B4

Our main goal is the modeling of a more fine-grained typology of non-default common ground (CG) updates by studying mirativity in Romance from a comparative and diachronic perspective, thus addressing the general research question QB of Area B (Grammar): How are linguistic devices put to use to perform non-default CG updates?

In Romance, mirativity – i.e., the linguistic encoding of the fact that an information is new or unexpected to the speaker (DeLancy 1997) – ‘hooks up’ to other grammatical and pragmatic categories. It is expressed via a wide array of linguistic devices (mirative markers, MMs), which interact with temporality-aspectuality, expressiveness, negation, mitigation and illocutionary mood and may correspond to different types of (non-)at-issue meaning. MMs express that an incoming or salient proposition p is unexpected, surprising or exceeds previous expectations. MMs have been characterized as triggering non-default updates as they expand the CG to “worlds that interlocutors have previously ruled out due to perceived outlandishness” (Beltrama & Hanink 2019: 1). They thus go beyond the Stalnakerian (1978) CG model, which recognizes that asserting a proposition may contradict what is in the CG, but which does not account for various types of non-default updates (e.g., Farkas & Bruce 2010; Rett & Murray 2013; Murray 2014; Rett 2021).

Combining a range of synchronic and diachronic corpus analyses with acceptability judgment experiments, we focus on selected MMs in Romance and address two main research questions:

(RQ1)   What are the defining features of mirativity?

(RQ2)  How do MMs develop and are there systematic, crosslinguistic correlations between the formal source of Romance MMs, their diachronic grammaticalization / pragmaticalization paths, and the specific mirative CG updates that they trigger?

Based on previous research and pilot studies (Dessì Schmid, Momma & Wiesinger 2025), our project’s first work package addresses RQ1. It challenges the received view that only (i) speaker-orientedness, and (ii) the recency restriction (cf. Rett & Murray 2013) are the defining features of mirativity and offers instead a more fine-grained typology of mirativity, for which a differentiated analysis of (un)expectedness as an epistemic andsocial stance as well as the question under what condition and for whom it is expressed play a crucial role. We will also argue that specific Romance MMs show different degrees of polyfunctionality with regard to different types of CG updates. Our project’s second work package addresses RQ2. For the study of the grammaticalization and pragmaticalization paths of MMs, the diachronically well-documented Romance languages provide an ideal comparative testing ground.