Expletives, referential pronouns and pro-drop: The Russian extraposition pronoun èto in light of the English it and the German es
Introduction
This paper examines the Russian propoun èto (‘it’, ‘this’) used in a construction similar to extraposition construction in the Germanic languages.1 This type of construction is illustrated by the examples in (1) for Russian, English and German, respectively2:
The correlates it and es that refer to the extraposed embedded clause are often assumed to be expletive. I will show that èto has striking similarities with it and es, which suggests that èto is also an expletive in some of its uses.3
What is unexpected about such an assumption is that the existence of an expletive pronoun in extraposition construction in English and German is commonly associated with the fact that both English and German have no pro-drop, that is, do not allow subjectless sentences. Now, Russian is assumed to be a partial null subject language (Franks, 1995, Roberts and Holmberg, 2010): Russian does have pro-drop, though pro-drop in Russian depends on discourse conditions that make it much less regular than, for instance, in Italian. See an example of zero anaphora in Russian, which is not due to syntactic ellipsis, in (2):
In extraposition construction, the pronoun èto can often be omitted. Cf. the examples (a) and (b) in (3) and (4):
This fits perfectly the analysis of Russian as a pro-drop language. In some other cases, however, èto is obligatory in extraposition construction, see (5), and this, by turn, is in line with characterizing pro-drop in Russian as partial.
Given the restricted use of zero anaphora in Russian, the existence of an expletive pronoun in Russian is not that surprising. Still, Russian differs significantly from English and German w.r.t. zero anaphora. Therefore, the fact that the formal features of èto are very similar to those of it and es requires an explanation.
It should be noted that the aim stated above – comparing extraposition constructions in Russian, English and German – is by itself not unproblematic, since what is compared is of different nature. In English and German, the presence of a pronoun in extraposition construction (at least a pronoun in the subject position) is obligatory, so the structure with a pronoun can be said to be an unmarked one. While in Russian, the structure that can be argued to be unmarked is the one without a pronoun, since the use of the pronoun, as I will show below, is subject to specific constraints.4
However, if one wants to find an explanation to the fact that èto behaves similar to it and es, this structural difference can be ignored. Moreover, given the structural difference, the similarities observed seem to be of particular interest, since they suggest that there are universal features shared by expletives of different types.
In what follows, I will propose an analysis based on the inherent properties of an expletive item. It is generally assumed that expletives are semantically empty or at least semantically impoverished (see, for instance, Svenonius, 2002), that they are non-arguments or at least quasi-arguments (borrowing the term from Chomsky, 1981). I will suggest that the formal features èto has in common with it and es follow automatically from èto being an expletive (that is, being semantically impoverished, etc.). The typological prediction of this account is that expletives in other languages are expected to display at least some of the same features.
A comment is needed concerning the theoretical framework of this paper. The reasoning I use below to elaborate my analysis is characteristic of the functionalist approach. However, a huge amount of work on expletives has been done within the generative grammar, so as far as possible, the results of this work are taken into account. This does not mean, however, that I proceed from the theoretical base of the generative tradition. In particalur, with the term pro-drop I refer here and below to the general phenomenon in which the subject NP may be unexpressed, rather then to the precise theoretical background of this term within the generative framework.
The rest of this paper is organized as follows. In Section 2, I review the main assumptions that can be found in the literature on extraposition it and es. In Section 3, I discuss the formal properties of the Russian èto, including those that èto has in common with it and es. I suggest on the basis of these properties that two types of èto have to be distinguished: an expletive and a referential èto. In Section 4, I summarize the parameters according to which the expletive and the referential èto differ and propose a kind of functionalist account for each parameter. More precisely, I argue that there is a link between the nature of an expletive, on the one hand, and the respective values of the relevant parameters, on the other. In Section 5, I specify the way in which different parameters interact when determining the categorial status of èto. Finally, in Section 6 I argue that the expletive èto is still less expletive than it and es in the subject position, thus suggesting “expletiveness” to be a scalar notion. Section 7 presents the main results of the study.
Section snippets
About the English it and the German es
In this section, I will briefly summarize what has been said on the extraposition it and es in the literature. In Section 2.1, I will review the proposals concerning the categorial status of it and es. In Section 2.2, I will consider the formal properties of it and es that are particularly important for my analysis of the Russian èto.
General information
The word èto (genetically NOM/ACC.SG.N of the demonstrative pronoun ètot) can be used in Russian as a pronoun (30), as a particle (31) and as a copula (32)7
Parameters that determine the status of èto: explaining their link to expletiveness
In the previous Section, I have suggested that there are five parameters relevant for the status of èto (expletive vs. referential) referring to an embedded clause. These are the following parameters:
- •
èto is preposed vs. postposed to the embedded clause;
- •
èto refers to new vs. given information;
- •
èto is marked for nominative vs. for an oblique case;
- •
the matrix predicate is impersonal vs. personal;
- •
èto refers to a complement vs. an adverbial clause.
The first value of each parameter (“preposed”, “new”,
Relevant parameters as a hierarchy
As I have shown in the previous two Sections, the categorial status of èto is determined by at least five parameters. I repeat them below:
- •
èto is preposed vs. postposed to the embedded clause;
- •
èto refers to new vs. given information;
- •
èto is marked for nominative vs. for an oblique case;
- •
the matrix predicate is impersonal vs. personal;
- •
èto refers to a complement vs. an adverbial clause.
An important question is whether and how these parameters interact, that is, what the general mechanism is according
A scale of expletiveness: is èto as expletive as it and es are?
I hope to have proved in the previous Sections that the Russian pronoun èto can be called expletive in some of its uses. A question arises, however, whether èto is expletive in the same sense or to the same extent the English it and the German es are. I suggest the answer to be negative: the subject it and es, which are the most indisputable instances of the expletive it and es, seem to be “more” expletive than the Russian èto. Below I will indicate a few facts that give evidence in favor of
Conclusions
In this paper I have tried to show that the Russian pronoun èto used in extraposition construction has important similiraties with the English it and the German es. Like the extraposition it and es, èto can refer to new information. The embedded clause cannot be questioned if it is preceded by èto, which, again, is parallel to how it and es behave. Based on these and similar facts, I argued that two types of èto have to be distinguished, an expletive and a referential one. I examined the
Acknowledgements
I am thankful to the anonymous Lingua reviewers, who provided a number of insightful corrections and comments. All errors are mine.
This research was supported by the Russian Foundation for Fundamental Research (grant no. 16-06-00226).
References (44)
- et al.
Cataphora, backgrounding and accessibility in discourse
J. Pragmat.
(2016) Two types of coordination in clause combining
Lingua
(2005)Systematic mismatches: Coordination and subordination at three levels of grammar
Journal of Linguistics
(2015)- et al.
Argument clauses and correlative es in German – deriving discourse properties in a unification analysis
- et al.
Why languages don’t like expletives
- et al.
Information structure and non-canonical syntax
Meaning and Form. English Language Series, 11
(1977)- et al.
Interclausal Cataphora in English
(1999) Impersonal Constructions and Sentential Arguments in German
(1990)Lectures on Government and Binding
(1981)