Sunteți pe pagina 1din 219

Q'eqchi' Grammar

Copyright 2015 by SOS Publishing Co.


All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in
any form or by any means, including photocopying, recording, or other electronic or mechanical
methods, without the prior written permission of the publisher, except in the case of brief
quotations embodied in critical reviews and certain other noncommercial uses permitted by
copyright law. For permission requests, write to the publisher, addressed Attention: Permissions
Coordinator, at the address below.
SOS Publishing Co
1250 James Cr.
Lafayette CO 80026
Ordering Information:
Quantity sales. Special discounts are available on quantity purchases by corporations,
associations, and others. For details, contact the publisher at the address above.
Orders by U.S. trade bookstores and wholesalers. Please contact the publisher at the address
above.

Printed in the United States of America


First Printing, 2015
ISBN 978-0-9962833-0-4
SOS Publishing Co 1250 James Cr., Lafayette CO 80026
www.sospublishingco.com

Art and Book Formatting by Lois Deveneau, Freelance Creative Support Services
Author photograph by Juan Ignacio Stewart

Table of Contents
The People and the Language____________________________________________________1
Research Environment____________________________________________________________8
Personnel__________________________________________________________________________8
Previous Studies__________________________________________________________________9
Scope of Present Study_________________________________________________________10

CHAPTER 2: PHONOLOGY__________________________________________________12
Phonemic Inventory_____________________________________________________________12
Consonants: Simple Occlusives_________________________________________________12
Glottalized Occlusives___________________________________________________________14
Fricatives_________________________________________________________________________16
Resonants________________________________________________________________________17
Glides_____________________________________________________________________________18
Minimal Pairs_____________________________________________________________________20
Vowels____________________________________________________________________________22
Origin of Long Vowels___________________________________________________________23
Minimal Pairs_____________________________________________________________________24

CHAPTER 3: THE PRONOUN________________________________________________25


The Pronominal Affixes__________________________________________________________26
Noun Possession: Set A_________________________________________________________26
Subject of Intransitive Verb: Set B_____________________________________________28
Stative Sentences: Set B________________________________________________________28
Agent of Transitive Verb: Set A / Patient of Transitive Verb: Set B________30
Independent Pronouns__________________________________________________________34
Use of Independent Pronouns__________________________________________________35

CHAPTER 4: VERBS_________________________________________________________36
Classes of Transitive Verb Stems_______________________________________________37
Classes of Intransitive Verb Stems_____________________________________________39
Class viR________________________________________________________________________________39
Class vin________________________________________________________________________________40
Class vik________________________________________________________________________________40
Class viV________________________________________________________________________________41
Class vi_________________________________________________________________________________43
Class vit_________________________________________________________________________________44
Minor Intransitive Classes: vir, vib', viw_____________________________________________45

Affective Intransitives___________________________________________________________47
Verb Inflectional Affixes_________________________________________________________48

Inflectional Affixes for Tense, Aspect, and Mood______________________________49


ta-: future definite.____________________________________________________________________49
nak- present habitual._________________________________________________________________50
x-: recent past.________________________________________________________________________52
ki- : narrative past.____________________________________________________________________53
chi- ~ : optative / imperative; mi-: negative optative / imperative______________54

Inflectional suffixes for Tense, Aspect, and Mood_____________________________56


non-future -k.___________________________________________________________________________56
Future (a)q:____________________________________________________________________________57

Imperatives______________________________________________________________________61
Inflection and Affixes for Direction and Manner_______________________________63
Voice and Ergativity_____________________________________________________________66
Active Voice____________________________________________________________________________67
Reflexive Voice_________________________________________________________________________67
Passive Voice___________________________________________________________________________68
Antipassive Voice______________________________________________________________________70
Uses of the antipassive._______________________________________________________________71

Absolutives and the formation of infinitives and participles.________________73


Non-specific Passive Voice______________________________________________________75

CHAPTER 5: NOUNS________________________________________________________76
Definitions________________________________________________________________________76
Simple Noun Classes____________________________________________________________77
Class I___________________________________________________________________________________77
Class II__________________________________________________________________________________77
Class III_________________________________________________________________________________78
Class IV_________________________________________________________________________________79
Class V__________________________________________________________________________________79
Class VI_________________________________________________________________________________80
Dialect Variation in Simple Noun Classes_____________________________________________80

Compounds_______________________________________________________________________81
Class
Class
Class
Class
Class
Class
Class
Class

I Compounds_____________________________________________________________________81
la Compounds.___________________________________________________________________82
lb._________________________________________________________________________________83
Ic._________________________________________________________________________________84
Id._________________________________________________________________________________85
Ie._________________________________________________________________________________86
If._________________________________________________________________________________86
Ig._________________________________________________________________________________87

Class II compounds______________________________________________________________88
Class III compounds_____________________________________________________________89
Compounding and Abstract Nouns_____________________________________________90
Numbers__________________________________________________________________________90
Numeral Roots___________________________________________________________________90
Cardinal Numbers________________________________________________________________91
Ordinal Numbers_________________________________________________________________93
Distributive numbers____________________________________________________________93

Groupative numbers_____________________________________________________________94
Unspecified Group: Cardinal Number plus -al.______________________________________94
Approximate Group: Cardinal Number plus -aq.___________________________________94
Definite Group: Numeral Root plus -ichal._________________________________________94
Cofradia Members: Ordinal Numbers plus il._______________________________________94
Measures and numbers________________________________________________________________95

Prepositional nominals (relational nouns)_____________________________________97


Complex prepositional expressions____________________________________________98
Individual prepositional nominals______________________________________________99

CHAPTER 6: STATIVES_____________________________________________________105
Types of statives_______________________________________________________________105
Adjectives_______________________________________________________________________105
Adjectives as statives________________________________________________________________106

Participles_______________________________________________________________________107
Transitive participle: bil____________________________________________________________107
Stative participle with vtR: -C1o_____________________________________________________107
Intransitive participle: - (j)enaq_____________________________________________________108

Positionals______________________________________________________________________108
Positional adjectives__________________________________________________________________109

Tense, aspect and mood with statives________________________________________109


Non-perfect stative____________________________________________________________110
Non-perfective statives and tense, aspect and mood______________________________110
Present tense: -k and ______________________________________________________________111
Future tenses: -(a)q, -kaq, -qaq___________________________________________________111
Recent past: raj (Coban), time adverb (others)__________________________________112
Remote past: raj (Coban), chaq (others)___________________________________________112

Perfective statives: Intransitive verbs_______________________________________113

Sources and References__________________________________________________117

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

CHAPTER 1: INTRODUCTION
The People and the Language
Qeqchi is a Mayan language spoken by 850,000
speakers in the Guatemalan departamentos
(provinces) of Alta Verapaz, Izabal, El Quiche, Peten,
and in the country of Belize (see Map I). About half of
this area is comprised of the highlands of Alta Verapaz,
while the rest is hot lowlands. The language is one of
the 24 to 30 extant Mayan languages (Kaufman, 1974:
34), and belongs to the Greater Kichean branch of
Mayan languages. Greater Kichean includes Kichean
Proper (Kiche, Kaqchikel, Tzutujil, Sipakapenyo, and
Sakapulteko), Uspanteko, Poqomchi, Poqomam, and
Qeqchi. Qeqchi has been a distinct language within the branch since about
300 A.D. (Kaufman, 1974: 85).
Qeqchi, in spite of the fact that it now occupies a geographic area larger
than any other Mayan language in Guatemala, has considerably less dialect
variation than others, such as Kiche and Mam (England, 1983: 6). Kaufman
(1976: 64) notes only two dialect groups or zones, Eastern and Western, the
former centered in the municipios (municipalities) of Lanquin, Chahal,
Cahabon, and Senah, while the latter covers the remaining area. The
reasons for the lack of characteristic dialect variation in Qeqchi, in spite of
its large areal spread, lie probably in two factors. First, Qeqchi until about
one hundred years ago occupied a much smaller area than it now occupies,
comprising not even all of the territory of Alta Verapaz that it now occupies
and none of the other departamentos surrounding it, which is primarily hot
lowlands. The vast majority of hot lowland Qeqchis today come from
Cobn, Carch, and Chamelco. These settlers began arriving in other areas
when new land tenure laws promulgated by Justo Rufino Barrios, President of
Guatemala from 1871 to 1885, allowed for the creation of large plantations
of coffee in Alta Verapaz and effectively cut the Qeqchis off from land used
to grow their traditional corn, beans, and squash. More recently, with the
coming of effective malaria control, the Qeqchis have continued to search
in new areas for available land (Adams, 1965: 13). Thus, all Qeqchis were
in much closer contact until recently, a factor which has reduced the
likelihood of extensive dialect differentiation.
Secondly, Cobn and the surrounding area, including two other major towns
within eight kilometers of Cobn, has enjoyed a preeminence among Qeqchi
towns unparalleled in the Kiche, Kaqchikel, or Mam areas. As a result the
Cobn area dialect is definitely the prestigious one, and people from other
towns work to learn and imitate it. It has become difficult to encounter young

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

people from eastern dialect zone towns, such as Lanquin and Cahabon, who
speak the unadulterated eastern dialect.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Map showing the Departamento de Alta Verapaz and the neighboring


Departamentos of Petn, Quich, Baja Verapaz, and Izabal, the boundaries of the
14 municipalities (municipios) that make up Alta Verapaz, plus the four
communities that are the focus of this grammar:

A Town of Cobn

B Town of Carch
C Town of Chamelco
D Village of Chami

Dialect area, including


the towns of Cobn, San
Juan Chamelco, San Pedro
3

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Carch and the village of


Chamil.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

However, the western or Cobn dialect is represented not just by Cobn but
by the other nearby towns of Carch and Chamelco. Nor is this dialect entirely
uniform: as this study will show, there are occasional clear differences in
vocabulary and grammar between Cobn proper and the other towns. But
considering the high degree of similarity and taking into account a desire to
study the dialect area as far as possible and not just one town, it is felt that
the consideration of these four towns, three municipios and one aldea, is
essentially the study of one dialect and not a dialect survey.
Both the history and the prehistory of the Qeqchi area may be characterized
as being somewhat isolated from the mainstream. Alta Verapaz was not
settled early, probably no earlier than 600 B.C. (King, 1974: 13), and was
never a center of Mesoamerican civilization, in spite of the fact that it lies
between other centers, such as Copn, Quirigu, Utatlan, and Tikal.
Historically as well, Alta Verapaz has enjoyed a uniquely isolated situation.
Shortly after the conquest of the highlands of Guatemala by Pedro de
Alvarado in 1524, attempts to penetrate the Qeqchi area were repulsed with
such ferocity that the area became known as Tuzulutlan, which means "the
land of war" in Nahuatl, the language of the Mexican mercenaries who
accompanied Alvarado. The Dominican Bartolom de las Casas then asked
permission to demonstrate his rather radical ideas in Tuzulutlan. Among these
ideas were propositions such as that the native inhabitants of America should
be treated with patience, that the only business the Spanish had there was to
convert the native Americans to Christianity, and that those Spaniards not
engaged in such activities should forthwith abandon the lands they occupied
and return to Spain. Permission was granted in 1537, and the Dominicans
prepared to enter the area. By 1550 the Dominicans were firmly ensconced in
the Verapaz, all other Spanish settlers prohibited from entering, and from that
time to independence in 1823, Alta Verapaz was further isolated from the
mainstream of Mesoamerican colonial life.
Following independence, non-Indians began moving into the area, but
immigration was slow, and no radical changes in the life of the Qeqchis
occurred until after 1871, when Barrios assumed the presidency. During
Barrios' tenure a series of laws were passed which abolished communal
landownership, sold such land to private individuals, and forced the Qeqchis
resident on the land to provide labor for the new plantation owners. Through
the application of these laws the Qeqchis lost the relative autonomy and
independence they had enjoyed during the colonial period. One of the most
important results of these events was to cause the Qeqchis to begin to
move east and north of their original area in search of land in which to
continue their traditional agriculture. The majority of the emigrants appear to
have left from near Cobn as evidenced by the predominance of the Cobn
area dialect in areas settled in the last hundred years by Qeqchis.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

The Qeqchis in recent times up to the present continue to emigrate from


the area around Cobn and for the same reasons: a desire for free land and
to avoid the obligations of working on large plantations (Carter, 1969: 5).
That the emigrations are from the Cobn area is evidenced by the fact that
the Garfunas of Livingston on the Atlantic coast regularly refer to the
Qeqchis as Cobneros (person from Cobn) or Sampedranos (person from
San Pedro, i.e., San Pedro Carch). Carter (1969: 10) noted that the rate of
population increase for Izabal between 1950 and 1964 was 5.1 percent and
that of the Peten was 4.0 percent, higher than the national average of 3.1
percent. At least part of this rate must be due to the Qeqchi immigrations
into both areas.
Ethnographic sources for the Qeqchi area are sparse. There are three
excellent studies, and while each sets for itself a very definite and limited
scope, the three together provide a balanced if still somewhat limited
ethnographic picture of the Qeqchi. Carter (1969) describes in detail a
small, recently settled community near Lake Izabal, concentrating primarily
on the agricultural practices and economics of the group. The study by
Arden King (1974) contrasts with that of Carter, being a detailed, statistics
laden investigation of Cobn, a study which he undertook in order to apply
the methods and goals of cultural anthropology to a more modern and
complex society than was common at the time when he began his fieldwork
in the 1950's. The third large study, by Cabarrus (1974), looks at religion
and world-view among the Qeqchis, though its value is perhaps mitigated
somewhat by having been produced primarily for the Catholic religious
community of Alta Verapaz by one who is himself a Jesuit. The brief
ethnographic survey presented here relies on the above sources and on
personal experiences in the field.
The Qeqchis are primarily farmers in the traditional Mesoamerican mold,
relying on corn, beans, and squash, plus lesser crops such as chili, sweet
potato, manioc, pineapple, sugarcane, garlic, onions, tobacco, peanuts,
tomatoes, and bananas (Carter, 1969: 88-89, 99). Many Qeqchis also grow
coffee, but the vast majority of lands devoted to coffee in the Qeqchi area
are owned by non-Qeqchis, though dependent on Qeqchi labor. For
example, in the municipio of San Miguel Tucur, 20,447 of the 27,869
manzanas (1 manzana = 1.73 acres) in cultivation are in the hands of 88 nonQeqchis while the remaining 7422 manzanas are in the hands of 2005
Qeqchis or Poqomchis, and the principal crop of the municipio is coffee, a
situation which demonstrates one type of land tenure and use of the Qeqchi
area (Alvarez de Stewart, 1976: 8). On land newly settled by the Qeqchis, in
contrast, coffee growing may be avoided altogether, as in the case of Carter's
study. In terms of agricultural methods the Qeqchis are primarily slash-andburn cultivators. This is apparently the case both in the lowlands, as detailed
by Carter, and in the highlands around Cobn, as evidenced by the quantity
of burning that is obvious preceding the rainy season, though there exists no
careful study of the highland methods.
6

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

All Qeqchi towns and most larger villages have regular markets, the largest
of these being located in Carch and Cobn. Chamil and Chamelco have
small daily afternoon markets, but large numbers of people from these towns
also do considerable trading in Carch and Cobn. All market towns share
one characteristic which constrasts sharply with other Guatemalan market
towns: all trade is carried on in Qeqchi if either buyer or seller is Qeqchi.
All Ladinos (Ladino = non-Indian) who do business in the Qeqchi area, and
close to 100 percent of the Ladinos who do not live in Cobn, are fluent in
Qeqchi, a situation unusual in other Mayan areas and towns, where normally
the Indians are forced to learn at least some rudimentary Spanish in order to
buy or sell to Ladinos. This situation was at one time common in Guatemala-the lingua franca of the colonial capital, Antigua, was Kaqchikelbut only in
the Qeqchi area does it still prevail, though it appears to be changing in
Cobn at present.
The religious and ceremonial life of the Qeqchis is a synthesis of Catholicism
and indigenous survivals, differing in this respect little from other Mayan
groups. Cabarrus (1979), for example, mentions a prayer prayed at a
roadside cross which mingles the cross, Christ, and tzuul-taqa (literally, hillvalley), this last representing a dominating earth spirit which pervades the
Qeqchi area. These spirits often have a special hill which they inhabit
particularly, on which now may be found the "calvary" chapels of the Catholic
church which are often located a short distance away, and elevated from, the
main churches in Guatemala. Religious ritual is inseparable from the
agricultural cycle, and here again can be witnessed the synthesis of the old
and new religions. Carter (1969: 71-73) records a preplanting prayer which
moves from Biblical figures to the tzuul-taq'a to the names of some preColombian gods and back to the names of apostles and saints.
As mentioned earlier, the Qeqchi area has always been one of the more
isolated in Guatemala. As a result of this isolation the area has been little
touched by governmental programs or other large scale development
projects, though with the recent construction of a good asphalt highway to
Carch from the capital, the opening of nickel mines in Izabal, and especially
the discovery of petroleum in northern Alta Verapaz, this isolation may begin
breaking up. But the past and present isolation coupled with the market
domination of Qeqchi means that few Qeqchis know Spanish or need to
know it, except on those rare (though crucial) occasions when they come into
contact with the Guatemalan government. Knowledge of Spanish does give
the individual some prestige, but there does not appear to be tremendous
pressure among the people to learn Spanish. In an informal survey by this
writer Qeqchis felt that the people of their area would rather learn to read
and write Qeqchi than Spanish, a situation which does not obtain in most
other Mayan Indian areas in Guatemala, where knowledge of Spanish is more
highly valued and learning to read and write the Indian language is not seen
as useful.
7

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

One reason for the interest among the Qeqchis in learning to read and write
their own language, apart from the previously mentioned monolingualism and
isolation of the people, is probably due to the fairly high quality of work of an
applied nature by the first linguists to work in the area, and particularly
William Sedat. Sedat (1955) produced a dictionary which fairly accurately
represents the phonemes of Qeqchi, thus replacing a number of poorly
phonemicized vocabularies by non-specialists which could only have hindered
literacy programs in Qeqchi. In 1961 he published a translation of the New
Testament which, if perhaps lacking in certain respects, was intelligible and
usable. The existence of these works, particularly the second, must have been
a powerful stimulus toward learning the written form of their language for the
Qeqchis.
Following Sedat, and always benefitting from his work, are Francis Eachus and
Ruth Carlson of the Summer Institute of Linguistics, whose contributions have
sustained and augmented the foundation laid by Sedat, both in continuing to
produce good usable translations of the Bible and stories based on the Bible
and in providing some good basic primers for use in literacy programs. These
latter have been used in a Guatemalan government program designed to give
pre-primary instruction, including literacy, in Qeqchi, although the program
has not been extended to but few schools in the area.
Contemporaneous for many years with Eachus and Carlson was the late Father
Esteban Haeserijn, a Belgian priest stationed first in the parish of Purulh and
then later in San Juan Chamelco. Father Haeserijn's primary applied contribution
lies perhaps in a small songbook in Qeqchi that he published, which has sold
more than 50,000 copies and apparently spurred tremendous interest in
learning to read among large numbers of Qeqchis. One account has it that a
clever young man in the municipio of Cahabon taught himself to read solely
with this songbook. Another of Padre Haeserijns contributions has been in
insisting that priests working among the Qeqchis learn the language and in
personally teaching it to them.
As a result of this early applied work in the Qeqchi language there is a small
but sizeable literate population, a growing demand for literacy training, and a
low but constant demand for reading material. Applied work involving the
author, the Qeqchi team of the Centro San Benito, and Father Ennio Bossu of
San Juan Chamelco dramatically demonstrated the demand for literacy
training. As soon as a new primer for teaching written Qeqchi produced in
the Centro San Benito became available, Father Ennio called together a group
of volunteer literacy teachers who had been recruited for an earlier, less than
successful literacy attempt. Fourteen volunteer teachers using the new primer
began teaching in their aldeas in classes involving over two hundred students.
The results were gratifying, apparently due for the most part on the improved
primer. Teachers reported little absenteeism, which had plagued earlier
efforts, and the teachers themselves appear more conscientious. Father Ennio
8

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

then succeeded in recruiting volunteer teachers from 28 more aldeas,


effectively establishing literacy programs throughout the entire municipio.
In view of the above it seems possible that Qeqchi may become more fully
developed as a written language, though predictions of this sort are difficult to
sustain. What can perhaps be stated with more certainty is that if any one of
the Mayan languages of Guatemala develops as a written language to any
degree, that language will be Qeqchi.
It is this last point which has spurred the present work. A grammar of any
relatively unstudied language like Qeqchi advances linguistic and anthropological knowledge, but not always does such a grammar have the
possibility of contributing, even in a small way, to the tentative development
of a spoken language to written status. The present work will be made
available to the Qeqchi community and to others working in the
development of written Qeqchi through a Spanish translation.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Research Environment
The fieldwork for the present study was undertaken in Guatemala from August,
1973 to September, 1977 while the author was working in two institutions as a
linguist, under the auspices of the U.S. Peace Corps. The first period, August,
1973 to December, 1975, was spent at the Proyecto Lingstico Francisco
Marroqun (PLFM) in Antigua, Guatemala, where the author taught linguistics to
small groups of students who were native speakers of Qeqchi, Tzutujil, Chuj ,
Qanjobal, Akateko, Awakateko, and Poqomchi; taught the Qeqchi students
how to produce materials for a Qeqchi-Spanish dictionary and supervised that
work; did the necessary linguistic analysis of Qeqchi for the dictionary and
teaching tasks; and completed 25 dialect surveys of eight hundred items in
Kaqchikel, Kiche, and Mam. The work at the PLFM provided the author with his
introduction to Mayan languages in general, Qeqchi in particular, and data in
the form of a preliminary stage of the dictionary consisting of some three
thousand words.
The second period, January, 1976 to September, 1977, was spent at the
Centro San Benito (CSB) in Cobn, a Benedictine mission with strong interest
in advancing Qeqchi as a written language. The responsibilities entailed in
this position were to teach linguistics to a group of Qeqchi students; to
upgrade the quality and quantity of literacy materials available in Qeqchi; to
supervise a project involving ethnobotanical research, translation of
educational pamphlets, production of texts, and the development of courses
in spoken Qeqchi and Poqomchi; and to do the necessary research for the
present study. The work at the CSB provided me with a great portion of my
field data, plus the opportunity to contribute to a variety of applied linguistics
projects.

Personnel
A large number of people have contributed to the realization of this study, both
in the PLFM and in the CSB. This section will list and describe these people,
though it will be impossible to determine exactly the relative contributions of
each in the analysis here presented. The reader is cautioned to regard the
number of years of education with some perspective: in Alta Verapaz six years
of education is considered very ample, three years of education very good
outside the larger towns, and simple literacy better than average.
Ernesto Chen Cao (PLFM) provided the author's first words of Qeqchi and
continued throughout the author's time in the PLFM to be a most valuable and
intelligent consultant. He is from San Juan Chamelco. Both his father and
mother are monolingual in Qeqchi, though his brothers and sisters have
learned Spanish. He is to be considered probably one of the most
sophisticated native experts on Qeqchi.
Miguel Sam Juarez (PLFM), who worked with the author in the PLFM, is also
from Chamelco. Jose Domingo Cuc Chen (PLFM), who worked with the author
10

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

at the PLFM, is from Chamil, an aldea in the jurisdiction of Chamelco. Pedro


Tiul Pop (PLFM), who worked with the author at the PLFM, is from Lanquin, a
town in the eastern dialect zone of Qeqchi not touched upon in this study.
Flora Ac Caal (CSB) worked with the author at the CSB, having previously
worked with both Campbell (1974) and the Texas group (Pinkerton, 1976).
She spent a year at Texas as consultant for classes at the University of
Texas, learned to write phonetically there, and must be considered a highly
sophisticated speaker of Qeqchi. She is from Cobn. Amanda Catun Caal
(CSB), who worked with the author at the CSB, had previously worked with
Campbell (1974). She is from Cobn. Francisco Tzul Tzub (PLFM, later CSB)
worked with the author for a short time in the PLFM and in the CSB. He is
from Chamelco. Domingo Cuc Xol (CSB) worked with the author at the CSB.
He is from Chamil, a relative of Cuc Chen at the PLFM. He is a highly
respected community leader, having served as assistant mayor for his
village. Manuel Tzib Cas (CSB) worked with the author at the CSB. He is
from Chamil. Rumaldo Choc Cacao (CSB) worked with the author in the CSB.
He is from San Pedro Carch and worked previously with Dutch botanists,
primarily Irene Kunkel, in projects involving collection of medicinal plants.
With the exception of Choc Cacao, all of the above consultants received
training in linguistics ranging from three to six months in basic phonology,
morphology, and syntax, especially as related to Qeqchi. Instruction was
directed primarily toward practical application, dictionary making in the case
of those working at the PLFM and text and literacy materials preparation in
the case of those working at the CSB. This instruction was valuable in that it
allowed the author to check and discuss with the consultants the aspects of
Qeqchi grammar presented in this study.
It is the author's opinion that a research strategy for linguistic description
which attempts to bring the consultant into the analysis process often helps in
avoiding certain pitfalls in analysis and quite definitely is a more satisfying
experience than one in which the consultant's role is limited to that of providing raw data. The basic linguistic training provided the consultants allowed
them to enter into and understand the problems under consideration at any
given moment in the analysis, thus aiding in a conscious way the analytical
process and at the same time providing insights into Qeqchi to the native
speaker consultant.

Previous Studies
Until quite recently Qeqchi had received comparatively little attention from
linguists. A few small sketches appeared around the beginning of the 20th
century, such as those by Stoll (1896), Burkitt (1902), and Sapper (1906), but
serious work on Qeqchi really began with the appearance of a dictionary by
William Sedat in 1955. Sedats dictionary is bilingual, Qeqchi- Spanish and
Spanish - Qeqchi, contains numerous example phrases, and has generally
been the foundation upon which other authors have worked. Sedat, as
11

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

mentioned above, also provided the first translation into Qeqchi of the New
Testament.
Francis Eachus and Ruth Carlson, mentioned above, published a sketch of
Qeqchi morphology in Mayers' Languages of Guatemala (1966), which
included a large number of derivational and inflectional affixes. They have
dedicated themselves primarily to Bible translations.
The late Father Esteban Haeserijn, also mentioned above, published the only
general description of Qeqchi. His Ensayo (1966) represents an excellent
attempt by a linguistically unsophisticated layman with a background in Latin
grammar to describe a previously unstudied language. Haeserijn then
published his Guia (1972), a companion volume to the earlier work containing
copious examples and exercises for the layman interested in learning to speak
Qeqchi. In addition, Haeserijn had nearly completed a new dictionary of
Qeqchi before his death in 1976, a work which was completed by his
assistants. This dictionary contains a wealth of examples of idiomatic
expressions which extend our knowledge of Qeqchi semantics considerably.
The only other grammar is a doctoral dissertation by Ray Freeze (1970),
entitled "Case in a Grammar of K'ekchi' (Maya)." Freeze's thesis contains a
valuable analysis of certain aspects of Qeqchi grammar using the case
grammar of Charles Fillmore as a theoretical base. Freeze did not attempt a
full description of Qeqchi, preferring to probe and illustrate Fillmores theory
using Qeqchi data rather than describe fully Qeqchi in light of Fillmores
work. In addition, Freeze (1976a) has examined the grammatical properties of
Qeqchi possession and how it relates to prepositional constructions,
alienability in noun possession, interrogation and negation, and
nominalization of verbs. Freeze also contributed texts he collected to a
collection of texts in various Mayan languages (1976b).
Campbell (1974) discussed the theoretical implications of some aspects of
Qeqchi phonology as they relate to rule ordering and formal simplicity in
phonological theory. Another article (1976) contains a study of lexical acculturation of bilingual speakers of Qeqchi, both Ladino and Indian, and its
cognitive implications. In addition, mention should be made here of
Campbells doctoral dissertation (1977) which discusses the position of
Qeqchi within the Quichean branch of Mayan.
Sandra Pinkerton (1976) became interested in Qeqchi contemporaneously
with the author, spending a summer in the field in Cobn in 1973 while still a
student at the University of Texas. She later took one of her consultants, Flora
Ac Caal, who later worked with the author at the CSB, to Austin for the 197475 academic year to work as a consultant in the departments of linguistics
and anthropology, particularly with a course in linguistic field methods. The
collected papers of the graduate students in the course, along with three
texts, a word list, and an appendix describing the historical and geographical
12

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

place of Qeqchi within the Mayan language family, were edited by Pinkerton
and published by the Texas Linguistic Forum as Studies in Qeqchi. Of
particular interest are the two papers by Pinkerton, one on phonology and the
other which discusses ergativity and word order, and one by Freund, which is
a sketch of Qeqchi verb morphology. Pinkerton also helped contribute texts
to the same collection to which Freeze contributed (Ac Caal and Pinkerton,
1976). Finally, due to Pinkerton's continued work, the author has seen
unpublished manuscripts produced in graduate courses taught by Pinkerton.
One of these papers (DeCormier, 1977), which explores aspects of the particle
chaq, was helpful to the author in this study.

Scope of Present Study


This work is meant to be a descriptive grammar of Qeqchi, represented by four
major towns in the western zone: Cobn, San Juan Chamelco, San Pedro Carch,
and Chamil. The language is treated throughout as uniform unless otherwise
indicated. Dialectal differentiation overall is slight, but in some cases becomes
important. The desire of the author is to provide a general description of a
dialect zone, and it has been seen as necessary to work with at least one
speaker from each of the above towns in order to be able to pinpoint those
areas in the grammar which may vary in contrast with those areas which are
uniform throughout the four town area.
In terms of organization this grammar may be seen to be organized around
the pronoun as the unifying principle. Pronominal inflectional affixes in this
study are the principal inflectional morphemes; tense, aspect, and mood
affixes are definitely secondary in the inflectional framework, and other
inflectional affixes run a distant third. Following the initial chapter on
phonology, the rest of the chapters take the pronoun as their point of
departure and defining principle. Chapter II surveys the pronoun itself and its
basic uses. Chapter III on verbs demonstrates the pronominal affixes as
agents and patients of verbs. Chapter IV is on nouns, defined as to their
occurrence with inflectional affixes. Chapter V is on statives, also defined as to
their occurrence with pronominal affixes. Future studies should include a full
treatment of the derivational morphology and a description and analysis of
Qeqchi syntax.

13

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

CHAPTER 2: PHONOLOGY
Phonemic Inventory
The phonemic symbols used here and throughout this study make up a practical
orthography officialized by the Academia de las Lenguas Mayas de Guatemala
(ALMG) [Academy of Mayan Languages of Guatemala].

Consonants: Simple Occlusives


The simple occlusive phonemes are a series of four simple (p, t, k, q) and two
affricated (tz, ch) stops each having a different point of articulation. Following
Kaufman (1976) and England (1983), it was decided to treat the affricates
together with the "traditional" occlusives. All stops occur initially, medially, and
finally. All simple occlusives have an aspirated allophone when they occur
before other consonants and when in final position. Each stop is described
below with examples of the allophones of each.

Phonemic Inventory of Consonant Symbols

Consonant
s

Bilabi
al

Dent
al

Alveol
ar

Palatal

Vela
r

Uvular

Simple
occlusives

tz

ch

Glottalized
occlusives

tz

ch

Fricatives
Resonants
Nasals
Lateral

n
l

12

Glottal

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Trill
Glides

r
w

Front

Central

Back

Vowels
Short

Long

Short

Long

Short

Long

High

ii

uu

Mid

ee

oo

Low

13

aa

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

/p/ is a voiceless, bilabial


stop.
/p/ [ph] /___C, ___#
[p]

/ ----------

/tap/

[taph]

crab

/chupchu/
/po/
/
koopopo
/

[uphu]
[po]

put out (fire)


moon

[ko:popo
]

toad

[a: th]

bed

/t/ is a voiceless, apico-alveolar stop.


/t/

[th]
[t]

/___C, ___#
/ ---------

/chaat/

/katko/
/tul/

burned
banana

/kanti/

[kat ko]
[tull]

[kanti]

/tikto/

[tikhto]

rapidly

snake

/tz/ is a voiceless, apico-alveolar, affricated stop. It has only one allophone.


/tz/

[]
/ -------/tzo]
[o]
rooster
/
motzo/ [moo]
caterpillar
/bitz/
[i]
a little
/ch/ is a voiceless, laminal, alveo-palatal, affricated stop. In some towns, notably
Chamil, /ch/ is retroflexed. It has only one allophone.
/ch/

[]
/cha/
[a]
ash
/kiche/ [kie]
forest
/akach/ [aka]
turkey
/k/ is a voiceless-dorso-velar
stop.
/k/

[kh]
[k]

/ ___ C, ___#
/ --------

/pek/
/tikto/
/kar/
/wakax/
/
kaqkaq/

[pekh]
h

stone

[tik to]
[karr]

k
[ waka]

rapidly
fish

[kaqh kaqh]

very red

cow

/q/ is a voiceless, dorso-postvelar or dorso-uvular stop.


/q/

[qh]

/ ___ C, ___#

/kaq/

[ka qh]

red

/tiqto/

[ti qhto]

dressed

14

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

[q]

/ -------

/qawa/
/tiqob/

[qawa]
[tiqom]

mister
sweat

Glottalized Occlusives
The glottalized occlusives are unit phonemes which contrast with the set of
simple occlusives plus glottal stop. They are of three types: glottal stop
//; implosive /b/; and ejective /t', tz', ch', k', q'/. Glottal stop contrasts with
its absence in all positions except initially, where it is nonetheless
phonetically present before vowels. The other glottalized occlusives may
occur phonemically in initial, medial, and final positions.
The glottalized occlusives are realized as implosives and ejectives in initial
positions and following consonants.
/C/ [ C ] / C___, #___
The glottalized occlusives are preglottalized in other positons. /b'/ may
optionally be laryngealized (see below).
/C/ [C ] / V___V, ___#
Each glottalized occlusive is described individually below with examples of its
allophones.
/b/ is a bilabial imploded stop.
/b/

[]

[b]

/ C___, #___

/ V___V

[aqh]

/baq/

/ -------

[iq ill]

cooked

/tzubuk/
/tzub/

[ubukh]
[um]

to kiss
kiss her!

[tiww]

[toto]
[met]

eagle

/torto/
/met/

/tz/ is a voiceless, apico-alveolar, glottalized, affricated stop.


/ C___,
/tz/ []
#___
/ tzi/
[i]
[]

/ --------

bone

/chiqbil/

/t/ is a voiceless, apico-alveolar, glottalized


stop.
/t/
[t]
/ C___, #___ /tiw/

[t]

round
dwarf

dog

/tzaptzo/

[ap o]

shut

/patzok/

[paokh]

to ask

15

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

/patz/

[pa]

ask it!

/ch/ is a voiceless, laminal, alveo-palatal, glottalized,


affricated stop.
/ch/
[o]

[]
/ C___, #___
/cho/
/chajcho/
[axo]
[] / -------/kachin/
[kainn]

/puyuch/
[puyu]

/k/ is a voiceless, dorso-velar glottalized stop.


/k/ [k]
/ C___, #___
/kim/

[k]

/ --------

rat
empty
small
parrot

[kimm]

tall grass

/katko/

[kathko]

burned

/sakok/
/sak/

[sakokh]
[sak]

to hit
hit it!

/q/ is a voiceless, dorso-postvelar or dorso-uvular, glottalized stop. In some speakers,


most notably those from Chamelco, the prevocalic allophones may be implosives or
clicks (Freeze, personal communication).
/q/

[q]
[q]

/ C___, #___

/qeq/

[qeqh]

/qetqet/

[qethqeth]

black
proud

/ --------

/jiqok/
[xiqok ]
to breath
/jiq/
[xiq]
breathe it!
/ / is a glottal stop. Although non-phonemic in initial position, it will appear after
certain prefixes, /atink/ to bathe and /xatin/ he bathed, but not after others,
/atinobaal/ bathing place and /watinobaal/ my bathing place. The official alphabet of
the ALMG has decided to write the few glottal stops that occur after these prefixes
using a dash, so that /xatin/ is written as /x-atin/. Non-initial examples are:
[la: ath]
[e]

/laaat
/che/

16

you
tree

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Fricatives
Fricatives occur in all positions. Words ending in vowels may optionally contain a
non-phonemic [h] which is phonemic in a few dialects of Qeqchi not considered
here, and, according to Kaufman (1969), represents proto-Mayan *h. But /h/ is
not phonemic in final position in the dialects here considered, though it may be
in Lanqun and Cahabn (Kaufman, 1976: 142-143).
/s/ is a voiceless, apico-alveolar
fricative.
/sib/
/bisok/
/us/

[sim]

smoke

[isokh]
[us]

to measure
good

/x/ is a voiceless, laminal, alveo-palatal fricative; it is retroflex in Chamil.


/xukub/
[ukum]
horn

/ixim/
[iimm]
corn

/pix/
[pi]
tomato
/j/ is a voiceless, dorso-velar fricative.
/jolom/
/kanjel/
/iyaj/

[xolomm]

[kanxell]

head

[i tyax]

seed

task

/h/ is a glottal fricative with a voiceless allophone in initial position and a somewhat
complicated distribution of voiced [] and voiceless [h] allophones (and, for some
speakers, ) in medial position.
(1)
[] / V(+) __V
/chahim/
[aimm]
star

[paokh ~
(2)
[ ~ : ] / V__+V /pahok/
pa:okh]
trailblaze
[h ~ ] / C+
[helho ~
(3)
__V
/helho/
helo]
extended
[huyhuth ~
/huyhut/
huyuth]
fish-like movement
[ha]
water
Initial /h/
[h]
/ha/

17

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Resonants
Resonants, which here include the nasals /m, n /, lateral / l / , trill / r /, and
glides / w, y /, are partially devoiced in final position. The onset of the
resonants is voiced, the release unvoiced or devoiced (Campbell, 1974: 271;
Pinkerton, 1976: 15). The rule is
/ R / [ RR ] / ____ #

All resonants occur in all positions.


/m/ is a bilabial nasal.
/mol/

[moll]
egg

/amoch/
[amo]
frog
/peepem/
[pe:pemm] butterfly

/n/ is an apico-alveolar nasal with a homorganic allophone [] before velar and


postvelar consonants.
/nim/
[nimm]
big

/inup/

[inuph]

[sakh]
[qann]

/l/ is a voiced, apico-alveolar, light lateral.

ceiba tree

/sank/
/qan/

ant
yellow

/laain/

I (first person singular


pronoun)

/sulul/

[la:inn]

[sulull]

mud

/r/ varies phonetically within the dialect area.


Cobn: /r/ is a voiceless, alveopalatal, retroflex fricative which may become
partially voiced internally, especially when surrounded by unstressed syllables
(Pinkerton, 1976: 16).
/rax/
[a]
green
/jor/
[xo]
break it!
/jor+ok/
/
x+oo+rap+
ok/

[xookh]

to break

[o:apokh]

we hit

Carch, Chamil, Chamelco: /r/ is an alveolar trill, with a devoiced allophone in


final position.
/rax/
[ra]
green

18

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

/jor/

[xorr]

break it!

/jor+ok/
/
x+oo+rap+
ok/

[xorokh]

to break

[o:rapokh]

we hit

Glides
Glides, or semivowels, undergo glide strengthening in certain environments
(Campbell, 1974: 270). /w/ has the allophones [w] and strengthened [gw ~ kw];
/y/ has the allophones [y] and strengthened [dy ~ ty ~ gy ~ ky]. The
phonological environment in which all strengthened allophones occur is initially
in a syllable, while the non-strengthened allophones appear elsewhere. The rule
is:
/w, y/

[ + strengthened] / (syllable begins) _____

When syllable boundaries are clearly marked, such as word initially or word
finally, or between the final consonant of a root and the initial consonant of a
following suffix, there is no question as to which allophone will be encountered.
/tuytu/

hanging

[tuytu]

/yokyo/

lying down

[dyokdyo]

/wiqwo/
/nawno/

kneeling
known

[gwiqgwo]
[nawno]

/wan/
/kaw/

there is
hard

[gwan]
[kaw]

/yuk/
/may/

goat
tobacco

[dyuk]
[may]

However, intervocalic /w, y/ appear to present Qeqchi speakers with less


distinct syllable boundaries, and varying pronunciations may be observed in the
realization of certain words containing these phonemes by the same speaker.
/xkawilal/ hardness may be pronounced [kagwilal] or [kawilal]. Native speaker
decisions on syllable boundaries in cases such as these vary from village to
village in the dialect area, from word to word within these villages, and as noted
above even from moment to moment for the same speaker.
This analysis differs from those of both Campbell (1974) and Pinkerton (1976).
Campbell said simply that glides are strengthened before vowels, but Pinkerton
(1976: 19) noted a large number of non-strengthened glides preceding vowels:
/mayink/

[mayinkh]

19

to smoke

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

/buyubank/
/awonel/

[uyuban
kh]
[awonell]

20

to pile
planter

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Pinkerton (1976: 16-21) argued for two sets of phonemic glides, a plain set and
an obstruentized set. She did note, however, examples in which the glide could
be either one glide or the other.

/xlawil/

/xyewil/

[lawill] or
[lakwill]

[tyewill] or
[tyekwill]

his key

his mare

Though her examples were only in borrowings from Spanish, other examples,
such as /xkawilal/ (mentioned above), may vary in the same speakers.
The analysis presented here better represents the data obtained from multiple
informants. Pinkertons analysis, while valid for her one informant with those
few exceptions, seems to represent an idiolect, one speakers decision as to
where syllable boundaries lie. Other speakers, even from Cobn, will vary to
some degree in using the simple and strengthened glides. They will be certain
which to use in some words and unsure in others, as was Pinkertons informant,
but the words in question differ.
In terms of dialect difference between the towns in the dialect area, Cobn
appears to strengthen intervocalic glides most and Carch least with Chamil
and Chamelco somethere between the two. But speakers in each of these other
towns will also differ among themselves as to where to strengthen intervocalic
glides.

21

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Minimal Pairs
p b /poq/
/boq/
t t /tob/
/tob/
tz tz /pitzok/
/pitzok/
ch ch /cham/
/cham/
k k /kok/
/kok/
q q /qix+bak/
/qix/
k q /ka/
/qa/
k q /kol/
/qol/
k ch /tikok/
/tichok/
k ch /kaj/
/chaj/
ch tz /loch/
/lotz/
ch tz /choq/
/tzoq/
ch tz /chap/
/tzap/
tz t /tzuul/
/tuul/
tz t /tzil/
/til/
t s /tib/
/sib/
s x ch r /sam/
/xam/
/cham/
/ram/
s j k /si/
/ji/
/ki/
j q /aj/
/aq/

whitish earth
call him!
loosen it!
undo it!
to jump
to smash
deep
rotten, spoiled
turtle
small
to burp
lukewarm
grinding stone
our legs
male (of animals)
necklace
to urge
to knock fruit from a tree
flour
wash it!
burn it!
clover (Cobn only
cut it! (not in Cobn)
zanate (type of bird)
grab it!
shut it!
hill
witch
strain it!
fasten it!
meat
smoke
snot
fire
deep
block it!
firewood
sharpen it!
sweet
cane for mats
thatch grass

22

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

j q /jun/
/qun/
m n /xam/
/xan/
h j /hal/
/jal/
h /ahin/
/ain/
h /ha/
/a/
q /si/
/siq/
/po/
/po/
/
t+aatinaq
C+ C /
/tanto/
/ch+elq/
/chelel/
w k + w /wib/
/kwib/
y t + y /yehok/
/tyos/

one
soft
fire
brick
corn ear
change it!
alligator
this
water
leg
firewood
bend it!
take it apart!
moon
hes going to speak
lying down
may he leave
kind of fruit (Spanish: paterna)
myself
two
to speak
God (Spanish: Dios)

23

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Vowels
There are ten vowels distinguished
by point of articulation (high, mid,
low, front, central, back) and by
duration (long, short).

Underlying short vowels have in


Cobn, and may have in other
towns, a long but non-contrastive
allophone preceding a resonant plus
another consonant word finally:
/sank/ ant is [sa:kh], /wark/ to sleep
is [gwa:rkh].

Long vowels may optionally be


realized as [Vh], a situation which is
phonemic in at least one of the
eastern dialects and which
represents proto-Mayan *Vh
(Kaufman, 1968).

Front vowels tend to be more lax


before velars and postvelars. /e/ is
always lax before glottal stop.

Segment /Vh+/ preceding a vowel


initial suffix becomes /V:/ before
consonant initial suffixes and
simply /V/ word finally: /tehok/ to
open, /teeto/ its open, /te/ open it!.

Short vowels occur in all positions.


Long vowels occur everywhere
except word finally and before word
final glottal stop.

Phonemic Inventory of Vowel Symbols

/i/

/i/ is a short, high-front,


unrounded vowel varying
between tense [i] and lax [I].

/ii/

/ii/ is a long, high-front,


unrounded vowel varying
between tense [i:] and lax
[I:].

/e/

/e/ is a short, mid-front,


unrounded vowel varying
between tense [e] and lax
[].

/ee/

/a/
/aa/

/i/ is always [I]


before /q/.
/ii/ is always [I:]
before /q/.

/e/ is always []
before // and /q/.

/oo/

/oo/ is a long, lax, mid-back


rounded vowel [:].

/u/

/u/ is a short, high-back,


rounded vowel varying from
tense [u] to lax [U].

/tib/ meat

/tiq/ hot

/iiq/ load

/kayiil/ market

/iitzinbej/ younger brother


/ewer/
yesterday
/ke/ cold

/mem/ dumb
/che/ tree

/eek/ feeling
/lajeeb/ ten
/ab/ hammock
/saq/ white
/cha/ ash
/aaq/ pig
/kaam/ rope
/ochoch/ house
/koj/ mask
/ko/ cheek
/oob/ five
/poop/ woven mat
/uk/ louse
/chu/ stinking

/aa/ is a long, low-central,


unrounded vowel [a:].
/o/ is a short, lax, mid-back
rounded vowel [].

/chaqi/ dry

/tzeq/ get rid of it!

/ee/ is a long, mid-front,


unrounded vowel varying
between tense [e:] and lax
[:].
/a/ is a short, low-central,
unrounded vowel [a].

/o/

/ik/ chile

/kuk/ water jar, squirrel

24

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

/uu/

/uu/ is a long, tense, highback, rounded vowel [u:],


with allophone [U:] before
/q/.

/uuq/ skirt
/tzuul/ hill

Origin of Long Vowels


Loss of h: phonemic long vowels
Speakers in our dialect area appear to have innovated in relation to the eastern
dialect area. Forms with the canonical form CVhC and CVhVC in the eastern
dialect area are CV:C in the western dialect area. Campbell (1977) shows that
the eastern forms are the most conservative, more closely representing what he
constructs for proto-Kiche. He also notes that eastern Qeqchi maintains the
old proto-Kiche and proto-Mayan long vowels, which in western Qeqchi were
lost, allowing for the development of new long vowels from the loss of h.
Proto-Kiche
*CVhC
*CVhVC

Eastern dialect
ahq
mahk
kaham
tzuhum

Western dialect
aaq
maak
kaam
tzuum

pig
sin
vine, rope
leather

The conversion of Vh to V: also occurs before morpheme boundary (+) plus


consonant and, in Cobn, before morpheme boundary plus vowel.
*CVh+Coh

tehtoh
rahro

teeto
raaro

Proto-Kiche
*CVh+ok

Eastern dialect
tehok

Western dialect
tehok

(all but Cobn)


(Cobn)
rahok

open
loved

to open

[teok ]
[te:ok]
rahok

to love

(all but Cobn)

[raok ]

(Cobn)

[a:okh]

Before final clusters of resonant plus C: non-phonemic V


In Cobn and occasionally in other towns in the dialect area, short vowels are
lengthened in word-final syllable which end in a consonant cluster where the
first consonant is a resonant or glide (n, m, l, r, w, or y).
These vowels are non-phonemic, occur primarily in Cobn, and as they are fully
predictable, they will not be written. In addition, in these segments it is not only
the vowel but both the vowel and the resonant that are lengthened.

25

Minimal Pairs
i ii
i e
ii ee
e ee
e a
ee aa
a aa
a o
aa oo
o oo
o u
oo uu
u uu

/mix(k)/
/miix/
/bisok/
/besok/
/tiix/
/teex/
/bek/
/beek/
/be/
/ba/
/beek/
/baak/
/al/
/aal/
/max/
/mox/
/baas/
/boos/
/chol/
/chool/
/ont/
/unt/
/toon/
/tuun/
/tul/
/tuul/

a moment ago
Catholic mass (Spanish misa)
to measure
to cut hair
old codger
roof tile (Spanish teja)
dig it up!
to walk
road
mole
to walk
cow (Spanish vaca)
young man
heavy
small monkey
a kind of herb
type of pot
pants pocket (Spanish bolsa)
open it
heart, soul
sling (Spanish Honda)
animal grease (Spanish unto)
tree trunk
a pile
banana
witch

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

CHAPTER 3: THE PRONOUN


The pronoun in Qeqchi may be
divided into two basic types:
independent pronouns and
pronominal affixes. Of the two
types the more basic and important
are the pronominal affixes, which
occur with the verb in all sentences
containing verbs, with nouns in all
occurrences of possession, and with
both nouns and adjectives in most
stative sentences. The present

chapter will briefly outline how


independent pronouns and
pronominal affixes function.
Succeeding chapters will elaborate
on the specific affixes and their cooccurrence with verbs, nouns, and
adjectives.

Chart III: Pronominal Allomorphs


Pronominal Affixes
Person

1st
singular
2nd
singular
3rd
singular
1st
plural
2nd
plural
3rd
plural

SET B
(absolutive)

SET A (ergative)

Independent
Pronouns

preconsonantal

prevocalic

- in

-w

- in

laain

-aw or -aaw

- at

laaat

-a or -aa

-x or -ix

-r or -ir

-qa

-q

-oh, -o or -oo

laao

-e or -ee

-er or -eer

- ex

laaex

x- ... -eb
or -ex

r- ... -eb
or -er

-eb or -e

aan or haan

heb aan,
aaneb or
haaneb

1. Vowel length in 2nd person singular and plural Set A: The short allomorph occurs with
transitive verbs where it immediately follows tense/aspect prefixes x-, ki-, and chi-. (The
explanation of these prefixes will be found in chapter 3). The long allomorph occurs in all
other environments.
2. Set A, 3rd person singular: In Coban some speakers may optionally insert an epenthetic
vowel -i before either the -x or -r allomorphs.
3. Initial h- in 3rd person singular and plural independent pronouns varies freely with its
absence.
4. Vowel length in Set B 1st person plural: Morpheme -oh- is realized /oh/ before a vowel, /oo/ as
subject of intransitive verbs, and /o/ in all other environments.

25

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5. Set B, 3rd person plural allomorphs: Allomorph eb is used finally and as patient of transitive
verbs preceding vowel-initial Set A agents -in and -aa. Allomorph -e is found in all other
environments.
6. Set A, 3rd person plural allomorphs: Allomorphs x-...-eb and r-...-eb are used with
nouns; allomorphs -ex and -er are used with verbs.

7.
8. The Pronominal Affixes

9.
In the pronoun chart above, it will be noted that the pronominal affixes
are divided into two "sets, A and B. Set A is further subdivided into variants
which occur before consonant-initial and vowel-initial stems, respectively. Set
A and Set B are also identified as ergative and absolutive, respectively,
terminology which reflects the ergative structure of Qeqchi and other Mayan
languages. Traditional terminology in Mayan studies is Set A and Set B,
however, and is the terminology which will be used throughout this grammar.
10.

11.

Noun Possession: Set A

12. The Set A pronominal affixes are used to indicate the possessor of a
noun, as demonstrated in Charts IV and V. Any word, therefore, which
contains a Set A prefix but not a tense/aspect prefix is a noun. It will be noted
in the chapter on Statives that nouns may also take the Set B pronominal
affixes, as suffixes, in the formation of stative sentences.
13.
14.
a-

20.
S
et A

16.

15.ye

22.
=

21.tail

17.

18.19.

23.y
o
u
r
24.25.
t
a
i
l

26.
27.
28.Chart IV: Set A as Noun Possessor Before Consonant-Initial Stem
30.
Set
31.
ste
32.
Mea
29. Person
A
m
ning
33. 1st person
35.
tzi
36.
my
34.
insingular

dog
37. 2nd person
39.
tzi
40.
your
38.
aasingular

dog
41. 3rd person
43.
tzi
44.
his/h
42.
xsingular

er/its dog
47.
tzi
48.
our
45. 1st person plural
46.
qa
dog
51.
tzi
52.
yall'
49. 2nd person plural
50.
ee
s dog
55.
tzi
56.
their
53. 3rd person plural
54.
x - eb'
dog
57.
58.
26

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

59.

Chart V: Set A as Noun Possessor Before


61.
Se
62.
st
60. Person
tA
em
64. 1st person
65.
66.
a
singular
waq
68. 2nd person
69.
70.
a
singular
aawaq
72. 3rd person
73.
74.
a
singular
raq
77.
78.
a
76. 1st person plural
qaq
80. 2nd person
81.
82.
a
plural
eeraq
84. 3rd person
85.
86.
a
plural
raq - eb'
88.
89.

27

Vowel-Initial Stem
63. Meaning
67.

my pig

71.

your pig

75.

his/her/it
s pig

79.
83.
87.

our pig
y'all's
pig
their pig

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

90. There are some exceptions to the above scheme in Charts IV and V.
Certain vowel-initial noun stems obligatorily take the consonant-initial series
of prefixes.
91.
92.inup

93.ceiba

96.imul

97.rabbit

94.in
in
u
p
98.in
im
ul

95.my ceiba

99.my rabbit

100.
101. A large number of common vowel-initial noun stems can take either
prevocalic or preconsonantal allomorphs.
102.
103.
watzam / inatzam

104.
salt

107.
w-iyaj /
in-iyaj

108.
my
seed

111.
w-ike /
in-ike
115.
w-ik / inik

112.
my
agave
116.
my
chile

my

105.
w-ukal /
in-ukal

109.

w-o / in-o

113.
w-iswa /
in-iswa

106.
my jug
110.
my
avo
cad
o
114.
my
tam
ale
118.

117.

119.
120. In these cases, if a vowel-initial noun stem occurs with the
preconsonantal series of Set A, the glottal stop which is present before all
vowel-initial stems is preserved. If the vowel-initial stem occurs with the
prevocalic series of Set A, the glottal stop is dropped.
121.
122.

anx

125.

w-anx

128.

in-anx

123.
[a:
n]
126.
[kw
a:n]
129.
[i
na:n]

124.

garlic

127.

my garlic

130.

my garlic

131.
132. Campbell (1974: 274) noted these exceptions and suggested that the
Set A preconsonantal set was the more basic set. He noted that frequently
occurring vowel-initial nouns usually take the vowel-initial series, while the
less frequently occurring vowel-initial nouns might take the preconsonantal
set, as in the above. This could mean that the prevocalic pronominal
28

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

allomorphs are learned piecemeal, while the preconsonantal set is the


productive, basic form.
133.
134.
w135.
my wife
136.
*in-ixaqil
ixaqil
137.
w138.
my older
139.
*in-as
as brother
140.
w141.
my younger
142.
*in-iitz in
iitz'in brother
143.
w-ix
144.
my back
145.
*in-ix
146.
w147.
my tongue
148.
*in-aq'
aq'
149.
150. All recent Spanish loans which are vowel initial take the preconsonantal
series.
151.
152.

in153.
fertilizer
ab'oon (Spanish abono)
155.
in156.
altar (Spanish
artal altar)
158.
in159.
sugar (Spanish
asuukr azucar)

154.*w-ab'oon
157.*w-artal
160.*w-asuukr

161.
162. This evidence, especially that of the Spanish loans, suggests that the
preconsonantal set is in fact the productive set and that the prevocalic set is
learned piecemeal and is limited to a small subset of noun stems.
163.

164.

Subject of Intransitive Verb: Set B

165. The Set B pronominal affixes are used to indicate the subject of
intransitive verbs, that is, verbs which denote either an "agent" who carries
out the action indicated by the verb or a "patient" who suffers the action of
the verb, but not both. By way of definition, any word which contains a Set B
prefix, but not a Set A prefix, is an intransitive verb. The examples below
demonstrate an intransitive verb with an agent subject and another
intransitive verb with a patient subject. Elements are:
tense/aspect
prefix - Set B - intransitive verb stem.
166.
167.
170.

175.

Agent subject
Patient subject

Subject

178.1st person

168.
x - at b'e
171.
x - at kam

169.

you walked

172.

you died

173.
174.Chart VI: Set B with an Intransitive Verb
176.
Tense/aspect - Set
177.
Mean
B - Intransitive
ing
179.

x - in - titz

29

180.

I got

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

singular
181.2nd person
singular
184.3rd person
singular

182.

x - at - titz

185.

x - - titz

188.

x - oo - titz

190.2nd person plural

191.

x - ex - titz

193.3rd person plural

194.

x - e - titz

187.1st person plural

196.

197.

bored
183.
you
got bored
186.
he/she
got bored
189.
we got
bored
192.
y'all
got bored
195.
they
got bored

Stative Sentences: Set B

198. The Set B pronominal affixes are used to indicate the subject or theme
of stative sentences. Stative sentences here refer to verbless, equative
sentences where the predicates indicate the identity, state, or condition of
the subject. A stative predicate may be defined as any word which contains a
Set B suffix and which, as noted above for nouns, may not contain a
tense/aspect prefix. In order to further distinguish statives from intransitive
verbs, which they may at times resemble, it is helpful to consider the statives
as describing "states" and the verbs as denoting some kind of action. The
following examples, showing an adjective used as a stative and an intransitive
verb derived from that adjective, are illustrative:
199.
200.
201.
y
at
205. 206.
s
Set
B
210.
211.
x
at215. 216.
p
Set
B

202.
ch
a
q
207.
pa
s
t
212.
ya
j
o

217.
get
s
i
c
k

203.

208.
=

204.

209.
you were sick (but the
action or process of getting sick
is not referred to).

213.

218.
=

214.

219.
you got sick (here the
process is referred to but not
the resultant state).

220.
221. Predicates may be nouns, adjectives, positional adjectives, or
participles. Examples with predicates indicating identity, state, and
condition:
222.
223.
Predic
ate - Set B
226.
winq

224.

Meaning

227.

you (are)

30

225.
Predic
ate type
228.
identit

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

at
229.
chaabi
l at
232.
yaj
at

a man
230.
you (are)
good
233.
you (are)
sick

y
231.
234.
on

state
conditi

235.
237.
ect
240.
person
244.
person
248.
person
252.
person
256.
person
260.
person

Subj
1st
singular
2nd
singular
3rd
Singular
1st
plural
2nd
plural
3rd
plural

236.
Chart VII: Set B with
238.
Predicat
239.
e
B
241.
242.
243.
yaj
- in
245.
246.
247.
yaj
- at
249.
250.
251.
yaj
-
sick
253.
254.
255.
yaj
-o
257.
258.
259.
yaj
- ex
261.
262.
263.
yaj
- eb

Stative Predicate
Meaning
I (am) sick
you (are) sick
he/she/it (is)
we (are) sick
y'all (are) sick
they (are) sick

264.
265. The following examples exhibiting Set B with other verbless predicates.
For more on these predicates, see sections on Nouns and Statives. For an
explanation of suffixes -k and -(a)q, see below.
266.
267.
Predic
ate
272.
yokyoo
277.
Predic
ate
282.
sak'b'il
287.
Predic
ate
292.
warena
q297.
Predic
ate

268.

onfuture
273.
k
278.
283.

269.
S
274.
in
279.
284.

in

270.
Positio
nal Adjective
275.
280.
Transiti
ve Participle
285.
290.

288.
293.

289.
294.

in

Intrans
itive
Participle

271.
276.
I (am/was)
lying down
281.
286.
hit
291.

295.
296.
300.

298.

299.

I (am/was)

Noun
(human
profession)

31

301.

I have slept

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

302.
aj
b'ak
'ol
si 307.
bakon
el
312.
aj
b'ak
'one
l317.
Predic
ate
322.
aj San
Juan
327.
aj
Cha
mil -

303.

308.

313.

304.
in

305.

306.
I am one
who ties up
firewood

309.

310.

311.

315.

316.
I am one
who ties (things)
up

314.

in

320.
318.

323.

328.

319.
324.

in
329.

in

Noun
(human
origin)

he-whoties-up (things)

321.

325.

326.
I am from
San Juan

330.

331.
I am from
Chamil

332.

333.

32

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

334. Agent of Transitive Verb: Set A / Patient of


Transitive Verb: Set B
335. In transitive verbs, which require both an agent who carries out the
action of the verb and a patient who suffers that action, both Set A and Set B
pronominal affixes are required. The Set A affixes indicate the agent; the Set
B affixes indicate the patient. The examples below demonstrate both the full
range of Set A agents, including the series occurring before vowel-initial
stems as well as the series occurring before consonant-initial stems, and the
Set B patients.
336.
337.
The normal order of elements in a transitive verb are
tense/aspect prefix - Set B patient - Set A agent - transitive verb stem, as
shown in nearly all the examples given in Charts VIII and IX. However, as
noted by Pinkerton (1976: 55), there are two variants to the above order:
the Set B patient may occur following the verb stem, or the Set B patient
may occur doubled, both following the tense/aspect prefix and also following
the verb stem. Thus the full range of possibilities for ordering elements in a
transitive verb follows.
338.
339.
340.
Exam
ple

(1) tense/aspect - Set B patient - Set A agent - verb stem


347.
341. 342. 343.
344. 345. 346.
350.
xati
ch'
x
at
in
nch'aj
348. 349.
352.
354.
356.
358.
361.
I
351.
p
353. Set
355. Set
357. wa
washed
B
A
359. 360.
you
362.
(2) tense/aspect - Set A agent - verb stem - Set B patient
368.
363.
364. 365. 366.
367. ch'
369. 370.
373.
xin
Exam
371.
x
in
a
at
ch'ajat
ple
j
372.
375.
377.
379.
381.
384.
I
374.
p
Set
was
Se 382.
washed
376.
A
378.
h 380.
you
383.
385.
(3) tense/aspect - Set B patient - Set A agent verb stem Set B
patient
386.
393.
395.
387. 388.
390. 391. 392.
394.
396.
xati
Exam
389.
ch'
a
x
in
nch'ajat
ple
at
398.
400.
402.
404.
406.
407.
I
397.
p
Set
Set
wa
S
washed
399.
B
401.
A 403.
405.
you

408.
409.
Order (1) is normally the most common of the three possible orders,
but there are exceptions. In Charts VIII and IX it will be noted that there
are two possibilities, one where the agent is 3 pl and the patient 2 pl and
another where both agent and patient are 3 pl, for which "No Form" is
entered. In these cases a passive construction results, removing the agent
33

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

from the verb entirely. The expected forms *x-ex-ex- b'oq they called y'all
and *x-eb'-ex-b'oq they called them (as well as the other variants) are
replaced by passives, which function in terms of inflection like intransitive
verbs:
410.
411.
x-

412.

416.
past

417.

421.

422.

426.
x-

427.

431.
past

432.

413.
boqe

418.
be called

xet B

414.
(xbaaneb)
419.
(bythem)

423.

428.
boqe

433.
be called

et B

424.

429.
(xbaaneb)
434.
(bythem)

415.
420.
Yall were
called (by them).
425.

430.
435.
They were
called (by them).

436.
437. Suffix -e is the suffix which normally derives passive stems from
transitive active stems and should not be confused with the homophonous 3rd
plural Set B allomorph. For an explanation of xb'aaneb', see section in Nouns on
relational nouns.
438.
439. The second exception involves all other forms with 3rd plural Set B
patients. It will be noted in Charts VIII and IX that all forms are given with Set
B patient following the verb stem and that some forms have only this ordering
of elements. Apparently this ordering, (2) above in this section, is the more
normal ordering with 3rd plural patients, while the ordering in which the Set B
patient follows the tense/aspect prefix, (1) above in this section, is impossible
where the agent is 3rd singular or 2nd plural and less normal with other agent
pronouns. Pinkerton (ibid.) pointed out that the 3rd singular agent with 3rd
plural patient (*x-e-x-b'oq) would be homophonous with a 3rd plural agent
with 3rd singular patient (x-ex-x-b' oq) and that this might explain why the
former did not occur.
440.
441. As a way of explaining the variation in Set B placement, Pinkerton
(1976: 56) had this to say:
442.
443.Marlys Bacon has told me that Mayan languages can be
roughly divided into three groups whose geographical location
correlates with the prefixing or suffixing of the absolutive (i.e., Set
B) pronouns to the verbs. The lowland languages suffix the
absolutive pronouns to the verb and the highland languages prefix
them to the verb. There are also "buffer zone" languages which
both prefix and suffix the absolutive pronouns to the verb. Qeqchi
seems to fall into the "buffer zone" group.
444.

34

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

445. In support of this statement it should be mentioned that Qeqchi, as a


member of the Kichean branch of Mayan languages, is historically a highland
language. But it can also be shown that at the time of the conquest, Qeqchi
was in contact with Chol and Lacandon (King, 1974), both lowland languages,
and Chol, at least, suffixes the Set B patients (Attinasi, 1973: 165).

35

1. Chart VIII: Set A Agents and Set B Patients with Vowel-initial Transitive Verb

Stem

QEQCHI GRAMMAR

3. P
2.
A

12.
2

21.
3

30.
2

39.
3

48.
1

57.
3

66.
1

75.
3

84.
1

93.
2

102.
3

111.
1

a
t
i
e
n
t

4.

14.

23.

32.

41.

50.

59.

68.

77.

95.

104.

121.

26.

ir-

113.
x-

19.

xi

27.

28.

xi

-il

33.
in-

34.
e

35.

36.
-il

42.
in-

43.
e

44.

51.
at-

52.
w

53.

at-

61.

62.

r-

at-

70.

71.

q-

at-

79.
e

80.

88.
w

89.

97.
a

98.

106.
r-

107.

72.

73.

115. 36
116.
q-

atril

atqil

81.

82.

90.

91.

99.

100.

108.

109.

xr

-il

114.

atwil

64.

-il

105.

neril

63.

-il

96.

neeril

55.

-il

87.

xi

54.

-il

78.

37.

xi

-il

69.

niril

46.

-il

60.

naawil

45.
-il

x-

112.
3
s
g

25.

x-

103.
3
s
g

in-

x-

94.

3
s
g

24.

x-

86.

18.
-il

x-

85.

3
s
g

17.

x-

76.

2
s
g

x-

67.

2
s
g

16.

J
oined
Word

10.

s
t
e
m

x-

58.

2
s
g

in-

8.

x-

49.

2
s
g

15.

x-

40.

1
s
g

e
r
b

7.

x-

31.

1
s
g

6.

22.

1
s
g

e
t

13.

1
s
g

5. S

9. v
STEPHEN O. STEWART, PH.D.

117.
-il

ateril

wil

awil

il

118.

qil

M
eanin
g

11.

Y
ou saw
me.

20.

29.

38.

47.

56.

65.

74.

83.

92.

e saw
me.
all
saw
me.
hey
saw
me.
saw
you.
e saw
you.
e saw
you.
hey
saw
you.
saw
him.
Y
ou saw
him.

101.

110.

119.

128.

e saw
him.
e saw
him.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

447.

Chart IX: Set A Agents and Set B Patients with Consonant-initial


Transitive Verb Stem

449.

448.
A

P
a
t
i
e
n
t

453.
S
450.
te

451.
Se

452.
S
454.
B

455.
v

456.
Join
ed Word

457.
Meanin
g

465.
xina
aboq

466.
You
called
me.

474.
boq

475.
He called
me.

459.
1
458.
2

s
g
468.

461.
in460.
x-

1
467.
3

s
g
477.

470.
in-

s
g
486.

479.
in-

s
g
495.

488.
in-

s
g
504.

497.
at-

512.
1

s
g
513.

498.
i
499.

496.
x-

2
503.
3

489.
e
490.

487.
x-

2
494.
1

480.
e
481.

478.
x-

1
485.
3

471.
i
472.

469.
x-

1
476.
2

462.
a
463.

506.
at505.
x514.
x-

515.
at-

507.
x
508.

516. 517.
q

37

464.
-

473.
-

482.
-

491.
-

500.
-

509.
518.
-

xinix

483.
xine
eboq

484.
Yall
called
me.

492.
xine
xboq

493.
They
called
me.

501.
boq

502.
I called
you.

xatin

510.
xatx
boq
519.
xatq
aboq

511.
He called
you.
520.
We

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2
s
g
522.

called
you.

2
521.
3

s
g
531.

524.
at523.
x-

3
530.
1

s
g
540.

533.

s
g
549.

542.

s
g
558.

551.

s
g
567.

560.

s
g
576.

569.

584.
2

s
g
585.

570.
e
571.

568.
x-

3
575.
3

561.
q
562.

559.
x-

3
566.
2

552.
x
553.

550.
x-

3
557.
1

543.
a
544.

541.
x-

3
548.
3

534.
i
535.

532.
x-

3
539.
2

525.
e
526.

578.

577.
x586.
x-

587.
oh

579.
e
580.

588. 589.
a

38

527.
-

536.
-

545.
-

554.
-

563.
-

572.
-

581.
590.
-

528.
xate
xboq

529.
They
called
you.

537.
oq

538.
I called
him.

xinb

546.
oq

xab

555.
q

xbo

564.
oq

573.
oq

xqab

xeeb

582.
xex
boq
591.
xoha
aboq

547.
You
called
him.

556.
He called
him.

565.
We
called
him.

574.
Yall
called
him.

583.
They
called
him.
592.
You
called

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

p
l
594.

us.

1
593.
3

p
l
603.

596.
o595.
x605.
oh

1
602.
2

p
l
612.

p
l
621.

614.
oh

p
l
630.

623.
ex

p
l
639.

632.
ex

p
l
648.

641.
ex

656.
1

p
l
657.

642.
q
643.

640.
x-

2
647.
3

633.
x
634.

631.
x-

2
638.
1

624.
i
625.

622.
x-

2
629.
3

615.
e
616.

613.
x-

2
620.
1

606.
e
607.

604.
x-

1
611.
3

597.
x
598.

649.
658.
x-

650.
No

651. 652.

659.
(e

660. 661.
i
(

3
p

39

599.
-

608.
-

617.
-

626.
-

635.
-

644.
-

600.
oq

xoxb

609.
xohe
eboq

610.
Yall
called
us.

618.
xohe
xboq

619.
They
called
us.

627.
xexi
nboq

628.
I called
yall.

636.
oq

637.
He called
yall.

xexb

645.
xexq
aboq

646.
We
called
yall.

654.
663.
xebi
nboq/xinb
oqeb

655.
They
called
yall.
664.
I called
them.

653.
N
662.
-

601.
He called
us.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

l
666.
3
665.
2

p
l
675.

p
l
684.

p
l
693.

p
l
702.

p
l

710.
711.
712.
713.
714.

715.

687. 688.
q
e

685.
x696. 697.
e
e

695.
694.
x704.
No

3
701.
3

678. 679.
x
e

686.

3
692.
2

677.
676.
x-

3
683.
1

669. 670.
a
(

667.
x-

3
674.
3

668.
(e

705. 706.

703.

671.
-

680.
-

689.
-

698.
-

672.
xeb
aaboq/xab
oqeb

673.
You
called
them.

681.
qeb

682.
He called
them.

xbo

690.
xqab
oqeb

691.
We
called
them.

699.
xeeb
oqeb

700.
Yall
called
them.

708.

709.
They
called
them.

707.
N

Independent Pronouns

716. Independent pronouns, as evidenced by their final syllables, appear to


be more closely related to the Set B pronominal affixes and were probably
derived from them originally.
717.
718.
P
erson
721.
1st
singular
726.
2n
d

singular
731.
3rd

719.

et B
722.

in

727.

at

732.

720.
723.
laain
728.
laaat
733.

40

Independent Pronouns
724. 725.
729. 730.
734. 735.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

singular
736.
1st
plural
741.
2n
d
plural
746.
3rd
plural

737.

(h)aa
n
738.
laao
743.
laaex
748.
(h)aa
ne
b

742.

ex

747.

eb

739. 740.
744. 745.
749.
o
750.
aan

(h)eb

751.
752. In some areas, according to Haeserijn (1966: 12), there exists what
appears to be an archaic set of pronouns more consistent with the present 3rd
person forms. This set is:
753.
754.
755.
1st
singular
757.
2n

756.

hain

singular
759.
3r

758.

haat

singular
761.
1st
plural
763.
2n
d
plural
765.
3r
d
plural

760.

haan

762.

hao

764.

haex

766.

haaneb

767.
768.
769.
The present set of independent pronouns appears to represent the
definite article li plus the non-3rd person pronouns of the archaic set frozen
into a single form. The vowel -a- in the first syllable of the archaic forms
has been lengthened to -aa- to compensate for the loss of the -i- of li and
the initial h- of the archaic forms.
770.
771.
772.
P
erson
775.
1st
singular
778.
2n
d

singular
781.
1st
plural
784.
2n
d
plural

773.
Article plus
archaic pronoun

774.
Prese
nt pronoun

776.

li + hain

777.

laain

779.

li + haat

780.

laaat

782.

li + hao

783.

laao

785.

li + haex

786.

laaex

41

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

787.
788.

789.
790.

42

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

791.

Use of Independent Pronouns

792. Independent pronouns are used to emphasize the agent or patient


when pronominal affixes are also present.
793.
794.
la
801.
yo
808.
815.
la

795.
x802.
pa
809.

816.
x-

822.
I

823.

829.

830.

836.
ch

837.
843.
ni
844.

796.
at

797.
in

798.
sa

803.
Se

804.
Se

805.
hi

810.

811.

812.

817.
at
824.
831.

838.
at
845.
Se

818.
in825.
832.

819.
sa
826.
833.

840.
la

839.
847.
yo
846.

799.

800.

806.
=

807.
I hit
(emphasis) you.

813.

820.
=

814.

821.
(emphasis)
I hit you.

827.
834.

828.
835.

841.

842.

848.
=

849.
(emphasis)
You are nice.

850.
851. Independent pronouns are used to identify people in response to "who"
questions.
852.
853. "Ani xsak?" "Laain."
"Who hit him?"
"(It was) I."
854.
855. Independent pronouns may also be used as the subject of stative
sentences in which the predicate identifies the subject, though not with the
other stative predicates mentioned in conjunction with Set B, state and
condition. In fact, independent pronouns are more common and have no
restrictions placed on them as subject of stative sentences with identity
predicates, whereas Set B in this context are less common and may
specifically not be used if the identity predicate is a compound.
856.
857. For example, it is possible to say both of the following sentences:
858.
859.
laain aj b'anonel
I (am a) medicine
man.
860.
aj b'anonelin
I (am a) medicine man.
861.
862. But to identify oneself as a tailor, raqol tikr (literally, cloth
measurer), it is necessary to use the independent pronoun.
863.
864.
laain aj raqol t'ikr
I (am a) tailor.
865.
43

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

866. The independent pronouns are also obligatorily used with the
absolutive voice.
867.
868.
I
874.
Ia

869.
p
875.
x-

881.

870.
S

871.

absolu
tive
intransitive

872.
Set
A

876.
in
877.

878.
aaw-

ab'in
880.

873.
dative
relational noun
879.

laain xin-ab'in aawe.

I (was the one who) heard you

44

882.
883.

884.

885.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

886. CHAPTER 4: VERBS


887.
888. Morphemes which can
be inflected for tense/aspect
are termed verbs, or verb
stems.
889.
890. A verb stem devoid of
derivational suffixes may also be
termed a verb root.
891.
892. Verb roots are always
monosyllables of the shape
(C)VC. Derived verb stems are
polysyllabic and contain a root
plus one or more derivational
suffixes, though for a few such
stems the root does not occur
elsewhere and its meaning may
be unrecoverable synchronically.
893.

894. Verb stems are either


transitive or intransitive.
895.
896. Transitive stems are those
in which Set A pronominal affixes
are employed to denote the
agent, who does the action
indicated by the verb stem, and
in which Set B pronominal affixes
are employed to denote the
patient, who suffers the action
indicated by the verb.
897.
898. Intransitive stems are those
in which only the Set B
pronominal affixes are found.
These denote the subject,
whether agent or patient, of the
action of the verb stem.
899.
900. The verb stem plus
pronominal affix(es) plus
tense/aspect inflection are
required to form the verb word,
though there are affixes for
direction and manner which may
optionally occur in the verb word.

901.
902.
903. Intransitive stems may be derived from transitive stems, and
vice versa.
904.
905. Derivation per se is not treated in depth in this grammar and no
attempt will be made to identify all the derivations which can occur in the
formation of a stem, though occasional analysis of certain derivations will be
required, such as those involved in the formation of passive and antipassive
(absolutive) voices.
906.
907. The classes of verb stems presented below are determined primarily on
the form of the derivational suffixes which derive those stems.
908.
909. The term infinitive, a verbal noun which "names" the action of
the verb, is used in a manner slightly different than is traditional in that the

47

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

infinitive may be inflected for person, whereas in normal usage infinitives are
uninflected.
910.

911.
912.

48

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

913.

Classes of Transitive Verb Stems

914. There are two classes of transitive verb stems, termed root transitives
(vtR) and derived transitives (vtD. The vtR stem is a monosyllabic, underived
root of the shape (C)VC-. The infinitive is formed by the root plus the
derivational suffix -b'al and is always possessed by the Set A pronominal
affixes.
915.
916.

917.
b'ek

921.

922.

927.
x-

928.
b'ek934.
vtR
s
t
e
m

933.
Set
A

918.

dig

it!
923.

919.
(stem =
imperative)

924.

925.

930.
b'al

935.

infi
nitive
suffix

926.

932.
931.

929.

920.

xb'ek
b'al

936.

937.

diggin
g it/its
digging

938.

939.
940. The vtD transitives are polysyllabic derived stems, although there is a
small number of stems in this class for which the derivation is not
synchronically recoverable in that the root does not occur elsewhere.
941.
942. These transitives normally end in a stem formative vowel, with the
following exceptions: transitives derived from positional roots, explained
below, drop the stem formative vowel -a in all inflected forms; causative
transitives may optionally drop the stem formative vowel -i of the
derivational suffix -si ~ -resi ~ -tesi ~ -b'esi.
943.
944. Infinitives are formed by first forming the absolutive intransitive
infinitive and then adding the derivational suffix -il plus the Set A possessive
prefix.
945.
946. Examples:
947.
948.
949.
Root
955.
b'aan961.
k'ay967.

(1)

Stems (= Imperative)
950.
Tra
nsitive
Derivatio
951.
nal Suffix

952.
vtD
ste
m

957.
=

958.
b'aanu

956.

962.

963.
=

964.
k'ayi

969.

953.

954.

960.
d

959.
do it!
965.

966.
w

ell it!

970.

968.

971.

49

972.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

973.

(2)

Infinitives

974.
Set A

975.

979.
x-

980.

vtD
stem
b'aa
nu

987.

985.

990.
996.
x-

976.
Absolut
ive
Intr
ansi
tive
981.
n
=

986.
992.

991.

997.

k'ay
i

1002.

977.
Infiniti
ve
982.
k
988.
xb'aanu
nkil
993.

998.
n
1003.
=

1006.
1007.

50

999.
k
1004.
xk'ayink
il

978.
vtD
infinitive
suffix
983.

il

984.

989.
doing it,
to do it
994.

1000.

il

995.

1001.

1005.
buying
it, to buy it

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1008.Transitives are derived from positional roots by the derivational


suffix V1b'a in which V1 represents the reduplicated root vowel. The final
-a is dropped in finite forms.
1009.
1010.
Posit
io
n
al
R
o
ot

1011.
Verb
Deri
vati
ona
l
Su
ffix

1018.
chun
-

1019.
ub a

1012.

1013.

1014.

1023.1024.
=
chunub
1020.

1021.

1022.

1028.

1029.
1037.
vtD
St
e
m
1045.
chunu
b

1030.

1027.
1026.

1034.
Past
1042.
x-

1035.
Set B
1043.
-

1016.
vtD
1015.
Stem

1036.
Set A
1044.
x-

1032.
1031.
1039.
1040.

1050.

1058.
Set
A
1066.
x-

1059.
vtD
ste
m
1067.
chunu
b'a -

1052.

1053.

1060.
Absol
ut
iv
e
1068.
n-

1061.
Absol
ut
iv
e
In
fin
iti
ve
1069.
k-

1074.

51

1025.
set it
dow
n!
1033.

1041.

1038.
1047.1048.
=
xchunub

1046.

1051.

1017.

1056.
1054. 1055.
1062.
vtD
I
n
fi
n
it
i
v
1064.
e
1063.
1070. 1071.1072.
-il
=
xchunub
ankil

1049.
He set
it
dow
n
1057.

1065.
1073.
setting
it
dow
n, to
set
it
dow
n

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1075.
Below are shown derived causative transitive verbs,
derived from an adjective, kaw hard.
1076.
1077.
ka
w
r
e
s
(
i
)

1083.
1078.
harde
n
it!
1086.

1085.
1093.
Pas
t
1101.
x-

1109.

1117.
Set
A
1125.
x-

1094.
Set B
1102.

1084.

1080.
1079.

1087.
1095.
Set A
1103.
x-

1110.

1082.

1089.

1090.

1088.

1096.
vtD
ste
m
1104.
kawres(
i)

1098.

1097.
1105.

1119.
absol
ut
iv
e
1127.
n-

1113.

1120.
absolu
tive
infi
niti
ve
1128.
k-

1133.

1134.

52

1121.
vtD
i
n
fi
n
it
i
v
e
1129.
il

1091.

1092.

1099.

1100.

1106.1107.
=
xkawre
s(i)

1112.
1111.

1118.
vtD
ste
m
1126.
kawres
i-

1081.

1108.
he
hard
ened
it

1115.

1116.

1123.

1124.

1114.

1122.
1130.1131.
=
xkawre
sinki
l

1132.
hardenin
g it,
to
hard
en it

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1135. Classes of Intransitive Verb Stems


1136.
Intransitive verbs are defined as those verbs which employ only Set
B pronominal affixes to denote the subject, whether agent or patient, of
the verb, but not Set A.
1137.
1138.
Intransitive verbs, like transitive verbs, are either non-derived or
derived. But the derived intransitives, unlike the derived transitives, have
a variety of stem formations.
1139.
1140.
The classes of intransitive verbs presented below are determined in
part by the form of the derivational suffix but also by the
interrelationships between certain intransitive stems and the transitive
stems mentioned above.
1141.
1142.
In presenting the intransitives mention will be made of these
interrelationships but detailed discussion will be reserved for a section
later on Voice.
1143.
1144. There are nine classes of intransitive verbs.
1145.
1146.
Class viR
1147.
Class viR are intransitives which have monosyllabic, non-derived stems.
Examples:
1149.

1148.
1156.
1157.
war1164.
viR
stem
(bound)
1171.
1179.
war1186.
Ste
m

1151.
1150.
1158.

1165.

1172.

1174.

1181.

1187.
Infi
nitive
Suffix

1188.

1194.

1196.

1193.

1195.

1201.
t-

1202.
in-

1209.
Futu
r

1210.
Set B

1154.

1155.

1160. 1161.

1162.

1163.

1167. 1168.

1169.

1170.

1175. 1176.

1177.

1178.

1166.

1173.

1180.

1152. 1153.
1159.

1203.
w
1204.
q
1211. 1212.
S
Future

1183.
1182.
war
=
k

1185.

1191.

1192.

1197. 1198.

1199.

1200.

1208.
I'm
going to sleep
1207.

1213.1214.

53

to sleep

1184.

1189. 1190.

1206.
tin
w
1205.
a
=
r
q

sleep

1215.

1216.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

e
1217.

1218.

1220.
1219.

1225.
war-

1226.
in

1233.
Ste
m

1234.
Imper
ati
ve

1227.

1235.

1242.
1241.

1249.
x-

1257.
Past

1221. 1222.

1228.

1230.
1229.war
=
i
n

1223.

1224.

1232.

sleep!

1231.

1236.
1244.

1243.

1250.

1251.
1252.
w
(k)

1258.
Set B

1260.
1259. NonS
fut
ure

1237. 1238.

1239.

1240.

1245. 1246.

1247.

1248.

1254.
xwa
1253.
r
=
(
k
)

1261.1262.

1265.

1266.

54

1256.
1255.

1263.

1264.

he slept

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1267.
Class vin
1268.
Class vin are intransitives which have stems ending in -n. Some
members of this class have a special relationship with transitive stems of class
vtD. The absolutive voice of vtD stems is formed by deriving an intransitive vin
stem from the vtD stem by means of suffix -n . Examples:
1276.

1269.
k'ulun1270.
1277.
vin stem
(bound)
1284.

1272.
1271.

1286.

1293.

k
1294.

1300.

Infinitive Suffix

1306.

1315.

1282. 1283.

1288. 1289.

1290. 1291.

1308.

n
1323.
S
et B

1316.
k
'ulun
1324.
S
tem

1330.

1317.
q
1325.
Future
1333.

1331.

1332.

1338.

1339.
k
'ulun -

1340.

1346.
Impera
tive

1347.
S
tem

1348.
F
uture

1354.

1341.
at
1349.
Set B
1357.

1355.

1356.

1362.
x-

1363.

1364.
k
'ulun -

1365.
(k)

1370.
Past

1371.
S
et B

1372.
S
tem

1373.
Nonfutu
re

55

1295.
=
1296.
k'ulun
k

1298.
to
c
o
m
1297. e

1302. 1303.

1304. 1305.

1310. 1311.

1312. 1313.

1301.
1309.

1307.

1314.
t
1322.
Future

1280. 1281.
1287.

1285.

1292.
k'ulun1299.
Stem

1275.

1279.

1278.

com
e

1273. 1274.

1318.1319.
=
tink'ul
un
q

1321.
I'll
c
o
m
1320.
e

1326.1327.

1328. 1329.

1334. 1335.

1336. 1337.

1350.1351.

1352. 1353.

1358. 1359.

1360. 1361.

1343.
1342.k'ulun
=
qa
t

1345.
com
e
1344.
!

1366.1367.
=
xk'ul
un
(k)

1369.
he
c
a
m
1368.
e

1374.1375.

1376. 1377.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1378.
1379.
Class vik
1380.
Class vik are intransitives which have stems ending in a short vowel.
When inflected in other than the future these intransitives obligatorily mark
nonfuture with suffix -k, while with all other intransitives the non-future suffix
-k is either optional or not used at all. Some members of this class have a
special relationship with transitive stems of class vtR parallel to the
relationship of vin to vtD mentioned above. The absolutive voice of vtR stems
is formed by deriving an intransitive viK stem from the vtD stem by means of
suffix -o or, if the vtR root vowel is -u-, by means of suffix u. Examples:
1381.

1389.
see -

1382.

1383.

1384.

1385. 1386.

1390.

1391.

1392.

1393. 1394.

1405.

1412.
see

1413.

1398.

1399.

1400. 1401.

1402.

1406.

1407.

1408. 1409.

1410.

1415.1416.
=
seek

1418.
to
lau
1417.
gh

1421.

1422. 1423.

1424.

1429.

1430. 1431.

1432.

1436.
s
ee

1437.
q

1439.
1438.
tinse
=
eq

1441.
I'll
lau
1440. gh

1444.
S
tem

1445.
Fut
u
r
e

1446.1447.

1448.

1453.

1454. 1455.

1456.

1460.

1461.1462.
=
seen

1464.
1463.laugh!

1467.

1468. 1469.

1470.

1475.

1476. 1477.

1478.

1485.
1484.
xsee
=
k

1487.
he
lau
gh
1486. ed

1492.1493.

1494. 1495.

k
1414.

1434.
t-

1442.
Future
1450.

1458.
see
1465.
Stem
1472.

1395. laugh

1403.

1404.

1426.

1388.
1396.

1397.
vik stem
(bound)

1419.
Stem

1387.

1420.

Infinitive Suffix

1427.

1428.

1435.

i
n-

1443.
S
et B
1451.

1452.

1459.
n
1466.
Imperative
Suffix
1473.

1474.

1480.
x-

1481.

1488.
Past

1489.
S
et B

1482.
s
ee 1490.
S
tem

56

1483.
k
1491.
Non
-

1411.

1425.
1433.

1449.
1457.

1471.
1479.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

f
u
t
u
r
e

1496.
1497.

57

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1498.
Class viV
Class viV are intransitives which are stems ending in a long vowel. When
inflected without a future suffix q, the long vowel VV becomes a short vowel
V. Passives related to vtD stems are viV intransitives when the VV remains in
all forms followed by future suffix q or non-future suffix k.
1501.
viV
1502.
1507.
stems (bound)
1503.
1504. 1505. 1506.
1513.
1514.
1508.
1512.kub'e
to go
kub'ee
=
e
dow
1509.
k
1510.
k
n
1511.
1521.
1520.
1515.
1519.
to get
kalaa
kalaa =
drun
k
1516.
k
1517.
k
1518.
1522.
1523.
infinitive
1528.
Stem
suffix
1524.
1525. 1526. 1527.
1499.
1500.

1529.

1530.

1531.

1536.

1532.

1533.

1537.
t-

1538.
in -

1539.
kub'ee -

1540.
q

1541.

1545.
t-

1546.
in -

1547.
kalaa -

1548.
q

1549.

1554.
Set B

1555.
Stem

1556.
Futur
e

1557.

1562.

1563.
1564.

1565.

1553.
Future
1561.

1569.
kub'
ee -

1570.

1576.
Stem

1577.

1583.

1584.

1591.

1598.
optative/
imperative
1605.

1606.

imperativ
e suffix

1571.

1534. 1535.

1543.
tinku
1542.
b'
=
e
e
q
1551.
tinka
1550.
la
=
a
q

1566. 1567.

1574.
1573.
1572.
kub'e
=
n
1579.

1580.
1581.

1586.

1587.

1588. 1589.

1592.
kalaa

1593.
q-

1594.
at

1599.
Stem

1600.
Futur
e

1601.
Se

1607.

1608.

1609.

58

1552.
I'll get
drun
k
1560.

1558. 1559.

1578.

1585.

1544.
I'll go
dow
n

1568.

1575.
go
dow
n!
1582.
1590.

1596.
1595.kalaa
=
q
at
1602.1603.
1610. 1611.

1597.
get
drun
k!
1604.
1612.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1613.
x1621.
x1629.
Past
1637.

1614.

1615.
kub'ee

1616.

1618.1619.
1617.
=
xkub'
e

1623.
kalaa

1622.

1630.
Set B

1631.
Stem

1638.

1639.

1624.

1625.

1626.
=
1627.
xkala

1632.

1633.

1634. 1635.

1640.

1641.

1642. 1643.

1645.
Examples of viV stems which are
passives derived from vtD stems.
1649.
ch'iila
a-

1650.

1651.

1657.
viV stem
(bound)

1658.

1664.

1666.

1672.
ch'iilaa

1665.

1673.

1652.

1646.1647.

1653.

1654. 1655.

1660.

1661. 1662.

1668.

1669. 1670.

1675.

1677.
ch'iil
1676.
a
=
a
k

1681.

1682.

1683. 1684.

1689.

1690.

1691. 1692.

1698.

1700.
tinch'
1699.
iil
=
a
a
q

1659.
1667.

k
1674.

1679.
Stem

1680.

1686.

1687.

1694.
t1702.
Future
1710.

Infinitive
Suffix
1688.

1695.
in -

1696.
ch'iilaa
-

1703.
Set B

1704.
Stem

1711.

1712.

1697.
q
1705.
Futur
e
1713.

1718.
x-

1719.

1720.
ch'iilaa-

1721.
k

1726.
Past

1727.
Set B

1728.
Stem

1729.
NonFu
tu

59

1706. 1707. 1708.


1714.

1715. 1716.

1722.

1724.
xch'ii
1723.
la
=
a
k

1730. 1731. 1732.

1620.
he went
dow
n
1628.
he got
drun
k
1636.
1644.
1648.

1656.
be
scol
ded
1663.
1671.

1678.
to
be
scol
ded
1685.
1693.

1701.
I'll be
scol
ded
1709.
1717.

1725.
he was
scol
ded
1733.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

re
1734.

1735.

1742.
But vtD
transitive:

1736.

1743.

1749.
x-

1750.

1751.
x-

1757.
Past

1758.
Set B

1759.
Set A

1741.
1737.

1738.

1739. 1740.

1744.

1745.

1746. 1747.

1753.

1754.
1755.
=
xch'ii
la

1748.

1752.
ch'iila
1760.
Stem

1765.
1766.

60

1761. 1762. 1763.

1756.
he
scol
ded
him
1764.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1767.

1768.
Class viV also contains an aberrant subset in the Chamil dialect.
Adjectives derived from transitive and positional roots of the form C 1VC-C1o or
C1VC-C1u, if the root vowel is -u-, may be inflected in the same way as other
intransitives of this class, with the following exceptions: the stem, as an
adjective, is not bound as in other verbs of this class, and there is also no infinitive. Examples:
1769.
1775.
1770.

1771.

1772.

1773.
1774.

1779.
1787.
Position
al Root
(Bound)

1780.

1781.

1782.
1783.

1788.

1789.

1790.
1791.

1778.
chun -

1784.

1795.
1804.

1813.
1819.

1797.

1798.

1799.
1800.

1805.

1806.

1807.

1808.
1809.

Position
al Root

1814.

1834.
t-

1843.
Future

1820.
Adjective
Derivational
Suffix
1827.

1835.
in -

1836.
chunc
hu
u-

1844.
Set
B

1845.
Stem
vi
V

1853.

1870.
Optati
ve/
Imp
erat
ive
1879.

1810.

1862.
chun
ch
u
u
1871.
Ste
m
vi
V
1880.

1854.

1829.
1830.

1837.
q

1839.1840.
=
tinchunc
1838.
huuq

1855.

1864.
at
1863.
q
1872.
Futur
e

1818.
sit
ting
1817. down

1847.
1848.

1823. 1824.
1832. 1833.

1842.
Ill
be
sitting
1841. down

1849.
1858.

1856.
1857.

1850. 1851.
1859. 1860.

1867.
1866.
chunchu
=
uqat
1865.

1869.
sit
1868. down!

1873.
Se
t
B

1881.

1802. 1803.

1822.

1821.

1828.

1846.
Fu
t
u
r
e

1793. 1794.

1811. 1812.

1831.

1852.

1861.

1801.

1815.1816.
=
chunchu

- chuu

1825.
1826.

1786.
sit
ting
down
1785.

1792.

1796.

chun

1776. 1777.

1882.

61

1874.
1875.

1876.

1883.
1884. 1885.

1877. 1878.
1886. 1887.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1888.
x-

1897.
Past

1889.
-

1898.
Set
B

1890.
chunc
hu
u
1899.
viV
st
e
m

1891.

1893.1894.
=
xchunc
hu
1892.

1900.

1901.
1902.

1906.
1907.
1908.
1909.

62

1896.
he
1895. sat down

1903.
1904. 1905.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

1910. Class vi
1911.
Class vi are intransitives with stems ending in -. In Coban the
combination -k in the formation of infinitives is realized -k. The - is lost before
the imperative suffix -n. Some members of this class, those in which - is
preceded by -e-, have a special relationship with transitive stems of class vtR,
parallel to the relationship between viV stems and vtD stems. The passive
voice of vtR stems is formed by deriving an intransitive vi stem by means of
the suffix -e .
1912.

1913.

1914.

1915. 1916. 1917.

1918.

1919.
In most of the other members of this class, the - is preceded by -o- and
the stems have a versive or inchoative semantic component, become. This
derivational suffix o is in allomorphic variation with -ob' . Suffix -o is found
stem finally and -ob' is found where the suffix is followed by another suffix
-b'esi ~ -tesi ~ -lesi, which derives a causative vtD stem from the in- choative
vi stem: josq'-o get angry, josq'-ob-resi make angry. Examples:
1926.
pas
1920.
nume1921.
s by
1922.
1923.1924. 1925.
1927.
josqo
1933.
get
1928.
angry
1929.
1930.1931. 1932.
1934.
1935.
1940.
1936.
1937.1938. 1939.
1941.
vi
stems
(bound)
1942.
1943.
1944.1945. 1946.
1947.
1948.

1949.

1950.

1955.

nume -

1956.

1961.

josqo -

1962.
k
1968.
in
finitive
suffix

1967.

stem

1973.

1980.
1987.
1994.

1974.

tt future

2001.

2008.
2014.
2020.

numestem

1981.
in-

1975.

1951.1952. 1953.

1959.
n
umek
1958.
~
1957.=
numek
1965.
jo
sqok
1964.
~
1963.=
josqok
1969.
1970. 1971.

1982.
nu

1988.
in1995.
Set
B

1983.
1984. 1985.
ti
q
=
numeq
1989.
1992.
ti
jos
1990.
1991.
njosqo
q
=
q
1996. 1997.
st
f
1998.1999.

2002.

2003.

2004.2005. 2006.
2011. 2012.

2010.
=

2021.

2023.2024. 2025.

2022.

2016.
2017. 2018.

63

1960.
to
pass by
1966.
to
get angry
1972.
1979.

1976.1977. 1978.

2009.
n
2015.
im
perative

1954.

n
umen

1986.
Ill
pass by
1993.
Ill
get angry
2000.
2007.

2013.
pas
s by!

2019.
2026.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2034.
optativ
e/imperative

2028.
josq

o
2035.
ste
m

2041.

2042.

2027.

2048.
2055.
2062.

xxpast

2049.
2056.
2063.
Set
B

2030.
2029. a
2031. 2032.
jo
q=
sqoqat
2036. 2037.
fu
S
2038.2039.
2043.

2044.2045. 2046.

2050.
nu

2052. 2053.

x
nume

2051.
=
2057.
jos

2059. 2060.

xj
osqo

2058.
=

2033.
get
angry!
2040.
2047.

2054.
he
passed by
2061.
he
got angry

2064.
st
2065.
2066. 2067.

2068.
2075.

2069.

2070.

2071.

2072.2073. 2074.

2091.

2092.

2093.

2094.2095. 2096.
2101. 2102.

2076.
Examples of vi stems which are passives derived from vtR
stems:
2083.
be
2077.
b'oqe
2078.
called
2079.
2080.2081. 2082.
2084.
vi
stem (bound)
2085.
2086.
2087.2088. 2089.
2090.
2098.
2104.

b'oqe stem

2110.
2117.
2124.

tfuture

2131.

2138.

2145.

2152.

2099.
k
2105.
in
finitive
suffix

2100.
=

2111.

2113.2114. 2115.

2118.
in2125.
Set
B
2132.

x-

past

2139.
2146.
Set
B

2106.
2107.

2112.

2097.

b'
oqek

2108.

2120.
2121. 2122.
ti
q
=
nb'oqeq
2126. 2127.
st
f
2128.2129.
2134.2135. 2136.

2140.
b'

2142. 2143.

2141.
=

2109.
2116.

2119.
b'

2133.

2103.
to
be called

x
b'oqe

2123.
I'll
be called
2130.
2137.

2144.
he
was
called

2147.
st
2148.
2149. 2150.

2151.

2153.
Class vit
2154.
Class vit is made up of intransitives which are composed of a C 1VC root
plus a derivational suffix C1ot or C1ut, where C1 represents the initial
consonant of the root. Allomorph C1ut occurs where the root vowel is -u-. This
class is defective in that normally it may only be inflected in the present

64

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

habitual, verbs which are termed affectives (see below).


2155.
Examp
2157.
les:
2156.
2158. 2159. 2160.
2162.

2163.

2169.
tentot
2176.
vit stem
(bound)
2182.

2189.
2195.

tentotstem

2215.
prese
nt habitual

2222.
2223.

2190.

2184.

2196.
infiniti
ve suffix
2202.

na -

2171.

2161.

2165. 2166. 2167.

2168.

2172. 2173. 2174.

2175.
pa
lpitate

2178. 2179. 2180.

2181.

2185. 2186. 2187.

2188.

2192. 2193.
t
2191.=
entotk

2194.
to
palpitate

2197.2198. 2199.

2200.

2204. 2205. 2206.

2207.

2213.
n
2212.
atento
2211.=
t

2214.
it
palpitate
s

2218.2219. 2220.

2221.

2177.
2183.

2201.

2208.

2170.

2164.

2209.

2216.
Set
B

2203.

2210.
tentot
2217.
stem

65

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2224.
2226. Minor Intransitive Classes:
vib', viw
2225.

vir,

2227.

2228.
2229.
The following three classes are minor, containing few members. It is felt
that the derivation of these classes represents morphological processes no
longer productive in Qeqchi though very productive in such sister
languages as Kiche, Kaqchikel, and Tzutujil and that these forms are a
residue of survivals from previously productive morphological processes.
2231.

2230.
2237.

2232.
2233.

2234. 2235.

2236.

Class vir are intransitives with stems ending in -r. Examples:


2239.

2240.

2245.
y
2246.
ajer 2252.
vir stem
(bound)

2247.

2259.

2260.

2238.

2258.

2265.
y
ajer
2271.
s
tem

2253.

2266.
2272.
2278.

infinitive
suffix
2279.

2277.

2284.

2241.

2242. 2243.

2248.

2249. 2250.

2251.
sick

2254.

2255. 2256.

2257.

2261.

2262. 2263.

2264.

2267.

2276.

2280.

2283.

2291.
f
uture
2298.

2305.

2312.
o
ptative
/
impera
tive

2285.
in -

2286.
yajer-

2292.
Set B

2293.
stem

2299.

2300.

2290.
I'll
get sick

2301.

2304.

2307.
q

2308.
at

2313.
stem

2314.
future

2315.
Set
B

2320.

2321.
2322.

2326.

x
-

2333.
p
ast

2327.

2334.
Set B

2328.
yajer
2335.
stem

2289.
2288. tinyaj
=
er
q

2287.
q
2294.
futu
r
e
2295.2296.

2306.
yajer

2319.

2281. 2282.

get

2270.
to
get sick

2273. 2274. 2275.

t
-

2269.
2268.
yajer
=
k

2244.

2329.

2302. 2303.

2310.
2309. yajer
=
qa
t

2311.
sickl

2316.2317.

2318.

2323. 2324.

2325.

2331.
2330.
xyaje
=
r

2336. 2337. 2338.

66

2297.

get

2332.
he
got sick
2339.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2341.

2342.

2340.

2343.

2344. 2345.

2346.

2351.
2357.

2352. 2353.
2358. 2359.

2354.

2363.

2364. 2365.

2366.

2370.

2371. 2372.

2347.
Class vib are intransitives with stems ending in -b. The Coban dialect
has lost this class, as all stems expected to be in the class have either lost the
b while retaining the glottalization (-b becomes -), resulting in verbs of class
vi, or lost the b altogether (b' becomes ), resulting in verbs of class vik.
Examples:
2349.

2348.
2355.
karab'- ~
karib2361.
vib' stem
(bound)

2368.

2367.

2374.
k
arab 2380.
s
tem

2393.

t
-

2400.
f
uture

2356.

2394.
in -

2369.

2388.

2395.
karab' -

2401.
Set B

2402.
stem

2408.

2409.

2407.

2414.

2421.
o
ptative
/
impera
tive

2415.
karab' 2422.
stem

2416.
q2423.
future

2429.

2430.

2428.

x
-

2442.
p
ast

2436.
2443.
Set B
2450.

2449.

2376.

2437.
karab
2444.
stem
2451.

2378.
2377.
karab
=
'k

2373.

2379.
to
go fishing

2382. 2383. 2384.

2385.

2389.

2392.

2390. 2391.

2398.
2397.tinkar
=
ab
'q

2396.
q
2403.
futu
r
e
2404.2405.

2399.
I'll
go fishing

2410.

2413.

2417.
at
2424.
Set
B

2411. 2412.

2419.
2418. karab
=
'q
at

2406.

2420.
Go
fishing!

2425.2426.

2427.

2432. 2433.

2434.

2445. 2446. 2447.

2448.

2452.

2455.

2431.

2435.

2360.
go
fishing

2362.

2375.
k
2381.
infinitive
suffix
2387.

2386.

2350.

2438.

67

2440.
2439.
xkara
=
b'

2453. 2454.

2441.
he
went
fishing

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2456.
Class viw are intransitives with stems ending in -w. The class is very
small and possibly disappearing, since one of its few members, xaaw-k to
vomit, has a more common variant, xawak, in the class vik. Examples:
2458.

2457.
2464.
s
2465.
aqew2471.
viw stem
(bound)
2478.

2477.

2459.
2460.

2461. 2462.

2467.

2468. 2469.

2470.
bec
ome dawn

2473.

2474. 2475.

2476.

2480.

2481. 2482.

2483.

2466.

2472.
2479.

2484.
s
aqew2490.
s
tem

2485.
k
2491.
infinitive
suffix
2497.

2498.

2496.

2503.
t
aa -

2510.
f
uture

2504.
-

2505.
saqew -

2511.
Set B

2512.
stem

2518.

2519.

2517.

2486.

x-

2531.
p
ast

2538.
2539.

2525.
2532.
Set B

2526.
saqew
2533.
stem

2489.
to
dawn

2492. 2493. 2494.

2495.

2499.

2502.

2500. 2501.

2508.
taasa
2507.
qe
=
w
q

2506.
q
2513.
futu
r
e
2514.2515.
2520.

2524.

2488.
2487. saqe
=
w
k

2463.

2527.

2521. 2522.

2529.
2528.
xsaqe
=
w

2534. 2535. 2536.

68

2509.
its
going to
dawn

2516.
2523.

2530.
it
dawned
2537.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2540.
2541.
2542.
Affective Intransitives
2543.
There is a defective subset of intransitive verbs to be mentioned which
is not determined by stem formation but rather by peculiarities of inflection
and use, called affectives. In terms of the previously mentioned classes these
verbs are found to be members of three classes: vin, vik, and vit. In terms of
inflection these verbs are defective in that they are almost always marked only
be for present-habitual tense/aspect. In marking person the vast majority mark
only third person singular, somewhat fewer mark third person plural as well,
while a much smaller number with a human subject may also mark non-third
person. In terms of use affectives may be used as main verbs but more usually
appear to act as adverbs or adjectives in describing the manner in which
something is done or the way something is acting. Examples:
2544.
2545. 2546.

2551.
na 2558.
presen
thabi
tual
2565.

2572.
nak 2579.
presen
thabi
tual
2586.

2593.
n2600.
presen
thabi

2553.
2552. ch'aj
ch
'o
t
2559.2560.
S
stem
(v
it
)
2566. 2567.

2574.
tub'u
2573.
b'
e
n
a
2580.2581.
S
stem
(v
ik
)
2587. 2588.

2595.
lubu
b
2594.
n
a
2601.2602.
S
stem
(v
ik

2547.

2548. 2549.

2556.
n
2555.
ach'ajc
=
h'ot
2554.

2550.

2557.
its empty,
unoccupied

2561.

2562. 2563.

2564.

2568.

2569. 2570.

2571.

2575.
k

2576.2577.
=
naket
ubub
nak

2578.
theyre
grouping together

2582.
nonfu
tu
re

2583.2584.

2585.

2589.

2590. 2591.

2592.

2596.
k

2597.
=
2598.
n
inlubu
bnak
2604.2605.

2603.
nonfu
tu

69

2599.
tired
2606.

Im feeling

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

tual

re

2607.
2608. 2609.

2614.
na 2621.
presen
thabi
tual

2616.
2615. boql

o
2622.2623.
S
stem
(v
ik
)

2610.

2617.
k
2624.
nonfu
tu
re

2611. 2612.

2618.2619.
n
=
ab'oq'l
ok

2625.2626.

2628.
2629.

2630.

70

2613.

2620.
He's
making the sound
of husking corn or
pulling feet out of
the mud.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2631. Verb Inflectional Affixes


2632.Verb inflectional affixes can be grouped into three types:
pronominal affixes, direction and manner affixes, and affixes which
denote notions of tense, aspect, and mood. Pronominal affixes were
presented in Chapter II.
2633.
2634.The affixes for tense, aspect, and mood (abbreviated T/A/M) are the
topic of the present section; the direction and manner affixes are treated
in the following section. However, it seems appropriate to provide at this
point an overview of how these three types of affixes may occur with verb
stems.
2635.
2636.
Verb inflection can be stated in terms of formulas. The charts
below show the inflectional elements which make up the verb word, though
without showing the requirements for co-occurrence.
2637.
2638.
2639.
T
/A/M
2648.

2657. i-

2666. -

2675.ak- t
2684.a- m
2693. i- c
hi- /

2702.
2703.

2640.
(D

2642.
2641.
(Di
S
r
)

2649.
ol2658.

2650.
in2659.

2651.
nu
2660.

ox2667.
laj2676.

at
2668.
2677.

kok
2669.
chi
2678.

2685.
2694.

Chart X showing Inflection of Intransitives


2644.
2643.
2645.
Im
2646.
St
T/A
2647.
p
S
e
/
(Dir)
v
m
M
.
2653.
2654.
2655. 2656.
2652.
-k
in-ke
2663.
2664.
2665.
2661. 2662.
-q
at
2673.
2674.
2670. 2671.
2672.
2682. 2683.
2679. 2680.
2681.
oo
2691.
2692.
2688. 2689.
2690.
ex
2700. 2701.
2697. 2698.
2699.
e-

oo
2686.
ex
2695.
e-

2687.
2696.

2704.
2705. 2706.
T/
(Dir
)

2707.
Set
B

2714.
ki-

2715.
ol-

2716.
in-

2723.
x-

2724.
ox-

2725.
at-

2732.
na

2733.
laj-

2734.
-

2708.
Set
A
2717.
in- /
w
2726.
aa- /
a
a
w
2735.
x- / r-

Chart XI showing Inflection of Transitives


2711.
2709.
2710.
2712.
2713.
Imp
(Dir
Ste
T/A/
Set
v
)
m
M
B
.
2718.
num
2720.
2721.
2722.
2719.
e
-om
-k
in2727.
kok2736.
chin
a

71

2728.

2729.

2730.
-q

2731.
at-

2737.

2738.

2739.

2740.
-

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2741.
ta2750.
mi

2759.
chi

2742.

2743.
oo-

2751.

2752.
ex-

2760.

2761.
e- /
e
b

2744.
qa- /
q2753.
ee- /
e
er
2762.
ex- /
e
r

2745.

2746.

2747.

2748.

2749.
oo-

2754.

2755.

2756.

2757.

2758.
ex-

2763.

2764.

2765.

2766.

2767.
e-

2768.

2769.

72

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2770. Inflectional Affixes for Tense, Aspect, and Mood


2771.Verb inflection includes affixes which express notions of tense, aspect
and mood. These affixes do not express all of the possibilities of tense,
aspect and mood, some of which may be expressed by means of particles or
phrases outside of the verb word. This section will analyze these notions as
expressed in inflectional affixes, leaving for a later section the analysis of
these notions as expressed in other ways. The inflectional affixes, six
prefixes and two suffixes, do not clearly distinguish between tense, aspect
and mood, but rather express various combinations of these notions.
2772.
2773.The prefixes will be considered first with brief mention of the suffixes,
where necessary. Then the suffixes will be considered in detail.
2774.
2775. ta-: future definite.
2776.
The term future definite will be used to label this particular
prefix. The prefix indicates: (1) a non-immediate future time as opposed
to an action which will take place immediately; and/or (2) an aspect of
definiteness or certainty that the action will take place; and/or (3) an
attitude of intention or purpose on the part of the agent of the verb.
Examples:
2777.
2778.
tooxik Were going (in a little while) .
2779.
tatinb'oq
I'll call you (intention/certainty) .
2780.
2781.
Synthesis of ta- with pronominal affixes
2782.In most cases ta- is shortened to t- before those person prefixes which
begin with vowels. Special cases involve the following:
2783.

ta-+ (3rd singular, Set B) taa- (Coban); ta- + ix- (3rd singular, Set A)
tix-; ta- + w (1st singular, Set A, prevocalic) tinw- (Coban), which
represents both preconsonantal and prevocalic Set A together.
In addition, Chamil generalizes the vowel dropping of ta- to t- before Set A
prefixes which are not vowel-initial.
Before Set A prevocalic prefixes for third singular and first plural, the short
vowel of ta- lengthens to taa-.

2784.A possible explanation for this and for the irregular first singular
prevocalic is analogy pressure from the other three pronominal prefixes, taaw(2nd singular), teer- (2nd plural), and ter- (3rd plural), which are longer or
"heavier" due to their tVVC- or tVC- composition. Prefixes tar- (3rd singular)
and taq- (1st plural) become taar- and taaq- through analogy pressure. Prefix
ta- + w- (1st singular), unable to follow the same process because of the
second singular form, appropriates the first singular preconsonantal -in, places
it before the -w, and thus produces tinw-, which appears weighted in a manner
similar to tVC-, the third plural form.
2785.
73

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2786.Examples with verb stems xik go, boq call, and il see:
2787.

74

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2788.
2789.
2790.
2795.

ta- + Set B
1st singular
2

nd

singular

2800.

3rd singular

2805.

1st plural
nd

2810.

2815.

3rd plural

2820.
2825.

2826.

2831.

2836.

2841.

2846.

2851.
2856.

plural

2791.
ta- +
in2796.
ta- +
at2801.
ta- + 2806.
ta- +
oo2811.
ta- +
ex2816.
ta- +
e2821.

2792.
= tin-

2793.
nxik

ti

2797.
= tat-

2798.
txik

ta

2802.
= taa~ t-

2803.
ta
axik ~
txik

2807.
= too-

2808.
oxik

to

2812.
= tex-

2813.
xxik

te

2817.
= te-

2818.
xik

te

ta- + Set A preconsonantal (Set B patient is )


2827.
ta- +
in2828.
2829.
ti
1st singular
+= tinnboq
2832.
ta- +
aa2833.
2834.
ta
2nd singular
+= taaaboq

3rd singular

1st plural

2nd plural

rd

3 plural

2847.
ta- +
ee+2852.
ta- +
ex+2857.

2809.
w
ell go
2814.
y
all
will go
2819.
t
heyll
go

2822.
2823.

2837.
ta- + x+2842.
ta- +
qa+-

2794.
I
ll go
2799.
y
oull
go
2804.
h
ell
go/sh
ell go

2838.
= tix2843.
= taqa~
tqa-

2848.
= tee2853.
= tex-

2824.

2854.
te
xboq

2830.
ll call
him
2835.
oull
call
him
2840.
ell
call
him
2845.
ell
call
him
2850.
all
will
call
him
2855.
heyll
call
him

2859.

2860.

2839.
ti
xboq
2844.
ta
qaboq
~
tqaboq

2849.
te
eboq

2858.

75

I
y

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2861.

2862.

2867.

2872.

2877.

2882.

2887.

ta- + Set A prevocalic (Set B patient is )


2863.
ta- +
2864.
2865.
ti
w= tinwnwil ~
1st singular
+~ twtwil
2868.
ta- +
aaw2869.
2870.
ta
2nd singular
+= taawawil
2874.
=
2873.
ta(a)r
ta- + r-~
2875.
ta
3rd singular
+traril ~ tril
2879.
2878.
= taaq2880.
ta
ta- + q~
aqil ~
1st plural
+tqtqil
2883.
ta- +
eer+2888.
ta- +
er+-

2nd plural

3rd plural

2884.
= teer2889.
= ter-

2885.
eril
2890.
ril

te

te

2866.
I
ll see
him
2871.
y
oull
see
him
2876.
ell
see
him
2881.
ell
see
him
2886.
all
will
see
him
2891.
heyll
see
him

2892.
2893. nak- present habitual.
2894.
The label present habitual will be used to represent the notions of
tense, aspect, and mood inherent in the prefix nak-. This prefix indicates: (1)
actions which are true at, but not limited to, the present time in that these
actions have a quality of stability; and/or (2) actions which are habitual or
customary. Examples:
2895.
2896.
La
2906.
I

2897.
ink'
a

2907.
not

2917.

2918.

2927.
Ju

2928.
cha

2898.
chik

2899.
sa

2908.
any
m
o
r
e

2909.
cl

2919.

2929.
li

2900.
n-

2901.2902. 2903.
2904. 2905.
I don't
i
il
=
see well any
more.

2915. 2916.
2910.
2912.2913. 2914.
(prese
prese
I
s
nt and stable)
nt
/
2911.
habit
u
al
2922. 2923. 2924.
2925.
2926.
2920.
2921.

2930.
r-

2931.
na-

2932.2933. 2934.
2935.

r
=

76

2936.
He
only wears

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2937.
O

a
b
'i
l
2938.
nice

2939.
the

2940.
hi

2941.
prese
nt
/
2942.
habit
u
al

fine clothes.

2943.2944. 2945.
2946.2947.
i
h
u

2948.
2949.

77

(habitu
al)

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

2950.
Synthesis of nak- with pronominal affixes:
2951.Nak- is reduced to na- before all person prefixes marking first person
singular and third person singular, and before Set A prefixes marking first
person plural. Na- then is reduced to n- before first person singular prefix in-.
In addition the vowel in nak- may optionally copy the vowel quality of the
following prefix.
2952.
2953.
Dialect variation: nak + in- is reported by Haeserijn to be
unreduced nakin- in some dialects, though it has not been possible to
verify which ones. In addition, Freeze (personal communication) asserted
that nak- + in may be realized as nan- in Cahabon, a town in the eastern
zone and outside the scope of this study, but it has not been possible to
verify in which other areas this happens.
2954.
2955.
Examples with verb stems xik go, boq call, and il see.
2956.
2957.
2958.
1st
sin
gul
ar
2963.
2nd
sin
gul
ar
2968.
3rd
sin
gul
ar
2973.
1st
plu
ral
2978.
2nd
plu
ral
2983.
3rd
plu
ral
2988.

nak- + Set B

2993.
2994.
1st
sin

nak- + Set A preconsonantal (Set B is )


2995.
2996.
nin
2997.
ninboq ~
nak- +
-~
nakinboq ~
innakin- ~
nanboq

2959.
nak- +
in-

2960.
nin
-~
nakin- ~
nan-

2961.
ninxik ~
nakinxik ~ nanxik

2962.
go

2964.
nak- +
at-

2965.
kat-

na
2966.

nakatxik

2967.
y
ou go

2969.
nak- +
-

2970.
-

na
2971.

naxik

2972.
h
e goes

2974.
nak- +
oo-

2975.
na
koo- ~
nokoo-

2976.
nakooxik ~
nokooxik

2977.
e go

2979.
nak- +
ex-

2980.
na
kex- ~
nekex-

2981.
nakexxik ~
nekexxik

2982.
y
all go

2984.
nak- +
e-

2985.
na
ke- ~
neke-

2986.
nakexik ~
nekexik

2987.
th
ey go

2990.

2991.

2992.

2989.

78

2998.
I
call him

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

gul
ar
2999.
2nd
sin
gul
ar
3004.
3rd
sin
gul
ar
3009.
1st
plu
ral
3014.
2nd
plu
ral
3019.
3rd
plu
ral
3024.
3029.
3030.
1st
sin
gul
ar
3035.
2nd
sin
gul
ar
3040.
3rd
sin
gul
ar
3045.
1st
plu
ral
3050.
2nd
plu
ral

3055.

nan3000.
nak- +
aa-

3001.
ka-

na

3005.
nak- +
x-

3006.
x-

na

3010.
nak- +
qa-

3011.
qa-

na

3015.
nak- +
ee-

3016.
na
kee- ~
nekee-

3020.
nak- +
ex-

3002.

3007.

3012.

nakaboq

3003.
y
ou call
him

naxboq

3008.
h
e calls
him

naqaboq

3013.
w
e call
him

3017.
nakeeboq ~
nekeeboq

3018.
y
all call
him

3021.
na
kex- ~
nekex

3022.
nakexboq
~ nekexboq

3023.
th
ey call
him

3026.

3027.

3028.

3025.

nak- + Set A prevocalic (Set B is )


3031.
nak- +
w-

3032.
nin
w- ~
naw-

3033.
ninwil ~
nawil

3034.
I
see it

3036.
nak- +
aaw
-

3037.
na
kaw-

3038.

nakawil

3039.
y
ou see it

3041.
nak- +
r-

3042.
r-

na
3043.

naril

3044.
h
e sees it

3046.
nak- +
q-

3047.
q-

na
3048.

naqil

3049.
w
e see it

3051.
nak- +
eer-

3052.
na
keer- ~
nekeer-

3053.
nakeeril ~
nekeeril

3054.
y
all see it

3056.

3057.

3058.

3059.

na

79

nakeril ~

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3rd
plu
ral
3060.
3061.

nak+
er-

ker- ~
neker-

nekeril

80

hey see
it

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3062. x-: recent past.


3063.The prefix x- indicates (1) that the action took place no later than
yesterday, and (2) that the action was completed. This prefix is thus
perfective. The translations of the examples below thus contain both the
notion I played and also that of I have played, a case similar to the present
perfect in Latin: portavi I carried, I have carried.
3064.
3065.
Synthesis of x- with pronominal affixes
3066.
x- undergoes only one change which
either shortens the combination x- past + x- third singular, Set A to x- or
inserts -i- between them: xix-. Some speakers may optionally insert a vowel
-a- between x- and -qa.
3067.
3068.
Examples with verb stems b'atz'un play, sak hit, eek'a feel.
3069.
3070.

x- + Set B

3076.

2nd singular

3077.
x- + at-

3078.
= xat-

3079.
x
atbatz
un

3081.

3rd singular

3082.
x- + -

3083.
= x-

3084.
x
batzun

1st plural

3087.
x- +
oo-

3088.
= xoo-

3089.
x
oobatz
un

2nd plural

3092.
x- +
ex-

3093.
= xex-

3094.
x
exbatz
un

3097.
x- + e-

3098.
= xe-

3099.
x
ebatz
un

3075.
I
pl
ay
ed
3080.
you
pl
ay
ed
3085.
he
pl
ay
ed
3090.
we
pl
ay
ed
3095.
yall
pl
ay
ed
3100.
they
pl
ay
ed

3102.

3103.
3104.

3105.

3071.

3086.

3091.

3096.
3101.
3106.

1st singular

3rd plural

3072.
x- + in-

3073.
= xin-

3074.
xi
nbatzu
n

x- + Set A preconsonantal (Set B is )

81

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3117.

3rd singular

3122.

1st plural

3118.
x- + x+ 3123.
x- +
qa+ -

2nd plural

3128.
x- +
ee+ -

3129.
= xe-

3130.
x
esak

3rd plural

3133.
x- +
ex+ -

3134.
= xex-

3135.
x
exsak

3111.
I hit
hi
m
3116.
you
hit
hi
m
3121.
he hit
hi
m
3126.
we hit
hi
m
3131.
yall
hit
hi
m
3136.
they
hit
hi
m

3138.

3139.
3140.

3141.

3107.

3112.

3127.

3132.
3137.
3142.
3143.

3148.

3153.

3158.

1st singular

3108.
x- + in+ -

3109.
= xin-

3110.
xi
nsak

2nd singular

3113.
x- +
aa+ -

3114.
= xaa-

3115.
x
aasak

3119.
= x- ~
xix3124.
= xaqa~
xqa-

3120.
x
sak ~
xixsak
3125.
x
aqasak
~
xqasak

x- + Set A prevocalic (Set B patient is )


3144.
x- + w3145.
3146.
x
st
1 singular
+ = xw weeka
3149.
x- +
3150.
3151.
x
aaw=
aaweek
2nd singular
+ xaawa

3rd singular

1st plural

3163.

2nd plural

3168.

3rd plural

3154.
x- + r+ 3159.
x- + q+ 3164.
x- +
eer+ 3169.
x- +

82

3155.
= xr-

3156.
x
reeka

3160.
= xq-

3161.
x
qeeka

3165.
= xeer3170.
= xer-

3166.
x
eereek
a
3171.
x
ereeka

3147.
I felt
it
3152.
you
fel
t it
3157.
he
fel
t it
3162.
we
fel
t it
3167.
yall
fel
t it
3172.
they

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

er+ -

fel
t it

3173.
3174.

83

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3175. ki- : narrative past.


3176.The prefix ki- indicates past time in complementwith the previously
mentioned prefix x-. Ki- is used to indicate: (1) action which took place in the
more remote past, and (2) action that is completed or perfective. To mark
remote past in Carcha and Chamil, the prefix oh- is used instead of ki-,
although the ki- prefix is known.
3177.
3178.Synthesis of ki- with pronominal affixes
3179.This prefix loses the vowel before person prefixes beginning with
vowels. It undergoes no change before consonants or . In some towns this
prefix appears to vary with ; in Coban it may optionally vary with before
-qa.
3180.
3181.Examples with verb stems lub get tired, loq buy, abi hear
3182.
3183.
3184.
st

ki- + Set B
1

singula
r
3189.
2

3185.
in-

ki- +

3186.
kin-

3187.
ki
nlub

3188.
I
got
tired

singula
r
3194.
3

3190.
at-

ki- +

3191.
kat-

3192.
k
atlub

3193.
y
ou got
tired

3195.
-

ki- +

3196.
ki-

3197.
lub

nd

rd

singula
r
3199.
1
st
plural
3204.
nd

3200.
oo-

ki- +

3201.
koo-

ki

3202.
k
oolub

plural

3205.
ex-

ki- +

3206.
kex-

3207.
k
exlub

3209.
3
3210.
ki- +
3211.
=
3212.
k
rd
plural
ekeelub
3214.
3215.
3216.
3217.
3219.
ki- + Set A preconsonantal (Set B is )
3220.
1

3198.
e got
tired
3203.
e got
tired
3208.
all
got
tired
3213.
hey
got
tired

h
w
y

3218.

3224.

st

singula
r
3225.
2

3221.
ki- +
in- + -

3222.
kin-

3223.
ki
nloq

3226.
ki- +
aa- + -

3227.
ka-

3228.
k
aloq

nd

singula
r

84

bough
t it
3229.
y
ou
bough
t it

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3230.
rd

singula
r
3235.
1
st
plural
3240.
nd

3231.
ki- +
x- + -

3232.
kix-

3233.
ki
xloq

3236.
ki- +
qa- + -

3237.
=
kiqa- ~
qa-

3238.
ki
qaloq ~
qaloq

3241.
ki- +
ee- + -

3242.
kee-

3243.
k
eeloq

plural

3245.
3
3246.
ki- +
3247.
=
3248.
k
rd
plural
ex- + kexexloq
3250.
3251.
3252.
3253.
3255.
x- + Set A prevocalic (Set B patient is )
3256.
1
st

singula
r
3261.
2

3257.
ki- +
w- + -

3258.
kiw-

3259.
ki
wabi

singula
r
3266.
3

3262.
ki- +
aaw- + -

3263.
=
kaw-

3264.
k
awabi

3267.
ki- +
r- + -

3268.
kir-

3269.
rabi

3272.
ki- +
q- + -

3273.
kiq-

3274.
ki
qabi

3277.
ki- +
eer- + -

3278.
ker-

plural

3279.
k
erabi

3281.
3
rd
plural

3282.
ki- +
er- + -

3283.
ker-

3284.
k
erabi

nd

rd

singula
r
3271.
1
st
plural
3276.
nd

ki

3286.
3287.

85

3234.
h
e
bough
t it
3239.
w
e
bough
t it
3244.
y
all
bough
t it
3249.
t
hey
bough
t it
3254.

3260.
I
heard
it
3265.
y
ou
heard
it
3270.
h
e
heard
it
3275.
w
e
heard
it
3280.
y
all
heard
it
3285.
t
hey
heard
it

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3288. chi- ~ : optative / imperative; mi-: negative optative /


imperative
3289.
3290.The combination of tense, aspect, and mood, labeled here
optative/imperative is the most complex T/A/M inflection. In terms of time this
inflection indicates action which is just about to occur or which begins at the
moment of speaking, thus dividing future time with the prefix ta-. This
inflection also may indicate an immediate desire on the part of the speaker
mixed with an element of doubt that the desire will be fulfilled, and in this
sense may be said to be optative or exhortative. Finally, this inflection may be
used to mark imperative, either alone or in conjunction with a special
imperative suffix.
3291.
3292.The optative/imperative is further complicated by having two prefixed
allomorphs, chi- and , and by having a separate prefix mi- to indicate the
inflection in the negative. All other inflections use separate negative words
that are outside the verb word.
3293.
3294.Charts similar to those for other inflections are presented below for the
chi- allomorph of the optative/ imperative and for mi-, the negative
optative/imperative, to show how these forms combine with the pronominal
affixes. The prefixes will be presented without example words, which will be
given in a later chart showing the interplay between the two allomorphs chiand .
3295.
3296.

86

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3297.
Synthesis of chi- with pronominal affixes
3298.In all cases these prefixes lose the i- (chi ch) before person prefixes
beginning with vowels. It undergoes no change before consonants or ,
except for chi- optionally becoming cha- before Set A -qa. The prevocalic
doubled Set A, -in-w, was mentioned for other inflections. Examples:
3299.
3300.
3306.
1st
sin
gu
lar
3312.
2nd
sin
gu
lar
3318.
3rd
sin
gu
lar
3324.
1st
pl
ur
al
3330.
2nd
pl
ur
al
3336.
3rd
pl
ur
al
3342.

3301.
c
hi- +
Set B

3303.

3304.
mi
- + Set
B

3305.

3308.

3309.
=
chin-

3310.
mi
- + in-

3311.
=
min-

3314.

3315.
=
chat-

3316.
mi
- + at-

3317.
=
mat-

3320.

3321.
chi-

3322.
mi
- + -

3323.
mi-

3326.

3327.
=
choo-

3328.
mi
- + oo-

3329.
=
moo-

3332.

3333.
=
chex-

3334.
mi
- + ex-

3335.
=
mex-

3337.
c
hi- +
e-

3338.

3339.
=
che-

3340.
mi
- + e-

3341.
=
me-

3343.

3344.

3345.

3346.

3347.

3352.
3357.
=
chin-

3353.
mi- + Set A
preconsonantal
3358.
mi
3359.
=
- + inmin+ -

3307.
c
hi- +
in3313.
c
hi- +
at3319.
c
hi- +
3325.
c
hi- +
oo3331.
c
hi- +
ex-

3349.
c
hi- +

3348.
3354.
1st
sin
gu

Set A
preco
nsona
ntal
3355.
c
hi- +
in- +

3302.

3350.
Se
3351.
is
3356.
-

87

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

lar
3360.
2nd
sin
gu
lar
3366.
3rd
sin
gu
lar
3372.
1st
pl
ur
al
3378.
2nd
pl
ur
al
3384.
3rd
pl
ur
al
3390.

3361.
c
hi- +
aa- +

3362.
-

3363.
cha-

3364.
mi
- + aa+ -

3365.
=
ma-

3367.
c
hi- +
x- +

3368.
-

3369.
=
chix-

3370.
mi
- + x+ -

3371.
=
mix-

3373.
c
hi- +
qa- +

3374.
-

3375.
=
chiqa- ~
chaqa-

3376.
mi
- + qa+ -

3377.
=
miqa-

3379.
c
hi- +
ee- +

3380.
-

3381.
che-

3382.
mi
- + ee+ -

3383.
=
me-

3385.
c
hi- +
ex- +

3386.
-

3387.
=
chex-

3388.
mi
- + ex+ -

3389.
=
mex-

3391.

3392.

3393.

3394.

3395.

3399.

3400.
m- + Set A
prevocalic

3404.
=
chiw- ~
chinw-

3405.
mi
- + w+ -

3409.
-

3410.
=
chaw-

3411.
mi
- +
aaw- +
-

3415.
3421.
-

3416.
=
chir3422.
=
chiq-

3397.
c
hi- +
3396.
3401.
1st
sin
gu
lar
3407.
2nd
sin
gu
lar
3413.
3rd
sin
gu
lar
3419.
1st
pl

Set A
prevo
calic
3402.
c
hi- +
w- +
3408.
c
hi- +
aaw+
3414.
c
hi- +
r- +
3420.
c
hi- +
q- +

3398.
Se

3403.
-

88

3417.
mi
- + r- +
3423.
mi
- + q+

3406.
=
miw~
minw3412.

maw-

3418.
=
mir3424.
=
miq-

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

ur
al
3425.
2nd
pl
ur
al
3431.
3rd
pl
ur
al

3426.
c
hi- +
eer- +
3432.
c
hi- +
er- +

3427.
-

3433.
-

3428.
=
cher-

3429.
mi
- + eer+ -

3434.
=
cher-

3435.
mi
- + er+ -

3437.
3438.
3439.
3440.
3441.

89

3430.
=
mer3436.
mer-

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3442. Inflectional suffixes for Tense, Aspect, and Mood


3443.Before continuing the discussion of the optative/ imperative inflection it
will be necessary to introduce the two T/A/M inflectional suffixes.
3444.
3445.
-(a)q:
3447.
-k:

3446.
3448.

future
non-future

3449.
3450.
These suffixes function to divide the
time of verb action into two segments: future time and non-future time,
including the present and the past. These suffixes in some instances must cooccur with the T/A/M prefixes presented above; in other cases these suffixes
are optional. In none of the examples given for the above prefixes, for
instance, were these suffixes required. Rules governing the occurrence of
these suffixes are given below.
3451.
3452.
Non-future -k is obligatory with certain intransitive stems, optional
with others, and is obligatory with stative participles and positional
adjectives in stative sentences. Future suffix -(a)q is obligatory with all
non-negative future expressions with intransitive stems, except some
imperatives. It is obligatory with all non-negative optative/imperatives of
transitive stems when marked by the allomorph; when the
optative/imperative is marked by allomorph chi-, future suffix -(a)q is
optional. Finally, stative sentences obligatorily mark future with suffix (a)q. Distribution of the allomorphs, -q and -aq, will be shown below, with
accompanying examples.
3453.
3454. non-future -k.
3455.
Non-future is obligatory with intransitive stems of class v.i.k.
and with viV passives derived from vtD.
3456.
3457.
e-

se

3464.
ch
iilaa3471.
3478.

na

3485.
pr
esent
habitua
l

3458.

3459.

3460.

3465.

3466.

3467.

3472.

3473.

3474.

3479.
3486.
Set B

3480.
see3487.
stem

3481.
-k
3488.
nonfut
ur

90

3462.
(bound
ste
m)
3461.
3469.
(bound
ste
m)
3468.
3475.

3476.

3483.
3482. nasee
=
k
3489. 3490.

3463.
(to)
laug
h
3470.
(to) be
scold
ed
3477.

3484.
he
laug
hs
3491.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

e
3493.

3494.

3495.

3492.

3499.

x-

3506.
re
cent
past

3500.
in-

3501.
see-

3507.
Set B

3508.
stem

3502.
-k
3509.
nonfut
ur
e

3514.

3515.

3516.

3513.

3520.

3497.
3496.

3504.
3503. xinsee
=
k

3510.

3527.
re
cent
past

3521.
in-

3522.
chiila
a-

3528.
Set B

3529.
stem

3523.
-k
3530.
nonfut
ur
e

3512.

3518.

3519.

3534.
3535.

91

3525.
3524. xinchiil
=
aak

3531.

3505.
I
laug
hed

3511.

3517.

x-

3498.

3532.

3526.
I was
scold
ed

3533.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3536.Non-future is optional with intransitive stems of classes viR, vin, viV,


vi, vir, and vib, except in Coban where non-future -k may not occur with
these stems.
3537.
3538.
r

wa

3545.

3539.

3540.

3541.

3546.

3547.

3548.

3542.

3543.
(bound
ste
m)

3544.
(to)
slee
p

3550.

3551.

3549.

3552.

na
-

3559.
pr
esent
habitual

3553.

3554.
war-

3560.
Set B

3561.
stem

3555.
(k)
3562.
nonfut
ure

3567.

3568.

3569.

3566.

3573.

3557.
3556. nawar(
=
k)

3563.

3558.
he
slee
ps

3564.

3565.

3571.

3572.

3570.

x-

3580.
re
cent
past

3574.
in-

3575.
war-

3581.
Set B

3582.
stem

3576.
(k)
3583.
nonfut
ure

3578.
3577. xinwar(
=
k)

3584.

3585.

3579.
I slept

3586.

3587.
3588.Non-future -k is obligatory in those stative sentences in which the
predicate consists of verbal or positional adjectives whose derivation is
root C1VC plus suffix -C1oh, or -C1uh if the root vowel is -u-, and in which
the topic or theme of the sentence is not third person singular.
3589.
3590.
y
ok3595.

3591.
3596.

3592.
3597.

3602.
y
ok3609.
p
osition
al root

3603.
yo
h3610.
de
rivation
al suffix

3604.
-
3611.
Set
B

3606.3607.
3605.
=
yokyo

3616.

3617.

3618.

3619.
3620. 3621.

3623.
y
ok3630.
p
osition
al root

3624.

yo
o-

3631.
de
rivation
al suffix

3593.
(posi
tional root)

3594.
lying
down

3598.
3599. 3600.

3601.

3612.
3613. 3614.
3628.
3626.
yokyo
3627.
oki
=
n
3
635.
3633.
3634.
S

3625.
k3632.
nonfu
tu

92

3608.
he/sh
e/it is lying
down
3615.
3622.

3629.
I am
lying down
3636.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

re

3637.
3638. Future (a)q:
3639.Future prefix ta- with intransitive stems requires the presence of the
-q allomorph of this suffix.
3640.
3641.
3642.
ar-

3649.

3656.

3643.

3644.

3645.

3650.

3651.

3652.

3647.
(bound
ste
m)
3646.
3654.

3653.

t-

3663.
fu
ture

3657.
in-

3658.
war-

3664.
Set B

3665.
stem

3659.
q
3666.
futur
e

3670.
3671.
3672.

93

3660. 3661.
=
tinwarq
3667.

3668.

3648.
(to)
sle
ep
3655.

3662.
I'll
sle
ep
3669.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3673.Future in stative sentences is marked by this suffix, with the following


distribution: -q is found with the previously-mentioned verbal and positional adjectives of the form C1VC-C1oh, while -aq is found with all other stative
predicates.
3674.
3675.
yok3682.
position
al
root

3677. 3678.
3679.3680.
y
q
=
okyooq

3681.
hell
be lying
down

3683.
de
rivation
al suffix

3684. 3685.
fu
S
3686.3687.

3688.

3690.

3691.

3695.

3676.

yo
h-

3689.

3696.
yok3703.
position
al
root

3697.

yo
h-

3699.
3701.
y
3698. 3700. okyooq
q=
in
3705. 3706.
fu
S
3707.3708.

3709.

3711.

3712.

3716.

3718.

3719.

3722.
(
noun
stem)
3720.3721.

3725.

3726.

3727.3728. 3729.

3713.3714. 3715.

3731.

3732.

3738.

3739.
st
em

3733.
a
3734.
3735.3736.
w

=
inqaq
3740. 3741.
fu
S
3742.3743.

3746.

3747.

wi
nq-

3745.

3748.3749. 3750.

3752.

3753.

3759.

3760.
st
em

3754. 3755.
3757.
w
a
a
3756. inqaqa
=
t
3761. 3762.
fu
S
3763.3764.

3767.

3768.

3774.

3775.

3778.
(
adjecti
ve
stem)
3776.3777.

3781.

3782.

3783.3784. 3785.

wi
nq-

3766.

3773.
chaab'il
3780.
3787.

3702.
Ill be
lying down

3704.
de
rivation
al suffix

3710.

3717.
winq3724.

3692.3693. 3694.

3788.
ch
aab'il-

3769.3770. 3771.

3789. 3790.
3791.3792.
c
a

=
haab'il

94

3723.

man

3730.

3737.
hell
be a man
3744.
3751.

3758.
youll
be a man
3765.
3772.

3779.
3786.

good

3793.
he'll
(itll) be

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3794.

3795.
st
em

aq
3796. 3797.
fu
S
3798.3799.

3800.

3802.

3803.

3807.

3801.

3808.

3809.
ch
aab'il-

3815.

3816.
st
em

3804.3805. 3806.

3810. 3811.
3813.
c
a
a
3812. haab'il
=
aqat
3817. 3818.
fu
S
3819.3820.

good

3814.
you'll
be good
3821.

3822.
3823.
3824. Future suffix -(a)q is used with the optative/ imperative inflection, chi- ~ .
The allomorph -q occurs with intransitive stems; the allomorph -aq occurs with
transitive stems, with one small exception to be noted below.
3825.
3826.
3827.

3828.

95

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3829. Interplay of Future Suffix -(a)q with Optative


3830.It was necessary to interrupt the presentation of the optative inflection
to present the suffixes for future and non-future because the presence of one
or the other of the two optative allomorphs, chi- and , determines the
presence of the future suffix -(a)q as well as the placement of the pronominal
affixes. If the chi- allomorph is used to mark optative, future suffix -(a)q is
obligatory with intransitive stems and optional with transitive stems. If is
used to mark optative, the future suffix is obligatory with both transitives and
intransitives. The different allomorphs for optative also determine the
placement of the Set B pronominal affixes. If chi- is used, Set B immediately
follows it in the normal pattern (i.e., the pattern used in the other inflections).
If is used, requiring the use of future suffix -(a)q, Set B will follow - (a)q.
3831.
3832.
The charts following exemplify the formation. Included in
the charts is the imperative, which will be explained more fully in
the following section.
3833.
3834.
3835.
Subje
ct
(1)

Chart XII: Intransitive stems with optative imperative allomorphs, chiand .


Stem is war sleep.
3836.
chi~

3837.
Set B

3854.
chi-

3855.
-in-

3863.

3864.

3872.
chi-

3873.
-at-

3881.

3882.

3890.

3891.

3898.
3rd
sin
g.

3899.
chi-

3900.

3908.

3909.

3916.
1st
pl.

3917.
chi-

3918.
-oo-

3926.

3927.

3935.
chi3944.

3936.
-ex3945.

3853.
1st
sin
g.

3871.
2nd
sin
g.
(2)

3934.
2ns
pl.

3838. 3839.
3840.3841.
S
I
S

3842.
J
oined
togethe
r

3856.
3857.3858. 3859. 3860.
c
-q
hinwarq
3865.
3867. 3868.
3866. 3869.
w
arqin
3874.
3876.
3878.
c
3875.-q
3877.
hatwarq
3883.
3885. 3886.
3887.
w
3884.
arqat
3892. 3893.
3894. 3895. 3896.
w
arin
3901.
3903.
3905.
c
3902.-q
3904.
hiwarq
3910.
3911.3912. 3913. 3914.
w
-q

arq
3919.
3921.
3923.
c
3920.-q
3922.
hoowarq
3928.
3930. 3931.
3929. 3932.
w
arqo
3937.
3939.
3941.
c
3938.-q
3940.
hexwarq
3946. 3947.3948. 3949. 3950.
w

96

3843.
Meaning
3861.
may I
sleep
3870.
3879.
may you
sleep
3888.
3897.
sleep!
3906.
may he
sleep
3915.
3924.
may we
sleep
3933.
3942.
may you
sleep
3951.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

(3)

3961.
3rd
pl.

3953.

3954.

3962.
chi-

3963.
-e'-

3971.

3972.

arqex
3955. 3956.3957. 3958.
3959.
w
arinqex
3964.
3966.
3968.
c
3965.-q
3967.
he'warq
3973.
3975. 3976.
3974. 3977.
w
arqeb'

3960.
sleep, y'all
3969.
may they
sleep
3978.

3979.
3980.
3981.
Notes - Chart XII
3982.
(1)Chart XII is arranged like Chart XIII minus the Set A pronominal affixes.
(2)Second person singular subject: Intransitives which have an imperative -Vn
in addition to the optative with chi- and generally use this -(V)in suffix as
the normal imperative.
3983. Second person plural subject: The -(V)n imperative in the singular is
also used in the plural, followed by the future suffix -q and the Set B
marker for second person plural -ex.

97

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

3984.

Chart XIII: Transtive stems with optative/imperative allomorphs, chi- and

with stems il see, b'oq call, ch'iila scold, and ab'i hear
3985. 3986. 3987.
3988. 3989. 3990.3991.3992.3993.
A
c
S
3994.
Joined
S
S
I
S
S
together
4008. 4009. 4010. 4011.
4013.
c
4012. (
4014. 4015. 4016.
chatwil
~ chatwilaq
4019. 4020. 4021. 4022.
4024.
4023. (
4025. 4026. 4027.
chatinb'o
q ~ chatinb'oqaq

4007. c
1

4032. 4033.
4035. 4036.
4030. 4031. w
4034. 4037.

4038.
4043. 4044.
4046. 4047.
4041. 4042. i
4045. 4048. 4049.

at

qilaqat
inb'oqaq

4052. 4053. 4054. 4055.


4057.
c
4056. (
4058. 4059. 4060.
chinaqil
~ chinawilaq

4051.
4063. 4064. 4065. 4066.
2

4068.
4067. (
4069. 4070. 4071.
chinab'o
q ~ chinab'oqaq
4077.
4080.
4074. 4075. 4076. 4078. 4079. 4081.

4082.
ilin
4088.
4091.
4085. 4086. 4087. 4089. 4090. 4092.

4093.
b'oqin
c

4096. 4097. 4098. 4099.


4101.
c
4100. (
4102. 4103. 4104.
chinril ~
chinrilaq
4107. 4108. 4109. 4110.
4112.
4111. (
4113. 4114. 4115.
chinxb'o
q ~ chinxb"oqaq

4095. c
3

4120. 4121.
4123. 4124.
4118. 4119. r
4122. 4125.

4126.
4131. 4132.
4134. 4135.
4129. 4130. x
4133. 4136. 4137.

4139.
4140. 4141. 4142. 4143.
1

rilaqin
xb'oqaqi

4145.
4144. (
4146. 4147. 4148.
chatqil ~
chatqilaq
4151. 4152. 4153. 4154.
4156.
4159.
chatgab'
c
4155. (
4157. 4158.
oq ~
chatqab'oqaq
4162. 4163. 4164. 4165. 4166. 4167. 4168. 4169. 4170.
qilaqat
c

98

3995.
Meanin
g
4017.
let me
see
you
4028.
let me
call
you
4039.
let me
see
you
4050.
let me
call
you
4061.
may you
see
me
4072.
may you
call
me
4083.
look at
me!
4094.
call me!
4105.
let him
see
me
4116.
let him
call
me
4127.
let him
see
me
4138.
let him
call
me
4149.
let's look
at
you
4160.
let's call
you
4171.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4175. 4176.
4178. 4179.
4173. 4174. q
4177. 4180. 4181.

at

qab'oqaq

4184. 4185. 4186. 4187.


4189.
c
4188. (
4190. 4191. 4192.
chineril
~ chinerilaq

4183. 4195. 4196. 4197. 4198.


4200.
2
c
4199. (
4201. 4202. 4203.

chineb'o
q ~ chineb'oqaq

4209.
4206. 4207. 4208.
4220.
4217. 4218. 4219.

4210. 4211. 4212.


o
4213.
4214.
4221. 4222. 4223.
o
4224. 4225.
gin

ilpmaqin
b'oqoma

4228. 4229. 4230. 4231.


4233.
c
4232. (
4234. 4235. 4236.
chine'ril
~ chine'rilaq
4239. 4240. 4241. 4242.
4244.
4247.
chine'xb
4243. (
4245. 4246.
oq ~
chine'xb'oqaq

4227. c
3

4252. 4253.
4255. 4256. 4257.
4250. 4251. r
4254. e
4258.

"

rilaqineb

4263. 4264.
4266. 4267. 4268.
4261. 4262. x
4265. e
4269.

neb'

xb'oqaqi

4274. 4275.
4277.
4272. 4273. c
4276. 4278. 4279. 4280.

aq

inch'iila'

4285. 4286.
4288. 4289.
4271.
4283. 4284. c
4287. 4290. 4291.
1

4315.
4316.
4317.

inch'iilaq

at

4296. 4297.
4299.
4294. 4295. w
a
4298. 4300. 4301.

4302.

wab'l'aq

4307. 4308.
4310. 4311.
4305. 4306. w
a
4309. 4312.

4313.

wab'iqat

let's see
you
4182.
let's call
you
4193.
may y'all
see
me
4204.
may y'all
call
me
4215.
look at
me
y'all
4226.
call me
y'all
4237.
let them
see
me
4248.
let them
call
me
4259.
let them
see
me
4270.
let them
call
me
4281.
let me
scold
him
4292.
let me
scold
you
4303.
let me
hear
him
4314.
let me
hear
you

Notes - Chart XIII

(1) Chart XIII is arranged to show the inflection with each of the six possible
Set A agents, which are identified in column one and then reproduced
as Set A in the fourth column, labeled "A", and in one other case, split
between column four and column eight.

99

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

(2) The second column contains either allomorph chi- or allomorph , both
with a vowel-initial stem, il see, and with a consonant initial stem, boq
call, found in the fifth column.
(3) The third and seventh columns contain Set B patients, depending upon
which of the optative/imperative allomorphs is used, chi- or .
(4) Agent second person singular: These forms are imperatives and,
when not inflected with the chi- allomorph, are unmarked for
agent.
(5) Agent second person plural: These forms are imperatives and, when
not inflected with the chi- allomorph, are unmarked for agent, and
have a suffix -om used only in plural imperative.
(6) The last section at the bottom contains some forms showing how
verb stems of class vtD differ in combining with suffix -aq. If -aq is
word final, a glottal stop is inserted between the stem vowel and
-aq: in-chilla-aq becomes inchiilaaq, w-abi-aq becomes wabiaq.
If -aq is followed by a Set B patient, -aq is shortened to -q: in-chiilaaq-at becomes inchiilaqat, w-abi-aq-at becomes wabiqat.

4318. Imperatives

4319.
In Charts XII and XIII in the previous section the imperative suffixes
were shown in relation to the optative construction. The formation of
imperatives is detailed in the following sections.
4320.
4321.
Transitive imperatives.
4322.
The second singular imperative of transitive stems is formed by the
transitive stem by itself plus the Set B patient.
4323.
4324.
il

4325.
+

4331.
look

4332.

4338.

4339.

4345.
il

4346.
+

4352.
look

4353.

4326.

4333.
Set B
hi
m

4328.
4327. =

4334. 4335.

4336.
4343.

4340.

4341. 4342.

4347.
-in
4354.
Set B
m
e

4349.
4348. =

4355. 4356.

4329.
il

4350.
ilin

4357.

4330.
look
at him!

4337.
4344.

4351.
look
at me!

4358.

4359.
4360.
The second plural imperative is formed by the stem plus
imperative suffix -om plus future suffix -aq plus the Set B patient.
4361.
4362.
il

4363.
+ -om

4369.
look

4370.
imperat
ive

4376.

4377.

4364.
+

4365.

4371. 4372.
fu
Set
B
4378. 4379.

100

4367.
4366.iloma
=
q

4368.
look at
him, yall

4373. 4374.

4375.

4380. 4381.

4382.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4383.
il

4384.
+ -om

4385.
+

4390.
look

4391.
imperat
ive

4386.
+ -in
4392. 4393.
fu
Set
B
m
e

4388.
iloma
4387.
qi
=
n

4389.
loo
k at me
y'all

4394.4395.

4396.

4397.
4398.
4399.
Intransitive imperatives.
4400.
The second singular imperative of intransitive stems is formed by
the suffix -(V)n or, occasionally, by the optative construction + stem + q
+ at. Of the few viR root stems, some form the imperatives with -(V)n and
others with the optative. Suffix -(V)n is always -Vn with these C1VC2 roots
except if C2 was originally -h. In this case the shape of the root is C1VV and
the form of the imperative suffix is -n. Where the imperative suffix is -Vn,
the quality of the vowel is not predictable.
4401.
4402.
4403.
war
4411.
sleep
4419.
b'ee
4427.
walk
4435.

viR imperatives with -(V)n:


4404.
4405.
in
4406.
+
4407.
4413.
imper
4412.
ative
4414. 4415.
4420.
4421.
n
4422.
+
4423.
4429.
impera
4430.
4428.
tive
4431.
4436.
4437.

4438.
4439.

4443.
viR imperatives with -aq (optative):
4444.
4445.
4447.
4446.
q
4448.
at
kam
+
+
4452.
4456.
Set
4454.
future
die
4453.
4455.
B you
4460.
4461.
4463.
4462.
q
4464.
at
b'ay
+
+
4468.
4472.
Set
be late
4469.
4470.
future 4471.
B you

4408.
4409.
w
=
arin

4410.
s
leep!

4416.
4417.

4418.

4424.
4425.
b
=
'een

4426.
alk!

4432.
4433.

4434.

4440.
4441.

4442.

4449.
4450.
k
=
amqat

4451.
ie !

4457.
4458.
4465.
4466.
b
=
'ayqat

4459.

4473.
4474.

4475.

4467.
b
e late!

4476.
4477.
Imperatives of vin stems appear to be the same as the stems
themselves. It is possible to imagine that -n as an imperative suffix is
added to these stems, all of which end in -n, and that the two n's are then
reduced back to just -n.
4478.
4479.
4482.

atin
bathe

4480.
(vin stem and
imperative)
4483.

101

4481.
4484. 4485.

take a bath!
4486.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4487.

4492.

4488.

b'atz'un

4493.
(vin stem and
imperative)

4489. 4490.

4491.

4494.

play!

4495.
play
4496.
4497. 4498.
4500.
4501. Imperatives of vik stems are all formed by the -n allomorph of the (V)n suffix added directly to the stem.
4502.
4503.

4509.

see-

laugh

4504.
+

4505.

4507.

-n
4506.
=

4511.
4510.

imperativ
e

4512.4513.

4515.

4516. 4517.

4518. 4519.

4521.

4522.
+

4525.
4524.
=

k'ale-

4527.
slash brush
for a field

4523.

4528.

-n

imperativ
e

102

see
n

4529.4530.

4499.

4508.
lau
g
h
!
4514.
4520.

k'ale
n

4526.
slas
h
!
4531.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4532.
4533.
Stems of class vi may form imperatives by means of either one or
the other of the suffixes depending on the stem. Those stems which use (V)n form the imperative by dropping the stem final glottal stop and then
adding the -n allomorph. Those stems which use the optative construction
add the -qat suffix directly to the unaltered stem.
4534.

4535.
n
ume

4536.
nu
m
e

4537.

4539.
=

4540.
numen

+
4538.
4542.
p
ass by
4549.
(s
tem)
4556.

4563.
k
aqo +

4570.
g
et red,
be
envious

4543.

4544.
imper
ative

4545.

4546.

4550.

4551.

4552.

4553.

4557.

4558.

4559.

4560.

4564.
q

4565.

4571.
fut
u
r
e

4566.
at

4567.
=

4541.
pass
by!

4547.

4548.

4554.

4555.

4561.

4562.

4568.
kaqoqat

4569.
be
envi
ous!

4575.

4576.

4573.
Set
B
y
o
u

4572.

4574.

4577.
(s
4582.
4583.
tem)
4578. 4579.
4580.
4581.
4584.
------------4585. Stems of class viV may form either one or the other imperative
depending on the stem. Those which use the -(V)n suffix shorten the long
stem vowel before adding the -n allomorph. Those with which the optative is
used do not shorten the stem vowel but simply add the -qat suffix to the
stem.
4586.
4587.
ku
b'ee- +

4588. 4589.

4594.
g
o down

4595.

4601.

4608.
ka
laa +

4590. 4591.
=

4602.

4609.
q

4596.
i
mpera
tive

4597.

4598.

4603.

4604.

4605.

4610.

4611.

103

4612.
=

4592.
kub'en

4593.
go
dow
n!

4599.

4600.

4606.

4607.

4613.
kalaaqat

4614.
get

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

dru
nk!

+
4615.
g
et drunk

4616.
fut
u
r
e

4617.
S
et B
you

4618.

4620.

4621.

4619.

4622.
4623.All other intransitive stems employ the optative construction to form
imperatives. The exception is the vit class, which has no imperative form.

4624. Inflection and Affixes for Direction and Manner


4625.The description of inflection given above contains all of the obligatory
elements found in the verb word: pronominal affixes, T/A/M affixes, and verb
stems. With the following description of directional and manner affixes, all of
the possible elements found in the verb word will be completely described.
There are seven direction and manner affixes: six prefixes and one suffix.
4626.
4627.
ol-/-ul-:
4629.
4630.
4631.
4632.
4634.
hither
4628.
4633.
4635.
Prefix -ol- occurs between the T/A/M prefix and the pronominal prefixes
and means that the agent comes from elsewhere to do the action denoted in
the verb. Prefix -ul occurs in some dialects, but -ol predominates in the
western dialect area.
4637.

4636.
4645.
x
4646.
ol 4655.4656.
p
directio
nal

4638.

4639.

4647. 4648.
inx4657. 4658.
S
Set
A
a
g
e
nt

4665.
4664.

4666.

4667.

4640.

4649.
b'oq
4659.
verb

4641.
4642. 4643.

4652.
x
4651.
olinxb'
4650.
=
oq
4660.

st
e
m

4644.

4653.
He
cam
e
4654.
to call
me

4661. 4662.

4668.

4669.

4670.4671.

4672.

4673.
-ox/-ux: thither
4674.
Prefix -ox- occurs between the T/A/M prefix and the pronominal prefixes
and means that the agent goes from here to elsewhere to do the action. Prefix
-ox- is thus effectively the opposite of -ol-. The Chamil dialect has the -olprefix but lacks the -ox- prefix.
4676.

4675.
4684. 4685.
x
ox -

4677.

4678.

4679.

4680.
4681. 4682.

4686.
in-

4687.
x-

4688.
b'oq

4689.
4690.4691.
=
xoxinx
b'oq

104

4683.

4692.
He
wen

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4694.4695.
p
directio
nal

4696. 4697.
S
Set
A
a
g
e
nt

4698.
verb
st
e
m

4700. 4701.

4704.
4703.

4705.

4712.

4706.

4699.

4707.

t
4693.
to call
me

4711.

4708.
4709. 4710.

-nume: passing by

4713.
Prefix -nume- occurs between the pronominal prefixes and the verb
stem and means that the agent does the action while passing by.
4715.

4714.
4723. 4724.
x
in -

4716.

4717.

4718.

4719.
4720. 4721.

4725.
x-

4726.
num
e
-

4727.
b'oq

4728.
4729.4730.
x
=
inxnu
meb'o
q

4733.4734.
p
Set B
pati
ent

4735. 4736.
S
direc
ti
o
n
al

4737.
verb

4738.

st
e
m

4739. 4740.

4744.

4746.

4747.
4748. 4749.

4742.

4743.

4745.

4751.
Prefix -nume- may also co-occur with
either -ox- or -ol-.
4755.

4756.

4764.
x
4765.
ol 4773.4774.
p
directio
nal
ther
e to
here

4757.

4758.

4766.
in

4767.
x4775. 4776.
S
Set
A
a
g
e
nt

4752.4753.

4759.

4760.
4761. 4762.

4768.
num
e
4777.
direc
ti
o
n
al
p
a
ss
in
g
b
y

4771.
x
4769.
olinxnu
b 4770.
meb'o
=
q
4778.
v

105

4779.4780.

4722.

4731.
Passing
by
4732.
he
calle
d
me

4741.
4750.
4754.
4763.

4772.
He
cam
e
and
calle
d
me,
pass
ing
by

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4783.

4790.

4782.

4784.

4791.
4792.
x
ox -

4793.
in

4800.4801.
p
directio
nal
here
to
ther
e

4785.

4794.
x4802. 4803.
S
Set
A
a
g
e
nt

4786.

4795.
num
e
4804.
direc
ti
o
n
al
p
a
ss
in
g
b
y

4809.
4810.

106

4787.
4788. 4789.

4798.
x
4796.
oxinxn
b 4797.
umeb'
=
oq
4805.
v

4806.4807.

4799.
He
wen
t
and,
pass
ing
by,
calle
d
me

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4811. Directionals and -q with Intransitives


4812. When the above described directionals occur with intransitive stems,
they are accompanied by a suffix -q in all forms.
4813.
4814.
x-

4815.
ul

4816.
-in

4823. 4824.
p
direc
ti
o
n
al

4825.
Set
B

4832.

4833.

4841.
x-

4842.
ox
4850. 4851.
p
direc
ti
o
n
al

4859.

4868.
x-

4834.

4843.
-in
4852.
Set
B

4860.

4861.

4869.
-in

4870.
num
e
4879.
direc
ti
o
n
al

4877. 4878.
p
Set
B

4817.
atin 4826.
intra
ns
iti
ve
st
e
m
4835.

4844.
atin 4853.
intra
ns
iti
ve
st
e
m
4862.

4871.
atin 4880.
intra
ns
iti
ve
st
e
m

4821.
4818.
4820.xulinati
-q
4819.
=
nq
4827.
direc
ti
o
n
al
4828.
4829.
4836.

4830.
4839.

4837.
4838.

4848.
4845.
4847.xoxinati
-q
4846.
=
nq
4854.
direc
ti
o
n
al
4855.
4856.
4863.

4831.
4840.

4849.
I
went to
bathe

4857.
4866.

4864.
4865.

4875.
xinnum
4872.
4874.
eati
-q
4873.
=
nq
4881.
direc
ti
o
n
al
4882.
4883.

4822.
I
came to
bathe

4858.
4867.

4876.
I
bathed
passing
by

4884.
4885.

4886.
4887.This -q suffix with directionals in intransitive constructions would appear
to be related to the -q suffix which denotes future constructions with
intransitive stems, especially since the -q suffix here noted occurs only with
intransitives. It is described here separately, however, since its occurrence
with past tense prefixes contrasts so obviously with the future meaning of the
-q suffix. Deeper analysis of these suffixes in the future may discover
connecting principles between them.
4888.
107

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4889. laj-: Repetitive Action


4890.Prefix -laj- occurs between the T/A/M prefixes and the pronominal
prefixes and means that the action denoted by the verb occurs over and over,
whether carried out by an agent or suffered by a patient.
4891.
4892.
x4899.
past

4906.

4893.
laj4900.
man
ne
r

4894.

4901.
Set
B

4902.
intran
siti
ve
ste
m
the drunkard over and over he
fall down

4910.
4917.
x4924.
past

4895.
- t'ane

4911.
4918.
laj4925.
man
ne
r

4897.
xlajt'
4896.
ane li
=
kalaajenaq

4898.
Th
e
drunkar
d kept
falling
down

4903.4904.
4907.4908.

4909.

4914. 4915.

4916.

4913.
4912.

4919.
in4926.
Set
B

4920.
b'oqe
4927.
intran
siti
ve
pa
ssi
ve
ste
m

4921. 4922.
xlaji
=
nb'oqe

4923.
I
kept
being
called/th
ey kept
calling
me

4928.4929.

4931.
4932.
(In the Carcha dialect an epenthetic
vowel -a- is inserted between x- and -laj-: xalajt'ane.)
4933.
4934.

108

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

4935. -kok and china: Diminutives


4936.The prefixes kok and china are termed diminutives in that they
indicate smallness in the verb action, though this may be interpreted in a
variety of ways semantically. These prefixes do not always occur with the
same stems. Even when they do, as with loq' buy, the semantic
differences between them when they occur with one stem, may not be
the same as when they occur with another stem.
4937.
4938.
x
4939.
in 4946. 4947.
p
Set
B

4940. 4941.
ixchina 4948. 4949.
Se
diminu
tive

4954. 4955.

4956.

4962. 4963.
x
ex
4970. 4971.
p
Set
A

4964. 4965.
ch'ina 4972. 4973.
Se
diminu
tive

4978. 4979.

4980.

4957.

5010. 5011.
x
ex
5018. 5019.
p
Set
A

4966.
loq
4974.
buy

4988. 4989.
kok' 4996. 4997.
Se
diminu
tive

4990.
loq
4998.
buy

5012.
-

4960.
4961.

4975.

4969.
Th
ey
bought it
(a little
one)

4968.
xexch'i
4967.
naloq
=

4976.
4984.
4983.

4985.

4999.

4993.
Th
ey
bought it
little by
little

4992.
4991.xexkokl
=
oq
5000.

5006.

5004.

5008.
5007.

5013.
kok5020. 5021.
Se
diminu
tive

5014.
piki
5022.
dig
more
than
one
thing

4945.
He
gave me
a tender
little call

4959.

4982.

5005.
5002. 5003.

4952.

4951.
4958.

4981.

4986. 4987.
x
ex
4994. 4995.
p
Set
A

4942.
-boq
4950.
call

4944.
xinixch'i
4943.
nab'o
=
q

5016.
5015.xexkok'
=
piki

5009.

5017.
Th
ey dug
little bits

5024.

5023.

5026.
5027. -ke: Immediately (Coban only)
5028.Suffix -ke occurs after the verb stem and after T/A/M suffixes, if any,
and means that the action denoted by the verb is done rapidly or
immediately.
5029.
5030. 5031. 5032. 5033.
n
ti
nume

5034. 5035.
qke

109

5036.
taak

5037.
we

5038.5039.
l
aataqla

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5040. 5041. 5042. 5043.


w
f
S
verb
ste
m
5050. 5051.
=
5052. 5053.

e
5044. 5045.
5046.
f
ma
youn
gi
n
ve
e
-it
r
naq tinnumeqke taak'e we li aataqlankil

5047.
tom
e

nkil.
5048.5049.
t
yourerr
and

When I come by right now, you can give me your errand (to do)

5054.

110

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5055. Voice and Ergativity


5056.
Qeqchi is an ergative language, which may be defined as a
language in which "the subject of an intransitive verb 'becomes' the object
of a corresponding transitive verb, and a new ergative subject is
introduced as the 'agent' (or 'cause') of the action referred to" (Lyons,
1969: 352). That Qeqchi fits these criteria may be noted from the fact
that the Set B pronominal affixes which denote the subject of an
intransitive verb are used to mark the objects or patients of transitive
verbs.
5057.
5058.
x5065.
past

5072.

5079.
x5086.
past

5059.
at 5066.
Set
B
su
bj
ec
t

5060.
war
5067.
intransi
tive
stem

5061.

5068.
5069.
5074.

5075.

5080.
at 5087.
Set
B
p
at
ie
nt
o
bj
ec
t

5081.
in -

5082.
warte
si

5088.
Set A
agen
t
subj
ect

5089.
transi
tiv
e
ste
m

5097.
in 5104.
Set B

5098.
war
5105.
intransi

5073.

5062.5063.
=
xatwar

5076.

5070.
5077.

5084.
5083.xatinwa
=
rtesi

5064.
You
slept

5071.
5078.

5085.
I
put you to
sleep,
made you
sleep

5091.

5090.

5093.
5094.As a result of this ergative construction and also because of the systems
of classes they fit into, it is possible to classify as transitive or intransitive all
inflected verbs. Every verb stem will occur with only Set B affixes or will occur
with Set A and Set B affixes. Ambiguous forms are limited; Set A first person
singular preconsonantal affix -in is identical to Set B first person affix -in, and
if the former co-occurs with Set B third person marker , is indistinguishable
from Set B -in by itself. But if the transitive agent and intransitive subject used
second person, the stem may be identified as either transitive or intransitive.
5095.
5096.
x5103.
past

5099.
5106.

111

5100. 5101.
x
=
inwar

5102.
I
slept

5107. 5108.

5109.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

tive
stem
5112.
5110.

5111.

5117.
x-

5118.
-

5124.
past

5131.

5114. 5115.

5125.
Set B

5119.
in5126.
Set A

5120.
boq
5127.
transi
tiv
e
ste
m

5132.

5133.

5134.

5138.
x-

5139.
at -

5145.
past

5146.
Set B

5140.
war
5147.
intransi
tive
stem
5154.

5152.

5159.
x5166.
past

5113.

5141.

5161.
a5168.
Set A

5123.

called
him

5128. 5129.

5130.

5135. 5136.

5137.

5142. 5143.
x
=
atwar

5144.
y
ou
slept

5149. 5150.

5151.

5156. 5157.

5158.

5148.
5155.

5153.

5160.
5167.
Set B

5121. 5122.
x
=
inboq

5116.

5162.
b'oq
5169.
transi
tiv
e
ste
m

5163. 5164.
x
=
ab'oq

5165.
y
ou
called
him

5170. 5171.

5172.

5173.
5174.The possibility of being able always to identify stems as either transitive
or intransitive becomes important in considering the role of voice as a
grammatical category in Qeqchi. Voice here refers to the distinctions of the
"form" of a verb and its "functions" in indicating the relationship of the
subjects and objects or agents and patients to the action denoted by verb.
The "forms" of the verbs in Qeqchi will always be either transitive or
intransitive and will always occur within the systems of verb classes and verb
inflection presented in preceding sections of this chapter. How voice "functions" and how it relates to these forms is the primary concern of this section.
5175.
5176.
As was mentioned above, the subjects of intransitive verbs may
be either agents or patients. Agent subjects do or commit the action
denoted by the verb. Patient subjects suffer or have done to them the
action denoted by the verb.
112

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5177.
5178.
x5184.
past

5179.
at5185.
Set
B agent
intransitiv
e subject

5190.

5180.
b'e
5186.
stem
5192.

5191.

5196.
x5202.
past

5197.
5203.
Set
B patient
intransitiv
e subject

5208.

5198.
q'a
5204.
stem
5210.

5209.

5214.
x5220.
past

5215.

in-

5221.
Set
B patient
intransitiv
e subject

5226.

5232.
x5238.
past

5216.
t'ane
5222.
stem
5228.

5227.

5233.
in 5239.
Set
B agent
intransitiv
e subject

5234.
wakli
5240.
stem

5181. 5182.
x
=
atb'e

5183.
y
ou
walked

5187. 5188.

5189.

5193.

5195.

5194.

5199. 5200.
x
=
qa

5201.
it
got
spoiled

5205. 5206.

5207.

5211.

5213.

5212.

5218.
x
intane
5217.

5219.
I
fell
down

5223. 5224.

5225.

5229.

5230.

5231.

5241. 5242.

5243.

5235. 5236.
x
=
inwakli

5237.
I
got up

5244.
5245.
Most intransitive stems require that the subject be always
clearly either agent or patient, though some stems might be ambiguous as
to whether the subject does the action or whether the action happens to the
subject, such as stems meaning dream, ripen, fear, and others. Intransitive
verb stems which plainly occur with agent subjects will be called active
intransitives, while those stems which plainly occur with patient subjects
will be called neutral intransitives (Chafe, 1970: 104). Although the
ambiguity of a few stems lends a lack of clarity to this system, its value
should be obvious in the following discussion of voice.
5246.
5247. Active Voice
5248.
The point of departure in discussing voice in Qeqchi is the
transitive verb in which both the agent and patient are identified within the
verb word as part of the normal inflection of a verb. The agent and patient
may be said to be in an active voice relationship with the verb stem wherein
113

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

the agent and patient are in a relationship which is unmarked with respect to
focus or emphasis. The verb stem is shown to be in the active voice by
simply remaining an unchanged transitive stem.
5249.
5250. Reflexive Voice
5251.
The reflexive voice in Qeqchi may be considered as a
subcategory of active voice in that the reflexive voice is not indicated by a
change in the verb stem, which remains as an ordinary transitive stem with
prefixed Set A agent and Set B patient. However, Set B may only be third
person singular which refers to the reflexive relational noun -ib' self,
immediately following the verb. This reflexive noun is formally possessed by
the same Set A pronominal affix as the agent of the verb with which it cooccurs, effectively identifying the patient of the verb as being also the
agent.
5252.

5253.
x
5254.
5264.5265.
p
S

5275. 5276.

5255.
i
5256.
tz'ap
5266. 5267.
S
trans
iti
ve
st
e
m
5277.

5278.

5286.
5288.
5287.
5289.
x
w
il
5297.5298. 5299. 5300.
p
S
S
trans
iti
ve
st
e
m

5257. 5258.
5262.
xi
w
i
5261.
ntz'ap
5259.5260. =
wib'
5
271.
5268. 5269.
5270.
S
s

5272. 5273.
5279. 5280.5281. 5282. 5283. 5284.

5290. 5291.
5292.5293.
5295.
x
w
i
s
le
5294.
wil wib
=
sa lem
5301. 5302.
5303.5304.
S
s
i
m

5305.5306.

5263.
I shut
m
ys
el
f
in

5274.
5285.

5296.
I saw
m
ys
el
f
in
a
m
irr
or

5308.
5309. Passive Voice
5310.
In the passive voice the active voice
relationships of subject/agent and object/patient with the verb are changed
in such a way that the normal object/patient becomes a subject/patient, the
verb stem becomes neutral intransitive, and the former subject/agent
becomes optional and is expressed by an agentive phrase following the verb
word.
114

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5311.

5312.
x5319.
past

5313.
at5320.
Se
t

5314.
in5321.
Set A
agen
t

5315.

tz'a
p

5316. 5317.
= ntz'ap

xati

5322.
tra
nsitive
stem

5318.
I
s
h
u
t

B
y
o
u

p
a
t
i
e
n
t
5326.

i
n
5323.5324.

5328.
5327.

5329.

5333.
x-

5334.
at-

5335.
tzape

5340.
past

5341.
Se
t

5342.
intransi
tive
stem

5330. 5331.

5336.
(inb'aan)
5343.
(
Set A
my
doing)

5338.
xatt
5337.
z'ape
=
(inb'aan)

5332.

5339.
Yo
u
w
e
r
e

s
h
u
t

p
a
t
i
e
n
t

5344.
i
n
(
b
y

5347.

5354.
x5361.
past

5345.

5349.
5348.

5355.
at5362.
Se
t

5356.
in
5363.
Set A
agen
t

5350.

5357.

ch'i
ila

5364.
tra
nsitive
stem

115

5351. 5352.

5358. 5359.
xati
=
nch'iila
5365.5366.

m
e
)
5353.

5360.
I
s
c
o
l

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

B
p
a
t
i
e
n
t
5368.

5375.
x5382.
past

d
e
d
y

5370.
5369.

5376.
at5383.
Se
t

5377.
ch'iila-a5384.
intransi
tive
stem

5371.

5378.

5372. 5373.

5379.5380.
xatc
=
h'iilaak

5385.
no
n-future

5374.

5381.
Yo
u
w
e
r
e

B
p
a
t
i
e
n
t

5386.5387.

5389.
5390. The intransitive stems in the above examples, tz'ape- and ch'iilaa-,
must be viewed both from the point of view of form and from that of function.
In form, tz'ape is derived from the transitive stem tz'ap- of class vtR into an
intransitive stem of class vi; ch'iilaa- is derived from the transitive stem
ch'iila- of class vtD into an intransitive stem of class viV. As intransitives of
their respective classes tz'ape- and ch'iilaa- may occur with all the
inflectional affixes normal for their classes and for intransitive verbs in
general. In form, therefore, tz'ape- and ch'iilaa- are simply ordinary
intransitive verbs.
5391.
5392. In terms of function tz'ape- and ch'iilaa- play the role of the passive
voice of the transitive verbs to shut (something) in and to scold
(someone), respectively. The passive in Qeqchi may be used in, though
not limited to, the following ways. The passive may be used to focus on
the patient who suffers the action denoted by the verb. The agent may not
be known and need not be expressed. If, however, the agent is expressed
in an agentive phrase following the verb, it may serve not so much to
highlight the patient as to further illuminate or define the nature of
agent's action.
5393.
116

s
c
o
l
d
e
d

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5394.
x5403.
pas
t

5412.

5421.
x5429.
pas
t

5437.

5446.
x5454.
pas
t

5462.

5395.
at

5396.
tz'ape

5404. 5405.
S
intra
ns
iti
ve
st
e
m

5413.

5414.

5423.
5422. tz'ape
at
5430. 5431.
S
intra
ns
iti
ve
st
e
m
5438.

5439.

5448.
5447. tz'ape
at
5455. 5456.
S
intra
ns
iti
ve
st
e
m

5463.

5464.

5398.

5400.
=

5397.

5399.
5407.

5406.

5408. 5409.
5416.
5417. 5418.

5415.

5424.
5432.
Set
A

5440.

5449.
5457.
Set
A

5465.

5425.
Inb
'
a
a
n
5433.
myd
o
i
n
g

5441.5442.

5450.
Inm
a
a
k
5458.
myf
a
u
l
t

5466.5467.

117

5426.
=

5401.
xa
ttz'ape

5410.
5419.

5427.
xa
ttz'ape
inb'aan

5434.

5435.

5443.

5444.

5451.
=

5452.
xa
ttz'ape
inmaak

5459.

5460.

5468.

5469.

5402.
You
we
re
sh
ut
in
(ag
ent
un
kn
ow
n)
5420.

5428.
you
we
re
sh
ut
in
by
me

5445.

5453.
you
we
re
sh
ut
in
be
ca
us
e
of
me
(it
s
my
fau
lt)
5470.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5471.
x5479.
pas
t

5473.
5472. tz'ape
at
5480. 5481.
S
intra
ns
iti
ve
st
e
m

5474.
5482.
Set
A

5475.
Ink

a
b

5483.
myn
a
m
e

5476.
=
5477.
xa
ttz'ape
inkaba

5484.

5487.
5488.

118

5485.

5478.
you
we
re
sh
ut
in
be
ca
us
e
of
me
(I
sai
d
to
do
it)

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5489. Antipassive Voice


5490.In the antipassive (or absolutive) voice the active voice subject/ agent object/patient relationship is altered in such a way that the active voice
subject/agent continues as subject/agent but the verb stem becomes
active intransitive and the object/patient becomes optional and, if expressed, occurs in a "patient" phrase following the verb. The patient
phrase uses the relational noun e mouth, edge, dative.
5491.
5495.
5492.
5494.
t
x
a
z
5493.
'
in
a
p
5503. 5504. 5505. 5506.
p
S
S
tran
s
i
t
i
v
e
s
t
e
m

5496.

5500.
5501.
=
xinaat
z'a
5497. 5498. 5499.
p

5507.

5508. 5509. 5510. 5511. 5512.

5516. 5517.
5527. 5528.
5526.
a
tz'ap
x5535. 5536. 5537. 5538.
in
p
S
tran
s
i
t
i
v
e
r
o
o
t
5545.
5547.
y
y
5546.
5548.

5518.

5519. 5520. 5521. 5522. 5523.


5533.
5530. 5531.
5532.
laaat
k
(w- e)
=
xat
5540.
tz'
n
ap
ok
5541.
(w
Set A
e)
pa
ti
en
t

5549.

5550.

5555.

5559.

5560. 5561. 5562. 5563. 5564.


5575.
5574.
xatwa
=
5571. 5572. 5573.
b'I
5582. 5583. 5584. 5585. 5586.

5514.
5525.
la

5515.

5556.
5567.
at

5557. 5558.
5568.
5566.
w
5569.
xab'i
5577. 5578. 5579. 5580.
p
S
S
tran
s

5529.
o5539.
intra
n
si
ti
v
e
st
e
m

5570.
5581.

119

5551.
my

5502.
You shut
me
in

5513.
5524.
5534.
You
(wer
e
the
one)
who
shut
(me)
in

5542.
5552. 5553.

5554.
5565.
5576.
I heard
you
5587.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

i
t
i
v
e
s
t
e
m
5588.

5599.
la

5589.

5590. 5591.

5601.
5600. i
5602.
xab'i5610. 5611. 5612. 5613.
in
p
S
tran
s
i
t
i
v
e
s
t
e
m
5620.
5622.
I
5621. I
5623.

5592.

5603.
n
5614.
intra
n
si
ti
v
e
st
e
m

5593. 5594. 5595. 5596. 5597.


5608.
laain
xin
5607.
ab'
=
in
5604. 5605.
(aa
(
e
we
5606.
)
5615.
Set A

5598.
5609.
I (was
the
one
who)
hear
d
(you
)

p
at
ie
nt
b
as
e
5616.5617. 5618.

5624.

5625.
your

5626. 5627. 5628.

5629.

5630.
5631.The antipassive voice, like the passive, must be viewed both in terms
of form and in terms of function. Looking at the above forms, it may be
noted that -tz'apo- and -ab'in- are intransitive stems of classes vik and
vin, respectively, and are derived from transitive stems -tz'ap- and -ab'i-,
which are of classes vtR and vtD, respectively. As intransitive stems of
their respective classes, -tz'apo- and -ab'in- may be inflected in a manner
undifferentiated from other intransitive verbs.
5632.
5633.

120

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5634. Uses of the antipassive.


5635.
In terms of function the antipassive may be used to
suppress the object/patient in order to form a simple active intransitive . The
antipassive may be used to express a generic (or non-specific) object.
5636.
5637.
x-

5638. 5639.
a5645. 5646. 5647.
p
S
Set A
ag
en
t

5640.
yok
5648.
tra
n
s
i
t
i
v
e

5641.
li si
5649.
the
fire
wo
od

s
t
e
m

5642.
=

5650.
5657.

5653.

5654.

5655.

5661.
but see
next example:

5656.

5662.

5675.
x-

5668.

5669.

5677.
5676. yok'atok
5683. 5684. 5685.
p
S
intra
ns
iti
ve
st
e
m

5670.

5678.
si
5686.
fire
w
o
o
d

5659.

5663.

5660.

5673.

5679.

5680.
=

5681.
xatyo
k'o
k
si

5687.

5688.

5689.

121

5652.

5665.
5664.
5672.

(
g
e
n
e
r
i
c

5644.
You
cut the
firewood

5651.

5658.

5671.
5667.

5643.
xayok
li
si

5666.
5674.

5682.
You
cut
firewood
5690.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

b
j
e
c
t
)

5691.
5692.
The second example may be paraphrased as you cut firewood
(for a living) or you cut firewood (your usual task) or you cut firewood
(among other things). It must be noted that other intransitive stems, not
just antipassives, may occur with this type of object.
5693.
5694.
The antipassive is used in relative clauses and questions to
indicate that it is the agent/subject of the clause or question that is being
relativized or questioned for. In the examples below five simple sentences
are provided followed by four complex sentences showing how the
antipassive is used.
5695.

122

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5696.
5697.
(
5698.

5705.

xka
msi

5699.
li

5700.
tzi

5701.
li

5706.
hekilled-it

5707.
the

5708.
dog

5709.
the

5715.

5716.

5702.
win
q
5710.
ma
n

5704.
T
he
5703.
man
=
killed
the
dog.
5711.

5718.

5719. 5720.

5717.
5713. 5714.

5721.
(
5722.

xka
msi

5723.
li

5730.
he5729.
killed-it

5731.
the

5724.
win
q
5732.
ma
n

5737. 5738.

5739.

5740.

5725.
li
5733.
the

5726.
tzi
5734.
do
g

5728.
T
he
dog
5727.
killed
=
the
man.
5735.

5742.

5743. 5744.

5741.

5745.
(
5746.

xbo
qeb'

5754.
hecalled5753.
them

5747.
li

5755.
the

5748.
kok

a
l
5756.
chil
d
r
e
n

5752.
T
he
man
5751.
called
=
the
childre
n

5749.
li

5750.
win
q

5757.
the

5758.
ma
n

5759.

5766.

5767. 5768.

5765.
5761. 5762.

5769.
(
5770.

xera
alina
5778.
theychased-it

5763.

5764.

5771.
li
5779.
the

5772.
tzi
5780.
dog

5773.
li
5781.
the

5777.

5774.
kok

a
l
5782.
chil
d
r
e
n

5776.
T
he
childre
5775.
n
=
chased
the
dog.

5783.

5789.
5785. 5786.

5787.

5793.
(
5794.
xraal
inaheb
5801. 5802.
itchasedthem

5795.
li
5803.
the

5788.

5796.
kok

a
l
5804.
chil
d
r

5790.

5797.
li
5805.
the

123

5798.
tzi
5806.
do
g

5791. 5792.

5800.
T
he dog
chased
5799.
the
=
childre
5807.
n.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

e
n

5809.
5810.
(
5811.
li
5822.
the

5821.
5832.5833.
5843.

5845.5846.
5857.
5856. xka
(
m
si
5868.
hek
ill
e
d
it
5867.
5878. 5879.
5889.

5891.5892.

5902.
(
5903.
li

5814.
xka
5812. 5813.
m
w
l
s
i
5823. 5824. 5825.
m
w
itk
i
l
l
e
d
h
i
m
5834. 5835. 5836.

5844.

5815.
li
5826.
the

5837.

5838. 5839.

5840.

5820.
kok

5819.
a
li
l
5830. 5831.
t
chil
d
r
e
n

5841.

5842.

= The man whom the dog killed had called the children.

5847. 5848. 5849.

5850.

5858. 5859.
l
t
5860.
li
5869. 5870. 5871.
t
d
the

5861.
win
q
5872.
ma
n

5880. 5881. 5882.

5883.

5890.

5816.
t
5817.
ak
5827.5828.
d
alre
a
d
y

5818.
xboq
eb

5829.
heca
lle
dth
e
m

5851. 5852.

5863.
xera
5862.
al
l
in
a
5873.5874.
t
they
c
h
a
s
e
d
it
5884. 5885.

5853.

5864.
li
5875.
the

5854.

5855.

5865.
k
5866.
5876.
childr
en

5877.
5886.

5887.

5888.

= The man killed the dog that the children chased.

5893. 5894. 5895.

5906.
xka
m
5904. 5905.
s
i
n

5896.

5907.
re

5897. 5898.

5899.

5900.

5908.
5909.
tzi

5910.
xboq
eb

5912.
kok

5911.
a
li
l

124

5901.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5914.
the

5915. 5916. 5917.


m
w
hek
i
l
l
e
d

5913.
5924.
5926.5927.

5938.
5937. xka
(
m
si
5949.
hek
ill
e
d
it

5925.

5959.

5939. 5940.
l
t
5941.
li
5950. 5951. 5952.
t
d
the

5960.

5919.5920.
t
dog

5921.
calle
d

d
a
t
i
v
e

5922. 5923.
t
chil
d
r
e
n

(
a
n
t
i
p
a
s
s
)
= The man who killed the dog called the children.

5928. 5929. 5930.

5948.

5918.
its
-

5931.

5942.
win
q
5953.
ma
n

5932. 5933.

5944.
xa
al
5943.
in
l
a
n
5954.5955.
t
itc
h
a
s
e
d

5934.

5945.
reheb
'
5956.
their
d
at
iv
e

5935.

5947.
kok

5946.
a
li
l
5957. 5958.
t
chil
d
r
e
n

(
a
n
ti
p
a
s
s
)
= The man killed the dog which chased the children.

5961.

125

5936.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

5962.In example (6) winq, subject/agent (S) of the main clause becomes the
object/patient (O) of the relative clause, represented by relative pronoun li,
while the verb (V) remains in active voice:
5963.SX (OX V S) V O. In example (7) tzi, the object/patient of the main clause,
becomes the object/patient of the relative clause, represented by li, while the
verb remains in the active voice: V OX S (OX V S). In example (8) winq, the
subject/agent of the main clause is also the subject/agent of the relative
clause, represented by li. Now, however, the verb changes to the antipassive
voice
5964.(VAP) and is accompanied by prepositional nominal re, glossed here as
dative, which marks the object (ORE ): SX (SX VAP ORE ) V O. In example (9) tzi,
the object/patient of the main clause, becomes the subject of the relative
clause, represented by li. In this case also the verb is changed to the
antipassive voice and the object is marked by the prepositional nominal: V OX
S (SX VAP ORE). Thus it can be seen that when either S or O in the main clause
become S in the relative clause, the verb in the relative clause must be
expressed in the antipassive voice and the object accompanied by the
prepositional nominal -e possessed by Set A. If, however, either S or O in the
main clause becomes O in the relative clause, the verb in the relative clause is
expressed in active voice.
5965.
5966.In questions, use of the antipassive depends on whether one is asking for
the subject/agent or object/ patient.
5967.
5968.
(
5969.
ani
5979.
w
h
o
5978.

5970.
x5980.
pa
s
t

5973.
sa

5971. 5972.
x5981. 5982.
S
Set A
he-

5998.
(
5999.
ani
6009.
wh
o

5990.

5991.

6000.
x6010.
pa
s
t

6001. 6002.
sakok
6011. 6012.
Se
intran
s.
(an
tipa
ss)
hit

5977.
W

5983.
tr
5984. 5985. 5986.

5992.
5988. 5989.

5976.
a
5975.
ni
5974. =
xsak?

5993.

5994.

5995. 5996.

b
a
s
e

6015.6016.

6006.
a
ni
6003. 6004. 6005.
xsako
re
=
k re?
6013. 6014.
Se
pat
i
e
n
t

6008.

126

5997.

6007.
W

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6018.
6019.In example (10) ani refers to the object/patient of the question and the
verb is thus in the active voice. In example (11), however, ani refers to the
subject/agent of the question and the verb is placed in the antipassive voice.
6020.

6021. Absolutives and the formation of infinitives and


participles.
6022. It was mentioned briefly above that the formation of infinitives of vtD
stems involved the derivational affix -n, which also derives antipassive
intransitive stems from vtD transitive stems. In this section we will review
both the two classes of transitive stems, vtR and vtD, and their
antipassive intransitive counterparts, vtk and vin, respectively, and how
each forms infinitives.
6023.
6024.
6027.
chap
6034.
kayi

transitive stem
6028.
6029.
grab
vtR
(something)
6035.
6036.
sell
vtD
(something)

6025.
6026.
6031.
6030.
chap-o
6038.
6037.
kayi-n

antipassive stem
6032.
6033.
vtK
rab
6039.
6040.
vin
ell

g
s

6041.
6042.Infinitives are formed on the antipassive stems, as with all other
intransitives, by the simple addition of the derivational suffix -k.
6043.
6044.
cha
p
o
k
6047.
k'ayi
n
k

6045.6046.
to
v
grab,
grabbing

6048.
6049.
to sell,
v
selling

6050.
6051.Infinitives of vtR transitive stems are formed by adding the
infinitive derivational suffix
6052.-b'al, this latter co-occurring with the Set A possessive prefix which
denotes the patient of the infinitive.
6053.
6054.
x-

6055.

6060.
Set A

6061.
vtR
stem

cha
p-

6056.
b'al
6062.
infiniti
ve

127

6058.
6057. xchapb'a
=
l
6063. 6064.

6059.
to
grab it/him,
grabbing
it/him

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6066.
his/it
s

6067.
to grab /
grabbing

6072.

6068.

6070.
6069.

6074.
6073.

6078.
in6085.
Set A
my

6076.

6075.

6079.

cha
p

6082.
6081. inchapb'
=
al

6080.
b'al
6087.

6086.

6077.

6083.
to
grab me,
6084.
grabb
ing me

6089.
6088.

6091.
6092.
It will be noted that xchapb'al, by containing an inflectional
prefix, seemingly violates the traditional definition of an infinitive which says
that an infinitive is an unmarked or uninflected verbal noun which is the
"name" of the action of the verb. The term infinitive has been chosen to
represent this form, notwithstanding the traditional definition, for
considerations of internal consistency in the grammar. Stems in Qeqchi are
rigorously transitive or intransitive, and in order to distinguish between them,
it is necessary that in the case of transitives the patient, represented by an inflectional prefix, be expressed.
6093.
6094.In the case of transitive infinitives, the normal marking of patients
by Set B affixes is replaced by Set A. The progressive construction, not
elaborated elsewhere in this grammar, illustrates one use of the
infinitives:
6095.
6096.
yoo6105.
progr
es
si
ve

6097.
k6106.
nonfu
tu
re

6098.
in
6107.
Se

6099.
chi
6108.
at

6100.
aa6109.
Se

6114.

6115.

6116.

6117.

6118.

6123.
yoo6132.
progr
es
si
ve

6124.
k6133.
nonfu
tu
re

6125.
at
6134.
Se

6126.
chi
6135.
at

6127.
in6136.
Se

6101.
6103.
yoo
sa
6102.
kin chi
=
aasak'b'al
6110.
hi

6104.
I

6111.6112.
6119.

6120. 6121.

6128.
6130.
yoo
sa
6129.
kat chi
=
insak'b'al
6137.
hi
6138.6139.

6141.
6142.
Infinitives of vtD transitive stems, in contrast, are formed by
first forming the absolutive intransitive infinitive and then adding the
inalienable possession suffix -il plus the Set A possessive prefix.
6143.
128

6122.

6131.
y

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6144.
k'ayi-

6145.

6150.
vtD.
Stem

6151.
a
ntipa
ssive

6147.
=

6152.
in
finitive

6153.

6146.

6156.

6148.
kayink
6154.
6160.

6157.

6158.

6162.
x-

6163.
k
'ayink-

6168.
Set A
prefi
x

6169.
i
ntran
sitive
antip
assiv
e

6159.

6164.

il

6170.
in
alienabl
e
infinitiv
e

6165.
=

6171.

6149.
to
sell,
selling

6166.
xk'ayin
kil

6155.
6161.

6167.
to sell it,
selling it

6172.
6173.

6174.
6175.
Past participles
derived from transitive stems have a similar derivation. Past participle
derivational suffix is -b'il. Participles formed on vtR transitive stems simply
add the suffix -b'il to the stem: chap- bil grabbed. The same participle for a
vtD stem requires that the suffix -n- be added making an absolutive stem to
which bil is added.
6176.
6177.
k'ayi -

6178.

6183.
vtD
stem

6184.
a
bsolu
tive

6179.
b'il

6185.
pa
rticiple
suffix

6180.
=
6186.

6181.
k'ayimb'
il

6182.
ld

so

6187.
6188.

6189.

6190. Non-specific Passive Voice


6191.
In the Coban dialect there exists a voice form in which the active
voice relationship of subject/agent and object/patient are changed in such a
way that the object/patient becomes subject/patient, the verb stem becomes
neutral intransitive, and the former subject/agent becomes an unspecified,
unidentified agent. Subject/patient is limited to third person singular.
6192.
6193.
x6202.
pas
t
6211.

6194.

6195.

6203.
S
6204.
S
et A
6212.

6218.

6219. 6220.

6196.
yok'
6205.
transiti
ve
stem

He-cut-it-down

6197.6198. 6200.
l
c 6199. xyok li
=
che
6206.6207.
t
t
6209.
6208.
6213.6214.
6215.

6221.

6222.6223.
6224.

129

6201.
He cut
dow
n
the
tree.

6216.
6225.

6226.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6227.
x-

6228.

6235.
pas
t

6236. 6237.
intransitive
S
non-specific
passive

6229.

yok man

6233.
6230.6231.
xyokma
l
c 6232.
n li
=
che
6238.6239.
t
t

6241.

6240.
6243.

6252.
na6260.
pre
s
/
h
a
b
.

6246.

6250.

6244.6245.

6247.6248.
6249.

6253.

6258.
nayok'm
6257.
an
6256.
=
che
6263.6264.
t
6255.
c

6254.
yok-man
6261. 6262.
intransitive
S
non-specific

6266.
6265.

6234.
The
tree
was
cut
dow
n/
The
y
cut
dow
n
the
tree.
6251.

6259.
Trees
are
cut
dow
n. /
The
y
cut
dow
n
tree
s. /
You
cut
dow
n
tree
s.

6268.
6269.The English translations with they and you are intended to indicate an
impersonal agent. In form, yokman in the above examples represents transitive
stem yok cut plus derivational suffix
6270.-man, which derives an intransitive stem of class vin. In terms of
function the non-specific passive voice is used to indicate a situation in
which the patient is known but the agent is either unknown or non-specific,
and in which the action is often habitual or routine.
6271.
6272.

cha
n ru
6276.
ho
w

6273.

nayeema
n..?

6277.

it is said

6280.

6281.

6284.
nati
wman
6288.
it-

6274.
6278.
6282.

6285.

wulaj
wulaj
6289.
every

130

6275.
How do
you say ?
6279.
6283.

6286.

6287.
You eat it
every day.

6290.

6291.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

is-eaten
6292.

6296.

aa
n

6300.

6297.

inka

th
is

6304.

day
6293.

6301.
negative
6305.

131

6294.

6295.

6298.
nab
aanuman
6302.
it-isdone
6306.

6299.
This is
not done.
6303.
6307.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6308.

132

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

6309.
6310.
6311.
6312.

6313. CHAPTER 5: NOUNS


Definitions
Nouns are defined as the class of stems which may occur prefixed by the Set
A pronominal affixes to indicate possession. The few nouns which cannot be
possessed nonetheless have the same distribution as those which can be
possessed. Nouns may also occur with Set B pronominal affixes in stative
sentences (see next section); Set B functions as the topic or subject and the
noun as the commentary or complement in this type of sentence. When Set B
is third person singular , therefore, the sentence consists only of the noun.
Nouns may also be modified by adjectives.
Examples:
Consonant-initial noun stem
tzi
dog
intzi
my dog
aatzi
your dog
xtzi
his/her dog
qatzi
our dog
eetzi
yalls dog
xtzi-eb
their dog

wawrqeerr-

Vowel-initial noun stem


aaq
pig
aaq
my pig
aaq
your pig
aaq
his/her pig
aaq
our pig
aaq
yalls pig
aaq-eb
their pig

Example of noun in stative sentence


winq
winqwinqwinqwinqwinqwinq-

in
at

o
ex
eb

man
Im a man
youre a man
hes a man
were men
yall are men
theyre men

Example of modified noun


nimla

tzi

big dog

133

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

saqi tzi
chaabil tzi

white dog
good dog

77

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Simple Noun Classes


Classes of simple nouns are determined by whether a noun can be possessed or
not, whether the possessor is known, whether a possessor is human or nonhuman, and the presence or absence of two suffixes, -Vl (any vowel plus -l) and
-(b)ej in both possessed and non-possessed forms. These suffixes are
inflectional, not changing the meaning of the noun stem nor the grammatical
class but rather adding information about possession or lack of it with certain
noun stems. Vaguely, -(b)ej indicates that the noun is of a type that ought to be
possessed, but the possessor is not known or is suppressed; -Vl indicates that
the possessor of the item is non-human. The inflectional suffix -Vl should not be
confused with other suffixes which have the same form but which are, however,
derivational. Chart XIV will provide a point of reference for the discussion of
noun classes to follow.
Class I
Class I nouns are nouns which may not be possessed. These nouns may be
underived roots or derived stems. The examples below demonstrate both types,
including two with the above-mentioned suffixes.
palaw
saqe
saqehil
nabejil

lake
sun
summer
motherhood

Saq'ehil obviously contains root saq'e(h) sun plus derivational suffix -il, which
appears to represent the -Vl inflectional suffix mentioned above, but in fact
derives an abstract state noun from a concrete noun.
Class II
Class II nouns are nouns which may be possessed by either humans or
non-humans, but may not appear with -(b')ej to indicate possessor
unknown nor may it appear with -Vl to indicate non-human possessor.
wakax cow
in-wakax my cow

78

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Class III
Class III nouns are nouns which occur with the suffix -(b')ej when not possessed
and without the suffix when possessed. The -(b')ej suffix, however, indicates not
that the noun is unpossessed but rather that it is a noun that normally should
be possessed by someone but the speaker cannot or does not identify who the
possessor is. The classic examples of Class III nouns are kinship terms. A person
cannot be a "father," for example, without being "someone's father." Or, in
other words, children "possess" their fathers. Thus, if a speaker refers to a man
as father without mentioning whose father he is, the speaker must add the (b')ej suffix to mark this.
chaabil-

good

yuwafather

bej
possess
or
unknow
n

tento
should

naq
that

Set B
he
li
The

yuwab
ej
father

= chaabil
yuwabej

He's a good father.

taatrab'aaji
k.
he-willwork

Fathers (or a father) should


= work.

Chart XIV Simple Noun Classes


Class

Example

saqe
saqehil
nabejil

II

wakax

III

sun
summer
motherhood

Presence of
suffixes
-(b)ej or Vl
-------------

cow

-----

In-wakax my cow

-----

na bej mother

-bej

in- na my mother
IV

xe root
in-xe my root
x- xe -el its root
cha ash

79

Possession Type
Unpossessable
Unpossessable
Unpossessable
Unpossessed
Possessed (+/human)

-----

Possessor unknown
Poss known (+
human)

---------Vl

Unpossessed
Possessed (+ human)
Possessed (- human)

-----

Unpossessed

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

x- chah -il its ash


VI

si
si- ej
in- si
x- si- il

firewood
firewood
my firewood
its firewood

80

-Vl

Possessed ( - human)

-----ej
-----il

Unpossessed
Possessor unknown
Possessor ( + human)
Possessor ( - human)

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Class IV
Class IV nouns are nouns which occur without suffixes when unpossessed or
possessed by humans, but occur with a -VI suffix when the formal possessor is
non-human. The relationship of possessor to the thing possessed is not always
semantically that of possession. The following examples will show something of
the range of this type of possession.
xxulel
ixim
= xxulel ixim
Set A possessor
its
animal
- human pos.
corn
= the bug or worm in the corn (the corn possesses or "has" the bug)
ocho
wch
=
Set A possessor
Set A
hous
- human pos.
its
Road
my
e
= the road to my house (house possesses the road)
x-

be(h)-

il

xSet A possessor
its

tib-

el

wa
tortill
-human pos.
Meat
a
= meat to go with the tortillas

xSet A possessor
its

wa(h)-

il

xSet A possessor
its

tib

xb'ehil
wochoch

xtibel wa

xwahil tib

-human pos.
Tortilla
meat
= tortillas to go with the meat

chakachBasket

il

kenq

xchakachil
= kenq

-human pos.

beans
= basket for the beans

Class V
Class V nouns are like those of Class IV with the exception that they may only
be possessed by non-human possessors with the -VI suffix and never by human
possessors. It is impossible to possess smoke, for example, though one may
possess another object, such as a cigarette, which formally possesses the
smoke.
sib'
sik

81

smoke
cigarette

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

x- sibits smoke

el
- human

inmy

sik
cigare
tte

pos.

82

xsib'el
insik'

my
cigarette
smoke

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Class VI
Class VI nouns are nouns which combine the qualities of Classes III and IV. Like
Class IV nouns they may occur unpossessed without a suffix, possessed by a
human possessor without a suffix, and possessed by a non-human possessor
with the -VI suffix. Like Class III nouns they may also occur unpossessed with
the -(b)ej suffix to indicate that the noun has or should have a possessor but
that the possessor is not known.
kam
bring-it

chaq
here

wan
there-is

li
the

us

xk'eebal

li
si
the
firewood!
= Bring the firewood here!
inmy-

si
firewood
= I have firewood.

x-si-il
li
ukal
itsgood
to-put-it
firewood
the
pot
= A piece of firewood should be put under the pot./ We should put a stick of wood
under the pot.
ma
kam
che
aan
li
abanan
siej
negativ
take-it
the
wood
because that(i
firewood
e
s)
= Dont take that wood because its firewood (and belongs to someone).

Dialect Variation in Simple Noun Classes


The problem with the above-presented system of noun classes is the extreme
degree of dialect variation involving the suffix -(b')ej among the pueblos of
Coban, Carcha, Chamelco, and Chamil, In Carcha and to a lesser degree in
Chamil and Chamelco the above system is valid. But in Coban and to some
extent Chamil, the -(b)ej suffix has been reduced to the status of a nonproductive derivational suffix with other than kinship forms and a few other
stems.
jolom
jolom-ej
oq
chi oq-ej

head
head cold
foot
on foot

There may be only one noun in the Coban dialect that fits the criteria for
Class VI nouns:
83

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

kaba
kaba:
kabaej:
xkaba:
xkabail
:

jun kab'a xwab'i chaq.


k' am chaq we chixjunil li kab'aej re
xnawbal aniheb xewulak anaqwan.
Ani xkaba a winq aan?
Aj elq kexke choq' xkabail xb'aan
naq jwal naelq'ak.

name
I heard a name.
Bring me all the names so
that (I can) find out who
came today.
What's that man's name?
They gave him as a
nickname "the robber"
because he robs a lot.

Compounds
The following section deals with a variety of noun constructions which will be
lumped together under the term compounds. In many cases these will
appear to be more like phrases than proper compounds and will at times
appear to vary between phrase and compound. What will not be considered
a compound is a simple noun modified by an adjective, such as nimla ochoch
big house, saqi t'ik'r white cloth, chaab'il winq good man.
Possession plays an important role in the analysis of compounds in Qeqchi.
It is often impossible to predict how a compound is to be possessed and,
without this knowledge, difficult to determine its meaning. Three major
classes of compounds are discussed below, the classification based on their
differences involving possession.
Class I Compounds
Class I compounds are the most varied and complex of the compounds. The
chart below provides a point of reference for the discussion which follows.
Each subclass of Class I is provided with a "formula," which contains the
following elements:
N1:

noun stem which is first member of the compound, reading left to right

N2:

noun stem which is second member of the compound

x-:

third person singular preconsonantal Set A prefix. In the formulas, xalways indicates possession of N1 by N2 and never possession of the
compound as a whole.

pos-:

Set A prefixes which in the formulas always indicate the possessor


of the compound as a whole. All examples are given in first person
singular.

Class I compounds are those in which the second member of the compound
N2 either directly modifies N1, or modifies N1 by formally possessing it. An
84

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

example of this in English showing both types of modification will be used as


an illustrative key in presenting the compounds of Class I. (NB: In English N1
and N2 are the reverse of Qeqchi.) The example occurs in Freeze (1976:
115), though it is expanded somewhat differently here.
(1) the table top
(2) the table's top = (3) the top of the table

In each of the three examples given above table modifies in some way top.
The difference between them is that (2) is a possessive or noun-of-noun
construction while (1), perhaps with a very slight difference in meaning, has
lost the possession marker of (2), resulting in a "frozen" lexical item.
A similar process occurs in Class I compounds in Qeqchi, but with the
difference that, under possession, compounds may flip-flop or alternate
between being like (1) and being like (2).
As mentioned, the basic structure is that N1 modifies N2 .
(3) naq

seed face = face-seed = eye


(4)x- naq
u
its-seed face = face its-seed =

face's seed = eye

In (3), N2 u face, modifies N1 , naq' seed. In (4), N2 formally possesses N1 , but


the resultant meanings of both are the same, or nearly so. Essentially, (3) is
parallel to (1) in English, while (4) corresponds to (2). However, rarely do both
(3) and (4) both occur as unpossessed compounds in the same dialect of
Qeqchi. Usually one or the other occurs as the unpossessed form while the
other occurs where the compound is possessed.
Class la Compounds.
x-N1 N2 / x- N1 pos- N2
In the above formula x- indicates Set A possession of N1 by third person
singular and is coreferential with N2 ; N2 possesses N1 . -pos indicates the Set
A possession of the whole compound. Diagonal [/] separates the unpossessed
form from the possessed form. Class la are of the structure N1 possesses N2 in
both possessed and unpossessed forms, as in (2) and (4) above. N1 always
occurs with Set A possessive prefix x- to indicate this relationship. When Class
la forms are possessed, the possession is marked by the appropriate Set A
prefix attached to N2 .
Examples of Class Ia:
x-

been

tel

its

space above

arm

85

shoulder

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

x-

been

in-

tel

its

space above

my

arm

r-

aq

xam

its

tongue

fire

r-

aq

in-

xam

its

tongue

my

fire

x-

yaal

tu

its

juice

breast

x-

yaal

in-

its

juice

my

r-

ismal

its

hair

eye

r-

ismal

w-

its

hair

my

eye

my shoulder

flame

my flame

breast milk

my milk (the mothers)

eye lash

my eye lash

tu
breast

N1 may contain inflectional suffix -Vl, indicating that the possessor, i.e., N2, is
non-human.
x-

sifirewo
od

il
non-human
pos.

its

sifirewo
od

il
non-human
pos

x-

be(h)-

its

road

il
non-human
pos

x-

be(h)-

w-

its

road

il
non-human
pos.

my

its
x-

ji

xsiil ji

oak

oak firewood

in-

ji

xsiil inji

my

oak

my oak firewood

ochoch

xbehil ochoch

house

the road to the house

ochoch

xbehil ochoch

house

the road to my house

N2 may contain the inflectional (b)ej, indicating that its possessor is not
known and thus that the possessor of the entire compound is not known. This
form is found only in Chamelco.
r-

alal

iitzin-

-bej

86

ralal iitzinbej

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

his/her

son

younger
brother

pos.
unknown

r-

alal

w-

his/her

son

my

-iitzin
younger
brother

nephew

ralal wiitzin

my nephew

Class lb.
Compounds of Class lb are identical with Class la compounds when possessed:
N2 possesses N1 while the possessor of the whole compound formally
possesses N2. However, in its unpossessed form N1 lacks the possessive
prefix of Set A, x- or r-, which would indicate that it is possessed by N1 . It
appears rather that N2 modifies N1 when the compound is unpossessed, but
that N2 possesses N1 when the compound is possessed. Thus, the
unpossessed forms are like (1) and (3); the possessed forms are like (2) and
(4).
mol
egg

akach
turkey

xits

mol
egg

wmy

rahil
pain

choolheart

ej
pos. unknown

xits -

rahil
pain

inmy

akach
turkey

chool
heart

turkey egg

my turkey egg

sadness

my sadness

A small subclass of Class Ib compounds are those few compounds in which N1


does not occur with the possessive prefix linking it to N2. It is felt that there are
at least two explanations to account both for considering these words as
members of Class Ib and also for the prefix lack.
(1
)

sa

spaceinside
sa
spaceinside

oq
foot
w-

oq

my

foot

87

sole of foot

my sole

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

(2
)

rahil

xul

pain

bug

mouth,
tooth

rahil

xul

w-

pain

bug

my

e
mouth,
tooth

toothache,
caries

my toothache

In the first example sa appears to be a noun, but it may also be used


commonly as a preposition meaning at, in, to (see 4.5.1). Since sa never is
possessed as a preposition, this may have been extended to its noun use in the
above and other similar compounds:
sa
space inside

tel
arm

armpit

sa
space inside

a
leg

crotch

In the second example, a rare type of compound because it involves three


nouns, both of the first two nouns begin with phonemes also used for
third person possession: r- and x- . It seems reasonable to allow one of
these to stand doubly for the third person marker, xul bug indicates that
the pain is caused by some sort of small animal, indicating an
understanding of what causes caries and thus toothaches.
Class Ic.
Compounds of Class Ic function exactly opposite of Class lb in that in the
unpossessed form of the compound N2 formally possesses N1 just as in Class
lb, but in the possessed compound this formal possessor is replaced by the
possessor of the compound. Thus, when the compound is possessed as a
whole it has the structure of N2 modifies N1 as in example (1), whereas when
unpossessed as a whole, N2 possesses N1 as in example (2).
rits

akach
turkey

wmy

akach
turkey

tzuul
the hill
tzuul
hill

=
=

turkey of the hill


wild turkey

my wild turkey

Class Id.
Compounds of Class Id have unpossessed forms like lb in which N2 modifies N1
but does not formally possess it. However, in contrast with Class lb, the
compounds of Class Id may be possessed in two ways: they may be possessed
like lb, where N2 possesses N1 and the possessor of the compound possesses
88

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

N2, or the possessor prefix may be attached to N1 . In this latter case it is still
understood that N2 modifies N1 , and the possessor possesses the entire
compound, which distinguishes Class Id from Class II compounds, to be
discussed below. Possessed Class Id compounds may differ in meaning in some
cases, depending on the type of possession employed. In the examples given
below differences of meaning will be noted when necessary. (First two examples
for Chamelco only.)
sok
nest

jolom
head

xits

sok
nest

inmy

inmy

sok
pad

jolom
head

ke
coldness

ixback

ej
posses
sor
unkno
wn

xits

ke
coldness

wmy

inmy

ke
coldness

ix
back

kaxon
box
(Sp. cajn)

seer
bee
(Sp. cera
wax)

xits

kaxon
box

inmy

inmy

kaxon
box

seer
bees

head-nest

jolom
head

= head pad for carrying


something on head

= my head pad

= my head pad

backcoldness

ix
back

= shiver

= my shiver

= my shiver

bee-box

seer
bees

= beehive

= my beehive
(for my own bees)

= my beehive
(but I dont have any
bees)

89

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Class Ie.
Possessed compounds of Class Ie exhibit the same possibilities as those of Class
Id. The unpossessed form of Ie, however, contrasts with unpossessed Id in that
N2 formally possesses instead of simply modifying it, as in class Id.
xits

kaam
string

-al
non-human

xaab
shoes

shoestring

my shoestring
(that goes with my
shoes)

my shoestring
(but not from my
shoes)

possessor
xits

kaam
string

-al

inmy

kaam
string

-al
my

inmy

xaab
shoe
s

xaab
shoes

Why in-kaam-al xaab does not reduce to in-kaam xaab when xaab ceases to
possess kaam-al is an irregularity. See the following example.
xits

sifirewood

il
non-human
possessor

ji
oak

xits

sifirewood

il

inmy

inmy

si
firewood

ji
oak

ji
oak

oak firewood

my oak firewood
(from my oak tree)

my oak firewood
(but not necessarily from
my tree)

Class If.
Compounds of this class are restricted to those in which N1 is vowel initial,
thereby using the r- allomorph of the Set A third person singular possessive
prefix. Class If compounds might be considered as an aberrant subclass of Ie,
differing only when possessed as N2 modifying N1 , but not as N2 possessing N1 .
Class If compounds, instead of replacing the Set A prefix indicating that N2
possesses N1 with the Set A possessor of the whole compound, retain the third
person prefix r- and then add another Set A prefix to it.
rits -

u
fruit

chaj
pine

rits -

u
fruit

inmy -

chaj
pine

90

pinecone

pinecone (from my pine

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

tree)
in-

r-

chaj

my

its

fruit

pine

my pinecone (not my
pine tree)

Class Ig.
Logically, it is possible that a compound exist in which the only possibility,
whether possessed or unpossessed, would be that N2 modify and never that N2
possess N1 . No such compounds exist, apparently.
Dialect Variation in Class I Compounds.
The preceding sections have detailed the processes underlying the subclasses of Class I compounds. But however valid these processes are for
the dialect area as a whole, it is impossible to fix the Class I compounds
once and for all into sub-classes since there is wide disagreement among
speakers from the different towns in the dialect area as to which of the
sub-classes is the correct one for a given compound. That is, a compound
may be of a certain sub-class for speakers from Carch, for example, but
be of another class for the speakers from Chamil.
One example of this dialect differentiation is the compound meaning
wrist. For speakers from Chamil and Chamelco the compound is Class la:
x-

kux

its

neck

xits

kux
neck

uq
(the)
hand
wmy

the hands neck

uq
hand

wrist

my wrist

For speakers from Carch and Coban, the compound is Class lb:
kux
neck

uq'
hand

hand-neck

wrist

xits

kux
neck

wmy

uq
hand

my wrist

Similar examples abound, and there seems to be no pattern that allows for
predicting where a particular compound will be classified in a given town.

91

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

92

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Class II compounds
Class II compounds are fundamentally different from Class I in that the first
member of the compound, which may be either a noun or adjective and
which will be referred to as Mod, modifies the second member, which is
always a noun.
Class II compounds seem to be derived from modifier plus noun in
ordinary noun phrases. What will be considered as Class II compounds
include:

Mod N, where Mod is a noun, and

Mod N, where Mod is an adjective, but through either lack of enclitic -i and/or
where the sum of these two as plain modifier plus noun is not the same as
the meaning of the compound.

Examples:
(1)

kum
pump
kin

chile

a type of fat, pumpkin-shaped


chile

sa
stoma
ch

dysentery

kenq
bean

ground-bean, variety that grows


on the ground, not on stalks

che
tree

roof beam

green

hal
cornear

but

rax-

kik
blood
choch
el
groun
d

(2)

ik

saq
white
rax

green
rapbil

green-corn-ear, fresh corn-ear


i
encliti
c

hal
corn
ear

=
green corn-ear (focus on color)

kab

93

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

whipp
ed

=
sweet

94

melcocha, a type of taffy-like


candy

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

It is possible for an unpossessed compound of Class II to appear the same


as an unpossessed compound of Class lb, but these will be disambiguated
when possessed.
chochel

kenq

ground

bean

xits

inmy

beanground
groundbean

choch
el
groun
d
choch
el
groun
d

inmy

kenq

bea
ns

ken
q
bea
n

=
=

ground for growing beans


bean variety which grows along
the ground

my bean-ground
(Class lb)

my ground-beans
(Class II)

Class III compounds


Class III compounds are those in which both N1 and N2 have equal status,
neither modifying the other. Possession of the compound involves
possessing both members of the compound equally. These compounds, if
modified, must also modify each member of the compound.
alal
son

kajol
son

w-alal
my son

in-k'ajol
my son

my children

tzuul
hill

taq'a
valley

hill-valley, the spirit of a place

loq'laj

tzuul

precious

hill

holy hill-valley spirit

pohol
moon

chahima
l
star

luck

loq'laj
precio
us

children

taq'a
valley

95

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

inmy

pohol
moon

inmy

chahima
l
star

96

my luck

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Compounding and Abstract Nouns


Abstract nouns derived from either adjective or noun are implicitly first
members of a noun-of-noun construction. Abstract nouns have the form
possessor + adjective/noun base + nominalizing suffix. The possessor in
these nouns, which is obligatory, refers to the second member of the Nplus-N construction. This latter may be said to possess the quality of the
former.
xits

qeqblack

al
suffix

inmy

tz'iibleb
pen

= The blackness of my pen -ormy pen's blackness

The abstraction produced in this way may semantically shift slightly in


such a way that it may be used as a vocabulary item without referring to a
noun which possesses the abstraction. For example, "blackness" may shift
to "darkness," meaning "night." Then the obligatory possession is dropped
and an independent noun or adjective is produced. Examples: kaq
red
(adjective)
x-kaq-al

redness

(literally, its redness) Must occur as N1 in Class la


compound.

my
redness
envy
my envy

in-kaqal
kaqal
in-kaqal

Numbers
Numbers are formally nouns, since they may be possessed by Set A
prefixes. This section will examine the various uses of numbers. The
number chart provided below will provide a point of reference for the
discussion of numbers.

Numeral Roots
Numeral roots are the nuclei of the various forms of numbers. With the
exception of jun one, which may occur without affixes both as a numeral
and as the indefinite article, all numeral roots are bound forms.
1
2
3
4

jun
ka- or
kaboxkaa-

6
7
8
9

97

waqwuqwaqxaqbele-

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

o- or ho-

10

98

laje-

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Cardinal Numbers
Cardinal numbers are used to count things, excluding those things counted
with special measure words. Normally, a cardinal number phrase is number +
chi + noun, but chi may be deleted. Cardinal numbers to ten are derived from
numeral roots by suffix Vb, a suffix probably related to the plural morpheme
-eb'. Cardinal numbers above ten are combined forms. Numbers 11-19 are
represented by the numeral roots from one through nine plus -laju, a variant of
the numeral root for ten. The cardinal numbers to twenty are:
1
2
3
4
5
6
7
8
9
10

jun
kaib / kwib / kiib / wiib
oxib
kaahib
oob / hoob / oob /
hoob
waqib
wuqub
waqxaqib
beleeb
lajeeb

11
12
13
14

junlaju
kablaju
oxlaju
kaalaju

15
16
17
18
19
20

olaju / holaju
waqlaju
wuqlaju
waqxaqlaju
beleelaju
jun may

Qeqchi, as other Mayan languages, has a vigesimal system of counting, which


is well preserved in comparison with other Mayan languages. In addition to the
numbers used at the present time (up to two hundred), numbers no longer in
use which would allow the system to count as high as 1,216,000,000 are
reported by Haeserijn (1966), reputed to have been found in a 17th century
manuscript from Cahabon, Alta Verapaz.
The system works as follows: all numbers are part of a series of twenty. The first
series ends in a unit of twenty, -may, though all the other units of twenty are
denoted by the morpheme -k'aal. We can thus count by twenties up to 380:
20
40
60
80
100
120
140
160
180

jun may
kakaal
oxkaal
kaakaal
okaal
waqkaal
wuqkaal
waqxaqkaal
beleekaal

200

otuuk

220
240
260
280
300
320
340
360
380

99

junlajukaal
kablajukaal
oxlajukaal
kaalajukaal
olajukaal
waqlajukaal
wuqlajukaal
waqxaqlajukaal
beleelajukaal

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

The form for two hundred should be lajek'aal. The root -tuuk appears to mean
forty but is unused except in this word. Haeserijn (1966: 118) glosses the morpheme as 40 corn ears, but it has not been possible to confirm this gloss
elsewhere. Brasseur (1961: 169) reported a similar form for Kiche, otuk
(orthography Brasseurs), in his grammar first published in 1862, but provided
no gloss. Unpublished data at the PLFM from Nahuala-Santa Catarina
Ixtahuacan collected by Will Norman contain otuk' or ootuk' 200 (T.Larsen,
personal communication).
To count the numbers between a particular series of twenty, one counts from
1-19 in that series, followed by the series number possessed by Set A third
person singular marker. Thus, 21 is jun x-kak'aal one of the 2 x 20 series. A
sampling of other numbers:
33
76
10
5
19
1
34
3

oxlaju xkak'aal
waqlaju xkaakaal

(13 of the 2 x 20 series)


(16 of the 4 x 20 series)

hoob' rok'aal

(5 of the 5 x 20 series)

junlaju rotuuk

(11 of the 5 x 40 series)

oxib xwaqxaqlajukaal

(3 of the 18 x 20 series)

Number 400 is oqob, where qob would apparently mean eighty; o-qob
appears to be 5 x 80. It is possible to count by 400's by prefixing numeral
roots to ten and cardinal numbers 11-19:
400
800
120
0
160
0
200
0
240
0
280
0
320
0
360
0
400
0

oqob
kaboqob

4400
4800

junlaju oqob
kablaju oqob

oxoqob

5200

oxlaju oqob

kaahoqob

5600

kaalaju oqob

ooqob

6000

olaju oqob

waqoqob

6400

waqlaju oqob

wuqoqob

6800

wuqlaju oqob

waqxaqoqob

7200

waqxaqlaju oqob

beleehoqob

7600

beleelaju oqob

lajeehoqob

Intermediate numbers work the same as previously:


100

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

80
1
23
53

jun roxoqob
oxlaju xwaqxaqlajukaal
xwaqoqob

(1 of the 3 x 400 series)


(13 of the 18 x 20 series of the 6 x 400
series)

Higher numbers should work the same way, though Haeserijn goes into no
detail about the intermediate numbers. 20 x 400 (8000) is jun chuy, which
Haeserijn glosses as one sack of cacao beans, or jun tuub, which means a
pile (monton in Spanish). Brasseur's (1961: 171) form for 8000 in Kiche is
chuvy (orthography Brasseurs), which Brasseur glossed like Haeserijn: "la
talega o el costal que contenia ocho mil almendras de cacao." Modern Kiche
has chuy (official orthography) balsa de tela para guardar dinero o cigarros.
Tom Larsen (personal communication) has noted that older orthographies
often used VV for V and that -u- and -v- were often interchangeable,
indicating that Brasseur's chuvy and the PLFM's chuy are the same. Further,
Qeqchi often has V where Kiche has V, such as in Qeqchi mol, Kiche
saqmol egg; Qeqchi sam snot, Kiche tzam nose.

Ordinal Numbers
Ordinal numbers two through ten are formed by attaching the Set A third
person singular prefix to the numeral roots. The form xb'een first is borrowed
from the relational nouns (see following section); as a relational noun xbeen
means top of. Ordinal numbers for 11-19 are formed by the same Set A
possession attached to the cardinal number plus suffix -il. All ordinal
expressions for numbers above 19 use the cardinal numbers. The ordinal
numbers are:
First
Second
Third
Fourth
Fifth
Sixth

xbeen
xkab
rox
xka
ro
xwaq

Seventh
Eighth
Ninth
Tenth
Eleventh
Twelth
Etc

xwuq
xwaqxaq
xbele
xlaje
xjunlajuhil
xkablajuhil

Distributive numbers
Distributive numbers are those which in English are expressed by each, every,
or each and every plus a cardinal number. The system as currently used
appears to represent a combination of two former systems, one involving
reduplication of the numeral root and another involving the morpheme -taq.
This -taq loses its vowel word-finally; it is doubtless related historically to the
Poqomchi plural, taq.
101

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

each one
each two
each three
each four
each five
each six
each
seven
each eight
each nine
each ten
each
eleven
each
twelve
etc.

junjunq
kakab
oxox
kaaka
ootq
waqitq

(junqal: Coban junjunqal: Chamil)


(kakabil: Chamil)
(oxoxil: Chamil)
(kaakahil: Chamil)

wuqitq
wajxaqitq
beleetq
lajeetq
juntaqlaju
kabtaqlaju

102

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Groupative numbers
There are three other formations used to denote groups of a certain
number.
Unspecified Group:

Cardinal Number plus -al.

In order to indicate a group of from two to ten, without specifying overtly what
the group is composed of, the cardinal number plus suffix al is used.
li
the

b'isok
tomeasur
e

kaam
tare
a

Approximate Group:

kwib'-al
twounspecified

naraj
itneeds
it

= To measure tareas you


need two.
(A "tarea" or "cuerda" is a
land measure of about 25
yards square.)

Cardinal Number plus -aq.

In order to indicate a group composed of an approximate number of things, the


cardinal number plus -aq is used. These numbers correspond to the indefinite
article.
Twaj

raj

I-want-it

would

Definite Group:

kwib'-aq
abouttwo

in-moos
my-worker

I'd like about two


workers.

Numeral Root plus -ichal.

In order to indicate a group of a certain number already established


between speaker and hearer, the numeral root plus -ichal, possessed by Set
A third person singular prefix, is used. This form in some ways parallels the
use of the definite article, li.
xinkayi
I-sold-it

xkabichal
the-two

eb
plural

li tzI.
the dog

I sold the two dogs.

This form is valid in Coban through number 4, in Chamil through number 6, in


Chamelco through number 7, and in Carcha through number 10. In Coban
number 3 may be doubly possessed: x-r-oxichal = xroxichal.
Cofradia Members: Ordinal Numbers plus il.
A special construction is used for members of cofradas, which are religious
brotherhoods found throughout the area populated by speakers of Mayan
languages and elsewhere (cf. Colby and van den Berghe, 1969). The ordinal
numbers plus the suffix -il are used to denote members who have taken on a
"cargo for a given year. Each cargo entails certain responsibilities, those with
lower numbers being more important than those with higher numbers.
beenil
xbeenil

cofrada
first mayordomo (with

xwaqil mertoom
xwuqil mertoom

103

sixth mayordomo
seventh

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

mertoom
xkabil
mertoom
roxil mertoom
xkaahil
mertoom
roil mertoom

the most
responsibilities)

mayordomo
xwaqxaqil
mertoom
xbeleehil
mertoom

second mayordomo
third mayordomo
fourth mayordomo
fifth mayordomo

eighth mayordomo
ninth mayordomo

xlajehil mertoom

tenth mayordomo

These terms may also be used to refer to the twelve apostles. Extra forms are
xjunlajuhil 11th apostle and xkab'lajuhil 12th apostle.
Measures and numbers
Qeqchi, as other Mayan languages, has a system of measures, often called
numeral classifiers (Day, 1973: 59-61; Kaufman, 1971: 91-93). These measure
words are used either with the numeral roots from one to five and with the
cardinal numbers over five to count certain objects and quantities, or with the
cardinal numbers only. It may be that the system of using the numeral roots
with measures is in the process of being replaced by the use of the cardinal
numbers only.
The measures that may occur with the numeral roots are taken from various
sources. A large number of measures are derived from positional (p) or transitive (vt) roots with the suffix Vl l:
mochmochol
kamochol
bisbisil
oxbisil

(p)

(vt)

in the shape of a fist


fistful
two fistfuls
to measure
a measure
three measures

Occasionally the roots are used without the V1l suffix:


tzoltzol
katzol

(p)

jach-

(vt)

yijach
oxjach

in rows
row
two rows
cut in parts (Numeral roots with this vt root produce
fractions)
one-half (yi = middle)
one-third

Another highly productive source of measures are the intransitive infinitives of


transitive verbs and other intransitive infinitives.
oxloqok

oxchiilan
k

three buys

104

three scoldings

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

oxboq
ok
oxkatok

three calls

oxaqink

three weedings

three burnings

oxkalek

three field clearings

Other measures used with numeral roots are the following:


-sut
-wa

occurrence
occurrence

yok(b)
kutub

105

step
handspan (cloth, wood)

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Chart XIV: Number chart


Indefini
te
Group

Cofra
de/
Apostl
e

junaq

xb'eenil

xkab'ichal

xkwib'aq

xkab'il

oxib'-al

roxichal

oxib'aq

roxil

kaaka(il)

kaahib'-al

kaahichal

kaahib'aq

xkaahil

r-o'

o'otq

oob'-al

ro'ichal

hoob'aq

ro'il

waqib'

x-waq

waqitq

waqib'-al

xwaqichal

waqib'aq

xwaqil

wuq-

wuqub'

x-wuq

wuqutq

wuqub'-al

xwuqichal

wuqub'aq

xwuqil

waqxaq-

waqxaqib'

x-waqxaq

waqxaqitq

waqxaqib'al

xwaqxaqic
hal

waqxaqib
'aq

xwaqxa
qil

b'elee-

b'eleeb'

x-b'ele

b'elleetq

b'eleeb'-al

xb'elehich
al

b'eleeb'a
q

xb'elehi
l

10

lajee-

lajeeb'

x-laje

lajeetq

lajeeb'-al

xlajehichal

lajeeb'aq

xlajehil

junlaju

x-junlajhil
(all but
Cobn;
Cobn =
cardinal
numbers)

juntaqlaju
(to 19,
then
cardinal
numbers)

junlajuhal
(11+ rare;
Chamil
only,
others
cardinal

Cardinal
numbers

junlajuha
q

xjunlaju
hil

Root &
w/Measu
re

Cardina
l

Ordinal

Distribut
ive

Unpseci
fied
Group

jun

jun

x-b'een

junjunq(al)
/
junqal

junesal

ka'-

kwib'/wiib
'/
ka'ib'

x-kab'

ka'kab'(il)

ka'ib'-al

ox-

oxib'

r-ox

oxox(il)

kaa-

kaahib'

x-kab'

o'-

o'ob'/hoo
b'/
oob'

waq-

11

junlaju

et
c

106

Definite
Group

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Prepositional nominals (relational nouns)


In order to introduce the prepositional nominals, also called relational nouns
(Kaufman, 1971), that small set of nouns in Qeqchi and other Mayan
languages so important to expressions of case, it is necessary to introduce
here the prepositions. The two important Qeqchi prepositions are chi at, to,
with and sa in, at, to, per. These prepositions may occur in simple
prepositional constructions involving only one or the other of these
prepositions and a noun serving as "object" of the preposition. The
preposition sa is the productive locative/temporal preposition in simple
prepositional constructions, and is the product of historical semantic
extension of the common noun sa meaning stomach, insides.
wan
he-is

sa
at/in

li santiglees
church

He is at/in church.

xinkul
I-received-it

sa
on

lunes
Monday

I got it on Monday.

While sa is largely limited to temporal and locative meanings in simple


prepositional expressions, chi has a much broader range of significance which
includes locative/temporal, instrumental, and manner.
xko

chi

junaq

he-went-away

for

one

naeekan
it-cries

chi
at

q'eq
black

xchal

chi

najt

he-came

from

far

wa
time

he left once and


for all.

It cries at night.

he came from far


away.

The use of chi and sa in the formation of place names will not be treated
here.

107

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Complex prepositional expressions


Complex prepositional expressions are those expressions which may or may not
contain one of the
above prepositions but which do contain a prepositional nominal. The
prepositional nominal, according to Freeze (1976: 115) usually names a specific
spatial or temporal position relative to the semantic prepositional object or
which names a part of that object. That nominal is formally possessed by the
prepositional object. Analogously, in the English prepositional phrase, on top of
the table, the noun top is the object of the preposition on and is formally
possessed by table. Table is in the genitive: of the table, a paraphrase for
which is tables, which plays a part in the paraphrase of the above: on the
table's top.
In the above example the role played by top in the prepositional phrase is
quite similar to the prepositional nominals in Qeqchi. Thus, often the
complex prepositional expression consists of one of the rather general
prepositions, a prepositional nominal with more specific temporal or locative
information, and the prepositional object which formally possesses the
prepositional nominal:
sa
at

x- b'een
its- top

meex
(the)
table

on top of the table

Complex prepositional expressions deal not only with the "concrete" or "local"
case distinctions illustrated above, but also with the "abstract" grammatical
case functions (Lyons, 1969: 295-302). In many cases a particular
prepositional nominal may perform both concrete and abstract case
functions; in some cases the distinction between these two may be rather
small.

108

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

Individual prepositional nominals


The following examples demonstrate each prepositional nominal as used in phrases and sentences, both with
and without the prepositions chi and sa when pertinent.

x been
LOCATIVE
sa

x-

been

kabl

at

its

over

house

chi

x-

been

li

ninqe

at

its

after

the

fiesta

above/over the house

after the fiesta

TEMPORAL

r ubel
LOCATIVE
chi
at

rits

ubel
under

li
the

tem
bench

under the bench

chi
at

rits

ubel
before

li
the

ninqe
fiesta

before the fiesta

chi
at

rits

u
surface

tzak
wall

xinixket
he-hit-me

chi
at

rits

u
surface

TEMPORAL

ru
LOCATIVE

109

che
tree

on the wall

he knocked me against the tree

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

TEMPORAL
chi
at

rits

COMPARATIVE

u
during

more/very big

aawoq
yourfoot

mas sa
more/very tasty

nawaak
he-eats

mas nim

li
the

chi

w-

ninqe
fiesta

during the fiesta

at

my-than

chi
at

w- u
my-than

You are taller than me.

He eats more than me.

BENFACTIVE
usgood-

it

chi
at

qour-

u
for

its good for us.

chaabil
nice- she at

chi
at

wu
my- for

To me, shes nice/ I like her.

Inka
natehon
no it-opens

chi
at

wu
my- for

It wont open for me/I cant open it.

xtehon
itopened

chi

aaw- u

at

your- for

naxik
He-runs-away

chi
at

wu
my- from

maq

chi

take-it

at

sa

in- kaba

at my-name/intervention

li
the

aa- kaax
yourbox

The box opened for me.

ABLATIVE

r-

his- from

re naq
so

inka

that

no

tixlow
he-willeat-it

He runs away from me.

Take it from him so that he wont eat


it.

ACCUSATIVE
naxnaw
he-knows-it
xwajsi

w- u
myacc.
r-

110

He knows me.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

hisacc.

I-woke-it-up

I entertained him.

re
LOCATIVE
chi
at

re
its-edge

nima
river

PURPOSIVE

= on/along the river bank

r- e naq
its- for that
not

inka
taachaqiq

xbaan

it-will-dry

its-act

tijok chi w- ix
praying at
my-for

r- e
maaka
its-for
nothing

taaoq
it-willenter

w- e
mydative

ak

xye

r- e

laj

already

he-said-it

his-dative

hon.

Mon
Raymo
nd

tintaqres i
I-will-water

li awimj
the plants

DATIVE
yookeb

chi

They-are-doing

at

li
saqe.
the
sun

Ill water the plants so that they dont dry up in


the sun.

They are praying for me so that nothing will


happen to me.

he already told it to Raymond.

POSSESSIVE
wmy-

e
possessive

li
the

tzi
dog

li chakach

wan
arin
there-is
here

r- e
herpossessive

li

the basket

= the dog is mine.


in-

the

na

my- mother

The basket here is my mothers.

r- ikin
LOCATIVE
taayaabaq
He-will-cry

r- ikin
its- near

li
the

aaw-ochoch
your-house

ASSOCIATIVE

111

= Hell cry near your house.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

ma
question

wan
there-is

tumin
money

aaw- ikin
your-with

= Do you have any money with you?

BENEFACTIVE
tinkanjelaq
I-will-work

aaw- ikin
your-with

= Ill work for you.

INSTRUMENTAL
nintooni
I-rent-it

li
xoral
the
lot

r- ikin
its- for

xinixket
he-hit-me

r- ikin
its- with

che
stick

sa tzalam
to jail

r- ikin
its- for

oxib ketzal
three Quetzal

= I rent the lot for 3 Quetzals.

= he hit me with a stick.

RESULTATIVE
xko
he-went

kalaak
getting-drunk

= he went To jail for getting drunk.

r - ix
LOCATIVE
chi
at

r- ix
its-outside

chi
at

q- ix
our-back

kabl
house

outside the house

behind us

after the fiesta

I wish I had some money


Id go with you to
Antigua.

TEMPORAL
chi
at

rits-

ix
after

li
ninqe
the fiesta

COMITATIVE
wan
there-is

ta
wan
doubt there-is

in- tumin --- tinxik


my-money
I-will-go

raj
chi
would at

112

eerix
your- with

Antigua
Antigua

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

x yi
LOCATIVE
sa
at

xits-

yi
between

li
the

che
trees

between the trees

x yanq
LOCATIVE
sa
at

eeyalls-

yanq
among

among yall

najt
far

qaour-

yanq
among

We live far apart.

x katq
LOCATIVE
chi
at

x
his-

katq
side

beside him

r ib
REFLEXIVE
kixkut
he-threw-it

rhis-

ib
self

sa
at

xher-

qour-

ib
self

been
over

he threw himself over her.

We dislike each other.

He went to jail for getting drunk.

RECIPROCAL
xiik
ugly

naqil
we-see-it

x baan
RESULTATIVE
xko
He-went

sa
to

tzalam
jail

x- baan
its-for

113

kalaak
getting drunk

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

nawulak
it-pleases

chi
at

ru
trabaajik
his-for
working

sa ortaliis x baan
in

garden its-for

naq nakehe

x-tumin

that it-is-given

his-money

114

He likes to work in the garden


because he makes money at it.

QEQCHI GRAMMAR STEPHEN O. STEWART, PH.D.

x maak
RESULTATIVE
xko
He-went

sa tzalam
to jail

x- maak
its- for

kalaak
getting drunk

He went to jail for getting drunk.

We found a replacement for


mayordomo (of the cofrada).

Give me the red one instead of the


black one.

r eeqaj
SUBSTITUTIVE
xqataw
We-found-it

q eetaj
our-substitute

ke

we

give-it

my- dative

chi
for

li

mertoomil
mayordomo

kaq chi

the red

at

ok
r eeqaj
enter itssubstitute

115

li

qeq

the black

CHAPTER 6: STATIVES

Statives are those stems which, as the


includes nouns possessable by Set A as well as
predicates of stative sentences, can be
statives which can normally not be possessed. The
inflected by Set B pronominal suffixes to form
concern of this chapter is not to place stems into
verbless, equative sentences in which Set B
rigid classes, but to describe stems as they interact
functions as the theme or subject of these
with Set B in stative sentences.
sentences. This class of morphemes is not
exclusive like the previous classes in that it
This section deals with two aspects of statives. The first of these is to identify the subclasses of stative stems,
specifically those which are non-nominal. These include adjectives, positionals, and a variety of participles. The
second aim of this section will be to examine how tense, aspect, and mood are marked in stative sentences, thus
complementing the exposition of these markers with verbs above.

Types of statives
Statives may be divided into two basic types: those in which the stative predicate basically identifies the theme,
in which case the stative is a nominal, and those in which the stative predicate describes the state or condition
of the theme, in which case the stative will be an adjective, participle, or positional. The distinction between the
two types is semantic and unmarked in Q'eqchi inflection, but it is mentioned here in partial justification for
treating nouns apart from other statives.

Adjectives
Adjectives were shown briefly, in the chapter on nouns, as modifiers of nouns, some of which require an
enclitic -i or -la when preceding the noun they modify.
saq

white

saqi tzi

nim

big

nimla tzi

white
dog
big dog

It is unclear what governs the use of the enclitic. In a list of thirty of the most common adjectives, sixteen
took the enclitic. Among those which did not were included all of the adjectives which ended in a vowel, of

which there were four. Further study will be necessary to determine how the enclitics function, with special
emphasis on the eastern dialects where these enclitics are more common.
Adjectives may be prefixed by the Set A pronominal prefixes under certain conditions. A noun phrase which
includes personal possession plus a modifying adjective may prefix the personal possession on the adjective preceding the noun.
Li
The

chinaa
l
boy

xk am

chaq

oxib'

in-saqi

t'ikr

brough
t

there

three

mywhite

cloth

= The boy brought me three white cloths.


In addition, a possessed noun phrase in which the noun has been deleted retains the Set A pronominal prefix
with the adjective.
Li chinaal xkam chaq
oxib

in-saq.
my-white
(ones).

= The boy brought me three


white ones.

Adjectives as statives
Adjectives, as statives, may be defined as stems inflected with Set B suffixes to form stative sentences. In
order to distinguish them from nouns, which identify the theme of the stative sentence, we may further define
the adjectives as predicates which describe the condition or state of the Set B theme. This stative adjective
class includes participles and positionals, derived from verb stems and positional roots, respectively, as well as
root adjectives and adjectives derived from another source.
adjective predicate
(condition):

yaj-

in

sick

(Set B) I

= I'm sick

noun predicate
(identity):

winq-

in

man

(Set B) I

= I'm a
man

Participles
A participle is a word derived from a verb and used as an adjective. The three participles thus derived in
Qeqchi fit this definition in that they may be used, both as modifiers of nouns and as predicates in stative
sentences. The following examples show the three types of participles as noun modifiers:
(1
)

(2
)

(3
)

Xkam

li

winq

b'ak'b'o

Hedied

the

man

tied-up

Xkam

li

winq

b'ak'b'il

Hedied

the

man

tied-up

Xkam

li

winq

warenaq

Hedied

the

man

sleeping

The tied-up man


died.

= The tied-up man


died.

(warjenaq in
Chamil)

= The sleeping
man died.

Transitive participle: bil


This suffix productively derives participles from transitive stems. They are passive in that, in a stative
sentence containing one of these participles, the Set B theme is understood to have suffered the action of the
verb as a patient. In order to distinguish between the passive verb constructions and participles, it is handy to
mark them [+/- action] and [+/- resultant state]:
(1
)

(2
)

x-

in-

sake

past

Set B
I

passive
stem

sakbil -

-in

hit

Set B

[+ action, resultant state]

= I was hit.

[+ action, +
resultant state]

= I am hit
(because I
was hit).

(participle)

In (1), which is the passive of the transitive stem sak hit, the action of hitting is stressed and the resultant
state is ignored. In (2) emphasis is on the state which results from having been hit, though it is understood
that the action must have taken place as opposed to being purely state. The derivation of these participles
differs between vtR transitive stems and vtD transitive stems. The vtR stems simply suffix bil to the root to
form the participle. As mentioned in the section on verbs, participles formed from vtD stems require that the
stems be in their absolutive form, vtD stem + -n, before adding the suffix bil. This -n often assimilates to -m
before -b'.
kayi

-n-

-bil

vtD stem
sell

absolutive
suffix

participle
derivational
suffix

kayinbil
/kayimbil/
sold

Stative participle with vtR: -C1o


This suffix exists only with vtR stems and there is no counterpart for most vtD stems, which have only -b'il
participles. The suffix is formed by reduplication of the initial consonant of the root + -o or, if the root vowel is
-u-, the suffix vowel is -u. For vtR stems this derivation provides a further possibility in terms of the factors of [+/action] and [+/- resultant state] mentioned above.
bak

- bo -

vtR stem
tied up

stative
participle

Set B
he

[- action, + resultant
state]

He is tied
up.

In this example, in contrast to those given in the previous section, the notion of the action required to
produce the state is absent and only the notion of the state is present. The only vtD stems which have a
corresponding form are those derived from positional roots. It appears that these participles may be recent
innovations by analogy with the positionals, especially since the deriving suffix -C1o is identical.
Intransitive participle: - (j)enaq
This suffix productively derives participles from intransitive stems. In terms of meaning these participles differ
from the bil participles derived from transitive stems in their lack of the notion of passive. Thus the Set B
theme or subject of a stative is not the underlying patient of a transitive construction, but rather the underlying

subject of an intransitive construction which might be either agent or patient. If, however, the subject is patientlike, there is no underlying notion of an agent acting upon the patient.
wartesi -

- mbil -

- in

vtD stem

ppl
derivational
suffix

Set B

- jenaq ppl
derivational
suffix

- in
Set B

put-tosleep
war
viR stem
sleep

= wartesimbi I am/was put to


lin
sleep
(result of an action
that someone did)

I
= warjenaqin

I am/was asleep
(not the result of
anothers action)

Positionals
While the other major word classes are fairly standard and traditional, needing little definition or
explanation, positionals are characteristic of Mayan languages and require some elaboration as to their
semantic and derivational characteristics. Positionals are a class of bound roots which must be derived to
form stems or words. As a root class the positionals thus contrast with other root classes such as vtR, viR,
and those nouns and adjectives that are underived roots. Stems derived from positionals may be
transitive, intransitive, nominal, or adjectival; some of the derivational suffixes employed are unique to
positionals.
sir sir -

-V ba

sirib

sir -

-laa

sirla

sir sir -

+
+

-V1l
-C1o

=
=

siril
sirso

(positional root)
(transitive
stem)
(intransitive
stem
(noun stem)
(adjective stem)

discoid
to make something
discoid
to become discoid
disc-like thing
discoid

The semantics of positionals are distinctive. The positional roots may be said to "describe the position, form, or
state of an object, and imply absence of movement" (England, 1975: 223). The transitive stems derived from

positional roots refer to making something of the shape described by the positional root or putting something of
that shape somewhere. Intransitive stems derived from positional roots refer to something becoming the form
described by the positional root. Nominal stems derived from positional roots are objects in the form or in the
position described by the positional root. Adjectival stems from positionals indicate that an object has the
position or form, or is in a state, described by the positional root. These last, as adjectivals occurring in stative
sentences, are the concern of this chapter. In order to give some idea of the semantic range a list of examples
in adjectival form is given below.
baqbo
barb'o
baqbo
bechbo
beqbo
betbo
birbo
buchbu
buqbu
butbu
buybu

(describes something) soft plopped on the ground, like cow shit


cylindrical and thin, like a pencil or like dough rubbed between the palms
ovoid, as an egg or potato
uneven, not straight, like a misbuttoned shirt or trousers badly put on
bug-eyed
with the tongue stuck out
(describes something) soft and skinny snapped in two, like a string bean
crouched or sitting, like a sleeping cat or a hen laying
bumpy, like a lump on the head from being hit
full of a liquid, like a jug
piled up, dry grain-like things like corn or sand

Positional adjectives
The derivational suffix which produces a positional adjective is composed of the reduplicated initial consonant
of the root plus -o or, if the root vowel is -u-,. -o becomes -u. It should be noted that this suffix is identical to
that suffix which derives a stative participle from a vtR transitive stem, which form has the same properties
as the positional adjectives: [- action, + state]. A few roots appear to be bi-valent, belonging both to vtR and
positional classes with no change in meaning, such as but -, while others such as baq- occur both as
positionals and as vtR roots, but without the same meaning:

an
d

butbu
butx-butub
but

(positional adjective or vtR stative participle)


(positional root)
(inflected transitive stem from positional)
(vtR root = imperative)

full
full
he filled it
fill it!

x-but
baq-

(inflected vtR stem)


(positional root)

xbaqab
baqbo

(inflected transitive stem from positional)


1- (positional adjective)
2- (vtR stative participle)

he filled it
soft, plopped like cow
shit
he plopped it
plopped
twisted like thread

baq
x-baq

(vtR root = imperative)


(inflected vtR root)

twist it!
he twisted it

Tense, aspect and mood with statives


In the section on verbs, it was briefly mentioned that the inflectional suffixes k non-future and
- (a)q future are used to denote notions of tense, aspect, and mood in stative sentences. In addition to these
suffixes there are also time adverbials and particles which provide additional tense information with all statives.
Two of the types of statives, those which have either the shape C1VC -C1o, or C1VC-C1u if V = -u, regardless of
whether derived from vtR roots or positional roots, are more complicated as regards tense in that these stems
have two additional suffixes, formed seemingly from -k and -(a)q, and also in that in towns other than Coban,
they may be inflected as intransitive verbs.

Non-perfect stative

This section will deal with notions of tense and mood which are aspectually non-perfective. That is, as is
traditionally accepted, non-perfective in the past tense indicates that the state or condition occurred in the
past but whether the state or condition terminated is not indicated nor when.
Spanish:

cocinaba
cocin

I cooked/I was cooking (non-perfective)


I cooked/I did cook (perfective)

Non-perfective with the future in Qeqchi is less common. Here when this aspect occurs in the future it will be
called non-perf because it parallels or complements the non-perfective of the past tense: in the past tense
non-perf indicates that whether the state ends or not is not known. In the future the non-perf indicates that the
beginning of the state is not known. The following chart provides a key to the non-perfective with statives.

Non-perfective statives and tense, aspect and mood


Remote
Past

Recent
Past

chiilambilin chaq

bakbookin chaq

yajin chaq

atijenaqin chaq

I was scolded

I was tied up

I was sick

I was bathed

chiilambilin
+ time adverb
I was scolded

bakbookin
+ time adverb
I was tied up

yajin
+ time adverb
I was sick

atijenaqin
+ time adverb
I was bathed

Present

chiilambilin
I am scolded

bakbookin
I am tied up

yajin
I am sick

atijenaqin
I am bathed

Future
optative
of doubt

chiilambilaqin
+ mare / ta na
I doubt Ill be scolded

baqbooqaqin
+ mare / ta na
I doubt Ill be tied up

yajaqin
+ mare / ta na
I doubt Ill be sick

atijenaqaqin
+ mare / ta na
I doubt Ill be bathed

Future
Optative

chiilambilaqin
I hope Ill be scolded

baqbooqaqin
I hope Ill be tied up

yajaqin
I hope Ill be sick

atijenaqaqin
I hope Ill be bathed

Future

No form
(= passive)

bakbooqin
Ill be tied up

yajaqin
Ill be sick

atijenaqaqin
Ill be bathed

Present tense: -k and


The present tense in stative sentences is denoted by with adjectival statives, -b'il transitive participles, and (j)enaq intransitive participles. With stative participles and positionals, both of the shape C1VC -C1o, present
tense is denoted by with Set B third person singular (also ) and by -k with non-third singular. Examples are:
adjective

yajsick

present

Set B he

He is sick.

chiilambilscolded

present

Set B he

He is scolded.

intransitive ppl

atijenaqbathed

present

Set B he

He is bathed.

stative ppl

bakbotied up

present

Set B he

chunchusitting

present

Set B he

kpresent

in
Set B I

I am tied up.

(1 singular)

bakbootied up

positional

chunchuu-

kpresent

in
Set B I

I am sitting.

transitive ppl

(3

rd

singular)

positional
(3

rd

singular)

stative ppl
st

(1st singular)

sitting

He is tied up.

He is sitting.

Future tenses: -(a)q, -kaq, -qaq


For statives which are adjectives, transitive participles, or intransitive participles there is only one future, -(a)q,
which is always realized as -aq.
yajsick

aqfuture

in
I

I will be sick.

chiilambilscolded
atijenaqbathed

aqfuture

in
I

I will be scolded.

aqfuture

in
I

I will be bathed

For statives which are stative participles or positionals there are three futures, differing in mood. Suffix -(a)q,
realized as -q, is indicative and states what is believed to be true. Suffix -kaq is future optative, stating a desire
on the part of the speaker relative to the future state. Suffix -qaq, which must occur with a particle expressing
doubt, such as maare maybe or ta na maybe (two particles together), expresses doubt on the part of the
speaker relative to whether the future state will occur. These moods with other statives are all expressed through
the single suffix noted above or, as will be shown, use verbs. Examples of these suffixes are given below.
Future
indicative

bakbootied up

qfuture indicative

-in
Set B I

I will be tied up.

Future
optative

bakbootied up

kaqfuture optative

-in
Set B I

May I be tied up.


(I hope so)

bakbootied up

qaqfuture doubt

-in
Set B I

Future doubt

I might be tied up.


(I doubt it, though.)

Recent past: raj (Coban), time adverb (others)


Recent past in stative sentences, as in sentences with finite verbs, refers to a time period no more distant
in the past than yesterday. Recent past is marked differently in Coban than in the other towns in the dialect
area. In Coban raj, a separate particle used in conjunction with the normal present tense forms, marks past
tense, both the recent past here discussed and the remote past mentioned below. In the other towns the
normal present tense
stative forms are used in conjunction with a time adverb that specifies recent
past, i.e., yesterday, a minute ago, this morning, and so on.
Coban:
Other:

yajin raj ewer


I was sick yesterday.
yajin ewer I was sick yesterday.

Remote past: raj (Coban), chaq (others)


As mentioned in the previous section the addition of the particle raj to the normal present tense forms
produces the Coban past tense forms with only the addition of a time adverb to distinguish between recent
and remote past.
Recent:

yajin
sick-I

Remote: yajin
sick-I

raj
ewer
past yesterday
raj
li
chihab
past the year

I was sick yesterday.


xnume =
it-passed

I was sick last year.

In the other towns in the dialect area the normal present tense forms are used in conjunction with a time
adverb and the particle chaq. Chaq in other contexts is locative or directional, indicating either that the
action occurs from there to here, hither or that the action occurs in some unspecified there (DeCormier,
1977).
Recent:

yajin

Remote:
Past

yajin

ewer

=
chaq

li

chihab

xnume.

I was sick yesterday.


=

I was sick last year.

Perfective statives: Intransitive verbs


Perfective, as used, with the past tense, indicates that the action or state terminates. Perfective with a
future will mean here that the beginning of the action or state occurs in the future. Perfectives, then, express actions or states which have a definite end or beginning as opposed to non-perfectives or
imperfectives, which do not express these notions.
However, a search for perfective aspect with statives leads to intransitive verbs. In other words, perfective
aspect was found to be peculiar to verbal forms which, in describing action as opposed to a state or
condition, may be thought to be more concerned with the beginning or end of the action or state. Support
for this argument arises from the fact that, in order to express perfective plus stative, it is necessary to
derive in some manner a verb from the stative. In the case of straight adjectives, the derivation is direct:
[1]

yajsick

-er
derivational suffix
yajin

ewer

yajervir stem

get sick

I was sick yesterday.

xinyajer
[2]

kaqred

ewer

I got sick yesterday.

-o
derivational suffix

kaqo-

kaqin ewer
xinkaqo ewer

=
=

redden

I was red (i.e. jealous) yesterday


I got jealous yesterday.

In the case of stative participles C1VC-C1o, which were noted above as being an irregular subset of viV
intransitives, the stative participle is simply inflected as if it were in fact an intransitive verb stem of class
viV:
b'ak'b'oostative stem
xpast

knon-future

in
Set B

ewer
yesterday

I was tied up yesterday

inSet B

b'ak'b'o
viV stem

ewer
yesterday

I got tied up yesterday

In the case of the intransitive participles, the derivation may be characterized as being reverse, in that, to
express the perfective, the participles simply revert back to the intransitive verb from which they were
derived originally:
atin- (bound stem)

bathe
atijenaqintrans. ppl.

in
Set B I

ewer
yesterday

I was bathed yesterday.

xpast

inSet B

atin
bathe

I (got) bathed.

In the case of the transitive participles the derivation occurs in two steps: (1) the participle stem reverts to the
original transitive stem, and (2) the transitive stem is passivized:
chiilambiltransitive

-in
Set B I

ewer
yesterday

= I was scolded yesterday.

participle
chiilachiilavtD stem

xpast

-a

chiilaaviV stem
(passive)

chiilaaviV stem

-k
non-future

derivational
suffix
inSet B I

(vtD stem)

= scold
= be scolded

ewer
yesterday

=
I was (or got) scolded yesterday.

The chart below summarizes the above findings and places them in perspective by including the data from
the previous chart. Placing statives, defined by their particular inflection, with intransitive verbs, also
defined by a distinct inflection, requires some further explanation. Informants generally characterized forms
like chiilambilin and xinchiilaak as meaning the same thing, since their Spanish glosses were the same:
fui reganado.
It was only after a good deal of reflection on their part that xinchiilaak was further specified as being I was
scolded (but the action or state ended, i.e., I am not now scolded) and ch'iilamb'ilin + time adverb as being I
was scolded (and it is uncertain whether I am still in this state or not).
This distinction holds for each of the cases in the past tense (see chart next page) where the only distinction
between the stative and verbal forms is whether or not the action was known to have ended. In looking at the
optative and future tenses the difference in meaning between the stative and verbal forms is whether or not
the beginning of the action, as opposed to the end, was known or expressed. Thus, yajaqin means I might be
sick (optative doubt), I hope I'll be sick (optative), or I will be sick (future), but in each of these cases there
exists the possibility that the speaker is sick right now. In the verbal forms, chinyajerq Hope Ill get sick and
tinyajerq Ill get sick, it is understood that the action or state has not yet begun but must rather begin at some
point in the future.
The forms given in the present tense present a slightly different explanation. An example was offered by the
informants using xaqxo standing up, a positional adjective.

Remote past
(perfective)

Remote past
(imperfective
)
Recent past
(perfective)

Recent past
(imperfective
)

Present
habitual
(perfective)
Present
imperfective
Optative

kinchiilaak
(= passive
viV)
I was scolded
(but not
now)
chiilambilin
chaq
I was scolded
(maybe still
am)
xinchiilaak
(= passive
viV)
I was scolded
(but not
now)
chiilambilin
+ time
adverb
I was scolded
(maybe still
am)
ninchiilaak
(= passive
viV)
I am scolded
(always)
chiilambilin
Im scolded
(now)
chiilambilaq

kinbakbo
(irregular
viV)
I was tied up
(but not
now)
bakbookin
chaq
I was tied up
(maybe still
am)
xinbakbo
(irregular
viV)
I was tied up
(but not
now)
bakbookin
+ time
adverb
I was tied up
(maybe still
am)
ninbakbo
(irregular
viV)
I am always
tied up
bakbookin
Im tied up
(now)
bakbooqaqi

kinyajer
(= versive)
I got sick
(but not
now)
yajin chaq
I was sick
(maybe still
am)
xinyajer
(= versive)
I got sick
(but not
now)

Chart XV
kinatin
(= normal
vin)
I bathed
atijenaqin
chaq
I was bathed
(maybe still
am)
xinatin
(= normal
vin)
I bathed

yajin + time
adverb
I was sick
(maybe still
am)

atijenaqin +
time adverb
I was bathed
(maybe still
am)

ninyajer
(= versive)
I always get
sick

ninatin
(= normal
vin)
I always
bathe
atijenaqin
Im bathed
(now)
Atijenaqaqin

yajin
Im sick
(now)
yajaqaqin +

doubt
(imperfective
)
Optative
(imperfective
)

Optative
(perfective)

Future
(imperfective
)

Future
(perfective)

in + maare
or ta na
Doubt Ill be
scolded
chiilambilaq
in
Hope Im
scolded
(maybe I am
now)
chinchiilaaq
(= passive
viV)
Hope I am
scolded (but
Im not yet)
chiilambilaq
in
Ill be
scolded
(maybe I am
now)
tinchiilaaq
(= passive
viV)
Ill be
scolded (but
I havent yet)

n + maare or
ta na
Doubt Ill be
tied up
bakbookaqi
n
Hope Im tied
up (maybe I
am now)

maare or ta
na
Doubt Ill be
sick
yajaqin
Hope Im sick
(maybe I am
now)

+ maare or
ta na
Doubt Ill be
bathed
atijenaqaqin
Hope Im
bathed
(maybe I am
now)

chinbakboo
q
(irregular
viV)
Hope I am
tied up
(but Im not
yet)
bakbookaqi
n
Ill be tied up
(maybe I am
now)

chinyajerq
(= versive)
Hope I get
sick
(but Im not
yet)

chinatinq
(= normal
vin)
Hope I bathe
(but I
havent yet)

yajaqin
Ill be sick
(maybe I am
now)

atijenaqaqin
Ill be bathed
(maybe I am
now)

tinbakbooq
(irregular
viV)
Ill be tied up
(but I
havent yet)

tinyajerq
(= versive)
Ill get sick
(but I
havent yet)

tinatinq
(= normal
vin)
Ill bathe (but
I havent yet)

(1)

xaqxoostanding up
=

(2)

na-

Set B he

present
habitual
=
(3)

chi
r-e
non-future
Set B
at
its-mouth
he is standing up (right now) in the doorway of the house.

napresent habitual

xaqxo
standing up

chi

re li

li
the

kabl.
house

kabl.
house

he is always standing (out of habit, say) in the doorway of the house.


Set B he

xaqli
viV stem
derived from
xaq-

chi re li

kab'l.

he always stands up (i.e., gets To his feet) in the doorway of the house.

In example (1) there is no intimation presented that the man spoken about actually goes through the motions
of standing up; all that is expressed is that he is standing. In (2) there is still nothing said of the motion of
standing but just that the man habitually stands in his doorway. In (3) the man goes through the motions of
standing up, a habitual action that might occur, for example, every time someone goes by his house.

Sources and References


Ac Caal, Flora and Sandra Pinkerton. 1976. Li kalek, li a;wk, ut li qolok. IJAL Native American Texts Series, 1:1.
University of Chicago Press: Chicago.
Adams, Richard N. 1965. Migraciones internas en Guatemala. Editorial Jos de Pineda Ibarra: Guatemala.
Alvarez de Stewart, Mara del Carmen. 1976. Investigacin monogrfica y social del municipio de San Miguel
Tucur, A.V. Editorial Rosales: Guatemala.
Attinasi, John J. 1973. Lak tan: a grammar of the Chol (Mayan) word. Unpublished doctoral dissertation,
University of Chicago.
Brasseur de Bourgourg, Charles Etienne. 1961. Gramtica de la lengua Quich. Editorial Jos de Pineda Ibarra:
Guatemala.
Burkitt, Robert. 1902. Notes on the Kekchi language. American Anthropologist 4: 441-463.
Cabarrs, Carlos Rafael. 1979. La cosmovisin Kekchi en proceso de cambio. San Salvador: Universidad
Centroamericana.
Campbell, Lyle. 1974. Theoretical implications of Kekchi phonology. IJAL 40, No. 4, Pt. 1: 269-278.
_______
1976. Kekchi linguistic acculturation: a cognitive approach. In Mayan linguistics,
No. 1, ed. Marlys McClaran. American Indian Culture Center, UCLA: Los Angeles.
_______
1977. Quichean linguistic prehistory. University of California Press: Berkeley.
Carter, William E. 1969. New lands and old traditions: Kekchi cultivators in the Guatemalan lowlands. Center for
Latin American Studies, University of Florida: Gainesville.
Chafe, Wallace L. 1970. Meaning and the structure of language. University of Chicago Press: Chicago.

Colbe, Benjamin N. and Pierre L. van de Berghe. 1969. Ixil country: a plural society in highland Guatemala.
University of California Press: Berkeley.
Cowan, Marion m. 1969. Tzotzil grammar. Summer Institute of Linguistics: Mexico.
Davidson, William, L. W. Elford, and Harry Hoijer. 1963. Athapaskan classificatory verbs. In Studies in the
Athapaskan languages, ed. Harry Hoijer, pp. 30-41. University of California Publications in
Linguistics, 29: Berkeley.
Day, Christopher. 1973. The Jacaltec language. Indian University Publications, Language Science Monographs,
vol. 12: Bloomington.
DeCormier, Christopher. 1977. The Kekchi particle aq. Unpublished manuscript.
Eachus, Francis and Ruth Carlson. 1966. Kekchi. In Languages of Guatemala, ed. Marvin Mayers, pp. 110-124.
Mouton: The Hague.
England, Nora C. 1975. Mam grammer in outline. Unpublished doctoral dissertation, Univeristy of Florida:
Gainesville.
__________

1983. A grammar of Mam, a Mayan language. University of Texas Press: Austin.

Freeze, Ray A. 1970. Case in a grammar of Kekchi (Maya). Unpublished doctoral dissertation, University of
Texas: Austin.
__________

1976a. Possession in Kekchi (Maya). IJAL 42, No. 1: 113-125.

__________

1976b. Kekchi texts. IJAL Native American Texts Series, 1:1. University of Chicago Press: Chicago.

Haeserijn, Esteban. 1966. Ensayo de la gramtica del Kekchi. Imprenta Suquinay: Purulh, B.V., Guatemala.
__________ 1967. Estudio sobre el estado religioso del indgena de Alta Verapaz (y Purulh), Guatemala.
Mimeographed: Purulh, B.V., Guatemala.

__________ 1972. Gua para prender a hablar, a leer y a escribir el Kekchi. Mimeographed: San Juan
Chamelco, A.V., Guatemala.
__________

1979. Diccionario Kekchi Espaol. Editorial Piedra Santa: Guatemala.

Kaufman, Terrence. 1969. Some recent hypotheses on Mayan diversification. Language-Behavior Research
Laboratory: Berkeley.
__________
Berkeley.

1971. Tzeltal phonology and morphology. University of California Publications in Linguistics, 61:

__________

1974. Idiomas de mesoamrica. Seminario de Integracin Social: Guatemala.

__________ 1976. Proyecto de alfabetos o ortografas para escribir las lenguas mayances. Proyecto Lingstico
Francisco Marroqun: Antigua, Guatemala.
King, Arden. 1974. Coban and the Verapaz: history and cultural process in northern Guatemala. Middle
American Research Institute, Publication 37, Tulane University: New Orleans.
Lyons, John. 1969. Theoretical linguistics. University of Cambridge Press: Cambridge, England.
McClaren, Marllys, ed. 1976. Mayan linguistics, No. 1. American Indian Studies Center, UCLA: Los Angeles.
Norman, William. 1973. Positional and transitive verb roots in Quich. Paper presented at the 1973 meetings of
the American Anthropological Association in New Orleans.
Pinkerton, Sandra, ed. 1976. Studies in Kekchi. Texas Linguistic Forum 3, Department of Linguistics, University
of Texas: Austin.
Sapper, Karl. 1906. Spiele de Kekchi-Indianer. Boas anniversary volume, pp. 283-289: New York.
Sedat, William. 1955. Nuevo diccionario de las lenguas kekchi y espaola. Alianza para le Progreso:
Guatemala.

___________ , trans. 1961. Li ac chak rab re li Kacua Jesucristo. Sociedad Bblica en Guatemala: Guatemala.
Stoll, Otto. 1896. Di Sprache der Kekchi-Indianer. Leipzig.

About the Author


Dr. Stephen O. Stewart is a linguist and anthropologist (Ph.D. University of
Colorado, 1978), who lived and worked for 25 years in Central America. Dr.
Stewart has worked as a social scientist and evaluation professional, and is an
expert in Mayan languages and bilingual education. He became interested in
cross-cultural work first as a VISTA volunteer working with white and black
sharecroppers in Missouri, two years as a Peace Corps volunteer in Nepal, and
a year in Paris teaching English as a Second Language to French businessmen.

stephen.o.stewart@gmail.com
stephenomerstewart.com

Q'eqchi' is a Mayan language spoken by perhaps 800,000 people in north-central


Guatemala and southern Belize. In Guatemala it is spoken primarily in the provinces
(de
partamentos in the Guatemalan national terminology) of Alta
Verapaz, Izabal, and Peten.

S-ar putea să vă placă și