A Historical Grammar of the Maya Language of Yucatan (1557-2000)
Page 11
vowel in one of them, instead of marking the vowel in one of them as short and the other as long (Ciudad
Real 1600?:fol. 424v):
(50)
toc.ah.ob. quitar tomar por fuerca, priuar arrebatar, robar y usupar casas y cosas muebles.
tooc.ah.ob. quemar abrasar, y cosa quemada.
An English translation of the quoted Spanish glosses appears in (51):
(51)
toc.ah.ob. ‘to take away, take by force, deprive, snatch, steal and usurp houses and movable
objects.’
tooc.ah.ob. ‘to burn, set on fire, and burned object.’
The reflex of toc in Modern Yucatec is tok ‘snatch away, defend’ (V. Bricker et al. 1998:279); its cognate in
San Quintín Lacandon (tak) has the same meaning (Canger 1995). The reflex of tooc is tóok ‘burn’ in Modern
Yucatec and San Quintín Lacandon (V. Bricker et al. 1998:279; Canger 1995). Itsaj has both tok ‘take away’
and took ‘burn’ (Hofling and Tesucún 1997:597, 601); Mopan has only tok ‘take away’ (Hofling 2011:409).
Kaufman and Justeson (2003) do not list these forms in their reconstructed vocabulary, but the presence
of tojclel (phonetic [tohk-l-el), glossed as ‘quemazón de monte, incendio’ (‘forest fire, conflagration’) in Chol
42 PHONOLOGY
(Aulie and Aulie 1998:117), suggests that *tohk was the earlier form of took. Because root-medial /h/ is
normally not represented in the Calepino, it is possible that the vocalic doubling in the second example
represents a long vowel, rather than preconsonantal /h/.
The stems in (44), (48), and (50) whose medial vowels are labelled as “long” or doubled in the Calepino
de Motul imply that preconsonantal /h/ had already disappeared, leaving high tone behind, because,
according to Terrence Kaufman (personal communication, 13 June 2010), it is likely that CVhC acquires
high tone first, followed by the conversion of /h/ to vowel length; alternatively, the changes could have
occurred simultaneously.
With high tone already in place in the late sixteenth century, the stage was set for low tone to develop
in the language, as the presence of both high and low tones in Modern Yucatec and the San Quintín dialect
of Lacandon suggests must have happened. But where did low tone come from? It may be the remains of
a consonant, which became phonologized as low tone. Or, because languages minimally have two tones —
High and Low — , it may have developed as the opposite tonal feature that allows the pitch space to be
divided into an upper (H tone) and lower (L tone) register. One can clearly rule out the first explanation
because unlike high-toned words, which had laryngeal /h/, there was no historical consonant present in
words with low tones. The second possibility is more plausible because there is evidence of the develop-
ment of low tone in common nouns with pre-existing vowel length after high tone appeared in Colonial
Yucatec. A few common nouns whose vowels are doubled or accented in the Calepino correspond to nouns
with low tone in Modern Yucatec:
(52)
Colonial
Gloss
Modern
Gloss
ác
grass
ʔàak
grass
baac
bone, horn (of deer, cow)
b’àak
bone
baach
type of pheasant
b’àač
chachalaca Ortalis vertula
baal
brother-in-law
b’àal
brother-in-law
beel
road, occupation; condition; sin
b’èel
road, occupation
fault, mistreatment
ceeb
belch
kèeb’
belch
ix coochol
large crickets
š kòočol
cricket
yeeb
fog, dew
yèeb’
fog, dew
nooy
pith, kernel, flesh
nòoy
pith, kernel
paal
young boy or girl
pàal
infant, child, minor
toon
penis
tòon
penis
beel, whose primary meanings are ‘road, occupation, or condition,’ deserves special mention because the
Calepino contains numerous examples of it in sentences, where it routinely appears with doubled vowels.
So, also, does the transitive stem derived from it — beeltic ‘do, make, cause’ (b’èeltik in Modern Yucatec,
b’etik in Itsaj) — , indicating a pattern consistent with a long vowel and low tone. The other examples pro-
vide additional evidence of the presence of nouns with long vowels in Colonial Yucatec that have cognates
with low tone in Modern Yucatec. Furthermore, the grammatical classes with the largest number of low-
tone stems in Modern Yucatec are antipassives and verbal nouns derived from root transitives and parti-
ciples, of course. I have already shown that participles had not yet developed long vowels and low tone in
Colonial Yucatec. The antipassive and verbal noun stems of Modern Yucatec correspond to stems in Itsaj
that have long vowels without tone, and for that reason alone we would expect to find them in Colonial
Yucatec. However, because they normally do not co-occur with -Vl suffixes and therefore are not likely to
be confused with other stems containing such suffixes, they only rarely appear with doubled vowels and/or
PHONOLOGY
43
accents in the Calepino de Motul. An interesting exception is cónol, the antipassive stem meaning ‘to sell,’
which is kòonol in Modern Yucatec (and kòon in San Quintín Lacandon [Canger 1995]). The Calepino con-
trasts it with conól ‘goods for sale’ (which is kòonol in Modern Yucatec). They are both konol in Itsaj (Hofling
and Tesucún 1997:359) and Mopan. These stems are unusual in Colonial and Modern Yucatec, Itsaj, and
Mopan; San Quintín Lacandon has adopted the general pattern for antipassive and verbal noun stems
(Canger 1995). Nevertheless, it is clear that Colonial Yucatec anticipated Modern Yucatec in lengthening
the vowel in the antipassive stem but had not yet extended it to the verbal noun stem.
Whether vowel length and tone developed simultaneously or in separate steps from the infixed /h/ that
preceded them, the changes did not affect all parts of the lexicon at the same time. Mediopassives were
among the first grammatical categories to undergo the changes, representing a large group of stems char-
acterized by high tone. Antipassives derived from root transitives were probably not far behind, represent-
ing an even larger group of stems characterized by low tone. Participles and verbal nouns derived from
root transitives were among the last to adopt vowel lengthening and ultimately low tone; these changes
did not begin until some time after the Calepino de Motul was compiled. In what may have been a less
systematic fashion, other kinds of nouns and adjectives acquired tone, some high and the others low, while
the other changes were taking place. Just when each class of stems crossed the tonal boundary cannot be
determined from the data available.
2.3.3. EVIDENCE FOR SCHWA IN COLONIAL YUCATEC. Modern Yucatec has five vowels: a, e, i, o, and u; the
other three languages in the family (Itsaj, Lacandon, and Mopan) also have a sixth vowel, a schwa. Because
the ma
jority of languages in the family have six vowels, it is likely that Yucatecan Maya originally had the
schwa too. The question is whether Colonial Yucatec had retained the schwa from earlier times, or had lost
it by the time the Calepino de Motul and other Maya documents came to be written in alphabetic letters.
The Calepino de Motul records only five vowels for Colonial Yucatec, the same five vowels as in Modern
Yucatec. However, this does not necessarily mean that Colonial Yucatec had only five vowels, just as the
absence of a symbol for the glottal stop in the Calepino and Colonial Maya documents does not mean that
the language lacked that consonant. Furthermore, the Calepino and other Maya language sources use a
single symbol for both laryngeal and velar /h/, as explained in 1.1. in this chapter. A symbol for one of the
five vowels may also have represented the schwa.
Hofling (2000:5, 2011:4)) describes schwa as a mid-to-high central vowel in Itsaj and Mopan, in both
cases “higher than schwas in English the and sofa.” Bruce (1968:24) says that the schwa in the Naha dialect of Lacandon is mid and central.
Hofling treats the schwa as the short grade of /a/ in his Itsaj and Mopan dictionaries (Hofling and
Tesucún 1997; Hofling 2011). It serves as the short form of /a/ in active transitive stems, contrasting with
the longer /a/ in the corresponding antipassive and verbal noun stems and as the /a/ allophone of the -Vl
suffix that marks intransitive imperfective stems. For these reasons, it seems likely that the letter “a” could
have represented the two vowels — /a/ and schwa — in Colonial Yucatec.
The most common type of inchoative verb in Colonial Yucatec was marked by an -h suffix. As intransi-
tive verbs, imperfective inchoatives suffixed -al to -h, yielding -h-al. If Colonial Yucatec had had six vowels,
the vowel in this suffix would have been a schwa, as it is in Itsaj today (Hofling and Tesucún 1997:158). The
Calepino de Motul sometimes lists two versions of the imperfective inchoative suffixes: -h-al and -h-il:
(53)
-h-al Variant
-h-il Variant
Gloss
bailhal
bailhil
continue, last, persist
caclam=pixhal
caclam=pixhil
kneel down
ceelhal
ceelhil
feel, be cold
44 PHONOLOGY
cohhal
cohhil
sink into vice
culucnachal
culucnachil
become very loud
çakhal çakhil
itch
chalanhal
chalanhil
become clear, settled (what is turbid)
chol=axhal
chol=axhil
blister, form warts
chuh=calhal
chuh=calhil
become acidic (stomach)
et=malhal
et=malhil
pass where others pass, participate
labacnachal
labacnachil
rot, spoil
talanhal
talanhil
become difficult, obscure, intricate
talanhal
talanhil
become decent, honest
tuhal
tuhil
stink, rot
utzhal
utzhil
recover, improve
ukucnachal
ukucnachil
become soaked, moist, damp
xibhal
xibhil
become sterile, impoverished (soil)
yanhal
yanhil
exist, have
Excluding the examples of -h-il that can be attributed to vowel harmony (bail-h-il, pix-h-il, and xib-h-il), this
data set suggests that the vowel in the suffix was somewhere between /a/ and /i/, in other words a schwa.
Not knowing exactly how to represent the vowel in these suffixes, the scribe wrote the suffixes sometimes
as -h-al and sometimes as -h-il, implying that the vowel was higher and more fronted than /a/. This scribal
indecisiveness can be interpreted as evidence that Colonial Yucatec had a schwa.
Colonial Yucatec lost the sixth vowel somewhere along the way to becoming Modern Yucatec. For
evidence of how and when the language lost schwa, I consider a set of examples drawn from other Colo-
nial and later sources. All Yucatecan languages have maʔ ‘no, not’ as a negative particle (it is spelled as
ma or maa in Colonial Yucatec) (V. Bricker et al. 1998:176; Canger 1995; Ciudad Real 1600?:fol. 277r; Hof-
ling 2011:298; Hofling and Tesucún 1997:433; Ulrich and Ulrich 1976:127). When prefixed to yan ‘exist,’ it
became manan or manaan ‘there is not’ in Colonial Yucatec (Ciudad Real 1600?:fol. 295r). The reflex of this
expression is mənaʔan in the Naha dialect of Lacandon (Bruce 1968:27) and mənaʔ in the San Quintín dia-
lect of the same language (Canger 1995), suggesting that the first vowel in manan/manaan was originally
schwa. Support for this inference lies in documents other than the Calepino de Motul, which has only the
manan/manaan spellings of this expression. During the sixteenth century and continuing into the first half
of the seventeenth century, only manan/manaan was recorded in other documents as well. But a minan
spelling appeared after 1650 and coexisted with the manan spelling, sometimes in the same document,
suggesting scribal uncertainty about how to represent schwa. A land document from Ebtun and dated
to 1670 contains examples of both spellings. At first, the manan spelling was more common than the
minan spelling (15 examples of manan versus two examples of minan during the seventeenth century). By
the end of the eighteenth century, there were almost four times as many minan spellings (60) as manan
spellings (only 14). The last manan spelling can be dated to 1816, with a total of 77 minan spellings for the
nineteenth century. By then, the schwa must have been raised to [i], accounting for the minaʔan form of
this expression in Modern Yucatecan Maya today (V. Bricker et al. 1998:185). This history of the changing
relationship between the manan and minan spellings suggests that the merging of schwa with either [a]
or [i] must have been completed during the first quarter of the nineteenth century.
The vacillation between manan and minan seems to have been limited to texts written by Yucatecan
Maya scribes. This is just as true of the Books of Chilam Balam as of notarial documents. For example,
the Book of Chilam Balam of Chumayel has 57 examples of minan and six examples of manan (V. Bricker
PHONOLOGY
45
1990a:315, 323–324), and the Book of Chilam Balam of Kaua has 28 examples of minan and three exam-
ples of manan (V. Bricker and Miram 2002); these proportions are in agreement with the late eighteenth-
century date attributed to these books on other grounds. On page 19 of the Chumayel (Gordon 1913),
three lines beginning with manan are immediately followed by five lines beginning with minan:
(54)
manan tun cħapahal
‘There was no sickness then;
manan tun chibil bac tiob
They had no aching bones then;
manan tun ɔam chacuil tiob
They had no high fevers then;
minan tun x pom kakil tiob
They had no pustule fever then;
minan tun elel tzemil tiob
They had no burning chests then;
minan tun yan akil tiob
They had no abdominal pains then;
minan tun tzentze
m cimil tiob
They had no consumption then;
minan tun cħibil pol tiob
They had no headaches then.’
Page 167 of the Kaua also has manan and minan in adjacent lines (Kaua n.d.:167):
(55) manan tun cħamac
‘There are no foxes then;
minan tun coohtacil bin chibalnaci
There are no pumas then that are going to bite.’
The structural parallelism in the lines of these examples keys on the equivalence of the two spellings of the
words that introduce them.
In their doctrinal and grammatical works, the Spanish priests did not distinguish between /a/ and /ə/,
writing both as /a/ (e.g., Coronel 1620a, 1620b). This pattern is consistent with the phonology of Spanish
vowels (Hualde 2005). In other words, these priests being Spanish and lacking a mid central schwa vowel in
their native language, lowered schwa to [a] in their writings. Only in the middle of the nineteenth century
did a Spanish priest use minan instead of manan in his sermons (Ruz 1846:I, 79–80), suggesting that by
then the raising of the schwa to [i] in this expression had become the spoken norm, as it is today (V. Bricker
et al. 1998:185). Only one schwa remains in Modern Yucatec, the one following the first person plural pro-
nominal prefix, which was never written in Colonial Yucatec (see Chapter 4).
3. SUMMARY OF PHONOLOGICAL CHANGES THROUGH TIME
The transition from infixed [h] to vowel length and (by implication) high tone is evident in one nominal root
and two transitive roots in the Calepino de Motul. During the intervening centuries between Colonial and
Modern Yucatec, these changes became more pervasive, affecting mediopassive and some intransitive
roots as well. Noun roots with long vowels in the Calepino eventually developed low tone. The one antipas-
sive stem with a long vowel in the Calepino also acquired a low tone, a feature that characterizes virtually
all antipassive stems based on transitive roots in Modern Yucatec. No participial roots in Colonial Yucatec
had long vowels or low tone, but all of them do in Modern Yucatec today.
Other documentable changes involved a reduction in the number of consonantal phonemes from 21
to 20 and in the number of vocalic phonemes from 6 to 5. Colonial Yucatec had two consonants, both rep-
resented orthographically as “h,” one laryngeal [h] and the other velar [x]. They began to merge during the
second half of the nineteenth century. “In Modern Yucatec, this distinction is still preserved in phonology,