CC BY
26
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
Olga V. Mitrenina
St Petersburg State University Philological Department,
St Petersburg
Синтаксис псевдокоррелятивных конструкций с местоимением который в старорусском
Митренина Ольга Владимировна
Санкт-Петербургский государственный университет, филологический факультет, Санкт-Петербург
Abstract
This article takes a close look at pseudo-correlatives: multiple sentences in Middle Russian with the pronoun kotoryj 'which' in the first clause. It will be argued that they lack correlatives features and that korotyj in such constructions was not a relative but an indefinite pronoun, like the Russian nekotoryj, koe-kakoj. The pseudo-correlatives of Middle Russian are the result of the intermediate stage of the process of the grammaticalization of kotoryj from indefinite to relative pronoun that caused the transformation of the compound constructions into the complex sentence.
Key words
correlatives, relatives, Middle Russian, grammaticalization, pseudo-correlatives.
2012 №1 Slovene
62 I
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
i. Introduction1
This paper is dedicated to syntactic constructions such as (1) that I will call pseudo-correlatives. They were observed in Old and Middle Russian starting from the first written texts of the 9th century and they died out at the beginning of the 18th century, although in Modern Russian they can occasionally be found in colloquial or dialectal speech:
(1) A kotoraja gsdr’ lošed poslanaja s nim <...> i ta lošed stala
v Volodimere.
and which master horse sent with him <...> and that horse stayed
in Vladimir
‘As for the horse that was sent with Stephan, that horse stayed in the city of Vladimir, master.’ (Gr 362)1 2
Pseudo-correlatives have been studied by many Russian linguists [Lom-TEV 1956; Sannikov 1965; Borkovsky 1979; Aksenova 1986 among many others]. They are traditionally described as follows: (a) the subordinate clause precedes the main clause; (b) the subordinate clause contains the relative/in-terrogative pronoun kotoryj ‘which’ with or without an NP3 and the main clause contains the demonstrative pronoun tot ‘that’ with or without the same NP or some other pronoun coreferential with the NP of the subordinate clause; and (c) the subordinate clause is preceded by the conjunction a (sometimes by i, da ‘and’ or no conjunction) and the main clause is preceded by the conjunction i (sometimes by a ‘and’ or in some cases by no conjunction4).
Thus, the basic schema of pseudo-correlatives is at (2), although it allows for some variation:
(2) [CorCP (subordinate clause) А kotoryj NPj...] j [ip (majn clause) i tot (NPi) •••]
[CorCP (subordinate clause) and which NPi*"] j [ip (main clause) and that (NPj) ...]
1 I am indebted to Ekaterina Lyutikova for her generous sharing of ideas during our numerous discussions of pseudo-correlatives. I am also grateful to Mihail Kopotev, Ahti Nikunlassi, and Dmitry Gerasimov for their helpful comments and suggestions. All errors are my own responsibility.
2 The following sources are used for the examples:
Gr: Грамотки XVII — нач. XVIII века, Москва, 1969
Kot: Котков С. И., Панкратова Н. П. Источники по истории русского народноразговорного языка XVII — начала XVIII века, Москва, 1964.
Mor: Хозяйство крупного феодала-крепостника XVII в., 1, 1933.
Mos: Московская деловая и бытовая письменность XVII века, Москва, 1968.
3 The question of whether Russian has DP or only NP is still an open problem, so I will generally use the term NP throughout the paper, but some linguists can take it as DP.
4 Svetlana Aksenova in [Aksenova 1986] has counted the conjunction used between two clauses in 257 cases of such constructions in the late period of Middle Russian (17th — beginning of the 18th centuries). In 183 (71%) of these cases, the conjunction i was used between the clauses and in 62 (24%) no conjunction was used.
Slovene 2012 №1
Olga Mitrenina
This schema is similar to a regular correlative construction that is widely used in Modern Russian [Mitrenina 2010]:
3) Kakuju mašinu uvidit, tu/takuju (mašinu) i prosit.
what car sees, that /such (car) and asks
‘Whatever car he sees, he asks for it.’
Correlative constructions such as (3) are classified as complex relative sentences with a preceding subordinate clause; the subordinate clause contains the relative phrase and the main clause contains the demonstrative phrase that is modified by the subordinate phrase [Bhatt 2003: 485-486].
Because of the similarity of these Middle Russian constructions (1) to regular correlative constructions (3), Rajesh Bhatt, following Edward Keenan, classifies them as correlatives [Bhatt 2003: 491]. Russian linguists have also traditionally classified such Middle Russian constructions as complex relative sentences with a preceding subordinate clause, although they have also noted that the relationship of subordination between the two clauses is weakened and both clauses are to a large extent independent [Borkovsky 1979: 59-60; Sumkina 1954: 177; Kačevskaja 1954: 212]. The reasons for considering the relationship of subordination between the two clauses to be weakened have traditionally been stated as follows: (a) the repetition of the same NP in the first and the second clauses; (b) the necessity of the demonstrative pronoun tot ‘that’ before the repeated NP that forms a sort of correlative connection with kotoryj in the preceding subordinate clause; and (c) the presence of the coordinating conjunction between two clauses.
None of these three reasons is enough to prove that the relationship of subordination between two clauses is weakened. The repetition of the same NP in two clauses is quite possible in correlative constructions such as (3) and the demonstrative pronoun is always needed in the main clause of regular correlatives [Bhatt 2003], so both of these features are typical for classical correlative constructions where the relationship of subordination between two clauses is not at all weakened. As for the functional word i ‘and’ between two clauses, it does not necessarily need to be a coordinating conjunction. There are two more functional words i: a so-called beginning (nacinatelnyj) conjunction i that was used in Old Russian to mark the beginning of the clause, and a particle i that is used in modern correlative constructions, as in example (3).
In the next section I will present some syntactic data showing that pseudo-correlatives differ from Modern Russian correlatives, pointing to the lack of subordination between the clauses of a pseudo-correlative.
I 63
2012 №1 Slovene
64 I
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
2. Structural Differences between Correlatives and Pseudo-correlatives
Pseudo-correlatives reveal a number of structural differences in comparison to modern Russian correlative constructions. In this section I will describe the most crucial of these differences. The data presented in this section proves that there was no strong correlative relationship between two clauses in pseudocorrelatives. All the examples are from Middle Russian (17th — beginning of the 18th centuries).
2.1. Third-person Pronouns instead of Demonstrative Pronouns
In pseudo-correlatives the demonstrative pronoun is used in most of the cases, although third-person pronouns can also be used in pseudo-correlatives in Middle Russian,5 as in (4); the demonstrative is required for the correlatives in Modern Russian [Bhatt 2003: 493]. The use of third-person pronouns is impossible in Modern Russian correlatives with the exception of few colloquial constructions [Lapteva 2003: 144].
(4) Da kotoryja lošedi i korovy prislany i ih" kormit’ nečim.
And which horses and cows sent and them feed nothing
‘As for the horses and cows that were sent here, there is no food for them.’ (Kot 41, Celishevy)
The third-person pronoun in Russian was historically a demonstrative pronoun and in Old Russian it was sometimes used as a demonstrative, but in Middle Russian it was not demonstrative but an anaphoric pronoun.
2.2. No Maximalizing Semantics Required
Mark de Vries (after Downing and others) defines as an implicational Universal G5 that correlatives have maximalizing semantics [Vries 2002: 38]. This is true for Modern Russian correlatives that can refer only to a unique individual or to a whole group. But pseudo-correlatives can refer to a part of some group; the word ‘many’ can be used in the second clause, a usage that violates Mark de Vries’s implicational universal:
(5) A kotorye de novye krest’janei prišli i u mnogih i dvory ne
and which PRT new peasants and came and at many and homesteads not
postavleny.
are.built
‘As for the new peasants who are said to have come, many of them have not even built their homesteads yet.’ (Mor 10)
5 According to [Aksenova 1986] in 257 cases of constructions with the preposition of the kotoryj-clause in Middle Russian (17th—beginning of the 18th centuries), third-person pronouns were used in 48 (19%) of them.
Slovene 2012 №1
Olga Mitrenina
l 65
And here is an example that Aksenova considered as the same type of construction [Aksenova 1986: 60]. The word “others” is used there in the second clause.
(6) Kotoryja narocetyja ljudi pap"latilisja a inym" i est" nečiva.
which best people paid and others and eat nothing
‘As for the best people, they have paid their debts, and the rest have nothing to eat.’ (Kot 43, Samariny)
This construction probably consists of three clauses: it includes a regular pseudo-correlative with the group ‘those people’ omitted in the second clause, and the third clause is inymi i est’ nečiva ‘and the rest have nothing to eat’; the English translation reflects this three-clausal structure very well.
2.3. The NP in the Main Clause can Differ from the NP in the Subordinate Clause
In some cases in pseudo-correlatives, the NP in the first clause with kotoryj is not the same as the NP in the second clause. These NPs are always semantically connected (in most cases the NP in the relative clause refers to a subclass of the NP in the main clause). Such usage is impossible in Modern Russian nor is it possible in other correlative constructions where only the same NP can be repeated [Bhatt 2003: 493].
(7) Da kotoruju ty gsdr’ izvol kupit belugu i toe gsdr’ rybu
and which you master pleased to.buy beluga and that master fish
poslana s seju ž ryboju.
sent with this same fish
‘As for the beluga (type of sturgeon) that you, Master, bought, that fish was sent together with this fish.’ (Gr 373).
(8) Kotoroj moj ržanoj hleb sejan byl na žreb’ju Alekseja Mescherinova
which my rye bread sowed was by lot of.Alexej Mescherinov
i on de tu rož’ požal i omolotil
and he PRT that rye reaped and threshed
‘As for the rye bread that I sowed by lot from Alexej Mescherinov, he has reaped and threshed that rye.’ (Mor 118)
(9) Kotoraja sol na Jung položena z gosudarevyh grebnyh
which salt on Jung put from masters’ rowing
strugov i tot anbar zapečatan <...> pečatju
boats and that barn sealed by.seal
‘As for the salt from masters’ rowing boats that was put at the place of Jung, that barn was sealed with a seal.’ (Gr 338)
2012 №1 Slovene
66 |
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
3. The Pronoun Kotoryj in Old and Middle Russian
In this section we will describe several other differences in behaviour of the pronoun kotoryj in Old, Middle, and Modern Russian.
3.1. Kotoryj-clause in Postposition
Middle Russian presents many examples of what Rajesh Bhatt calls English-type relative clauses [Bhatt 2003]. These are sentences such as (10) that are identical to English relative clauses:
(10) Volynskomu byt’ s prežnim polkom kotoroi u nego nne.
Volynsky should.be with previous regiment which at him now
‘Volynsky should stay with the same regiment which he has now.’
(Mos 14)
There was also another type of construction with kotoryj-clause in postposition used in Old and Middle Russian. These are the sentences with the same NP repeated in both clauses; the second NP is repeated right after the korotyj:6
(11) Prislali k nam <...> celoval’nuju zapis’, po kotoroj
(they) sent to us notary certificate, according to which
zapisi vy <...> krest celovali.
certificate you cross.ACC kissed.
‘They sent us a notary certificate, according to which you swore an oath by kissing the cross’. (Пам. ист. см. вр. [Lomtev 1956: 560])
Sentences such as (11) cannot be considered as headed relative clauses. Following Srivastav, Rajesh Bhatt describes structural differences between headed relative clauses and correlatives. He mentions that in relative clauses, the relativized NP cannot be repeated in the relative clause [Bhatt 2003: 492], see also [Vries 2002: 36]:
(12) *On kupil mašinu, kotoruju mašinu/Tojotu hotel.
*He bought the.car which car/Toyota he.wanted.
6 These constructions recall the unusual strategy of relativization in Old Russian (before the 15th century) described by Andrey Zaliznyak in [Zaliznyak 1980]. The relative pronoun was formed by adding the “relativizator” to or že to the interrogative pronoun: Posla Vsevolod" Svjatopolka <...> smolvjasja s nov’gorod’ci kotoryh to byl" prijal" (Киевская летопись по Ипатьевскому списку, л. 114 об.б from [Zaliznyak 1980: 98]).
Sent Vsevolod Svjatopolk talk with novgorodians which TO was accepted ‘Vsevolod sent Svjatopolk to talk to the people of Novgorod that he has accepted.’ These “relativizators” to or že disappear in Old Russian before the 15th century, but they were very popular in the oldest Russian texts of the 9th-13th centuries. The nature if these “relativizators” is not clear, but it is important to note that that word to was also used as a demonstrative in Old Russian and the need to use it with kotoryj as a “relativizator” may be related to the need to repeat the NP after the kotoryj in Old and Middle Russian. But definitely this phenomena needs to be studied more carefully.
Slovene 2012 №1
Olga Mitrenina
l 67
Such sentences with kotoryj in postposition and the same NP repeated in both clauses were used quite often in Middle Russian texts of the 15th — 16th centuries, although they were less frequent than the English-type relative clauses. But sentences with kotoryj in postposition were very rare in the early period of Russian (9th — 14th centuries) [Borkovsky 1979: 76, 82].
Borkovsky’s observations are supported by the remarkable and growing corpus of birchbark documents from Novgorod and its environs (11th — 15th centuries). They present several usages of the pronoun kotoryj, but the only sentence with kotoryj in postposition is presented in a late document dated to the beginning of the 15th century, illustrated in example (13), birchbark number 310 [Zaliznyak 2004: 670].
(13) ...ot Vavuly i ot tvoih" hrestijano kotorye hrestijani s Ylova prishli za tebja
.from Vavula and from your peasants those peasants from Ilovo came to you.
‘.from Vavula and your peasants from Ilovo that (peasants) came to you.’ or
‘.from Vavula and your peasants, those peasants came to you from Ilovo.’ [Zaliznyak 2004: 301]
So, the usage of kotoryj in postposition shows that the relative properties of this pronoun were expanding. Although they were expressed in only limited fashion in Old Russian, they began to achieve some power in Middle Russian, and, finally, in Modern Russian kotoryj is used as a regular relative pronoun. Thus, in constructions such as (11) and (13) in Old and Middle Russian, the pronoun kotoryj was used as a modifier of NP.
3.2. Grammaticalization of the Pronoun kotoryj
Grammaticalization is a historical process leading from lexemes to grammatical formatives; a sign is grammaticalized to the extent that it is devoid of concrete lexical meaning and takes part in obligatory grammatical rules [Lehmann 2002: VII].
The pronoun kotoryj reveals some of the criteria of grammaticalization described by Christian Lehmann [Lehmann 2002]. These criteria show that in Middle Russian the pronoun kotoryj was more lexical, while in Modern Russian the relative pronoun kotoryj is more grammaticalized, in the following ways:
a) A decrease in integrity. This includes a decrease in semantic integrity (desemanticization) as well as a decrease in phonological integrity (phonological attrition).
2012 №1 Slovene
68 I
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
The desemanticization of kotoryj is connected with the loss of the “which one of many” selective meaning. This meaning of the interrogative kotoryj occurs in Middle and Modern Russian,7 although the original meaning of the interrogative pronoun with its root was “which one of two” [Lomtev 1965: 558]:
(14) Kotoraja iz nih okažetsa samoj udačnoj?
Which of them will.turn.out most successful
‘Which one of them will turn out to be the most successful?’
The pseudo-correlative pronoun kotoryj in Middle Russian presents some of the meaning “which one of many,” which is why Kačevskaja suggests calling them not “relative”, but “selective-relative” pronouns [Kačevskaja 1954: 212]. The relative pronoun in Modern Russian is co-referenced with the relativized NP and it is not used with the meaning “which one of many.”
b) The decrease in phonological integrity is connected with the possibility of being stressed. The relative kotoryj in Modern Russian cannot bear phrase stress [Paducheva 1985: 121]. In Middle Russian the pseudo-correlative pronoun can bear phrase stress because it can be separated from its NP by intervening material. For example, in (1), the first phase before an address is A kotoraja, where kotoraja is the only word that can be stressed. In addition, in an example such as (5) we have a strong proof that kotoryj was stressed. In (5) kotoryj is followed by the clitic de that obeys Wackernagel’s Law, which requires clitics to appear in the so-called second position, after the first syntactic phrase or the first stressed word in a clause [Zaliznyak 2008]. It proves that kotoryj is stressed in such sentences.
b) A decrease in structural scope. The pronoun kotoryj in Old and Middle Russian was able to unite with an NP, as in example (11), whereas the relative pronoun kotoryj in Modern Russian usually cannot do this, as shown in example (12).
c) A decrease in syntagmatic variability, i.e., the decrease of the ease with which a word can be shifted around in its context. In Middle Russian when the pronoun kotoryj was used in postposition, it was also possible to use it at a distance and not immediately following its head noun, as shown in example
(15) . In Modern Russian such sentences sound awkward, because in Modern Russian the regular position of the relative kotoryj is right after the head NP.8
(15) Pismo vše ja polučil za kotoroe i blagodarstvuju. letter your I got for which thank.you
‘I have received your letter that I thank you for’
7 The popular Russian question Kotoryj čas? ‘What time is it now?’ (lit.: ‘Which hour?’) is an idiomatic construction that comes from Old Russian, its usage in Old Russian is not clear.
8 In the case of pied piping kotoryj can move to the right within the relative clause, but the relative clause still follows the head NP.
Slovene 2012 №1
Olga Mitrenina
l 69
This data shows that in Old and Middle Russian the pronoun kotoryj was in the process of grammaticalization and the result of this process was the Modern Russian relative pronoun kotoryj
4. The Evolution of Pseudo-Correlative Constructions
In this section I suggest an approach that explains the unusual properties of kotoryj in Old and Middle Russian. They can be explained if, in Old Russian, kotoryj was used mostly as an interrogative or indefinite pronoun. In Middle Russian kotoryj started to be used also as a relative pronoun, so all three possible usages of kotoryj were observed.9 In Modern Russian we can see only the relative kotoryj, with rare reminders of the old indefinite usage of kotoryj.
4.1. Transformation of the Compound Construction into the Complex Sentence
Most of the scholars who study pseudo-correlative constructions believe that they are transitional constructions between coordination and subordination. At first there were two independent clauses united by the conjunction i, which was either a coordinator or an element of the so-called chain threading (cepočečnoe nanizyvanie) that was a regular way to join sentences in Old Russian. Then the pronoun kotoryj began to participate in the syntactic relationship by acquiring some relative properties [Kačevskaja 1954]. Together with the demonstrative of the second clause, they formed a correlative conjunction that was used together with the coordinating conjunction i [Borkovskij 1979: 58-59].
Pseudo-correlatives recall one construction that is quite popular in Modern Russian colloquial speech. It is described in [Lapteva 2003: 144] as a nominative topic construction, modification number 5. The subordinate clause of this type includes kotoryj ‘which’ + NP that is the topic of the sentence. The main clause follows the subordinate clause and includes a personal pronoun (or sometimes a demonstrative pronoun) co-referential to the NP of the subordinate clause. The example of such sentence is in (16), from [Lapteva 2003]. Ekaterina Ljutikova also mentions such constructions as topicalized NPs with an anaphoric pronoun in the main clause [Ljutikova 2009].
(16) A kotorye rebjata, byli tam, očen’ nesladko im, vsem prišlos’.
And which guys were there, very tough to.them all fell
As for the guys that were there, they all had a hard time.
Aksenova in [Aksenova 1986] considered pseudo-correlatives as nominative topic constructions that have some similarity to relative clauses. That is
9 According to Lujan, *kwo- words have three main uses in Old Indo-European languages: interrogative, indefinite, and relative [Lujan 2009: 222].
2012 №1 Slovene
70 I
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
why they were not relative, but compound constructions that were later transformed into the complex sentence. In this case the question arises as to what happened to the coordinative conjunction i between two clauses.
In Modern Russian kotoryj is used only as an interrogative or relative pronoun. Correlative constructions with kotoryj similar to pseudo-correlatives are used in Modern Russian [Zaliznyak, Paducheva 1979; Lyutikova 2009], but they sound as syntactic archaisms, although all the other relative words can be used in Modern Russian correlatives [Mitrenina 2010]. I can suggest that the relative kotoryj does not produce correlatives in Modern Russian because it still preserves a trace of its original selective meaning. This meaning is not allowed in correlative constructions, which should always have maximalizing semantics, but it is allowed in pseudo-correlatives whose structure is close to the nominative topic.
4.2. The Functional Word i between two clauses
The study of pseudo-correlatives in Middle Russian shows that while the compound sentences were transforming into the relative construction, the coordi-native conjunction i was transforming into the particle i that is similar to the particle i10 used in modern correlative constructions, as in example (3). The particle i can have several meanings, but according to classification of Elena Uryson, the one used in modern correlatives is “the anaphoric particle referring to the mentioned situation” [Uryson 2011: 273-275]. For example sentence (17) presupposes that it was already mentioned in the previous part of the text that the man started to have problems at some point in his life. Moreover, in such context it is almost impossible to omit this particle i.
(17) V derevne i načalis’ ego nesčastja.
In village PRT started his misfortunes
‘It was in the village where his problems started.’
The only difference between the functional i between two clauses of the pseudo-correlative and the particle i in the Modern Russian correlative is the location. In correlatives i is usually located before the verb or before some other important member of the clause. In pseudo-correlatives i is usually located between the clauses. But we can also find examples of pseudocorrelatives where they are used simultaneously in both places: between the clauses and before the important member of the second clause (see example
(5) above, where the last i is used as in Modern Russian). There are even examples where i is used in the same way in which it is used in the modern correlative; that use is only before that part of the sentence referring to the mentioned situation:
10 Some attempts to study particles within the framework of generative grammar are summarized in [Haiden 2005].
Slovene 2012 №1
Olga Mitrenina
l 71
(18) I kotoroi sluga <...> delo <...> poterjal tovo slugu za tem"
And which servant documents lost that servant for that
delom" i prišlite hodit’.
documents PRT send to.go
‘Send the servant who lost that documents for it.’ (ГПБ, ОЛДП 128, п. II № 22 from [Aksenova 1986: 127])
Examples such as (5) and (18) show that the functional word i was not strictly confined to the place between the clauses but as the pseudo-correlative construction was transforming into the relative construction, the functional word i was shifting to the phrase that refers to the mentioned situation in the main clause.
4.3. Kotoryj as Indefinite Pronoun in Terms of Formal Semantics
If the Middle Russian kotoryj is an indefinite pronoun similar to the Modern Russian nekotoryj/nekotorye or koe-kakoj/koe-kakie, then it can have two different analyses in terms of formal semantics.
First of all, the indefinite NP can undergo existential closure in the first clause and therefore be referential; in this case, the demonstrative in the second clause is a definite description:
(19) A kotoraja lošed poslanaja i ta lošed stala v Volodimere.
and which horse sent and that horse stopped in Vladimir
‘As for the horse that was sent, that horse has stopped in the city of Vladimir.’
(= ‘Some horse was sent and it has stopped in the city of Vladimir.’)
(20) 3 x horse (x) a was_sent (x)
stopped_in_Vladimir (lx. horse (x) A was_sent (x))
Alternatively, pseudo-correlatives with the first part in present or future tense have the additional meaning of conditionals (see e.g. [Lomtev 1956: 560; Borkovsky 1973: 8]). Conditionals may contain a covert quantifier over situations (see [Lewis 1975; Kratzer 1986]). Both the variable corresponding to the indefinite DP with kotoryj as well as the demonstrative are bound by this quantifier, which ensures their coindexing in all the situations or worlds.
(21) A kotoraja šerst’ ne goditca v sukna i toe šerst’ peredelat’ v vojloki.
And which wool not good to cloth and that wool to-use to thick felts
‘Use for thick felts the wool that is not good for cloth.’ (=‘If some wool is not good for cloths then use it for thick felts.’)
(22) ALWAYS (s, x) [there is s which contains x such that wool (x) л not_good_for_cloth (x)] make felts of x in s
The difference between the two cases is how exactly the first clause restricts the demonstrative in the second clause.
2012 №1 Slovene
72 I
The Syntax of Pseudo-Correlative Constructions with the Pronoun Kotoryj ('Which') in Middle Russian
5. Conclusions
The pseudo-correlatives of Middle Russian are the result of the intermediate stage of the process of grammaticalization of kotoryj from indefinite to relative pronoun. This process caused the transformation of the compound constructions into the complex sentence; the coordinative conjunction i between the clauses was evolving into the particle i and moved to the phrase that refers to the mentioned situation in the main clause.
References
Aksenova 1986
Аксенова С. С., Полипредикативная конструкция с местоимением КОТОРЫЙ в русской речи XVII-XVIII вв.: дисс. канд. филол. наук, Томск.
Bhatt 2003
Bhatt R., “Locality in Correlatives”, Natural Language and Linguistic Theory, 21, 485-541. Borkovsky 1973
В. И. Борковский (ред.), Сравнительно-исторический синтаксис восточнославянских языков. Сложноподчиненные предложения, Москва.
Borkovsky 1979
В. И. Борковский (ред.), Историческая грамматика русского языка: Синтаксис, сложное предложение, Москва.
Haiden 2005
Haiden M., “Verb Particle Constructions”, in: M. Everaert, Henk C. van Riemsdijk (eds.), The Blackwell Companion to Syntax, 5, chapter 76, 344-375.
Kačevskaja 1954
Качевская Г. А., “К истории сложноподчиненных предложений с придаточным определительным”, Труды Института языкознания АН СССР, 5, 203-223.
Kratzer 1986
Kratzer A., “Conditionals”, Chicago Linguistics Society, 22/2, 1-15.
Lapteva 2003
Лаптева О. А., Русский разговорный синтаксис, Москва.
Lehmann 2002
Lehmann Ch., Thoughts on Grammaticalization, vol. 2 (revised edition) (= Arbeitspapiere des Seminars für Sprachwissenschaft der Universität Erfurt, No. 9), Erfurt.
Lewis 1975
Lewis D., “Adverbs of quantification”, in: E. L. Keenan (ed.), Formal Semantics of Natural Language, Cambridge, 3-15.
Lomtev 1956
Ломтев Т. П. Очерки по историческому синтаксису русского языка, Москва.
Slovene 2012 №1
Olga Mitrenina
l 73
Lujan 2009
Lujan E. R., “On the grammaticalization of *kwi-/*kwo- relative clauses in Proto-Indo-European”, in: V. Bubenik, J. Hewson, S. Rose (eds.), Grammatical change in Indo-European Languages, Amsterdam, 221-234.
Lyutikova 2009
Лютикова Е. А., “Относительные предложения с союзным словом который: общая характеристика и свойства передвижения”, в: К. Л. Киселева, В. А. Плунеян,
Е. В. Рахилина, С. Г. Татевосов (ред.). Корпусные исследования по русской грамматике. Москва, 436-511.
Mitrenina 2010
Mitrenina O., “Correlatives: Evidence from Russian”, in: G. Zybatow et al. (eds.), Formal Studies in Slavic Linguistics: Proceedings of Formal Description of Slavic Languages 7.5, Frankfurt am Main, 135-151.
Paducheva 1985
Падучева Е. В., Высказывание и его соотнесенность с действительностью, Москва. Sannikov 1965
Санников В. З., Сложноподчиненное предложение в древнерусских памятниках XI-XIVвв.: дисс. канд. филол. наук, Москва.
Sumkina 1954
Сумкина А. И., “К истории относительного подчинения в русском языке XIII-XVII вв.”, Труды института языкознания АН СССР, 5, 139-202.
Vries 2002
Vries, M. de, The Syntax of Relativization, Ph.D. dissertation, University of Amsterdam. Uryson 2011
Урысон Е. В., Опыт описания семантики союзов: лингвистические данные о деятельности сознания, Москва.
Zaliznyak, Paducheva 1979
Зализняк А. А., Падучева Е. В. “Синтаксические свойства местоимения который”, в: Категория определенности-неопределенности в славянских и балканских языках, Москва.
Zaliznyak 1981
Зализняк А. А., “Противопоставление относительных и вопросительных местоимений в древнерусском”, в: Балто-славянские исследования 1980, Москва, 89-107.
Zaliznyak 2004
Зализняк А. А., Древненовгородский диалект, Москва.
Zaliznyak 2008
Зализняк А. А., Древнерусские энклитики, Москва.
Ольга Владимировна Митренина, к.ф.н.
Санкт-Петербургский государственный университет, филологический факультет, кафедра математической лингвистики 199034 Санкт-Петербург, Университетская наб., д. 11 Russia / Россия mitrenina@gmail.com
