UD for Wolof
Tokenization and Word Segmentation
- Following most tokenization patterns, words are delimited by whitespace.
- Clitics that attach to orthographic words are systematically segmented. Function words such as prepositions, conjunctions, auxiliaries and determiners can attach to other function or content words. For instance, a word like cib “in a” is segmented into the preposition ci “in” and the indefinite article ab “a”. A similar segmentation is done for focus markers, since these usually constitute an amalgam of a subject pronoun (e.g. mu “he/she”) and a morpheme marking focus (e.g. a which mark subject focus), resulting in a complex form (e.g. moo).
Morphology
Tags
- Wolof uses all 17 universal POS categories, including particles (PART). The most common usage of PART is for infinitive particles a (to).
- Wolof uses a wide range of auxiliary verb (AUX), including:
- a: subject focus marker
- avoir: French auxiliary avoir contained in some foreign material
- bu: negative imperative marker
- da: verb focus marker
- di: imperfective marker
- être: French auxiliary être contained in some foreign material
- la: non-subject focus marker
- na: neutral (i.e. non-focus) and perfective marker
- ngi: progressive marker
- woon: past tense marker
- The AUX category also includes copulas. Wolof hhas three copulas:
- di: is used in identificational sentences.
- la: is used in predicational sentences
- ngi: is used in presentative/progressive constructions.
- In addition the data contain the French copula être (to be) (French foreign material)
- The tag DET is used for definite/indefinite articles, quantifiers and determiner words used with a determiner function; they can precede or follow the noun they quantify. The tag PRON is used for subjects or objects of a noun phrase and to show possession. In addition, possessive, reflexive, relative, interrogative, demonstrative, and indefinite pronouns are also grouped under the PRON tag.
- Since Wolof lacks adjectival modifiers, the tag ADJ was used only to annotate foreign material (i.e. some of the French adjectives that occurred in the data).
Features
- Determiners agree with noun in Number and Noun class.
- Definite articles and pronouns may specify deixis information. Also, demonstratives encode information about deixis, including reference to the speaker and/or addressee.
- There is no morphological Case; instead, adpositions (ADP) are used as case markers and specify the role of a noun in a phrase.
- Focus is marked morphosyntactically. Focus on the subject, verb, or non-subject (i.e. any constituent which is neither subject nor main verb) is signaled by special morphemes. The encode the focus type respectively as Subject focus, Verb Focus and Non-subject focus.
- The tense feature distinguishes three values: past, present and future. Tense is typically marked on auxiliaries (e.g. dina “he will”) and particles (woon “past tense particle”) which follow the verb.
Syntax
- Wolof is an SVO language with a rigid word order.
Subjects
- Nominal subjects are in initial position.
- Nominal subject (nsubj) is a noun phrase without preposition.
- A (nonfinite) subordinate clause may serve as the subject and is labeled csubj.
Objects
- Objects follow the main verb.
- When a verb has two objects, the one immediately folowing the verb is the primary object, the second one treated as the secondary object.
Wolof uses the following relation subtypes:
- compound:prt to attach verbal particles to verbs
- compound:svc to connect verbs in a serial verb construction
- acl:relcl: relative clause modifier
- obj:caus: causative object (caused agent)
- obj:appl: direct applicative object
- iobj:appl: indirect applicative object
- obl:appl: oblique applicative argument
Copular constructions
- Wolof distinguish between three copulas: di, la, and ngi. These appear
- di appears in identificational sentences like Góor gi di njiit (The man is a leader). Such sentences follow the standard SVO worder order (i.e. Nominal Subject + Copula + Nominal Predicate). The predicate is typically a nominal one.
- la is used in predicational sentences like Góor gi njjit la (The man is a leader). In contrast to identificational sentences, Wolof predicational sentences do not show the SVO word order, but rather SOV order. The predicate, which is typically a nominal one, precedes the copula.
- ngi is mainly used in presentative/progressive sentences. Unlike the two previous copulas, the predicate of the ngi copula can only be a manner or a locative (e.g. in the house) as in Góor gi mu ngi ci kër gi (The man is in the house). Also, such copular constructions have an SVO word order.
Treebanks
There is only one Wolof UD treebank at present: