Sketch of Lakhota, a Siouan Language

DAVID S. ROOD AND ALLAN R. TAYLOR

Handbook of North American Indians, Vol.17 (Languages), pp.440-482.

Copyright 1996 by Smithsonian Institution. All rights reserved.

Short Table of Content      Full T.O.C.

Introduction

1. Phonology
2. Vowels
3. Stress and Pitch
4. The Phonemic Structure of Words

Grammar
5. Interjections
6. Conjunctions
7. Adverbs
8. Nominals
9. Verbs
10. Enclitics
11. Selected Vocabulary


Full Table of Content

Introduction

1. Phonology
1.0. Consonants
1.1. Stops
1.1.1. Aspirated Stops
1.1.2. Glottalized Stops
1.1.3. Voiced Stops
1.2. Fricatives
1.2.1. Sibilants
1.2.2. Velar Fricatives
1.3. Sonorants
1.4. Glides
1.4.1. /w y h/ at the Beginning of a Words
1.4.2. Glides between Vowels
1.4.3. Before and between Nasalized Vowels
2. Vowels
2.1. Oral Vowels
2.2. Nasalized Vowels
2.2.1. Nasalized vowels after Nasal Consonants
2.3. Diphthongs and Contracted Vowels
3. Stress and Pitch
3.1. Secondary Stress
3.2. Pitch on Long Vowels
4. The Phonemic Structure of Words
4.1. Segmental Sequences
4.2. Syllabification
4.3. Phonological Changes in the Basic Form of Words
4.3.1. Phonological Changes that Result from Speaking Style
4.3.1.1. pi in Rapid Speech
4.3.1.2. Vowel apocope
4.3.1.3. Dropping of Glides and /?/
4.3.2. Phonological Changes that Result from Grammatical Processes
4.3.2.1. Stress shift
4.3.2.2. Insertion of /?/
4.3.2.3. Nasalization Spread
4.3.2.4. Change of /k/ to/č/
4.3.2.5. The Affix ma-
4.3.2.6. Final Vowels in A-words
4.3.2.7. Change of /p t č k/ to/b/ /g/
4.3.2.8. Reduplication

Grammar

5. Interjections
6. Conjunctions
7. Adverbs
7.1. Single-word Adverbs
7.2. Adverbial Phrases
7.2.1. Adverbial Phrases of Time
7.2.2. Adverbial Phrases of Place
7.2.3. Adverbs of Manner
7.2.4. Instrumentality
7.3. Adverbial Clauses
8. Nominals
8.1. Pronouns
8.2. Nouns
8.2.1. Compound nouns
8.2.2. Derived nouns
8.3. Determiners
8.3.1. Articles
8.3.2. Demonstratives
8.3.3. Quantifiers
8.3.4. Partitive Markers
8.3.5. Summary of determiners
8.4. Modified Nouns
8.4.1. Possession
8.4.2. Other Modified Nouns
8.4.2.1. Definite Modified Noun Phrases
8.4.2.2. Indefinite Noun Phrases with Modifiers
8.4.2.3. Relative Clauses
8.5. Sentences as Nominals
8.5.1. Marked Nominalized Sentences
8.5.2. Unmarked Nominalized Sentences
8.5.3. Other Nominalized Sentences
9. Verbs
9.1. Sentence and Affix Types
9.1.1. Impersonal Verbs
9.1.2. Stative Verbs
9.1.3. Stative transitive Verbs
9.1.4. Active intransitive Verbs
9.1.5. Active transitive Verbs
9.1.6. Verbs Requiring Three Complements
9.2. Verbal Derivation
9.2.1. Complex Stem Formation
9.2.2. Class-changing Processes
9.2.2.1. Causatives
9.2.2.2. Inceptives
9.2.2.3. Deactivization or Stativization
9.3. Verb Inflection
9.3.1. Stative Affixes
9.3.2. Object Affixes
9.3.3. Active Subject Affixes
9.3.4. Two-affix Combinations
9.3.5. Irregular Verbs
9.3.6. Reflexive Verbs
9.3.7. Reciprocal Verbs
9.3.8. Reflexive Possessive, Dative, and Benefactive Verbs
9.3.8.1. Reflexive Pssessive
9.3.8.2. Dative
9.3.8.3. Benefactive
10. Enclitics
11. Selected Vocabulary

Introduction

Lakhota is one dialect of a language widely spoken in the northern plains. It is not easy to find a universally acceptable designation for this language, given the fact that there is no unambiguous native name for it. Nineteenth and early twentieth-century scholars (Stephen R. Riggs, Franz Boas) used the term Dakota both for the language and for its eastern dialect. This is obviously awkward and liable to confusion. Here the designation Sioux is used for the language, reserving Dakota for the dialect. Many speakers of the language dislike the term Sioux because of its foreign origin (cp. Goddard 1984), its use primarily by non-Indians, and because some do not recognize that all the dialects represent the same language.

The Sioux language is the first or second language of about 10,000-12,000 people in the northern plains and contiguous areas of the United States and Canadian prairie provinces. Some speakers of the language are to be found in other places in both countries as well, such as Los Angeles and Toronto. This is one of the largest surviving native language communities in North America.

Lakhota (Teton Sioux) is one of the five closely related dialects. Parks (1990), based on extensive surveys of all the Sioux-speaking reservations and reserves in the late 1970s, identifies these as Santee-Sisseton, Teton, Yankton-Yanktonai, Assiniboine, and Stoney. The easternmost of these is Santee-Sisseton. Nineteenth-century scholars, following native usage, referred to this dialect as Dakota. The westernmost of the dialects, Teton, is designated by its native name, Lakhota or Lakota. Speakers of the Assiniboine and Stoney dialects call their language Nakoda. The remaining dialect, Yankton-Yanktonai, also located geographically between the Santee-Sisseton and Teton dialects, show affinities with both Dakota and Nakoda, although speakers call their language Dakota.

Each of these dialects has reservation- or reserve-based subdialects, some quite different from the others. The subdialects of Teton Sioux oppose the southwest reservations (Pine Ridge and Rosebud) to those on the Missouri River (Cheyenne River, Lower Brule, Standing Rock). The populations on these reservations reflect earlier band divisions among the Teton Sioux, so the present linguistic differences quite likely reflect differences older than the reservation period, which dates only from the last third of the nineteenth century.

The Sioux language, in one or another of its dialects, but chiefly Dakota, has been written for over 150 years by missionaries, anthropologists, educators, and native speakers, using a variety of writing systems, all based on the Roman alphabet. Not surprisingly, there exists a sizeble corpus of Sioux writings (see de Reuse 1987, 1990), some favoring a broad rendering of the language, others a fairly narrow rendering. In most cases there is no indication with or in the document of the intended degree of phonological exactitude, although most are more broad than narrow.

1. Phonology

Table 1 gives the consonant phonemes of Sioux, and table 2, the vowels. The discussion that follows specifies the significant variants of these phonemes for the Teton dialect. Although the transcription is phonemic, it uses and adapts some conventions and diacritics from earlier practical orthographies and uses capitalization and punctuation in accordance with written English norms.

Table 1. Lakhota Consonants
  Labial Dental Palatal Velar Post Velar Glottal
Obstruents
stops voiceless plain p t č =č k   ? =glot.gif (71 bytes)
voiceless aspirated ph th čh =čh kh    
voiceless glottalized p? t? č? =č? k?    
voiced b     g    
fricatives voiceless plain   s š =š   ĥ =ĥ  
voiced   z ž =ž   ğ =ğ  
Resonants
nasals m n   ň =ň    
lateral   l        
Glides w   y     h
 
Table 2. Lakhota Vowels
  Front Central Back
high oral i   u
nasal į =į   ų =ų
mid   e   o
low nasal   a  
oral   ą =ą  

Stress and Pitch
        ' relative loudness, high pitch

Diacritics used in phinetic transcriptions
        ` - less loud than vowels stressed
        ^ - fading loudness, high fall
        v - increasing loudness, high rising pitch

1.0. Consonants

There are either 26 or 28 consonant phonemes in Lakhota, depending on how [b] and [g] are counted (1.1.3.).

1.1. Stops

/p, t, k, ?/ are plain oral stops articulated at labial, dental, velar, and glottal positions. /č/ is a palatal affricate, but it patterns with the stops. The sounds represented by p, t, k can be compared with the English sounds represented by the letters p, t, and c in the words spar, star, and scar. English has no exact equivalent for the sound represented by Lakhota č, but the reader can easily create the sound by inventing an English word such as schar, pronounced to rhyme with spar, star, and scar: ch is here pronounced like Lakhota č. Lakhota words that contain these sounds are:

sápa ‘black’
yúta ‘to eat’
sáka ‘to dry’
hečá ‘buzzard’.

/?/ represents a soundless hiatus betweeen other sounds: the hiatus occurs when  the air  stream is interrupted by tight closure of the glottis. The characteristic "sound" of this consonant is the sharp interruption of the air stream that occurs when the glottis is abruptly closed or an audible pop when the closure is released. An English example is the interjection oh oh! ?o] uttered by English speakers when they are confronted by something unexpected. Lakhota examples are:

a?ú ‘to bring here’
a?í ‘to take there’.

1.1.1. Aspirated Stops

The aspirated stops of Lakhota consist of a stop closure followed by a release accompanied by either glottal or velar friction. When the release has glottal friction, the sound is very much like the English aspirated stops found before stressed vowels as in the following English words: pick, tick, chick, and kick. Release with glottal friction occurs in Lakhota with čh and with the other aspirated stops before the vowels /i u į/. Glottal friction is also sometimes heard before the vowel /ų/. Lakhota words in which the aspirated stops have glottal friction are the following:

phí ‘liver’
thí ‘to live’
čha ‘so’
khí ‘to reach home there’
phú ‘rotten (wood)’
nuphį' ‘both’
phuté ‘upper lip’.

Before other vowels except /e/ the aspiration is velar in release, rather thatn glottal. There are no English equivalents to the sound of the aspirated stops of this kind; they can be compared with the sound of the Navajo aspirated stops if these are know to the reader. Examples of velar release are the following:

[pxá] ‘head’
[txápa] ‘ball’
[kxáta] ‘hot’
[thó] ‘blue, green’
[kxá] ‘to signify’
[txezí] ‘stomach’
[othų'wahe]  ‘village’.

The distribution of the two kinds of release described above is not absolute: both kinds of release occur before the vowel /e/, and before this vowel, the distinction is phonemic in particular ideolects. It should be noted that all speakers agree on the nature of aspiration except before the vowel /e/ and that in these cases there is no variation: a speaker pronounces a given word with either one or the other kind of release quite consistently. The distinction is thus more lexical than phonological. Examples of this kind are:

[pheží, pxeží] ‘grass’
[phehį', pxehį'] ‘hair of the head’.

Some speakers pronounce these words with glottal friction, others with velar friction, as indicated. Because the disctinction between [h] and [x] is so nearly predictable and because speakers disagree about those places where it is not predictable, all aspirated stops are presented here as consonant plu [h]: phá ‘head’, thápa ‘ball’, kháta ‘hot’, etc.

1.1.2. Glottalized Stops

The glottalized stops of Lakota are ejectives; these sounds are not found in English or in any Western European languages. Ejective sounds are formed by the near simultaneous release of two closures, one in the mouth at the position of the stop, the other in the larynx at the glottis. Some compression of the air in the mouth occurs due to the double closure, and it is the release of this compressed air that gives the characteristic ‘crack’ when the ejective is released. Lakhota words containing ejective stops are:

p?ó ‘fog’
t?á ‘dead’
?į' ‘harness’
k?ú ‘to give’.

1.1.3. Voiced Stops

The phonemes [b] and [g] represent the same sounds as in English bet and get. They have a very restricted distribution in Lakhota, occurring only before sonorant consonants (l m n), the voiced glides (w y), and in various kinds of vowel-dropping situations (Section 4.3.1.2.) and reduplication (4.3.2.8.). Since /p/ and /k/ never occur in these positions, /b/ and /g/ are actually positional variants of the plain voiceless stops. There is neverthelessan extremely small number of words where the /b/ is not predictable, so it seems best to consider that sound marginally phonemic. Moreover, there is a long tradition of writing both b and g when they occur, despite there theoretical status, and that tradition is followed here.

When followed by a sonorant, the stop closure is released before the articulation of the sonorant is begun, giving a voiced, vowellike transition to the sonorant.

The first two of these examples sound very much like the English words below and (the first part of) galore:

bló ‘potato’
gló ‘to grunt’
tób ‘four’
sabsápa ‘severally black’
sabyá ‘to blacken’
pathág ‘stopping, halting abruptly’
sagsáka ‘dry’
nagwáka ‘to kick out the foot’.

1.2. Fricatives

Lakhota has a more extensive fricative system than does English, so some of the Lakhota fricatives have no English counterpart.

1.2.1. Sibilants

Lakhota /s z š ž/ are postdental and palatal in articulation: these sounds can be compared with the highly similar English sounds found in the words seal, zeal, rasher, and azure. Lakhota words that contain these sounds are:

Hásapa ‘Black person’
máza ‘metal’
maší ‘he ordered me’
pheží ‘grass’.

1.2.2. Velar Fricatives

The Lakhota velar fricatives /ĥ ğ/ [x ğ ] have no English equivalents, but they can be compared with sounds found in close relatives of English such as German and Spanish. /ĥ/ represents the final sound in the German word Bach or the initial sound Spanish jota. Lakhota examples of this phoneme are:

ĥé ‘mountain’
iĥá ‘to laugh’
?é ‘rock’.

The phoneme /ğ/ has two positional variants, one an uvular tap, the other, a midvelar voiced fricative. The former sound occurs in Lakhota before the vowel /i/; an example is the word

ğí ‘yellow, brown’.

Compare this to the sound represented by r in French word Henri when this name is pronounced very quickly. Examples of the other variant of /ğ/ are the Lakhota words

ğú ‘burned’
káğa ‘to make’.

This sound is identical to Spanish g between vowels as in the word pagar   ‘to pay’.

A lengthened form of the voiced fricatives is found when these are initial before a stressed vowel. Examples of this lengthened fricative sound are:

‘foot’
šá ‘red’
ĥóta ‘gray’.

1.3. Sonorants

Lakhota /m n ň/ are respectively labial, dental, and velar in articulation. All have ready English equivalents: him, sin, and sing. A Lakhota example containing all three sounds is:

šųňmánitu ‘coyote’.

For most speakers, Lakhota /l/ has a "clear" rather than a "dark" timbre. It is most like the "clear" l of Spanish or Italian ala ‘wing’ or French elle ‘she’. It never sounds like English "dark" l after a vowel, e.g. ball, well. However, there are reports that some speakers have the same distribution of "clear" and "dark" l in Lakhota that they do in English. Lakhota examples of /l/ are:

lél ‘here’
thaló ‘meat’
Lakhóta ‘Indian’.

1.4. Glides

The glides /w y h/ have more conditioned variants than any of the other consonant phonemes of Lakhota. This is probably due to the fact that their status as semivowels makes them highly sensitive to the vocalic environments in which thev occur. /w y/ are voiced: /h/ is voiceless.

1.4.1. /w y h/ at the Beginning of a Words

/w y h/ are most clearly pronounced at the beginning of a word. Lakhota examples are:

‘snow’
‘to go’
‘skin’.

Compare these with equivalent English sounds in the same position, as in the words wet, yet, and hot.

1.4.2. Glides between Vowels

When the glides come between vowels they are pronounced weakly, if at all. This is particularly the case with /w/ and /y/. Examples are the Lakhota words:

wayáwa ‘he reads’
iyápi ‘they say’
othų'wahe ‘town’
yuwášte ‘to make good’
hąhépi ‘night’
čhąhą'pi ‘sugar’.

1.4.3. Before and between Nasalized Vowels

Before and between nasalized vowels all three Lakhota glides are pronounced with heavy nasalization. English equivalents for the nasalized glides are rare, although the sounds are easy to produce if the velic is open and air passes out of the nasal cavity during their articulation. Examples of nasalized /h/ do occur in English in the informal affirmative and negative particles uh-huh [NhN'] ‘yes’ and uh-uh [(h)N'?N] ‘no’. Lakhota examples are:

wį'yą ‘woman’
chąhą'pi ‘sugar’.

2. Vowels

The vowel system of Lakhota has five oral vowels and three nasal vowels. They are typically rather short in duration. However, lengthened versions of all the vowels may occur lengthened due to contraction of identical vowels, and two additional long vowels having no short analogues result from the contraction of unlike vowels. Vowel contraction is described in sections 2.3. and 4.3.

2.1. Oral Vowels

The five oral vowels are very comparable in their typical value to ihe five "cardinal" vowels, for example, as these are realized in Spanish. All are "pure"; that is, there is no shift in articulators or articulatory position during their articulation. Pure cardinal vowels are difficult for English speakers to produce because precisely these vowels begin in English with a "pure" vowel and end with a glide pronounced in the same general area. This English glide is usually written with the midvowels /e/ and /o/ but not otherwise. Compare the following:

Lakhota English
‘foot’ see (si+y-glide)
‘seed’ sue (su w-glide)
blé ‘lake’ play (ple +y-glide)
bló ‘potato’ below (belo +w-glide).

The vowels /i/ and /u/ are front unrounded and back rounded, respectively. Compare their sounds to the same vowels in Spanish or Italian. Lakhota examples of these vowels are:

‘tooth’
‘leg’.

The vowels /e/ and /o/ are lower-mid; /e/ is front unrounded; /o/ is back rounded. There are no good phonetic analogues for these Lakhota vowels in English or other Western European languages. The sounds of these vowels are somewhat more open than the accepted pronunciation of the cardinal vowels e and o. Lakhota examples of these vowels are:

‘horn’
‘voice’.

The vowel /a/ is low and central, pronounced in Lakhota with'it .cardinal value: that is, it is pronounced with close to maximum opening. An adequate English analogue is the vowel in the first syllable of the word father. A Lakhota example is:

‘skin’.

All Lakhota oral vowels are partially devoiced when they are stressed and in utterance-final position. This devoicing makes it sound like a final h follows the vowel. This is especially noticeable when words are pronounced very carefully, as in citation.

2.2. Nasalized Vowels

Nasalized counterparts exist for the two highest and the single lowest vowel: /į, ų, ą/. The nasalized vowels are phonetically lax. The sounds of these vowels are comparable to, but not identical with, certain English vowels followed in the same syllable by a nasal consonant. The sound of /į/ and /ą/ may be compared with the vowels in sin and nun. (Note that the comparison is to the sound of the vowel only, without the following nasal consonant.) American English has no equivalent for /ų/. Here are Lakhota examples of the nasalized vowels:

hį' ‘hair’
hą' ‘yes’
hų'ku ‘his mother’.

While most Lakhota speakers agree as to which vowels are pronounced with nasalization, there are some vowels that are nasalized by some speakers, but not by others. Such cases no doubt represent doublets: both pronunciations are correct. (Compare this with a word such as which in English, where some speakers preaspirate and-devoice the /w/. while others pronounce voiced /w/ alone.) Lakhota examples of such doublets are

ki, kį ‘the’
náži, nážį ‘to stand’.

2.2.1. Nasalized vowels after Nasal Consonants

The vowels /i, u, a/ are always pronounced with some nasalization when they follow a nasal consonant. However, some speakers have a phonemic contrast between nasalized and nonnasalized vowels following nasal consonants. For these speakers, strong nasalization indicates that the vowel is phonemically nasalized while weak nasalization indicates that the vowel is phonemically oral.

Speakers who have phonemic contrast after nasal consonants probably continue an earlier pattern in the language whereby there was full phonemic contrast in oral and nasal vowels after nasal consonants. That this is not an idiosyncratic feature of some persons' speech is shown by their agreement with speakers of other Sioux dialects such as Nakoda, where full contrast is found after nasal consonants. Some examples of contrasting nasality after nasalized consonants are:

mąká ‘I sit’ versus maká ‘skunk’
gmų'za ‘slimy’ versus gmúza ‘closed, as the fist’
nįyą' ‘cause to live’ versus niyá ‘to breathe’

At an abstract level, Patterson (1991) argues for three kinds of vowel nasalization in Lakhota: some vowels are [+nasal]. some are [-nasal], and some are unmarked for this feature.

This chapter represents speech in which vowel nasalization is neutralized after nasal consonants, which appears to be the usage of the majority of Lakhota speakers. In this environment only oral vowels are written.

2.3. Diphthongs and Contracted Vowels

There is only one diphthong in Lakhota. the sequence /au/. This occurs in a single Lakhota word, one of .whose functions is as a greeting to a man:

Hau! ‘greeting to an adult male’

The pronunciation of this word is identical to that of the English word how. This word may be a loan from a non-Siouan language.

While the case of diphthongs is quite simple in Lakhota, that of contracted vowels is not. Contracted vowels result from the conflation of syllables through the collapse of a syllable boundary or from the vocalization of consonantal elements followed by conflation with a preceding vowel. Contraction happens most often in rapid, colloquial speech, although there are a few examples where the contraction has become the standard form. A prominent example of this is the word [a:'ta:] ‘entirely’, which must have an underlying form with normal short vowels. However, contemporary speakers cannot supply an underlying form for this word, which is unusual. Long contracted vowels of this sort are phonemic: and written double: /áataa/.

In almost all cases, the collapse of a syllable or word boundary results from the disappearance of a glide between vowels (section 1.4.2.). After the loss of the glide, the vowels in hiatus contract. The contracted vowel is nasalized if either of the uncontracted vowels was nasalized.

When the original vowels were of the same height, or if assimilation occurs before contraction, a simple long vowel results. Here are examples of this kind of contraction, with pitch marked as explained in section 5.2.:

[hav:pi] ‘clothing’, < hayápi

[mi^:hakab] ‘immediately after me’ < mí ihakab

[ke^:] ‘he said that’ < kéye

[u^:kte] ‘they will come’ < ú pi kte

[o^:'na] ‘they wounded him and’ < ó pi na

[čhNv:pi] ‘sugar’ < čhąhą'pi (‘tree juice’).

When the uncontracted vowels were of different heights or when feature contrasts exist between the vowels and the glide, the contracted vowel is qualitatively different from the uncontracted sequence. Here are Lakhota examples of the two vowels that result from this kind of contraction:

[iyae^:] < iyáye ‘he left for there’

[wakhae^:ža] < wakhą'yeža or wakhą'heža ‘child’

[mith^:] < mitháwa ‘it is mine’

[ųyN^:kte]< ųyą' pi kte ‘we will be going’

3. Stress and Pitch

Vowels in all languages are pronounced with some kind of accompanying melody (loudness, pitch). The word is the domain of stress in Lakhota. In Lakhota words the first (or only) stressed vowel has higher pitch and greater loudness than all other vowels in that word. Most of the time the stressed vowel is the second one in the word, but this is not always so. It is therefore necessary to write stress on every word. Compare the following Lakhota words from the same verbal paradigm:

iyáye ‘he set out to go there’
éyaye ‘they set out as a group to go there’.

A rare example with stress on other than the first or second syllable is tuktená "which ones’".

3.1. Secondary Stress

There is usually only one stressed vowel in each word. Exceptions to this are compound words, which usually retain the stressed vowels of the originally separate words. In a compound word, therefore, there can be two or more stressed vowels: in such cases, the first stressed vowel has higher pitch and greater loudness than any subsequent stressed vowel. These two stress levels are referred to as primary and secondary stress, and when it is desirable to distinguish them the acute accent (') can be used for primary stress and the grave accent (`) for secondary stress. Examples are:

mázaskà ‘money’ (lit.‘while metal, silver’)
šų'kawakhą` ‘horse’ (lit. ‘mystery dog’).

Ordinarily this distinction is not marked, and the acute accent is used for both types of stress.

3.2. Pitch on Long Vowels

Lakhota long vowels, which always result from vowel contraction, may be pronounced with rising or falling pitch, depending on the stress pattern of the original, uncontracted vowels. If the first uncontracted vowel had stress, the contracted sequence has higher pitch and greater loudness at the beginning of the contracted sequence, with a fall in pitch and loudness as the sequence continues. If the uncontracted sequence had stress on the second vowel, then the contracted sequence shows a rise in pitch and loudness toward the end of the sequence. Compare the following examples, where a circumflex (^) marks a contracted vowel with the high point toward the beginning, and a wedge (v) marks a contracted vowel with the high point toward the end of the sequence:

[u^:kte] ‘they will be coming’

[Ov:phe] ‘I hit him’.

In phonemic writing, long vowels are written as geminates, with the appropriate one stressed. For example, [i^:] is /íi/ and [uv:] is /uú/.

4. The Phonemic Structure of Words

Lakhota phonemes are combined into words according to very regular rules, as is true of all languages. Some consonant and vowel sequences are exceedingly frequent, others are rare, and some theoretically possible combinations never occur.

4.1. Segmental Sequences

Lakhota words in their basic form almost always end in a vowel. Of the consonants, only /l/ is at all frequent as the final sound in a word. Also occasionally found in word-final position are /b/, /g/, /m/, /n/, /s/, /š/, and /ĥ/. Of these, /n/ is most frequent.

Lakhota words begin, as a rule, with one or two consonants. Words written with an initial vowel are usually pronounced with a preceding [?], although this sound may be omitted. Any vowel can follow any consonant except that nasalized vowels never follow /l/. In some dialects an /l/ following a nasalized vowel in the same syllable is replaced by /n/: hehą'l in these dialects is pronounced hehą'n. There are a number of restrictions on the makeup of consonant clusters; for example, there are no clusters with one fricative directly adjacent to another.

Within words, vowels and consonants almost always alternate, with each vowel separated by one, two, or (very rarely) three consonants.

There are nevertheless a few words in the language where vocalic prefixes are added to vowel-initial stems with no intervening consonant or consonantal element. Examples are:

naíč?ikeğa  ‘to scratch oneself with the foot’
míoglas?į ‘mirror’
kaížu ‘to pay off one’s debts’
paóskiča ‘to cram things into something’,
(e.g., too many clothes into a washing machinge)
paótkuğa ‘push, close, and lock with a motion,
as a door with an automatic lock’
yuótkuğa ‘pull, close, and lock’ (as above)

A similar phenomenon is more difficult to deal with, and speakers disagree about some of these examples. Vowels that are the same and that have no intervening consonant, especially /ii/, are frequently pronounced as a series of rearticulated vowels. Where one of the clustered vowels has stress, there is a rising (or falling) pitch on the phonetically lengthened vowel. Examples are:

lową' ‘to sing’
ílową ‘to sing about something’
íič?ilową ‘to sing about oneself’
íiič?ilową ‘to sing about one’s impending death’
sabmíč?iye ‘I blackened myself, made myself black’
sabmíič?iye ‘I blackened myself because of some obligations or for some ceremonial purpose’.

Usually native speakers do not pronounce any of these vowel sequences with intervening glides or [?], and they reject the insertion of such elements. It is unclear to what extent there may have been glide neutralization in such cases (see 4.3.1.3.) or whether a phonemic element was ever present between these vowels.

Refer to the sample vocabulary for examples of consonant and vowel sequences.

4.2. Syllabification

The number of syllables per word is determined by the number of vowels present in the word: there is one syllable for each vowel. Taylor and Rood disagree about the accuracy of this statement for the identical adjacent vowels described just above: Taylor hears a simple long vowel in these contexts, while Rood believes that the vowels are separately rearticulated and thus constitute separate syllables. It is likely that speakers differ in the phonetic realization of these sequences.

The only other exception to this rule is the word hau, which includes the only diphthong known in the language. This word is a monosyllable.

Most syllables begin with one or two consonants and end, wherever possible, with a vowel:

[wa-yá-wa] ‘he reads’
[la-kxó-ta] ‘Indian’
[spa-yó-la] ‘Mexican’
[o-í-yo-ki-phi] ‘be happy’
[lél] ‘here’
[nų'm] ‘two’

The syllable always ends in a vowel or a single consonant, but when two or more consonants come between vowels, it is not always easy to know whether the syllable boundary will come before all or between the first two.

If the consonants belong to the same morpheme (meaningful sequence of sounds), the syllable break comes before the consonants:

[šičá-mna] ‘it stinks’ (cf. mná ‘to stink’)
[yu-ptą'-ptą] ‘to rock back and forth with the hand’
(cf. -ptą- ‘unsteady, rocking’).

But if the two consonants belong to different morphemes, the syllable boundary comes between the two:

[nų'm-nų-pa] ‘two by two’ (cf. núpa ‘two’)
[tób-to-pa] ‘four by four’(cf. tópa ‘four’).

Except before /m, n, l/, /b/ and /g/ occur only in morpheme-final position. Hence there will always be a syllable boundary after these sounds if any other consonant follows.

When three consonants are found between vowels within utterances, the consonant cluster is always divided by the syllable boundary, since such sequences are found only in compound words:
 

[šųň-bló-ka] ‘male horse or dog’.

4.3. Phonological Changes in the Basic Form of Words

Pronunciation of isolated words is often different from the pronunciation of the same words in phonological contexts of larger size such as word compounds, phrases, or sentences. In some cases also the change of the grammatical form of a word can cause phonological changes in the word itself. In other words, the form of a word, or the forms of related words, are often affected by the presence of other linguistic elements.

4.3.1. Phonological Changes that Result from Speaking Style

In all languages, rapid, colloquial speech often differs markedly from slow. careful speech. As a rule, rapid speech is a reduced form of slow speech. English examples of this are spose for ‘suppose’, gotcha for ‘I’ve got you’, and gonna for ‘going to’.

There are many changes of this kind in colloquial Lakhota, and Lakhota speakers are themselves aware of the difference. Precise (unchanged) speech is called yat?į'sya wóglakapi ‘firm or clear speech’, while rapid, slurred speech is called ikčéya-wóglakapi ‘ordinary or normal speech’.

4.3.1.1. pi in Rapid Speech

One of the most striking differences between precise and rapid speech is the replacement of the enclitic (suffixlike word: see 10.0.) pi in rapid speech by a vowel before the enclitics kte, , kštó, na, and possibly others. The vowel that replaces pi is determined by the height of the vowel that immediately precedes pi. If the vowel is high (/i/, /į/, /u/, /ų/), pi is replaced by /u/. If the vowel is mid or low (/e/, /o/, /a/, /ą/), pi is replaced by /o/. The replacing vowel is nasalized if the preceding vowel is nasalized. Here are some examples of this change:

Slow speech Fast speech Meaning
Hí pi kte. [hí u kte] ‘They will arrive here.’
Ú pi kte. [u^: kte] ‘They will come.’
Čhį' pi kte. [čhį' ų kte] ‘They will want.’
Olé pi na [olé o na] ‘They looked for him and...’
Yatką' pi na [yatką' oN na] ‘They drank it and...’
Oyúspa pi kštó. [oyúspa o kštó] ‘They caught him.’

4.3.1.2. Vowel apocope

Also characteristic of rapid speech is the dropping of unstressed, word-final vowels. For example, in the enclitic pi /i/ is frequently lost when other enclitics follow. If a nasalized vowel precedes, and a fricative follows, /p/ is then changed to /m/:

Slow speech Fast speech Meaning
Awíčhayuštą pi šni. [awíčhayuštą m šni] ‘They aren't leaving them alone.’
Awíčhayuštą pi he? [awíčhayuštą b he] ‘Are they leaving them alone’?

/į/ in the article (8.3.1.) is also frequently dropped:

Slow speech Fast speech Meaning
hokšíla kį lé [hokšíla g lé] ‘this boy’

In these examples, note that p and k are voiced to b and g when they come to stand before a consonant. A similar process is described in 4.3.2.7. below.

Dropping of word-final vowels is particularly frequent when the following word begins with a vowel; note that such newly word-final /p/ and /t/ do not become /b/ or /l/. (The loss of /w/ is described in the next section.)

Slow speech Fast speech Meaning
Ağúyapi etą' ophéthų wo! [ağúap et óphethų o] ‘Buy some bread.’

4.3.1.3. Dropping of Glides and /?/

Another frequently encountered phenomenon is the dropping of glides and /?/ when these phonemes are located between vowels. The vowels left without a separating consonant are sometimes then contracted into a single long vowel having some of the features of both original vowels (2.3.).

/w/ and /y/ are weakly pronounced, or dropped, when one of the neighboring vowels shares positional features with the glide: /u/, /ų/, or /o/ with /w/; /i/, /į/, or /e/ with /y/. An example of loss of /w/ is seen in the fast speech form given above for the (enclitic wo), pronounced [o]. An example of loss of /y/ is the word áye ‘he took it’, phonetically [ae].

Glides are also regularly dropped when the vowels on either side are /a/ or /ą/.

When glides are dropped from the sequences /aya/, /aye/, and /awa/ (/ą/ could appear instead of /a/ in any of these), the vowels usually further contract into a single long vowel. These resulting long vowels are [a:], [ae:], and [o:], respectively, or if /a/ is present, [:], [aeN:], and [N:].

The same vowels result from the dropping of /h/:

othų'wahe > [othų'wae:] ‘town’

hąhépi > [haeNv:pi] ‘night’

chą-hą'pi > [čhNv:pi] ‘sugar’

Situations also arise where a morpheme that ends in a vowel comes before one that begins with /?/. Very often, both the vowel and the /?/ are then dropped, and native speakers prefer the fast speech form as the citation form. When this rule will apply and when the underlying /V?/ sequence will be retained is not presently predictable. Examples of the loss are:

|ečhá?ų| > ečhų' ‘he did it’

|tha?óyąke| > thóyąke ‘his agency’

|wa?íyatke| > wíyatke ‘cup’

|ištá?othó| > ištótho` ‘he has a black eye’.

4.3.2. Phonological Changes that Result from Grammatical Processes

Since a large part of the grammar of Lakhota concerns verbs, a number of phonological changes are restricted to, or at least most noticeable in, the verbal processes of the language. Some of them are productive (apply in all words); some are unproductive (apply in only some specific words).

4.3.2.1. Stress shift

A frequently observed sound change is the shift of stress from one syllable to another. When prefixes are added to the basic form of a word, the stress is moved as far forward as is necessary to prevent it from standing on a syllable later than the second syllable of the new construction. Compare:

yuhá ‘he has’ ųyúha pi ‘we have’
wašté ‘to be good’ wayúwašte ‘to make things good’.

4.3.2.2. Insertion of /?/

Another very simple rule whose effect is most often seen in verb conjugation is the insertion of /?/ whenever grammatical processes place an element beginning with a vowel immediately after a boundary between elements within the word. Examples are:

a?í ‘they arrived there’
(í ‘to reach a place away from here’; a ‘collective plural’)
iyé?ųyą pi ‘we found him’
(iyéyA ‘to find’; ų ‘we’)
slol?ų'yą pi ‘we know him’
(slolyÁ ‘to know’; ų ‘we’)
theb?ų'yą pi ‘we ate it up’
(thebyaÁ ‘to eat up; ų ‘we’).

This inserted /?/ can of course be dropped if it follows a vowel (4.3.1.3.).

4.3.2.3. Nasalization Spread

Still a third phonological change associated with verb conjugation is a phenomenon that can be called nasalization spread. When a nasalized and an oral vowel are separated by a glide it is quite usual for both of the vowels and the glide to be pronounced with nasalization; this does not happen if the spread is from left to right across the /y/ of a prefix (cf. Patterson 1991):

iyé?ųyą pi ‘we found it’,
    compare iyéya pi ‘they found it’

wąyą'ke ‘he saw it’,
    compare wąbla'ke ‘I saw it’

wąwą'yąke ‘he saw something’,
    compare wa?ónaĥ?ų ‘he heard something’

ómakįyį kte ‘it will help me’,
    compare ómakiye ‘it helped me’.

Compare these examples where there is no spread; here -ya- is a prefix:

ųyáĥtaka pi ‘we bit him’

ųyákağa pi ‘you made us’.

There seems to be some complex ordering between the nasal spread rule and the insertion of inflectional affixes, since the secondary nasalization of -yą- in ‘?o see’ also occurs in wąyéchiyąka pi ‘you saw each other’, but not in the simple wąláke ‘you saw him’.

4.3.2.4. Change of /k/ to /č/

When grammatical processes place /i/ (sometimes /e/) before k (whether plain, glottalized, or aspirated) plus a vowel (khV, k?V, or kV), the k frequently becomes /č/, but not always. Compare mak?ú ‘he gave it to me’, čhičú ‘I gave it to you’, kú ‘he is coming home’, gličú ‘he reached home here’, ómakiye ‘he helped me’, óničiya he? ‘did he help you?’ khiya ‘over yonder’, and čhiya ‘over there’.

Exceptions to this rule are stative verbs, such as okháyakA ‘to have things (such as leaves, burrs) stuck on’ (oníkhayake ‘you have things stuck on you’); adverbs; dependent verbs (verbs that require another verb in the same sentence) such as kapį' ‘be reluctant to’; and a few exceptional transitive verbs (cf. Boas and Deloria 1941:14). When a derivational prefix is added to a root and the resulting verb is stative (kat?át?A ‘to fall down by accident’, from t?Á ‘be dead’), the k does not change (nikát?at?a ‘you fell down by accident’). But if the new verb is transitive, k changes to č providing that the pronominal prefix precedes the derivational one (kat?át?a can also mean ‘to knock someone out’, and ‘she knocked you out’ * is ničát?at?a). If the pronoun precedes the root, k still does not change, even if the verb is transitive. Note nikį'za ‘you squeaked’ and naníkįza ‘he made you squeak by stepping on you’.

*In this sketch. Lakhota third-person singular pronouns referring to people are translated at random as either ‘he’ (‘him’, ‘his’) or ‘she’ (‘her’) when there is no determining context; either translation is correct.

4.3.2.5. The Affix ma-

The affix (prefix or infix; see 9.3.1.) ma- ‘I, me’ loses its vowel when it is added to a stem that begins with /i/. Compare mawášte ‘I am good’, mak?ú ‘he gave it to me’ with mitháwa ‘it is mine’ miglúkse ‘I cut myself’, mihákab ‘behind me.’ Some verbal .stems that have initial /i/ drop the /i/ when no affix precedes. ‘It is his’, for example, is tháwa. (Note that the position of stress on the old second syllable is evidence for the original presence of the initial /i/.)

4.3.2.6. Final Vowels in A-words

The final vowel of a large group of Lakhota verbs is subject to regular changes, depending on what follows the verb. A small number of enclitics also share this feature. For convenience of reference words with these vowel alternations are called "A-words," and such words are cited with a written final A or Ą. The form of words with these alternations is correct in each instance with a particular one of the alternating vowels. The vowels that alternate are /a/ or /ą/, /e/, and /į/ (for some speakers also /i/). When no element follows, /e/ is always found. When the enclitics ktA and na follow, the vowel is almost always /į/, but some speakers use /i/ before na, at least sometimes. Otherwise, either /a/ (/ą/) or /e/ is found, with each context calling for one or the other. Examples are yatké ‘he drank it’ and t?é ‘he is dead’; yatkį' kte ‘he will drink it’ and t?į' kte ‘he will die’; yatką' he ‘did he drink it?’ and t?a he ‘is he dead?’

Speakers do not always agree on which verbs show this kind of alternation, particularly when the verb in question is somewhat rare. For example, the verb olúluta ‘be sweltering hot’ is treated as nonalternating by some speakers but as alternating by others. The same thing is true for speakers of the other Dakota dialects. It appears that any verb that ends in -a in its basic form may be regarded by some as an alternating verb. Because of this, verb-final vowel alternation will probably become more widespread in the future and may possibly result eventually in a state where all historically -a verbs will become -A verbs.

The terminal vowel is lost entirely when verbs of this kind are reduplicated (4.3.2.8.) or incorporated as any but final member into a word compound. The consonant immediately before the dropped vowel may also change (4.3.2.7.). Examples are sabsápA ‘black’ (sapA ‘to be black’). yulphíčA ‘edible’ (yútA ‘to eat’), čhebyÁ ‘to fatten’ (čhépA ‘to be fat’), and kaĥkhíyA ‘'to cause someone to make something’ (kágA ‘to make’).

A number of nouns also lose their final vowel in the same kinds of constructions (though reduplicated nouns are rare): nabkózA ‘to beckon’ {napé ‘hand’), čheĥ?íkhą ‘bucket handle’ (čhéğa ‘kettle’), čhašthų' ‘to make a name for oneself’ (čhažé ‘name’).

Another kind of vowel loss is frequently seen in compounds also. When one member ending in a vowel stands before another that begins with a vowel, the first of the two vowels is ordinarily dropped. Examples are exceedingly numerous: nab?ágle ‘to lay hands on’ (napé ‘hand’, aglé ‘to place on’, itípakhįte ‘face towel’ (ite ‘face’, ipákhįta ‘to wipe with’; note that /t/ is not replaced by /l/ - 4.3.2.7.), makhóĥloka ‘cave’ (makhá  ‘earth’, oĥlóka ‘hole’). kéyA ‘to say that’ ( ‘that’; eyÁ ‘to say’), and wiglóčhethi ‘gas stove’ (wígli ‘oil’, očhéthi ‘stove’).

4.3.2.7. Change of /p t č k/ to /b l l g/

When vowel dropping (of any origin except possibly the fast speech phenomena illustrated in section 4.3.1.2.) places /p t c k/ in word-final position or at an internal boundary between linguistic elements, these become [b], [l], [l], [g], respectively. When a nasalized vowel precedes these sounds, they may further shift to a nasal consonant: [m], [n], [n], or [ň], respectively. Note that these shifts do not occur before vowels unless [?] is inserted to mark the boundary. Compare the examples in the preceding paragraph of itípakhįte versus nab?ágle. Examples of such consonant changes have already been seen. Further examples are b ‘four’ (shortened from tópa) and m ‘two’ (shortened from núpa); khal ‘to heat’ (cf. khátA ‘to be hot’) and čhąňmášiče ‘I am sad’ (shortened from čhąté-mašíče); psípsil ‘skipping’ (shortened from psípsičA); pathág ‘stopping short’ (shortened from patháka) and šųňwį'yela ‘mare, bitch’ (cf. šų'ka ‘dog’).

When the voiced fricatives /z ž g/ come to stand at a boundary they are replaced by /s š ĥ/ respectively. Examples are skoza ‘waving’ (cf. kózA ‘to wave’), khuš ‘to make nauseous’ (cf. khúžA ‘to be nauseous’), and piĥ ‘to boil’ (cf. píğA ‘to be boiling’).

4.3.2.8. Reduplication

One of the most productive grammatical processes in Lakhota is reduplication, the repetition of a portion of a word. The repetition creates a new word whose basic meaning is similar to the unreduplicated form but whose grammatical meaning is different. The meaning of reduplication is variously plurality, repetition, distribution through space ("here and there"), or intensity. Although most words have just one correctly reduplicated form, the part to be repeated can be anywhere in the word. An example of full reduplication is zizí ‘yellow’ (cf. ‘yellow’); an example with initial reduplication is ĥolĥóta ‘gray’ (cf. ĥóta ‘gray’); with final reduplication wašté ‘good’ (inanimate plural; cf. wašté ‘good’); with medial reduplication, napčį'yųňyųka ‘nine by nine’ (cf. napčį'yųka ‘nine’). Note that consonant changes of the kind mentioned in 4.3.2.7. are very frequent in reduplication.

The part of a given word that is reduplicated can generally be predicted if enough is known about the etymology of the word: it is usually the last full syllable of the root. But this is nevertheless one of the more difficult parts of the grammar of Lakhota, even for native speakers. Refer to Carter (1974), Shaw (1980), and Patterson (1990:89-99) for details; Patterson reviews several other theoretical studies of Lakhota reduplication.


Grammar

Lakhota sentences can be described as consisting of a series of optional and obligatory slots, each slot filled by a particular type of word or phrase. The maximum structure is

(interjection) (conjunction) (adverb(s))
(nominal) (nominal)(nominal)
(adverb(s)) verb (enclitic(s)) (conjunction)

(Parentheses imply optionality; (s) means there is no theoretical limit to the number of like elements that can occur in this position.)

Note that the only obligatory slot is that of the verb; every other position is optional.

Discussion of Lakhota grammar from the point of view of formal linguistic theory can be found in Van Valin (1985, 1987) and references there. Van Valin argues that a careful and accurate account of even the simplest Lakhota sentences requires major revisions in the Chomskyan models that dominated linguistic theory from the 1960s through the 1980s.

Discussion of Lakhota grammar will be organized around the sentence slots enumerated above; each 7 slot, and its possible fillers, will be discussed in turn. Given first however, is a brief definition of the terms:

interjection: exclamation expressing surprise, hesitation, disgust, etc.

conjunction: connector, such as ‘and,’ ‘but,’ ‘however,’ ‘furthermore.’

adverb: expression of time, place, manner (including instrument), or cause.

postposition: a word that relates a nominal to a verb; compare English prepositions. In Lakhota the relating word follows the element it governs.

nominal: a naming word or phrase: noun, pronoun, modified noun, or another element used as a noun.

verb: core word, predicator, word that says something about a nominal.

enclitic: almost a suffix, but actually a separate word; expresses tense, mood. aspect, and other similar grammatical notions.

The first two categories (interjections and conjunctions) are functionally distinct, but it is sometimes difficult to decide whether a particular "sentence launching word" belongs to one category rather than the other. Some of the words cited as examples in 5.0. or 6.0. might therefore be switched to the other category upon further investigation. However, it is certain that all words in either category will precede either adverbs or nominals.

5. Interjections

Some examples of interjections include:

Ma líla olúluta.Gee, it’s hot (and humid).’ (spoken by a woman)

lé ąpétu kį líla kháte!Boy but it’s hot (and dry) today!’ (spoken by a man)

Į'ska / Eyá ‘Well...’; often used as a pause filler.

E'yaš ‘Well, but’

In the case of į'ska, the first vowel may be lengthened quite extensively for added effect:

į'įįįįįįska ‘Weeeeeell’

There are no interjections equivalent to English profane or curse words.

See Buechel (1939: 126-127 and 266-267) for a list of interjections in use in the early twentieth century.

6. Conjunctions

Conjunctions connect two sentences, often showing something about the relationship between the sentences as well. They occur in two possible positions: in the second slot from the beginning or in the last slot in the sentence. The more common position is last in the sentence.

Several different Lakhota words translate the English word ‘and’; but these words do not all mean the same thing in Lakhota. Examples are na or nahą' ‘and also’; čha, chąkhé ‘and so’, yųkhą' ‘and then (rather unexpectedly)’. Other conjunctions include éyaš ‘but’, na?į'š ‘or’, and ho (or honá) ‘furthermore’.

Tákuni tokhų'k?ų pi šni éyaš ųkóyuspa pi.
nothing we.did pl. not but they.arrest.us pl.
‘We didn’t do anything, but they arrested us.’

Of course, conjunctions such as na or na?į'š can join two of the parts of a sentence, such as nominals or verbs. In this case, they occur in the sentence position appropriate to the major part:

Čík?ala na?į'š thą'ka čha yačhį' he?
little or big a you.want question
‘Do you want a little one or a big one?’

Some words that are conjunctions in English do not occur as such in Lakhota. The conjunction pair ‘if...then’, for example, is expressed in hypothetical sentences by simply nominalizing (see 8.5.) the if sentence and adding the then sentence:

Máza-ska ųyúha pi , iyéčhįkįyąke wąží ophé?ųthų pi kte.
money we.have pl. the car a we.buy pl. Pot.
‘If we had money, (then) we would buy a car.’

7. Adverbs

Adverbs may occur either before or after the nominals of the sentence. In theory, there is no limit to their number, nor is there any preferred sequence or position for the various types: any number of adverbs may occur in any order in either place in the sentence.

Formally, adverbial expressions are of three possible types: words, phrases, or sentences.

7.1. Single-word Adverbs

Examples of single word adverbs:

ĥeyáta ‘out in the country’
hį'hąni ‘this (past) morning’
ĥtálehą ‘yesterday’
áataa ‘all; completely’
iníla ‘quietly’
?ą'khoya ‘quickly, without wasting time’
éna ‘right there’
tąyą' ‘well’.

Deictic adverbs are formed by adding a demonstrative (8.3.2.) to an adverb or a postposition (7.2.):

héna ‘right there’ (, éna). létu ‘here’ (, étu). katą' ‘from yonder’ (, etą'hą), hél ‘there’ (, él).

Interrogative adverbs are also single words for the most part (note that all Lakhota interrogative words  begin with t-):

tóhą ‘when?’ (referring to a realized event)
tohą'l ‘when?’ (referring to an unrealized event)
tuktél ‘where?’
tukté él ‘at which place,’ ‘whereabouts?’
tókhiya ‘where (in or to what region)?’
tákuwe ‘why?’
tókheškhe
(tóškhe)
‘how?’

In some cases, one-word adverbs are words whose principal use is as some other part of speech. For example, hį'hąni is a noun or verb meaning ‘morning’ or ‘be morning’; áataa is a pronoun meaning ‘all (of something)’; and tąyą' is a verb meaning ‘to be well’.

Adverbs may be marked as intensive or repetitive by reduplication (4.3.2.8.) or by the addition of suffixlike particles such as ĥča/ĥči and šna:

Héktaktakiya wačhí. ‘She kept dancing backward.’
E'nagna hiyéye. ‘They [inanimate] are lying here and there.’
Tąyé ĥči ečhų'. ‘He did it very well.’
E'na šna yąké. ‘He is always there.’
Hąhépi iyóhila él thiwáhepi kį ob thimá šna hiyú pi. ‘Every night they would come in with their families.’

7.2. Adverbial Phrases

Adverbial phrases generally contain a nominal (8.0.), sometimes accompanied by a postposition. Other adverbs may also participate in addition to the nominal.

7.2.1. Adverbial Phrases of Time

In adverbial phrases of time, nominals are usually accompanied by the articles (8.3.1.) and k?ų. in such phrases marks the phrase as referring to ‘hypothetical’ (unrealized) time, while k?ų marks ‘actual’ (realized) time.

When the nominal expression does not include a postposition, its use is absolute (not grammatically marked):

Lé ąpétu kį maĥpíyaya. ‘It is cloudy today.’

Hąhépi mağážu. ‘It rained last night.’

English analogs of these are seen in the sentences ‘He came this morning’, and ‘It rained last night.’

Other examples of this kind are hį'hąni kį ‘tomorrow’, hąhépi kį ‘this evening’.

Nominals used absolutely as adverbs are often followed by a true adverb:

Okó núpa k?ų héhą
week two the.past then
‘two weeks ago’
hináphe šni hąní
sun come.out not not.then
‘just before sunrise’

Examples of postpositional phrases with temporal meaning are:

ĥta?íyokpaza óhakab
darkness after
‘after dark’
ąpétu él
day the that at
‘on that day’
ąpétu-wakhą' núpa okó
day-holy two between
‘between two Sundays’.

7.2.2. Adverbial Phrases of Place

Numerous postpositions are also used in adverbial phrases of place.

Mas?óphiye él ačhíphį kte
Store at I.await.you potential
‘I’ll wait for you at the store.’
Wačhípi ektá pi kta he?
Dance to you.go pl. potential question
‘Are you (pl.) going to go to the dance?’
Othų'wahe etą'hą wahí
town from I.arrive
‘I arrive from town.’
Thípi ókšą énažį
house the around they.stand
‘They (collective) stood around the house.’

The line between adverbs and postpositions is sometimes difficult to draw, chiefly because the same words are often used both ways. English adverbs and prepositions show the same kind of interchangeability. ‘Come on out from down in under there!’ has six adverb/prepositions in this kind of ambiguous function. A Lakhota example is:

Owóte-thípi wígli-o?ínažhį hél isákhib hé.
eating.place-house the oil-stopping.place the there beside stand
‘The cafe is there beside the gas station.’

In this example the adverb isákhib functions nearly as a postposition.

Very often, a noun will combine with a postposition to form a compound; the result is the conversion of a phrase into a single word adverb:

thimáhel ‘in the house’
thí-mahél ‘house’-‘inside’
čhą?ákhotąhą ‘across the woods’
čhą'-?akhótąhą ‘tree’-‘across’
thilázata ‘behind the house’
thí-lazáta ‘house’-‘behind’

Most specific locational adverb/postpositions of place begin with i:

ilázata ‘behind’
isákhib ‘beside’
ihúkhuta ‘below’
ihákab ‘after’
ithókab ‘in front of’

For some of these there is a corresponding word without the i, which is only used as a postposition. For all of the i- adverbs there is an alternative form with stressed i, which marks the location as very close to or against the object:

ílazata ‘right close behind’
ísakhib ‘right next to’
íhukhuta ‘just below’
íhakab ‘right after (also refers to time’
íthokab ‘right before (also refers to time)’

Both these stis of forms take personal object inflections (see 9.3.2.) when the object is an animate pronoun:

milázata ‘behind me’
wičhíthokab ‘in front f them’

The stressed í appears as íi in inflected forms:

níisakhib ‘very close to you’
ųkíihukhuta ‘right below us’

7.2.3. Adverbs of Manner

Adverbs of manner are often single words formed from other classes of words. A frequently used formative of such adverbs is the suffix -ya:

wašíčųya ‘in English’ (wašíču, wašíčų ‘White man’)
šičáya ‘badly’ (šíčA ‘be bad’)
wakhą'yą ‘mysteriously’ (wakhą' ‘wondrous, awesome’).

Other suffixes are also used.

léčhel ‘thus, in this way’ ( ‘this’)
naĥmála ‘secretly’ (naĥmá ‘to hide’)
mánikhel ‘on foot’ (máni ‘to walk’)

There are also postpositional phrases that express manner:

Wasnásnaheča čhįčála s?e ipáĥlala yąká pi
kindbird offspring like in.a.row sit pl
‘They are sitting in a row like young kingbirds.’
Čhįča pi ob
offspring pl. with come
‘He came with their children.’
Čhąpágmiyąpi ogná iyáye
wagon in go
‘He went in a wagon.’

7.2.4. Instrumentality

Instrument is sometimes expressed in the verbal prefix (see 9.2.1.). When a more specific instrumentality needs to be indicated, the postposition ų' is used with an appropriate noun:

Iyéčhįkįyąke ų' nat?é
car with die.by.foot.action
‘He was run over and killed by a car.’
Ağúyapi ečéla ų' wičháša pi šni
bread the alone with man the live pl. not
‘Men do not live by bread alone.’

7.3. Adverbial Clauses

Sentences used as adverbs (adverbial clauses) are first nominalized; they are then the equivalent of a noun in absolute (see 7.2.1.) use. Following this nominal comes a time adverb or a postposition. (Compare this with the similar construction and use of phrases described in 7.2.)

Sentential time adverb:

Ĥ?okhá pi k?ų héhą, wačhípi iyáye.
Singers the arrive pl. the.past then dance the start
When the singers came, the dance began.’

Sentential place adverb:

wa?ų'yąka pi okíhi šni él iyáye.
He he.see.us pl. he.be.able not a to he.go
‘He went to (a place) where he could not see us.’

Sentential manner adverb:

Táku iyéya okíhi ų' čhą' kaksáksa.
What he.find he.be.able the that with wood the he.chop
‘He chopped the wood with whatever he could find.’

Buechel (1939 passim) contains a more complete description of adverbs than is presented here. Note that many of the words he calls ‘adverbs’ are called ‘enclitics’ in this sketch.

8. Nominals

There are from zero to three nominal expressions (apart from verbal affixes) in every Lakhota sentence. The three slots provided in the original sentence diagram serve three possible roles: subject, indirect object, direct object. If more than one nominal occurs in a sentence, the order will ordinarily be subject first, then either of the objects. For some speakers, there is no required order between direct and indirect objects; thus ‘the men gave the boy to the bear’ and ‘the men gave the bear to the boy’ are identical: Wičháša kį hokšíla kį mathó kį k?ú pi or Wičháša kį mathó kį hokšila kį k?ú pi. Very rarely, however, does this cause any misunderstanding, since in most such sentences potential ambiguities are resolved by the meanings of the particular words. Other speakers insist that the order is subject-indirect object-direct object; for them. the two sentences above are not synonymous.

The presence of an indirect object is marked in the verb except for a handful of special verbs such as k?ú ‘give’, ‘ask for’ or iyų'ğA ‘to ask someone something’ (see 9.1.6. and 9.3.8.).

The grammatical roles of subject, object, or indirect object may be indicated by verbal affixes (prefixes or infixes; see 9.3.) instead of by overt nominals. If there are separate nominal expressions, they may be any of four types: pronouns, nouns, modified nouns, and sentences.

8.1. Pronouns

Independent pronouns are rarely used in ordinary Lakhota but are available for emphatic expressions or to serve as the objects of postpositions such as kičhí ‘together with’. There are two sets. The first is simply emphatic: the second is used to contrast one referent with others.

Set1:

miyé ‘I’
ųkíye ‘we’
niyé ‘you’
iyé ‘he, she, it, they’

Set 2:

míš ‘I’
ųkį'š ‘we’
níš ‘you’
į'š ‘he, she, it, they’

Examples of pronouns used as nominals:

Subject

Miyé, wačhį' šni. Me, I don't want to.’
Miš táku ophéwathų kta he? ‘And what shal I buy (now that the others have decided)?’

Object

Hená ųkíye wa?ų'yąka pi šni. As for us, they didn't see us.’

Postpositional Object

Tuwá niyé kičhí wačhí pi he? ‘Who danced with you?’

The two sets can be used together, for example, in the expression míš-miyé   kį ‘as for me; in my opinion; for my part’. Compare the expression Níš ehą' ‘your turn’, using the postposition ehą' ‘at a time’ with a contrastive pronoun.

8.2. Nouns

Lakhota nouns are either simple or derived; derived nouns may be either compounds or affixed forms. Examples of simple nouns include čhą' ‘wood, tree’; natá ‘head’; šų'ka ‘dog’, and maĥpíya ‘cloud, sky’.

8.2.1. Compound nouns

Compound nouns consist of two (or more) nouns, or of a noun plus a verb. (De Reuse 1994 discusses the degree of "tightness" in noun-verb compounds.) The elements that enter into the compound may exist as independent words, or they may be compounding forms (essentially roots) that never appear in that form outside of compounds. Where all compounded elements occur alone as words, the compound is written with a hyphen between the elements. Where one or more of the compounded elements is in root form, the compound is written without separation of the constituent elements. Stresses after the first in any word should be read as secondary (see 3.1.).

In noun-noun compounds, the earlier element usually modifies the later. When non-nominal elements are present in the compound, these usually follow the nominal elements and modify them, but they precede in some cases.

Noun-noun compounds
1. Modifier-modified

wígli-o?ínažį ‘gas station’
wígli ‘grease’
o?ínažį ‘stopping place’
čhą-hą'pi ‘sugar’
čhą' ‘tree’
hą'pi ‘juice’
Khąğí-wičháša ‘Crow Indian’
khąğí ‘crow’
wičháša ‘man’
mas?óphiye ‘store’ (originally ‘cash register’)
máza ‘metall’
ophíye ‘box, storage place’

2. Modified-modifier

šųňmánitu ‘coyote’
šų'ka ‘dog’
manítu ‘wilderness, wild place’

Noun-verb compounds

pte-yúha ‘rancher’
pté ‘buffalo’
yuhá ‘to have’
Mní-šoše ‘Missouri River’
mní ‘water’
šošé ‘be turbid’
šųňkhíyuĥa ‘stud, stallion’
šų'ka ‘dog, horse’
khiyúĥa ‘to breed’
máza-ská ‘money’
máza ‘metall’
ská ‘be white’
šų'ka-wakhą' ‘horse’
šų'ka ‘dog’
wakhą' ‘be awesome, be marvelous’
hąm?íkčeka ‘mocassin’
hą'pa ‘shoe’
ikčéka ‘be common’

Verb-noun compounds

tųwéya-wičháša ‘scout’
tųwéyA ‘crow’
wičháša ‘man’
wayáwa-thípi ‘school’
wayáwa ‘count things; read things’
thípi ‘house’

Compounds of the verb-noun type may in reality be examples of noun + noun, if the first element is actually a nominalized verb (cf. wayáwa wą ‘a student’). There are no known reliable criteria that can distinguish these possibilities.

Compounds that consist of more than two included elements also exist. These compounds have an internal hierarchy indicated by underlining in the examples:

Mni-lúzahe-othų'wahe
water-run.swiftly-town, rapids town
‘Rapid City, South Dakota’

šųk?áką-yąkápi
horse/dog.on-sit, horseback sit
‘to ride horseback’

natá-yazą'pi-phežúta
head-ache-medicine
‘aspirin’

šųň-sí-maza
dog/horse-foot-iron
‘horseshoe’

máza-ská-zí
iron.be.white-be.yellow, silver-be.yellow
‘gold’

8.2.2. Derived nouns

Nouns derived by affixation may have either prefixes or suffixes. Elements used as .suffixes are usually identical to enclitics (10.). Prefixes tend to have fairly specific meanings, while (lie meanings of the suffixes are more general, though related to the meaning of the same element used as a verbal enclitic.

Some prefix examples:

o- ‘place where’

o?ínažį ‘station’ (cf. inážį ‘to stop’)
oyą'ke ‘sitting place; agency’ (cf. yąkÁ ‘to sit’)
othí ‘den’ (cf. thí ‘to dwell’)
ognáke ‘container’ (cf. gnákA ‘to put away’)

i- ‘instrument for’

wa?íyatke, wíyatke ‘cup’ (cf. wayátkĄ ‘to drink things’)
wa?íkhalye, wíkhalye ‘coffee pot’ (cf. wakhályA ‘to heat things’)
?į' ‘harness’ (cf. k?į' ‘to pack on the back’)

Some suffix examples:

thípi ‘house’ (cf. thí ‘to dwell’)
wakhályapi ‘coffee’ (cf. wakhályA ‘to heat’)
yazą'pi ‘pain’ (cf. yazą' ‘to hurt or ache’)
wa?éčhųčhųka ‘jack-of-all-trade’ (cf. wa?éčhų ‘to do things’)
wamánus?a ‘thief’ (cf. wamánu ‘to steal things’)

8.3. Determiners

Determiners are a class of words that terminate nominal expressions. There are three kinds: articles, demonstratives, and quantifiers. Determiners occur in the order:

(quantifier) (article) (demonstrative) (quantifier).

Nominals may also appear without a determiner.

8.3.1. Articles

Words that function as articles include , k?ų, , wąží, wąžíni, eyá, etą', etą'ni, tákuni, tuwéni, čha.

The choice of the article depends on various features of the noun and of the sentence in which it occurs. The noun may be generic, that is, may refer to all or any of a class of objects, such as ‘dogs’, in ‘dogs bark’ or ‘coffee’ in ‘coffee is brown.’ Such nouns generally have no article in Lakhota no matter what the rest of the sentence may be. In addition, countable nouns used generically always take a plural verb:

wičháša iyéčhįkįyąke ophéthų.
than man the car buy
‘That man buys cars.’
Igmúla sįté yukhą' pi.
cat tail have pl.
Cats have tails.’
Hokšíla thąkthą'ka čhéya pi šni.
boy big cry pl. not
‘Big boys don’t cry.’
Wakhályapi wačhį'.
coffee I.want
‘I want coffee.’

A determiner is used in a generic construction when the construction could otherwise be understood as a word or phrase rather than as a sentence:

Lakhóta kį wačí pi. Indians dance.’
cf. Lakhóta-wačhípi ‘(an) Indian dance’
Thaspą' kį šašá. Apples are red.’
cf. thaspą' šaša ‘red apples’

If a noun is not generic, it must be either definite or indefinite.

If the noun is definite, the article is either or k?ų; ‘the’ is the English equivalent of both of these. The difference between and k?ų seems to be that k?ų marks more emphatically definite nouns. Often, therefore, k?ų can be translated as ‘the aforementioned’, although this is usually abbreviated ‘the.past’ in glossing the examples:

Sífka k^H thal? ki thebye.

Šų'ka hél yųké.
(dog the there lie)
The dog is lying there.’
Šų'ka k?ų thaló thebyé.
(dog the.past meat the eat.up
‘The aforementioned dog ate up the meat.’

This example is from a traditional tale.

Recall that nominals in adverbial functions are nominalized by if the reference is to hypothetical time, but by k?ų if the reference is to real time. Sentences containing k?ų are always translated with the English past tense. This is evidently an attempt by Lakhota speakers to render the hyperreality of the Lakhota sentences with k?ų. In fact, at present is used regularly in real as well as hypothetical sentences. The difference between and k?ų may have been sharper at an earlier time, since different forms of the indefinite article are used in sentences with real versus hypothetical meaning.

All the other words listed above are indefinite articles. The choice of indefinite article is made on the basis of a number of covert classes to which nouns belong (table 3). These include mass form (that is, whether the object named by the noun can be counted. like houses, or not, like soup), human, and non-human. Moreover, there are different forms depending on whether the sentence in which they appear is negative or affirmative; and, if it is affirmative, whether it refers to real or to hypothetical things.

Table 3. Indefinite Articles
  Real Hypothetical Negative
Singular ‘a, an’ wąží ‘a, an’ wąžíni ‘not...a, no, not any’
Plural  Animate Human eyá ‘some’ etą' ‘some, any’ tuwéni ‘no one, not..any’
Nonhuman eyá ‘some’ etą' ‘some, any’ tákuni ‘no, not...any’
Inanimate eyá ‘some’ etą' ‘some, any’ tákuni ‘no, not...any’
Noncountable eyá ‘some’ etą' ‘some, any’ etą'ni ‘no, not...any’

Note: An older form of eyá in all its uses is k?eyá. Some speakers still use this word.

The following examples illustrate the use of the indefinite articles. Note that although nonhuman and inanimate plurals are marked in the same way, sentences with nouns from these categories differ because all animate (human and nonhuman) plural objects require -wičha- in the verb, and animate plural subjects require pi, while inanimates never occur with -wičha- or pi (see 9.3.2.-9.3.4.). In the examples below, then, the verbs used with ‘houses’ differ from those used with ‘birds’, even though the articles are the same:

Iyéčhįkįyąke ophéwathų. ‘I bought a car.’

Iyéčhįkįyąke wąží ophéwathų kte. ‘I’m going to buy a car.’

Iyéčhįkįyąke wąžíni ophéwathų šni. ‘I didn’t buy a car.’

Lakhóta eyá hí pi.Some Indians have come.’

Lakhóta etą' hí pi he? ‘Have any Indians come?’

Lakhóta tuwéni hí pi šni.No Indians have come.’

Wakhą'yeza eyá wąwíčhablake. ‘I saw some children.’

Zįtkála eyá wąwíčhablake. ‘I saw some birds.’

Čhą'-thipi eyá wąbláke. ‘I saw some houses.’

Wakhą'yeža etą' wąwíčhalaka he? ‘Did you see some/any children?’

Zįtkála etą' wąwíčhalaka he? ‘Did you see some/any birds?’

Čhá-thipi etą' wąláka he? ‘Did you see some/any houses?’

Wakhą'yeza tuwéni wąwíčhablake šni. ‘I didn’t see some/any children.’

Zįtkála tákuni wąwíčhablake šni. ‘I didn’t see some/any birds.’

Čhą'-thipi tákuni wąláke šni. ‘Didn’t you see any houses?’

Wahą'pi eyá očhéthi aką'l hé. ‘There’s some soup on the stove.’

Wahą'pi etą' yačhį' he? ‘Do you want some/any soup?’

Wahą'pi etą'ni yatké šni. ‘He didn't eat (drink) any soup.’

A special construction exists to make a nominal emphatic (whether or not to call it "topicalized" depends on future studies of Lakhota discourse structure). This is often translated into English as ‘It was a/the NOUN who/which VERB’. Emphatic nominals of this kind are marked by the article čha; if the noun is indefinite, no further determiner is used:

Hokšíla čha šų'ka-wakhą' kį iwíčhaču pi.
'It was some boys who took the horses.’

If the emphatic nominal is also definite, the verb é ‘be a certain one’ precedes čha:

Hokšíla kį é pi čha šų'ka-wakhą' kį iwíčhaču pi.
It was the boys who took the horses.’

This use of čha is probably very closely related to the relative clause marker čha (8.4.2.3.).

8.3.2. Demonstratives

Lakhota has three demonstrative roots: ‘this,’ ‘that.,’ ‘yonder.’ Each of these can function in numerous ways: alone as a singular pronoun or as a noun modifier; with the suffix -na as a plural pronoun or modifier (lená ‘these’, hená ‘those’, kaná ‘the ones yonder’); or with the suffixes -na and -?ųs (some speakers use -yos) as dual pronouns or modifiers (hená?ųs, henáyos ‘those two’). Various adverbials are formed from demonstratives (7.1.)

Semanticaily, is the most neutral. Once a noun has been located, either by pointing or by description, in space or in the listener's mind, can then be used. Before that. or is usually used to demonstrate exactly what is meant, although may also be used while pointing.

šų'ka-wakhą' wąláka he? tuwá tháwa he?
that.(yonder) horse the you.see question that who it.belongs.to question
‘Do you see that horse? Who does he belong to?’

When demonstratives are used as nominal markers, they are usually accompanied by an article. They indicate the number (singular, dual, or plural) of the noun: hé wį'yą kį ‘that woman’, hená?ųs wį'yą kį ‘those two women’, hená wį'yą kį ‘those women’. The demonstrative may either precede the noun or follow the article. If the demonstrative precedes the noun, an article must occur after the noun: if the demonstrative follows, the article may be omitted. By far the most common article used is . but expression such as hená wį'yą eyá ‘these (indefinite) women’ may be used, usually in relative clauses (see 8.4.3.).

The difference between singular and plural is always indicated when demonstratives are used with countable nouns. This differs from the use of pi and wičha to mark plural with verbs, since these verbal elements refer only to animate nouns (see 9.3.1.).

8.3.3. Quantifiers

This class of words includes the numbers (‘one’-is the same as the singular indefinite article; table 3) núpa ‘two’, yámni ‘three’, tópa ‘four’, etc., a handful of indefinite numerals, such as óta ‘many’, hų'ĥ ‘some’ (note: this is not the same ‘some’ as those translated by eyá or etą', see 8.3.1. and 8.3.4.). čónala ‘few’, iyúha ‘all of a group’ (distributive), oyás?į ‘all of a group’ (collective), and áataa ‘all of a mass’, and the interrogative words tóna, tónakeča, and tohą'yą ‘how much? how many?’.

Many quantifiers may also function as stative verbs. Observe these examples:

Wičháša óta t?á pi.
man the many die pl.
Many of the men-died.’
Wičháša óta pi.
man the many pl
There are many men.’
Wičháša yámni iyéwičhaya pi.
man the three find.them pl.
‘They found three of the men.’
Wičháša yámni pi.
man the three pl.
There are three men; the men are three.’

Quantifiers may occur with or without articles or demonstratives, and either before or after them, but the meanings differ depending on order:

wičháša óta ‘many men’
wičháša hená óta ‘many of those men’
wičháša óta hená ‘those many men’
wičháša šakówį ‘seven men’
wičháša kį yámni ‘three of the men’
wičháša tópa kį ‘the four men’.

8.3.4. Partitive Markers

As can be seen from the examples, when used alone or before an article or a demonstrative, the quantifiers specify the size of the group. Used after the determiner, they indicate that the predicate refers to a specified part of the subject. The notion of partitive touched on here has complexities that call for further comment.

‘Some’ (= part of) is expressed in several different ways in Lakhota, depending on several factors: the nature of the whole, the nature of the part, and whether the part is positive, negative, or interrogative.

The whole may consist of separate, identifiable individuals (such as persons in a group), a single individual (for example, a watermelon), or an undifferentiated mass (flour). The whole may be either generic or specific. The part may be individuals, a portion of a single individual, or a portion of a mass.

Given these parameters, the choice of partitive marker is as given in table 4.

Table 4. Lakhota Indefinite Partitive
Affirmative Negative(4) Interrogative Meaning
1. hų'ĥ wąžíni tóna or
tónakeča
Some individuals from a group or individuals
2. hąké hąkéni tohą'yą Some of a single individual
3. hų'ĥ or
etą'
etą'ni tohą'yą or
tónakeča
Some of an undifferentiated mass

(4) The notions generic and negative are incompaiible. The negative partitive can thus be used only with specific reference.

Examples

1.

Oyáte hų'ĥ wičhášiče. Some people are evil.’
Lakhóta tóna wąwíčhalaka he? How many Indians do you see?’
Ĥ?okhá kį hų'ĥ hí pi. Some of the singers have come.’
Ĥ?okhá kį tóna hí pi he?. How many of the singers have come?’
Ĥ?okhá kį wąžíni hí (pi) šni. None of the singers has come.’

2.

Špą'-šni-yútapi hąké ųyúta pi s?a. ‘We (habitually) eat some watermelon.’
Špą'-šni-yútapi tohą'yą yáta pi s?a he? How much watermelon do you eat (habitually)?’
Ağúyapi-skúyela kį hąké ųyúta pi. ‘We ate some of the cake.’
Ağúyapi-skúyela kį hąkéni ųyúta pi. ‘We didn’t eat any of the cake.’
Ağúyapi-skúyela kį tohą'yą yáta pi he? How much of the cake did you eat?’

3.

Phežúta hų'ĥ phá. Some medicine is bitter.’
Phežúta tohą'yą nič?ú he? How much medicine did he give you?’
Ağúyapi-blú kį etą' ų' we. ‘Use some of the flour.’
Ağúyapi-blú kį etą'ni ų' šni. ‘She didn’t use any of the flour.’
Ağúyapi-blú kį tohą'yą nú he? How much of the flour did you use?’

8.3.5. Summary of determiners

To summarize the discussion of determiners: the complete set of possible slots in the nominal composed of a noun and its determiners is as follows:

(demonstrative) noun (quantifier) (article) (demonstrative) (quantifier).

8.4. Modified Nouns

8.4.1. Possession

Possession is marked in one of three ways: by special affixes in the verb, by an appropriate modifying form of the staiive verb itháwa ‘belong to, own’, or by special prefixes on the noun. Moreover, many (perhaps most) nouns, including some body parts such as á ‘armpit’ and ablóhu ‘shoulder blade’ cannot be formally marked for a possessor anywhere in the sentence. At present it appears impossible to predict whether a noun will be possessable or not, so this information must be pan of each noun’s dictionary entry. The marking of possession in the verb is discussed in 9.3.8.; the other expressions of possession will be described here.

If a noun can be possessed, the form of the possessive prefix differs depending on whether the noun is alienably or inalienably possessed. Alienably possessed nouns are things that can be acquired or given away; inalienably possessed nouns are understood as an inherent part of the owner’s person; they include body parts, many relatives. and (formerly, at least) some essentially personal things such as tools, clothing, and pets. The prefixes are:

alienably
possessed nouns

inalienably
possessed nouns

mithá- ‘my’ ma-/mi-
ųkítha- ‘our’ ųkí-
nithá- ‘your’ ni-
tha- ‘his, hers’ 0.

If the possessor is plural, pi follows the prefixed noun. It is also possible to analyze the alienable prefixes as complex, consisting of a stem-derivational element itha, prefixed to the noun, to which stative verb affixes are then prefixed.

Because possessed nouns are always definite, a definite article almost always occurs with them. The only exceptions are in partitive usage, for example, mithášųka wąží ‘one of my dogs’, where a quantifier replaces the article. The choice between ma- and mi- is semantically determined in Oglala speech: ma- is used of concrete visible possessions, mi- of intangibles: manáği kį ‘my shadow’, mináği kį ‘my spirit’. Speakers from other Lakhota-speaking groups differ as to their use of ma- and mi-.

Prefixation to show possession is not used with some possessable nouns. These are generally nouns thal are not normally possessed, such as rocks or trees. For such nouns, the possessive construction uses the stative verb itháwa. Itháwa is nearly unique among stative verbs for two reasons. First, it carries possible double affixation (see 9.1.3.). Second, it occurs as a noun modifier in its inflected as well as uninflected forms.

The possessive construction with itháwa has the following structure:

noun + itháwa form + .

The itháwa form depends only on the possessor:

mitháwa
‘my’
ųkíthawa
‘belonging to you and me’
ųkíthawa pi
‘our’
nitháwa
‘your’
nitháwa pi
‘your(pl.)’
tháwa
‘his, her’
tháwa pi
‘their’

Examples:

Šų'ka-wakhą' tháwa pi iyéwaye.
horse poss. pl. the I.find.it
‘I found their horse.’
Phežúta nitháwa líla waš?áke.
medicine your.sg..poss. the very be.strong
Your medicine is very strong.’

The criteria for choosing between the itháwa construction and the prefixed forms are not well understood.

Kinship terms generally follow a separate paradigm. They are inalienably possessed, but they often take a suffix -ku when the possessor is a third person; moreover, many of the forms have separate roots for the various forms. Here, as samples, are paradigms for ‘father’ and ‘mother’:

até, atéwaye ‘my father’ iná, ináwaye ‘my mother’
niyáte ‘your father’ nihų' ‘your mother’
atkúku ‘his, her father’ hų'ku ‘his, her mother’
até?ųyą pi ‘our father’ iná?ųyą pi ‘our mother’
niyáte pi ‘your father’ nihų' pi ‘your mother’
atkúku pi ‘their father’ hų'ku pi ‘their mother’

Buechel (1939:101-107) has extensive lists of paradigmatic forms for kin terms.

8.4.2. Other Modified Nouns

There are no words in Lakhota strictly equivalent to English adjectives; nevertheless, Lakhota nouns can be modified. The modifiers are usually stative verbs (like itháwa). but the constructions used for modification differ for definite and indefinite noun phrases.

8.4.2.1. Definite Modified Noun Phrases

If a modified noun is definite (marked by or k?ų), it is usually possible to form a phrase with this structure:

noun + modifier + .

Examples are:

wičháša thą'ka kį (man + big + ) ‘the big man’

wakhą'yeža čík?ala kį (child + small + ) ‘the little child’

iĥ?é ská kį (rock + white + ) ‘the white rock’

If the noun is plural, the modifier is reduplicated (4.3.2.8.):

wakhą'yeža čigčík?ala kį (child + small + ) ‘the little children’.

Phrases of this kind have a special intonation: only the noun receives primary stress; the modifier receives secondary stress (3.1.).

If the collocation of noun and modifier is very common, Lakhota speakers usually form a compound of the two, rather than use the phrasal construction. Thus šųňwášte kį ‘the good horse’ is preferred over the possible, but unusual šų'ka-wakhą' wašté kį.

8.4.2.2. Indefinite Noun Phrases with Modifiers

There are two constructions for modifying indefinite nouns. The choice between the two is chiefly idiomatic.

The less frequent construction is exactly like the construction used for modified definite nouns: noun + modifier - article (, eyá, etc.) This construction is used only when the modifier states an inherent quality that is expectable for the given noun: ‘a warm blanket’ or ‘a tall tree’ are examples of expectable inherent qualities. Lakhota examples are:

thípi thą'ka wą ‘a large house’
mní sní eyá ‘some cold water’.

Such phrases are often replaced by compound nouns: mni-sní eyá, sųňwášte wą ‘a good horse’.

If the modifier attributes to the noun a meaning that is unusual or unexpected, another construction is used instead:

noun + indefinite article + modifier + čha.

Čha is sometimes omitted. The position of the indefinite article immediately following the noun nevertheless contrasts the present construction with the one last described. Examples:

Šų'ka-wakhą' hušté (čha) wąbláke.
horse a limp I-saw
‘I saw a lame horse.’ (Compare Šųňhúšte wą wąbláke. ‘I saw a crippled horse.’)
Zįtkála eyá thothó (čha) wąwíčhablake.
bird some blue/green I-saw
‘I saw some blue birds.’
Wį'yą líla thą'ka (čha) amáphe.
woman a very big hit-me
‘A very big woman hit me.’
Mni-píğa etą' šiča (čha) ųyátką pi.
beer indef.-art bad we-drink pl.
‘We drank some bad beer.’

The construction with čha may actually be used correctly for any indefinite modified noun, including collocations that are semantically expectable combinations.

Modified generic nouns follow the indefinite pattern with čha, but without articles, of course:

Thaló ağúyapi čha waĥtéwala šni.
meat burned I dislike
‘I don't like burned meat.’

This leads naturally to a discussion of nouns that are modified by more than a stative verb, that is, nouns modified by a whole sentence. Sentence modifiers are called relative clauses.

8.4.2.3. Relative Clauses

A relative clause is a sentence that modifies a noun. The noun being modified must also occur as part of the modifying sentence. An English example would be ‘They arrested the man who hit me.’ In this sentence ‘who hit me’ modifies man. If this sentence were not acting as a modifier, it could not have who as its subject. Instead, it would have to be ‘A man hit me.’ The original sentence, then, contains two sentences, one included in the other: They arrested the man [a man hit me]. English grammar requires that the modifying sentence follow the noun it modifies, and furthermore that the noun in the modifying sentence be replaced by a relative pronoun, in this case who. (Which and that are also relative pronouns.)

In Lakhota relative clauses, the modifying sentence comes before the noun it modifies; the noun in the main sentence is then dropped, but the rest of the sentence remains. To construct the Lakhota for the English example above, begin with the modifying sentence (in which the shared noun is always indefinite):

wičháša amáphe
man a hit.me
‘a man hit me’

Then add the main sentence:

wičháša oyúspa pi
man the arrest.him pl.
‘They arrested the man.’

The result is:

Wičháša wą amáphe wičháša kį oyúspa pi.

Now drop the second wičháša, but keep its article:

Wičháša wą amáphe kį oyúspa pi. ‘They arrested the man who hit me.’

Even if the noun in the modifying sentence is the object of its sentence, it must come first in the complex sentence. Hence Wičháša wą hokšíla kį wąyą'ke kį can mean either ‘the man who saw the boy’ or ‘the man whom the boy saw’.

If the article in the second sentence is indefinite, it changes to čha:

Wičháša wą amápha čha oyúspa pi. ‘They arrested a man who hit me.’

The similarity between indefinite modified nouns and nouns modified by relative clauses should now be obvious:

Wičháša wą Iíla thą'ka čha oyúspa pi. ‘They arrested a very big man.’

Wičháša wą amápha čha oyúspa pi. ‘They arrested a man who hit me.’

Moreover, this construction is also reminiscent of the emphatic construction (8.3.1.):

Hokšíla čha owíčayuspa pi. ‘It was boys whom they arrested.’

Hokšíla kį é pi čha owíčayuspa pi. ‘It was the boys whom they arrested.’

A relative clause always comes first in a sentence, whether it modifies the subject or object; the resulting ambiguities are rarely a problem, since contexts or probabilities will clarify nearly anything.

Here are some more complex examples of relative clause constructions:

Hokšíla eyá šų'ka-wakhą' iwíčhaču pi
boy some horse take.them pl. the
čhąksá-yuhá owíčhayuspa pi.
police the arrest.them pl.
‘The police arrested the boys who took the horses.’

 

Šų'ka-wakhą' eyá hokšíla iwíčhaču pi
horse some boy the take.them pl the
wičháša iyéwičháya pi.
man the find.them pl.
‘The men found the horses the boys took.’

The two improbable sentences ‘The boys who took the horses arrested the policemen' and ‘The horses the boys took found the men' would be exactly like those two examples, but their improbable meanings would keep them from being understood in that way without elaborate explanatory context.

To summarize modified nominals, including those with relative clause modifiers:

(a) definite nominals: noun + modifier + { }
k?ų
(b) indefinite nominals
1) semantically expectable collocations: noun + modifier + article
2)others: noun+article+ { modifying word }+{ }
k?ų
modifying sentence
čha

For a different approach to this and many other subordinate clause constructions, see Simons (1989).

8.5. Sentences as Nominals

Many verbs permit whole sentences to serve as their subjects or objects. English sentences used as nominals are marked in one of three ways: with for...to; with that; with ’s...-ing. Secondarily, for, that, ’s, and to are sometimes omitted. Observe these examples:

It's time for him to take his medicine.

I told him to come.

She helped me (to) find the right page.

I said (that) he should call me.

The cat(’s) scratching (of) the couch annoys mother.

In Lakhota, too, sentences can serve as the subjects or objects of verbs. These sentences may be marked by , k?ų. or by no article at all. If or k?ų is used, a demonstrative (see 8.3.2.) may also be used. Usually the need for or k?ų is determined by which main verb is used.

8.5.1. Marked Nominalized Sentences

The following verbs require a determiner on their complements: slolyÁ ‘know’, iyókiphi ‘be happy that,’ šíčA ‘be bad that,’ šičáya ‘be too bad that’, wašté ‘be good that’, waštélakA ‘like (it) that’, waĥtéla šni ‘dislike (it) that’, wąyą'kA ‘see that’, yawá ‘read that,’ naĥ?ų' ‘hear (something)’ (note: not ‘hear that’) and héčha ‘be necessary that ...right away'. Examples include:

Wąčhį'yąke wašté.
I.see.you the this good
‘It’s good to see you.’
Kičhíwomayaglaka pi iyómakiphi.
you.discuss.with.me pl. the I.am.glad
‘I’m glad you (pi.) discussed it with me.’
Mas?óphiye ektá théhą ų' pi šni wašté.
store at long stay pl. not the good
‘They better not stay at the store long.’
Agnes mni-píğa yatké waĥtéwala šni.
Agnes beer drink the I.like not
‘I don’t like (for) Agnes to drink beer.’

The word k?ų can substitute for under certain circumstances. K?ų identifies a strongly asserted factual statement: it can only be used if the assertion is known to the speaker to have been an accomplished fact when the action described by the main verb began. Consequently it is never used with verbs expressing question. doubt, or value judgments.

8.5.2. Unmarked Nominalized Sentences

Verbs that require their complement sentence to be unmarked include čhí ‘want’, eyÁ (kéyA) ‘say’, okíhi ‘be able to’, iyúkčą ‘to think, plan, expect’, onáĥ?ų ‘to hear about’, kéya naĥ?ų' ‘to hear that...’ and okíyakA ‘to tell someone (that)’. Here are some examples:

Šų'ka-wakhą' wąží ophéyathų yačhį' he?
horse one you.buy you.want question
‘Do you want to buy a horse?’
Šų'ka-wakhą' wąží ophéthų onáyaĥ?ų he?
horse one he.buy you.hear.about question
‘Did you hear abou: his buying a horse?’
Waná wačhípi ektá ųkíyaya pi ųkókihi pi.
now dance to we.leave pl. we.can pl.
‘We can leave for the dance now.’
Ožą'žąglepi wąží bluğą' yačhį' he?
window a I.open you.want question
‘Do you want me to open a window?’

8.5.3. Other Nominalized Sentences

A few verbs permit sentential complements that are like infinitive or participle constructions in English. An example of such an English sentence is He told me to come, in which the infinitive construction me to come represents a reduced sentence that is seen more clearly in ‘He said that I should come.’

In Lakhota constructions of this kind, the complement verb precedes the principal verb. If the complement verb is transitive it may have object affixes (9.3.2.), but otherwise it has no affixes. The principal verb has the expectable affixes (i.e.. subject and/or object affixes). The close relationship between verbs in this construction may be shown in ways other than the unusual distribution of personal affixes. Stress patterns typical of compound words and phonological changes of the kind described in 4.3.2.6. and 4.3.2.7. are two such ways.

Examples of this Kind of construction are as follows:

Wówapi kIn yawá maší.
book the read me.he.commanded
‘He told me to read the book.’
Owíčhale ómayakiya wačhį'.
look.for.them you.help.me I.want
‘I want you to help me to look for them.’

(This sentence contains two subordinate verbs, but only the first is of the type described here; cf. 8.5.2. for čhį'.)

Nahą'ĥči škal yúštą pi šni he?
yet play finish pl. not question
‘Haven't they finished playing yet?’

(Note that skal yúštą is stressed like a compound.)

Two other uses for such reduced sentences are for purpose clauses with verbs of motion and for loosely joining sentences expressing sequential events, usually in stories. Here are examples:

Thaló ophéthų mní kte.
meat buy I.go potential
‘I’ll go buy meat.’
Wičháša etą' wąníyąg ú pi kte.
man som see.you come pl. potential
‘Some men will come to see you.’
Tákuni isą'm tókhų šni, ųglíyaču pi.
nothing more do not we.come.home pl.
‘Doing nothing more, we came home.’

In very colloquial speech, some verbal constructions of this kind may be reduced even further by completely omitting affixes from the first verb, and indicating all grammatical relations in the principal verb. If the principal verb is ordinarily intransitive, it nevertheless appears here with transitive affixes. Thus Wičháša wą wąníyąg hí ‘A man came to see you’ may become Wičháša wą wąyą'g nihí.

9. Verbs

9.1. Sentence and Affix Types

9.1.1. Impersonal Verbs

9.1.2. Stative Verbs

9.1.3. Stative transitive Verbs

9.1.4. Active intransitive Verbs

9.1.5. Active transitive Verbs

9.1.6. Verbs Requiring Three Complements

9.2. Verbal Derivation

9.2.1. Complex Stem Formation

9.2.2. Class-changing Processes

9.2.2.1. Causatives

9.2.2.2. Inceptives

9.2.2.3. Deactivization or Stativization

9.3. Verb Inflection

9.3.1. Stative Affixes

9.3.2. Object Affixes

9.3.3. Active Subject Affixes

9.3.4. Two-affix Combinations

9.3.5. Irregular Verbs

9.3.6. Reflexive Verbs

9.3.7. Reciprocal Verbs

9.3.8. Reflexive Possessive, Dative, and Benefactive Verbs

9.3.8.1. Reflexive Possessive

9.3.8.2. Dative

9.3.8.3. Benefactive

10. Enclitics

11. Selected Vocabulary