Deriving Discourse Configurationality of (East) Slavic

One of the most easily observable ways in which human languages differ from each other is their word order, or the order in which a sentences subject, object, verb and other syntactic constituents are arranged relative to each other. Identifying the principles underlying such cross-linguist ic differences is one of the tasks of modern linguistics. Another is to explain the relation between observed word orders of a language and the semantics they express. Free word order languages such as Ukrainian and Russian, in which permutation of syntactic constituents within a sentence is extremely common, along with other Slavic languages belong to the group of so-called discourse configurational languages. These are languages where permutations in word order bring about changes in the sentences information packaging, encoding such notions as topicality, givenness, specificity or presuppositionality. A curious challenge for linguistic theory is linked to the well-known empirical observation that such changes in word order, despite being important for encoding information structure- relevant notions, are nevertheless optional, i.e., the same semantics can be expressed without permuting the word order of a sentence. This latter property, known as the optionality of syntactic movement, which has been especially difficult to capture in formal linguistic frameworks, is the focus of the present project, which hypothesizes several reasons for such optionality. We hypothesize that for various word orders in Ukrainian and Russian, which we concentrate on, there is an additional means of signaling the same information structural notions, namely prosody, or intonation. In other words, if this hypothesis is correct, there may not be true optionality of syntactic movement to speak of, there is only a choice for the speaker regarding how to encode the intended semantics, via syntactic means (permutation of word order), or via prosodic means (adjustment of prosody). That this hypothesis may be correct is suggested by recent work on Ukrainian Object Shift, a syntactic operation that obligatorily encodes partitivity/specificity semantics for the shifted object, where prosodic encoding is shown to function as an alternative to the syntactic operation of Object Shift. Another hypothesis that could account for at least some of the empirical data is that syntactic labels such as SVO, SOV, OVS, VSO (which are used to mark different relative ordering of Subject, Object and Verb as the main sentence constituents) are non- unique identifiers, meaning that behind each label there could be multiple or at least non-unique mappings of syntactic structure to prosodic contour and information structure. Finally, we hypothesize that two syntactic operations, Argument Inversion and Object Shift, identified through earlier work on East Slavic, constitute the syntactic core of the permuted word orders in question, thus to a large extent accounting for the greater flexibility of word order in these and other discourse configurational Slavic languages that set them apart from fixed word order languages such as English. The findings and deliverables of the project are expected to have important implications not only for the rest of the Slavic language family, but also for languages with the more restricted freedom of word order permutations, by improving our understanding of the mechanisms underpinning word order permutability and their restrictions.

The project proposed a number of hypotheses in order to address the optionality-of-syntactic- movement problem and to probe the property of Discourse Configurationality. Among them is the possibility that the Slavic languages differ from the rigid word order languages in terms of having a more articulated phrase structure, i.e., more landing sites as targets for syntactic movement. Relatedly, we hypothesized that labels such as SOV, by which we mean a particular linearization of Subject, Object, and Verb, are cover terms for a number of constructions with this linearization, but with constituents occupying different structural positions in the syntactic tree. Thus, we hypothesized that in Ukrainian and Russian, the SOV linearization corresponds to more than one construction, which can be teased apart by pairing the linearization with a distinct interpretation and prosodic contour. The project investigated these and other related hypotheses via a combination of theoretical, empirical corpus and experimental prosodic research. Theoretical work was assisted by a quantificational diagnostic test, the Scope Freezing Diagnostic, developed by the PI and refined in the course of the project. The novel empirical evidence contributed by the diagnostic provided strong support for the Split Voice view of Ukrainian, according to which the phrase structure includes a causative projection, vCauseP. This strand of research helps us better understand the syntactic implications of the Split Voice architecture, yielding the finding that the causative head comes in two varieties, vCauseTR(ANSITIVE) and vCauseIND(IRECT). A strand of research concerned with word order and argument structure has provided evidence for a (raising) Applicative projection, ApplP, in the clausal spine of East Slavic. This head is shown to have phasal properties, and argued to be responsible for the greater permutability of word order in the postverbal field in East Slavic. Corpus research into the SOV word order has yielded a number of results. First, it provided independent support for the existence of Object Shift in Ukrainian. In the microvariation perspective on SOV, a novel finding we report is that Ukrainian allows a type of object movement operation familiar from Russian ("salient objects"), previously not posited for Ukrainian. Furthermore, we show that Ukrainian allows bare nouns to raise to the preverbal position, while Russian does not. This research has thus yielded a confirmation of our hypothesis that labels such as SOV are cover terms. Taken together, these findings suggest that the greater flexibility of word order in the postverbal field in (East) Slavic is indeed due to the more developed phrase structure. The microvariation in preverbal object movement discovered in our study, on the other hand, preliminarily suggests differences between Ukrainian and Russian in relying on the prosodic recontouring strategy for deriving the same interpretation.