home edit page issue tracker

Specific constructions

Clausal structures

In Russian, the standard case pattern of a predicate-argument construction is as follows: * the first argument (subject) is in the Nominative case; * the second argument (direct object) is in the Accusative case; * all other arguments are coded in other cases or in the prepositional phrase depending their semantics. However, there is a number of constructions which have non canonical case patterns.

Constructions with Dative subject

The constructions are instantiated by an infinitive verb (usually under negation) or a predicative, the first argument of which is in the Dative case and not in the canonical Nominative case. The Dative argument is labeled iobj.



  
    input
    brat
    info
  
Роботу не угнаться за собакой . \n Robot.Dat cannot-keep-pace.Inf with dog .
iobj(угнаться, Роботу)
iobj(cannot-keep-pace.Inf, Robot.Dat)
{
    "text": "Роботу не угнаться за собакой . \n Robot.Dat cannot-keep-pace.Inf with dog . ",
    "entities": [
        [ "embedded-1-T1", "token", [ [ 0, 6 ] ] ],
        [ "embedded-1-T2", "token", [ [ 7, 9 ] ] ],
        [ "embedded-1-T3", "token", [ [ 10, 18 ] ] ],
        [ "embedded-1-T4", "token", [ [ 19, 21 ] ] ],
        [ "embedded-1-T5", "token", [ [ 22, 29 ] ] ],
        [ "embedded-1-T6", "token", [ [ 30, 31 ] ] ],
        [ "embedded-1-T8", "token", [ [ 34, 43 ] ] ],
        [ "embedded-1-T9", "token", [ [ 44, 64 ] ] ],
        [ "embedded-1-T10", "token", [ [ 65, 69 ] ] ],
        [ "embedded-1-T11", "token", [ [ 70, 73 ] ] ],
        [ "embedded-1-T12", "token", [ [ 74, 75 ] ] ]
    ],
    "relations": [
        [ "embedded-1-R1", "iobj", [ [ "arg1", "embedded-1-T3" ], [ "arg2", "embedded-1-T1" ] ] ],
        [ "embedded-1-R2", "iobj", [ [ "arg1", "embedded-1-T9" ], [ "arg2", "embedded-1-T8" ] ] ]
    ],
    "attributes": [],
    "error": false,
    "sentlabels": [ "1" ]
}



  
    input
    brat
    info
  
Мне стыдно за вас . \n I.Dat am-ashamed.PRAEDIC for you .
iobj(стыдно, Мне)
iobj(am-ashamed, I.Dat)
{
    "text": "Мне стыдно за вас . \n I.Dat am-ashamed.PRAEDIC for you . ",
    "entities": [
        [ "embedded-2-T1", "token", [ [ 0, 3 ] ] ],
        [ "embedded-2-T2", "token", [ [ 4, 10 ] ] ],
        [ "embedded-2-T3", "token", [ [ 11, 13 ] ] ],
        [ "embedded-2-T4", "token", [ [ 14, 17 ] ] ],
        [ "embedded-2-T5", "token", [ [ 18, 19 ] ] ],
        [ "embedded-2-T7", "token", [ [ 22, 27 ] ] ],
        [ "embedded-2-T8", "token", [ [ 28, 46 ] ] ],
        [ "embedded-2-T9", "token", [ [ 47, 50 ] ] ],
        [ "embedded-2-T10", "token", [ [ 51, 54 ] ] ],
        [ "embedded-2-T11", "token", [ [ 55, 56 ] ] ]
    ],
    "relations": [
        [ "embedded-2-R1", "iobj", [ [ "arg1", "embedded-2-T2" ], [ "arg2", "embedded-2-T1" ] ] ]
    ],
    "attributes": [],
    "error": true,
    "sentlabels": [ "2" ]
}

Cf. “дат-субъект” in SynTagRus.

Adjectival and adverbial constructions

Comparative constructions

Most Russian qualitative adjectives and adverbs have both morphological and analytic comparative and superlative forms, e. g. умный “smart”, умнее (умней) “smarter”, более умный “smarter”, (наи)умнейший “smartest”, наиболее умный, самый умный “smartest”. The synthetic comparative forms are assigned the Cmp Degree feature while the synthetic superlative forms are treated as separate lemmata. The most frequently used comparative constructions are the following:

Миша.Nom умнее брата.Gen. “Misha is smarter than his brother.” (only with morphological comparatives)
Миша.Nom более умный / более умен / умнее, чем брат.Nom. “Misha is smarter than his brother.” (with both types of comparatives)
Миша.Nom самый умный / умнейший из.PREP всех.Gen. “Misha is the smartest of them all.” (with both types of superlatives)
Миша.Nom такой же умный / так же умен / столь же умен, как (и) его брат.Nom. “Misha is as smart as his brother.” (equality comparison)

The “lesser degree” comparison (expressed periphrastically) is marked the same way: * Миша.Nom менее умный / менее умен, чем его брат.Nom. “Misha is less smart than his brother.” (with both types of comparatives) * Миша.Nom наименее глупый из.PREP всех.Gen. “Misha is the least stupid of them all.” (with both types of superlatives)

Миша умнее брата . \n Misha is-smarter than-his-brother .
nmod(умнее, брата)
nmod(is-smarter, than-his-brother)

Миша умнее , чем брат . \n Misha is-smarter , than his-brother .
amod(умнее, брат)
amod(is-smarter, his-brother)
case(брат, чем)
case(his-brother, than)

Миша самый умный из всех . \n Misha is-the-most smart of them-all .
nmod(умный, всех)
nmod(smart, them-all)
case(всех, из)
case(them-all, of)
amod(умный, самый)
amod(smart, is-the-most)

Миша так же умен , как и его брат . \n Misha as then smart , as his brother .
amod(умен, брат)
amod(smart, brother)
case(брат, как)
case(brother, as-6)
advmod(умен, так)
advmod(smart, as-2)
discourse(так, же)
discourse(as-2, then)

Noun phrases with quantifiers

See nummod:gov for numerals governing the case of the noun: две.Nom жены.Gen, пять.Nom жен.Gen. See nummod for numerals not governing the case of the noun: с двумя.Ins женами.Ins. See det:numgov for the pronominal quantifiers governing the case of the noun: сколько.Nom жен.Gen. See det:nummod for the pronominal quantifiers agreeing in case with the noun: со сколькими.Ins женами.Ins. See compound for the compound numerals: двадцать два.

Other types of QP: * две.Nom целых.Gen.Sg четыре.Nom десятых.Gen.Sg миллиона.Gen.Sg рублей.Gen.Pl “2.4 million rubles” (i.e. “four tenth parts of million” with the ellipsis of “parts”) * к двум.Dat целым.Dat четырем.Dat десятым.Dat миллиона.Gen.Sg рублей.Gen.Pl “to 2.4 million rubles” ~~~ sdparse Две целых четыре десятых миллиона рублей . \n Two whole-parts four tenth million rubles . nummod:gov(рублей, миллиона) nummod:gov(rubles, million) nummod:gov(миллиона, десятых) nummod:gov(million, tenth) nummod:gov(десятых, четыре) nummod:gov(tenth, four) nummod:gov(целых, Две) nummod:gov(whole-parts, Two) conj(целых, десятых) conj(whole-parts, tenth) ~~~