home edit page issue tracker

This page pertains to UD version 2.

It appears that you have Javascript disabled. Please consider enabling Javascript for this page to see the visualizations.

UD for Tatar

This is a work-in-progress overview of the UD annotation for Tatar.

Tokenization and Word Segmentation

In general, words are delimited by white spaces or punctuation, Whitespaces or punctuation may appear in some abbreviations or numeric expressions.

Morphology

Tatar is an agglutinative language with suffixation, and has a rich inflectional and derivational morphology. This morphological complexities cause some conflicts with the current UD v2 feature inventory. For example, a verb’s voice can contain more than one values. To this problem, the treebank temporarily employs a manner proposed by the Turkish treebanks, and just concatenates the values together in the alphabetical order.

Tags and Features

This is an overview only. For more detailed discussion and examples, see the list of Tatar POS tags.

PART is not currently used in the treebank.
Negation particle түгел is marked as AUX.
Copular and modal иде is marked as AUX.
Non-firsthand evidentiality particle икән is marked as AUX.
In a certain set of grammaticalized converb construction, the finite verb is marked as AUX.
There are four main (de)verbal forms, distinguished by the value of the VerbForm feature. They are always tagged as VERB:
- Finite verb Fin
- Participle Part
- Converb Conv
- Infinitive Inf
- Verbal noun Vnoun

Syntax

Subject nsubj is typically an actor of the predicate and is marked with a nominative case.
Object obj is typically an undergoer of the transitive predicate and is marked with an accusative case when it is definite, but is not marked (bare form) when it is indefinite.
Oblique obl is an argument that directly modifies the predicate and is marked with dative, locative, ablative, and equative case. Also, time expressions and other adverbial expressions in nominative (e.g., ике тапкыр “two times”) are categorized as obl. Note that this notion of oblique is rather optimized for UD, and is not precisely the best categorization with regard to descriptive linguistics of Tatar.
Tatar is a partly zero-copula language; the copula is usually zero in the present tense, but in the past tense it is expressed by means of an auxiliary AUX иде and in the future tense a verb VERB бул-. Note that the latter бул- is tagged as VERB, not AUX, and the relation from nominal or adjectival predicate is xcomp, though this needs cross-linguistic discussions in detail.

Relations Overview

The following relation subtypes are currently used in Tatar:
- compound:lvc

However, more relation subtypes are considered to be used in later versions, such as:

advmod:emph
nmod:poss
obl:tmod

Treebanks

There is one Tatar UD treebank:

Tatar-NMCTT