UD for Chechen 
Tokenization and Word Segmentation
- Words are generally delimited by whitespace.
-
Multiword tokens in Chechen are formed with clitics that attach phonologically to the following or preceding host word. The clitics are:
- The proclitic ma, which is used to negate imperatives and desideratives; otherwise it is used with emphatic functions
- The enclitic ’a, which has coordinating, additive, and emphatic functions
- The emphatic enclitic q
- Punctuation marks are attached to neighboring words. They are tokenized as separate tokens.
- There are no multiword tokens written with whitespace.
Morphology
Tags
- Chechen-MottDT uses 16 tags, with the exception of SYM.
- The tag PART is applied to the proclitic ma, the enclitic q, and the quotatives boox and baax
- The tag AUX is applied to two lexemes: d.u, which functions as a copula, and xülu, which serves as an auxiliary in complex temporal-aspectual constructions. Both items are inflected for grammatical agreement, tense, and have finite, participle, and free relative forms. In addition, the tag AUX is applied to the lexical root wa ‘stay’ when used as an auxiliary in complex temporal-aspectual constructions to contribute imperfective reading.
- The tag DET is applied to pronominal items used attributively. One exception is possessive pronouns, which are all tagged as PRON.
- Predicative uses of pronominal items are tagged as PRON.
Features
Nominal features
-
Nouns, adjectives, determiners, and numerals up to ‘five’ inflect for Case and Number.
- Nouns and personal pronouns have extensive case paradigms:
Nom,Gen,Dat,Erg,Ins,Loc(nouns only),All,Abl,Ins,Lat, andCmp. - Adjectives and determiners make a
Nomnominative /Accaccusative (oblique) distinction.
- Nouns and personal pronouns have extensive case paradigms:
- There are two language-specific nominal features: [Log]{} and NounClass.
- Personal pronouns have a feature Log, applied for uses of third person reflexive forms in direct speech to refer to the reported speaker.
Verbal Features
- Verbs inflect for Tense, Aspect, Mood, and NounClass agreement.
- Verbs are annotated for VerbForm:
Conv,Inf,Part,Noun. - Auxiliaries inflect for Tense and NounClass agreement.
- There are four NounClass types:
J-class,V-class,D-class, andB-class.
Syntax
-
Chechen has ergative-absolutive alignment, ergative is marked, while absolutive is unmarked. Ergative-marked arguments are syntactic subjects of transitive verbs. Absolutive arguments are syntactic subjects of intransitive verbs and are syntactic objects of transitive verbs. Predicates agree with the absoultive argument; the case of the argument, the agreement pattern, and the valence of the predicate are used to distinguish between the subject and the object argument.
-
The main copula construction in Chechen uses a form of the copula d.u. In finite clauses, the copula is inflected for subject agreement, tense, and aspect (as relevant).
1 so so PRON _ Case=Abs|Number=Sing|Person=1 5 nsubj
2-3 juq _ _ _ _ _ _
2 ju d.u ADV _ NounClass=Jclass|Tense=Pres 5 cop
3 q q PART _ _ 2 advmod:emph
4 hwa hwa PRON _ Case=Gen|Number=Sing|Person=2 5 nmod
5 jow jow NOUN _ Case=Abs 15 root
so ju =q hwa jow
1SG.ABS J-be.PRS EMPH 2SG.GEN girl
"I am your daughter."
-
There two language-specific relations: advmod:emph, ccomp:reported, and obj:caus.
-
The subtype relations used are as follows:
Treebanks
There is one Chechen UD treebank: