Participating researchers: Anke Holler , Markus Steinbach & Hedde Zeijlstra

The project will investigate the emergence of (partly) iconic spatial agreement, classifier or comparative constructions in sign languages and co-speech gestures and compare the results to the emergence of corresponding arbitrary constructions in spoken languages. Since the spatial constructions of many established sign languages combine arbitrary morphosyntactic with transparent (iconic) semantic properties, we expect a combination of language-specific linguistic and general gestural restrictions to constrain the emergence of spatial constructions in the visual modality.

Participating researchers: Marco Coniglio & Stavros Skopeteas

The expression of interrogatives with verb fronting in polar questions is an areal property of the European languages. This project will account for the diachronic development of this syntactic operation in three language families (Germanic, Romance, Greek), in which verb fronting is optional for questions in earlier stages and develops to an obligatory operation during language change. This project will integrate the phenomenon of verb fronting for clause-typing to the larger picture of V-fronting operations in these languages (also including the role of information structure and the properties of wh- questions).

Participating researchers: Nivedita Mani & Markus Steinbach

Most approaches to language assume arbitrariness in the extent to which form and meaning overlap in the lexicon. Thus, typically, there is no intrinsic relationship between the form and the meaning of the word. However, recent models of language learning highlight how children can leverage their existing lexical knowledge of how words sound or what words mean to acquire new similar-sounding or similar-meaning words. This project will examine the extent to which form-meaning overlap in the early lexicon guides early word learning and lexical processing in a series of eye-tracking studies with young infants and children.

Participating researchers: Marco Coniglio & Götz Keydana

Hyperbata are showcase examples of many-to-one relations between (surface) form and meaning. Especially in early Indo-European languages, hyperbata are more frequent than in later stages (Lühr 2016) and appear to be able to trigger Left Branch Extraction (LBE):

(1) […] ut meus victor vir belli
that my:NOM conqueror:NOM husband:NOM war:GEN
clueat
be.called:PRS.SUBJV.3SG
‘[…] that my husband shall be hailed the conqueror in the war.’ (Plaut.Am.2.2.16)

Depending on the language(s) investigated, we intend to focus on the following questions:

• Is LBE really a case of extraction, or is it base-generated?
• What are the conditions triggering LBE that were lost diachronically? Why do languages with massively restricted hyperbata end up with the patterns currently attested?
• How does information structure influence LBE? How and why does this interaction change?
• How do these factors change and how are they related to other syntactic changes, e.g. shifts in the distribution of subject pro, change from OV to VO word orders etc.?
• What are the consequences of LBE and its loss for syntactic theory?

The projects aims to define the properties of LBE in hyperbata and to determine the factors leading to its gradual loss in selected languages.

Participating researchers: Uwe Junghanns & Thomas Weskott

We start from the observation that languages vary with respect to whether wh-words can or have to remain in situ, or whether they can or have to move to the left periphery. Languages with relatively free word order like, e.g., Slavic languages, take middle ground here in that the placement of wh-elements seems to be underdetermined by purely syntactic factors, as witnessed by the lack of superiority, or weak cross-over effects. For a double wh-question in (e.g., who ... what?) there are, in principle, four possible linearizations corresponding to one core meaning. This raises the question what exactly the additional (semantic and/or pragmatic) meaning contributed by the different orders may be: scope properties, information structural properties, echo interpretations, etc.
The project aims to investigate these additional contributions by means of controlled acceptability judgment studies; these off-line studies will be accompanied by on-line studies of processing difficulty (e.g., reading times) to target the processing preferences associated with the different meanings.

Participating researchers: Marco Coniglio & Hedde Zeijlstra

Many languages exhibit null subjects, but it is still not clear what properties exactly license such null subjects. Traditionally, a connection with rich verbal agreement has been assumed, but that turns to be at least insufficient. This project will investigate other factors that may play a role in the licensing of null subjects.

Participating researchers: Götz Keydana & Guido Mensching

Second position pronominal (2P) clitics are a clear case of a form-meaning mismatch. In this project, the focus is on the development of 2P clitics over time. Depending on the language(s) under investigation we want to look into various aspects of the diachrony of 2P clitics:

• their emergence or their loss,
• changes in the domain of placement (from sentence to smaller domains like the VP, e.g. in Tocharian, Lithuanian and Old Irish, the latter two with further grammaticalizaiton into affixes),
• changes in placement within the domain (as e.g. in the development of Latin weak enclitic pronouns into Romance proclitics),
• extensions in clitic placement like clitic climbing (e.g. Romance),
• changes in the nature of clitic placement, e.g. from prosodic to syntactic placement or vice versa (like in some Slavic languages),
• changes in possible hosts, e.g. from word to constituent (again as in some Slavic languages),
• changes in clitic templates (i.e. the number and order of pronominal clitics admissible in a given language, e.g. in the Tocharian languages),
• the emergence of clitic doubling (e.g. some Slavic languages, Italian, Spanisch),
• the grammaticalization of 2P clitics into affixes (e.g. Old Irish, Lithuanian, see above).

The aim of the project is to determine the factors triggering changes in clitisization and to embed these changes in a broader picture of syntactic change.

Participating researchers: Anke Holler , Markus Steinbach & Nina-Kristin Meister

This project will build on the results of a previous project on imperatives and imperative speech acts in sign languages and broadens the view in two dimensions: (i) the new project will include other sentence types and speech acts such as biased questions and exclamatives and (ii) it will compare IFIDs in sign languages with corresponding manual and nonmanual gestural markers in spoken languages.

Participating researchers: Uwe Junghanns & Clemens Steiner-Mayr

In natural languages there is a connection between questions and conditionals. This is evident in languages where the two types of clauses share material as, e.g., English if or Czech li. The ties lead to questions on (i) the status of the involved items (conjunction or particle), (ii) their meaning (interrogative or conditional), and (iii) compositional semantics. Further issues arise with respect to restrictions on distribution (root and indirect questions, conditional clauses). Both single-language and cross-linguistic approaches may be chosen for an investigation on the topic of questions and conditionals. Apart from semantics, syntactic issues constitute another focus. At least for some languages, insights may be obtained by considering diachronic facts. The topic offers a range of options as to mismatches between form and meaning as well as to empirical domains (synchronic data, language variation, and language change).

Participating researchers: Guido Mensching & Stavros Skopeteas

This is a follow-up project on presentational particles in French and Italian. An initial observation is that presentational particles can have diverse diachronic sources, such as verbal forms, demonstratives, and locative expressions (and often combinations thereof). A first and very rough working hypothesis is that despite the multiple origins, such particles develop a similar syntax and similar meanings. The project will investigate such particles in a large number of languages of different families.

Participating researchers: Clemens Steiner-Mayr & Thomas Weskott

This project will investigate clausal constructions that express an exception to a predication from a previous utterance, as in Peter watered all the plants, but he didn't water the cacti. Exceptions of this type can be expressed using various forms, while the underlying semantics is rather uniform and seems to be dependent on the ordering of the propositions (cf. #Peter didn't water the cacti, but he watered all the plants). The project aims at (i) providing a (language specific, or cross-linguistic) description of the forms that participate in clausal exceptive constructions; (ii) the intrasentential semantic building blocks of the constructions; and (iii) account for the intersentential restrictions the constructions seem to have.