonology and Morphology. Publications in Linguistics, vol. 61. Berkeley: University of California Press.
Kaufman, Terrence S. 2002. A Preliminary Mayan Etymological Dictionary. With the assistance of John Justeson. Accessed April 27, 2009. http://www.famsi.org/reports/01051/index.html.
Kaufman, Terrence S., and William M. Norman. 1984. “An Outline of Proto-Cholan Phonology, Morphology, and Vocabulary.” In Phoneticism in Mayan Hieroglyphic Writing, ed. John S. Justeson, 77–166. Institute for Mesoamerican Studies, State University of New York Publication 9. Albany: State University of New York.
Keller, Kathryn C., and Luciano G. Plácido. 1997. Diccionario Chontal de Tabasco. Serie de vocabularios y diccionarios indígenas, Mariano Silva y Aceves, no. 36. Tucson, AZ: Summer Institute of Linguistics.
Knowles, Susan. 1984. “A Descriptive Grammar of Chontal Maya (San Carlos Dialect).” PhD dissertation, Department of Anthropology, Tulane University, New Orleans, LA.
K’ulb’il Yol Twitz Paxil. 2004. U’tirache Ojroner Maya Ch’orti’: Gramática Descriptiva Ch’orti. Guatemala City: Academia de las Lenguas Mayas de Guatemala.
Lacadena, Alfonso. 2011. “Maya Hieroglyphic Texts as Linguistic Sources.” In New Perspectives in Mayan Linguistics, ed. Heriberto Avelino, 343–73. Newcastle upon Tyne: Cambridge Scholars Publishing.
Law, Danny. 2014. Language Contact, Inherited Similarity, and Social Difference: The Story of Linguistic Interaction in the Maya Lowlands. Amsterdam: John Benjamins.
Lehmann, Winifred P. 1962. Historical Linguistics: An Introduction. New York: Holt, Rinehart, and Winston.
Macri, Martha J. 1991. “Prepositions and Complementizers in the Classic Period Inscriptions.” In Sixth Palenque Round Table, 1986, ed. Virginia M. Fields, 266–72. Palenque Round Table Series, vol. 8, Merle Greene Robertson, gen. ed. San Francisco: Pre-Columbian Art Research Institute.
Macri, Martha J. 2005. “Nahua Loanwords from the Early Classic: Words for Cacao Preparation on a Río Azul Ceramic Vessel.” Ancient Mesoamerica 16: 321–26.
Macri, Martha J., and Matthew G. Looper. 2003a. The New Catalog of Maya Hieroglyphs, vol. 1: The Classic Period Inscriptions. Norman: University of Oklahoma Press.
Macri, Martha J., and Matthew G. Looper. 2003b. “Nahua in Ancient Mesoamerica: Evidence from Maya Inscriptions.” Ancient Mesoamerica 14 (2): 285–97. http://dx.doi.org/10.1017/S0956536103142046.
Macri, Martha J., Matthew G. Looper, and Gabrielle Vail. 2001–2012. The Maya Hieroglyphic Database. Current beta-version available at the Native American Language Center. Davis: University of California.
Macri, Martha J., and Gabrielle Vail. 2009. The New Catalog of Maya Hieroglyphs, vol. 2: The Codical Texts. Norman: University of Oklahoma Press.
Martínez Hernández, Juan. 1929. Diccionario de Motul, Maya-Español Atribuido a Fray Antonio de Ciudad Real. Mérida, Mexico: Compañía Tipográfica Yucateca.
Molina, Fray Alonso de. 1944. Vocabulario en lengua mexicana y castellana. Facsimile of 1571 edition. Colección de Incunables Americanos, vol. 4. Madrid: Ediciones Cultura Hispanica.
Smailus, Ortwin. 1975. El Maya-Chontal de Acalan: analisis lingüístico de un documento de los años 1610–1612. Centro de Estudios Mayas, Cuaderno 9. Mexico City: UNAM.
Stuart, David. 2000. “ ‘The Arrival of Strangers’: Teotihuacan and Tollan in Classic Maya History.” In Mesoamerica’s Classic Heritage: From Teotihuacan to the Aztecs, ed. David Carrasco, Lindsay Jones, and Scott Sessions, 465–513. Boulder: University Press of Colorado.