Below I describe current and past strands of my research program.
universals and variation in the evidential domain
Evidentials are expressions that encode the source of semantically determined information for an utterance, such as English reportedly or allegedly. Over the years, I have worked on grammatical evidentials in several languages (Bulgarian, Georgian, German, Turkish).
An ongoing debate in this area concerns the relation between evidentiality and epistemic modality. According to one line of research, evidentials are garden variety epistemic modals. According to another, evidentials interact with the structure of speech acts. By discussing analytical options proposed for evidentials, I show that the debate is lacking formally-explicit tools that would differentiate between the two classes. Current theories, even though motivated by superficially different data, make in fact very similar predictions. I reduce the cases of apparent semantic variation to factors independent from evidentiality, such as the syntax of clausal complementation, and show that these cases do not resolve the modal-illocutionary debate.
Based on data from a range of languages, I argue that evidentials have a unified semantics of first-person mental states. Drawing on philosophical research on self-knowledge, I further argue that some linguistic properties of evidentials are rooted in the non-linguistic properties of the cognitive processes they describe, and show that it explains in a uniform fashion the behavior of evidentials across enviornments.
Currently I am also interested in constructions dealing with evidence more generally.
My joint work with Pranav Anand (UC Santa Cruz) focuses on various issues related to subjective language. It explores linguistic and non-linguistic properties of predicates of personal taste (e.g. delicious) and other subjective expressions (e.g. delightful). Recent work in formal semantics and philosophy of language shows that linguistic behavior of PPTs differs from that of other predicates, like round or popular, both in grammatical distribution and conversational dynamics: for example, disagreements over tastiness are seen as matters of opinion, not fact. We aim to determine whether PPTs form a natural class across languages and conceptual domains, and whether other predicates involving judgment -- aesthetic, moral, value -- also are PPTs. The overarching goal is to understand how natural language conceptualizes taste and what makes PPTs special: the semantics, the pragmatics, or the epistemology and psychology of taste. To explore this broader issue, we look at a series of less-studied puzzles and compare PPTs with other taste ascriptions (e.g. to like).
Our work is fully collaborative and the contributions of both authors are equal. The order of authors in published work alternates from publication to publication.
non-canonical interrogatives across slavic
Slavic languages show great propensity for non-canonical interrogatives, whose primary goal is not a naive inquiry for information but a different pragmatic function. In the past, I have worked on constructions consisting of two interrogative clauses, a clause with the fronted wh-adverbial ('what' or 'how') and a wh-clause with a fronted wh-phrase, which have been previously analyzed as an instance of wh-scope marking and which I have argued are interrogative parentheticals. Currently I am exploring biased questions in across Slavic. Questions with espitemic bias, which are used to confirm or rejct an idea the speaker has, have attracted a lot of attention in the recent literature. However, with few exceptions, much of the work in this area has been conducted on the material of English, German and French. Cross-linguistic research is pivotal for semantics and pragmatics, and in this strand, I seek to redress this empirical gap by looking at discourse particles associated with question bias across Slavic.
This paper is devoted to the conversational dynamics of Russian questions with the left-periphery particle razve. Razve-questions seem to convey negative bias in conﬂicting evidence scenarios, thus bearing superﬁcial resemblance to questions with English really (cf. Repp and Geist., forth.). I argue that razve-questions signal that the speaker is in a situation with belief revision potential, facing a conﬂict between a prior belief and a current abductive inference. Depending on context, such questions receive (i) an information-seeking intepretation or (ii) what I will call a 'point-making' interpretation. I propose a uniﬁed semantics for razve, while also showcasing the limitations of current approaches to question bias.
Evidentiality often clusters in certain regions, and where it does, it has a dominant morphosyntactic form, e.g. most evidentials of Western Europe are modal auxiliaries. Another frequent make-up is (present) perfect morphology, referred to as "Perfect of evidentiality". It happens to encode some sort of evidentiality all over the globe: Dogon, Newari, Scandinavian languages, Spanish of La Paz, Northern Ostyak, Komi Zyryan. The highest concentration of the perfect-evidential overlap is found in the Balkan-Caucasus region, sometimes referred to as the Old World evidential belt, including but not limited to Balkan Sprachbund. Most famously described for Turkish as well as other Turkic languages, it in fact pervades the area and is a feature of the following languages: Balkan Romance: Aromanian, Daco Romanian, Megleno Romanian; Iranian: Farsi, Ishkashim, Tajik; Kartvelian: Georgian; Indo-Aryan: Romani; South Slavic: Bulgarian, Macedonian; Daghestanian: Agul, Archi, Bagvalal, Dargwa, Hunzib; Indo-European isolates: Eastern Armenian and Albanian.
Such perfects express interesting commonalities: (1) they express indirect evidentiality and are often analysed as two accidentaly homophonous markers, conjectural and reportative and (2) they encode non-trivial temporal relations between the speech situation and the situation of evidence acquisition.
I've done fieldwork on Georgian, which shows interesting (dis)similiarities with Turkish and Bulgarian both in the evidential domain and interaction of evidentiality with tense and aspect. I'm not actively working on the topic but am still interested in the nature, historic development and georgraphic distribution of the perfect-evidential overlap and in how to derive the meaning of indirect evidentiality without postulating accidental homophony.
Adyghe is a Northwest Caucasian language, spoken mostly in the Caucasus. I'm proud to have participated in a series of fieldwork trips to the national republic Adyghea organised and funded by the Russian State University for the Humanities in 2004-2010. Such trips, or expeditions, are a hallmark of Russian typological tradition, largely thanks to the efforts of late A.E. Kibrik. Instead of bringing a consultant to the classroom, which is a common practice for the Field Methods in the U.S., a group of people goes into the field to work on a language. In such a setting, one gets acquainted not only with the language itself, but also with the culture of its speakers, which is an essential part of being a fieldworker. Theoretical research is combined with the language documentation component, including, as the ultimate goal, writing a reference grammar with a significant amount of elicited texts. It's in these expeditions where I got the training I have now.
Adyghe is a polysynthetic language with complex and fascinating morphological patterns. Over the years, I have mainly worked on the tense and aspect system, and constraints on affix ordering. I have been investigating semantics of various tense and aspect suffixes, whose inventory in Adyghe goes beyond the familiar and includes things like 'pretend that p'. In particular, I was working on the so-called double past, which led me to the exploration of semantic typology of pluperfects, their non-temporal values and possible application of inertia worlds apparatus in this domain.
In my own work and in joint work with Yury Lander (Higher School of Economics, Moscow), I showed that Adyghe affix ordering is compositional in the suffix domain and templatic in the prefix domain.
In 2009, I contributed to the collection of papers (in Russian) that serves as a basis for a theoretically-informed reference grammar of Adyghe, and currently I am not working on this.