NON-SPATIAL SETTING IN WHITE HMONG

by
NATHAN M. WHITE

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS
in
THE FACULTY OF GRADUATE STUDIES
LINGUISTICS

We accept this thesis as conforming to the required standard

...............................................................................
Sean Allison, PhD; Thesis Supervisor

................................................................................
Kenneth Gregerson, PhD; Second Reader

...............................................................................
Ken Manson, PhD; Third Reader

TRINITY WESTERN UNIVERSITY
September 16, 2014
© Nathan M. White

Contents
Abstract ......................................................................................................................... iii
Acknowledgements ........................................................................................................ iv
Abbreviations .................................................................................................................. v
1. Introduction ................................................................................................................. 1
1.1 Source of Data ....................................................................................................... 1
1.2 Summary of Conclusions ....................................................................................... 2
1.3 Structure of Thesis ................................................................................................. 2
2. Background ................................................................................................................. 3
2.1 Summary of Non-spatial Setting ............................................................................ 3
2.1.1 Dixon’s verb classification .............................................................................. 6
2.2 Grammaticalization and morphological marking .................................................... 8
2.3 Regional Typology .............................................................................................. 11
2.3.1 Mandarin Chinese ......................................................................................... 13
2.3.2 Guizhou Dong ............................................................................................... 20
2.3.3 Green Hmong................................................................................................ 24
2.4 Basic grammar ..................................................................................................... 34
3. Grammatical elements marking Non-spatial Setting in White Hmong ........................ 45
3.1 Pre-Subject Position ............................................................................................ 47
3.1.1 Temporal shifters and other time markers ...................................................... 47
3.1.2 ntshai ‘maybe’ .............................................................................................. 50
3.1.3 tej zaum ‘maybe’ ........................................................................................... 51
3.2 Pre-modal position............................................................................................... 52
3.2.1 tseem ‘still’ ................................................................................................... 52
3.2.2 twb ‘already, indeed’ ..................................................................................... 55
3.2.3 nyuam qhuav ‘just, a moment ago’ ................................................................ 58
3.2.4 mam (li) ‘will’ ............................................................................................... 60
3.2.5 sij ‘continually, repeatedly’ ........................................................................... 65
3.2.6 kiag ‘really’................................................................................................... 66
3.2.7 tiag ‘really’ ................................................................................................... 69
3.2.8 yeej ‘already, certainly, can’ .......................................................................... 70
3.2.9 tab tom ‘begin to, IMPERFECT’ ....................................................................... 75
3.2.10 maj mam ‘slowly’........................................................................................ 78
3.3 Modal position..................................................................................................... 80
3.3.1 yuav ‘POSITIVE IRREALIS, intend’ ................................................................... 80
3.3.2 yuav tau ‘must, have to’ ................................................................................ 85
3.3.3 yuav tsum ‘must, have to’ .............................................................................. 87
3.4 Pre-main-verb position ........................................................................................ 88
3.4.1 tau ‘ATTAINMENT, can’ .................................................................................. 88
3.4.2 tsum ‘can, have to’ ........................................................................................ 97
3.4.3 txawj ‘know how to’ ..................................................................................... 98
3.4.4 txhob ‘NEGATIVE IRREALIS, should not’ ....................................................... 101
3.4.5 chiv ‘begin’ ................................................................................................. 107
3.4.6 pib ‘begin’ .................................................................................................. 108
3.4.7 nyim ‘keep -ing’ .......................................................................................... 110

i

3.4.8 pheej/pej ‘keep, continue to’ ....................................................................... 110
3.4.9 rau siab ntso ‘continue to, commit oneself to’ ............................................. 112
3.5 Post-main-verb position ..................................................................................... 113
3.5.1 Attainment Serial Verb Constructions ......................................................... 113
3.5.2 Reduplication .............................................................................................. 115
3.5.3 ntua ‘just, immediately’ .............................................................................. 117
3.5.4 taus ‘be physically/materially able to’ ......................................................... 118
3.5.5 dua ‘EXPERIENTIAL MARKER’....................................................................... 119
3.6 Phase of activity—Finishing position ................................................................. 122
3.6.1 tas ‘finish’ ................................................................................................... 122
3.6.2 tiav ‘finish’ ................................................................................................. 123
3.7 Final adverbs and particles position ................................................................... 125
3.8 lawm ‘COMPLETION MARKER’............................................................................. 125
3.9 xwb ‘indeed’ ...................................................................................................... 128
3.10 Morphemes not belonging to Non-spatial Setting ............................................. 129
3.10.1 kam ‘be willing to, have the habit of’ ........................................................ 129
3.10.2 kheev ‘be willing, be inclined to’ ............................................................... 131
3.10.3 xav ‘want’ ................................................................................................. 132
4. Non-spatial Setting in White Hmong ....................................................................... 134
4.1 Tense ................................................................................................................. 134
4.2 Reality ............................................................................................................... 134
4.3 Degree of Certainty ........................................................................................... 135
4.4 Phase of Activity ............................................................................................... 136
4.5 Completion ........................................................................................................ 136
4.6 Speed and Ease .................................................................................................. 137
4.7 Summary of system ........................................................................................... 137
4.8 Further directions of inquiry in White Hmong.................................................... 138
4.9 Typological implications for Basic Linguistic Theory ........................................ 139
5. Conclusion .............................................................................................................. 141
References ................................................................................................................... 142
Appendix A ................................................................................................................. 146
Appendix B ................................................................................................................. 158
Appendix C ................................................................................................................. 163

ii

Abstract
Dixon (2010a,b, 2012) presents an excellent introduction to a framework for
documenting a language’s grammar. One portion of this framework is the marking of
Non-spatial Setting, that is, the grammatical coding of marking of time, aspect, and other
material in the verbal system. The primary aim of this thesis is to apply this portion of
Dixon’s framework to White Hmong (Hmong-Mien, Laos) by describing the system of
Non-spatial Setting in this language. The thesis first looks at the Non-spatial Setting
systems of typologically-similar languages from the region, continues by considering
what it means for a word to be grammaticalized, and then provides a lengthy discussion
on the Non-spatial Setting system of White Hmong itself. It is found that White Hmong
possesses a robust system of Non-spatial Setting markers. These include several classes
of Lexical Time Words as well as positive and negative Irrealis marking intertwined with
a system of marking Modality, which itself includes a number of Modals and Semimodals as well as one adverb. In addition, there is a set of Degree of Certainty markers
made up of seven morphemes at two levels of certainty, high and moderate, and there is a
group of Secondary verbs (following the terminology of Dixon 2006) that mark Phase of
Activity. There are also five Completion morphemes—three of which mark distinct types
of Perfect and two Imperfect—and two Completion-marking strategies, namely, an
Attainment Serial Verb Construction (terminology following Jarkey 2004) for the Perfect
and reduplication for the Imperfect. Finally, there is one Speed and Ease morpheme that
marks slowness. Some implications that the system of White Hmong has for Non-spatial
Setting in general are also briefly discussed.

iii

Acknowledgements
First of all, I would like to greatly thank my language consultant, Jay Yang, for the
countless hours he selflessly gave for the project and continual willingness to help,
without which this thesis would have been impossible.
Second, I thank my wife Helen for her support, advice, and devotion as we managed
school, work, travel, hosting relatives, and having our first child, Stephen, during the
process of research and writing of this thesis over the last nine months. I also thank
Stephen for being such a calm, well-behaved son, making this process that much easier.
Third, I want to thank my advisor, Dr. Sean Allison, for his helpful advice, willingness to
freely discuss the thesis at any time, and his dedication to see the project through even
while in the field.
Finally, I would like to thank my thesis committee members, Dr. Kenneth Gregerson and
Dr. Ken Manson, for agreeing to take part in this process with me, their helpful
comments, and, in the case of Dr. Ken Manson, his great flexibility time-wise for the
thesis defense.

iv

Abbreviations
1DU
1PL
1SG
2DU
2PL
2SG
3DU
3PL
3SG
ADP
ATT
CLF
COMP
CONJ
COP
CP
CRS
CVB
EMPH
EXP
IEP
INDF.PRON
IP
IPFV
IPRF
IRR

MOD
NEG
NEG.IRR

NEGP
NMLZ
PC

PF
PFV
PL.CLF
POS.IRR
POSS

PostV
PPART

PreMOD
PreS
PreV
Q
RECP

first person dual
first person plural
first person singular
second person dual
second person plural
second person singular
third person dual
third person plural
third person singular
final adverbs and particles position
attainment marker
classifier
complementizer
conjunction
copula
completion particle
Currently Relevant State
coverb
emphatic marker
experiential marker
interrogative and exclamatory particle
indefinite pronoun
intensive particle
imperfective
imperfect
irrealis
modal position
negative
negative irrealis
negative position
nominalizer
Potential complementizer
phase of activity—finishing position
perfective
plural classifier
positive irrealis
possessive particle
post-main-verb position
pause particle
pre-modal position
pre-subject position
pre-main-verb position
question particle
reciprocal

v

REDUP
REL
TOP

reduplication
relative pronoun
topic marker

vi

1
1. Introduction
In his three-part work on Basic Linguistic Theory, Dixon (2010a,b, 2012) presents an
excellent introduction to a useful system for analyzing and documenting the grammar of
a language. One part of this is Non-spatial Setting, that is, the grammatical coding of
marking of time, aspect, and other material in the verbal system. The goal of this thesis is
to apply Dixon’s framework to White Hmong (Hmong-Mien, Laos) by describing the
system of Non-spatial Setting in this language. It is found that White Hmong has a robust
system of marking various elements of Non-spatial Setting, which is made up of Lexical
Time Words1,2 as Tense, Irrealis and, within it, Modality markers, Degree of Certainty
markers, Phase of Activity markers, a relatively large set of Completion markers, and one
Speed and Ease marker.
1.1 Source of Data
While a number of scholarly sources were relied on at times for examples throughout this
work, a significant amount of data has been obtained from Tzerge (Jay) Yang, a native
speaker of White Hmong originally from Xieng Khouang province, Laos who currently
lives in Fresno, California. The data gathered includes three texts—a narrative that tells
about Mr. Yang’s background, a procedural text that gives instructions on how to cook
eggs, and a hortatory text in the form of a Christian religious speech3—as well as a
number of other examples, adapted either directly or indirectly from these texts.

1

Please see below for a brief description of this term from Dixon (2012:20).
Specific terms from Dixon (2012)’s framework for Non-spatial Setting are generally capitalized to
indicate that these are being used in Dixon’s specialized sense.
3
Please see the appendices below for interlinear transcriptions of these.
2

2
1.2 Summary of Conclusions
White Hmong possesses a system of Non-spatial Setting that encompasses several of
Dixon (2012)’s categories. Among these are Lexical Time Words, markers of positive
and negative Irrealis intertwined with a system of Modality markers that includes a
number of Modals and Semi-modals and one adverb. In addition to these, White Hmong
has a system of Degree of Certainty marking that includes seven morphemes at two levels
of certainty and a set of Secondary verbs (following the terminology of Dixon 2006)
marking Phase of Activity. Furthermore, the language has a set of five Completion
morphemes, three which mark distinct types of Perfect and two Imperfect, two
Completion-marking strategies, namely, reduplication for the Imperfect and an
Attainment Serial Verb Construction4 for the Perfect, and one Speed and Ease morpheme
that marks slowness.5
1.3 Structure of Thesis
This thesis is made up of five sections. Section 1 is a brief introduction. Section 2
presents relevant background information, including a basic introduction to Non-spatial
Setting, typological patterns in the grammar of nearby languages, and a brief presentation
of the basics of White Hmong Grammar. Section 3 covers the individual words and
constructions that constitute Non-spatial Setting in White Hmong. Section 4 summarizes
the findings, suggests further directions of inquiry for Non-spatial Setting in White
Hmong, and provides a brief discussion on the implications of the system found in White
Hmong for the system as a whole, while Section 5 provides a brief conclusion.
4

Term following Jarkey (2006).
The scope for this thesis is to describe how Non-spatial Setting is indicated in White Hmong and what
role these markers play at the verb/clause level, with brief reference to sentence level (e.g., in regard to
perfective-imperfective relationships, see below), as is relevant. The role that the Non-spatial Setting
markers found in this thesis play at the sentence and discourse level is thus generally left for further
research.
5

3
2. Background
This section is divided into three subsections, dealing first with a presentation of the
basics of Dixon (2012)’s Non-spatial Setting, then continuing with a discussion on what
constitutes genuine morphological marking apart from normal lexical items. This is
followed by a discussion of a cross-linguistic typology for Southeast Asian languages,
which includes a discussion on the Non-spatial Setting systems of three languages from
the region. The section then finishes with a brief introduction to White Hmong grammar.
2.1 Summary of Non-spatial Setting
Dixon (2012)’s Non-spatial Setting takes the form of a framework of eleven categories
that organize marking of tense, aspect, and modality as they appear across languages
(2012:3, 5-6). These categories are presented below in Table 1.
Category
Tense

Content
past, present, and possibly future, if
functions as tense; also words referring to
time
Reality
realis and irrealis, which includes modality
and possibly future time, if functions as
irrealis
Degree of Certainty
level of probability
Phase of Activity
stage of an action
Completion
perfect and imperfect
Boundedness
telic and atelic
Temporal Extent
progressive, punctual, etc.
Composition
perfective and imperfective
Degree or Frequency
number of times an event occurs
Speed and Ease
rate of an activity and its degree of
difficulty
Evidentiality
source of evidence
Table 1. Categories of Dixon (2012)’s Non-spatial Setting (based on Dixon 2012:5-9, 25).
The first category, Tense, is rather straightforward, marking the time at which an
event happened in reference to the current time. Dixon’s category also includes time
adverbials, or in his terminology, Lexical Time Words, belonging to one of five

4
categories: Duration, Frequency, Specific Time Spans, With Respect to Expectation,
which includes notions such as ‘soon’ and ‘already’, and Temporal Shifters, which
includes ‘yesterday’ and ‘tomorrow’ (2012:20). Future time is a special case, since it can
fall into either the Tense category or the Reality category, as Table 1 shows, depending
entirely on whether forms that indicate future events function as a part of the tense
paradigm or are themselves varieties of Modality (Dixon 2012:7-8).
Dixon’s Reality category contains an Irrealis subcategory that includes more
typical notions of irrealis as well as Modality (following Dixon 2012:22). Modality itself
can be one of several semantic types, of which the most prominent cross-linguistically are
Prediction, Obligation, Necessity, Ability, and Imminent Action (Dixon 2012:26-27).
Others are also possible, including Desire, Intention, Scheduled Activity, Achievement,
and Inevitability (Dixon 2012:26-27). Cutting across these semantic categories are two
morphosyntactic categories of Modality marking, namely, Modals and Semi-modals,
where Modals have reduced morphosyntactic features such as a lack tense marking (that
is, in languages that make tense distinctions) and a requirement that they precede all other
verbs in a series, while Semi-modals generally function as regular verbs (2012:26). For
example, Dixon’s modals for English include words such as will, should, and must
(2012:26)—a class of words that must precede the verb they modify, as in will go, should
go, must go, and have no marking for tense—that is, to indicate past time requires the use
of another auxiliary such as have, as in must have gone. In contrast, Dixon’s Semimodals for English include phrasal verbs such as be going to and get to (2012:26), which
can have other verbal content preceding them, as in I would like to get to go, and inflect
for tense, as in I got to leave early.

5
Categories of Desire and Intention can also be part of Dixon’s system of Modality,
though only as long as other arguments are not allowed to appear (2012:27-28). For
example, English allows sentence constructions such as I want him to go, where want
takes him as an argument; in this case, want is not a Modality marker. On the other hand,
in some languages the verb want can only be followed by a verb phrase or a complement
clause, and an object argument is not allowed; in this case, the verb want would be a
modality marker of Desire.
Dixon’s Degree of Certainty includes modal-like words and other morphological
elements that “describe the chance of some action or state eventuating” (2012:29). Phase
of Activity covers the morphological marking of the beginning, finishing, and
continuation of activities. Completion is the category for perfect and imperfect. For this
category, Dixon provides the general definition of the perfect as “‘an action which is
completed before the present time’ to which is often added ‘and which has present
relevance’” (2012:31). He defines imperfect as “refer[ring] to something which began
before the present and is still continuing” (2012:31).
Dixon provides the Boundedness category for overt marking of telicity and
phenomena associated with the inherent telicity of certain verbs (following Dixon
2012:33). Overt marking can include affixes, while phenomena can include a verb’s
behavior in regard to whether it is telic semantically (Dixon 2012:33).
Temporal Extent focuses on the explicit grammatical indication of “punctual” and
“durative” (Dixon 2012:34). Composition is the domain of the distinction between
Perfective’ and Imperfective. Perfective functions such that “the event is regarded as a

6
whole, without respect for its temporal constituency,” while imperfective is concerned
with “the temporal make-up of the event” (Dixon 2012:35).
The category of Frequency and Degree covers marking of senses such as
‘habitual’, ‘reiterative’, and ‘repeated’, as well as ‘a little bit’ and ‘a lot’ (Dixon 2012:36376). The category of Speed and Ease includes the morphological marking of concepts
exemplified by words such as ‘quickly’, ‘rapidly’, and ‘slowly’, as well as ‘with ease’
and ‘with difficulty’ (Dixon 2012:37-38). Evidentiality considers the ‘system’ of
“obligatory specification of the source of information on which” a ‘statement’ “is based”
(Dixon 2012:38).
While the Non-spatial Setting framework appears to categorize most forms of
marking on verbs, Dixon notes that other elements such as physical setting, mood, and
“clause linking” (marking of relationships between clauses) are explicitly excluded from
the Non-spatial Setting portion of his framework (2012:1-3). In addition, this set of
eleven categories may not include all of the possibilities found in any given language,
which, as Dixon notes, may “include…a variety of other types of Non-spatial Setting”
(2012:6). Thus, Dixon (2012) provides a near-comprehensive system for marking
temporal, modal, and aspectual information on verbs which is left open for new
possibilities in specific languages.
2.1.1 Dixon’s verb classification
One important aspect of Non-spatial Setting in Dixon (2012) is the use of “Secondary
verbs” as a part of the grammar. Thus, a brief overview of Dixon’s system of classifying
verbs is a necessary part of considering Non-spatial Setting, and is presented here.
6

Dixon (2012) cites a number of sources that describe the grammar systems of several languages that
possess the categories listed here for Frequency and Degree as well as Speed and Ease. The reader is
directed to Dixon (2012:36-38) for more information.

7
First, verbs are divided into two categories: Primary and Secondary. With Primary
verbs, their “arguments can all be just [noun phrases],” while with Secondary verbs, their
“arguments cannot all be just [noun phrases] or pronouns. That is, one argument must be
a clause” (Dixon 2006:9). Primary verbs fall into Primary-A and Primary-B categories;
the distinction is based on the fact that Primary-A verbs cannot have a complement as an
argument while Primary-B verbs can (Dixon 2006:9). More important for Non-spatial
Setting are Secondary verbs, which represent what Dixon (2006) calls “Secondary
concepts,” which can be marked by full verbs, affixes, or some other strategy, depending
on the language (Dixon 2006:9, 11). These Secondary verbs cannot occur on their own,
but must be associated with an additional verb, which is found in a complement clause or
as part of a “complementation strategy” (Dixon 2006:11, 12).
Secondary verbs fall into three categories, namely, Secondary-A, Secondary-B,
and Secondary-C (Dixon 2006:12-13). Secondary-A verbs can be one of four types: 1)
“Negators;” 2) “Modal-type” verbs, which includes modals; 3) “Beginning-type,” which
includes verbs with semantic values associated with Phase of Activity; or 4) “Trying-type”
(Dixon 2006:12-13). For a verb to be a Secondary-A verb, “the main and complement
clauses must have the same subject” (Dixon 2006:13). Secondary-B verbs are verbs such
as “want,” “intend,” or “hope,” which are characterized by the trait “that even when the
subjects may differ, the expectation is that they are most likely to be the same, and the
subject token in the complement clause is then generally omitted” (Dixon 2006:13).
Secondary-C verbs are verbs with meanings like “let,” “make,” and “help,” where “main
and complement clauses are likely to have different subjects” and with which “it is

8
unlikely that the subject token in the complement clause can be omitted” (Dixon
2006:13).
Important to Non-spatial Setting is category Secondary-A, which includes modals
and potentially many other markers of Non-spatial Setting (Dixon 2012:26). At the same
time, category Secondary-B is relevant to Non-spatial Setting in that if a verb could
belong to Modality, such as “want,” but allows more than two arguments (one being the
subject and the other a complement), it belongs to Secondary-B rather than Secondary-A,
and is in such a case not a Modality-marking verb (Dixon 2012:27). Dixon’s system of
verbal classification will be referenced below in the discussion on the Non-spatial Setting
of White Hmong.
2.2 Grammaticalization and morphological marking
The process of grammaticalization is important to Non-spatial Setting as many of the
grammatical elements in a language like White Hmong—i.e., an isolating, generally
monosyllabic language that is completely lacking in any overt affixation7—will have
some connection with lexical words. Distinguishing between genuinely normal lexical
items and grammatical ones is not an easy task when dealing with this kind of language,
and so a clear, even if brief, presentation of the nature of grammaticalization is necessary
for arguing whether a certain morpheme is actually operating as part of the grammar. 8
Here, grammaticalization will be defined and briefly discussed, and a set of processes

7

See below for the discussion on linguistic typology for the region.
Dixon generally recognizes a basic distinction between lexicon and grammar in his version of Basic
Linguistic Theory, where grammar is comprised of systems of small, closed sets of morphemes (2010a:47
ff., 214 ff.). It should be noted, however, that Dixon explicitly describes an expectation in the grammar that
“each [member of a system] may be exhaustively listed, each being fully defined by the exclusion of all
others” (2010a:47). At the same time, examples from specific languages may present difficulties for this
definition, such as the English prepositions on and upon—noting that Dixon explicitly recognizes on as part
of the grammar (2010a:214).
8

9
associated with grammaticalization will be presented with the ultimate goal of showing
below which morphemes in White Hmong are connected with the language’s grammar.
First, a definition of grammaticalization is needed. Hopper & Traugott provide a
helpful definition:
Grammaticalization is the change whereby in certain linguistic contexts speakers
use parts of a construction with a grammatical function. Over time the resulting
grammatical item may become more grammatical by acquiring more grammatical
functions and expanding its host-classes. (2003:99)
In addition to this, Brinton & Traugott make an important observation about semantic
changes associated with grammaticalization:
There is no doubt that, over time, meanings tend to become weakened during the
process of grammaticalization. Nevertheless, all the evidence for early stages is
that initially there is a redistribution or shift, not a loss, of meaning. (2005:94)
As a result, it is clear that when a word is grammaticalized, it is affected both
syntactically in terms of where it can appear, and semantically in terms of a change of
meaning. It can be concluded from this that if a language such as White Hmong contains
two homophonous words with slightly different semantic and syntactic properties from
one another, and if one of these two words serves a more grammatical function than the
other, then it can be suspected that the one is more likely a grammatical element. 9
Second, there are several processes distinctly connected with grammaticalization
presented by Brinton & Traugott, namely, decategorialization, bleaching, subjectification,

9

In fact, Bisang (1996:533-534) points out that grammaticalized elements are generally those used to serve
the role of marking tense, aspect, and/or mood, or in his terms, “mak[ing] a concept or an action/process
more concrete” (1996:533), in Southeast Asian languages. Please see section 2.3 below for the larger
discussion.

10
productivity, frequency, and typological generality (2005:110). Each of these will be
briefly introduced in turn.
Decategorialization is discussed by Brinton & Traugott as follows:
Hopper [1991:22] defines decategorialization as the process by which forms ‘lose
or neutralize the morphological markers and syntactic privileges characteristic of
the full categories Noun and Verb, and ... assume attributes characteristic of
secondary categories such as Adjective, Participle, Preposition, etc.’
Decategorialization is the defining characteristic of grammaticalization since it is
the mechanism by which lexical items become functional. (2005:107)
From this, it is clear that a grammaticalized element will have morphological and/or
syntactic irregularities when compared to the “full categories,” which serve as the most
prominent sign that a word has been grammaticalized—a trait that will be especially
important for an isolating language such as White Hmong.
Bleaching is defined by Brinton & Traugott as “weakening of meaning through
generalization, most especially loss of contentful meaning” and is associated with
grammaticalization “at least in late stages” (2005:108).
Subjectification is defined as “the anchoring of meaning in the speaker's
assessment of the situation,” where the word becomes increasingly connected with the
expression of “the speaker’s perspective...or to get others to do things” (Brinton &
Traugott 2005:108).
For Productivity, Brinton & Traugott state that “items that grammaticalize
become more productive in the sense that the grammaticalizing element occurs with

11
increasingly large numbers of categories, i.e., with increasing type frequency. The shift is
from a less to more productive pattern...” (2005:109).
In regard to Frequency, Brinton & Traugott comment that “items that
grammaticalize are used ‘in more contexts and for a larger set of lexical items’; therefore
grammaticalizing items always become more token frequent than their source”
(2005:109).
Typological generality is connected with grammaticalization in that
“grammaticalization patterns tend to be cross-linguistically replicated...and may affect
whole semantic classes...” (Brinton & Traugott 2005:109).
Altogether, grammaticalization is seen as a process that affects a word both
syntactically and semantically, and changes its function such that it increasingly serves a
grammatical purpose. In addition, grammaticalization is distinctly associated with six
processes, namely, decategorialization, bleaching, subjectification, productivity,
frequency, and typological generality.
2.3 Regional Typology
Before considering Non-spatial Setting in White Hmong, we first turn to typological
considerations from languages in the region as a helpful background.
The majority of languages in Southeast Asia share a number of typological
features concerning their verbal systems. First, they exhibit a general absence of
inflection, most notably in the areas of grammatical gender, number, marking or
agreement for subject or object, or tense (Goddard 2005:3-5). Second, their primary word
order is SVO, though the order is generally flexible (Goddard 2005:7-8).
Third, a number of languages in the region have particles that possess an emotional or
affective value, termed ‘sentence-final particles’ (Goddard 2005:24).

12
Fourth, languages in the region typically have serial verb constructions (Goddard
2005:17). A number of these serial verb constructions have historically produced
grammatical items that indicate causation, direction, grammatical case, and the result of
the action of another verb, as well as tense, aspect, and mood, and even conjunctions
(Bisang 1996:534, 563, 570). Verbs may also undergo ‘synchronic derivation’ to produce
‘locus prepositions’ and adverbs (Clark & Prasithrathsint 1985:34, 38), or, in other words,
prepositions and adverbs that indicate the location of something can be derived from
verbs.
Finally, these languages are characterized by what Bisang (1996) refers to as
“indeterminateness.” While Bisang does not provide a clear definition of
indeterminateness, it appears to refer to the degree of underspecification of grammatical
notions that are often mandatory in other languages. In relation to the indeterminateness
of verbs, Bisang states that “[a] verb in the languages to be described...merely posits an
action or a state. One verb—which is often just one syllable—can show enough
information in a given context...” (1996:532). He goes on to list areas of
indeterminateness for verbs, including tense/aspect/mood marking, ‘role assignment’,
‘valency’ and ‘complex sentences’ (1996:532). Bisang also discusses how ‘discourse
pragmatics’ may impact the general indeterminateness found in these languages, in that,
in certain cases, increasing the degree of definiteness of a word by grammatical means
may be required (1996:533). To fill this need, grammaticalized elements such as those
marking causation, result, tense, aspect, and mood are used (1996:533-534).
As it will be apparent below, several of these typological traits of Southeast Asian
languages will feature prominently in White Hmong. Before launching into the details of

13
White Hmong itself, three other Southeast Asian languages will be considered in terms of
Dixon’s framework for Non-spatial Setting: Mandarin Chinese, Guizhou Dong, and
Green Hmong.
2.3.1 Mandarin Chinese
Mandarin Chinese (Sino-Tibetan, China) is a language with strong typological
similarities to Hmong and other languages in the region. Some scholars, e.g. Li (1991),
argue that Chinese has historically influenced the verbal morphology of the Hmongic
languages, and so Chinese is directly relevant to the Non-spatial Setting system of White
Hmong. The grammar features that are especially relevant to Dixon’s framework are
Mandarin’s auxiliary verbs, modal-like full verbs, and its system of marking aspect—or
what Dixon (2012:5-6) calls “Completion” (Perfect vs. Imperfect), Composition
(Perfective vs. Imperfective), and Degree. Each of these will be discussed below.
Mandarin auxiliary verbs are distinguished from normal verbs by their syntactic
limitations and are rather abundant in number. Li and Thompson (1981:174-175) describe
six limitations: 1) they must appear with a main verb (though a main verb may be implied
by context); 2) they do not have their own aspectual marking; 3) they do not appear with
adverbial “intensifiers;” 4) they are not available for nominalization processes; 5) they
cannot be preposed in sentence-initial position; and 6) they do not appear with objects
(1981:174-175). However, like verbs, they can undergo negation and ‘A-not-A’ type
questions10 (1981:172-173).
The auxiliary verbs that fit the above definition have several interesting features.
Li & Thompson (1981:182-183) provide a list of these auxiliary verbs, reproduced below.

10

This type of question construction is a variety of polar question, and takes the form of ‘Subject Verb not
Verb?’.

14
yīnggāi, yīngdang, gāi
néng, nénggòu, huì, kěyi
néng, kěyi
gǎn
kěn
děi, bìxū, bìyào, bìděi
huì

‘ought to, should’
‘be able to’
‘has permission to’
‘dare’
‘be willing to’
‘must, ought to’
‘will, know how to’ (Li & Thompson 1981:182-183)

Several interesting observations are to be made here. First, gǎn ‘dare’ and kěn ‘be willing
to’ are auxiliary verbs, which are interesting as these fit Dixon (2012)’s definition of
Modal syntactically, though semantically they represent categories of Modality in
addition to the set he presents in his work. Second, while Dixon (2012:26-27) provides a
single category of Ability, Li & Thompson split the concept into three categories, glossed
as ‘be able to’, ‘has permission to’, and ‘will, know how to’, which contain unique words
not shared by the other two categories. While some auxiliary verbs such as néng, kěyi,
and huì show overlap between these, nénggòu does not, and none appear in all three
categories. This shows a clear ‘specialization of labor’ between these auxiliaries of ability.
Lastly, two degrees of obligation are signaled by the categories of ‘ought to, should’ and
‘must, ought to’, with no overlap in terms of category membership.
Mandarin also has several verbs with meanings similar to the above auxiliary
verbs, though unlike auxiliary verbs or Dixon’s modals, they allow events with additional
arguments to appear as their objects (Li & Thompson 1981:175ff.). These include yào
‘want’, qíngyuàn ‘wish, prefer’, jìxù ‘continue’, xūyào ‘need’, xīwàng ‘hope’, xiǎng
‘think, miss’, and biǎoshì ‘express’ (1981: 175-177). While some of these are
semantically similar to the examples of Modality markers that Dixon provides, they do
not qualify as true Modality since they can have additional arguments (Dixon 2012:2728).

15
Other potential cases for auxiliaries or modals found in Li & Thompson (1981)
include yào ‘be going to, in the immediate future’, kěnéng ‘possible, likely’, and several
kinds of adverbs (1981:175-176, 181). First, the word yào ‘be going to, in the immediate
future’ does not qualify as a true auxiliary since it cannot appear in negative or the ‘Anot-A’ question constructions (Li & Thompson 1981:175-176). However, this does not
disqualify it from Modality status by Dixon’s definitions, since Modality can include
syntactic particles (following Dixon 2012:27). Note that while this word is identical in
form as yào ‘want’ above, the distribution of the ‘be going to’ meaning is comparatively
more restricted; there is also some cases where ambiguity between the two is possible (Li
& Thompson 1981:175-176). In regard to Dixon’s categories of Modality marking, while
Li & Thompson provide no further information that would help elucidate the category
membership of yào, the glosses they provide, namely, ‘immediately, in the immediate
future, am going to’, suggest that yào belongs to the Imminent Activity category, and this
interpretation is provisionally adopted here.
Second, the word kěnéng ‘possible, likely’ is similar to auxiliary verbs
semantically and syntactically in some contexts, such as in sentences where it
immediately precedes a verb, though it itself patterns as a verb, or specifically, an
‘adjectival verb’ (Li & Thompson 1981:179-180). This word seems to fall into Dixon’s
category of Degree of Certainty, except that kěnéng is a full verb, while morphological
markers for this category should be either modals or morphological affixes; thus, this
verb would be a Semi-modal at best (following Dixon 2012:26, 29).
Third, several kinds of adverbs behave in ways similar to auxiliaries, such as
dàgài ‘approximately’, yídìng ‘definitely’, and kuài(yào) ‘soon’. These adverbs fail the

16
‘A-not-A’ question test as well as a second test where auxiliaries appear as an acceptable
response to a question, showing that they are not auxiliaries. The semantic domain of
Modality in Mandarin is thus comprised of a complex system of auxiliaries, full verbs,
and possibly adverbs.
Aspectual particles, or what Dixon considers Completion and Composition, also
form an important portion of Mandarin grammar. These include the ‘Currently Relevant
State’ marker le (‘CRS’), the ‘Perfective’ marker –le (‘PFV’), ‘Imperfective’ markers zài
and –zhe ‘IPFV’, and the ‘Experiential’ marker –guo (‘EXP’) (Li & Thompson 1981:185,
240).
The ‘Currently Relevant State’ marker, le, is defined by Li & Thompson
(1981:240) as denoting:
...that some state of affairs is current with respect to some particular situation.
When no other situation is mentioned, then it is always assumed that the statement
signaled by the sentence with the le is relevant to now, that is, to the situation of
the speech context in which the speaker and hearer are engaged.
Li & Thompson further provide five types of contexts where this kind of definition
applies, namely, if it expresses a situation that “is a changed state,” “corrects a wrong
assumption,” “reports progress so far,” “determines what will happen next,” and/or “is
the speaker’s total contribution to the conversation at that point” (1981:244).
The general idea of having implications for the present makes this seem identical
to Dixon (2012)’s perfect, for which he adopts the general definition that it refers to “‘an
action which is completed before the present time,’ to which is often added, ‘and which
has present relevance’” (2012:31). However, this particle does not necessarily mark

17
completed action on the verb with which it appears since le can refer to present relevance
even with stative verbs to indicate a change in state (per Li & Thompson 1981:244), often
in cases where the cause—that is, the completed action or event that brings about the
change of state—is not mentioned. 11 In addition, Dixon argues against the English
have...-en construction’s status as a perfect when it refers to a continuing event (2012:32),
suggesting that the ‘progress so far’ context of le is equally unacceptable as a perfect.
Unfortunately, since Dixon does not provide a clear category for the English
construction12, and since the particle le has no relevance to tense, it is unclear how le can
be categorized in the Non-Spatial Setting framework. The solution here is to honor Li &
Thompson’s descriptive term, ‘Currently Relevant State’.
Similar to le is the ‘Perfective’ particle –le ‘PFV’. Li & Thompson define the
usage of this affix as that “it indicates that an event is being viewed in its entirety or as a
whole” (1981:185). This matches Dixon’s definition for Perfective within his
Composition category perfectly (2012:35). Li & Thompson further develop this idea for –
le by finding four types of possible boundedness, namely, “...being a quantified event,”
“...being a definite or specific event,” “...bounded because of the meaning of the verb,”
and “...being the first event in a sequence” (1981:185-186). Thus, the particle –le is
characterized as a variety of perfective and is associated with an event that is somehow
bounded. At the same time, Li & Thompson (1981:215) demonstrate that –le cannot
indicate simple completion, as there are situations where it is used that have little to do
with a completed action. Thus, -le should be placed within Composition as a Perfective.

11

Please see Li & Thompson (1981:245 ff.) for examples of this kind of construction, virtually all of which
lack any mention of a completed action that brings about the change in state.
12
Apart from possibly treating one form of it as a “relative tense;” see Dixon (2012:32) for details.

18
The ‘Imperfective’ or ‘Durative’ (Li & Thompson 1981:185, 217) particles zài
and –zhe are additional aspectual markers in Chinese, which “signal the ongoing, or
durative, nature of an event,” much like English be -ing (Li & Thompson 1981:217).
These two morphemes have specific distributions in the grammar, where zài appears only
with verbs of ‘activity’ (Li & Thompson 1981:218), while –zhe only appears in the
standard language with states, but not with adjective-type verbs (Li & Thompson
1981:219-222). In either case, while Dixon’s framework provides two potential matches
for an ‘ongoing, durative’ marker, namely, Imperfect as part of Completion and Durative
as part of Temporal Extent, the fact that Li & Thompson (1981) links these two forms
with English be –ing suggests that zài and –zhe mark the Imperfect.
On the other hand, the behavior of zài and –zhe is such that it cannot co-occur
with –le ‘PFV’, due to the fact that they disagree semantically, where the former indicate
“unbounded” action and the latter indicates “bounded” action (Li & Thompson 1981:203).
This suggests a paradigmatic relationship in Mandarin that involves Completion and
Composition at the same time.
The affix –zhe also functions as a marker that “signal[s] that one event provides a
durative background for another event” (Li & Thompson 1981:223). With this usage,
–zhe can occur with activity verbs as well as state verbs (Li & Thompson 1981:224). This
form of –zhe matches Dixon’s examples for ‘Imperfective’ rather well, since they also
signal backgrounded events (2012:35). However, the forms used in Chinese for the other
event, which would then be “regarded as a whole” (Dixon 2012:35), are not obligatorily

19
marked with the –le suffix13. In general, though, this form is a good fit within
Composition.
The last important aspectual element under discussion for Mandarin is the
‘Experiential’ –guo affix. Li & Thompson define the usage of this affix as follows:
The aspect suffix –guo means that an event has been experienced with respect to
some reference time. When the reference time is left unspecified, then –guo
signals that the event has been experienced at least once at some indefinite time,
which is usually the indefinite past. (1981:226, emphasis original)
They further go on to contrast this affix with –le, stating that “the perfective –le signaling
a bounded event typically conveys the message that the event took place, while –guo
signals that an event has been experienced at least once” (1981:227). This form seems to
fall into Dixon’s Completion category, since –guo effectively expresses “‘an action
which is completed before the present time’...‘and which has present relevance’”
(2012:31), because –guo indicates that an event has occurred and is relevant to some
point in time. At the same time, -guo ‘EXP’ cannot co-occur with –le ‘PFV’, providing
further evidence that Completion and Composition form a paradigmatic relationship with
one another.
A final aspectual construction found in Mandarin is verbal reduplication, which
indicates “doing an action ‘a little bit,’ or for a short period of time” with activity verbs
(Li & Thompson 1981:232, 234). This construction can also optionally include the word
yi ‘one’, which is placed in the middle of the construction, with the two parts of the
reduplicated verb on either side (Li & Thompson 1981:232). The meaning ‘a little bit’

13

See Li & Thompson (1981:223-226) for examples where –le does and does not appear.

20
here fits perfectly within Dixon’s Degree category (following Dixon 2012:36), while the
‘short period of time’ meaning can probably be treated as an extension of ‘a little bit’.
Altogether, Mandarin has a number of morphological and syntactic forms that fit
well into Dixon’s Non-Spatial Setting in the areas of Irrealis (Modality), Completion and
Composition (as a single category), and Degree, as well as the distinctive category
‘Currently Relevant State’, summarized below in Table 2.
Category
Modality as part of Irrealis

Subcategory
Obligation
Ability

Imminent
Activity
Other
Completion-Composition

Perfect
Imperfect
ImperfectImperfective
Perfective

Examples
yīnggāi, yīngdang, gāi ‘ought to, should’
děi, bìxū, bìyào, bìděi ‘must, ought to’
néng, nénggòu, huì, kěyi ‘be able to’
néng, kěyi ‘has permission to’
huì ‘will, know how to’
yào ‘in the immediate future, be going to’
gǎn ‘dare’
kěn ‘be willing to’
-guo ‘EXP’
zài ‘be -ing’
-zhe ‘be –ing, IPFV’

-le ‘PFV’
Degree
reduplication: ‘a little bit, for a short period of
time’
Currently Relevant State
le ‘CRS’
Table 2. Summary of Non-spatial Setting elements in Mandarin Chinese. 14
As it will be seen, the findings here for Mandarin will have parallels in White
Hmong.
2.3.2 Guizhou Dong
Guizhou Dong15 (Tai-Kadai, China) is another language that is typologically similar to
White Hmong. Historically, at least one language of the family of which Dong is a part

14

Undoubtedly, Mandarin Chinese has a number of other morphemes and processes relevant to Non-spatial
Setting; however, what is displayed here is the forms discussed in Li & Thompson (1981).

21
has served as a ‘superstratum language’ in the Hmong homeland (Li 1991:43), suggesting
the possibility of direct influence on the development of Hmong, and Guizhou Dong is
currently spoken in areas in contact with various Hmong dialects. As a result, Dong
provides useful parallels for the grammar of White Hmong. Grammatical points of
interest available include modals, aspectual markers, and marginal reduplication, each of
which is discussed below in reference to Dixon’s Non-spatial Setting.
Guizhou Dong has a number of “modals,” including wo31 ‘know’, ju53 ‘want’,
haŋ13 ‘agree’, ȵon33 ‘be willing’, ʔam323 ‘dare to’, and li323 ‘have N to V’ (Long & Zheng
1998:120, 122). Long & Zheng’s modals typically appear before a verb, though some
modals, such as wo31 ‘know’ and ȵon33 ‘be willing’, can be followed by a pronoun or
noun instead (1998:122-123). This suggests that at least these last two do not belong to
Dixon (2012)’s Modality, as they behave as Primary verbs in the verbal classification
system of Dixon (2006). Likewise, the gloss for li323, ‘have N to V’, suggests the
mandatory presence of a different subject for the second verb, which suggests that it is a
Secondary-C verb (per Dixon 2006:13), and is thus not a marker of Modality (following
Dixon 2012:26). Information on how the other ‘modals’ behave is unavailable, so their
status is unknown. As for the specific semantic categories of Modality, such as Necessity
or Obligation, ju53 ‘want’ can be provisionally placed in the Desire category, while the
semantics of haŋ13 ‘agree’ and ʔam323 ‘dare to’ are such that they do not neatly fit into the
canonical categories provided by Dixon (2012:26), and so can be provisionally placed in
an “Other” category.
15

The source for this language, Long & Zheng (1998), provides data from both ‘southern’ and ‘northern’
dialects; generally, the southern forms are the ones cited in the discussion unless indicated otherwise; this
tendency is reflected in the discussion here.

22
Aspectual markers in Guizhou Dong include to323 ‘action in progress’, ta33
‘completed action’, ljeu31 ‘completion, change, or past event’, khwən35(ljeu31) ‘finished
action’, and ȶən212ma35 ‘beginning/continuation of an action’ (Long & Zheng 1998:148).
The first of these, to323, expresses ongoing action; it follows the verb (Long & Zheng
1998:148). Though there is relatively little data to determine the exact status of this form,
Long & Zheng (1998:148) relate it to the Mandarin –zhe, suggesting that it would qualify
as an imperfective under Dixon’s system.
The second aspectual marker, ta33 ‘completed action’, is further defined as
expressing previous experience, and is paralleled by the Mandarin –guo; it appears after
the verb and sometimes after the object as well, depending on the dialect (Long & Zheng
1998:149). As an experiential following the pattern of –guo, it appears that this form
would be a type of perfect, as discussed for Mandarin above.
The third and fourth markers, ljeu31 ‘completion, change, or past event’ and
khwən35(ljeu31) ‘finished action’, mark completion and follow the verb (Long & Zheng
1998:150). The marker ljeu31 is suggested to parallel Mandarin le16 (Long & Zheng
1998:150). As a result of its semantic value of completion, it appears that this form may
be a perfect within Dixon’s system, because of the notion of being “completed before the
present time” (Dixon 2012:31). The other marker here, khwən35(ljeu31), seems to fall into
the same category for the same reason.
The fifth marker, ȶən212ma35 ‘begin –ing’, is used to signal the starting or ongoing
status of an activity, and follows the verb (Long & Zheng 1998:150). As such, this falls

16

Note, however, that it is unclear which version of le is intended. It is assumed here that the perfective
form is in fact the one under consideration.

23
into Dixon’s Phase of Activity category, which includes notions of “beginning” and
“continuing” (Dixon 2012:30).
In addition to these, one remaining marker is found in Long & Zheng (1998):
kən11 ‘finished action’17 (1998:150). It follows the verb, and when it appears with lja31
‘completion’,18 an emphatic sense of ‘already’ is expressed (Long & Zheng 1998:150).
Whether this would belong to Dixon’s Phase of Activity (as ‘finishing’) or Completion as
a perfect is uncertain, due to the lack of sufficient data.
Reduplication in Guizhou Dong is only marginally attested and is likely the result
of influence from Chinese; the attested examples are either: 1) reduplicated
monosyllables that can signal ‘try out’ or ‘for a short time’, or 2) reduplicated bisyllabic
words that indicate repetitive action (Long & Zheng:121-122). It seems that these uses
would likely make reduplication fit into Dixon (2012)’s category of Frequency and
Degree.
As a whole, Guizhou Dong has a system of Modality belonging to Dixon’s
Irrealis, a system of aspect marking that seems to belong to his Phase of Activity,
Completion, and Composition categories, and a marginal reduplication process; these are
summarized in Table 3 below.

17
18

This is a northern dialect form. It is uncertain whether this is related to the southern khwən35 found above.
This form is the northern version of ljeu31 (Long & Zheng 1998:150).

24
Category
Modality as part of Irrealis

Subcategory
Desire
Other
Beginning
Perfect

Elements
53

ju ‘want’
haŋ13 ‘agree’, ʔam323 ‘dare to’
Phase of Activity
ȶən212ma35 ‘begin –ing’
Completion
ljeu31 ‘completion’,
khwən35(ljeu31) ‘finished
action’, ta33 ‘EXP’
Composition
Imperfective
to323 ‘action in progress’
Degree
reduplication: ‘try out, for a
short time’, ‘REPETITIVE
ACTION’
Table 3. Summary of Non-spatial Setting in Guizhou Dong.
2.3.3 Green Hmong
Green Hmong (Hmong-Mien, China/Laos) represents a group of Hmongic dialects
belonging to the Chuanqiandian cluster. Two of these dialects are considered here: one
from Xiaodala village, Honghe prefecture, Yunnan province, China 19 (Xiong & Cohen
2005:7), and the other from Laos.
2.3.3.1 Xiaodala Honghe dialect
The Xiaodala variety of Green Hmong has a number of aspectual particles, modal verbs,
and Lexical Time Words. The particles include leuf20 ‘complete situation’, lak ‘change in
progress’, dangl ‘finish’, dluat ‘experienced action’, and zhenf ‘in the process of’ (Xiong
& Cohen 2005:54-56). The phrase zhit dout ‘has not, did not’ also appears (Xiang &
Cohen 2005:56-57). Each of these is discussed below in turn.
The word leuf can either follow the verb or be sentence-final, and indicates that
“the speaker is looking at the action of the verb as a total event or a complete situation,
without attention to the internal structure of the event” (Xiong & Cohen 2005:54). This
19

The dialect represented in Xiong & Cohen (2005) is specifically the variety spoken in Xiao Dala village
within Honghe prefecture.
20
In the orthography of the Honghe dialect, tones are represented with final consonants. For this dialect, <b>
represents high falling tone, <t> mid-high level tone, <l> low falling breathy tone, <d> mid rising tone, <x>
mid falling tone, <k> mid level tone, <s> mid breathy tone, and <f> low falling creaky tone (Xiong &
Cohen 2005:12).

25
definition fits Dixon’s definition of the perfective perfectly and so belongs to the
category of Composition.
The word lak is sentence-final and “indicates that a change of state is currently in
progress or that an action has begun to happen” (Xiong & Cohen 2005:55). As such, this
seems to be some sort of progressive marker, which would place it in the category of
Completion as an imperfect.
The particle dangl either follows the verb or is sentence final; it only accompanies
an ‘action verb’ (Xiong & Cohen 2005:55). It “shows that the action of the preceding
verb has been finished;…a speaker uses [it] to show that a process has been completed or
that a series of repeated events has come to an end” (Xiong & Cohen 2005:55). This
particle seems to fit into both the perfect Completion (as marking a completed process)
and Phase of Activity categories.
The word dluat marks ‘experiential aspect’ and follows the verb (Xiong & Cohen
2005:56). It “indicates that the action of the main verb took place at least once in the past
or with respect to some reference time” and “implies that the person referred to in the
subject of the sentence has had the experience of undergoing or accomplishing this type
of action, and that the person’s past experience still has current relevance” (Xiong &
Cohen 2005:56). The “current relevance” status makes this particle a perfect within
Dixon’s Completion, and specifically one of marking past experience.
The word zhenf is a preverbal particle and marks ‘progressive aspect’ (Xiong &
Cohen 2005:56). Its possible translations include “in the process of” and “in the course of”
(Xiong & Cohen 2005:56). From these traits, it appears that this particle fits Dixon’s
category of Imperfect as a part of Completion. Zhenf can also be used in the construction

26
zhenf zhit, where it is combined with the negative zhit to indicate ‘not yet, still not’
(Xiong & Cohen 2005:56-57), making it extend semantically into the domain of Dixon’s
Lexical Time Words, specifically With Respect to Expectation (2006:20).
The phrase zhit dout ‘has not, did not’ provides an example of a marker that
signals that “an expected action has not taken place or did not take place” (Xiong &
Cohen 2005:56). It seems that this would likely fall into the category of Perfect as part of
Completion in Non-spatial Setting.
Modal verbs in Honghe Green Hmong include yuad ‘will, want, need’, dout ‘be
able’, dout ‘get the chance, have the opportunity to’, dout ‘must’, doul ‘be physically
able’, sangd ‘want’, and nchait ‘probably’ (Xiong & Cohen 2005:54-56, 81-82). The
phrase zhit xaob ‘need not’ also appears (Xiong & Cohen 2005:81). Each of these is
discussed in more detail below.
The word yuad is a preverbal modal marking ‘prospective aspect’ and indicates
“that someone is intending to do some action, or is on the point of doing it” (Xiong &
Cohen 2005:54). This is clearly a Modal in Dixon (2012)’s sense from a semantic
standpoint; the few examples available suggest that it is a regular Modal. This would
place it within Dixon’s Irrealis, as a marker of the Modality of Intent.
Yuad also serves a second purpose as a marker of Obligation or Necessity, where
it is translated as ‘need’ or ‘want’; this use appears before the verb (Xiong & Cohen
2005:81). This usage would fall into the categories of the same names, Obligation and
Necessity, in Dixon (2012).
In the negative, yuad as part of the phrase zhit yuad indicates the notion expressed
by the phrases ‘need not, there’s no need to, don’t have to’ (Xiong & Cohen 2005:81).

27
Xiong & Cohen (2005:81) explicitly note here that this serves as a “negation of necessity
or obligation.” In the framework of Dixon (2012), this is straightforward: zhit yuad is the
negative form for Obligation and Necessity as categories.
The modal dout ‘be able’ is postverbal and indicates “that the action of the verb is
permitted or capable of being accomplished” (Xiong & Cohen 2005:55). While it is
unclear whether this is operating as a full verb21, semantically, this is clearly Ability
Modality within Dixon’s Irrealis.
The modal dout can also be used preverbally with the meaning ‘get the chance,
have the opportunity to’ and signifies “that the person referred to in the subject of the
sentence has received the opportunity or privilege of doing some desired action” (Xiong
& Cohen 2005:55). While the nuance is different from dout ‘be able’ above, there is still
a semantic trait of Ability present, and so appears to fit within Dixon’s Ability Modality.
A third use of dout is to indicate the notion of ‘must’, and precedes the verb
(Xiong & Cohen 2005:81). Xiong & Cohen comment that dout here indicates obligation
and necessity, placing it in the corresponding categories provided by Dixon (2012).
Dout can also appear in the phrase yuad dout, where the resulting combination can
be translated ‘need to, should, have to’ and is placed before the verb (Xiong & Cohen
2005:81). Like dout, this serves as marking Obligation and Necessity (Xiong & Cohen
2005:81), and falls into the Modality categories in Dixon (2012) with the same names.
The word doul ‘be physically able’ is postverbal and accompanies ‘verb[s] of
physical activity’ (Xiong & Cohen 2005:55). It signals “that a person is physically

21

The data from the parallel word in White Hmong, tau ‘can,’ suggest that it is a regular verb and thus a
Semi-modal; see below for details.

28
capable of accomplishing the action” (Xiong & Cohen 2005:55). This appears to be
another Modal belonging to Dixon’s Ability category.
The word sangd ‘want’ is an auxiliary verb which precedes the matrix verb; it
indicates “desire or intent” (Xiong & Cohen 2005:56). While the data here does not
include tests for direct objects, it seems that this is a likely candidate for Dixon’s Desire
category of Modality, due to its semantic value and status as an auxiliary verb.
The word nchait ‘probably’ is a probability marker that Xiong & Cohen consider
a modal (2005:81-82). It precedes the verb and other auxiliaries such as yuad (following
Xiong & Cohen 2005:82). If this is in fact a modal (and not an adverb), then it fits well
within Dixon’s Degree of Certainty category.
The phrase zhit xaob is used as a ‘negative command’ and signifies the idea of
‘need not, do not need to’ (Xiong & Cohen 2005:81). As such, it would be a sort of
Modality marker indicating negative Necessity in the framework of Dixon (2012).
A few ‘lexical time words’ are also attested in Xiong & Cohen, including uat nax
‘usually’, zeuf ‘already’, and let ‘finally’ (2005:71).
As it has been seen, the Honghe dialect has morphology belonging to Dixon
(2012)’s Tense (in the form of lexical time words), Irrealis, Degree of Certainty, Phase of
Activity, and Completion categories. These are summarized in Table 4 below.

29
Category
Lexical Time Words

Subcategory

Elements
uat nax ‘usually’, zeuf
‘already’, let ‘finally’
Modality as part of Irrealis
Necessity-Obligation
yuad ‘need’, dout ‘must’
Negative Necessity
zhit xaob ‘need not, do not
need to’
Ability
dout ‘be able’, dout ‘get the
chance’, doul ‘be physically
able’
Desire
sangd ‘want’
Intent
yuad ‘will, want’
Phase of Activity
Finishing
dangl ‘finish’
Completion
Perfect
dangl ‘finish’, dluat
‘experienced action’, zhit
dout ‘has not, did not’
Imperfect
lak ‘change in progress’,
zhenf ‘in the process of’
Composition
Perfective
leuf ‘complete situation’
Table 4. Summary of Non-spatial Setting in the Honghe dialect.
2.3.3.2 Laotian Green Hmong
The Laotian dialect of Green Hmong shows striking similarities to the Honghe dialect
above, especially in the area of grammatical markers. Li provides information on yuav22
‘Future tense’, tau ‘get to’, tau ‘potential mode’, tau ‘Attainment aspect marker’, 23 lawm
‘Completion aspect marker’, and taabtom ‘Progressive aspect marker’ (1991:32, 35, 37,
52).24 Each of these is considered in turn.
The word yuav ‘Future tense’ is considered by Li to be an optional future tense
marker (1991:52). It has a phonologically identical counterpart yuav ‘buy’ with which it
cannot co-occur (Li 1991:52-53). It is likely to be left out if a time adverb signaling the

22

In the orthography of the Laotian dialects of Hmong, including Green and White Hmong, tones are
represented with final consonants, like in the Honghe dialect discussed above. For these dialects, <b>
represents high level tone, <Ø> mid level tone, <s> low-mid level tone, <v> mid rising tone, <j> high
falling tone, <g> low falling breathy tone, and <m> short low tone (Mottin 1978:15).
23
These three instances of tau are treated as differing uses of the same word by Li (1991).
24
Other relevant Laotian Green Hmong forms mentioned by Li include yuavtsum ‘must, should,’ pheej
‘reiterative,’ taag ‘finish,’ and xaav ‘want’ (1991:39, 52, 54). Coverage of these in the article, however, is
rather limited.

30
future is present, and, according to Li, yuav is mandatory in the apodosis of conditional
statements if no such adverbs are present (1991:53). Li’s description of yuav suggests that
it belongs to Dixon’s Tense category, although this assessment contrasts with that of yuad
‘prospective aspect’ in the Honghe dialect above, where it is found to be a modal. As a
result, it is not completely clear here whether yuav should be considered Tense or Irrealis
in Dixon’s model, or if the two categories should be treated as one concept in this
dialect.25
The word tau ‘get to’ appears immediately before the verb, and Li appears to
consider it a regular verb (following Li 1991:32, 34). This form may occur with the
future yuav in the form yuav tau with the meaning ‘will get to’ (Li 1991:31). Li does not
provide significant information for this form apart from his analysis of the ‘Attainment’
meaning of tau, so the exact status of this form is somewhat unclear. However, since Li
treats it as a verb—with no further indication otherwise—it should be treated as falling
outside the realm of Non-spatial Setting for the purposes of the discussion here.
The word tau with the function of ‘potential mode’ can appear in one of two
places: 1) immediately after the verb, or 2) after the direct object following the verb (Li
1991:35). While Li does not appear to provide a significant amount of additional
information about this use of tau, his treatment of it as marking ‘potential mode’ strongly
suggests that it serves as an Ability-type Modal or Semi-modal, an analysis adopted here.
The particle tau ‘Attainment aspect marker’ precedes the verb and the sentence in
which it appears must have some sort of phrase or marker that “bounds” the action, such
as a direct object, a time phrase, a “direction/location phrase,” or the “Completion aspect
25

It will be found below that the synonymous form yuav in White Hmong is a Modal, rather than tense
marking.

31
marker” (Li 1991:29, 34). The particle signals “the attainment of the event signaled by
the sentence,” where the event could “be a hypothetical or future event,” (Li 1991:29). Li
argues strongly against a common analysis of tau as past tense, and gives a conclusive
example where tau appears in a hypothetical protasis of a conditional statement (1991:2629). As a result, it is clear that tau is not a tense marker, and likely falls into the category
of Perfect as a part of Completion, as is found for its White Hmong counterpart below.
The particle lawm ‘Completion aspect marker’ appears at the end of the clause
and marks “the completion of an event” (Li 1991:37, 39). In terms of its distribution, its
appearance with the future yuav is allowed if the negative marker tsi is also present, since
“an overtly specified future event in Hmong cannot be seen as completed” (Li 1991:4041). At the same time, Li asserts that lawm can occur in an “irrealis” context, and
provides an example where lawm appears in a ‘when’ time clause marking a yet-to-becompleted action (1991:39, 41). In either case, lawm clearly follows Dixon (2012)’s
definition of Perfect, and so belongs to his category of Completion.
The word taabtom ‘Progressive aspect marker’ appears after the subject noun and
only accompanies ‘activity verbs’ (Li 1991:46). It can appear with yuav ‘Future aspect
marker’; the resulting construction means ‘be about to’ and is allowed to appear
preceding ‘stative verbs’ (Li 1991:46-48). The word taabtom is never accompanied by
lawm ‘Completion aspect marker’ or tau ‘Attainment aspect marker’ (Li 1991:47). While
it cannot appear with adverbs of past time in isolation, it can appear in a context giving
information on a background activity in relation to another event (Li 1991:49-50). The
restriction against taabtom appearing with perfect markers such as tau or lawm and

32
especially its use with past time adverbials to signal a background activity show that
taabtom is an imperfective marker belonging to Dixon’s Composition category.
The Laotian dialect of Green Hmong thus has markers belonging to Irrealis,
Completion, and Composition, and, if Li (1991) is correct, Tense. A summary of these
elements is shown below in Table 5.
Category
Tense
Modality as part of Irrealis
Completion

Subcategory
Future
Ability
Perfect

Examples
yuav ‘Future tense’
tau ‘potential mode’
tau ‘Attainment aspect marker’, lawm
‘Completion aspect marker’
Composition
Imperfective taabtom ‘Progressive aspect marker’
Table 5. Summary of Non-spatial Setting in Laotian Green Hmong.
The findings for Non-spatial Setting within the grammar of the three languages,
Mandarin Chinese, Guizhou Dong, and the two dialects of Green Hmong, are shown
below in Table 6.
Category

Mandarin
Chinese

Guizhou Dong

Honghe
Green
Hmong

Tense: Future
Tense: Lexical Time
Words

Modality:
yīnggāi,
Necessity/Obligation yīngdang, gāi
‘ought to,
should’
děi, bìxū, bìyào,
bìděi ‘must,
ought to’
Modality: Negative
Obligation

uat nax
‘usually’,
zeuf
‘already’, let
‘finally’
yuad ‘need’,
dout ‘must’,

zhit xaob
‘need not, do
not need to’

Laotian
Green
Hmong
yuav ‘Future
tense’

33
Category

Modality: Ability

Modality: Imminent
Activity

Mandarin
Chinese

Phase of Activity:
Beginning
Phase of Activity:
Finishing
Completion: Perfect

Honghe
Green
Hmong
dout ‘be
able’, dout
‘get the
chance’, doul
‘be
physically
able’

ju53 ‘want’

sangd ‘want’
yuad ‘will,
want’

néng, nénggòu,
huì, kěyi ‘be
able to’
néng, kěyi ‘has
permission to’
huì ‘will, know
how to’
yào ‘in the
immediate
future, be going
to’

Modality: Desire
Modality: Intent
Modality: Other

Guizhou Dong

gǎn ‘dare’
kěn ‘be willing
to’

Laotian
Green
Hmong
tau ‘potential
mode’

haŋ13 ‘agree’,
ʔam323 ‘dare to’
ȶən212ma35
‘begin –ing’
dangl ‘finish’

-guo ‘EXP’

ljeu31
‘completion’,
khwən35(ljeu31)
‘finished
action’, ta33
‘EXP’

-le ‘PFV’

Composition:
Imperfective

-zhe ‘IPFV’

tau
‘Attainment
aspect
marker’,
lawm
‘Completion
aspect
marker’

lak ‘change
in progress’,
zhenf ‘in the
process of’
leuf
‘complete
situation’

Completion:
Imperfect

Composition:
Perfective

dangl
‘finish’, dluat
‘experienced
action’, zhit
dout ‘has not,
did not’

to323 ‘action in
progress’

taabtom
‘Progressive
aspect
marker’

34
Category

Degree

Mandarin
Chinese
reduplication:
‘a little bit, for
a short period
of time’

Guizhou Dong

Honghe
Green
Hmong

Laotian
Green
Hmong

reduplication:
‘try out, for a
short time’,
‘REPETITIVE
ACTION’

Currently Relevant
le ‘CRS’
State
Table 6. Non-spatial Setting in Mandarin Chinese, Guizhou Dong, and Green Hmong.
With the above findings as a background, we will now briefly touch on the basic
grammar of White Hmong before considering its Non-spatial Setting in detail.
2.4 Basic grammar
White Hmong is an isolating language similar to other Southeast Asian languages, with a
basic word order of SVO (following Fuller 1988:25). An example of the basic word order
appears in (1).
(1)

Tsov26 tom twm
tiger
bite water.buffalo
‘Tigers bite water buffalo.’

(Fuller 1988:18)

Here, the subject tsov ‘tiger(s)’ appears first, followed by the verb tom ‘bite’, which is
then followed by the object noun twm ‘water buffalo’.
Grammatical marking is done through the use of additional words, such as
particles or adverbs, rather than verbal affixes or inflection.27 This is shown in (2) below.

26

Throughout this thesis, White Hmong is written using the native orthography. For a description of this
orthography, please see Fuller (1988), Heimbach (1969), or Jarkey (1991).
27
As mentioned as a general trait for Southeast Asian languages in section 2.3 above.

35
(2)

Lawv
tau
ntshai.
28,29
3PL
ATT
afraid
‘They were afraid.’

(Fuller 1988:20)

In (2), lawv ‘3PL’ serves the role of sentence subject and ntshai ‘afraid’ as predicate, and
the notion of perfect is conveyed by the separate word tau ‘ATT’.
Noun phrases typically consist of either a single noun, a construction with a noun
and an adjective, a construction including a noun classifier, or a possessive construction.
When adjectives are used, the word order is generally Noun Adjective (Fuller 1988:13;
Jarkey 1991:33), as in (3) below.
(3)

lub

tsev
loj
house big
‘the big house’

(Mottin 1978:48)

30

CLF

Here, the head noun tsev ‘house’ is followed by the adjective loj ‘big’. A small set of
adjectives do not follow the Noun Adjective pattern, but immediately precede the noun,
such as niaj ‘great’, nyuam ‘little’, and qub ‘old’ (following Fuller 1988:13; Jarkey
1991:33).
Noun classifiers are ubiquitous in White Hmong, and classifiers are generally
assigned to nouns based on some semantic component of the noun. For example, if the
noun is animate, the classifier tus ‘animate CLF’ will be the one used. Some common
classifiers appear in Table 7 below.

28

Elements of Non-spatial Setting are glossed throughout based on findings presented in section 3 below.
In some cases, these may rely in part on outside sources; the reader is directed to the relevant parts of
section 3 below for citations of these sources.
29
Glosses for content from outside sources are sometimes altered from the original to be consistent with
other examples throughout, especially as different sources tend to gloss Non-spatial Setting markers in
different ways.
30
See Table 7 below for the citation for glosses referring to classifiers.

36
Classifier
cov
lub
rab
tus

Category
plurals of an unspecified number
round, bulky, hollow objects; abstract nouns
tools and weapons
animate beings; long, slender, cylindrical objects;
abstract nouns
txoj
objects that come in lengths
yam
kinds, classes
Table 7. Some common classifiers in White Hmong. (quoting Jarkey 1991:35-37)
Noun classifiers generally appear when numerals, demonstratives, or some
possessives are used (see Fuller 1988:19). With numerals, the word order is Numeral
Classifier Noun (Fuller 1988:13; Jarkey 1991:32). With demonstratives, the order is
Classifier Noun Demonstrative (see Jarkey 1991:32). Examples of classifiers with a
numeral and a demonstrative appear in (4) and (5), respectively.
(4)

ib
lub
teb chaws
one CLF country
‘one country’

Here, the numeral ib ‘one’ is followed by the classifier lub ‘round objects/abstract CLF’,
which in turn is followed by the noun teb chaws ‘country’ to produce the meaning ‘one
country’.
(5)

lub

ntiaj teb
CLF
world
‘this world’

no
this

In (5), the classifier lub ‘round objects/abstract CLF’ appears first, followed by the noun
ntiaj teb ‘world’, followed by the demonstrative no ‘this’.
There are three varieties of possessive constructions: 1) construction with a
classifier; 2) a comparatively rare construction made up of possessor noun followed by
possessed; 3) construction with the li possessive marker. The construction with a

37
classifier takes the form of ‘Possessor Classifier Possessed’ (see Fuller 1988:19), as
shown in (6) below.
(6)

peb lub
hom phiaj
1PL CLF purpose
‘our purpose’

In (6), the possessor peb ‘1PL’ appears first, followed by the relevant classifier for the
possessed noun, in this case lub ‘round objects/abstract CLF’, followed by the possessed
noun hom phiaj ‘purpose’.
A construction made up of the possessor noun followed by the possessed noun is
occasionally seen, though this is restricted to specific possessed nouns (Mottin 1978:30).
One example is kuv txiv ‘my father’, where kuv is ‘1SG’ and txiv is ‘father’; adding the
relevant classifier, tus ‘animate CLF’, results in the meaning changing to ‘my husband’
(Fuller 1988:19; Mottin 1978:30). Another rare construction is one which uses the
particle li in the pattern ‘Possessor li’ and tends to be used with the copula in a
predicative sense (Heimbach 1979:111; Mottin 1978:46), as shown in (7).
(7)

tej no31
yog kuv
these
CLF
1SG
‘These are mine.’33

li
32

POSS

Here, the phrase kuv li ‘mine’ is made up of kuv ‘1SG’ and the particle li, and appears
after the copula yog, serving a predicative purpose.
Relative clauses are one of three types: 1) a clause headed by uas ‘that’ after the
head noun; 2) a clause headed by the pronoun/classifier qhov ‘thing’;34 and 3) a verbal
phrase headed by a classifier. The first type is a clause comprised of a head noun, the
31

Word grouping and gloss based on Heimbach (1979:313).
Gloss following Heimbach (1979:111).
33
Translation modified from Jay Yang’s original to make grammatical in English; follows Heimbach
(1969:111)’s treatment of kuv li as ‘mine.’
34
Gloss following Mottin (1978:35).
32

38
subordinating conjunction uas, and the phrase that is relativized. This is shown in (8)
below.
(8)

lub

hom phiaj uas peb los
nyob
purpose
REL 1PL
come live35
‘...the purpose that we live in this world...’36
CLF

rau
to

hauv
in

ntiaj teb
world

no
this

In (8), the head noun is hom phiaj ‘purpose’, which is followed by the relative clause uas
peb los nyob rau hauv ntiaj teb no ‘that we live in this world’.
The second type of relative clause is comprised of a classifier such as qhov ‘thing’
or tus ‘animate CLF’ followed by the clause, and functions as a nominalized entity
(following Mottin 1978:25). An example appears in (9).
(9)

Tiam sis qhov kuv
coj
tuaj
qhuab qhia
but
CLF
1SG take come preach
‘But what I bring to preach to us tonight...’

peb
1PL

hmo
night

no...
this

In this case, the relative clause is formed from the classifier qhov ‘thing’ combined with
the clause kuv coj tuaj qhuab qhia peb hmo no... ‘I bring to preach to us tonight’, where
qhov fills the slot of the otherwise missing argument.
Pronouns in White Hmong are split into singular, dual, and plural varieties, with
one additional ‘impersonal pronoun’37 (Mottin 1978:44), as shown in Table 8 below.
Singular
Dual
Plural
First Person
kuv ‘1SG’
wb ‘1DU’
peb ‘1PL’
Second Person
koj ‘2SG’
neb ‘2DU’
nej ‘2PL’
Third Person
nws ‘3SG’
nkawd ‘3DU’
lawv ‘3PL’
Impersonal
yus ‘one’
Table 8. Pronouns in White Hmong. (Adapted from Mottin 1978:44)

35

It should be noted that the word nyob ‘live, stay’ also has a grammaticalized use as a coverb meaning ‘in’;
both uses appear extensively below.
36
A more idiomatic English translation would be ‘the purpose for which we live in this world,’ though the
translation provided above, while somewhat awkward in English, accurately represents the structure of the
phrase in White Hmong.
37
English terminology here following Jarkey (1991).

39
Verb phrases in White Hmong represent one of the most complex aspects of the
language. This includes word order and morphological phenomena. First, the basic word
order is the following:
(Adverb) (Modal) (Negative) Verb(s) (Adverb) lawm
where lawm is a marker signaling one variety of perfect.38 Verb phrases also exhibit a
number of important phenomena, including grammaticalized markers of aspect, modals,
serial verb constructions, the use of adverbs, and complementation constructions.
Grammaticalized markers of aspect and modals will be dealt with in detail in the
discussion on Non-spatial Setting, as they are an integral part of the system. The others
will be briefly discussed in turn below.
Serial Verb Constructions (SVCs) 39 are extremely common in White Hmong.
These can take any one of a large number of forms. Important construction types include
“Cotemporal,” “Causative,” “Accomplishment,”40 and “Disposal” serial verb
constructions (Jarkey 1991). Accomplishment SVCs are important to Non-spatial Setting,
and will be discussed in detail below. For specific information on the other SVCs, the
reader is directed to Jarkey (1991).
Adverbs and adverbial phrases generally occur in one of three places: 1) at the
beginning of the clause; 2) at the very beginning of the verb phrase, after the subject; or 3)
near or at the end of the clause, immediately before the final perfect marker lawm if it is

38

See the discussion on Non-spatial Setting below for arguments and examples that show this.
Abbreviation in line with the usage found in Jarkey (1991).
40
Also known as “Attainment” in Jarkey (2004).
39

40
present.41,42 Examples of each of these sentence patterns are shown in (10), (11), and (12),
respectively.
(10) Hnub no
kuv
coj
Vaj Tswv
day
this 1SG take God
‘Today I will preach God’s word…’

txoj lus
word

tuaj…
come

Here, the adverbial phrase hnub no ‘today’ precedes the subject pronoun kuv ‘1SG’,
demonstrating the use of an adverbial at the beginning of the sentence.
(11) Tej zaum
peb twb
muaj coob
tug sim
maybe
1PL already have many CLF try
‘Maybe we have many people who have tried it already.’

tas
finish

lawm.
CP

In (11), the adverb twb ‘already’ appears between the subject pronoun peb ‘1PL’ and the
verb muaj ‘have’, showing that this kind of adverb can occur between the subject and the
rest of the verb phrase of which the adverb is a part.
(12) Peb twb
nrhiav tau peb lub
1PL already search ATT 1PL CLF
‘We already found our purpose yesterday.’

hom phiaj
purpose

nag hmo
yesterday

lawm.
CP

In this case, the adverb nag hmo ‘yesterday’ appears after the object peb lub hom phiaj
‘our purpose’ but before lawm ‘CP’, showing the possibility of adverbs in this position.
Though adverbials can occur in each of these positions, there are restrictions on
the positioning of certain adverbs and adverbial phrases. While such consideration is
generally beyond the scope of this thesis, one restriction is important to Non-spatial
Setting: the position immediately after the subject is the location where grammaticalized
adverbials regularly appear, such as the Imperfect tab tom ‘currently’, or the Modality

41

One example that appears in the data from Jay Yang suggests that there may be a fourth option, where
the adverbial phrases appear after the verb but before content indicating the direct object, all inside a larger
nominalized verb phrase. More research is necessary to determine why this one example stands out from
the others and whether it is indeed indicative of a fourth possibility for adverbial placement.
42
See below for the discussions on adverbs and lawm.

41
marker mam li ‘will’. Please see section 3.2 below for the discussion on such adverbials
in the post-subject, pre-modal position.
Complementation is widely attested in White Hmong, of which there are several
different types. These are: 1) a “Serial-like complementation strategy,” where the
complement content immediately follows the head verb without an explicit marker but
does not function as a full complement syntactically; 2) a “Potential” complement that
begins with the conjunction kom ‘PC’43; 3) a “Fact” complement beginning with (hais)
tias ‘that’; 4) “Topic” complement beginning with txog (hais) tias ‘about’; 5) “Indirect
Will” beginning with tias kom ‘that’; and 6) “Activity,” which lacks an explicit marker
and functions as an argument of the head verb (Jarkey 2006:123, 124, 126, 127, 132-133).
Each of these six complementation patterns is shown below in turn.44
(13)

43

(a)

Yog li ces peb thiaj li
txiav txim siab xam phaj tuaj
45,46
so
then 1PL consequently
decide
interview come
‘So then we made the decision to interview to come here to…’

Gloss based on Jarkey (2006:124). In line with Jaisser (1984), Jarkey (2006:124) points out three distinct
uses of the word kom as a distinct entity. In addition to its use as a “Potential complementizer,” it can be
used as a conjunction meaning ‘so that, in order that’ as well as a verb meaning ‘order’ (or ‘tell’ in Jaisser
1984). Kom is also glossed by Heimbach (1979:85) as ‘cause’, suggesting a fourth option. Throughout this
thesis, identification of kom as a Potential complementizer, as opposed to these other uses, is generally
based on Jarkey (2006:132-133).
44
The classification of verbs found in Jarkey (2006:132-133) is relied upon in part for the examples found
below.
45
The language consultant, Jay Yang, at times consulted outside sources in the course of glossing certain
words, especially function words. These have generally been replaced with glosses and/or accompanied by
citations from appropriate academic works when they appear in this thesis. The citations will be provided
where the word in question first appears. At the same time, some of these glosses from outside sources
ultimately influenced a number of his free translations; most of these free translations have been left as is,
along with the understanding that the language consultant formulated the ultimate translation based on his
intuition, and incorporated any glosses that he found agreeable as a native speaker idiomatically into the
translation. The outside sources included a Green Hmong dictionary named English-Mong-English
Dictionary, written by Lang Xiong, William J Xiong & Nao Leng Xiong, Bing Translator, and several
Hmong community websites.
46
Thiaj and thiaj li function generally in the same manner and can be translated as ‘so,’ ‘consequently,’
‘thereby,’ or ‘then’ (Heimbach 1969:340). Heimbach’s glosses are used throughout for these two words.

rau
to

42
(b)

teb chaws A mi kas
country
America
‘…America.’

no.
here

Here, the verbal phrase txiav txim siab ‘make a decision to’ is accompanied by the phrase
xam phaj tuaj rau teb chaws A mi kas no ‘to interview to come here to America’, without
an explicit marker serving as introduction. 47
(14) (a) …ces nws thiaj
tau hais
then 3SG consequently ATT say
‘…then he told them to take me…’
(b)

kom

lawv
3PL

PC

thiab kuv
tus
muam wb
mus rau
and
1SG CLF sister
2DU go
to
‘…and my sister—us—(and) go to Thailand.’

los
come

coj
take

kuv
1SG

Thaib Teb.
Thailand

In this case, the Potential, or ‘Intention/Will’ complement (Jarkey 2006:124) kom lawv
los coj kuv thiab kuv tus muam wb mus rau Thaib Teb ‘to take me and my sister—us—
(and) go to Thailand’ accompanies the verb hais ‘say’ and begins with the word kom ‘PC’.
(15) ...los sis tsis
paub
hais tias Vaj Tswv
or
NEG
know that
God
‘...or do not know that God created all things...’

tsim
create

ib puas tsav yam...
everything

Here, the head verb paub ‘know’ is followed by the complementizer hais tias ‘that’ and
the rest of the complement clause Vaj Tswv tsim ib puas tsav yam ‘God created all things’,
which signifies the “asserting” of “a fact” (Jarkey 2006:120).
(16) Yog peb xav paub txog hais tias
if
1PL want know about
‘If we want to know why we...’

47

yog
COP

vim
because

li cas
what

peb
1PL

ho
so48

Following Jarkey (2006:133), the verb txiav txim ‘decide’ takes a “serial-like strategy” when the subject
is the same for both main and complement verbs.
48
Ho is a “connective particle indicating sequence and carrying forward the action of the verb” and can be
translated as ‘and,’ ‘but,’ ‘and then,’ ‘again,’ or ‘so’ (Heimbach 1969:53). Heimbach’s glosses are used
throughout.

43
los
nyob rau hauv lub
come live
to
in
CLF
‘...came to live in this world...’

ntiaj teb
world

no...
this

In this case, the verb paub ‘know’ is followed by the complementizer txog hais tias
‘about’ and the rest of the complement phrase yog vim li cas peb ho los nyob rau hauv
lub ntiaj teb no ‘why we came to live in this world’, which indicates the Topic connected
with the verb paub.
(17) Yog li
kuv
txiv hlob
thiaj li
then
1SG uncle
consequently
‘Then my uncle said that…’

tau
ATT

cov
uas peb hos
me
los sis
PL.CLF
REL
1PL and.then little
or
‘…those who are little and the old people who…’

hais
say

tias kom
that

cov
PL.CLF

laus
old

peb
1PL

uas
REL

mus
go

tsis

tau
ces
rov qab los
mus thawj
nyab laj
nyob
can then return
come go
surrender Vietnamese in
‘…cannot walk should go back and surrender to the Vietnamese…’
NEG

rau tom
to
at
‘…in town.’

zos.
town

In the example above, the long complement clause tias kom peb…rau tom zos ‘that
we…in town’ is introduced by tias kom ‘that’ serves as a complement to the verb hais
‘say’ and expresses the Indirect Will of kuv txiv hlob ‘my uncle’.
(18) …peb lub hwj chim uas peb
1PL
CLF glory
REL 1PL
‘…our glory that we dream of…’

niaj
hnub
every49 day

ua
do

npauv suav xav
dream
think

yuav xav
tau nyob rau yav pem ntej
want think get in
to
in.the.future
‘…having in our lives in the future.’
Here, the Activity complement xav yuav xav tay nyob rau yav pem ntej ‘of having in our
lives in the future’ immediately follows the predicate ua npauv suav ‘dream’ without the
49

Gloss confirmed by Heimbach (1979:140).

44
use of a distinct complementizer (following Jarkey 2006:132). The reader is directed to
Jaisser (1984) and Jarkey (2006) for more details on complementation in White Hmong.
One important discourse-related entity is the topic-comment construction in
White Hmong. This takes the form:
Topic (mas) Comment
where mas is a topic marker (Fuller 1988:65 ff.). An example appears below in (19).
(19) Lub

teb chaws Sam-Neua mas peb
country
Sam-Neua TOP 1PL
‘The Sam-Neua area, we let you rule over...’
CLF

cov
Hmoob
PL.CLF Hmong
‘...the Hmong.’

cia
let

koj
2SG

saib
look

xyuas
visit

(Fuller 1988:37)

Here, the topic, lub teb chaws Sam-Neua ‘the Sam-Neua area’ is followed by the topic
marker mas. This is then followed by the comment peb cia koj saib xyuas cov Hmoob
‘we let you rule over the Hmong’, which refers to a situation involving the location
mentioned in the topic.
There are also a large number of markers associated with exclamations and other
similar effects on the sentence. The reader is directed to Mottin (1978) for examples of
these.

45
3. Grammatical elements marking Non-spatial Setting in White Hmong
White Hmong has a large number of words and phrases that belong to Non-spatial Setting.
These grammatical markers are shown in their proper category in Table 9 below.
Category
Tense: Lexical Time Words

Irrealis: Positive
Irrealis: Negative
Irrealis: Modality

Level of Certainty: Absolute

Level of Certainty: Moderate
Phase of Activity: Beginning
Phase of Activity: Continuing

Phase of Activity: Finishing

50

Examples
Specific Time Words: nyuam qhuav ‘moment’, hnub
‘day’, vas nthiv ‘week’, hli ‘month’, xyoo ‘year’, sauv
ntxov ‘morning’, hnub tseg ‘noontime’, hnub qaij
‘afternoon’, tsaus ntuj ‘night’
With Respect to Expectation: twb ‘already’, yeej ‘already’,
tseem ‘still’
Temporal Shifters: tag kis ‘tomorrow’, neeg kis ‘the day
after tomorrow’, puag nraus ‘three days from now’, puag
nag nraus ‘four days from now’, nag hmo ‘yesterday’,
hnub hmo ‘the day before yesterday’, hnoob hnub ‘three
days ago’, puag hnoob hnub ‘four days ago’, ntua ‘just’,
nyuam qhuav ‘just, a moment ago’, maj mam ‘after a
while’
yuav ‘POS.IRR, intend’, mam (li) ‘will’ (adverbial)
txhob ‘NEG.IRR, should not’
yuav ‘will, intend’, txhob ‘should not’
yuav tsum ‘should, have to, must’
yuav tau ‘must, have to’
tsum ‘can, have to’
tau ‘can’
taus ‘be physically/materially able to’
txawj ‘know how to’
(tsis) yeej ‘can(not)’
kiag (li)50 ‘really’
tiag ‘really’
twb ‘indeed’
yeej ‘certainly, definitely’
xwb ‘indeed’
ntshai ‘maybe’
tej zaum ‘maybe’
pib ‘begin to’, tab tom ‘begin to’, chiv ‘begin to’
pheej/pej ‘continue to V, keep Ving’,
nyim ‘keep Ving’,
rau siab ntso ‘keep Ving, commit oneself to V’
tas ‘finish Ving, be done Ving’
tiav ‘finish Ving’

In some ways, kiag behaves like a particle, and in others like an adverb, as discussed above.

46
Category
Completion: Perfect

Examples
tau ‘have Ved (ATT)’, lawm ‘have Ved (CP)’, tau...dua,
dua...lawm ‘have Ved (EXP)’, Attainment Serial Verb
Construction51
Completion: Imperfect
tab tom ‘IMPERFECT’,
sij ‘continually, repeatedly’,
Reduplication
Speed and Ease: Slowness
maj mam ‘slowly, progressively, gradually’
Table 9. Grammatical markers of Non-spatial Setting in White Hmong.
These markers fit into the basic clause structure as follows:
PreS NP PreMOD MOD NEGP52 PreV Verb PostV NP PF ADP lawm xwb53
Each of the markers of Non-spatial Setting below is presented in order based on the
earliest possible slot in which they appear in a typical clause.54 This presentation is
followed by a section on those forms that are treated as grammatical markers in other
sources but for various reasons do not fulfill Dixon (2012)’s criteria for the relevant
category of Non-spatial Setting.

51

Each of these forms, while marking the Perfect as part of Non-spatial Setting, represents a distinct
category of Perfect—at least some of which are able to co-occur both semantically and in actual sentences.
As a result, these should not be regarded as forming a natural paradigm, but as constituting unique
categories of Perfect. At the same time, it should be noted that tau ‘ATT’ can take part in the Attainment
Serial Verb Construction, as mentioned in section 3.5.1 below.
52
Note that this is the location for sentence-level negation; negation of individual verbs, subordinate
clauses, and some elements of Non-spatial Setting such as tau ‘be able’ is done through the use of negation
markers immediately before the verb or grammatical marker.
53
The ordering presented here is the most general for purposes of clause-level marking of Non-spatial
Setting. It should be noted, however, that at least some of the markers can appear in other positions, usually
later in the clause, due likely to emphasis, local scope, or clausal subordination. Considerations of this
nature, however, are generally outside the scope of this thesis.
54
Note also that a small subset of Non-spatial Setting markers, namely tau ‘ATT,’ tas ‘finish,’ and lawm
‘CP,’ can also combine with Lexical Time Words and a grammaticalized version of the verb los at the
beginning or end of a sentence in unusual configurations to indicate time since (following Mottin
1978:106). Please see section 3.1.1 where this is discussed. Nevertheless, for purposes of this thesis, the
various morphemes are generally considered in relation to how they mark Non-spatial Setting on verbs
directly.

47
3.1 Pre-Subject Position
Grammatical items in this position include Lexical Time Words of the “Temporal shifter”
variety55 and the markers of moderate Degree of Certainty, ntshai ‘maybe’ and tej zaum
‘maybe’.
3.1.1 Temporal shifters and other time markers
White Hmong has a class of adverbs that do the work of Dixon (2012:20)’s “Temporal
shifters.” As stated above, Temporal shifters are a category of Lexical Time Words in the
larger category of Tense in Non-spatial Setting. These adverbs include nag hmo
‘yesterday’, hnub hmo ‘the day before yesterday’, tag kis ‘tomorrow’, neeg kis ‘the day
after tomorrow’, and puag nraus ‘three days from now’. Mottin (1978:104)56 provides a
list of these that also includes hnoob hnub ‘three days ago’, puag hnoob hnub ‘four days
ago’, and puag nag nraus ‘in four days’. More specific periods of a day can be added
after these, such as sauv ntxov ‘morning’, hnub tseg ‘noontime’, hnub qaij ‘afternoon’,
and tsaus ntuj ‘night’. These adverbs generally appear in one of two specific places: at
the very beginning of a clause, or in the “final adverbs and particles” position. First, they
can appear at the very beginning of a sentence, as in (20).
(20) Tag kis
peb yeej
nrhiav tau peb lub
hom phiaj.
57
tomorrow 1PL certainly search can 1PL CLF
purpose
‘Tomorrow we will be able to search for our purpose.’
Here, tag kis ‘tomorrow’ appears at the very beginning of the sentence.

55

Note that two special Temporal shifters, namely, nyuam qhuav ‘just, a moment ago’ and ntua ‘just,
immediately,’ canonically appear in positions other than the pre-subject position, and they are addressed at
length in sections 3.2.3 and 3.5.3, respectively.
56
Several of the terms that Mottin (1978:104) provides differ from the forms given by my language
consultant, Jay Yang. These are hnub hmos ‘the day before yesterday,’ nag kis ‘the day after tomorrow,’
and nag nraus ‘three days from now.’
57
Please see Mottin (1978) or Jarkey (1991) for a discussion on classifiers in White Hmong. The gloss CLF
found throughout this thesis is in alignment with their work.

48
Second, time adverbs can appear in the “final adverbs and particles” position.
Typically, this will be at the end of the clause unless the perfect marker lawm also
appears in the sentence, which will then follow the time adverb. Examples include (21)
and (22) below.
(21) Peb yeej
yuav
nrhiav tau
peb
1PL certainly POS.IRR search can/ATT 1PL
‘We will be able to find our purpose tomorrow.’

lub
CLF

hom phiaj
purpose

tag kis.
tomorrow

In this case, tag kis appears at the end of the sentence, without any clear difference in
terms of time or scope when compared with (20).
(22) Peb nrhiav tau
peb lub
hom phiaj
1PL search ATT 1PL CLF purpose
‘We have found our purpose yesterday.’

nag hmo
yesterday

lawm.
CP

Here, nag hmo ‘yesterday’ appears, followed by the perfect marker lawm.
These time adverbs behave similarly to full adverbial phrases indicating other
periods of time in terms of their sentence position. This is shown by (23) and (24) below.
(23) Tam sim no peb nrhiav tau
peb lub hom phiaj.
right.now
1PL search can
1PL CLF purpose
‘Right now we can look for our purpose.’
In (23), the phrase tam sim no ‘right now’ appears sentence-initially like tag kis does in
(20).
(24) Koj noj tau
hmo
2SG eat
can
dinner
‘You can eat dinner now.’

tam sim no
now

lawm.
CP

Here, tam sim no appears near the end of the sentence and immediately before lawm,
paralleling nag hmo in (22) above. Thus, time adverbs and their phrasal counterparts
behave in the same manner, appearing in the same positions in the sentence.

49
Adverbial phrases of time also present an interesting feature: they often take the
same markers as verbs. These include tau ‘ATTAINMENT’58 and lawm ‘COMPLETIVE
PARTICLE’

(following Mottin 1978:106). Verbs such as tas ‘finish’ or los ‘come’ may also

appear (following Mottin 1978:106). Some examples appear below.
(25) Kuv
nyob
Thaib Teb tau
1SG
live
Thailand
ATT
‘I lived in Thailand for three years.’

peb
three

xyoos
year

(Mottin 1978:106)

Here, the perfect marker tau appears with the phrase peb xyoos ‘three years’ to indicate
elapsed time.
(26) Kuv twb
yuav
ua kws kho mob
1SG almost POS.IRR do doctor
‘I almost became a doctor five years ago.’

tsib
five

xyoo
year

tas
los
finish come

lawm.
CP

In this case, the time phrase tsib xyoo ‘five years’ is followed by the phrase tas los lawm,
made up of the verb tas ‘finish’, the verb los ‘come’, and the perfect marker lawm,
representing the meaning of the English ‘ago’. This combination can also be found in the
sentence topic, as in (27) below.
(27)

peb59 xyoos tas
los
lawm
ATT three year
finish come CP
‘Three years ago60, I was in…’

(a)

Tau

(b)

Thaib Teb
Thailand
‘…Thailand.’

kuv
1SG

nyob
in

(Mottin 1978:106)

Here, tau ‘ATT’, tas ‘finish’, and lawm ‘CP’ appear with the time phrase 3 xyoos ‘three
years’ in the topic to place the comment at a point in time. Thus, the markers tau and

58

Mottin (1978) gives the French equivalent for the English ‘for’ with a time phrase in his treatment of this
use of tau. For purposes of this thesis, this use of tau is also treated as ‘ATTAINMENT’, to show the
connected relationship in meaning between attainment of an event and attainment of time.
59
This was the numeral “3” in Mottin (1978)’s original example. It has been replaced here with the actual
Hmong word for “three” written out.
60
Mottin (1978)’s work is written in French, and I at times have relied on outside sources to assist in
translating the French. Here, the translation for il y a in reference to years is drawn from Kellogg (2014).

50
lawm as well as the verbs tas ‘finish’ and los ‘come’ may appear with phrases to place an
event in time in both sentence topic and comment.
As part of a full system of Lexical Time Words, Temporal shifters can combine
with some words belonging to the Specific time span category, as in (28) below.
(28)

nag hmo hnub qaij
yesterday afternoon
‘yesterday afternoon’

Here, the Temporal shifter nag hmo ‘yesterday’ is combined with the Specific time span
word hnub qaij ‘afternoon’ to produce the time phrase ‘yesterday afternoon’. Temporal
shifters can also co-occur with With Respect to Expectation adverbs, as in (29) below.
(29)

Peb twb
nrhiav tau peb lub hom phiaj
1PL already search ATT 1PL CLF purpose
‘We have already found our purpose yesterday.’

tas
nag hmo lawm.
finish yesterday CP

In this case, the With Respect to Expectation morpheme twb ‘already’ appears in the
same sentence as the Temporal shifter nag hmo ‘yesterday’. As a result, it is clear that the
different classes of adverbs in the Lexical Time Words category can co-occur.
3.1.2 ntshai ‘maybe’
The morpheme ntshai ‘maybe’61 expresses a moderate level of certainty (following
Mottin 1978:115) as part of Degree of Certainty. It is identical in form to the verb ntshai
‘be afraid’,62 and is placed at the beginning of its clause, before the subject. An example
appears in (30).
(30) Ntshai nws yuav
tuaj
maybe 3SG POS.IRR come
‘Maybe he will come to visit me.’

xyuas
visit

kuv
1SG

(Mottin 1978:115)

Here, ntshai ‘maybe’ appears at the beginning of the clause, before the subject nws ‘3SG’.

61
62

Gloss following Mottin (1978:115).
Gloss following Fuller (1988:20).

51
This form of ntshai is a grammaticalized item since its meaning has undergone
semantic bleaching to become a Degree of Certainty marker. To be sure, examples such
as (30) above could be interpreted as ‘(I) fear he will come to visit me’, though Mottin
(1978:115) points out that this form can be used “with or without [the] idea of fear 63.”
Since the ‘fear’ meaning has been lost in sentences such as (30) above, and the ‘maybe’
sense is the result, it is clear that semantic bleaching has occurred. On the basis of this
and the resulting semantics, it is certain that ntshai ‘maybe’ marks Degree of Certainty.
As for the larger system of which ntshai ‘maybe’ is a part, Mottin (1978:115)
demonstrates that words at this level of certainty can often be used with others, such as
tej zaum ‘maybe’. As a result of this, it is clear that at least some markers of Degree of
Certainty can co-occur and are thus not part of a paradigm that requires the use of only
one marker at a time.
3.1.3 tej zaum ‘maybe’
Tej zaum ‘maybe’64 is a Degree of Certainty marker that indicates a moderate level of
certainty (following Mottin 1978:115). It appears in clause-initial position. An example
appears in (31) below.
(31) Tej zaum
nws yuav
tuaj
maybe
3SG POS.IRR come
‘Maybe he is going to come.’

(Mottin 1978:115)

Here, tej zaum ‘maybe’ appears before the subject pronoun nws ‘3SG’, demonstrating its
meaning and its placement at the beginning of the clause before the subject pronoun.
Tej zaum is clearly a grammatical marker because its semantics have undergone a
shift from its literal part-by-part meaning, showing signs of semantic bleaching. This is

63
64

Translation of French crainte from Kellogg (2014).
Gloss following Mottin (1978:115).

52
evidenced by the fact that its individual parts are ‘some’ 65 and ‘time’, 66 which does not
have an immediate relation to ‘maybe’, 67 showing a change in meaning towards one with
a more grammatical function. Altogether, since tej zaum indicates a medium level of
certainty with the meaning ‘maybe’ and is known to be a grammaticalized element, it
belongs to Non-spatial Setting as a marker of Degree of Certainty.
In terms of the larger system of Degree of Certainty, Mottin (1978:115) mentions
that certain multiple forms of Degree of Certainty can be used together. This may also be
true even with greater Degrees of Certainty, such as yeej ‘certainly’.68
3.2 Pre-modal position
Grammatical items in this position include a wide range of adverb-like morphemes,
namely, the With Respect to Expectation adverbs tseem ‘still’ and twb ‘already’, the
Temporal Shifter nyuam qhuav ‘just, a moment ago’, the Irrealis adverb mam (li), the
Continuing action marker sij ‘continue to’, the Degree of Certainty markers kiag (li)
‘really’, tiag ‘really’, and yeej ‘certainly’, the Imperfect marker tab tom ‘IMPERFECT’, and
the Speed and Ease marker maj mam ‘slowly, progressively, gradually’.
3.2.1 tseem ‘still’
The word tseem is glossed as ‘still’ by Jarkey (1991:75) and is often regarded by scholars
to be a type of grammatical marker, indicating “imperfective aspect” (Jarkey 1991:75), a
“temporal/aspectual distinction” (Fuller 1988:20), or “action still in progress” (Heimbach

65

This is the gloss that Jay Yang generally provided, and is corroborated by Mottin (1978:54). It is notable,
however, that Mottin gives the relevant construction as tej CLF (NOUN), while zaum ‘time’ appears without
a classifier in tej zaum. However, the texts obtained from Jay Yang have several examples where the
construction is tej + noun.
66
Gloss in line with Heimbach (1969:435).
67
Tej zaum can also mean ‘sometimes’ (Mottin 1978:115), which suggests that it has undergone mild
grammaticalization (in this case decategorialization from a noun phrase to an adverbial phrase) in other
directions as well.
68
Please see example (70) in section 3.2.8 below for a case where tej zaum co-occurs with yeej ‘certainly.’

53
1969:350). The semantics of tseem as ‘still’ suggest that this morpheme is a Lexical Time
Word of the With Respect to Expectation variety.
Examples of tseem in use appear in (32) and (33) below.
(32) Kuv
tseem
1SG
still
‘I still run.’

khiav.
run

(Fuller 1988:20)

In (32), tseem appears with khiav to indicate the ongoing situation translated as ‘still run’.
(33)

Lawv
haj tseem noj mov.
3PL
still
eat rice
‘They are still eating.’

(Heimbach 1969:350)

Here, tseem appears with noj mov ‘eat rice’ to produce ‘still eating’, that is, the ongoing
situation where the agents lawv ‘3PL’ continue to eat. Note here the additional element
haj, which, Heimbach (1969:350) suggests can optionally appear with tseem, without a
noticeable change in meaning.
Tseem is often used in negative contexts in conjunction with tau ‘ATTAINMENT’ to
express unrealized situations. Mottin (1978:102) even asserts that whenever tsis tau ‘have
not’ is used, tseem also appears. This usage is shown below in (34) and (35).
(34)

(a)

...rau qhov
tshuav peb cov nplej nyob rau tom teb
because
have
1PL CLF rice in
to
there farm
‘...because we still had unhulled rice in the farm that...’

(b)

tseem
ua tsis tau
still
do NEG ATT
‘...still needed to be done.’

tiav
finish

Here, the phrase ua tsis tau tiav ‘is not done’ is accompanied by tseem ‘still’ to indicate
an ongoing lack of realization, in this case, of the completed action of threshing rice.

54
(35) Nws tseem
tsis tau
3SG
still
NEG ATT
‘He still has not gone.’

mus.
go

(Heimbach 1969:350)

In this example, tseem appears with tsis tau and the verb mus ‘go’ to express the meaning
‘still has not gone’—that is, the ongoing lack of realization of the agent’s going.
Grammatically, tseem behaves similarly to tab tom ‘Imperfect’ and twb ‘already’
as a morpheme that is placed between the sentence subject and the negative (following
Jarkey 1991:75) or a Modal. 69 This is demonstrated by the four examples below.
(36) Peb tseem
tab tom
nrhiav
1PL still
IPRF
search
‘We are still looking for our car.’

peb
1PL

lub
CLF

tsheb.
car

In (36), tseem is placed after the subject peb ‘1PL’ and before tab tom ‘IMPERFECT’ and
the verb nrhiav ‘search, look for’, producing a well-formed sentence.
(37)

*Tseem
peb tab tom
nrhiav
still
1PL IPRF
search
‘Still we are looking for our car.’

peb
1PL

lub
CLF

tsheb.
car

In (37), tseem has been placed at the beginning of the sentence, where other classes of
adverbs can appear. However, this results in an ungrammatical sentence.
(38)

*Peb tab tom
nrhiav peb
1PL
IPRF
search 1PL
‘We are looking for our car still.’

lub
CLF

tsheb
car

tseem.
still

Here, tseem appears at the end of the sentence—a possible place for other varieties of
adverbs—but this renders the sentence ungrammatical. Thus, tseem must be placed
between the subject and the verb.

69

Note that one instance of tab tom following yuav was elicited, though this ordering is not preferred by
Jay Yang.

55
(39) Nws
tseem
tsis
3SG
still
NEG
‘He still has not gone.’

tau
ATT

mus.
go

(Heimbach 1969:350)

In this case, tseem appears before tsis tau mus ‘has not gone’, showing that tseem is
placed before the negative.
(40) Peb tseem
yuav
mus peb txoj kev.
1PL still
POS.IRR go
1PL CLF
way
‘We will go our own way (but we haven’t left yet).’
In (40), tseem appears before the Modal-type word yuav ‘POSITIVE IRREALIS, intend’. As
it has been seen, tseem must be placed between the subject and either the first verb, the
negative marker tsis, or a Modal-type word such as yuav, whichever occurs first. This
suggests that tseem is in the same class of words syntactically as tab tom ‘IMPERFECT’ and
twb ‘already’. However, since tseem can co-occur with tab tom, the two do not form part
of a single grammatical paradigm. In addition, while tab tom fills the role of Imperfect,
tseem fits naturally into the With Respect to Expectation subcategory of adverbs in Dixon
(2012)’s Tense category, along with twb, with which it shares at least some of its
grammatical tendencies. 70
3.2.2 twb ‘already, indeed’
The morpheme twb is an adverb meaning ‘already’, though it serves several other
functions, with possible meanings such as ‘indeed, even, really’ (Heimbach 1969:330).
With the meaning ‘already’, twb is a Lexical Time Word, of the With Respect to
Expectation variety. In its extended meaning as ‘indeed’, it is a Degree of Certainty
marker indicating strong certainty.
An example of twb in use with the meaning ‘already’ appears in (41) below.
70

It is notable here that there is a syntactic unity in White Hmong between markers of Imperfect as part of
Completion and Lexical Time Words of the With Respect to Expectation variety, such that the distinction is
blurred, at least syntactically.

56
(41) Tej zaum
peb twb
muaj coob
tug
sim
maybe
1PL already have
many CLF try
‘Maybe we have many people who have tried it already.’

tas
lawm.
finish CP

Here, twb appears before the verb muaj ‘have’ and after the subject pronoun peb ‘1PL’,
producing ‘we have...already’.
My language consultant, Jay Yang, suggested that twb has two meanings: one
being ‘already’, and the other providing a nuance with a range that includes ‘already’ but
could mean ‘indeed’ as well. An example of each of these uses is found below in (42) and
(43), respectively.
(42) Peb twb
tab tom
nrhiav peb lub
1PL already IPRF
search 1PL CLF
‘We are currently looking for our car already.’

tsheb.
car

In (42), twb appears after the subject pronoun peb ‘1PL’ and before tab tom ‘Imperfect’
and the verb nrhiav ‘search’, forming the phrase ‘we are currently looking...already’.
(43) Peb tab tom
twb
nrhiav peb lub
tsheb.
1PL IPRF
indeed search 1PL CLF car
‘We are currently looking for our car.’ or ‘We have been looking for our car.’
In (43), twb appears in a different position—after the Imperfect marker tab tom and
immediately before the verb nrhiav ‘search’, and the ongoing process of the search is
reflected in ‘we are currently looking’ or ‘we have been looking’, whereas the notion of
‘already’ is not represented in the language consultant’s translation. 71 In addition, Jay
Yang’s initial intuition is that twb does not mean ‘already’ in this case, though he
confirmed that ‘already’ could be a component; he confirmed that it could mean ‘indeed’,
and suggested that this example is referring to the confirmation that the action is already
happening as a confirmation to the hearer. In the literature, Heimbach (1969:330) defines

71

The language consultant later said that twb here gives the nuance ‘already’ in that the event expressed by
the sentence is already in progress, rather than that it has already happened.

57
twb as a “preverbal particle drawing attention to the state of affairs at the time,” while
Jarkey (1991:68) calls it a “status morpheme” that “is used when the speaker wishes to
indicate certainty concerning the truth of the proposition.” This confirms the secondary
reading for twb as ‘indeed’, showing that it functions in an extended use to indicate a
high Degree of Certainty. It is clear, then, that twb’s semantics, while canonically
centered around ‘already’, can also serve to indicate a sort of emphatic confirmation that
an event is actively occurring.
Syntactically, twb can only appear in pre-modal position, using the ordering
scheme presented above. Attempting to place it in pre-subject position or “final adverbs
and particles” position results in an ungrammatical sentence. This is shown in (44) and
(45) below.
(44)

*Twb
peb nrhiav peb lub
already
1PL search 1PL CLF
‘Already we are searching for our car.’

tsheb.
car

Here, twb has been placed in clause-initial position, before the first subject, resulting in
an ungrammatical sentence.
(45)

*Peb nrhiav peb lub tsheb twb.
1PL
search 1PL CLF car
already
‘We are searching for our car already.’

In this case, twb has been placed at the end of the clause, again resulting in an
ungrammatical utterance. Thus, twb needs to be placed in the attested position, that is,
between the subject noun and the verb.
The semantics of twb suggest that it is a Lexical Time Word belonging to the
category With Respect to Expectation, per Dixon (2012:20). As such, with the meaning
‘already’, it indicates that an event has taken place at a time other than what is expected.
From this, the notion of an emphatic ‘indeed’ is a secondary extension. Its syntactic

58
distribution, however, is similar to that of tab tom ‘IPRF’ or tseem ‘still’ in that it only
appears after the subject and before the first verb. Drawing on its semantics and its status
as an adverb, twb should be treated as 1) a Lexical Time Word within the Tense category
of Dixon (2012) meaning ‘already’ and 2) a Degree of Certainty marker indicating strong
certainty with the meaning ‘indeed’, while noting that twb generally forms a natural class
syntactically with tab tom ‘IMPERFECT’ and tseem ‘still’.
3.2.3 nyuam qhuav ‘just, a moment ago’
The phrase nyuam qhuav72 generally means ‘moment, short time’ (Heimbach 1969:222).
As a grammaticalized element 73, it appears before the verb and means ‘just’ or ‘a moment
ago’ (gloss following Heimbach 1969:222), and is considered a marker of “immediate
past” by Mottin (1978:84). As such, nyuam qhuav is a Lexical Time Word of the
Temporal shifters type as part of Dixon (2012)’s category of Tense.
Examples of the grammaticalized form of nyuam qhuav appear below.
(46) Nws
nyuam qhuav
3SG
just
‘He just arrived.’

tuaj
come

txog.
arrive

(Mottin 1978:84)

In (46), nyuam qhuav appears after the subject pronoun nws ‘3SG’ and before the verb
complex tuaj txog ‘arrive’, providing here the meaning ‘just’.
(47) nyuam qhuav
mus
just/a.moment.ago go
‘Just went, went just a moment ago.’

(Heimbach 1969:222)

In (47), nyuam qhuav is placed before the verb mus to give the meaning ‘just went’. In
both cases, nyuam qhuav marks a recent event.
72

Nyuam can appear in a number of time phrases, such as ob peb nyuag hnub ‘a few days’ or ib nyuag qhov
‘a little bit, a little place’ (Heimbach 1969:222), where nyuam is undergoing a standard White Hmong tone
sandhi change. Its usage with numerals here and with no ‘this’ in example (48) suggest that it sometimes
serves as some sort of time classifier. Such considerations, however, are outside the scope of this paper.
73
Note that this analysis as a grammaticalized element is based on the available limited data. It was not
attested in the data elicited from my language consultant, Jay Yang.

59
Nyuam qhuav is grammaticalized, because 1) it does not appear in its normal
location as a noun phrase (decategorialization), and 2) its meaning has been weakened to
that of a grammatical entity (semantic bleaching). First, nyuam qhuav does not appear in
its usual location. A full noun phrase with non-grammaticalized nyuam qhuav ‘moment’
(following Heimbach 1969:222) can appear sentence initially as a sort of topic, as in (48)
below.
(48) Nyuam qhuav
no,
moment
this
74
‘He just arrived.’

nws
3SG

tuaj
txog.
come arrive

(Mottin 1978:84)

Here, nyuam qhuav appears with the demonstrative no ‘this’, and the whole phrase has
been placed as a topic for the sentence nws tuaj txog ‘he arrived’. With the natural
English translation providing the meaning ‘just’, it would appear that nyuam qhuav no
‘this moment’ is being used adverbially in this context as well, though it is otherwise
behaving as a normal noun phrase in a normal location. On the other hand, the more
grammaticalized equivalent in (46) and (47) is placed in an unusual location for a normal
noun phrase, that is, between the subject and the verb. Thus, nyuam qhuav’s unusual
syntactic behavior in cases like (46) and (47) above suggests that it has been
grammaticalized.
Second, the meaning of nyuam qhuav in this grammatical usage has been
weakened from its literal meaning. In a typical noun phrase, nyuam qhuav means
‘moment’ or ‘short time’, as shown in (49) below.

74

Here, note that while the noun phrase nyuam qhuav no appears in a normal position for noun phrases,
showing no signs of decategorialization, the English gloss suggests that the phrase has still undergone
semantic bleaching even in this position.

60
(49) ib
nyuam qhuav
one moment/short.time
‘A moment, a short time.’

(Heimbach 1969:222)

Here, nyuam qhuav is used with the numeral ib ‘one’, to produce the meaning ‘a moment’
or ‘a short time’, presenting itself as a noun phrase. However, in the examples above,
each was rendered into English with the words ‘just’ or ‘a moment ago’, suggesting a
shift and possibly weakening in meaning, and so it is clear that semantic bleaching has
occurred. As a result, there is clear evidence that this phrase has undergone
grammaticalization, where it is a noun phrase operating as an adverbial with a slightly
weakened meaning. In sum, nyuam qhuav, when it appears alone, is a slightly
grammaticalized adverbial that ultimately functions as a Lexical Time Word for purposes
of Non-spatial Setting.
3.2.4 mam (li) ‘will’
The morpheme mam75 or its phrasal equivalent mam li is an adverb that can denote a
range of modalities. These modalities include future actions, 76,77 Necessity, and
Permission in Dixon (2012)’s system of Modality. Examples of each of these three uses
are given below.
(50) Kuv mam (li)
1SG will
‘I will go’

mus.
go

In (50), mam appears with the verb mus ‘go’, together meaning ‘will go’. This clearly
indicates future action (or possibly intent).

75

The existing scholarly literature only treats mam as a form meaning ‘then,’ which is attested in the data I
collected but is beyond the scope of this paper. Please see Mottin (1978) and Heimbach (1969) for this use
of mam.
76
My language consultant, Jay Yang, used Bing Translator in order to gloss mam, which provided the
translation ‘will’; it does not appear, however, that his free translations necessarily took this gloss into
account, since he provides his own translations of several varieties.
77
I double-checked this with two online Hmong-authored dictionaries, Xiong (2012) and Xiong (2014).

61
(51)

(a) Ces
koj
mam (li)
muab
then 2SG need
take.hold.of78
‘Then, you need to stir until…’

do
stir

rau
to

kom
so.that

(b) zoo zoo
tib si
tas.
really.good together complete
‘…it is mixed really well.’
In (51), mam li appears with muab do ‘stir’ to make ‘will need to stir’. This use of ‘need’
shows that mam li here expresses Necessity.
(52) ...ces koj mam (li) noj mas zoo
nkaus li qab
79
then 2SG may
eat IP
good look.like tasty
‘...and you may eat, it looks really good.’

kawg li
very80

In (52), mam li combines with noj ‘eat’ and the resulting form is translated ‘may eat’—a
form of permission. Thus, mam and mam li can express a range of Modal meanings.
Though mam (li) has some range of meaning, the canonical meaning of the
morpheme is that of marking general future action. This is evidenced by the intuitions
and translations provided 81 by my language consultant, Jay Yang, as well as online
Hmong dictionaries made by members of the Hmong community82, which, more often
than not, translate mam (li) as ‘will’. 83
Syntactically, mam (li) is an adverb, rather than a Modal or Semi-modal. Modals,
like yuav, typically appear before the first verb or the negative marker tsis, though they
can never be followed by an adverb. Semi-modals, such as tau, behave to some extent
like regular verbs (following Dixon 2012:26), in that they appear after the negative
marker tsis and cannot under normal circumstances be followed by an adverb. Mam,
78

Gloss from Heimbach (1969:132).
Gloss following Fuller (1988).
80
Gloss following Mottin (1978:110).
81
Note that Jay Yang relied on outside sources for this, but chose ‘will’ as his gloss of choice.
82
These include Xiong (2012) and Xiong (2014), both of which define mam as ‘will.’
83
It should be noted here that Jay Yang also gave data for mam (li) that gave it the nuance ‘you are the one
to V.’ While this usage does not appear to belong to Non-spatial Setting, but rather marking exclusivity, it
is worth mentioning in reference to this morpheme.
79

62
however, must be placed in pre-modal position, in a pattern identical to a number of other
morphemes that belong to Non-spatial Setting such as yeej ‘certainly’, tab tom
‘Imperfect’, and tseem ‘still’. This is shown in (53) and (54) below.
(53) Kuv
mam yuav
tso
koj
1SG
will
POS.IRR permit 2SG
‘I will be letting you go.’

tseg.
leave84

Here, mam ‘will’ appears before the Modal-type word yuav ‘POSITIVE IRREALIS’ and
follows the subject pronoun kuv ‘1SG’. This shows that mam is placed after the subject
and before yuav, like the other adverbs mentioned above.
(54) *Koj yuav
mam muab ob lub qe tsoo tso rau hauv.
2SG POS.IRR will get
two CLF egg break put to
in
‘You will get two eggs and break them and put them in there.’
In this case, mam ‘will’ is placed after yuav ‘POSITIVE IRREALIS’, rendering the sentence
ungrammatical. Thus, mam must precede yuav, similar to adverbs such as tseem ‘still’
and twb ‘already’. 85
Furthermore, mam can both precede and follow other adverbs, much like the
adverbs tseem and twb. This is shown in (55) and (56) below.
(55) Kuv niaj hnub mam li86 muab
87
1SG everyday will
CVB
‘I will let you go everyday.’

koj tso
2SG permit

tseg.
leave

In (55), niaj hnub ‘everyday’ precedes mam li, showing that mam li can follow another
adverb.
84

Glosses for tso and tseg are from Heimbach (1979:354). Also in (55) and (56).
Note that the Imperfect marker tab tom also belongs to this class of words, though grammatically it
signals the Imperfect; please see below for the larger discussion.
86
Note that in this context, mam li is preferred to mam by the language consultant Jay Yang. In contrast, in
(50), either can occur. On the other hand, when it appears with the adverb yeej, mam can appear both before
and after yeej without any issue grammatically. The reason for these co-occurrence restrictions is unknown,
but since such nuances involve aspects of the language other than Non-spatial Setting, this is beyond the
scope of this thesis.
87
Gloss following Jarkey (1991). Also found in following examples. Note that muab can be either a coverb,
as in this example, or a full verb meaning ‘get,’ as in the previous example above.
85

63
(56) Kuv
mam (li) niaj hnub
1SG
will
everyday
‘Everyday, I will let you go.’

muab
CVB

koj
2SG

tso
tseg.
permit leave

In (56), mam (li) precedes niaj hnub, demonstrating that mam (li) can also precede
another adverb. Thus, mam li can appear both before and after other pre-verbal elements,
like tseem and twb can. As a result, the word order also suggests that mam, like tseem and
twb, is an adverb.
In terms of distribution, mam does not act like a Modal. Jaisser (1984) provides an
excellent set of seven tests to determine whether a certain word belongs to the class of
verbs. In general, true verbs can co-occur with each of seven grammatical items—one in
each test—while words belonging to other classes will either fail some of these tests or a
different meaning may emerge when the tests are applied (following Jaisser 1984:50).
These tests can be extended to compare words of similar classes, such as adverbs and
Modals. The basic idea here is that if two words are the same kind of adverb, for example,
then they should have more or less the same results with these seven tests. The seven
tests are the following: 1) the negative tsis; 2) the question particle puas; 3) the “past
tense marker” tau; 4) the “future tense marker” yuav; 5) the negative imperative with
txhob; 6) the post-verbal ‘can, be able to’ “modality” tau; and 7) the question
construction with los tsis V at the end, where V is a repetition of the word under
consideration (Jaisser 1984:48-50).
Applying the tests from Jaisser (1984), mam fails the tests in ways that are distinct
from the Modal-type word yuav ‘POSITIVE IRREALIS, intend.’ The test results for mam in
comparison with yuav are shown in Table 10 below.

64
Test Type

mam
yuav
negative tsis
yes
no
question particle puas
yes
yes
“past tense” tau
no
no
“future tense” yuav
no
no (illogical)
negative imperative txhob
no
questionable
post-verbal ability tau
questionable
yes*88
question construction with los tsis V
questionable
yes*
Table 10. Comparison of verbal test results with mam and yuav.
In test 1, mam and yuav clearly behave differently, where mam is allowed after the
negative tsis while yuav is not. In addition, in test 7, mam produces a questionable result,
while yuav produces a grammatical sentence. Thus, on the basis of the verbal tests, mam
behaves in some ways that are different from yuav. This shows that, on the basis of
restrictions on distribution, mam is not a modal, nor is it a regular verb, since it would
otherwise pass all seven tests, as Jaisser (1984:48-50) explains. Altogether, from these
two strands of evidence, the syntactic and the distributional, it can be concluded that mam
is an adverb.
The fact that mam is an adverb presents interesting implications for Non-spatial
Setting in White Hmong, in that while mam is semantically a type of Modality,
grammatically it is not a verb or Modal in Dixon (2006, 2012)’s frameworks as would be
expected. This suggests that, for purposes of analyzing the system of Modality in White
Hmong, grammatical elements other than verbs should be considered. As it will be seen
below, the inclusion of non-Modal, non-verbal elements in the analysis of Modality in
Non-spatial Setting will prove fruitful. In either case, the analysis that mam is an adverb
that marks semantic Modality is explicitly adopted here.

88

Asterisks mark cases where more explanation is needed to fully reflect the effects of the test. This
explanation is provided in prose immediately following the table in each case. This method of marking
such instances is found throughout this thesis.

65
As part of the larger system of Modality in White Hmong, mam (li) ‘will’
performs an interesting role: it can mark Modality or Irrealis by itself, as in (50) above,
where it qualifies the word mus ‘go’ to indicate a future event, or it can co-occur with the
Modal-type word yuav ‘POSITIVE IRREALIS, intend’, as in (53), where the two together
indicate a future event. As a result, it is clear that mam (li) does not have a paradigmatic
opposition to the other morphemes that serve a Modality-marking role, but in fact
reinforces the other morphemes’ Modality- or Irrealis-marking function.
3.2.5 sij ‘continually, repeatedly’
The morpheme sij means ‘continually, repeatedly’ (Heimbach 1969:293) and marks a
variety of Imperfect 89 as part of Completion. An example of sij appears in (57).
(57) sij
ua
continually/repeatedly do
‘To do continually or to do repeatedly.’

(Heimbach 1969:293)

In (57), sij appears before the verb ua ‘do’, resulting in the meaning ‘to do continually’ or
‘do repeatedly’, showing the two possibilities for the meaning of sij.
The morpheme sij is an adverb, as shown in (58).
(58) Kuv
sij
yuav
vamkhom.
1SG
continue.to POS.IRR trust
‘I will continue to trust.’

(Heimbach 1969:293)

In (58), sij appears after the subject pronoun kuv ‘1SG’ and before the Modal-type word
yuav ‘POSITIVE IRREALIS, intend’—in the pre-modal slot. Note also here that sij provides
the meaning ‘continue to’, suggesting some potential connection with Phase of Activity. 90

89

In parallel to this, Jarkey (1991:75) states that sij marks “imperfective aspect,” though her imperfective at
times seems to cover both Imperfect and Imperfective in Dixon (2012)’s framework.
90
Consistent with this is the suggestion by Heimbach (1969:293) that sij is comparable to pheej ‘continue
to,’ and the grouping together of pheej and sij by Jarkey (1991) as part of her set of “imperfective aspect”
morphemes.

66
Additional corroborating evidence that sij is an adverb is found in Jarkey
(1991:75), where she considers that “imperfective aspect” is marked 91 with adverbs92 in
White Hmong, of which sij is one, and she considers that these markers appear in one
location: “after the subject but before the negative.”
In terms of Non-spatial Setting, sij should be treated as an adverb of the same
class as tab tom ‘IMPERFECT’—in line with Jarkey (1991:75)—for two reasons: 1) it
appears syntactically in the same position as tab tom, and 2) it possesses the same
semantic property of unfinished, ongoing action as tab tom. In regard to the category of
Non-spatial Setting, while sij does have aspects of the notion of ongoing action as other
adverbs of its class, its semantics also spread into Phase of Activity with the meaning
‘continue to’ and Frequency and Degree with the meaning ‘repeatedly’. However, since
this semantic spread is similar to that of reduplication, another form of the imperfect, as
well as the fact that it forms a natural class, semantically and syntactically, with tab tom
‘IMPERFECT’, its canonical meaning and use is that of an Imperfect marker as part of
Completion. As part of the larger system, however, no examples from the data are known
that provide any examples where sij and tab tom co-occur, suggesting a potential
paradigmatic relationship; however, further research is necessary to determine this with
certainty.
3.2.6 kiag ‘really’
The word kiag, and its variant form kiag li (Heimbach 1969:84), is an adverb supplying
the notion ‘really’ or ‘completely’ (glosses following Heimbach 1969:84, Mottin
91

A number of the morphemes Jarkey (1991) lists as “imperfective aspect,” such as tab tom ‘currently,’
tend to primarily perform the function of the Imperfect in the data I obtained from Tzerge Yang.
92
Tab tom, tseem, pheej, and yeej are also treated as members of this class of adverbial “imperfective”
markers (Jarkey 1991:75). As seen above, at least tab tom, pheej, and yeej do not function like other
adverbs. However, it is notable that in Jarkey (2006), pheej is treated as a verb; see above for the discussion.

67
1978:114).93 It belongs to Degree of Certainty in Dixon (2012)’s Non-spatial Setting, and
can occur in one of several places in the clause, namely, pre-modal, post-verbal, and
“final adverbs and particles” positions, as well as between verbs.
An example of kiag (li) in context is (59) below.
(59) Tsis

tau
kev
mus qhov twg
can
road go
where
‘I couldn’t really go anywhere.’
NEG

kiag li.
really

(Mottin 1978:114)

In (59), kiag li follows the phrase tsis tau kev mus qhov twg ‘couldn’t go anywhere’, and
supplies the meaning ‘really’.
Kiag, like tiag (tiag), has been treated in the literature as a grammatical element
of sorts that can express a level of certainty. Heimbach (1969:84) calls kiag a “post verbal
intensive indicating decisiveness and completeness,” while Mottin (1978:109,114)
regards it as an adverb that “expresses certainty.” Mottin’s assessment seems to be
correct at first, since kiag generally behaves like an adverb syntactically: when it is used
with the meaning “really,” it appears in the “final adverbs and particles” position, just
before lawm. An example of this appears in (60) below.
(60) ...kom
cov roj
so.that
CLF
oil
‘...until the oil is hot.’

ntawd
there

kub
hot

kiag
really

lawm.
CP

In (60), kiag appears at the end of the sentence immediately before lawm, like other
adverbs.
Two unusual syntactic traits that distinguish kiag from normal adverbs do appear,
however: 1) kiag can be placed immediately after the copula verb in a sentence

93

Kiag also serves a second purpose as a marker that apparently expresses emphasis, which came up a
small number of times in the secondary data obtained from Jay Yang. This second meaning may be more
salient in the examples given below.

68
containing a nominalized verb phrase, and 2) kiag can occur between verbs. This is
shown in (61) and (62) below, respectively.
(61) Qhov

nov yog
kiag
qhov
this COP the.one NMLZ
‘This is the one that I want.’
CLF

kuv
1SG

xav
want

tau.
get

In this case, kiag is placed immediately after the copula yog and before the noun phrase
complement qhov kuv xav tau ‘the one that I want’. Thus, unlike other adverbs, it can
occur in post-verbal position, before the object.
(62) nqos
kiag
mus
swallow whole/right.down go
‘To swallow whole’ or ‘To swallow right down.’

(Heimbach 1969:84)

Here, kiag appears after the verb nqos ‘swallow’ and before the verb mus ‘go’. Thus, kiag
can occur between two verbs. From these two examples, it is clear that kiag has a wider
distribution than that of normal adverbs—showing that it has undergone an even greater
level of grammaticalization than its adverbial counterparts.
Since kiag behaves as something other than a typical adverb, and since its
meaning in a number of contexts contributes a high level of certainty to what is being
said, the morpheme belongs to the Degree of Certainty category in Dixon (2012)’s
framework. As part of the system of Degree of Certainty in White Hmong, the data
shows that kiag generally does not co-occur with other members of the set, and seems to
be functionally very similar to tiag ‘really’. However, further research may prove
beneficial.

69
3.2.7 tiag ‘really’
The word tiag and its reduplicated version tiag tiag means ‘really’ 94 and is an adverb that
is placed either before the verb or at the end of a sentence. Tiag belongs to Degree of
Certainty, as expressing a high level of certainty.
Examples appear in (63) and (64) below.
(63) Peb lub hom phiaj yog
1PL CLF purpose
COP
‘What is really our purpose?’

dab tsi
what

tiag tiag?
really

Here, tiag tiag occurs at the end of the sentence, adding the meaning ‘really’.
(64) Peb lub hom phiaj tiag tiag
1PL CLF purpose
really
‘What, really, is our purpose?’

yog
COP

dab tsi?
what

In this case, tiag tiag ‘really’ appears after the subject phrase peb lub hom phiaj ‘our
purpose’ and before the copula verb yog ‘be’, with the same resulting meaning. Thus,
tiag tiag can occur in two different positions, both of which are typical for adverbs. It
should be noted however that tiag tiag most often appears at the end of the sentence; in
every case of tiag tiag in the three texts I collected, tiag tiag always occurred near the end
of the sentence.
Tiag tiag is relevant to the discussion of Non-spatial setting since some authors,
have treated it as a particle of emphasis meaning ‘really, truly’ (Heimbach 1969:319), or
of certainty (Mottin 1978:114). Heimbach (1969)’s treatment is especially notable, since
in this usage he regards it as a “final emphatic particle,” suggesting a grammaticalized
status. However, since examples such as (64) are possible, it is clear that tiag tiag has the

94

It should be noted here that in the Hmong-community dictionary “Hmong Dictionary Online” (Xiong
2014), this is glossed as the adjective ‘real.’ Whether this operates as an adjective as well as an adverb is
unknown, since all academic sources consulted treat tiag as a grammatical entity of some sort.

70
syntactic distribution properties of an adverb. 95 At the same time, the fact that tiag is
similar in function—as well as to a significant extent, distribution—to kiag (li), combined
with the fact that tiag can be reduplicated as tiag tiag with no change in meaning
(suggesting semantic bleaching to some extent), suggests that tiag is a function word,
though admittedly with the syntactic distribution of an adverb.
As part of the larger system of Degree of Certainty, tiag generally does not cooccur with its very similar counterpart kiag (li) ‘really’, suggesting a paradigmatic
relationship of sorts. However, since these two words have basically the same function
and generally occur at similar places in the sentence, it could be that their phonetic
similarity, rather than the existence of a contrastive paradigmatic relationship, prevents
them from co-occurring. However, considerations of how similarity in form and meaning
could affect the grammar of White Hmong are outside of the scope of this thesis, and are
left for further research.
3.2.8 yeej ‘already, certainly, can’
The word yeej is a morpheme that is likely derived from the verb yeej ‘to win in an
encounter, have success’96 and appears in grammaticalized form as either an adverb or a
Semi-modal verb with limited distribution. The adverb will be dealt with first, while the
Semi-modal will be discussed further below.
The adverb has functions associated with two meanings: 1) ‘already’ (Fuller 1988)
or ‘originally’ (Heimbach 1969; Jarkey 1991), and 2) ‘certainly’ (following Mottin
1978:114) or ‘definitely’ (Jarkey 2006). With the second function, yeej denotes a high
Degree of Certainty. Each of the two possibilities is shown in (65) and (66) below.
95

The one unique thing about tiag (and kiag) is that it is one of a very small number of adverbs that can
appear both at the beginning of the verb phrase and at the end of the sentence.
96
Gloss per Heimbach (1969:424).

71
(65) Nws
yeej
tuaj.
3SG
already come
‘S/he already came.’

(Fuller 1988:20)

In (65), the word yeej supplies the meaning of ‘already’ and appears between the subject
nws ‘he, she’ and the verb tuaj ‘come’.
(66)

(a)

(b)

(tias-)kom Lis mus mas, kuv yeej
(that-)PC
Li
go
TOP
1SG definitely
‘To have Lee go, I definitely forced them.’
lawv mas
3PL
IP
‘...them.’

tau
ATT

yuam
force

(Jarkey 2006:127)

In (66), yeej supplies the meaning of ‘definitely’ (per Jarkey 2006:127), and appears after
the subject pronoun kuv ‘1SG’ and before the rest of the verb phrase tau yuam ‘forced’.
As a marker of certainty, yeej patterns in two ways depending on the point in time
relative to the present: if in past situations, it is commonly used for certainty regarding
one’s own actions (following Jarkey 2006:127), while in non-past situations, it can
express certainty in relation to existence, ability, or outcome. An example of certainty in
the past is found above in (66). There, yeej appears with tau yuav ‘forced’ to express the
speaker’s absolute certainty over his/her past action, in effect assuring the listener that
this was in fact the case.
When used as an indicator of high certainty for non-past events, yeej is often used
in one of three contexts: certainty of existence, certainty of ability, and certainty of
outcome. Examples of each of these three are shown below.
(67) Nws yeej
muaj
lub
hom.phiaj zoo
3SG
certainly have
CLF
purpose
good
‘He has a good purpose for you and me...’

rau
for

koj
2SG

thiab
and

kuv...
1SG

In (67) above, yeej is used with muaj ‘exist, have’ to express certainty over the existence
of the hom.phiaj ‘purpose’.

72
(68)

Peb yeej
yuav
nrhiav tau
peb
1PL certainly POS.IRR search can
1PL
‘We will be able to find our purpose tomorrow.’

lub
CLF

hom phiaj
purpose

tag kis.
tomorrow

In (68), yeej appears with the sequence yuav nrhiav tau ‘will be able to find’, expressing
certainty over the ability.
(69)

(a)

...ces
peb yuav tsum rau siab ntso mus khwv tiag tiag...
then
1PL must
commit.to
go
work really
‘...then we must commit ourselves to working really hard...’

(b)

...ces peb yeej
yuav tsum
then 1PL certainly must
‘...then we will have money.’

muaj
have

xwb97
indeed

nyiaj
money

In (69), yeej appears with yuav tsum muaj nyiaj ‘will have money’ to express the certain
outcome of the action expressed in the previous clause by yuav tsum rau siab ntso mus
khwv ‘must commit ourselves to work’. Thus, in non-past contexts, there are three main
uses for the degree of certainty variety of yeej.
A final interesting trait about yeej when marking expectation of outcome is that it
occasionally can be used with lesser degrees of certainty, with the notion of outcome left
intact. This is exemplified in (70) below.
(70)

97

(a)

...tej zaum nws yeej
pab tau yus
maybe
3SG certainly help can INDF.PRON
‘...maybe it will be able to help you...’

(b)

ua tau
lub
neej vas meej
do get
CLF
life
successful
‘...to become successful if you’re....’

yog
if

kom
yus
so.that INDF.PRON

hais tias
that

yus
INDF.PRON

Gloss following Strecker & Vang (1986:5); please see section 3.9 below for the discussion on this
morpheme.

73
(c)

rau siab ntso
ua tiag tiag.
commit.oneself.to do really
‘...willing to commit yourself to doing it.’

In (70), yeej occurs with tej zaum ‘maybe’, creating the idea of ‘maybe it will...’, perhaps
as a form of concession about the likelihood of a potential result 98, that is, while the
speaker expects that the outcome is guaranteed, it is not the action or goal for which he is
arguing. While this example seems at first glance to produce a contradiction of certainty,
where ‘maybe’ contradicts the ‘definitely’ (as glossed above) of yeej, the notion of
certain outcome of a process still holds with yeej here.
In terms of Non-spatial Setting, the first meaning, ‘already’, falls into the Lexical
Time Words of Dixon (2012:20), while the second meaning, that marking a high level of
certainty, belongs naturally in the category of Degree of Certainty. As a part of the larger
system of Degree of Certainty, yeej ‘certainly’ can co-occur at least with tej zaum
‘maybe’, as in (70) above, as well as with xwb ‘indeed’.99 This suggests that yeej does not
form a paradigmatic opposition to other markers of Degree of Certainty, and the exact
relationships between yeej and the others is unknown, and left to further research.
The word yeej as a Semi-modal marks the Modality of Ability when it appears
with the negative, in a manner very similar to tau ‘can’ above (following Heimbach
1969:424). An example appears below in (71).
(71) Kuv
yuav
tsis yeej.
1SG
obtain NEG can
‘I cannot obtain it (because of some hindrance).’

(Heimbach 1969:424)

In (71), yeej follows the verb yuav ‘obtain’, producing the meaning ‘cannot obtain’,
showing the Modality of Ability in a negative sense.
98

That is, based on the context from where this example sentence comes. Please see Appendix C for the
full text.
99
Please see section 3.9 below for examples of the co-occurrence with xwb.

74
However, in the positive, yeej often simply means ‘win’ or, as Heimbach
(1979:424) defines it, ‘to win in an encounter’, as in (72) below.
(72)

ntaus yeej
fight
win
‘To win a fight.’

(Heimbach 1979:185)

Here, the verb ntaus ‘hit’ is followed by yeej, resulting in the meaning ‘win a fight’. Thus,
it would appear that yeej can serve as a full verb meaning ‘win’.
However, there are three reasons why yeej should be treated as a grammaticalized
element when used as in example (71) above: 1) it follows another verb when it signals
Ability, and 2) the Ability meaning surfaces clearly only when the sentence is negated,
and 3) the notion of Ability expressed when yeej is used can refer to an unrealized
situation.
First, yeej always follows another verb in the data when it indicates Ability. 100
While this pattern is seen in example (71) above, Heimbach (1969) provides three other
examples where yeej indicates ability, in all of which yeej follows another verb. Thus, in
the available data, yeej always follows another verb when it indicates Ability.
Second, the Ability meaning appears with yeej only when the sentence is negated.
As shown in (72) above, where yeej follows ntaus ‘hit’ to produce ‘win in a fist fight’,
the meaning ‘win’ is prevalent when the sentence is not negated; this is also true of the
other positive example Heimbach (1969) provides. However, all four of the negative
examples Heimbach provides indicate a lack of ability. Thus, there is a correlation
between negation and the Ability meaning with yeej.

100

Note, however, that while all attested instances of yeej signaling ability are when it appears after a verb,
not all instances of post-verbal yeej signal Ability, as in example (72) above.

75
Third, the idea of Ability with yeej can refer to a hypothetical situation. This is
shown in (71) above, where the presence of some ‘hindrance’ prevents the action
represented by yuav ‘obtain’ from being realized—in fact, there is no indication that it is
even attempted. This contrasts with the usage of yeej as a normal verb, in that the
‘Accomplishment Serial Verb Construction’ (following Jarkey 1991), which would be
made up of the normal verb yeej preceded by another verb, would be expected to refer to
an attempted action that did not succeed. Since this is not what is found with tsis yeej,
which indicates general inability rather than a lack of success, it appears that its usage
with the negative tsis produces a somewhat different nuance, in line with what is
observed with tau when it marks Ability. Altogether, it is seen that while yeej when it
indicates Ability appears to be identical to yeej with the full verbal meaning of ‘win, have
success’, the two usages differ slightly on syntactic and semantic grounds. As a
grammatical marker, this places yeej ‘can(not)’ in the system of Modality as being
associated with negative Ability. In terms of a paradigm, however, there is no clear
evidence that yeej can co-occur with other Modality markers, but since co-occurrences
are possible elsewhere in the system, this cannot be ruled out, and this possibility is left to
further research.
3.2.9 tab tom ‘begin to, IMPERFECT’
The word tab tom ‘begin to101, IMPERFECT’ has two significant functions relevant to NonSpatial Setting: it can 1) express beginning of an action, and 2) it can express a current or
ongoing action. The first use of tab tom belongs to Phase of Activity as Beginning action,
while the second use marks Imperfect as part of Completion.
Examples of the two uses appear in examples (73) and (74) below.
101

Gloss based on Heimbach (1979:303).

76
(73) (a)

(b)

Kuv tab tom
hais lus
nws txawm
1SG begin.to say
word 3SG thereupon102
‘I had no sooner begun to speak than he...’
khiav lawm
run
CP
‘...ran off.’

(Heimbach 1969:303)

In (73), tab tom expresses that the action hais ‘speak’ had recently begun.
(74) ...lub

sij hawm uas nyab laj tab tom
nrhiav cov
CLF
time
that Vietnam IPRF
search PL.CLF
‘...the time that the Vietnamese were searching for those...’

neeg...
people

In (74), tab tom is used with nrhiav ‘search’ in a way equivalent to the English
progressive ‘were searching’.
An example where both uses of tab tom are possible appears in (75) below.
(75) Peb tab tom
nrhiav peb lub
hom phiaj
1PL begin.to/IPRF search 1PL CLF purpose
‘We begin to search for our purpose.’
or ‘We currently are looking for our purpose.’
In this case, tab tom is ambiguous—it can either signal the beginning of the action of
nrhiav ‘search’, producing ‘begin to search’, or it can signal that nrhiav refers to a current,
ongoing affair, resulting in the meaning ‘currently are looking’. Thus, both meanings can
be possible in at least some circumstances.
In addition to activities, tab tom can be used with current states, as in (76) below.
(76) Tab tom

muaj ntau phau ntawv...
have
many book
‘Currently, there are many books...’
IPRF

Here, tab tom is expressed in English as ‘currently’,103 and combines with muaj to
express a current state.

102
103

Gloss from Heimbach (1979:375).
My language consultant, Jay Yang, only sometimes used this gloss to bring out the nuance of tab tom.

77
Tab tom can also serve a role that extends from its use as an Imperfect marker to
indicate a nuance of Imperfective with a state (following Jarkey 1991:75), in Dixon
(2012)’s system. An example of this appears in (77) below.
(77)

(a)

Nws tab tom mob
thaum nws
3SG
IPFV
be.sick time
3SG
‘He was sick when his relatives came...’

(b)

...xyuas nws
visit
3SG
‘...to visit him.’

kwv tij
relative

tuaj...
come

(Jarkey 1991:75)

In this case, tab tom ‘IMPV’ appears with mob ‘be sick’ to indicate the ongoing state at the
time when the other event, nws kwv tij tuaj xyuas nws ‘his relatives came to visit him’,
occurred, showing that at least with the state mob ‘be sick’, tab tom can be used with an
extended meaning to indicate an ongoing state as a backdrop to a second event. For
purposes of Non-spatial Setting, however, this usage, when compared with the rest of the
data, forms a sort of ‘secondary’ usage for this form, where the primary use is the
Imperfect sense of an ongoing event or action, and the secondary use is the mere
juxtaposition of a single event with the Imperfect form to create an Imperfective sense.
Thus, this Imperfective form should be considered a subtype of the Imperfect use of tab
tom.
As it has been shown, tab tom can serve one of two uses, of which the first is
marking the beginning of an action and the second signaling that an action is currently in
progress. As such, as a part of Non-spatial Setting, the first belongs to Phase of Activity
as a marker of Beginning, and the second belongs to Completion as an Imperfect marker.
As part of the larger system of Phase of Activity, there is no known evidence in the data
showing that tab tom could co-occur with other markers of Phase of Activity, suggesting
a paradigmatic relationship with other members such as pheej ‘continue to’ or tiav

78
‘finish’. Likewise, as part of Completion, there are no known examples from the data
where tab tom can co-occur with other Completion markers such as lawm ‘COMPLETIVE
PARTICLE’,

though this possibility should not be ruled out.

3.2.10 maj mam ‘slowly’
The word maj mam is an adverb that conveys one of several meanings, including
‘progressively’, ‘slowly’ (Mottin 1978:119), and ‘gradually’ (Heimbach 1969:124).
Examples appear in (78) and (79) below.
(78)

Maj mam
ua
progressively do
‘Do progressively.’

(Mottin 1978:119)

Here, maj mam conveys the idea of ‘progressively’, qualifying the verb ua ‘do’.
(79) Maj mam mus.
slowly
go
‘Go slowly.’

(Heimbach 1969:124)

In this case, maj mam contributes the idea of ‘slowly’ to the verb mus ‘go’.
Maj mam also can serve as a Temporal Shifter (Dixon 2012:5-6) meaning ‘after a
while’ (Heimbach 1969:124). An example appears in (80) below.
(80) Maj mam104 kuv
yuav
mus
after.a.while 1SG POS.IRR go
‘After a while I’ll go too.’

thiab.
too

(Heimbach 1969:124)

In (80), maj mam expresses the notion ‘after a while’. Altogether, maj mam can serve two
purposes: expressing ‘progressively, gradually, slowly’, and operating as a Temporal
shifter with the meaning ‘after a while’.
Syntactically, though the available evidence from scholarly sources is scant, maj
mam seems to appear in different locations when it provides a different function: it

104

It should be noted that maj mam here could be appearing in either a topic slot or the pre-subject slot. As
other Temporal shifters often appear in the topic slot, this is assumed here as well.

79
appears between the subject noun and the verb when it expresses ‘progressively,
gradually’ and before the subject noun when it appears as ‘after a while’. The positioning
with the meaning ‘progressively, gradually’ is demonstrated by (81) below.
(81) Nws
maj mam tuaj
ze
me ntsis.
3SG
gradually come close105 a.little.bit 106
‘He gradually came closer.’

(Heimbach 1969:124)

Here, maj mam appears after the subject pronoun nws ‘3SG’ and before the verb tuaj
‘come’, and expresses the notion of ‘gradually’. The positioning of maj mam at the
beginning of the sentence with the meaning ‘after a while’ appears in (80) above, where
maj mam precedes the subject noun kuv ‘1SG’. Thus, the use of maj mam with the
meaning ‘progressively, gradually’ is associated with the position in between the subject
noun and the verb, similarly to morphemes such as tab tom ‘imperfect’ and tseem
‘still’,107 while the use as ‘after a while’ is associated with sentence-initial position, much
like many other adverbs.
In terms of Non-spatial Setting, maj mam, when it indicates ‘progressively,
gradually’, appears to belong to the class of adverbs represented by tab tom ‘IMPERFECT’
and tseem ‘still’. At the same time, it is not a Lexical Time Word, nor a clear indicator of
the Imperfect, since some of the actions, such as in examples (78) and (81) above, are not
good examples of ongoing action. However, in all of the examples, there is an element of
slow or gradual continuing action toward a goal, which is corroborated by Mottin
(1978:119)’s treatment of this morpheme as indicating “progression,” along with another

105

Gloss following Heimbach (1969:435).
Gloss following Heimbach (1969:125).
107
Please see the discussion below concerning these two words.
106

80
adverb indicating slowness. 108 As such, this first use of maj mam properly belongs to the
category of Speed and Ease as a marker of slow action. The second use of maj mam with
the meaning ‘after a while’ is clearly a Lexical Time Word, fitting specifically into the
category of Temporal shifters provided by Dixon (2012:20).
3.3 Modal position
The modal position is the location where the positive irrealis marker and true modals in
Dixon (2012)’s Modality framework are found. The three markers in this position are
yuav ‘POSITIVE IRREALIS, intend’, yuav tau ‘must, have to’, and yuav tsum ‘must, have to’.
3.3.1 yuav ‘POSITIVE IRREALIS, intend’
As a grammaticalized morpheme, the word yuav ‘will, intend’109 is a positive Irrealis
marker with an extended function as a Modal signifying Intent, with a strong preference
for referring to future action. An example appears in (82) below.
(82)

...wb
yuav
sib
tham
me ntsis txog kev ua noj
2DU110 POS.IRR RECP talk111 a.little.bit about cooking
‘...we will talk a little bit about cooking.’

Here, yuav precedes the phrase sib tham me ntsis ‘talk a little bit’, indicating a future
action and translated by the English ‘will’.
Though at first glance the word yuav seems to simply mark future time, in line
with the position taken by a number of authors, including Jaisser (1984) and Fuller
(1988), the word yuav actually expresses Modality as well as broader Irrealis. This is

108

The other adverb in question is qeeb qeeb ‘slowly’ (Mottin 1978:119), which does not appear to be the
product of grammaticalization, but rather a grammatically productive form in the language based on qeeb
‘slow, slowly’ (Heimbach 1979:262).
109
Gloss based on Jarkey (2006:130).
110
Gloss following Mottin (1978:44).
111
The glossing and treatment here of the phrase sib tham relies on Jarkey (1991:351) and Jarkey
(2006:129).

81
known from two pieces of evidence: the fact that yuav can refer to situations that do not
necessarily refer to future time, and that yuav can refer to Irrealis situations in the past.112
First, yuav can be used to refer to situations that do not require a future time
interpretation. This can be found in (83) below.
(83) Yus

muaj ib
tug
tub, yuav
muab nyab
have one CLF boy intend take
son’s.wife
‘(Suppose) one has a son, (who) intends to take a wife.’

(Jarkey 1991:70)

INDF.PRON

Here, the idea expressed is one of present intent—marked by the word ‘intends’—with no
indication that the sentence refers strictly to future time.
Second, yuav can be used to refer to Irrealis situations in the past, as in (84) below.
(84)

Kuv twb
yuav
ua kws kho mob
1SG almost POS.IRR do doctor
‘I almost became a doctor five years ago.’

tsib xyoo
five year

tas
finish

los
come

lawm
CP

In (84), yuav is used to express an unrealized, but possibly intended event, a context
where a notion of irrealis is prevalent.
Though the above evidence shows that yuav is not a future tense marker
(following Clark 1982, Jarkey 1991, etc.), many authors have treated it as such (including
Jaisser 1984 and Fuller 1988, for example). However, Li (1991) suggests this kind of
misanalysis is due to data-gathering methodologies that focus on translating English
phrases, line by line, that have resulted in yuav being used as an approximation for the
English word will in reference to the future.113
Since the word yuav is not a tense marker, it must be something else. Dixon (2012)
provides two possibilities for future time: Tense marking and Modality. Yuav is a Modal,
since it conveys the “Secondary” concept of Intent (following Jarkey 1991:71, Jarkey
112

The argument presented here generally follows Jarkey (1991:69-71).
It is notable here, however, that while Li (1991) argues against this kind of methodology, he still
concludes that yuav is a future tense marker, probably due to a relative lack of evidence to the contrary.
113

82
2006:130), fulfilling Dixon’s expectation that Modality be expressed through ‘Secondary’
verbs (2012:25-26).
An example of this “Intent” status appears in (83) above, repeated as (85) below.
(85) Yus

muaj ib
tug
tub, yuav
muab
have one CLF boy intend take
‘(Suppose) one has a son, (who) intends to take a wife’
INDF.PRON

nyab
son’s.wife

(Jarkey 1991:70)

Here, the notion of Intent is clearly expressed in the translation of yuav as ‘intends’.
Another example of yuav expressing Intent is found in (86).
(86) Kuv
yuav
ua kws kho mob
tiam sis tsis
1SG
POS.IRR do
doctor
but
NEG
‘I would become a doctor, but I don’t have money.’

muaj
have

nyiaj
money

In (86), yuav refers to a stated intent, though current circumstances (‘but I don’t have
money’) may prevent this intent from being realized. The consultant, Jay Yang,
suggested that (86) expresses a situation where if there is no money, then the idea
expressed by ‘become a doctor’ will not be possible in the future. Thus, yuav expresses a
speaker’s intent for the future.
Dixon provides two morphosyntactic categories for Modality: Modals and Semimodals, where the former is required to be VP-initial and is limited morphologically,
while the latter can behave as a typical verb (2012:26). The word yuav is a Modal, rather
than a Semi-modal, since: 1) it must occur in initial position in a verbal phrase; 2) it must
appear in an unusual position—before the negative marker tsis ‘not’; and 3) it does not
function as a regular verb.114 First, when it affects the main verb of a clause, it can appear
only in initial position of a verb phrase 115, as shown by (87), (88), and (89) below.

114

In line with the expectations for a Modal in Dixon (2012:26).
Counterexamples are possible, such as Koj ua yuav tas ‘You’re almost done’; cases such as this are
likely the result of yuav operating directly on tas ‘finish,’ and yuav tas is perhaps subordinated in some sort
of complement construction to ua.
115

83
(87) Peb yuav
tau
ua
1PL POS.IRR ATT do
‘We will become a doctor.’
(88) *Peb
1PL

tau

(89) *Peb
1PL

tau

ATT

ATT

kws kho mob
doctor

yuav
ua
POS.IRR do
ua
do

kws kho mob
doctor

yuav
kws kho mob
POS.IRR doctor

Here, (87) is grammatical when yuav occurs at the beginning of the verbal phrase.
However, in (88) and (89), the sentence is ungrammatical when yuav appears in a
position other than at the beginning of the verbal phrase. Thus, yuav must occur in first
position, consistent with Dixon’s modal category.
Second, yuav must appear in an unusual position compared to regular verbs: it
must always come before the negative marker tsis ‘not’. This is shown in (90) and (91)
below.
(90)

...peb
yuav
tsis
paub
hlub...
1PL
POS.IRR NEG
know love
‘...we will not know how to love...’

Here, yuav precedes the negative marker tsis, resulting in a grammatical sentence.
(91)

*Kuv
tsis
yuav
ua
1SG
NEG
POS.IRR do
‘I will not become a doctor.’

kws kho mob.
doctor

In this case, the sentence is ungrammatical due to the fact that tsis precedes yuav; the
language consultant, Jay Yang, suggested the ordering yuav tsis to fix the sentence. This
shows that yuav must precede tsis, unlike regular verbs, which allow tsis to precede them.
Third, yuav does not function as a regular verb. When the seven tests for
verbhood proposed by Jaisser (1984) are applied to yuav (expanding on the discussion on
the placement of tsis above), this word fails on several points, as shown in Table 11.

84
Test Type

yuav
negative tsis
no
question particle puas
yes
“past tense” tau
no
“future tense” yuav
no (illogical)
negative imperative txhob
questionable
post-verbal ability tau
yes*
question construction with los tsis V
yes*
Table 11. Test results for verbhood of yuav.

typical verb
yes
yes
yes
yes
yes
yes
yes

As seen in Table 11, tests 1, 3, and 4, i.e., preverbal negation, tau perfect (or “past tense”
in Jaisser 1984), and yuav Positive Irrealis/Modality (or “future tense” in Jaisser 1984) all
fail, while for 5, the language consultant, Jay Yang, suggests that it could be used, though
he never would. Test 6 works, though whether the ability marker tau is being applied to
yuav or what follows is not completely certain, and test 7 is allowable though not
preferred. As a result, since only test 2 passes without further issue, it is clear that yuav is
not a normal verb since it is subject to a number of restrictions. Thus, since yuav must
occur first in a series of verbs and it does not behave as a normal verb, it is a Modal in
Dixon (2012)’s sense of the term.
Yuav can also be used in an irrealis sense, where “the speaker seems to be
asserting his/her belief in the likelihood of the truth of the proposition” (Jarkey 1991:70),
in an unrealized context, as in (92) below.
(92)

(a)

(b)

Ib
yam
li
qhov
peb hais
one kind as thing 1PL say
‘For example, if we say “he picked...”’

tias
COMP

ib
lub taub”
yuav
tsis
meej
one CLF pumpkin POS.IRR NEG clear
‘...a pumpkin” it would not be as clear as…’

“nws
3SG

npaum
equal

de
pick

li
as

tau
get

qhov
thing

85
(c)

“nws de
tau ib
lub
taub
3SG
pick get one CLF punpkin
‘…“he picked a yellow pumpkin.”’

dag.”
yellow

(Jarkey 1991:70)

Here, yuav is placed in front of the phrase tsis meej npaum li ‘not be as clear as’ and
supplies the meaning ‘would’, in a sentence signaling a hypothetical situation. Thus, it
takes on a general irrealis notion, marking the fact that the sentence does not refer to a
realized situation.
In sum, yuav is a positive Irrealis marker that has an associated function as a
Modal expressing Intent. As part of the fuller system of Modality in White Hmong, yuav
can co-occur freely with other forms such as tau ‘can’ in its positive-irrealis-as-future
sense, 116 which allows it to appear with such markers of Modality.
3.3.2 yuav tau ‘must, have to’
The Modality marker yuav ‘POSITIVE IRREALIS, intend’ and the Perfect marker tau
‘ATT’117 can combine into a single unit to signal Obligation, for at least some speakers
(Clark 1982:130).118 Examples of this usage appear in (93) and (94) below.
(93)

Koj yuav tau
ua.
2SG must
do
‘You must do (it).’

(Heimbach 1969:431)

In (93), yuav tau appears after the subject koj ‘2SG’ and before the verb ua ‘do’ to form
the phrase ‘you must do’, signaling obligation for the hearer.

116

Please see section 3.33 below for further details.
It should be noted here that Clark (1982:130) treats this morpheme in her work as an auxiliary version of
the verb tau ‘get,’ rather than referring to this grammaticalized form as an Attainment marker.
118
Clark (1982:130) states it thus: “Some dialects or individuals allow yuav-tau as a compound auxiliary
verb to denote obligation.”
117

86
(94)

Kuv yuav tau
kawm lus
no.
1SG have.to
study word this
‘I have to study these words.’

(Clark 1982:130)

In (94), yuav tau follows the subject kuv ‘1SG’ and precedes the verb kawm ‘study’ to
express the meaning ‘I have to study’, here marking obligation for the speaker. Thus, as a
unit, yuav tau signals Obligation.
However, the series yuav tau can also function as the sum of its parts, namely that
yuav signals its normal Modal function of ‘will, intend’ and tau as either the Semi-modal
meaning ‘get to’, or the regular verb ‘get’. Examples for each of these appear in (95) and
(96) below, respectively.
(95)

...peb
yeej
yuav
tau
los
ua xwb.
1PL
certainly POS.IRR can/ATT come do indeed
‘...we will become one.’ or ‘...we will be able to become one.’

In (95), yuav tau appears with the verb series los ua ‘become’ to form the phrase meaning
‘will become’ or ‘able to become’. In the former interpretation, tau is serving in its role
as a Perfect marker, while in the latter case, tau marks the Modality notion of Ability.
Yuav refers to a future event.
(96)

Nyab laj
luag
yuav
tau txhom txhua
Vietnamese others POS.IRR get capture all
‘The Vietnamese will capture all the people...’

leej...
people

(Fuller 1988:20)

In (96), yuav and tau appear together before the verb txhom ‘capture’. Here, yuav and tau
are distinct elements: yuav is a Modal indicating the future here, while tau is likely the
general verb ‘get’ that is associated with the verb txhom, where tau txhom indicates an
idea of ‘getting’ and ‘capturing’. Thus, yuav and tau can also appear together while
maintaining their own distinct meanings. As a whole, yuav tau can either operate as a
single unit signaling Obligation or appear together with their distinct meanings and
functions intact. As a Modality morpheme indicating Obligation, yuav tau generally does

87
not co-occur with other Modality markers nor with either the general irrealis or the more
specific Modality nuance of yuav ‘POSITIVE IRREALIS, intend’. This suggests a
paradigmatic distinction on the part of morphemes indicating Modality, and a prohibition
against like forms co-occurring in the case of yuav, roughly similar to the restriction
against yuav ‘will’ co-occurring with yuav ‘buy’ in Laotian Green Hmong (Li 1991:52).
3.3.3 yuav tsum ‘must, have to’
The morpheme yuav tsum is a Modal of Necessity and Obligation meaning ‘must, have
to’. An example appears below in (97).
(97) (a)

(b)

..peb yuav tsum los
pib
nhriav
1PL
must
come start
search
‘...we must start searching for ourselves...’
...rau
ntawm
to
there119
‘...in God.’

peb
1PL

tus
CLF

kheej
self

nyob...
in

Vaj Tswv.
God

Here, yuav tsum ‘must’ appears with the phrase beginning with los pib nrhiav... ‘come
start looking...’ to indicate Necessity.
Yuav tsum is a Modal on the basis of its position before the negative marker tsis,
and as such, before other verbs—a required trait of Modals (Dixon 2012:26). An example
of the placement of yuav tsum appears in (98).
(98) ...yam uas peb yuav tsum
type120 REL 1PL must
‘...the one that we must not do...’

tsis
NEG

txhob
ua...
IRR.NEG do

Here, yuav tsum ‘must’ precedes the phrase tsis txhob ua ‘not do’, showing that yuav
tsum must appear before the negative marker and other verbs, and is thus a Modal.

119

Gloss based on “there-nearby” found in Jarkey (1991:42). Note that this word is technically a “spatial
deictic,” a variety of word in White Hmong that plays the role of a demonstrative while sometimes serving
a function of spatial indication similar to some prepositions (Jarkey 1991:42-43).
120
Gloss from Heimbach (1979:419).

88
As part of the greater system of Modality marking, yuav tsum generally does not
co-occur with other markers of Modality in the data nor with the irrealis marker yuav
‘POSITIVE IRREALIS’, likely for the same reasons as with yuav tau ‘must, have to’ in
section 3.3.2 above. This suggests a paradigmatic relationship with the other Modality
categories. In comparison with other markers of Obligation and Necessity such as yuav
tau ‘must, have to’ or tsum ‘can, have to’, there is no evidence that yuav tsum belongs to
a distinct category, suggesting that it is another option with more or less the same
function as the other markers of its category.
3.4 Pre-main-verb position
The “Pre-main-verb position” is the location where many Secondary-A verbs in Dixon
(2006)’s verbal classification scheme are found. These include the Irrealis and Modality
markers tau ‘can, be able to’, tsum ‘can, have to’, txawj ‘know how to’, and txhob
‘should not, NEGATIVE IRREALIS’, as well as the Phase of Activity markers chiv ‘begin to’,
pib ‘begin to’, nyim ‘continue to’, pheej ‘keep, continue to’, and rau siab ntso ‘continue
to, commit oneself to’.
3.4.1 tau ‘ATTAINMENT, can’
The word tau ‘ATTAINMENT, can’ as a grammatical morpheme indicates one of two things:
1) Attainment as a type of Perfect, belonging to the category of Completion, and 2)
Ability as an indicator of Modality. 121
However, tau has been typically considered in the scholarly literature to serve
four different functions: 1) before a verb, as a marker of the perfect (following Jarkey
1991:76, 2006:116-117122); 2) before a main verb, as a modal expressing ‘get to, manage

121
122

It should also be noted here that tau is also a full verb meaning ‘get’ (Clark 1982:133).
Specifically as ‘perfective aspect,’ which corresponds more or less to the ‘perfect’ of Dixon (2012).

89
to’ (Jarkey 1991:74); 3) after a verb, as a marker of ‘Attainment’ in a serial verb
construction (Jarkey 1991:73); and 4) after a verb, expressing ability or “potential (mood)”
(Heimbach 1979:307; Jarkey 1991:72).123 Each of these roles considered in the literature
will be presented in turn, followed by a more cohesive analysis of tau.
First, tau appears pre-verbally to indicate the perfect. This assessment is based on
Jarkey (1991:76), who calls this “perfective aspect.” However, Jarkey’s defines this
variety of ‘perfective aspect’ by stating that “tau indicates that the event or state
described by the verb is completed or realised, or will be completed or realised, at the
time referred to” (1991:76). Thus, this “perfective” is the Perfect of Dixon, which is
defined as expressing “an action which is completed before the present time” (2012:31).
Thus, tau is treated as a Perfect marker here. Examples of the use of tau in this pre-verbal
role are given below in (99), (100), and (101).
(99)

Ces
nws kuj
then 3SG consequently124
‘Then he said to my dad...’

tau
ATT

hais
say

rau
to

kuv
1SG

txiv...
father

In (99), tau is applied to hais ‘say, tell’, producing the translation ‘told’.
(100)

Vaj Tswv tau qhia tseeb tseeb
txog
tsib yam
125
God
ATT teach the.truth
about five type
‘God has taught us the truth about five types of purpose...’

hom phiaj...
purpose

In (100), tau appears with qhia ‘teach’, resulting in the translation ‘has taught’.
(101)

123

(a)

Vaj Tswv yeej
tsis
tau tso
peb
God
certainly NEG ATT let.go 1PL
‘God does not/won’t let us stay in the dark...’

(b)

paub
know

dab tsi
what

nyob
live

tsaus ntuj nti
darkness

hlo li...
at.all

It is notable here that Jarkey (1991:228-229) considers that both uses of post-verbal tau stem from the
same notion of achievement.
124
Gloss from Heimbach (1969:88).
125
Gloss corroborated by Heimbach (1969:350), who gives “spoken truth.”

90
‘...knowing nothing at all.’
In (101), tau appears in a negative clause with tso ‘let go’, leading to the translation of
tsis tau tso as ‘does not let us’ or ‘won’t let us’. This variability in translation is found in
multiple sources, suggesting that either tau can mean multiple things, or it has no direct
equivalent in English.
Jarkey (1991)’s definition above can shed light on the situation: the completion of
the event is in focus in regard to some time. In addition, unlike with other perfect forms
in White Hmong, such as lawm, the optional ‘present relevance’126 portion of the notion
of the perfect may be absent. As a result, this construction with tau can freely be placed
in any time period as seen in the above examples, as long as the action is completed
before a certain, even implied, time. This is manifested in the past tense translation in
(99), with a probable reference time in the present, the present perfect in (100), again
with a probable reference time in the present, and the negated present/future form in
(101), with a probable reference time either in the present or at some point in the future.
Thus, this usage of tau can be treated as a Perfect marker of Completion, in connection
with some other moment in time.
A large number of authors, most notably Mottin (1978), Heimbach (1969), Jaisser
(1984) and Fuller (1988)127, treat this usage of tau as a past tense marker. However, as
shown in example (101) above, tau can be used in reference to events in non-past time.
Thus, while tau in this role more often than not appears with events that took place in the
past, it does not actually mark the past tense, but rather the Perfect.

126
127

Note that for Comrie (1976:56), this “continuing relevance” idea is central to the notion of the perfect.
See Li (1991:26) for a more complete list of authors taking this position.

91
Second, tau can appear before a main verb as a type of modality expressing ‘get
to, manage to’ (Jarkey 1991:74). An example of this is (102) below.
(102)

Yog li ntawd sawv daws kuj
so
there
everyone
also
‘And also we got to go back home.’

tau
get.to

rov qab
mus
going.back go

tsev
home

Here, tau appears with the phrase rov qab mus tsev ‘go back home’ to express ‘got to go
back home, where tau means ‘got to’.
Third, when it appears after a verb, tau serves as a marker of ‘Attainment’ in a
serial verb construction (Jarkey 1991:73). Jarkey (1991:229) describes this usage as
follows: “a verb such as tau ‘get’ in the V2 slot in this construction indicates that some
goal is achieved, but is extremely imprecise regarding exactly what kind of goal that is.
The only possible interpretation is that this “goal” is the performance of the activity itself.”
An example of this usage of tau as accomplishing a goal is shown in (103) below.
(103)

...peb thiaj li
nrhiav tau
1PL
consequently search ATT
‘...in order for us to find our purpose.’

peb
1PL

lub
CLF

hom phiaj.
purpose

Here, tau follows the verb nrhiav ‘search’ to produce the meaning ‘find’. In this case, the
combined use of nhriav tau marks the successful completion of the action associated with
nrhiav (following Jarkey 1991, 2004).128
Fourth, tau can appear after a verb to express ability. This is shown in (104) and
(105).

128

It is notable here that it is exactly this verb, nrhiav ‘search,’ that Jarkey (1991, 2004) uses to support her
points for this use of tau. As a result, the point made with this example is basically the same as what is
found in her works.

92
(104) Tus

nai hais tias koj noj tau
CLF
boss say
that 2SG eat
can
‘The boss said that you can eat dinner now.’

hmo
tam sim no.
dinner now

In (104), tau appears after the verb noj ‘eat’ to express ‘can eat’, in the context of a boss
giving permission.129
(105) Tej
lus
pab no tej zaum nws yeej
pab
these word help this maybe
it
certainly help
‘All these teachings, maybe it will be able to help you...’

tau
can

yus...
INDF.PRON

In (105), tau follows the verb pab ‘help’ to produce ‘be able to help’, where the phrase
expresses the speaker’s expectation about the ability of the ‘teachings’ to be of benefit.
Thus, tau can serve to mark ability in post-verbal position.
Jarkey (1991) takes the third and fourth usage above and appears to combine them,
where the ‘attainment’ definition above is the basic meaning for post-verbal tau, with
“permission” and “ability” as “follow[ing]” from it (Jarkey 1991:73). In this case,
achievement is in focus, with ability implied (Jarkey 1991:228). At the same time,
alternations showing both a distinction and an overlap are found in the data, shown in
(106), (107), and (108) below.
(106) Nws hais tias tag kis
peb mam li130 nrhiav
3SG say that tomorrow 1PL will
search
‘He said that tomorrow we can search for our purpose.’

tau
can

peb
1PL

lub
CLF

hom phiaj
purpose

In (106), the combination nrhiav tau appears, with the resulting meaning ‘can search for’.
(107) ...peb thiaj li
nrhiav tau
1PL
consequently search ATT
‘...in order for us to find our purpose.’

peb
1PL

lub
CLF

hom phiaj.
purpose

Here, the same phrase appears with the meaning ‘find’.

129

Jarkey (1991:73) notes the possible use of pre-verbal tau as indicating permission.
See section 3.2.4 above for the meaning and use of mam li. It does not appear to affect the meaning here,
as other examples, such as Tag kis peb yeej nrhiav tau peb lub hom phiaj ‘Tomorrow we will be able to
search for our purpose,’ without mam li, were found.
130

93
(108)

...peb thiaj li
yuav
1PL
consequently POS.IRR
‘...we can find ourselves...’

nrhiav
search

tau
can/ATT

peb
1PL

tus
CLF

kheej...
self

In (108), nrhiav tau appears again, but with the meaning ‘can find’. This can be treated as
ability with ‘search for’ in (106), Jarkey (1991)’s ‘attainment’ with ‘search for’ in (107),
and ability and ‘attainment’ with ‘search for’ in (108). As a result, tau post-verbally can
express ability and attainment, either one or the other or both. This suggests a
complicated relationship, one that fits somewhat within Jarkey (1991)’s explanation
described above, though at the same time, one that presents an enigma when it comes to
discerning when ability is actually implied. 131
This possibility of alternation can even be found in the same environment, as
shown in (109) below.
(109) Nws hais tias peb nrhiav tau
peb
3SG say that 1PL search can/ATT 1PL
‘He said that we can search for our purpose.’
or ‘He said that we have found our purpose.’

lub
CLF

hom phiaj
purpose

lawm
CP

In this case, both ability and attainment interpretations are possible with nrhiav tau,
where the phrase can mean ‘can search for’ or ‘have found’, apparently depending on
whether the search took place in past time or if it is yet to be realized. As a result, an
interpretation where two distinct forms are being considered is preferable for post-verbal
tau, though with the concession that both meanings could be intended at the same time or
even that the attainment meaning implies ability in at least some cases.
Altogether, past scholarly works have typically suggested four different uses for
tau. However, one significant aspect of tau not previously considered in the literature is a
131

It should be noted here that Li (1991:32-34) presents several instances for Laotian Green Hmong where
tau in preverbal position can only mean ‘get to’ and other examples where it can mean both ‘get to’ and
“Attainment aspect,” with the defining difference being the presence of additional content in the sentence
that “serve[s] to bound the event.”

94
profound one: the position of tau in the sentence is not actually related to its basic
meaning. This can be concluded from two factors: 1) pre-verbal and post-verbal tau
function as signaling ‘Attainment’, as described in the literature by different sources, and
2) pre-verbal and post-verbal tau can both signal Ability. Each of these factors is
discussed below.
First, tau can function as signaling ‘Attainment’ either before or after a verb. This
is demonstrated in the scholarly literature, where Li (1991) treats pre-verbal tau as
“Attainment,”132 while Jarkey (1991) treats post-verbal tau as the same. Since both
appear to make valid points in regard to the use of tau, it seems that a unified treatment is
merited. Consider (110) and (111) below.
(110)

Kuv tau
kawm los
ua kws kho mob.
1SG
ATT
learn
come do doctor
‘I have/had learned to become a doctor.’

Here, tau precedes the verb kawm ‘study, learn’ to signal the completion of the action and
the attainment of the goal of becoming a doctor.
(111)

Nag hmo
peb nrhiav tau
yesterday 1PL search ATT
‘We found our purpose yesterday.’

peb
1SG

lub
CLF

hom phiaj
purpose

lawm.
CP

In this case, the verb nrhiav ‘search, look for’ is followed by tau, where tau provides a
notion of completion and ‘attainment’, that is, the action has achieved the desired goal
(following Jarkey 2004). Thus, tau in the ‘Attainment’ sense provides the same meaning
both before and after the main verb, while there is a perceived distinction in nuance that
may be related to either the relationship of tau to the main verb or the construction used.

132

It is notable that Li (1991) calls pre-verbal tau ‘Attainment’ in reference to Green Hmong, though the
demonstrated function of tau in his work is identical to that of White Hmong as described above.

95
Second, tau can signal Ability both before and after a verb. This is shown in (112)
and 113) below.
(112) Yog
koj
tuaj
no
peb tau
noj hmo
if
2SG come this 1PL can
eat
dinner
‘If you come, we will be able to eat dinner together.’

ua ke.
together

Here, tau precedes the verb noj ‘eat’ and contributes the idea of ‘will be able to’, showing
that tau can signal ability before the main verb.
(113) Txhua txoj kev...
puav leej
pab
133
every
CLF
road
all
help
‘All the roads...won’t be able to help us...’

tsis
NEG

tau
can

peb...
1PL

In this case, tau follows pab ‘help’ and the negative tsis to signal a lack of ability. Thus,
it is clear that tau can appear on either side of the main verb to indicate a notion of ability.
It is significant here that the language consultant, Jay Yang, pointed out that the
difference in translation into English between ‘can/be able to’ and ‘get/manage to’—
typical translations in the literature for postverbal and preverbal tau, respectively—is one
of personal preference.
In addition to the above data, Mottin clearly defines one sense of both preverbal
and postverbal tau as being one of signaling ‘possibility’ (1978:102). At the same time,
Mottin suggests that preverbal tau also signals ‘have the opportunity’ and postverbal tau
signals ‘have permission’ or ‘have the right to’. Thus, while both preverbal and
postverbal tau share a common meaning, that of indicating possibility, different nuances
are possible depending on where tau appears. In either case, it is clear that tau can signal
Ability both before and after a verb.

133

Gloss and parsing of puav leej from rest of sentence here deviates slightly from the original obtained
from Jay Yang, and is confirmed by Heimbach (1969:238).

96
Thus, it is seen that the inherent meaning of tau, whether it is used in the sense of
Ability or Attainment/Completion is not affected by whether it precedes or follows a
given verb, though a distinction in nuance can be detected. Considering that tau in postverbal position takes part in an Accomplishment Serial Verb Construction, regardless of
the meaning realized (per Jarkey 1991:73), the semantics of this type of construction may
be a major contributing factor that affects the nuance of tau in this position. In contrast to
this, constructions that involve pre-verbal tau can be taken to be a different type of
construction altogether.134 Thus, the distinction made by many scholars between preverbal and post-verbal tau is likely due to the semantics provided by the relationship
established by a certain class of Serial Verb Construction rather than a genuine
morphological distinction between the two syntactic varieties of tau. Altogether, it is
concluded here that, regardless of where it appears in relation to the main verb, tau serves
two functions: marking 1) Attainment and 2) Ability. 135 The first of these serves as a
Perfect as part of Completion, while the second is a Semi-modal marking Ability; the
Semi-modal status here is drawn from the fact that tau effectively acts like a verb in
Serial Verb Constructions, following Dixon (2012:26). As part of the larger system of
Perfect markers in White Hmong, tau can co-occur with other types of Perfect marking,
suggesting that these belong to slightly different categories but are related by both
meaning and usage, as discussed in reference to lawm ‘CP’ in section 3.8 below.
134

It is possible that this may be a distinct Serial Verb Construction of its own, or a special class of an
Accomplishment Serial Verb Construction. However, it is notable that Jarkey (1991) found four types of
Serial Verb Constructions in White Hmong, and it is questionable whether a fifth type exists specifically
for pre-verbal tau, and so treating this as a special type of Accomplishment Serial Verb Construction would
be the preferable option, where tau is the initial verb marking Attainment, and the second verb indicates the
outcome of that attainment, namely, the action itself.
135
It should be reiterated here that Jarkey (1991) considers the postverbal usage of tau to be one of
achievement, and ability as a form of that achievement; this interpretation could potentially hold
preverbally as well. Nevertheless, for purposes of Non-spatial Setting, the two nuances of tau are separated
out here for simplicity, while it is explicitly noted that the two functions of tau are indeed related.

97
3.4.2 tsum ‘can, have to’
The morpheme tsum, when used alone, is a Semi-modal that often appears with the
negative tsis. It can be used for Ability or Necessity136 (following Heimbach 1969:357,
Clark 1982:130), as in (114) and (115), respectively.
(114) Muab tsis
tsum.
get
NEG
can
‘(I) cannot get (reach) it.’

(Heimbach 1969:357)

In (114), tsum appears after the verb muab ‘get’ and the negative tsis, producing ‘cannot
get’—denoting inability.
(115) ...tsis

tsum
nrhiav lev
pua
chaw.
have.to search mat spread place
‘...don’t have to search for a mat to spread out.’

(Clark 1982:130)

NEG

In (115), tsum again appears after tsis, but before the verb nrhiav ‘search’, resulting in the
meaning ‘don’t have to search’ and marking the lack of obligation or necessity. In either
case, tsum clearly marks Modality of some sort.
Syntactically, tsum is a Semi-modal, as opposed to a Modal, because of two
factors: 1) it can be directly preceded by the negative, and 2) it is not always the first verb
in the series. First, the negative can appear before it, unlike with true Modals such as
yuav ‘POSITIVE IRREALIS, intend’. Both examples (114) and (115) above show that the
negative tsis consistently appears before it. In addition, sentences containing both yuav
and the negation marker tsis consistently display the ordering yuav tsis, as shown above
in section 3.3.1. This shows a contrast between the Modal-type word yuav and other
words such as tsum, suggesting that these belong to different classes.

136

For Necessity, Clark (1982:130) notes that this usage is attested in a variety of White Hmong spoken in
Thailand, and that in Xieng Khouang, Laos, the compound form yuav tsum is more widely attested.

98
Second, tsum is not always the first verb in the series. In (114) above, the verb
muab ‘get’ precedes the negative tsis and tsum, showing that tsum is not always the first
verb in the series. Since Dixon’s categories for Modal and Semi-modal are distinguished
in part by whether the verb always comes first (2012:26), as is the case for modals, tsum
is a Semi-modal. Thus, since tsum regularly appears in syntactic positions unsuitable for
modals, it must be a Semi-modal in Dixon (2012)’s framework.
In terms of the relationship of tsum ‘can, have to’ to other Modality morphemes,
the available data suggests that it does not co-occur with other Modality morphemes,
except when combined with yuav ‘POSITIVE IRREALIS, intend’ to make the compound
Modal yuav tsum meaning ‘must, have to’. In terms of the meaning of tsum, it appears to
overlap somewhat with other Modality items such as yuav tsum ‘must, have to’, and tau
‘can’. However, the rarity of the data for tsum limits what can be known about its
relationship to these other forms, and further research is necessary. 137
3.4.3 txawj ‘know how to’
The morpheme txawj is a Modal that denotes ability that results from past learning,
translated as ‘know how to’ (Heimbach 1969:375, Jarkey 1991:74). It fits into Dixon
(2012)’s framework as a Semi-modal marking Ability. Examples appear in (116) and
(117) below.
(116) Koj puas txawj
hais
lus
2SG Q
know.how.to speak word
‘Do you know how to speak Hmong?

Hmoob? (Jarkey 1991:74)
Hmong

In (116), txawj precedes the verb hais ‘speak’ to produce ‘know how to speak’,
expressing ability based on knowledge.

137

My language consultant, Jay Yang, never used tsum ‘can, have to’ as a morpheme distinct from yuav
tsum ‘must, have to’ in any of the data obtained.

99
(117)

(a)

Kuv yuav tsum
1SG must
‘I need to see…’

saib
look.at

seb
whatever

(b)

txawj
ua yam
twg.
know.how.to do kind which138
‘…what I know how to do.’

kuv
1SG

In (117), txawj appears with ua ‘do’, and in this case txawj was rendered as the English
‘skill, talent’, again expressing ability from knowledge. Thus, txawj marks Ability
Modality, of especially the knowledge variety.
Txawj patterns with regular verbs in some ways and with modals in others.
Txawj is similar to regular verbs in that it is placed after the negative marker tsis, like
regular verbs, as shown in (118).
(118) Kuv
tsis
txawj
kho
1SG
NEG
know.how.to heal
‘I don’t know how to heal patients.’

mob.
patient

Here, txawj follows the negative particle tsis, like regular verbs, rather than before it, as
with the Modal-type word yuav. However, txawj does not pattern like verbs in other ways.
First, it is helpful to compare txawj with the semantically similar verb paub, ‘know (how
to)’. Paub generally serves the same purpose as txawj in that it expresses ability deriving
from knowledge. An example of this verb being used this way appears below.
(119) Kuv
tsis
paub
kho
mob.
1SG
TSIS
know heal patient
‘I don’t know how to heal patients.’
In this case, paub serves the same role as txawj in (118), and the resulting translation is
also the same in both examples. This shows that, at least in this context, txawj and paub
mean effectively the same thing.

138

Gloss from Heimbach (1969:330).

100
However, the two semantically differ in several significant ways. The results of
the seven tests from Jaisser (1984) with txawj ‘know how to’ and paub ‘know (how to)’
appear in Table 12.
Test Type

txawj
paub
negative tsis
yes
yes
question particle puas
yes
yes
“past tense” tau
no
yes*
“future tense” yuav
no
yes*
negative imperative txhob
no*
yes, but strange*
post-verbal ability tau
yes, but strange
yes, but strange
question construction with los tsis V
yes
yes
Table 12. Results of grammatical tests for txawj and paub.
Tests 1, 2, 6, and 7 produced identical results—that is, 1, 2, and 7 resulted in perfectly
grammatical sentences while 6 resulted in an acceptable sentence that sounded somewhat
strange to the consultant. 3 and 4 were not acceptable with txawj, but acceptable though
somewhat strange with paub; with slight modifications, such as the addition of los lawm
with 3 and yeej with 4, the sentences were perfectly fine with paub. With 5, txawj could
not co-occur with txhob alone, whereas with paub, it could, though it sounded strange. If
the sentence is changed to the negative imperative with tsis txhob, both txawj and paub
could be used. As a result, in at least some cases, while paub can be used even if the
result sounded strange, txawj cannot be used. Thus txawj does not behave consistently as
a regular verb, and thus must be some variety of Modal.
Since txawj cannot be a regular verb, it must be something else. However, it does
not behave exactly like a true Modal-type word such as yuav ‘POSITIVE IRREALIS, intend’,
either. Jaisser (1984)’s tests for verbhood for txawj and yuav demonstrate small
differences, and these are shown together in Table 13 below.

101
Test Type

txawj
negative tsis
yes
question particle puas
yes
“past tense” tau
no
“future tense” yuav
no
negative imperative txhob
no*
post-verbal ability tau
yes, but strange
question construction with los tsis V
yes
Table 13. Tests for verbhood for txawj and yuav.

yuav
no
yes
no
no (illogical)
questionable
yes*139
yes*

In this case, test 1 has a different result, likely due to the order of the negative: yuav must
precede the negative, while txawj follows it. Also, while txawj cannot be preceded by
txhob alone, it can be preceded by tsis txhob, while yuav produces questionable results in
either case. Finally, txawj can be used comfortably in the case of test 7, while yuav is not
the preferable form. Thus, while the results are similar for txawj and yuav, they are not
exactly the same syntactically. As a result, txawj can be treated as a special variety of
Modal, distinct from yuav. In sum, txawj is a special Modal marking Ability.
As part of the system of Ability Modality, txawj marks acquired ability or skill, as
opposed to tau, which marks general ability, and taus, which marks physical or material
ability, tsum, which marks some form of physical ability (based on the scant evidence
available), and yeej ‘can(not)’, which is associated with negative ability. As such, txawj
generally does not co-occur with these other types of Ability, but rather forms a distinct
category to specifically distinguish acquired ability from other types.
3.4.4 txhob ‘NEGATIVE IRREALIS, should not’
The morpheme txhob ‘do not, should not’ marks negative irrealis (following Jarkey
2006:125), placing it as a type of irrealis marker in the Reality category of Dixon
(2012)’s framework. As such, it serves as a negation particle that often occurs in irrealis

139

See the discussion on yuav above for the special situations associated with tests 6 and 7 here.

102
contexts, such as the imperative, expressing a purpose or desire (following Jaisser
1984:60-61), negating the subordinate clause of a modal, or marking negation in the
protasis of counterfactual statements. It can also sometimes mark negative modality.
Examples of each of these five uses appears in (120), (121), (122), (123), and (124),
respectively.
(120) Txhob
txhawj
NEG.IRR worry
‘Don’t worry.’

(Jaisser 1984:59)

In (120), the negation of the imperative txawj ‘worry’ is achieved with the presence of
txhob, producing the meaning ‘don’t worry’.
(121) Nws pab kom
tus tsov txhob
tom
3SG help so.that CLF tiger NEG.IRR bite
‘He helped me so that the tiger wouldn’t bite me.’ 140

kuv
1SG

(Jaisser 1984:59)

In (121), the potential event tus tsov txhob tom kuv ‘the tiger wouldn’t bite me’ is marked
by txhob to show the actor’s goal of the event not happening.
(122) ...yam uas peb yuav tsum
type
REL 1PL
must
‘...the one that we must not do...’

tsis
NEG

txhob
ua...
NEG.IRR do

In (122), txhob, with the regular negative marker tsis, negates the verb ua ‘do’, and the
whole phrase, tsis txhob ua, is subordinate to the Modal yuav tsum ‘must’.
(123) (a)

140

Yog kuv (tsis)141 txhob
lem, peb twb
if
1SG (NEG)
NEG.IRR turn 1PL sure
‘If I hadn’t turned, we...’

Note that the translation provided here by Jaisser (1984) includes ‘me’ twice, while kuv ‘1SG’ appears
explicitly in the Hmong sentence only once.
141
Parentheses here and following in an example indicate that the word can be optionally included with no
effect on the grammaticality of the sentence.

103
(b) tsis

sib

tsoo.
NEG
RECP
smash
‘...wouldn’t have had the accident.’

(Jarkey 1991:68)

In (123), txhob marks the negation the counterfactual if-then statement’s protasis yog kuv
lem ‘if I had turned’ to express an idea of what could have been.
(124) Nws txhob
ua kws kho mob.
3SG
NEG.IRR do
doctor
‘He shouldn’t be(come) a doctor.’
In (124), txhob is placed before ua kws kho mob ‘be(come) a doctor’ to signal a negative
modality translated as ‘shouldn’t’.
One important semantic context where txhob is often used where the regular
negative marker tsis sometimes cannot appear alone is in purpose complements
introduced by kom (Jaisser 1984). An example of this is (125) below.
(125) Kuv kho lub rutsev kom
nag (tsis)
1SG fix CLF roof
so.that rain NEG
‘I fix the roof so that the water doesn’t come in.’

txhob
xau (Jaisser 1984:60)
NEG.IRR leak

In this case, the presence of txhob in the kom clause—with or without the negative
marker tsis—ensures that the sentence is grammatical. Compare this with (126) below.
(126) *Kuv
1SG

kho
fix

lub
CLF

rutsev
roof

kom
so.that

nag
rain

tsis
NEG

xau
leak

(Jaisser 1984:60)

In this case, the presence of tsis alone in the kom clause renders the sentence
ungrammatical. Thus, in at least some contexts, txhob often must appear as the negative
marker in purpose clauses, further suggesting that it marks negative irrealis.
Txhob can also be accompanied by the normal negative marker tsis without
affecting the meaning (Jarkey 1991:68). Compare example (120) above with (127):

104
(127) Tsis

txhob
txhawj
NEG
NEG.IRR worry
‘Don’t worry.’

(Jaisser 1984:59)

In (127), the same sentence appears as in (120) above, though with the addition of the
negative tsis ‘not’, without any change in meaning. This optional usage of tsis is also
manifested in (123) above, where tsis is placed in parentheses to show that it is optional,
again without any clear effect on the meaning. This same possibility exists for many of
the other examples above, such as the optional tsis in (125) above, and (124) as in (128)
below.
(128) Nws tsis txhob
ua kws kho mob.
3SG
NEG NEG.IRR do
doctor
‘He should not become a doctor.’
Here, tsis can appear before txhob, again with no change in meaning. Altogether, in many
cases, tsis can optionally accompany txhob without impacting the semantics of the
sentence, suggesting that txhob has an additional role outside of simply marking the
negative, that of irrealis.
While the above treatment as an Irrealis marker with a possible modal use
satisfactorily characterizes the use of txhob, Jarkey (1991:68), on the other hand, argues
that this particle marks a level of certainty in which “the falsity of the proposition is not
certain.” She further suggests that it could be a “subject of speculation” or a matter of
‘hope’ (Jarkey 1991:68). However, her example expressing this “speculation” is the
counterfactual if-then example in (123) above, and her example representing ‘hope’ is a
negative imperative. Both of these uses are irrealis in nature, and neither necessarily
expresses an inherent lack of certainty: in the case of (123), certainty does not appear to
be a salient aspect of the meaning of the counterfactual, and in the case of the negative

105
imperative, certainty over the event’s happening is not relevant. Thus, txhob should be
treated as a negative irrealis marker.
Syntactically, txhob shares features with regular verbs such as following the
negative marker tsis, as in example (128), and appearing in first position of a verbal
sequence, as in all of the examples above. However, when considering the seven tests for
verbal status in White Hmong proposed by Jaisser (1984), some differences emerge.
Table 14 shows the results of the seven tests with txhob.
kom ‘tell’
(Jaisser 1984)
negative tsis
yes*
yes
question particle puas
questionable
yes
“past tense” tau
yes, but strange
yes
“future tense” yuav
questionable
yes
negative imperative txhob
no
yes
post-verbal ability tau
yes*
yes
question construction with los tsis V yes, but strange
yes
Table 14. A comparison of verbal status test results for txhob and the regular verb kom.
Test Type

txhob

The tests with txhob generally showed different results compared to normal verbs such as
kom ‘tell’. In the case of test 1, in which tsis ‘not’ is placed before the element to test, tsis
txhob produces effectively the same meaning as txhob alone, since txhob is inherently
negative; in the case of test 6, while the sequence txhob tau V is possible, it seems that
tau actually patterns with the following verb, with ‘shouldn’t have V’ as the most salient
meaning, where tau does apply to the following verb, not to txhob. The translation
‘shouldn’t be able to V’, the expected outcome of test 6, is an ambiguous case, since tau
can either precede or follow the verb to express ability of differing varieties. 142 All of the
other tests produced either allowable but strange results, genuinely questionable results,

142

See above for the discussion on the uses of tau.

106
or unacceptable results.
Thus, txhob is not functioning in the same way as a verb like kom, and must
therefore not be a regular verb. Syntactically, it can either be considered a Modal, though
not of the same class as yuav, which precedes the negative, or merely an Irrealis marker
that can express Modality. Since Dixon (2012)’s Modality markers are all considered
irrealis markers, txhob should be treated as a Modality marker for purposes of Nonspatial Setting, and this analysis is adopted here. The fact that txhob is inherently negative,
and when used as a Modal it expresses a possibility that should not come to be, however,
seems very unusual for Modals, and provides an interesting insight into what is possible
for categories of Modality.
As part of the larger system of Reality and its subcategory of Modality in White
Hmong, txhob can generally co-occur with morphemes marking Modality, with the
pattern in the data being that txhob ‘NEGATIVE IRREALIS’ modifies the verb found in the
complement of the Modal or Semi-modal, as discussed above. Another possible pattern is
for txhob to precede the special Modal txawj ‘know how to’ in order to produce a
negative imperative, as in (129) below.
(129)

Tsis

txhob
txawj
kho
NEG.IRR know.how.to heal
‘Do not know how to heal patients!’
NEG

mob.
patient

In this case, tsis ‘not’ and txhob ‘NEGATIVE IRREALIS’ precede the special Modal txawj
‘know how to’, resulting in a negative imperative. Whether this pattern can be extended
to Semi-modals is unknown, and is left for further research.

107
3.4.5 chiv ‘begin’
The morpheme chiv expresses the meaning ‘to begin, originate, start’ and is regarded by
Heimbach to be comparable to the verb pib ‘start’ (1969:24, 229). This morpheme
belongs to the Beginning action category of Phase of Activity.
An example of the use of chiv appears in (130).
(130) chiv
ua
begin do
‘To begin to do.’

(Heimbach 1969:24)

Here, chiv appears with the verb ua ‘do’, providing the ‘to begin to’ part of the translation.
Note that in the data Heimbach (1969) provides, chiv only appears before verbs.143 With
nouns, chiv is used in a compound form chivkeeb ‘begin’, 144 as in (131) below.
(131) Chiv keeb
mob
li cas?
begin145
ailment how
‘How did the ailment begin?’

(Heimbach 1969:24)

In this case, the compound chivkeeb is followed by the noun mob ‘ailment’ and the
interrogative phrase li cas ‘how?’ and has the meaning ‘begin’. Thus, it seems that on the
basis of the scant data available, chiv can only appear with verbal complements, and
chivkeeb appears in other contexts.
This apparent limitation suggests that chiv has some sort of grammatical
restriction compared to normal verbs, and, when considering its meaning, it falls
naturally into the Secondary-A verb category to which other Phase of Activity verbs
belong (following Dixon 2006:13). As such, it is clear that chiv itself belongs to the
Phase of Activity category. As part of the larger system of Phase of Activity, the

143

The morpheme chiv did not appear in the data I obtained from my language consultant, Jay Yang.
Gloss from Heimbach (1969:24).
145
Heimbach (1969) did not provide a clear gloss for the morpheme keeb here; it may be connected with
the meaning ‘basis, basic fact, the root of things’ (Heimbach 1979:82).
144

108
available data shows chiv ‘begin to’ occurring by itself without other Phase of Activity
morphemes present, suggesting that these form part of a paradigm; however, due to the
scarcity of data for this morpheme, further research is necessary.
3.4.6 pib ‘begin’
The verb pib signals the beginning of an event (following Jarkey 2006:133). It appears
before the verb expressing the event that is beginning, as a Secondary-A verb of
Beginning (Jarkey 2006:133). This morpheme belongs to the category Phase of Activity
as an indicator of Beginning action.
An example of pib appears in (132) below.
(132) Peb pib
nrhiav peb lub
hom phiaj
1PL start search 1PL CLF purpose
‘This is how we start searching for our purpose...’

li
like

no...
this

Here, pib accompanies the verb nrhiav ‘search’ to express ‘start searching’.
Jarkey (1991:74) regards pib to be a variety of deontic modality, where such
modals are accompanied by an unmarked complement, that is, no complementizer is
present. This is the case in example (132) above, where pib is immediately followed by
the rest of the verb phrase without anything in between. Jarkey further argues that pib
with a following verb do not compose a serial verb construction because they have no
arguments in common; rather, the verb phrase following serves as a second argument for
pib (1991:350). Her research is notable in that pib does take the same variety of
complement common to modals (1991:74).
Nevertheless, the semantic notion of ‘start’ hardly fits into a natural category
marking deontic modality, considering Dixon (2012)’s treatment of it as a type of Irrealis,
in spite of Jarkey (1991:350)’s observation that pib is similar syntactically to txawj ‘know

109
how to’.146 This issue is solved by Jarkey (2006), where she treats pib as belonging to a
category other than deontic modality. There, she treats pib as a ‘Secondary verb’ of
‘Beginning type’ in the verbal framework provided by Dixon (2006) (Jarkey 2006:133).
As such, pib is syntactically restricted in that it must be accompanied by another verb
(Dixon 2006:30).
In addition to this restriction, Jarkey (1991:351-352) also points out that time
adverbs associated with the complement verb that appears with pib cannot differ from
those that have sentence scope. This suggests that pib and the verb that follows are
associated more closely together than a normal verb-verb sequence, further reinforcing
the understanding that pib is grammatically less than a full verb.
Since pib ‘begin to’ manifests these grammatical restrictions that suggest that it is
not a full verb, and since ‘Secondary verbs’ are often associated with grammatical
functions in certain languages (following Dixon 2012:26), pib should be understood to be
a grammatical element. From its semantics, it is clear that it naturally belongs to Phase of
Activity, since it marks the beginning of an action. The morpheme pib is therefore a
Phase of Activity marker indicating Beginning action as part of Non-spatial Setting.
As part of the larger paradigm of Phase of Activity in White Hmong, it is unclear what
relationships or restrictions exist between pib ‘begin to’ and the other members of the
category, apart from the fact that pib generally does not co-occur with other members
such as chiv ‘begin to’, pheej ‘keep, continue to’, or tas ‘finish’ in the data.

146

See above for the discussion that places txawj into the category of Modal, though with a special
syntactic status, a verb that Jarkey (1991:350) includes with pib as a syntactic category. It is notable here
that in Jarkey (2006:133), she places pib and txawj in differing categories based on their semantics,
following Dixon (2006).

110
3.4.7 nyim ‘keep -ing’
The word nyim appears before the verb and contributes a sense of “continuous action,”
often translated as ‘keep’ (Heimbach 1969:219). It belongs to the Phase of Activity
category of Non-spatial Setting as an indicator of Continuing action.
An example of nyim appears in (133) below.
(133) Nws
nyim
quaj.
3SG
keep
cry
‘He kept crying.’

(Heimbach 1969:219)

Here, the verb quaj ‘cry’ is preceded by nyim, contributing the idea of continuing action
indicated by the word ‘kept’ in English.
Syntactically, Heimbach (1969:219) calls nyim a “preverbal intensifying particle,”
which suggests that it is a grammatical element. However, since no other data is known
to be available, it is assumed here that Heimbach’s analysis is correct, and so nyim is
treated below as a grammatical element for purposes of Non-spatial Setting. As part of
the larger system, no data is known to be available to determine whether nyim
participates in a paradigmatic relationship with other morphemes in the Phase of Activity
category, and thus further research is needed.
3.4.8 pheej/pej ‘keep, continue to’
The morpheme pheej (or pej) ‘keep, continue to’ supplies a notion of continuing action to
the main verb (following Heimbach 1969, Jarkey 1991, and Jarkey 2006). This idea of
continuing can also take on an iterative meaning such as ‘repeatedly’ (Heimbach
1969:242). In terms of word class, according to Jarkey (2006:133), pheej (or pej) is a

111
Beginning type Secondary-A verb, following Dixon (2006)’s verbal categories. 147 As
such, it has a limited distribution as it must occur with a second verb (Dixon 2006:11). As
a grammaticalized morpheme, it belongs to the Continuing action class of morphemes of
the Phase of Activity category of Non-spatial Setting.
An example demonstrating the two nuances associated with pheej—continuation
and repetition— appears in (134) below.
(134) Nws pheej noj tshuaj
pheej tsis
3SG
keep
eat medicine keep
NEG
‘He keeps taking medicine but gets no better.’

zoo.
good

(Heimbach 1969:242)

Here, the first instance of pheej appears with noj tshuaj ‘take medicine’ to signify that the
person represented by nws ‘he/she’ takes medicine ‘repeatedly’, while in the second
instance, pheej implies that the state expressed by tsis zoo is ongoing, without change. As
a result, it is clear that pheej covers notions of continuing states as well as repeated
actions.
A possible variant pronunciation of pheej is pej148, as shown in (135) below.
(135) Nws
pej
do
tsis
paub
3SG
continue.to stir NEG know
‘He continued to stir without stopping.’

tsum.
stop

Here, pej supplies the notion of ‘continued to’ for the verb do ‘stir’. In this case, the word
indicates a continuing action.
Altogether, pheej/pej marks continuing action, and thus naturally belongs to the
Phase of Activity category within Non-spatial Setting. As part of the larger paradigmatic
system of Phase of Activity in White Hmong, it is unknown whether pheej/pej can co147

A discussion on Dixon’s verbal categories is beyond the scope of this paper, though it can be noted that
virtually all verbs that fit within Dixon (2012)’s Non-spatial Setting appear to be Secondary verbs of some
sort.
148
This appears to be either dialectal or idiolectal variation: the language consultant, Jay Yang, rejected
pheej as a possible form, and consistently used pej. However, I was unable to find another source in the
literature that attests pej as an independent form.

112
occur with other Phase of Activity items, though the semantics involved suggest
otherwise; in either case, further research is needed.
3.4.9 rau siab ntso ‘continue to, commit oneself to’
The phrase rau siab ntso serves to mark Continuing action as part of the Phase of
Activity category of Non-spatial Setting. Examples appear in (136), (137), and (138)
below.
(136) ...kuv yeej
yuav tsum
ua tau xwb
ces rau siab ntso
1SG
certainly must
do can indeed then continue.to
‘...I will be able to do this, then, continue to do this...’

ua...
do

In (136), rau siab ntso appears with ua ‘do’, producing ‘continue to do’.
(137) Kuv
rau siab ntso
1SG
continue
‘I continued eating.’

noj
eat

mov.
rice

In (137), rau siab ntso appears with noj ‘eat’, resulting in the meaning ‘continued eating’.
(138) ...yog hais tias yus
rau siab ntso
ua
149
if
that
INDF.PRON
commit.to
do
‘...if you’re willing to commit yourself to doing it.’

tiag tiag
really

In (138), rau siab ntso again appears with ua, though a new nuance appears: ‘commit
yourself to doing’, also having a notion of continuity, though of a markedly volitional
nature. Thus, rau siab ntso expresses an idea of continuing action with a connotation of
intentional willfulness.
Jarkey (2006:133) provides a gloss for rau siab ntso, “put (one’s) heart into it,” or
more exactly, “put in liver steadily,” that reveals an idiomatic quality, providing support
for its existence as a unit as well as the degree of volition found in example (138) above.
In contrast to the treatment above as continuity, however, Jarkey (2006:133) treats rau
siab ntso as a verb of Trying type, rather than Beginning type like pheej. This suggests
149

Gloss following Jarkey (1991:70). Also in another example below.

113
that rau siab ntso expresses a degree of effort. While example (138) would agree with
this (as well as at least one other example in my data), examples (136) and (137) are not
completely explained. If the idea of continuing action salient in (136) and (137) is
combined with the degree of effort found in Jarkey (2006) and example (138), a solution
is found: treat rau siab ntso as continuing action with a high degree of effort. This would
then naturally contrast with pheej/pej ‘keep, continue to’, which expresses simple
continuing action. Adopting this analysis, White Hmong has two varieties of continuing
action, where rau siab ntso expresses continuing action with willful effort.150
3.5 Post-main-verb position
The “Post-main-verb position” represents the placement of Attainment SVC markers, the
second element in reduplication, the Temporal shifter ntua ‘immediately, just (now)’, the
Semi-modal taus ‘be physically/materially able to’, and the Perfect marker dua ‘EXP’.151
Other elements already mentioned, such as tau ‘ATT, be able’ can be placed in this
position as part of an Attainment SVC.
3.5.1 Attainment Serial Verb Constructions
The Accomplishment (Jarkey 1991) or Attainment Serial Verb Construction (Jarkey 2004)
is a construction that is characterized by having multiple verbs where “the final verb in
the series serves to describe the successful accomplishment or achievement of the action
or event depicted by the preceding verb” (Jarkey 1991:214). In other words, this type of

150

It is notable here that in at least some contexts, the language consultant, Jay Yang, rejected pej but
accepted rau siab ntso unless a clause stating a reason for continuing was provided.
151
The exact placement of dua in the larger sentence is ambiguous, due to a lack of clear data
demonstrating the exact ordering of dua with other post-verbal elements. It is placed in the “post-main-verb
position” provisionally, as the other non-grammaticalized morphemes that have the form dua are all
verbs—one of which means ‘to pass on, to go’ (Heimbach 1979:38)—and the post-main-verb position is a
very common position for other verbs to appear. This analysis is corroborated by the fact that the
experiential perfect marker in Mandarin, -guo, is grammaticalized from a verb meaning ‘to pass,’
suggesting a link through the typological generality aspect of grammaticalization.

114
construction serves as a form of indicating the Perfect, since it signals “successful”
completed action. While postverbal tau ‘ATT, be able’152 often occurs in this construction,
other verbs, such as cuag ‘reach’ or txog ‘arrive’ can serve in this role as well (Jarkey
1991:214, 229). An example of this construction with one of these other verbs appears in
(139) below.
(139) Kuv
mus raws
cuag
1SG
go
pursue reach
‘I caught up with them.’

lawv.
3PL

(Jarkey 1991:214)

In this example, the verb raws ‘pursue’ is followed by the second verb cuag ‘reach’ to
signal the idea of ‘catch up’, that is, expressing the successful completion of the action
indicated by raws.
As an indicator of completed action, this Attainment Serial Verb Construction
(SVC) belongs to Completion as an indicator of the Perfect, along with the more general
marker tau ‘ATT, be able’, which participates in the construction. As for its relationship
with other Completion markers, the Attainment SVC can co-occur with the marker lawm
‘CP’, as in (140) below.
(140) Nws
mus txog
tom
3SG
go
arrive over.there
‘She has arrived at the market.’

khw
market

lawm.

(Jarkey 1991:214)

CP

Here, mus ‘go’ and txog ‘arrive’ are combined to produce the meaning ‘arrive’, and
appear along with the Perfect marker lawm ‘CP’, showing that these can co-occur. As a
result, it is clear that the Attainment SVC can occur with at least one other Perfect marker,

152

Tau ‘ATT, be able’ is distinctive because of both its semantic and syntactic complexity and the fact that
the function of tau in this construction is such that the “goal” indicated by tau “is the performance of the
activity itself,” that is, the activity indicated by the main verb (Jarkey 1991:229). In other words, tau is
much more semantically bleached than other verbs used in this construction, which generally “[express] a
fairly precise goal” (Jarkey 1991:229). For a detailed discussion on tau, please see section 3.4.1 above.

115
though further research is necessary to determine the relationship of the Attainment SVC
to other Completion markers.
3.5.2 Reduplication
Reduplication of verbs generally reflects ongoing action in White Hmong, and as such,
marks Imperfect as part of Completion in Non-spatial Setting. More specifically,
reduplication of verbs produces several distinct nuances, including the imperfective,
habitual action, repetitive action, or action lasting for a period of time (Jarkey 1991:75;
Owensby 1986:237-239). Each of the functions is discussed in turn below.
First, reduplication can signal the imperfective, as in (141) below.
(141) Nws txham nws noj~noj
(Owensby 1986:239)
3SG
choke
3SG
eat~REDUP
‘He choked while he was eating.’
Here, the verb txham ‘choke’ signals an action that takes place momentarily while the
event signaled by the reduplicated verb noj ‘eat’ is happening. This usage matches Dixon
(2012:35)’s description of the imperfective, in that the “temporal make-up of the
event”—namely, that of eating in this case—is in view, and the act of choking is placed
as a single event within the larger time frame of eating.
Second, reduplication can mark habitual action, as in (142) and (143) below.
(142)

(a)

Tus

lw zaj mas nws da~da
dej
tau
CLF
lake
TOP 3SG immerse~REDUP water
ATT
‘That lake, he’s been taking his bath (in it) for several...

(b)

xyoo
no,...
year
this
...years now...’

ob peb
two three

(Jarkey 1991:75)

Here, the verb da ‘immerse’ has been reduplicated to signal a habitual action, ‘taking his
bath’, over a period of time, i.e. ob peb xyoo no ‘for several years now’.

116
(143) Nws ua~ua
paj ntaub.
(Owensby 1986:239)
3SG
do~REDUP pa.ndau
‘She makes pa ndau153 (more than she does anything else).’
Here, ua ‘do’ has been reduplicated to indicate that the making of paj ntaub is a habitual,
common activity on the part of nws ‘3SG’.
Third, reduplication can signal a non-habitual, repetitive action. This is shown in
(144) below.
(144) dhia~dhia
(Owensby 1986:238)
jump~REDUP
‘jumped many times’
Here, dhia ‘jump’ has been reduplicated to signal the notion of ‘many times’.
Fourth, reduplication can indicate an event’s taking place for an amount of time.
Examples appear in (145) and (146).
(145) Txiv Nraug Ntsuag pw~pw
ib tsam
ov,...
the.Orphan.Boy
rest~REDUP little.while IP
‘(After) the Orphan Boy slept for a while,...’

(Jarkey 1991:75)

In this case, the verb pw ‘rest, sleep’ is reduplicated and appears with the phrase ib tsam
‘a while’, signaling an event that took place for a period of time.
(146)

tau

qw~qw
(Owensby 1986:238)
ATT
shout~REDUP
‘shouted (for a long time)’

Here, the verb qw ‘shout’ has been reduplicated to indicate that it was an ongoing event
‘for a long time’, perhaps as a series of shouts over an extended period of time. Note also
the particle tau here indicating the repetitive action’s completion. Thus, reduplication can
serve to mark the imperfective, habitual action, repetitive action, and an action that lasted
for a certain amount of time.

153

Heimbach (1979:223) defines paj ntaub as ‘embroidery.’

117
In regard to the placement of reduplication into Non-spatial Setting, the four uses,
when considered separately, would fall into several different categories, namely,
Composition (Imperfective), Frequency and Degree (habitual, repetitive), and
Completion (Imperfect marking duration). However, all of the four uses can be treated as
different aspects of a single notion: ongoing action. In fact, for several of the above
categories, Owensby (1986:238) notes that “the function of reduplication of verbs may
vary according to semantic classes of verbs,” and so it should be expected that, while on
the surface there are several different functions, underlyingly there is a common element.
In this case, the imperfective meaning is drawn from the notion that the action is ongoing
rather than representing a single point in time, in contrast to its unmarked perfective
counterpart. The habitual and repetitive meanings would stem from an action that is
repeated on an ongoing basis through a longer period of time. The use associated with
duration is ongoing action for an unspecified amount of time, with no further
specialization in meaning. Thus, all of the common meanings associated with
reduplication can be connected to ongoing action. As such, it is an Imperfect as part of
Completion, in the sense that the action is ongoing and therefore incomplete. This is the
analysis adopted for purposes of Non-spatial Setting below.
3.5.3 ntua ‘just, immediately’
The word ntua “adds the idea of immediacy to the action of the verb” (Heimbach
1969:193). Heimbach (1969) treats it as a ‘restricted post verbal intensifier’, and provides
glosses such as ‘just’, ‘straight’, or ‘right’, with a nuance that the event takes place in
immediate proximity to something else, whether physical or temporal. Thus, part of the
function of this word falls into the semantic realm of Non-spatial Setting, as a Lexical

118
Time Word in Dixon (2012)’s category of Tense, while part of it does not. Examples of
ntua where it provides its temporal sense appear in (147) and (148) below.
(147) los
txog ntua
come arrive just
‘Just came, just arrived, just returned.’

(Heimbach 1969:193)

In (147), ntua appears with the verbs los ‘come’ and txog ‘arrive’ and provides the
meaning represented by the English ‘just’, that is, immediate proximity in time.
(148) Lawv
teb
ntua
kuv
3PL
answer immediately 1SG
‘They answered me immediately.’154

(Mottin 1978:118)

Here, ntua follows teb ‘answer’ to produce the meaning ‘immediately’, again indicating
proximity in time.
Heimbach (1969)’s treatment of ntua as belonging to the class of “restricted post
verbal intensifier[s]” means that he regards it as a particle (1979:468) and is “restricted”
in terms of which verbs with which it can appear. 155 Thus, it serves a grammatical role,
and belongs to Non-spatial Setting.
3.5.4 taus ‘be physically/materially able to’
The morpheme taus ‘be physically/materially able to’ is a post-verbal marker that
signifies ability (Mottin 1978:103).156 As such, it marks one type of Ability Modality in
Non-spatial Setting.

154

The gloss and translation of the French aussitôt here and in the prose following relies in part on Kellogg
(2014).
155
It should be noted here, however, that many of Heimbach (1979)’s “restricted post verbal intensifiers”
may actually be morphemes belonging to a class of onomatopoeia-like words (following Ratliff 1986),
rather than genuine grammatical elements.
156
It is notable here that, after consulting with a relative, the language consultant, Jay Yang, suggested that
the difference between taus and tau is one of euphony rather than function. Nevertheless, taus is given its
own section here as it is treated by scholarly works as a separate word. Further research is necessary to
determine this morpheme’s exact status.

119
Examples of taus marking physical ability and material ability appear in (149) and
(150) below, respectively.
(149) Ces cov uas muaj zos
thiab mus taus
kev ces
then CLF REL have strong and
go
able.to road then
‘Then, those that had strength and were able to walk, then,...’
lawv thiaj li
los
3PL
consequently157 come
‘...they went to Thailand.’

mus
go

rau
to

Thaib Teb
Thailand

lawm.
CP

In (150), taus accompanies mus kev ‘walk’ to signify ‘were able to walk’, denoting
physical ability in this case.
(150) Kuv
pab
tsis
taus
koj
1SG
help NEG able.to 2SG
‘I don’t have the means to help you.’

(Mottin 1978:103)

In (149), taus is negated with tsis, and the resulting phrase tsis taus ‘don’t have the means’
qualifies pab ‘help’ to express the idea that the speaker lacks the material ability to help.
Syntactically, taus generally behaves like post-verbal tau ‘can’ in that it follows
the verb it modifies and can take the negative marker tsis. Thus, on the basis of limited
evidence, it is likely a verb of the same class as tau, indicating Ability Modality. As part
of the larger system of Modality marking in White Hmong, there is no evidence that
would suggest that taus can co-occur with other Modality markers, suggesting a
paradigmatic relationship with these others; however, more research is necessary to prove
this.
3.5.5 dua ‘EXPERIENTIAL MARKER’
The morpheme dua generally means ‘again’, though in conjunction with tau or lawm
indicates past experience (Heimbach 1969:38, Mottin 1978:85). This latter construction

157

Gloss per Heimbach (1969:340).

120
marks a distinct variety of Perfect as part of Dixon (2012)’s Completion category, namely,
the Experiential Perfect (following the terminology of Comrie 1976:58).
Examples of this construction with dua combined with tau or lawm appear in
(151), (152), and (153) below.
(151)

Kuv tau mus dua lawm.
1SG ATT go
EXP CP
‘I’ve already been there.’

(Mottin 1978:85)

In (151), tau and dua are placed on either side of the verb mus ‘go’ and this phrase is
followed by the perfect marker lawm. This results in the meaning ‘have already been
there’.
(152) Kuv tsis
tau
mus
1SG NEG ATT go
I’ve never been there.’

dua
EXP

(ib
zaug).
(Mottin 1978:108)
(one time)158

In (152), tau and dua are again placed on either side of mus, though the negative marker
tsis appears in front of this phrase, producing the meaning ‘have never been’. The phrase
ib zaug ‘one time’ can also appear with this example, with no apparent change in
meaning.
(153) ua dua
lawm (Heimbach 1969:38)
do EXP
CP
‘To have done before.’
In (153), dua and lawm appear together with ua ‘do’, though without tau, yet still giving
the meaning ‘have done before’. In each of these cases, the use of dua with another
marker expresses past experience.
Comrie (1976:58) considers constructions marking past experience to be a form
of perfect within his ‘experiential perfect’ category. He defines this category as
“indicat[ing] that a given situation has held at least once during some time in the past
158

Gloss following Heimbach (1969:38).

121
leading up to the present” (1976:58). This falls within Dixon (2012:31)’s broader
definition of the perfect, namely, that “’perfect’ is taken to mean ‘an action which is
completed before the present time’, to which is often added ‘and which has present
relevance’,159 in that the situation in the past is relevant to the present situation in terms
of one’s personal experience. Thus, this construction with dua and tau or lawm (or both)
should be considered as a variety of perfect.
In addition, this construction is slightly different in meaning from those without
dua in that they do not treat as salient this notion of experience, whereas the construction
with dua does. In conclusion, the dua construction should be treated as distinct for
purposes of the various forms of the perfect in relation to Non-spatial Setting.
One final interesting trait of dua is the unity in its use between Comrie’s
experiential perfect and its use with the meaning ‘again’,160 shown in (154) below.
(154) ua dua
ib
zaug
do again one time
‘To do a second time.’

(Heimbach 1969:38)

Here, dua appears with the verb ua ‘do’ and is accompanied by the phrase ib zaug ‘one
time’. In (152) above, ib zaug could appear with the experiential perfect with no apparent
change in meaning compared to its absence. In both cases, the full phrase dua ib zaug can
be used, showing a distinct unity in form between the two meanings, with the difference
stemming from the presence of tau or lawm, as is the case in (152) and (153).

159

It should also be noted that Dixon (2012)’s broader definition here is remarkably similar to the general
definition of the perfect in Comrie (1976:52), which states that “the perfect indicates the continuing present
relevance of a past situation,” where the ‘present relevance’ is a primary element in focus.
160
Gloss following Heimbach (1969:38).

122
3.6 Phase of activity—Finishing position
This is the position where Finishing action is indicated by the verbal items tas ‘finish’
and tiav ‘finish’.161
3.6.1 tas ‘finish’
The morpheme tas ‘finish’ 162 is a verb that functions as a marker denoting the finishing
of an action (following Heimbach 1969:304). This fits into Non-spatial Setting as a Phase
of Activity morpheme denoting Finishing action.
An example of this notion of finishing appears below.
(155)

(a)

...ces thaum koj
muab do
tas
lau
then
when
2SG get
stir finish EP163
‘Then, when you are done stirring, then you wait...

(b)

li... [about] ib
na thi li
tej
ntawd
for
one minute around those there
‘...for about a minute or so.’

ces
then

koj
2SG

tos
wait

In this example, tas expresses the idea of ‘done’ in relation to the phrase muab do ‘stir’.
In this case, it conveys the idea that one event needs to finish before the next one begins,
in this case koj tos ‘you wait’.
Tas regularly appears with lawm ‘CP’ (in line with Jaisser 1984:15, Mottin
1978:85), as in (156) below.

161

It should be noted here that this position is more or less the same as the post-main-verb position, with
the exception that tas ‘finish’ and tiav ‘finish, complete’ are likely part of a distinct serial verb construction,
perhaps the Disposal SVC, where the object noun appears before the second verb (Jarkey 1991:240).
162
Gloss following Fuller (1988). It should be noted that tas can also very occasionally appear by itself
without an object as a regular verb. The only case of this in the data I obtained from Jay Yang may actually
be a set phrase. Further research is needed.
163
The treatment of this particle as an ‘exclamatory particle’ is based in part on Jay Yang’s description of it.
Mottin (1978:146) provides a description of a particle lauj, for which he describes its range of meaning as
including “very great joy, surprise.” These are likely the same particle.

123
(156) Kuv noj mov tas
lawm.
1SG
eat rice
finish CP
‘I’m finished eating.’
Here, the phrase noj mov ‘eat (rice)’ is followed by the complex tas lawm, producing the
idea of both ‘finishing’, supplied by tas, and a notion of current relevance, supplied by
lawm.164
Tas can also indicate when an event is not finished as part of the phrase tsis tau
tas ‘not done’. An example of this appears in (157) below.
(157) Tam sim no
kuv
tab tom
ua tsis
right.now
1SG IPRF
do NEG
‘Right now, I’m not done working on it yet.’

tau
ATT

tas
finish

In this case, the event expressed by ua ‘do’ is ongoing, and the phrase tsis tau is
accompanied by tas to mean that completion has not occurred. Thus, tas can be used with
the negative to indicate the lack of completion.
As part of the larger system of Phase of Activity, the data does not provide any
known instance where tas ‘finish’ can co-occur with other Phase of Activity morphemes
such as pib ‘begin to’, pheej ‘continue to’, or tiav ‘finish’. As a result, it is assumed here
that tas forms a paradigmatic relationship with the other members of this set, though
further research would prove useful.
3.6.2 tiav ‘finish’
The morpheme tiav ‘finish’165 is very similar to tas above in that it conveys an idea of
finishing (following Heimbach 1969), and appears to function in more or less the same
way. As such, it is an indicator of Finishing action as part of Phase of Activity.

164

See above for the discussion of Jarkey (1991) and the meaning and function of lawm.
All of the glosses from available sources, including the language consultant, make this an adjective or
past participle in English, such as ‘finished’ or ‘done’ (Heimbach 1979:320), though it will be seen below
that this is actually a verb in Hmong.
165

124
An example of tiav appears below in (158).
(158) Tam sim no
kuv
tab tom
ua tsis
tau
tiav
right.now
1SG IPRF
do NEG ATT complete
‘Right now, I’m still doing it, it’s not complete yet.’166
Here, tiav appears in place of tas in example (157) above, and there is no change in
meaning. In this case, tsis tau conveys the notion of ‘not yet’ and tiav supplies the idea of
‘complete’.
Another example, in what appears to be a serial verb construction, is (159) below.
(159) Yuav
ua tiav
huv huv li
POS.IRR do
finish completely
‘It must be completely done./(You) must finish it.’

(Heimbach 1969:320)

In this example, the idea of ‘done’ or ‘finish’ is supplied by tiav, following the verb ua
‘do’.
In terms of word class, while tiav generally appears after another verb in the data,
it passes the seven verbhood tests of Jaisser (1984). The results of these tests appear in
Table 15.
Test Type

tiav

negative tsis
yes
question particle puas
yes
“past tense” tau
yes
“future tense” yuav
yes
negative imperative txhob
yes*
post-verbal ability tau
yes
question construction with los tsis V
yes
Table 15. Test results for verbhood of tiav.

kom ‘tell’
(Jaisser 1984)
yes
yes
yes
yes
yes
yes
yes

The verb tiav passes all of the tests. One case, test 5, provided a situation where tiav
could be used by itself with the negative irrealis marker txhob, though not as the second

166

This appears to be one clause in the original, though attempting to render this into English as a single
clause would likely have a rather awkward result, hence Jay’s translation here.

125
verb in the sequence; in all other cases, tiav could follow another verb, such as ua ‘do’, as
was done with these tests. Thus, it can be readily concluded that tiav is in fact a verb.
In contrast to this, Heimbach (1969:320) provides the definition “finished, sign of
completed action” for tiav, which appears to suggest that tiav may be a grammatical
marker of some sort. However, while the definition is accurate semantically, the tests
above clearly conclude that tiav is a verb, suggesting a more complicated usage for this
morpheme. Nevertheless, tiav is most often used as a verb after another verb, and when
used after another verb, it cannot appear with txhob ‘NEGATIVE IRREALIS’,167 suggesting
altogether that in most circumstances, it is being used in a manner akin to tas above. As a
result, it can be concluded that while tiav does have properties of regular verbs, it is also
used to mark Phase of Activity in the same way as its counterpart tas, indicating
Finishing action. In terms of the larger category of Phase of Activity, tiav generally does
not appear with tas, nor with markers of Beginning and Continuing action, suggesting a
paradigmatic relationship with these others; however, more research would be necessary
to make further conclusions.
3.7 Final adverbs and particles position
The previously discussed Degree of Certainty markers kiag (li) ‘really’ and possibly also
tiag ‘really’ may appear in this position, as well as a large number of Lexical Time
Words and phrases.
3.8 lawm ‘COMPLETION MARKER’
The morpheme lawm is typically a clause-final marker that conveys an idea of
‘completion’ (following Heimbach 1969). In this sense, lawm can serve the role of a

167

There is the possibility that this restriction is due to a prohibition on txhob ‘do not, should not’ appearing
in certain positions of the construction in which tiav ‘finish, complete’ generally takes part.

126
perfect marker, where the action is finished, and may also be optionally connected to
relevance to another time (following Jarkey 1991, Dixon 2012). 168 It can also function as
a marker of finishing, where an action is signified to be completed before another begins.
In either case, it performs the role of a Perfect of completed action as part of Dixon’s
Completion category, similarly to its cognate counterpart lawm ‘Completion aspect
marker’ in Laotian Green Hmong.
The role of lawm as a Completion marker is shown below in (160).
(160) ...ces lawv thiaj li
then
3PL
consequently
‘...then they went to Thailand.’

los
come

mus
go

rau
to

Thaib Teb
Thailand

lawm
CP

In (160) above, lawm is used with a past event, namely, los mus rau Thaib Teb ‘go to
Thailand’, to signify that the action was completed (with potential continuing relevance).
It is clear that the primary meaning of lawm is to mark completion.
An extended use of lawm is when it is used to refer to a future event for which a
decision has been made. This idea of ‘already decided’ was the intuition of my language
consultant, Jay Yang, for this kind of usage. An example appears in (161) below.
(161) Kuv
yuav
ua
1SG
POS.IRR do
‘I will do (this).’

lawm.
CP

Here, the future is indicated by the Modal-type word yuav ‘POSITIVE IRREALIS’ (see
below), yet the completion marker lawm also appears, producing the meaning of ‘will do’
for yuav ua lawm. However, with the consultant’s intuition that a decision already has
been made, lawm can be taken to mean that something has in fact already occurred.
Drawing from typological considerations from the region (see above), particles like lawm
often reflect a change of state or have ongoing relevance. In this case, the change of state
168

Jaisser (1984) uses the term ‘perfective’ to characterize the usage of lawm.

127
would be the decision made, and the ongoing relevance would be connected with the
future act. Thus, lawm allows for reference to time other than the past when expressing
an idea of completion.
The morpheme lawm can also be used in conjunction with time adverbs to express
completion of time (following Mottin 1978). An example of this usage appears in (162).
(162)

Yav tag los
lawm
kuv
the.past come CP
1SG
‘In the past, I always ate dinner.’

yeej
always

noj
eat

hmo
dinner

tas
finish

li
169

PART

In (162), the phrase yav tag ‘the past’ is accompanied by los lawm, literally, ‘came’, to
signal that the period of time indicated is in the past, and has thus come to completion.
Altogether, lawm signals the completion of an act or of a period of time. As such,
it is a Perfect marker signaling completed action in the category of Completion in the
framework of Dixon (2012). In terms of the larger framework of the Perfect, however,
lawm can co-occur with the other types of Perfect, namely, the Experiential Perfect dua
‘EXP’ and the Attainment Perfect tau ‘ATT’. This co-occurrence appears in (151), repeated
in (163) below.
(163)

Kuv tau mus dua lawm.
1SG ATT go
EXP CP
‘I’ve already been there.’

(Mottin 1978:85)

Here, lawm ‘COMPLETIVE PARTICLE’ appears in the same sentence as dua ‘EXPERIENTIAL
PERFECT’ and

tau ‘ATTAINMENT PERFECT’. As a result, it is clear that these three

morphemes do not form a paradigm together, but each makes its own unique contribution
to the sentence. In other words, Dixon (2012)’s Perfect as manifested in White Hmong is
actually three distinct categories that can co-occur. However, the fact that dua in the
169

Heimbach (1969:111) lists this as a particle with a meaning such as “like, as, after the manner of.” In
final position in contexts like this, the meaning and function is not necessarily clear, and determining its use
is outside the scope of this paper. As such, when its meaning and function are unclear, it is glossed as PART.

128
Perfect sense must co-occur with with one of the other Perfect markers such as lawm
suggests that these three words are connected to one another, and each of the three should
be seen as individual parts of a more cohesive system of Perfect in White Hmong.
3.9 xwb ‘indeed’
The word xwb, as a grammatical morpheme in Non-spatial Setting, indicates a high
Degree of Certainty, and can be glossed as ‘indeed’ (Strecker & Vang 1986:5) or ‘truly,
for sure’ (Mottin 1978:114).170 It often co-occurs in the data from Jay Yang with yeej
‘certainly’, and generally appears at the end of the clause. An example appears in (n)
below.
(164) kuv yuav tsum ntseeg hais tias kuv yeej
1SG must
believe that
1SG certainly
‘I must believe that I will be able to do this...’

yuav tsum
must

ua
do

tau
can

In this case, xwb accompanies the subordinated clause kuv yeej yuav tsum ua tau ‘I will
be able to do this’ to indicate a high Degree of Certainty.
This use of xwb can also be extended to indicate emphasis, as in (n) below.
(165) ces
tas
li
then finish like
‘And this is the end.’

no
this

lawm
CP

xwb.
indeed

Here, xwb emphasizes the meaning signaled by the phrase ces tas li no lawm ‘and this is
the end’ to indicate that the end has indeed been reached.
In cases such as (n) above, xwb clearly marks a high Degree of Certainty within
Non-spatial Setting, and the emphatic use in (n) can be seen as an extension of it. As part
of the larger system of Degree of Certainty, xwb is allowed to co-occur with at least yeej

170

Note that xwb can also mean ‘only’. There are some cases in the available data where it is ambiguous
whether xwb means ‘only’ or ‘indeed’; for the purposes of this thesis, only those cases that unambiguously
mark Degree of Certainty are glossed ‘indeed’, and ‘only’ appears in all other cases.

xwb...
indeed

129
‘certainly’, while more research is necessary to determine its relationship with other
elements.
3.10 Morphemes not belonging to Non-spatial Setting
Several morphemes have been treated in the literature as belonging to categories that
form a part of Non-spatial Setting, such as Modality, but do not fulfill the definitions and
expectations put forth by Dixon (2006, 2012) for their respective category as they possess
either semantic values or syntactic behavior that deviates from both the definitions and
the tendencies of their Non-spatial Setting counterparts. The candidates closest to true
Non-spatial Setting in terms of semantics include kam ‘be willing to’, kheev ‘be willing
to’, and xav ‘want’, and these will be shown not to belong to Non-spatial Setting below.
Others also exist such as laj ‘be tired of doing’, maj ‘be in a hurry to/be busy doing’,
nyiam ‘like to’, and sim ‘try doing’ (Jarkey 1991:74), though treatment of these are
outside the scope of this thesis. 171
3.10.1 kam ‘be willing to, have the habit of’
The morpheme kam is a verb with two uses: one with the meaning ‘be willing to’ (Jarkey
1991:74), or ‘consent, agree, allow’ (Jaisser 1984)172 and precedes the main verb (Mottin
1978:97), and the other with the meaning “have the habit of” (Mottin 1978:97) or
“accustomed to” (Heimbach 1979:75) when used with the negative marker tsis after the
main verb (following Mottin 1978:97). While neither use belongs to Non-spatial Setting,
the first is relevant as it is regarded by Jarkey (1991) to be a deontic modality marker.
Each of the two uses is discussed below in turn.

171

The primary reason that these are not considered here is the fact that their semantics stray far from the
notion of Irrealis that Modality markers in Dixon (2012)’s framework contribute. It is notable, however,
that these look like what Dixon (2006) refers to as Secondary-A verbs.
172
Heimbach (1978:75) treats this as ‘willing, agreeable.’

130
The first usage of kam, which has the meaning ‘be willing to, consent, agree’, is
as a Secondary-C verb (following Jarkey 2006:133) and, as such, does not mark Modality
(in line with Dixon 2012:26). This first usage is demonstrated with the examples (166)
and (167) below.
(166) Koj puas kam mus tag kis?
(Jarkey 1991:350)
2SG Q
KAM go
tomorrow
‘Are you willing to go tomorrow?’
In (166), kam appears with the subject koj ‘2SG’ and the complement verb mus ‘go’,
where koj is the subject of both.
(167) Nws kam
kuv
txiv
ua
3SG willing
1SG
father do
‘She is willing (for) my father (to) do (it).’

(Jarkey 1991:348)

In (167), kam has the subject nws ‘3SG’, while the complement has its own subject, kuv
txiv ‘my father’. Thus, kam can have a complement where the subject is the same for both
kam and the complement verb, or a different subject for each.
As mentioned above, this verb is relevant to the discussion on Non-spatial Setting
since Jarkey (1991:350) treats the ‘be willing to’ variety of kam as a type of deontic
modality when the subject of kam and that of its complement refer to the same entity.
However, since this variety of kam can also take part in constructions such as (167), this
suggests that kam is not a Secondary-A verb in Dixon (2006)’s verbal classification
framework at all, since Secondary-A verbs must have the same subject for both the verb
itself and for its complement. As a result, since Dixon (2012:26) requires Modalitymarking morphemes to be Secondary-A verbs, it is clear that this first use of kam does
not mark Modality.
The second usage of kam, with the meaning “have the habit of” or “accustomed to”
and following the main verb, is shown in (168) below.

131
(168) Kuv
mus tsis kam
1SG
go
NEG have.the.habit.of
‘I don’t have the habit of going there.’

(Mottin 1978:97)

Here, kam follows the main verb mus ‘go’ and the negative marker tsis, producing the
meaning ‘don’t have the habit of going there’. This usage, while behaving similarly in
terms of syntax to Modality markers such as tau ‘be able’, does not appear to have any
bearing on the Non-spatial Setting of the sentence, especially when considering that an
alternative translation for kam in this kind of context is ‘accustomed to’. As a result, this
second usage should likely be treated as a grammaticalized morpheme of some sort, but
not one indicating Non-spatial Setting.
3.10.2 kheev ‘be willing, be inclined to’
The morpheme kheev meaning ‘to be willing, to be inclined to’ (Heimbach 1969:97) also
appears in the literature and is a verb similar to kam ‘be willing to, consent’ (following
Jarkey 2006). As such, it does not operate as a member of Non-spatial Setting. Also like
kam, it is relevant to the discussion here as at least one usage of it has been treated by
Jarkey (1991:350) to indicate modality.
Examples where kheev appear in a modal-like context are given in (169) and (170)
below.
(169) Kuv
kheev
mus.
1SG
be.willing.to go
‘I’m willing to go.’

(Heimbach 1969:97)

In (169), kheev appears with the single argument kuv ‘1SG’ and the verb mus ‘go’,
forming the phrase ‘I’m willing to go’.

132
(170) Kuv txiv
kheev
kuv mus tag kis
1SG
father consent 1SG go
tomorrow
‘My father consents (for) me (to) go tomorrow too.’

thiab
also

(Jarkey 2006:130)

In (170), kheev appears with the subject argument kuv txiv ‘my father’ and the
complement mus tag kis ‘go tomorrow’, where the complement’s subject, kuv ‘1SG’,
differs from the subject of the main clause. Thus, like kam ‘be willing to’, kheev allows
either a single argument or both a subject and object argument, though, unlike kam, the
two resulting meanings, namely, ‘be willing to’ and ‘consent’, are semantically much
similar, suggesting that the actual difference between the use of kheev in (169) and (170)
is merely syntactic. In this case, the primary difference between the two construction
types is the use of a different variety of the “serial-like complementation strategy”
(following the terminology of Jarkey 2006) that permits an additional subject in the case
of (170). As a result, it is reasonable that kheev does not mark Modality in Dixon
(2012)’s Non-spatial Setting, since, as stated earlier, morphemes that truly mark Modality
must be Secondary-A verbs (Dixon 2012:26), and as such, they cannot have a
complement clause with a different subject (Dixon 2006:13).
3.10.3 xav ‘want’
The word xav ‘want’ is a Secondary-B verb within Dixon (2006)’s verbal classification
scheme (Jarkey 2006:133), and as such, it is not part of the Modality system the
language. 173 This verb is relevant to Non-spatial Setting as it has been treated in the past
as an indicator of modality by Jarkey (1991). The most important piece of evidence
against xav as being part of the system of Modality of White Hmong is that, while xav
can take the “serial-like complementation strategy” when its subject and that of its

173

The following discussion on the status of xav ‘want to’ relies to a certain extent on the ideas expressed
in Jarkey (2006:131-134).

133
complement are the same, it can also take a complement when the two subjects differ
(Jarkey 2006:133). This is shown in (171) and (172) below.
(171) Kuv xav
tau
ib
1SG
want get
one
‘I want to get a book.’

phau
CLF

ntawv
book

(Jaisser 1984:84)

Here, xav is followed by the complement tau ib phau ntawv ‘get a book’, where the
subject of tau ‘get’ is the same as that of xav ‘want’.
(172) Nws
xav
kom
kuv
yuav
3SG
want PC
1SG marry
‘He wants me to marry you.’

koj
2SG

(Jarkey 2006:133)

In this case, xav ‘want’ is followed by the complement kom kuv yuav koj ‘that I marry
you’. In this case, the complement subject, kuv ‘1SG’, differs from the subject of xav, nws
‘3SG’. The fact that the subject of the complement can be different from the subject of
xav itself disqualifies it from Secondary-A verb status (Dixon 2006:13). As such, since
Dixon (2012:26) requires Modality-type verbs to be Secondary-A verbs, by definition,
xav is not part of the system of Modality in White Hmong.

134
4. Non-spatial Setting in White Hmong
This section covers the full system of Non-spatial Setting in White Hmong, organizing
the above morphemes. It is found that White Hmong has a system that contains Lexical
Time Words as part of Tense, markers of Reality, including Modality, Degree of
Certainty, Phase of Activity, Completion, and Speed and Ease. Each of these categories is
presented below in turn.174
4.1 Tense
White Hmong does not mark tense. However, it does have a system of Lexical Time
Words, as shown in Table 16.
Category of Lexical Time Words
Specific Time Spans

Examples
hnub ‘day’, vas nthiv ‘week’, hli ‘month’, xyoo
‘year’, sauv ntxov ‘morning’, hnub tseg ‘noontime’,
hnub qaij ‘afternoon’, tsaus ntuj ‘night’
With Respect to Expectation
twb ‘already’, yeej ‘already’, tseem ‘still’175
Temporal Shifters
tag kis ‘tomorrow’, neeg kis ‘the day after
tomorrow’, puag nraus ‘three days from now’, puag
nag nraus ‘four days from now’, nag hmo
‘yesterday’, hnub hmo ‘the day before yesterday’,
hnoob hnub ‘three days ago’, puag hnoob hnub ‘four
days ago’, ntua ‘just’, nyuam qhuav ‘just, a moment
ago’, maj mam ‘after a while’
Table 16. Lexical Time Words in White Hmong.
4.2 Reality
There is no distinct marker for Dixon (2012)’s Realis category in White Hmong.
However, the language has a robust system of Irrealis and Modality markers. This is
shown below in Tables 17, 18, and 19. One point of interest here is the division of labor
between Modal type markers and Semi-modal type markers, where markers of a certain
variety usually do not have equivalents of the other syntactic type.
174

For each subsection, please see the discussion for each word above for the source of the glosses found.
As mentioned above, the adverbial tab tom ‘currently’ and other Imperfect markers fit with twb and
tseem syntactically.
175

135
Category
Positive Irrealis (incl. Future
and many modalities)
Negative Irrealis

Modal type
yuav ‘POSITIVE IRREALIS,
intend’
(none)

Semi-modal type
(none)
txhob ‘NEGATIVE IRREALIS,
should not’

Table 17. Basic irrealis system in White Hmong.
Category
Necessity/Obligation
Necessity/Ability
General Ability

Modal type
yuav tsum ‘should, have to, must’,
yuav tau ‘must, have to’
(none)
(none)

Physical Ability

(none)

Acquired Ability
txawj ‘know how to’ (special)
Desire/Intent
(yuav)
Table 18. Modality system in White Hmong.

Semi-modal type
(txhob)
tsum ‘can, have to’
tau ‘can’
(tsis) yeej ‘can(not)’
taus ‘be
physically/materially able
to’
(none)
(none)

Category
Adverbial
Positive Irrealis (and Future)
mam (li) ‘will’
Table 19. Adverbial and other marking of irrealis in White Hmong.
4.3 Degree of Certainty
White Hmong has a set of words that signal high and moderate levels of certainty, as
shown in Table 20.
Level of Certainty
High

Adverb/Particle
kiag (li)176 ‘really’
tiag (tiag) ‘really’
twb ‘indeed’
yeej ‘definitely’
xwb ‘indeed’
Moderate
tej zaum ‘maybe’
ntshai ‘maybe’
Table 20. Degrees of Certainty in White Hmong.

176

In some ways, kiag behaves like a particle, and in others like an adverb, as discussed above.

136
4.4 Phase of Activity
White Hmong possesses three ways to mark beginning action, three ways to mark
continuing action, and two ways to mark finishing action, as shown in Table 21.
Phase type
Beginning
Continuing

Continuing,
Willful Effort
Finishing

Pre-verbal
pib ‘begin to’, tab tom ‘begin to’, chiv
‘begin to’
pheej/pej ‘continue to V, keep Ving’,
nyim ‘keep Ving’

Post-verbal
(none)

rau siab ntso ‘keep Ving, commit oneself
to V’
(none)

(none)

(none)

tas ‘finish Ving, be done
Ving’
tiav ‘finish Ving’

Table 21. System of Phase of Activity in White Hmong.
An interesting observation here is the ‘iconic’ nature of Phase of Activity in White
Hmong: the marking of beginning and continuing always precedes the main verb, while
finishing is always indicated after the verb. It seems here that beginning is placed before
the verb and finishing after the verb to iconically place the action’s taking place relative
to its beginning and ending in the actual ordering of the words themselves.
4.5 Completion
Completion is represented by three perfect forms, one perfect construction, two imperfect
forms, and one imperfect construction, as shown in Table 22.

137
Type of completion
Perfect, Completion with
Present Relevance
Perfect, ‘Attainment’
Perfect, ‘Experiential’177

Form
lawm ‘CP’

Construction
(none)

tau ‘have Ved’
tau...dua, dua...lawm ‘have
Ved’
tab tom ‘IMPERFECT’
sij ‘continually, repeatedly’

Attainment SVC
(none)

Imperfect
Imperfect,
Ongoing/Repetitive
Table 22. Completion markers in White Hmong.

Reduplication
(none)

4.6 Speed and Ease
White Hmong has one clear marker of speed, maj mam ‘slowly, progressively, gradually’,
marking slowness of an action.
4.7 Summary of system
Altogether, White Hmong possesses a system of Lexical Time Words, Reality and
Modality, Degree of Certainty, Phase of Activity, Completion, and Speed and Ease. As
stated above, the constituents of Non-spatial Setting in White Hmong take the following
order:
PreS NP PreMOD MOD NEGP IRR PreV Verb PostV NP PF ADP lawm xwb
In this ordering, 178 the Lexical Time Word category of Temporal Shifters generally fits
into both PreS and PreMOD positions, where some words have a preference for one or
the other position, while With Respect to Expectation adverbials always fit into the
PreMOD slot. Reality is represented by: 1) mam (li) ‘will’, which fits into the PreMOD
slot; 2) Modals, which fit into the MOD slot; 3) postverbal tau ‘can’ and taus ‘able to’,
which fit into the PostV slot; and 4) other Semi-modals, which belong to the PreV slot

177

Experiential Perfect, as a category of the perfect, following Comrie (1976).
It should be reiterated here that while the ordering presented here is the most general, exceptions can be
made for purposes of emphasis, local scope, or clausal subordination, where markers may appear in
positions other than the ones presented here. This usually results in markers being found further to the right
than as presented in this scheme.
178

138
(including txhob ‘should not, NEGATIVE IRREALIS’). Degree of Certainty is represented by:
1) kiag ‘really’, which is found in PreMOD, PostV, and ADP positions, as well as
between verbs in a serial verb construction; 2) tiag ‘really’, which seems to appear in
PreMOD179 as well as ADP position; 3) yeej ‘certainly’, which fits into PreMOD position;
4) ntshai ‘maybe’ and tej zaum ‘maybe’, which are found in the PreS slot; and 5) xwb
‘indeed’ is clause-final. Phase of Activity markers appear in several positions: all of the
Beginning and Continuing action markers are found in PreV position, and the Finishing
action markers tas ‘finish’ and tiav ‘finish’ are placed in the PF slot. As for Completion,
all Imperfect morphemes appear in the PreMOD position, while reduplication involves
the Verb and PostV positions. Perfect markers can appear in one of several locations,
depending on the marker: tau ‘ATT’ appears either in PreV or PostV position, dua ‘EXP’
appears in PostV position, and lawm appears in its own slot near the end of the clause.
The remaining means to mark the Perfect, the Attainment SVC, takes place in the Verb
and PostV positions. The slow Speed marker, maj mam ‘slowly, progressively, gradually’,
appears in PreMOD position.
4.8 Further directions of inquiry in White Hmong
While this thesis generally covers all of the common markers of Non-spatial Setting in
White Hmong, a few more potential candidates do appear in the literature, all of which
either have very little available data or their membership in Non-spatial Setting is
doubtful as they may not actually be grammatical markers or at least markers that
genuinely form a part of Non-spatial Setting. These include sam sim ‘still, in the process
of’ (Heimbach 1979:287), nyaj ‘without [a] doubt’ (Mottin 1978:115), and sub

179

This seems to be the case on the basis of example (64) above combined with the similar data on kiag,
though more research is needed to determine whether it actually appears in PreV position in that instance.

139
‘interrogative and completive particle used to express probability with some doubt
attached’ (Heimbach 1979:299). In addition, as mentioned above, further research is
needed to prove what co-occurrence restrictions and paradigmatic relationships exist
between a number of markers, especially those of the Degree of Certainty and Phase of
Activity categories, as well as the Imperfect markers of the Completion category.
4.9 Typological implications for Basic Linguistic Theory
While most of the above findings fit neatly into what would be expected for markers of
Non-spatial Setting, several important features are notable: the existence of the modallike adverbial particle mam (li) ‘will’, the three-way distinction marking various types of
Ability, the three-way distinction marking the Perfect—including an experiential
perfect—with co-occurring markers, and the iconic nature of beginning and continuing
versus finishing in terms of verbal placement as part of a larger system.
First, mam (li) is significant in that it is semantically a type of Modality marker,
but syntactically it is an adverb. As Dixon (2012:26) points out, Modality markers are
typically morphological markers or Secondary verbs of some sort. This would suggest
that a language can have a full system of Modality markers as verbs, while also
containing a morphological marker in the form of an adverb.
Second, the existence of a three-way distinction between various types of
ability—in the form of tau, taus, and txawj, as shown above—is significant in that
multiple possible forms of ability can distinctly be marked, rather than all falling into one
general category of Ability, as Dixon (2012:26) presents it for English. It is also notable
that this division of labor in the area of Ability is paralleled by Mandarin Chinese and
Honghe Green Hmong.

140
Third, the three-way contrast in the Perfect, especially the existence of the
existential perfect as a distinct category with the addition of an adverbial, is significant in
that there exists a three-way distinction in Perfect types, and more significantly, that these
three Perfect types can co-occur. The existential perfect is also a category that needs to be
considered cross-linguistically as part of Non-spatial Setting in general, especially as it is
one of Comrie (1976)’s basic forms of the perfect and because it is found as a distinct
form in many languages in East Asia.
Fourth, the iconic nature of Phase of Activity in White Hmong in its placing of
markers of Beginning, Continuing, and Finishing action is important as it shows how a
grammatical system that uses Secondary verbs can have a split where the items belonging
to the set fall into different positions (in line with Dixon 2012:40) based on the semantic
perceptualization of an event, as a part of Non-spatial Setting. In addition, it shows how
the existence of other constructions, such as the Accomplishment or Disposal Serial Verb
Constructions, may have an influence on where markers of Finishing appear in the larger
system, since as the verbs indicating the accomplishment or result of the disposal follow
the main verb, so do the markers of Finishing.
In sum, the grammatical forms for Non-spatial Setting in White Hmong present
an intriguing set of implications for this part of a grammar cross-linguistically.

141
5. Conclusion
Altogether, White Hmong possesses a system of Non-spatial Setting that includes Lexical
Time Words in the place of a tense system, Irrealis and Modality markers, Degree of
Certainty markers indicating a high level of certainty, Phase of Activity markers that
appear on one side of the verb or the other, depending on its type, a robust system of
Completion markers, and one Speed and Ease marker.

142
References
Bisang, Walter. 1996. Areal typology and grammaticalization: Processes of
grammaticalization based on nouns and verbs in East and mainland South East
Asian languages. Studies in Language 20(3). 519-597.
Brinton, Laurel J. & Elizabeth Closs Traugott. 2005. Lexicalization and language change.
Cambridge: Cambridge University Press.
Comrie, Bernard. 1976. Aspect. Cambridge: Cambridge University Press.
Clark, Marybeth. 1982. Some auxiliary verbs in Hmong. In Bruce T. Downing &
Douglas P. Olney (eds.), The Hmong in the West, 125-141. Minneapolis:
Southeast Asian Refugee Studies Project, Center for Urban and Regional Affairs,
University of Minnesota.
Clark, Marybeth & Amara Prasithrathsint. 1985. Synchronic lexical derivation in
Southeast Asian languages. In Ratanakul, Suriya, et al. (eds.), Southeast Asian
Linguistic Studies Presented to André-G. Haudricourt, 34-81. Institute of
Language and Culture for Rural Development, Mahidol University.
Dixon, R. M. W. 2006. Complement clauses and complementation strategies in
typological perspective. In Dixon, R. M. W. & Alexandra Y. Aikhenvald (eds.),
Complementation: A cross-linguistic typology, 115-136. Oxford: Oxford
University Press.
Dixon, R. M. W. 2010a. Basic Linguistic Theory, Vol. 1: Methodology. Oxford: Oxford
University Press.
Dixon, R. M. W. 2010b. Basic Linguistic Theory, Vol. 2: Grammatical topics. Oxford:
Oxford University Press.

143

Dixon, R. M. W. 2012. Basic Linguistic Theory, Vol. 3: Further grammatical topics.
Oxford: Oxford University Press.
Fuller, Judith Wheaton. 1988. Topic and comment in Hmong. Ph.D. Thesis, University of
Wisconsin-Madison. Bloomington, IN: Indiana University Linguistics Club.
Goddard, Cliff. 2005. The languages of East and Southeast Asia. Oxford: Oxford
University Press.
Heimbach, Ernest. 1969. White Meo-English dictionary. Ithaca, NY: Southeast Asia
Program, Cornell University.
Heimbach, Ernest. 1979. White Hmong-English dictionary. Ithaca, NY: Southeast Asia
Program, Cornell University.
Hopper, Paul J. & Elizabeth Closs Traugott. 2003. Grammaticalization. Cambridge:
Cambridge University Press.
Jaisser, Annie Christine. 1984. Complementation in Hmong. Unpublished MA thesis, San
Diego State University.
Jarkey, Nerida. 1991. Serial verbs in White Hmong: A functional approach. Unpublished
Ph.D. thesis, University of Sydney.
Jarkey, Nerida. 2004. Process and goal in White Hmong. In Tapp, Nicholas & Gary Yia
Lee (eds.), The Hmong of Australia: Culture and diaspora, 175-189. Canberra:
Pandanus Press.
Jarkey, Nerida. 2006. Complement clause types and complementation strategy in White
Hmong. In Dixon, R. M. W. & Alexandra Y. Aikhenvald (eds.), Complementation:
A cross-linguistic typology, 115-136. Oxford: Oxford University Press.

144
Kellogg, Michael. 2014. WordReference English-French Dictionary.
http://www.wordreference.com/fren/. (June-August 2014.)
Li, Charles N. 1991. The Aspectual System of Hmong. Studies in Language 15(1). 25-58.
Li, Charles & Sandra A. Thompson. 1981. Mandarin Chinese: A functional reference
grammar. Berkeley: University of California Press.
Long, Yaohong & Guoqiao Zheng. 1998. The Dong language: In Guizhou Province,
China. Translated by D. Norman Geary. Dallas: Summer Institute of Linguistics,
Inc. and University of Texas, Arlington.
Mottin, Jean. 1978. Éléments de grammaire Hmong Blanc. Bangkok: Don Bosco Press.
Owensby, Laurel. 1986. Verb serialization in Hmong. In Glenn L. Hendricks, Bruce T.
Downing & Amos S. Deinard (eds.), The Hmong in transition, 237-243. Staten
Island, NY: Center for Migration Studies of New York; Minneapolis, MN:
Southeast Asian Refugee Studies of the University of Minnesota.
Ratliff, Martha. 1986. Two-word expressives in White Hmong. In Glenn L. Hendricks,
Bruce T. Downing & Amos S. Deinard (eds.), The Hmong in transition, 219-236.
Staten Island, NY: Center for Migration Studies of New York; Minneapolis, MN:
Southeast Asian Refugee Studies of the University of Minnesota.
Strecker, David & Lopao Vang. 1986. White Hmong dialogues. Minneapolis, MN: Center
for Urban and Regional Affairs, University of Minnesota.
Xiong, James B. 2014. Hmong dictionary online. http://hmongdictionary.us/. (AprilAugust 2014.)
Xiong, Jay. 2012. Hmong dictionary. http://www.hmongdictionary.com/. (December
2013-August 2014.)

145
Xiong, Yuyou & Diana Cohen. 2005. Student’s practical Miao-Chinese-English
handbook. Kunming: Yunnan Nationalities Publishing House.

146
Appendix A
Narrative Text
(1) nyob zoo .

hello
Hello.
(2) ntawd no kuv
180

here

1SG

yog Txawj Zeb Yaj .
COP Txawj Zeb Yang

My name is Txawjzeb Yang.
(3) kuv zoo siab hais tias hnub no

Nej Teem wb tau tuaj nyob ua ke
hauv
1SG happy that
day this Nathan
1D can come gather.together in
kuv tsev no sib tham me ntsi txog kuv lub neej nyob tim ub tuaj txog
1SG house this talk.about a.little.bit about 1SG CLF story over.there come arrive
rau tim no.
to here
I’m very happy that today Nathan and I would be able to meet together in my house to
talk a little about my story from over there to here (Laos to America).
(4) yog li

kuv xav piav rau nws mloog .
that.is.why 1SG want tell to 3SG listen

That’s why I would like to tell it to him.
(5) kuv lub npe

1SG

CLF

hu ua Txawj Zeb .
name call do Txawj Zeb

My name is Txawjzeb.

180

The reader is referred to the relevant section of the body of the thesis for citations of the glosses of
morphemes discussed there. Other citations, when needed, are given once for the first instance and apply to
the remainder of the three appendices.

147
(6) kuv yug

1SG

nyob lub zos toj pob xeev khuam
teb chaws plog .
was.born in
CLF town Tojpob Xieng houang Laos

I was born in the town of Tojpob, in the province of Xieng Khouang, Laos.
(7) kuv txiv

1SG

lub npe hu ua tooj khwb .
father CLF name call do Toojkhwb

My father’s name was Toojkhwb.
(8) kuv niam

1SG

lub npe hu ua maum tswb vaj .
mother CLF name call do Maumtswb Vang

My mom’s name was Maochue Vang.
(9) kuv nco qab

muaj ib zawg
1SG remember one.time
los raws
kuv niam cov
come live.with 1SG mother CLF
ntau .
more

peb tsiv los nyob rau hauv lub zos naj loom
1PL move in
to in
CLF town Nalong
neej tsa peb los
nyob hauv tau li
ob xyoo
in-laws 1PL come live in
more.than two year

I remember one time we moved to live in the town of Nalong to live with my mom’s side
of the family.
(10) kuv xav

hais tias yuav
yog xyoo xya caum tsib mus txog rau xya
1SG think that
POS.IRR COP year seven ten
five go about to seven
caum xya .
ten seven
I think it was from '75 to '77.

148
(11) thaum peb los

nyob hauv muaj ib xyoo thaum peb noj peb cawg tag
when 1PL come live in
have one year when 1PL eat New.Year s finish
ces kuv txiv cias li tau ib twj
mob
tsis zoo .
181
then 1SG father have
one CLF sickness NEG good
When we came to live there, after we celebrated New Year’s, then my father had a very
bad sickness.
(12) tus mob
CLF

ntawv ua rau kuv txiv pom wb pom no tua kuv txiv .
sickness that do to 1SG father seeing.things
harm 1SG father

Then that sickness made my dad see things and came to harm him.
(13) nws pom poj koob yawm ntxwv los

nrog nws nyob .
come with 3SG live

3SG see ancestor

He saw his ancestors come to live with him.
(14) qhov nov ua rau kuv txiv

thing this do to 1SG

feeb tsis meej .
father crazy

This sickness made my dad go crazy.
(15) yog li peb nyob rau lub zos

so
1PL in
to CLF
nrog peb nyob rau ntawd .
with 1PL in
to there

naj loom ntawd peb tsis muaj kwv tij txeeb zes
town Nalong there 1PL NEG have close.relative

So, where we lived in the town of Nalong, we did not have any close relatives living with
us there.
(16) kuv txiv

1SG

cov kwv tij hos nyob rau puag tom lub zos kiaj ma na lawm .
father CLF relative then live to far there CLF town iamana PRF

My father’s relatives lived in a town called Kiamana.

181

This gloss is tentative, as other sources do not appear to contain it. The status of twj as a classifier is
based on syntax and the fact that Jay Yang treated it in glossing in a manner similar to known classifiers. It
is possibly related to Mottin (1978:67)’s twg ‘that, whichever,’ though the syntax of twj here differs
noticeably from what Mottin provides.

149
(17) kab

hais tias mus li
ib hnub ke thiaj
mus txog .
182
figure that
go about one day road consequently
go about

I guess it would take about one day’s walk to get there.
(18) yog li

thaum kuv txiv tus tij laug hlob hnov -- ua yog kuv txiv tus
and.then when 1SG father CLF older.brother hear
do COP 1SG father CLF
tij laug hlob hlob hnov -- hais tias kuv txiv mob
mob
.
eldest.brother
hear
that
1SG father sickness sickness
And then, when my dad’s older brother—that is, my dad’s oldest brother—heard of my
dad’s sickness...
(19) yog li

nws
that.is.why 3SG
thiab nws tuaj ua
and 3SG come do

thiaj li
tuaj nrog kuv txiv nyob tau ib vas nthiv
consequently come with 1SG father live ATT one week
neeb kho kuv txiv tau ib vas nthiv .
jingle heal 1SG father ATT one week

...that’s why he came to stay with my dad for a week and he (my uncle) came to “make
jingle” to heal my dad for one week.
(20) ces

ua ua
rau ces kuv txiv kuj
has
nyob
183
then do repeatedly to then 1SG father consequently
get.a.little.better in
tu zog
lawm.
not.worse.or.better CP
Then, he (my uncle) kept doing it. Then my dad calmed down.
(21) ces

nws kuj
tau hais rau kuv txiv hais tias nyob rau hauv
then 3SG consequently ATT say to 1SG father that
live to in
naj loom ntawd ces yog raws
neej tsa xwb .
Nanglong there then COP live.with in-laws only
Then he said to my dad, “Living in Nanglong, you live only with your wife’s relatives.”

182
183

Glosses for thiaj and thiaj li are from Heimbach (1979:340).
Gloss from Heimbach (1979:88).

150
(22) ces

yog li
then and.then
lub zos es nws
CLF town so 3SG
txiv .
father

nws yuav
coj kuv txiv mus nrog nws nyob tom nws
3SG POS.IRR take 1SG father go with 3SG in
there 3SG
thiaj
yuav
muaj caij nrhiav neeb nrhiav yaig kho kuv
consequently POS.IRR have time find.a.shaman
heal 1SG

Then he took my father to live with him in his town, so he would have time to go find a
shaman to come to heal my dad.
(23) yog li

nws thiaj
coj kuv txiv thiab kuv
COP about 3SG consequently take 1SG father and 1SG
tus kwv yaus
thiab kuv tus muam yaus
yog li
CLF youngest.brother and 1SG CLF youngest.sister and.then
mus lawm .
go CP

niam nrog rau kuv
mother with to 1SG
lawv thiaj li
3PL consequently

So, he took my dad and my mom with my youngest brothers and my youngest sister and
left.
(24) yog li

lawv
COP about 3PL
thiab kuv wb nyob
and 1SG 1D in
tom teb tseem ua
there farm still do

mus tau ib vas nthiv thaum ntawd tshuav kuv tij laug hlob
go ATT one week
then
have 1SG oldest.brother
tom qab rau qhov tshuav peb cov nplej
nyob rau
back
because have 1PL CLF unhulled.rice live to
tsis tau tiav .
NEG ATT done

So, they went for a week; at that time my oldest brothers and I would live back home
because we still had unhulled rice in the farm that still needed to be done.
(25) ces

yog li
wb nyob tso wb mus ntaus nplej
.
then COP about 1D in
so 1D go thresh unhulled.rice

Then that’s why we stayed so we could go finish the unhulled rice.

151
(26) yog li
wb thiaj
nyob wb nyob tau ib vas nthiv ces
that.is.why 1D consequently live 1D live ATT one week
then
thiaj li
hnov xov xwm tsis zoo tuaj txog rau wb hais tias kuv txiv
consequently hear news
NEG good come arrive to 1D that
1SG father
tau muab phom tua kuv niam thiab kuv niam tau tuag lawm .
ATT get
gun kill 1SG mother and 1SG mother ATT die CP
That’s why we stayed behind. We stayed behind for about a week, then we heard the bad
news coming to us saying that my dad got a gun and killed my mom and she died.
(27) ces

nws ho rov
muab phom tua nws tus kheej .
then 3SG turned.back get
gun kill 3SG CLF self

Then he turned the gun against himself and killed himself.
(28) ces

nkawv ob leej
tau tuag tas
sim neej nyob rau hauv lub zos
then 3D
two person ATT die finish forever in
to in
CLF town
kiaj ma na lawm .
iamana CP
Then both of them were dead forever in the town of Kiamana.
(29) yog li

qhov nov thiaj
ua rau kuv tu siab heev thiab
that.is.why thing this consequently do to 1SG sad
very and
tij laug
wb qhua.
184
older.brother 1D cry
Then that’s why this news made me very sad and my brother and I cried.

184

Gloss relying on Heimbach (1979:316).

152
(30) yog li

thaum kuv niam thiab kuv txiv nkawv tuag tas
ces peb
that.is.why when 1SG mother and 1SG father 3D
die finish then 1PL
hlob tuaj coj peb cov me nyuam mus nrog nws nyob ces yog li
peb
uncle come take 1PL CLF children go with 3SG live then that.is.why 1PL
thiaj li
tau tsiv tawm hauv naj loom ntawd thiaj
mus nrog kuv
consequently ATT move out
Nanglong there consequently go with 1SG
txib hlob nyob rau tom kiaj ma na lawm .
uncle
live to there iamana CP
When my mom and dad died, our uncle came to take us children to go live with him.
That’s why we moved out of Nanglong to go live with my uncle in Kiamana.
(31) ces

yog li
thaum peb mus nyob tod tau li
ib tog xyoo xwb ces
then and.then when 1PL go live there ATT about one half year only then
nyab laj
kuj hos tuaj tua hauv zos kiaj ma na ces peb thiaj li
tau
Vietnamese also while come kill in
town iamana then 1PL consequently ATT
tawg mus nyob hav zoov hav tsuag lawm .
out go live jungle
CP
Then when we came to live there for only about six months, then the Vietnamese
(soldiers) came and then we moved out of Kiamana to hide in the jungle.
(32) nyob rau lub caij nyoog nov yog lub caij nyoog uas mi kas

in
to CLF time
this COP CLF time
tsov rog nyab laj ntawd lawv tau swb lawm .
Vietnam war
there 3PL ATT lose CP

thiab
that America and

At this time, it was the time when America—and the Vietnam War was going on—and
they lost the war.
(33) es mi kas

tau swb rog nrog nyab laj
lawm .
so America ATT lose war with Vietnamese CP

And America lost the war with Vietnam.

153
(34) es mi kas

tau rov qab mus tsev thiab tsis tas li
ntawd lub
so America ATT go.back go house and NEG done about there CLF
tsov rog nyob rau teb chaws plog kuj tau tag
nyob rau lub sij hawm ntawd
war
in
to Laos
also ATT finish in
to CLF time
that
thiab .
also
And they went back home and also the war in the country of Laos ended at that time too.
(35) yog li
COP

ntawd sawv daws kuj tau
rov qab mus tsev .
about there everyone also get.to go.back go house

And also we got to go back home.
(36) lub sij hawm no

yog lub sij hawm uas nyab laj
tab tom nrhiav cov
CLF time
this COP CLF time
that Vietnamese IMPF
find CLF
neeg uas ua thab ham nrog rau a mi kas ntawd
tua pov tseg
.
person that do soldier with to America at.that.time kill throw.away
This time was the time that the Vietnamese were searching for those that served with
America at that time to kill them.
(37) yog li ces

peb thiaj li
tau tawg
so
then 1PL consequently ATT out
vim
kuv txiv hlob kuj yog ib tug
because 1SG father uncle also COP one CLF
nai phoo lawv thiab .
general 3PL also

mus nyob hav zoov hav tsuag lawm
go live jungle
CP
nai
thab ham ua tsov rog nrog rau
leader soldier do war
with to

And then therefore we had to move and hide in the jungle because my uncle was a
captain working with General Vang Pao too.

154
(38) yog li thaum peb mus nyob hav zoov hav tsuag tau muaj li

rau hlub hli
so
when 1PL go live jungle
ATT have about six month
mus rau ib xyoos ntawd ces nyuab heev vim
muaj me tub me nyuam thiab
go to one year there then hard very because have children
and
poj niam tub se nrog sawv daws mus nyob rau tom hav zoov hav tsuag ntawd tsis
women
with everyone go in
to there jungle
there NEG
muaj noj muaj haus txom nyem .
have eat have drink hunger
Therefore, when we moved to live in the jungle for about six months to a year, then it
was very hard because we had women and children with us hiding there in the jungle; we
were very hungry because there was nothing to eat or drink.
(39) yog li kuv txiv

hlob thiaj li
tau hais tias kom
then 1SG father uncle consequently ATT say that
hos me los sis cov laus uas mus tsis tau ces rov qab
so185 small or
CLF old that go NEG able then go.back
nyab laj
nyob rau tom zos .
Vietnamese in
to there town

peb cov uas peb
1PL CLF that 1PL
los
mus thawj
come go surrender

Then my uncle said that those who are little and the old people who cannot walk should
go back to town and surrender to the Vietnamese.
(40) ces

cov uas muaj zos
thiab mus taus
kev ces lawv thiaj li
then CLF that have strong and go able.to road then 3PL consequently
los mus rau thaib teb lawm .
come go to Thailand CP
Then those that had strength and were able to walk, then they went to Thailand.
(41) yog li thaum lawv los

mus rau thaib teb tag , ces peb thiaj li
so
when 3PL come go to Thailand end then 1PL consequently
rov qab los
mus thawj
nyab laj
.
go.back come go surrender Vietnamese
So when they had gone to Thailand, then we went and surrendered to the Vietnamese.

185

Gloss from Heimbach (1979:53).

155
(42) ces

los mus nyob hauv zos kiaj mas nas tau li
ib xyoos ntawd
then come go in
in
town iamana
ATT about one year there
ces kuv tij laug
thiaj li
rov qab txib neeg , nws muaj neeg
then 1SG older.brother consequently go.back send person 3SG have person
rov qab los
ces nws thiaj
tau hais kom lawv los
coj kuv thiab
186
go.back come then 3SG consequently ATT say PC
3PL come take 1SG and
kuv tus muam wb mus rau thaib teb .
1SG CLF sister 1D go to Thailand
Then, we stayed in the town of Kiamana for about one year. Then my brother sent
people—there were some people returning—then he told them to take me and my
sister—us—(and) go to Thailand.
(43) yog li

wb thiaj li
tau
mus rau thaib teb lawm .
and.then 1D consequently get.to go to Thailand CP

So then, we got to go to Thailand.
(44) ces

peb mus txog rau thaib teb ces
then 1PL go arrive to Thailand then
hais tias nyob rau lub teb chaws tid
that
in
to CLF country over.there
teb chaws uas nws muaj tsov muaj rog .
country that 3SG have.war

peb thiaj li
txiav txim siab
1PL consequently make.decision
txog nyem thiab nws yog ib lub
poor
and 3SG COP one CLF

Then, we got to Thailand. Then we made a decision not to stay there because that country
is poor and there were wars going on.
(45) ces

yog li
peb yuav
tau tuaj rau teb chaws a mi kas nov rau qhov
then and.then 1PL POS.IRR ATT come to country America this to thing
teb chaws a mi kas nov nws yog ib lub teb chaws tsis muaj tsov muaj rog thiab
country America this 3SG COP one CLF country NEG have.war
and
nws kuj muaj noj muaj haus zoo heev .
3SG also have eat have drink good very
That’s why we came here to America, because America is one of the countries that does
not have war and it also has plenty to eat and drink.

186

Gloss following Jarkey (2006:124).

156
ces peb thiaj li
txiav txim siab187 xam phaj tuaj rau
and.then then 1PL consequently make.decision
interview come to
teb chaws a mi kas no .
country America here
(46) yog li

So then we made the decision to interview to come here to America.
(47) yog li nyob rau lub ib

so
in
to CLF
nyob tom koos thej .
stay there Bangkok

hlis
ntuj ces peb thiaj li
sawv kev tuaj
one month sky then 1PL consequently get.up road come

So, in the month of January we came to stay in Bangkok.
(48) peb tuaj

nyob rau tod tau li
ib hlis
ces lub ob hlis
1PL come stay to there ATT about one month then CLF two month
vas nthib nee kaum plaub ib txhiab
cuaj pua
yim caum ntawd ces peb
date
twenty
four one thousand nine hundred eight ten there then 1PL
thiaj li
tuaj poob rau pem
Nebraska ped
.
consequently come fall to over.there Nebraska over.there
We came to stay there about a month, then, on February 24, 1980, we came to Nebraska.
(49) peb tuaj

1PL

nyob rau ped
tau li
tsib xyoos .
come live to over.there ATT about five year

We came to live there for five years.
(50) ces

peb thiaj li
khiav los
nyob rau hauv California nov .
then 1PL consequently move come live to in
California this

Then we moved to live here in California.

187

My transcription here corrected from txiav tsis siav to match line 44, based in part on several sources
(Heimbach (1979), Jarkey (2006), Xiong (2012)).

157
(51) thaum peb los

nyob
when 1PL come live
ntawv
tiav nyob hauv
education done in
in
nyob hauv Fresno nov .
in
in
Fresno here

hauv California nov ces thiaj li los
mus kawm
in
California this then then come go learn
nov thiab ces thiaj li
los yuav poj niam
here and then consequently come marry wife

When we came to live here in California, then I finished my education here and got
married here in Fresno.
(52) tam sim nov kuv muaj ob

tug me nyuam tub
right.now
1SG have two CLF children son
ib tug hu ua ywg pheej nrog rau kuv poj niam
one CLF call do Youpheng with to 1SG wife
no .
this

ib tug hu ua tub tsheej hos
one CLF call do Toucheng and
nws npe hu ua cua
lis
3SG name call do Choua Lee

Right now I have two children: one of them is named Toucheng, and the other one is
named Youpheng, with my wife—her name is Choua Lee.
(53) yog li

tam sim no peb nyob rau hauv Fresno.
so.then right.now 1PL live to in
Fresno

Right now, we live here in Fresno.
(54) nov qhov nov yog lub neeg

me me ntawm kuv lub neeg kuv
here thing here COP CLF person small small here188 1SG CLF person 1SG
piav qhia rau Nathan paub li
ntawd .
tell teach to Nathan know about that
This is a little story about my life that I told Nathan.

188

Gloss based on “there-nearby” found in Jarkey (1991:42). Found here and below.

158
Appendix B
Procedural Text
(1) nyob zoo .

hello
Hello.
(2) hnub no

yog lub ob hlis
ntuj vas nthiv nees nkaum plaub ob phav
day this COP CLF two month sky week
twenty
four two thousand
kaum plaub .
ten four
Today is February 24, 2014.
(3) Neb tee rov qab tuaj

rau hauv kuv tsev thiab wb yuav
sib tham
Nathan go.back come to in
1SG house and 1D POS.IRR RECP talk189
me ntsis txog kev ua noj yog li hnub no qhov kuv txiav txim siab los mus
little
about way do eat so.then day this thing 1SG make.decision come go
tham txog me ntsis yog tham txog kev ua qe noj .
talk about little
COP talk about way do egg eat
Nathan came back to my house and we will talk a little bit about cooking, so then, today
I’m going to make a decision about how to cook eggs.

189

The glossing and treatment here and in Appendix C of the phrase sib tham relies on Jarkey (1991:351)
and Jarkey (2006:129).

159
(4) qhov nov yog hmoob li

ua qe noj xwb xwb thawj koj muab dos
thing this COP Hmong POSS do egg eat first
lead 2SG get
onion
coj los
hlais kom
mos mos tas
ces yog koj hos nyiam qhiav
191
192
take come slice so.that
fine
fine finish then COP 2SG while like ginger
nrog thiab no koj muab ib qho
me me qhiav los tsoo kom
with also this 2SG get
one little.piece small small ginger come crush so.that
mos mos and then koj muab tsos cia .
small small
2SG put.it.away
190

This is the Hmong way of how to cook eggs. First, you need to get some onion and slice
it into thin (pieces), then, if you like ginger with it, you can get a little bit of ginger and
crush it into very small pieces, and then you put it away.
(5) ces

koj muab lub yias thiab koj muab li
ob diav roj coj los
then 2SG get
CLF pan and 2SG get
about two spoon oil take come
tso rau hauv lub yias .
drop to in
CLF pan
Then, you get a pan and you get about two tablespoons of oil and put it into the pan.
(6) thiab koj muab tso

and

2SG get

rau ntawm lub qhov cub
.
drop to there CLF hole cooking.place

And, you put the pan on the stove.
(7) ces

koj muab taws
rau Medium High .
then 2SG get
turn.on to

And then, you turn to Medium High.
(8) ces

koj tos li
ib mi pliag zoj li
nees nkaum xis koos kom cov roj
then 2SG wait about a.little.while about twenty
second so.that CLF oil
ntawd kub kiag lawm .
there hot really CP
Then, you wait for about twenty seconds until the oil is hot.

190

Glosses for li rely on Heimbach (1979:111).
Gloss following Jaisser (1984:58).
192
Gloss from Heimbach (1979:130).
191

160
(9) koj mam
193

2SG will

muab ob lub qe coj los
tsoo tso rau hauv .
get
two CLF egg take come crush drop to in

Then, you get two eggs and break them and put them in there.
(10) ces

koj muab tso rau hauv .
then 2SG get
drop to in

Them. you put them in there.
(11) ces

koj tso me ntsis ntsev rau thiab .
then 2SG drop little
salt to also

And then, you put a little salt in there.
(12) ces

koj muab cov dos uas koj tsuav ntawd thiab cov qhiav koj tsoo
then 2SG get
CLF onion that 2SG chop there and CLF ginger 2SG crush
ntawd muab tso rau hauv .
there get
drop to in
Then, you put the onions that you chopped there, and the ginger that you minced, put in
there.
(13) ces

koj mam li muab do rau kom zoo zoo tib si
tas
.
then 2SG need get
stir to so.that good good together finish

Then, you will need to stir until it is mixed really well.
(14) ces

thaum koj muab do tas lau , ces koj li
about ib na thi
194
then when 2SG get
stir done EP
then 2SG about
one minute
li tej ntawd .
around there
Then when you are done stirring, then you wait for about a minute or so.

193

Glosses for mam and mam li in line with Xiong (2014).
The treatment of this particle as an ‘exclamatory particle’ is based in part on Jay Yang’s description of it.
Mottin (1978:146) provides a description of a particle lauj, for which he describes its range of meaning as
including “very great joy, surprise.” These are likely the same particle.
194

161
(15) ces

koj saib cov qe yog koj nyiam noj kom siav
siav
ces
then 2SG watch CLF egg COP 2SG like
eat so.that well-done well-done then
koj ua rau kom nws siav
siav
.
2SG do to so.that 3SG well-done well-done
Then, you watch the eggs. If you like to eat them well-done, then you need to make it
well-done.
(16) thiab si li kuv no

but

as 1SG

ces kuv tsuas nyiam noj kom siav
kiag xwb .
195
this then 1SG only
like
eat so.that well-done really only

But for me, I only like to eat them just right.
(17) ces

koj ua li
ntawd ces nws zoo qab
dua
rau kuv .
196
then 2SG do about there then 3SG good taste.good more
to 1SG

Then you do it this way, it tastes better to me.
(18) qhov ntawv lau koj ua tas

, ntawm no ces koj muab coj los mus
thing paper EP 2SG do done right.here then 2SG get
take come go
tso , hais ib tais mov thiab koj muab cov qe ntawd tso rau saum toj cov
drop say one plate rice and 2SG get
CLF egg there drop to over
CLF
mov ces koj mam li noj mas zoo nkaus li qab
kawg li .
197
rice then 2SG may
eat IP good look.like taste.good very198
This is when you are done here, then you scoop it out, and put the eggs on top of the rice.
And you may eat, it looks really good.
(19) ces

kuv qhia li no rau koj paub hais tias ua qe hmoob yog ua
then 1SG teach like this to 2SG know that
do egg Hmong COP do
li cas .
how199
Then, I’m letting you know how to make Hmong eggs.

195

Gloss corroborated by Heimbach (1979:359).
Gloss following Mottin (1978:56).
197
Gloss following Fuller (1988).
198
Gloss following Mottin (1978:110).
199
Gloss following Mottin (1978:64, 67).
196

162
(20) ces

tas
li
no lawm xwb
.
200
201
then finish like
this CP
indeed

And this is the end.
(21) ua tsaug

.

thank.you
Thank you.

200
201

Gloss relying in part on Heimbach (1979:111).
Gloss following Strecker & Vang (1986:5).

163
Appendix C
Hortatory Text
(1) Nyob zoo .

hello
Hello.
(2) Ib tsoom

niam txiv kwv tij neej tsa hmov thiab tshua ntawm no kuv yog
everybody mother father brothers in-laws all.my.beloved right.here 1SG COP
xib fwb txawj zeb yaj .
Teacher Txawj Zeb Yang
All my beloved, this is Teacher Tzerge Yang.
(3) kuv yog tus yuav

tuaj muab Vaj Tswv txoj lus rau peb sawv daws sib
1SG COP CLF POS.IRR come get
God
word to 1PL everyone RECP202
cog mloog nyob rau lub caij lub nyoog no .
bring listen in
to at.this.time.and.hour this
I am the one that will come to preach to you at this time.
(4) kuv vaj

thiab cia siab hais tias nej
sawv daws yuav
koom nrog kuv
203
1SG hope and let liver that
2PL
everyone POS.IRR join with 1SG
thiab nrog kuv nyob es peb sib cog kawm Vaj Tswv txoj lus seb
and with 1SG in
and 1SG RECP bring learn God
word whether
Vaj Tswv txoj lus hais li cas rau peb es peb thiaj li
yuav
paub coj
God
word say what to 1PL and 1PL consequently POS.IRR know take
los ua peb lub neej kom muaj nuj nqis nyob rau ntawm Vaj Tswv .
come do 1PL CLF life so.that have important in
to there God
I hope that you will join with me and stay with me so we can learn the word of God
together, to see what God’s word means to us so we know that it’s important to live our
lives according to God.

202
203

Gloss based on Heimbach (1979:292) and Jarkey (2006:129).
Gloss supported by Mottin (1978:44).

164
(5) hnub no

kuv coj Vaj Tswv txoj lus tuaj yog muaj lub ntsiab hais li no
day this 1SG take God
word come COP have CLF title say as this
hais tias txhua yam tsav
yog pib ntawm Vaj Tswv los
.
204
that
each type kind
COP start here God
come
Today I will preach God’s word on the title of “Everything Starts with God.”
(6) yog nej muaj Vaj Tswv txoj lus thov

caw nej muab Vaj Tswv txoj lus
COP 2PL have God
word please invite 2PL get
God
word
los thiab peb nthuav rau phauv ntawv hau Lau Xais tshooj ib nqi kaum
come and 1PL open to book
Colossians
chapter one verse ten
rau es peb mus nyeem ua ke
seb Vaj Tswv txoj lus hais li cas rau peb es
six and 1PL go read together see God
word say what to 1PL and
peb mam li ho pib kawm ua ke
mus nawb yom .
205
1PL will
so
start learn together go TAG
If you have the word of God, please, I invite you to get the word of God and open to
Colossians chapter 1, verse 16, so we can read together to see what the word of God says
to us so we can learn and respond together.
(7) nyob rau hauv

hau Lau Xais tshooj ib nqi kaum rau nws hais li no
in
to in
Colossians
chapter one verse ten six 3SG say as this
hais tias : [ Text read from Colossians 1:16]
that
In Colossians chapter 1, verse 16, this is what it says: [Text read from Colossians 1:16]
(8) Ib tsoom

niam txiv kwv tij
everybody mother father brothers
rau hauv hau Lau Xais tshooj ib
to in
Colossians
chapter one

neej tsa , Vaj Tswv txoj lus hais li cas nyob
in-laws God
word say what in
nqi kaum rau ?
verse ten six

All my brothers and sisters, what did God’s word say to us in Colossians chapter 1, verse
16?

204
205

Gloss from “Hmong Dictionary Online by James B Xiong.”
Gloss from Heimbach (1979:53).

165
(9) Vaj Tswv txoj lus tau hais tias Vaj Tswv yog tus tsim

ib puas tsav yam
God
word ATT say that God
COP CLF create everything
tag nrho
huv tib si
tsis hais saum tej
nruab ntug los sis nyob hauv
206
207
completely
all
together NEG say above some sky208
or
in
in
lub ntiaj teb no , tsis hais tej
qhov muag pom thiab tej
qhov muag tsis
CLF earth
this NEG say some thing eye see and some thing eye NEG
pom , tsis hais tej
tim tswv los sis tej
muaj koob meej thiab cov uas muaj
see
NEG say some gods
or
some have fame
and CLF that have
meej mom .
powerful
God’s word told us that God is the creator of all things, the things that are in the sky, and
also things here on earth, and the things that the eye can see, and the things that the eye
cannot see, all the gods, and also those that are powerful.
(10) Vaj Tswv yog tus tsim

God

COP CLF

ib puas tsav yam .
create everything

God is the creator of all things.
(11) Yog li qhov ntawd qhia rau peb hais tias txhua yam tsav nyob hauv lub

so
thing there tell to 1PL that
each type kind in
in
CLF
ntiaj teb no tsis yog av luaj dag tawg es
cia li tsim muaj los xwb tiam sis
209
earth
this NEG COP come.by.itself so
let as create have come only but
nws yeej
muaj ib tug tswv yog tus tsim ib puas tsav yam ntawd tsis
3SG certainly have one CLF lord COP CLF create everything
there NEG
tas
li
ntawd xwb txawm yog peb los
Vaj Tswv yog tus tsim peb huv
210
finish about there only even
COP 1PL come God
COP CLF create 1PL all
tib si
thiab yom .
together and TAG211
So this tells us that all the things in this world did not just come up by themselves, but
there is a God who created all things—not just those things, but us also, OK?

206

Gloss and treatment based on Heimbach (1979:182).
Gloss from Heimbach (1979:56).
208
Gloss and treatment based on Xiong (2014).
209
Gloss from Xiong (2014).
210
Gloss following Xiong (2014).
211
Gloss corroborated by Heimbach (1979:429).
207

166
(12) Yog li ib tsoom

niam txiv kwv tij neej tsa sawv daws , yog hais tias
so
everybody mother father brothers in-laws everyone
COP that
koj thiab kuv peb tsis ntseeg los sis tsis paub hais tias Vaj Tswv tsim
2SG and 1SG 1PL NEG believe or
NEG know that
God
create
ib puas tsav yam cais
peb yuav
saib
peb lub neej tsis muaj nuj nqis
212
everything
then
1PL POS.IRR look.at 1PL CLF life NEG have important
thiab peb yuav
saib ib puas tsav yam hauv lub ntiaj teb no tsis muaj
and 1PL POS.IRR watch everything
in
CLF earth
this NEG have
nuj nqis dab tsi rau peb hlo li .
important what to 1PL at.all
So, all my brothers and sisters, if, you and I, we do not believe or know that God created
all things, then we will look at our lives as not important, and we will look at all things
here on earth as not important at all.
(13) Yog hais tias thaum zoo

li no lawm cais peb yuav
tsis paub hlub
COP that
when good as here CP
then 1PL POS.IRR NEG know love
thiab txuag
ib puas tsav yam uas tshwm sim nyob rau hauv lub ntiaj teb no .
and take.care everything
that appear
in
to in
CLF earth
here
When we see things like this, then we will not know how to love and take care of all the
things that appear here on earth.
(14) Peb yuav

1PL POS.IRR
hnub no xwb .
day this only

muab siv
ua lwj ua liam pov tseg
kom tag
rau peb
213
214
get
use do run do ruin
throw.away so.that finish to 1PL

We will waste all these things, just on ourselves, only for today.

212

Mr. Yang glossed the word cais as ‘then’ in two other locations, consulting an unknown source when
glossing the second of these two others (in line 30); none of my sources had this gloss for cais, however.
This gloss is applied here as well, based on the free translation.
213
Gloss supported by Heimbach (1979:295).
214
Gloss from Heimbach (1979:113).

167
(15) Peb tsis pom tag kis

hais tias muaj nuj nqis dab tsi ntxiv
rau koj
215
1PL NEG see tomorrow that
have important what anymore to 2SG
thiab rau kuv peb lawm .
and to 1SG 1PL CP
We cannot see that tomorrow is important to you or me—us—anymore.
(16) Ib tsoom

niam txiv kwv tij neej tsa kuv xav kom koj thiab kuv peb
everybody mother father brothers in-laws 1SG want PC 2SG and 1SG 1PL
paub zoo li no hais tias lub hom phiaj uas peb los
nyob rau hauv ntiaj teb
know good as here that
CLF purpose
that 1PL come live to in
earth
no tsis yog rau peb tus kheej nyob ib pliag
xwb nawb mog .
216
this NEG COP to 1PL CLF self live one moment
only TAG
My brothers and sisters, I really want you and me to understand that the purpose for
which we live in this world is not just to live for ourselves only temporarily, OK?
(17) Vaj Tswv lub hom phiaj rau peb lub neej nws zoo
God
CLF purpose
to 1PL CLF life 3SG good
tsav uas nyob rau hauv peb lub neej uas peb niaj
hnub
218
kind that in
to in
1PL CLF life that 1PL every
day

tshaj
txhua yam
217
more
each type
ua nov .
do this

God’s purpose for our lives is much better than what we have everyday here (on earth).
(18) nws zoo

tshaj peb tsev neeg , nws zoo tshaj peb lub
3SG good more 1PL family
3SG good more 1PL CLF
lub hwj chim uas peb niaj hnub ua npauv suav xav yuav
CLF glory
that 1PL every day do dream
think want
yav pem ntej .
in.the.future

hom phiaj thiab peb
purpose and 1PL
xav tau nyob rau
think get in
to

It’s better than our family. It’s better than our purpose and our glory that we dream of
having in our lives in the future.

215

Gloss from Heimbach (1979:212).
Gloss from Heimbach (1979:251).
217
Gloss following Mottin (1978:56).
218
Gloss confirmed by Heimbach (1979:140).
216

168
(19) yog peb xav

paub
if 1PL want know
hauv lub ntiaj teb no ,
in
CLF earth
this
ntawm Vaj Tswv .
there God

txog hais tias yog vim
li cas peb ho
about that
COP because what 1PL so
peb yuav tsum los
pib nrhiav peb tus
1PL must
come start search 1PL CLF

los
come
kheej
self

nyob
live
nyob
in

rau
to
rau
to

If we want to know why we came to live in this world, we must start searching for
ourselves in God.
(20) vim

hais tias nws tsim peb los ntawm nws lub hom phiaj thiab tsim
because that
3SG create 1PL come there 3SG CLF purpose and create
peb los
rau nws lub hom phiaj , tsis yog los
rau peb lub hom phiaj
1PL come to 3SG CLF purpose
NEG COP come to 1PL CLF purpose
nawb mog .
TAG

Because God created us for him and for his purpose, and it is not for our purpose, OK?
(21) yog peb nrhiav ntawm peb xwb yeej

if 1PL search there 1PL only have.success
yog tsim peb los
rau peb lub hom phiaj .
COP create 1PL come to 1PL CLF purpose

219

tsis tau vim
peb tsis
NEG able because 1PL NEG

If we look only in ourselves, we won’t be able to find it, because we are not created for
our purpose.

219

Gloss from Heimbach (1979:424).

169
(22) peb niaj

hnub nrhiav peb lub hom phiaj ntawm
1PL every day search 1PL CLF purpose there
rau hauv ntiaj teb no yog dab tsi tiag tiag tau ntau
to in
earth
this COP what really220 ATT many
lawm peb nrhiav tsis tau
thiab tsis paub txog
CP
1PL search NEG can/ATT and NEG know about

peb lub neej los
nyob
1PL CLF life come live
xyoo ntau tiam
los
year many generation come
li .
as

We are searching for our purpose in life everyday in this world. What is it that we are
looking for—for so many years and so many generations—that we cannot find and have
no knowledge of?
(23) yog vim
COP

li cas cov kwv tij es
?
221
because what CLF brothers PPART

Why is that, brothers and sisters?
(24) vim

peb nrhiav yuam kev lawm peb thiaj
nrhiav tsis
because 1PL search wrong.way CP
1PL consequently search NEG
los sis tsis paub hais tias yog vim
li cas peb hom los
nyob rau
222
or
NEG know that
COP because what 1CP so
come in
to
ntiaj teb no .
earth
this

tau
can/ATT
hauv lub
in
CLF

Because we are searching in the wrong place—that’s why we cannot find it or have no
knowledge of it—why we came to live in this world.
(25) peb pib

nrhiav peb lub hom phiaj li no peb nrhiav ntawm peb tus kheej
1PL start search 1PL CLF purpose like this 1PL search there 1PL CLF self
ua ntej .
do first
This is how we start searching for our purpose: we search in ourselves first.

220

Gloss corroborated by Heimbach (1979:319).
Gloss from Heimbach (1979:2).
222
Gloss following treatment for ho ‘so’ above.
221

170
(26) peb nug peb tus kheej li

1PL ask 1PL

CLF

self

no tias kuv xav ua dab tsi nes ?
like this that 1SG want do what IEP223

We are asking ourselves like this: “What do I really want to do?
(27) kuv yuav

1SG

POS.IRR

ua kuv lub neej zoo li cas ?
do 1SG CLF life good what

What do I really want my life to be?
(28) kuv cov hom phiaj los sis kuv cov kev

1SG CLF purpose or
1SG CLF
yav tom ntej nes yog dab tsi tiag ?
future
IEP COP what
really

224

NMLZ

npau suav rau kuv lub neej nyob
dream
to 1SG CLF life in

What are my goals or my dreams for my future?”
(29) yog hais tias peb tsuas

nrhiav peb lub neej ntawm
if that
1PL only
search 1PL CLF life there
kuv qhia rau nej hais tias peb yeej
nrhiav tsis tau
1SG teach to 2PL that
1PL certainly search NEG can
225

peb tus kheej xwb cais
1PL CLF self only then
ib zaug hlo li .
one time at.all

If we’re searching only in ourselves, then let me tell you, we won’t be able to find it.
(30) Vaj Tswv txoj lus hais li

no tias , Vaj Tswv yog tus tswj hwm peb lub
God
word say like this that God
COP CLF control
1PL CLF
neej thiab Vaj Tswv yog tus tswj hwm ib puas tsav yam uas muaj sia
nyob
226
life and God
COP CLF control
everything
that have life
in
rau hauv lub ntiaj teb no nyob rau hauv yauj tshoom kaum ob kaum .
to in
CLF earth
this in
to in
Job chapter ten two ten
This is what God’s word said: God is the one who controls our lives and God is the one
who controls all things that have life in this world, in Job chapter 12, verse 10.

223

Gloss from Heimbach (1979:136), based on ne/nej/nev, with the same meaning.
Gloss corroborated by Mottin (1978:25).
225
Gloss from Heimbach (1979:359).
226
Gloss from Heimbach (1979:298).
224

171
(31) yog li kuv xav

kom peb nkag siab
rau nqi vaj lu kub no hais tias
227
so
1SG want PC 1PL understand
to verse Bible
this say that
Vaj Tswv yog tus ua ib puas tsav yam ntawd .
God
COP CLF do everything
that
So, I would like us to understand this Bible verse that says that God is the one who
created all things.
(32) yog li peb yuav tsum los

so
1PL must
tau peb lub hom phiaj .
ATT 1PL CLF purpose

nrhiav ntawm nws xwb peb thiaj li
nrhiav
come search there 3SG only 1PL consequently search

So, we must search in God in order for us to find our purpose.
(33) twb

yeej
muaj ntau phau ntawv , ntau liab nam thiab ntau
already certainly have many book
many CLF228 movie and many
lub rooj cob qhia rau peb hais tias peb yeej
nrhiav tsis tau peb lub
CLF conference
to 1PL say that 1PL certainly search NEG can 1PL CLF
hom phiaj ntawm peb lub neeg los
ntawm peb tus kheej ib zaug hlo li .
purpose there 1PL CLF person come there 1PL CLF self one time at.all
Already there are many books, many movies, and many teaching conferences for us that
we absolutely cannot search our purpose of our lives in ourselves at all.
(34) Tej zaum peb twb

maybe

1PL

muaj coob tug sim tas lawm .
already have many CLF try229 done CP

Maybe we have many people who have tried it already.

227

Gloss and treatment following Heimbach (1979:151).
The gloss here is uncertain. It follows Jay Yang’s tendency to gloss classifiers as “the,” as he did with
liab here.
229
Gloss following Heimbach (1979:294), Jarkey (2006:133).
228

172
(35) peb twb

1PL
tau
can/ATT

tsis yog tus tsim peb nes es peb yuav
ua cas nrhiav
already NEG COP CLF create 1PL IEP and 1PL POS.IRR how230 search
peb lub hom phiaj ntawm peb tus kheej .
1PL CLF purpose there 1PL CLF self

We are not the ones who created us. How can we find our purpose in ourselves?
(36) pib txwv : koj sim saib

seb
ib tug neeg uas nws tsim ib
example 2SG try watch whether.or.not one CLF person that 3SG create one
rab riam los sis ib rab taus los sis ib rab
hlau .
231
CLF knife or
one CLF axe or
one CLF hoe
Let’s take a look at one individual who created a knife or an axe or a hoe.
(37) yog vim
COP

li cas nws ho
tsim tej
cuab yeej
ntawm ?
232
because what 3SG how.come create those household.equipment
there

What did he create all those things for?
(38) nws tsim

3SG

rab taus los
mus ndov ntoo txib taw .
create CLF axe come go chop tree split wood

He created an axe for chopping trees, splitting wood.
(39) qhov ntawd yog nws lub hom phiaj nws thiaj li

thing that COP 3SG
rau qhov ntawm .
for thing there

CLF

purpose

3SG

That is its purpose, so that he created an axe for it.

230

Gloss from Heimbach (1979:5).
Gloss confirmed by Heimbach (1979:274).
232
Gloss from Heimbach (1979:18).
231

tsim rab taus los
consequently create CLF axe come

173
(40) Nws tsim

rab hlau yog los ncaw ab thiab los
khaw nroj los ua
3SG create CLF hoe COP come till
soil and come dig.up weed come do
tej yam ntawd .
that.kind this
He created a hoe to till soil and to dig up weeds, and that’s the kind of job that it does.
(41) yog li nws lub hom phiaj nws tsim

so

3SG

CLF

purpose

3SG

rab hlau los ua qhov ntawd .
create CLF hoe come do thing that

So its purpose that the creator created the hoe for is to do that kind of job.
(42) yog li rab hlau tshwm sim vim

tus tswv uas tsim rab hlau ntawd nws
so
CLF hoe appear
because CLF lord that create CLF hoe there 3SG
muaj lub hom phiaj rau rab hlau ntawd .
have CLF purpose for CLF hoe this
So, the hole exists because the Lord who created this hoe—he has a purpose for this hoe.
(43) yog vim
COP

li cas Vaj Tswv tsim peb ?
because what God
create 1PL

Why does God create us?
(44) rau qhov vim Vaj Tswv tsim

because

God

peb los
rau nws lub hom phiaj .
create 1PL come for 3SG CLF purpose

Because God created us for his purpose.
(45) nws yeej

3SG

muaj lub hom phiaj zoo rau koj thiab kuv peb .
certainly have CLF purpose good for 2SG and 1SG 1PL

He has a good purpose for you and me—us.

174
(46) yog li Vaj Tswv thiaj li

so
xwb .
only

tsim peb los rau Vaj Tswv lub hom phiaj
consequently create 1PL come for God
CLF purpose

God

So, God created us for his purpose only.
(47) yog li koj thiab kuv peb thiaj li

nrhiav tsis tau
lub hom phiaj
so
2SG and 1SG 1PL consequently search NEG can/ATT CLF purpose
ntawm peb tus kheej vim
peb tsis yog tus tsim peb .
there 1PL CLF self because 1PL NEG COP CLF create 1PL
So, you and I—we—cannot find our purpose in ourselves because we are not the one
who created us.
(48) peb tsis muaj hom phiaj .

1PL

NEG

have purpose

We do not have a purpose.
(49) yog li peb yuav

so

1PL

POS.IRR

ua cas nrhiav tau
lub hom phiaj ntawd ?
how search can/ATT CLF purpose there

How can we find this purpose (in ourselves)?
(50) yog li Vaj Tswv txoj lus hais nyob rau hauv Yauj tshooj

so
God
word say in
to in
Job
thiaj li
hais tias : [ quote from Job 12:10]
consequently that

kaum ob kaum
chapter ten
two ten

So, God’s word says in Job chapter 12, 10, that: [quote from Job 12:10]
(51) yog li peb yuav tsum pib

nrhiav peb tus kheej nyob rau ntawm Vaj Tswv .
so
1PL have.to
start search 1PL CLF self in
to here God
vim
Vaj Tswv yog tus tsim peb rau nws lub hom phiaj .
because God
COP CLF create 1PL to 3SG CLF purpose
So, we have to start searching for ourselves in God, because God is the one who created
us for his purpose.

175
(52) qhov uas peb tshwm sim los

ua neeg nyob rau hauv lub ntiaj teb no ,
thing that 1PL appear
come do person in
to in
CLF earth
here
vim
twb
yog Vaj Tswv tsim peb thiaj li
muaj peb .
because already COP God
create 1PL consequently have 1PL
How we came about in this world is because God is the one who created us to be here.
(53) Vaj Tswv yog tus tsim

God
COP CLF
uas nws muaj ntawd .
that 3SG have here

peb thiab nws tsim peb los
rau lub hom phiaj
create 1PL and 3SG create 1PL come for CLF purpose

God is the one who created us and he created us for the purpose that he has.
(54) tsis yog rau peb lub hom phiaj yog hais tias peb tsis nkag siab

zoo li
NEG COP to 1PL CLF purpose
COP that
1PL NEG understand good like
no ces peb yeej
tsis paub lub hom phiaj uas peb los ua nyob rau
this then 1PL certainly NEG know CLF purpose that 1PL come do in
to
hauv lub ntiaj teb no yog dab tsi hlo li nawb mog , peb cov niam txiv
in
CLF earth
this COP what at.all TAG
1PL CLF mother father
kwv tij neej tsa e
.
233
brothers in-laws PPART
It is not for our own purpose; if we do not understand this, then we won’t know what we
came to live in this world for. OK, my brothers and sisters?

233

Gloss here based on treatment of es ‘PPART’ above.

176
hov uas peb yuav
nrhiav tau peb lub hom phiaj ntawm peb lub neeg
thing that 1PL POS.IRR search ATT 1PL CLF purpose there 1PL CLF person
thiab peb txoj kev uas peb yuav
mus ntawd , peb yuav tsum pib nrhiav
234
and 1PL CLF road that 1PL POS.IRR go that
1PL must
start search
ntawm Vaj Tswv mus xwb peb thiaj li
yuav
paub hais tias peb yuav
there God
go only 1PL consequently POS.IRR know that
1PL POS.IRR
mus qhov twg peb lub hom phiaj yog ua dab tsi tiag tiag .
go where
1PL CLF purpose COP do what really
(55)

The way for us to find our purpose in life and the way that we are going, we must start
searching only in God so that we will know where we are supposed to go, what is really
our purpose.
(56) txhua txoj kev uas nyob hauv ntiaj teb no

puav leej
each CLF road that in
in
earth
this all
kom peb paub txog peb lub hom phiaj uas peb los
so.that 1PL know about 1PL CLF purpose that 1PL come
ntiaj teb no .
earth
this

pab tsis tau peb
help NEG can 1PL
ua neeg nyob hauv
do person live in

All the roads in this world won’t be able to help us to know our purpose why we came to
live in this world.
(57)

uv tau nyeem ntau phau ntawv los pab kuv ua kuv lub neej .
1SG ATT read many book
come help 1SG do 1SG CLF life

I have read many books to help my life.
(58) txhua phau ntawv uas kuv nyeem no

each book
that 1SG read
neej ib yam
nkaus .
235
life the.same
way

pua leej qhia kuv nrhiav kuv lub
this all
teach 1SG search 1SG CLF

Every book that I read, they all teach me how to search for my life the same way.

234
235

Gloss relying on Mottin (1978:37).
Gloss here relying on Heimbach (1979:419).

177
(59) cov ntawv no

qhia kuv hais tias kuv yuav
muab kuv cov kev
CLF paper this teach 1SG that
1SG POS.IRR get
1SG CLF NMLZ
npau suav coj los saib seb
kuv nyiam ua dab tsi .
dream
take come watch whether.or.not 1SG like do what
These books taught me that I will have to look at my dreams to see which one I like more.
uv yuav tsum muab kuv cov kev npau suav ntawd coj los
theej
1SG must
get
1SG CLF NMLZ dream
there take come record
tseg
saib yog dab tsi tiag tiag .
236
write.down watch COP what really
(60)

I must take all my dreams and write them down to see what my dream really is.
(61) kuv yuav tsum saib

1SG must

seb
kuv txawj
ua yam twg
.
237
look.at whatever 1SG know.how do type which

I need to see what I know how to do.
(62) kuv yuav tsum ntseeg hais tias kuv yeej

yuav tsum ua tau xwb ces
1SG must
believe that
1SG certainly must
do can indeed then
rau siab ntso ua mus tsis txhob tas
kev cia siab ces ib hnub kuv
continue.to do go NEG NEG.IRR finish NMLZ hope
then one day 1SG
yeej
yuav
ua tau raws li
kuv txoj kev npau suav thiab raws li
certainly POS.IRR do can according.to 1SG NMLZ dream
and according.to
qhov kuv xav yuav xav tau .
thing 1SG think want think get
I must believe that I will be able to do this, then, continue to do this, do not lose hope,
and one day you will be able to have your dream the way you always wanted and wanted
to have.

236

Mr. Yang may have obtained this gloss from an outside source, or read material from an outside source
and thought of this gloss; in either case, none of my sources contained it.
237
Gloss based on Heimbach (1979:330) and Mottin (1978:67).

178
(63) tej

ntawv nyob hauv
those paper in
in
yus
mus txog rau
238
INDF.PRON
go about to

ntiaj teb no nws puav leej txawj
qhia
earth
this 3SG all
educated teach
tej
ntawd .
those that

Those books that are in this world, it all teaches you about all these things (dreams).
(64) kuv tsis tau hais tias tej

1SG

NEG ATT

ntaub ntawv no pab tsis tau kuv .
say that those book
this help NEG can 1SG

I did not mean that all these books cannot help me.
(65) tej

lus pab no tej zaum nws yeej
pab tau yus
kom
these word help this maybe 3SG certainly help can INDEF.PRN so.that
yus
ua tau lub neej vas meej yog hais tias yus
rau siab ntso ua
INDF.PRON do get CLF life successful COP that
INDF.PRON continue.to do
tiag tiag .
really
All these teachings, maybe it will be able to help you to become successful if you’re
willing to commit yourself to doing it.
(66) cov kwv tij

yog hais tias peb xav muaj nyiaj ces peb yuav tsum
CLF brothers COP that
1PL want have money then 1PL must
rau siab ntso mus khwv tiag tiag tas hnub tas hmo xya hnub xya hmo
continue.to go work really
done day done night seven day seven evening
ces peb yeej
yuav tsum muaj nyiaj xwb .
then 1PL certainly must
have money indeed
My brothers, if we want to have money, then we must commit ourselves to working
really hard day and night, seven days and seven nights. Then we will have money.

238

Gloss following Jarkey (1991).

179
(67) yog peb xav

ua ib tug
COP 1PL want do one CLF
yeej
yuav
tau los
ua
certainly POS.IRR ATT come do

kws kho mob ces peb rau siab mus kawm peb
doctor
then 1PL aspire! go learn 1PL
xwb .
indeed

If we want to become a doctor, then we aspire to learn, (then) we will become one.
(68) tiam sis qhov kuv coj

tuaj qhuab qhia peb hmo
no los sis hnub no
but
thing 1SG take come preaching 1PL evening this or
day this
los sis lub sij hawm no tsis yog tej
kev vas meej nyob rau hauv lub
or
CLF time
this NEG COP these NMLZ successful in
to in
CLF
ntiaj teb no xwb los sis nyob rau peb lub neej no xwb .
earth
this only or
in
to 1PL CLF life this only
But what I bring to preach to us tonight or today or in this hour, it is not about the success
in this world or in our life here.
(69) kuv hais txog

lub hom phiaj uas yog vim
li cas
1SG say about CLF purpose that COP because what
koj thiab kuv peb los nyob rau hauv lub ntiaj teb no
2SG and 1SG 1PL come live to in
CLF earth
this
tiag tiag .
really

Vaj Tswv ho tsim
God
so create
nws yog dab tsi
3SG COP what

What I’m talking about: what really is the purpose that God created you and me—us—to
live in this world for.
hov
thing
txog yog
about COP
(70)

ntawd lau
yog qhov kuv xav coj koj thiab kuv peb sib tham
this
EMPH COP NMLZ 1SG want take 2SG and 1SG 1PL RECP talk
qhov ntawd .
thing here

This is it! It is what I want to take you and me—us—into discussion about what I was
saying before.

180
(71) tsis yog tej

kev

vas meej nyob rau ntawm yus
tus kheej nyob
NEG COP those NMLZ successful in
to there INDF.PRON CLF self in
rau hauv lub ntiaj teb no xwb .
to in
CLF earth
this only
This is not about the success in yourself in this world.
(72) rau qhov hais tias lub hom phiaj uas kom

yus

lub neej vas meej
because that
CLF purpose
that cause239 INDF.PRON CLF life successful
nyob hauv lub ntiaj teb no thiab lub hom phiaj ntawm yus
lub neej yog
in
in
CLF earth
this and CLF purpose there INDF.PRON CLF life COP
vim
li cas yus
ho los nyob hauv ntiaj teb no txawv txav tsis zoo
because what INDEF.PRON so come in
in
earth
this different
NEG good
ib yam nawb mog cov kwv tij .
the.same TAG
CLF brothers
Because the purpose that makes your life a success in this world and the purpose why you
are in this world are different—not the same. OK, my brothers?
(73) kuv xav

kom peb paub li no peb thiaj li
paub los mus nrhiav
1SG want PC 1PL know like this 1PL consequently know come go search
peb lub hom phiaj nyob rau ntawm Vaj Tswv .
1PL CLF purpose in
to there God
I want us to know like this so that we know to search for our purpose in God.
(74) tawm yog koj thiab kuv peb twb

ua tau peb lub neej vas meej nyob
even COP 2SG and 1SG 1PL already do get 1PL CLF life successful in
rau hauv lub ntiaj teb no lawm los peb yeej
tsis tau paub hais tias yog
to in
CLF earth
this CP
come 1PL certainly NEG ATT know that
COP
vim
li cas Vaj Tswv ho tsim peb los nyob rau hauv lub ntiaj teb no .
because what God
so create 1PL come live to in
CLF earth
this
Even you and I, we already have a successful life in this world now, but we still don’t
know why God created us to live in this world.

239

Gloss from Heimbach (1979:85).

181
(75) Vaj Tswv txoj lus hais li

God

no tias : [ quote from Matthew 16:26].
say like this that

word

This is what God’s word says: [quote from Matthew 16:26].
(76) nyob rau hauv ma thais tshooj

in

to in

kaum rau nqi nee nkaum rau hais li no
Matthew chapter ten six verse twenty
six say like this

.
In Matthew chapter 16, verse 26, it says like this.
(77) nqi

vaj lub kub no tau hais tias yog peb nrhiav peb lub hom phiaj ntawm
verse Bible
this ATT say that COP 1PL search 1PL CLF purpose there
peb tus kheej xwb peb yeej
nrhiav tsis tau
li nawb mog .
1PL CLF self only 1PL certainly search NEG can/ATT TAG
This Scripture has said that we cannot find our purpose in ourselves. We will not be able
to find it. OK?
(78) peb yuav tsum los

nrhiav
1PL must
come search
yuav
nrhiav tau
peb tus
POS.IRR search can/ATT 1PL CLF

ntawm Vaj Tswv txoj kev
there God
CLF way
kheej thiab qhov
tseeb
240
self and thing
true241

xwb peb thiaj li
only 1PL consequently
ntawm peb tus kheej .
there 1PL CLF self

We must search only God’s way so that we can find ourselves and the truth about
ourselves.

240
241

Gloss following Mottin (1978:25).
Gloss following Heimbach (1979:350).

182
(79) qhov nov lau

cov kwv tij es qhov kuv qhia hnub no los sis lub
thing this EMPH CLF brothers PP NMLZ 1SG teach day this or
CLF
sij hawm no rau koj thiab kuv peb kawm no tsis yog qhia li tej
time
this to 2SG and 1SG 1PL learn this NEG COP teach as those
ntau ntawv hauv ntiaj teb no qhia los
nrhiav ib txoj hauj lwm zoo rau koj
book
in
earth
this teach come search one CLF job
good for 2SG
, los
pab kom koj tau raws
li koj txoj kev npau suav , los
come help so.that 2SG get in.accordance.with as 2SG NMLZ
dream
come
pab koj npaj
koj lub neej hauv ntiaj teb no .
help 2SG prepare 2SG CLF life in
earth
this
This is it, my brothers! What I’m teaching today or at this time for you and me and us to
learn is not the same as what the books of this world teach you—how to find a good job
for you, help you fulfill your dreams, teach you how to help prepare for your life in this
world.
uv tsis yog yuav
qhia koj nrhiav lub nras242 hnyav ntxiv los
rau
1SG NEG COP POS.IRR teach 2SG search CLF burden heavy another come to
koj lub neej .
2SG CLF life
(80)

I’m not trying to teach you to find another burden for your life.

242

Mr. Yang originally gave this word as nra, though after checking an outside source, he decided on nras.
This appears to be a case of tone sandhi in any case (following Jarkey 1991:25, Mottin 1978:18, etc.).
Heimbach (1979:173) glosses nra as ‘goods, baggage’—a clear match.

183
(81) tiam sis qhov kuv qhia hnub no

los sis hmo
no los sis lub sij hawm
but
thing 1SG teach day this or
evening this or
CLF time
no yog qhia hais tias peb yuav
nrhiav li cas thiaj li
paub nrhiav lub
this COP teach that
1PL POS.IRR search how consequently know search CLF
nras
kom sib los
rau peb lub neej los ntawm kev saib
seb
burden so.that light come to 1PL CLF life come there NMLZ look.at whatever
yam twg yog yam tseem keeb tshaj es ua yam ntawd mus xwb .
type which COP type important more then do type that243 go only
But what I’m teaching you today or this evening or at this time is teaching us how to find
a lighter burden for our lives by looking at which one is the more important thing, then do
that one only.
(82) txhob

mus lob yam ub yam no yam pev
yam nrav
los
rau
NEG.IRR go grab type this type this type over.there type over.here come to
peb tus kheej .
1PL CLF self
Don’t go grab things from here and there, over there and over here, for ourselves.
(83) qhov ntawd nws yuav

ua lub nras
hnyav
thing that 3SG POS.IRR do CLF burden heavy
kev nyuab siab rau peb nyob rau hauv lub ntiaj teb
trouble
to 1PL live to in
CLF earth

thiab yuav
ua
and POS.IRR do
no .
this

These things, it will make a heavy burden and will make troubles for us to live in this
world.
(84) yog qhia kom peb paub ua peb lub neej los sis ua ib

tug neeg
COP teach PC
1PL know do 1PL CLF life or
do one CLF person
raws li
Vaj Tswv lub hom phiaj uas nws tsim peb los ua neeg nyob
according.to God
CLF purpose
that 3SG create 1PL come do person live
rau hauv lub ntiaj teb no xwb .
to in
CLF earth
this only
It is teaching us to know our lives or to become a person according to God’s purpose
alone, for which he created us to live in this world.

243

Gloss following Heimbach (1979:186).

184
(85) yog li peb yuav

ua li cas peb thiaj li
paub txog peb lub hom phiaj
so
1PL POS.IRR do what 1PL consequently know about 1PL CLF purpose
hais tias yog vim
li cas Vaj Tswv thiaj li
tsim peb los nyob rau
that
COP because what God
consequently create 1PL come live to
hauv lub ntiaj teb no ?
in
CLF earth
this
So, what do we need to do so that we know about our purpose why God created us to live
in this world?
(86) lub hom phiaj ntawm peb lub neej uas Vaj Tswv tsim

purpose there 1PL CLF life that God
hauv ntiaj teb no yog dab tsi tiag tiag ?
in
earth
this COP what really
CLF

peb los nyob rau
create 1PL come live to

What is really the purpose in our lives that God created us to live in this world for?
(87) qhov yooj yim tshaj plaws ces

rov qab mus cuag
tus uas tsim koj xwb
thing easy
more.than then go.back go contact CLF that create 2SG only
, nws thiaj
yuav
paub hais tias yog vim
li cas .
3SG consequently POS.IRR know that
COP because what
The easiest way is to go back to the one who created us—he will know why.
(88) Vaj Tswv yeej

tsis tau tso peb nyob tsaus ntuj nti tsis paub dab tsi
God
certainly NEG ATT drop 1PL live darkness
NEG know what
hlo li thiab niaj hnub tso peb xav hais tias peb lub neej ces yog li no
at.all and every day drop 1PL want that
1PL CLF life then COP like this
los sis yog li tod xwb .
or
COP like that only
God does not let us stay in the dark knowing nothing at all, and does not let us think that
our lives are just like this or like that.

185
(89) Vaj Tswv tau qhia

tseeb tseeb txog
244
God
ATT teach the.truth
about
lub neej nyob rau hauv Vaj Tswv txoj lus
CLF life in
to in
God
word

tsib yam hom phiaj nyob rau hauv peb
five type purpose in
to in
1PL
uas yog phau vaj lus kub .
that COP book Bible

God has taught us the truth about five types of purpose in our lives in God’s word which
is the Bible.
(90) nyob rau hauv Vaj Tswv txoj lus tau qhia rau peb hais tias yog vim

in
to in
God
word ATT teach to 1PL that
COP because
li cas peb thiaj li
muaj sia nyob , peb yuav
ua peb lub neej li cas ,
what 1PL consequently have life live
1PL POS.IRR do 1PL CLF life what
yam twg yog yam uas peb yuav tsum tsis txhob ua thiab nws muaj dab tsi
which
COP type that 1PL must
NEG NEG.IRR do and 3SG have what
tos peb nyob rau peb lub neej yav pem suab .
wait 1PL in
to 1PL CLF life in.the.future
In the word of God, it teaches us why we are able to live, how we will live our lives,
which one is the one that we must not do, and what he has waiting for us in the future.
(91) Vaj Tswv txoj lus hais li

God

no hais tias : [ quote from 1 Corinthians 2:7].
say like this that

word

God’s word says like this, that: [quote from 1 Corinthians 2:7].
(92) Nyob rau hauv ib

in

to in

khau li thaus tshooj ob nqi xya hais li ntawm .
one Corinthians chapter two verse seven say like there

In 1 Corinthians chapter 2, verse 7, it says like that.
(93) Vaj Tswv tsis yog txoj sia uas xub pib tuaj

God
NEG
hauv paus siab .
foundation life

COP CLF

life that start

xwb tiam sis nws yog lub
come only but
3SG COP CLF

He is not only the starting point of life, but he is the founder of life.

244

Gloss is Mr. Yang’s original, which is corroborated as “spoken truth” in Heimbach (1979:350).

186
(94) yog peb xav

paub txog peb lub hom phiaj ntawm peb lub neej hais tias
COP 1PL want know about 1PL CLF purpose
there 1PL CLF life that
yog vim
li cas peb ho los
nyob hauv lub ntiaj teb no peb yuav tsum los
COP because what 1PL so come live in
CLF earth
this 1PL must
come
nrhiav ntawm Vaj Tswv uas yog tus tsim peb nawb mog cov kwv tij .
search there God
that COP CLF create 1PL TAG
CLF brothers
If we want to know about our purpose in our lives—why we are here in this world—we
must come search in God, the one who created us, my brothers.
(95) peb tsis txhob

1PL

NEG NEG.IRR

mus nrhiav ntawm neeg lub tswv yim .
go search there person CLF wisdom

We shouldn’t go search in the wisdom of man.
(96) peb yuav tsum ua peb lub neej nyob rau ntawm txoj kev tseeb uas nyob

1PL must
do 1PL CLF life live to there NMLZ
true that lives
tas mus ib txhis tsis zoo li neeg ntiaj teb lub tswv yim es
nyob ib pliag
245
forever
NEG like
person earth
CLF wisdom so
live temporarily
xwb .
only
We must live our lives in the truth that lives forever, not like the wisdom of man in this
world that lives only temporarily.
(97) thaum muaj tus neeg

when have
li no lawm .
like this CP

CLF

txawj
tshiab los ces nws yuav
hais tias yog
person educated new come then 3SG POS.IRR that
COP

When there is a new educated person coming, then he will say that it is like this now.
(98) Vaj Tswv txoj lus hais li

God

word

no ntxiv : [ quote from Ephesians 1:11].
say about this add

In addition, this is what God’s word says: [quote from Ephesians 1:11].

245

Gloss from Xiong (2014).

187
(99) nyob rau hauv E fe xaus tshooj

in

to in

ib nqi kaum ib hais li ntawm .
Ephesians chapter one verse ten
one say like there

In Ephesians chapter 1, verse 11, (it) says like that.
(100) nyob rau nqi

in

Vaj lub kub no nws muaj peb lub ntsiab lus qhia peb .
to verse Bible
this 3SG have 1PL CLF meaning teach 1PL

In this Bible verse here, it has three meanings to teach us.
(101) lub ntsiab lus ib

peb
CLF meaning one 1PL
hom phiaj los
ntawm peb
purpose come there 1PL
.

yuav tsum nrhiav tau peb tus kheej
must
search can 1PL CLF self
muaj txoj kev sib rauv
zoo nrog
have NMLZ
relationship good with

thiab peb lub
and 1PL CLF
Ye Xus he Tos
Jesus Christ

The first meaning: we must be able to search ourselves and our purpose by having a good
relationship with Jesus Christ.
(102) qhov ob

tuaj mus Vaj Tswv twb
paub peb ua ntej peb paub nws .
thing two come go God
already know 1PL first
1PL know 3SG

Here is the second meaning: God already knew us before we knew him.
(103) Vaj Tswv twb

npaj
peb lub neej thiab peb lub hom phiaj thaum tsis
God
already prepare 1PL CLF life and 1PL CLF purpose when NEG
tau muaj peb lawm .
ATT have 1PL CP
God already prepared our lives and our purpose when we did not yet exist.
(104) lub ntsiab lus thib peb hais li

no hais tias
CLF meaning three
say about this that
peb tseem loj tshaj li
qhov peb xav yuav xav
1PL still big more than thing 1PL think want think

Vaj Tswv lub hom phiaj rau
God
CLF purpose
to
tau .
get

The third meaning says like this: God’s purpose for us is bigger than what we think and
want.

188
(105) Vaj Tswv lub hom phiaj yog kom

God
mus ib txis .
go forever

CLF

purpose

COP

peb muaj txoj sia nrog Vaj Tswv nyob
so.that 1PL have CLF life with God
live

God’s purpose is for us to have life with him forever.
(106)

hov no lau
!
thing this EMPH

This is it!
(107) Cov kwv tij

yog lub hom phiaj uas Vaj Tswv xav kom koj thiab kuv
CLF brothers COP CLF purpose
that God
want PC 2SG and 1SG
peb tau thiab yog qhov Vaj Tswv tsim peb los nyob rau hauv lub ntiaj teb
1PL get and COP thing God
create 1PL come live to in
CLF earth
no .
this
My brothers, this is the purpose that God wants you and me—us—to have and is what
God created us to live in this world here for.
(108) Yog li kuv xav

kom koj thiab kuv peb txhob niaj hnub mus nrhiav
so
1SG want PC 2SG and 1SG 1PL NEG.IRR every day go search
peb lub neej nyob rau ntawm peb tus kheej lawm los sis peb lub hom phiaj rau
1PL CLF life in
to there 1PL CLF self CP
or
1PL CLF purpose to
ntawm peb tus kheej .
there 1PL CLF self
So, I do not want you and me—us—to go search for our lives in ourselves or our purpose
in ourselves everyday.
(109) tiam sis peb yuav tsum mus nrhiav rau ntawm Vaj Tswv peb thiaj

but
1PL must
go search to there God
paub hais tias lub hom phiaj nyob rau ntawm Vaj Tswv .
know that
CLF purpose
in
to there God

1PL consequently

But, we must search in God. So, we know that our purpose is in God.

189
(110) los sis peb lub hom phiaj yog dab tsi tiag tiag .

or

1PL

CLF

purpose

COP

what

really

Or we know what is really our purpose.
(111) nawb mog .
TAG

OK?
(112) Es

kuv vam thiab siab hais tias nej sawv daws yuav
nkag siab zoo es
246
and 1SG hope
that
2PL everyone POS.IRR understand good and
nej yuav
muab qhov no coj los ua ib txoj kev rau nej npaj
nej lub
2PL POS.IRR get
thing this take come do one CLF way to 2PL prepare 2PL CLF
neeg nyob rau hauv lub ntiaj teb no nawb mog .
person live to in
CLF earth
this TAG
And I hope that everyone will understand well and will take this as one way to prepare
your lives here in this world. OK?
(113) Es

yog li
and so
pub koob hmoob
bless

thov Vaj Tswv nrog nraim nej thiab thov Vaj Tswv
please God
with
2PL and please God
nej cov uas nej koom nrog kuv nyob rau lub sij hawm no .
2PL CLF that 2PL join with 1SG in
to CLF time
this

And may God be with you and may God bless you—those that join with me at this hour.
(114) Es

los
come

peb mam li rov
sib ntsib dua nawb mog yog Vaj Tswv tsis tau
and 1PL will
return meeting again TAG
COP God
NEG ATT
.

And we will meet again—OK?—if God has not come.

246

Mr. Yang found the gloss ‘hope’ for vam in an outside source, and decided that ‘hope’ best matched the
full phrase here. The gloss for vam as ‘hope’ is corroborated by Jarkey (2006:133).