In two recent studies Krifka (2019, 2020) has postulated the hypothesis that assertive clauses involve propositions, judgements, commitments and acts. These different layers correspond to Tense Phrase, Judgment Phrase, Commitment Phrase, and Act Phrase. We will investigate whether the last three layers are constitutive of all assertions: not only those that profile some aspect or aspects of the assertive act itself but also those that merely convey/introduce an informative update. We will investigate whether questions and exclamatives also require the three layers postulated for assertions. We will investigate the role of left peripheral first person singular strong pronouns in prodrop languages such as Catalan, Spanish, European Portuguese, Italian and Greek (e.g., Spanish yo me parece que tenemos Covid para tiempo ‘It seems to me that Covid will be among us for quite a while’; yo me duele que estés triste ‘It upsets me that you are sad.’), and will hypothesize that experiencers allow different profilings (as speakers, committers, and judgers), depending on the type of verb (psychological predicates), the type of sentence (impersonal sentences with and without datives), and access to commitment states, commitment spaces, and their development in a dynamic semantics. We will also investigate the use of 1st and 2nd person pronouns in dislocations, in comparison to vocatives; the expression of emotions through language (Trotzke 2017, 2019; Trotzke & Villalba 2019, 2020, forthcoming), by means of function words (interjections, intensifiers) in languages such as English and German; and a multimodal dimension that combines language and gesture.

This specific goal will be devoted to investigating whether a unifying account of expletiveness in terms of expressivity is possible following a Speech Act and a Commitment-based Semantics approach. We will focus on the expressive role of polydefinites, plural mass nouns, expletive negation both in root clauses and in subordinate clauses, expletive voice (Schäfer 2008), expletive coordination (Poletto 2005), and expletiveness in clitics (Cuervo 2003). We will argue in support of the hypothesis that at the time of communicating an utterance that contains any of these categories the speaker performs an expressive speech act by which (s)he publicly commits to an emotive stance towards the content of a proposition φ. This emotive stance is analysed as a proposition ψ that is directly transferred from the immediate context of utterance into the speaker and addressee’s common ground and, subsequently, their shared commitment space. To take an example, in the case of Greek plural mass nouns, ψ appears to convey the expressive meaning that the speaker feels dislike with respect to the situation described by the proposition φ. We aim to investigate whether this analysis can be extended to achieve a better understanding of phenomena such as so-called expletive negation in natural languages such as Greek and Catalan (Espinal 1992, 2000). This study will be contrasted with recent work on expletive negation in Italian (Delfitto & Fiorin 2014, Greco 2020a, Delfitto 2020).

We will focus on presumptive questions and expressive meanings associated with response particles (Holmberg & Wu 2020), as well as on rhetorical questions (Farkas 2020) and rhetorical exclamatives. What is the syntax of high negation (Ladd 1981, Romero & Han 2004, among others) and how does it connect to the speaker’s expectations? Goodhue (2019) has recently suggested that in English high negation –which in questions biases the speaker towards the proposition embedded under high negation– does not contribute propositional negation, with the bias resulting from high negation scoping over an epistemic operator. We would like to explore whether there is comparable evidence for this approach to be maintained in other languages as well. Neg-raising (Fillmore 1963 and ff.) has been widely addressed in the syntax and semantics literature (Fillmore 1963, Lakoff 1969, Horn 1971, 1972, Seuren 1974, McCawley 1998, Collins & Postal 2014, 2017), but even the most recent accounts raise some theoretical concerns yet to be solved, such as the stipulation of multiple negative heads across clauses in the syntax that are subject to a dedicated deletion rule so that negation is interpreted only once in the semantics. It is the aim of this specific goal to explore the compositionality of this phenomenon to gain insight on the nature of the functional category Neg(ation) and its interaction with certain predicates. Concerning English tag questions, these have been extensively described and analysed in the literature (Quirk et al. 1985, Cattell 1973, McCawley 1988/1989 among others), but it is yet unclear how exactly they are derived syntactically, and how the polarity-reversal mechanism –in those that involve reversing the polarity of the antecedent clause– can be characterised and formalised. In addition, it is standardly assumed that prosody and syntax interact at the time of assigning question tags different expressive meanings. We will try to understand such a connection better by exploring it experimentally in the grammar of both native and non-native speakers (e.g., Catalan/Spanish bilinguals acquiring English). Concerning rhetorical questions and rhetorical exclamatives, we aim to investigate how their discourse contribution to common ground as assertions combines with their expressive import, and whether the latter is encoded in a functional layer responsible for (i) expressing subjective commitments (Biezma & Rawlins 2017), (ii) licensing expletive negation and/or NPIs (Han 2002, Andueza, 2011), and (iii) triggering a special prosody (Dehé & Braun 2019, Braun et al. 2019). We will also focus on rhetorical questions that occur in Latin (i.e., those introduced by Nonne and Numne; (Ernout & Thomas 1953; Greco 2020b).

Syntactic grammaticalization processes usually go side-by-side with loss of semantic content. Grammaticalization processes of the lexical category verb have been usually related to the emergence of auxiliary verbs and  omplementizers (and, therefore, to the emergence of functional categories such as Aux and C). When lexical verbs cease to be members of the category V, they lose conceptual content and acquire procedural meaning (Cuenca & Marín 2000; Traugott 2010, 2014; Escandell-Vidal et al. 2011; Tanghe 2016, a.o.). However, lexical verbs may turn out into deverbal discourse markers that do not routinely restrict the logical relationships between accessible propositions, that is, they do not introduce constraints on implicatures (Sperber and Wilson 1986/1995, Wilson and Sperber 1993), but rather usually introduce restrictions on the judgements of the speaker, and/or the listener, regarding the propositional content asserted. Consequently, we will study the expressive meaning associated with deverbal discourse markers in Catalan and other Romance languages, devoting special attention to the Spanish discourse marker vaya (lit. go) in two constructions: in combination with an indefinite DP (vaya un lío ‘what a mess’, vaya un huerto ‘what a garden’), and in combination with a P con ‘with’ + definite DP (vaya con la niña ‘that damn girl’, vaya con el Manneken pis ‘that damn Manneken pis’). We will put forward the hypothesis that deverbal discourse markers, which introduce either expressive subjectivization or intersubjectivization meanings, support an analysis that postulates different speech act layers in the force field (Krifka 2015, 2017, 2019, 2020), and various construals of these speech act layers. 

The question of what (in)definiteness means and how it is expressed in languages that do not possess overt articles still presents an unresolved problem in formal approaches. Based on our previous research on definiteness in Russian (as a representative language without articles) (see Seres 2020, Seres & Borik to appear, Borik et al. to appear, Seres et al. 2019) definiteness in articleless languages cannot be straightforwardly associated with uniqueness in a strict semantic sense and hence cannot be accounted by uniqueness-based theories of definiteness (Frege 1879; Elbourne 2005, 2013; Heim 2011; Dayal 2018). Seres (2020) and Seres & Borik (in press) argue that the default interpretation of bare nominals in Russian is indefinite, whereas definiteness results from a pragmatic strengthening mechanism. This hypothesis is corroborated by existing (although still scarce) experimental evidence (see Borik, Borràs-Comes & Seres 2020; Šimík & Demian 2020). This project will investigate in detail what kind of mechanisms give rise to this pragmatic strengthening and how precisely they bring out a definite interpretation for a bare nominal. The factors that might be of interest are the following: anaphoricity, as it establishes co-reference with a previously introduced discourse entity and hence, indirectly, enforces unique/maximal reference for a bare NP; ontological uniqueness (i.e. existence of one single object in the actual world, satisfying a given description); topicality, due to its strong links with definiteness cross-linguistically. Moreover, if definiteness is indeed a largely pragmatic phenomenon in (at least some) languages without articles, other pragmatic effects associated with a definite interpretation can possibly be discovered that have not been noticed before. Thus, our main goal will be to determine, describe and analyze those pragmatic effects that are linked to a definite interpretation of bare nominals in Russian and possibly other languages without articles. Furthermore, if indefiniteness is the basic reading in languages without articles, what repercussions will this hypothesis have for the syntax of an indefinite NP? And how is the indefinite interpretation derived in terms of compositional semantics? Can a hierarchy of interpretations be proposed for languages without articles, e.g., indefinite non-specific (as default) > specific / definite? And can it be comparable to languages without articles? Another relevant question is the relationship between definiteness and expressivity in languages with and without articles, with a special reference to Latin and Russian. In languages with articles, expressivity may be encoded by determiner spreading / by the expletive definite article. What are the means for expressing the same meaning in languages without articles? 

We will explore a new, unified, syntactically-driven approach to the compositionality of indefiniteness, (anti-)specificity and partitivity in Romance (Catalan, Spanish, French, Italian and Brazilian Portuguese), and its potential extension to Basque. As for Romance, we will explore, following previous work by Cyrino and Espinal (2020), the idea that the availability of an indefinite reading on plural and mass nouns might be accounted for by postulating an abstract operator DE that applies to definite determiners. This operator can be overtly or covertly instantiated (des/de N, dei/di N, and bare plurals/mass nouns). DE-phrases can merge with a quantifier that is inherently specific (e.g., certains, unos ‘certain, some’) or non-specific (e.g., plusieurs, varios ‘some’), with the result that it turns a property into a generalized quantifier. Anti-specificity would be derived by adjoining an abstract operator ALG to a quantifier inherently encoding specificity, with the result that a modified generalized quantifier is obtained: ALG shifts its meaning by introducing referential vagueness (e.g., quelques, algunos ‘some’) (Etxeberria & Giannakidou 2017). This analysis will be compared to the analysis of epistemic indefinites postulated by Gianollo (2018, 2020). Partitivity would be derived by merging a specific/non-specific/anti-specific QP (denoting a subset; i.e., a part) in the specifier position of a bi-relational partitive RELATOR head (spelled-out by de in all the languages we investigate) that takes a definite DP (denoting a set; i.e., the whole) in complement position. If confirmed, the output of this analysis would be an indefiniteness hierarchy that is syntactically grounded. We will also investigate whether the existential interpretation of the DP in Basque can be explained following Etxeberria (2005, 2010, 2012), or rather following Etxeberria (2014), by postulating that the definite article is a null D that allows the existential interpretation of direct objects in Old Basque.

In this specific goal we will address the question of how interrogatives and exclamatives restrict or widen the speaker/addressee’s commitments. The influential work by Krifka (2001, 2004, 2011) on the topic of questions, as well as his recent work on Commitment-based Semantics (Krifka 2016, 2017, 2019, 2020; see also Geurts 2019) has paved the way for a better understanding of the role of interrogative sentences at the time of different sorts of updates of the common ground, and involving various sorts of commitments of speakers and a variety of forms to restrict future discourse developments. However, such seminal work has not yet been extended to provide a detailed analysis of the full gamut of interrogative sentences in Romance and Germanic languages, where interrogatives display a rich set of syntactic forms, interpretative nuances and prosodic contours. We aim at investigating how different interrogatives should be encoded in the speech act domain, for this will offer us a better explanation for the expressive content linked to noncanonical interrogatives. Apart from interrogatives, we will study the contribution of exclamatives to discourse interaction and speaker/hearer’s commitments (Trotzke & Villalba 2000, forthcoming) in German, Spanish and Catalan. To do so, against received common wisdom, we will focus on the fact that exclamatives can be used as responses to questions in certain contexts, and will investigate the hypothesis that exclamatives, when used as responses, endorse a commitment with respect to a previous speech act. Consequently, we will argue for the hypothesis that exclamatives, at this level, instantiate functional categories at both information structure and the speech act level of the clause.