Search Results

This chapter is taken up with three separate concerns: what are the relationships between sound‐class verbs and the other classes of perception verbs; are these verbs evidential, and how is the evidentiality to be analysed; and how should we understand the phenomenon of what has come to be known as copy raising? It is argued that the network structure surrounding these verbs' senses includes part of the semantic structure of their sound‐class counterparts, and this provides a basis for the understanding of evidentiality; an understanding of evidentiality and the semantics of as if clauses also underscores the treatment of copy raising.

Previous work on Korean grammar has claimed that one person cannot “directly” speculate about another person's thoughts, intentions, and sensations (Choi 1991; Kim 1978; Lee 1993; Sohn and Park 2003). By looking at how a teacher at a Korean heritage language school represents her access to students' thoughts and feelings, I argue that expressions of epistemic stance relate to moral evaluation. Speakers portray their access to the thoughts and sensations of individuals who they deem morally worthy as more distant and uncertain. When individuals are evaluated as morally suspect, however, speakers represent their access to these persons' emotions, thoughts and sensations as direct and unmediated. Evidential marking thus serves as a resource for indexically constructing others as different kinds of moral beings and for constituting respect, power, and authority.

In this chapter, a diachronic approach is adopted in the analysis of the two large SiBol newspaper corpora, in order to examine and compare the expression of evidentiality (Bednarek, 2006; Dendale and Tasmowski, 2001; and Chafe, 1986) – that is, how the writer's knowledge is marked as having been ‘seen’ or ‘heard’, etc., how the knowledge is attributed, and how it is passed on to the reader. Findings show an increased use of evidential markers over the thirteen-year period studied, and, at the same time, a shift in reporter usage of evidentiality towards hearsay evidence and the reporting of knowledge acquired by speculation. This is in keeping with other observations regarding an increased ‘vagueness’ in contemporary journalism (see Duguid [a], this volume),which is counterbalanced at times with an elevation of the newsworker's presence.

This work further investigates the use of a complementizer-like particle que in Spanish with a reportative meaning. The reportative QUE in Mexican Spanish shows an unparalleled behavior as an evidential: it is the only Spanish variety—considering Latin America and the Iberian Peninsula—in which que may appear preceding nonclausal constituents DP, NP, and PP. It will be further shown here that this unexpected property must obey two prosodic restrictions: a pause and a certain intonational pattern are mandatory when the nonclausal constituent is postverbal. Even though the reportative QUE may acquire modal overtones (of doubt or [ad]mirativity), it is not inherently modal, unlike the adverbial dizque and the Mexican innovation según que, which are inherently modal. It is proposed here that the Mexican reportative QUE is the natural extension of the complementizer que of verba dicendi; it is also contended that QUE becomes a grammaticalized reportative evidential.

Chapter 6 is the conclusion. It gives a summary of the analysis and the empirical coverage, but also discusses extensions of the proposed analysis. The proposed semantics for all sentence types is quite general, though designed for a compositional semantics for evidentials and mood. Many other phenomena may be seen as contributing similar kinds ofmeaning, including appositives, non‐restrictive relative clauses, and (slifting) parentheticals, in addition to other expressions of evidentiality.

The Germanic languages have a number of different verbs of perception such as ‘see’, ‘hear’, and ‘look like’, and these verbs can appear in different syntactic frames. The literature on these verbs point to many similarities and also interesting and subtle differences between verbs and constructions. This chapter specifically focuses on English ‘look like’ and its Swedish counterpart ‘se ut som’. Specifically, copy-raising examples like ‘Mia looked like she was sleeping’ are compared to expletive examples such as ‘It looked like Mia was sleeping’. A comparison between new psycholinguistic study of Swedish and similar recent studies on Swedish and English lends support to the hypothesis that copy-raising and expletive examples are more similar to each other in English than they are in Swedish: in Swedish the embedded pronoun is less likely to be interpreted as co-referential with the matrix subject.

A morphosemantic proposal is made for the unification of the passive participle of the previous chapter with the participle found in the perfect construction. Previous semantic analyses of the perfect are presented and discussed, and a new proposal is made for the semantic contribution of the perfect auxiliary have.

Every language has a way of speaking about how one knows what one says, and what one thinks about what one knows. In any language, there are ways of phrasing inferences, assumptions, probabilities and possibilities, and expressing disbelief. These epistemiological meanings and their cultural correlates are the subject matter of this chapter. We briefly revisit the relationship between evidentiality, that is, grammaticalized marking of information source, and information source, and attitudes to information, expressed through other linguistic means. Evidentials may have non-evidential extensions. Their use tends to reflect cultural norms and conventions. And their meanings may change as new techniques of acquiring information become available. The expression of knowledge and attitudes to it correlate with various cultural requirements, including the requirement to be explicit and precise, value of knowledge of different kind, and beliefs. Means of expressing knowledge are prone to diffusion in language contact. The last section of the chapter contains a brief outline of this volume.

Maaka, a Chadic language of north-eastern Nigeria, has a variety of means expressing notions of knowledge and truth. The language has a complicated system of expressing information source within a noun phrase, and also within a clause. This is in addition to further evidentiality strategies with their own epistemic overtones, by means of evidential markers and epistemic and modal verbs, which help estimate the reliability of information. In Maaka, knowledge and truth can be conceptualized as the result of direct observation and affectedness (vision, joint perception, intuition); belief and conviction; possession and ownership of epistemic resources; control over information; and shared insight and common experience. A tendency towards exclusive control of knowledge among the Maaka may be responsible for the development of its complex system of evidentials and epistemics.

The chapter focuses on the interaction of future-time reference with modality and evidentiality in predictions. Predictions are a type of speech act that (a) requires the propositional content to be a future eventuality and (b) presupposes that the speaker has indirect evidence at his/her disposal to assert the propositional content. The paper is aimed at exploring how future reference in prediction affects the use of modal expressions as evidential strategies in Italian, introducing specific constraints and affordances. More specifically, it presents an investigation of the behaviour of the Italian modal verbs potere (‘can’, ‘may’) and dovere (‘must’, ‘have to’) in predictions, based on a large corpus of Italian economic-financial news articles. Adopting a relational analysis of modality in the broad lineage of Relative Modality, and focusing on the contextual saturation of the conversational background of the modals, the chapter examines how the two modal verbs can function as evidential strategies in predictions, manifesting the kind of evidence the prediction is based on, and acting as anaphorical or cataphorical pointers towards other discourse utterances where the evidential basis of the prediction is made explicit.

This chapter provides an account of the Spanish ‘mirative future.’ Starting from the assumption that the Spanish future is an inferential evidential (Escandell-Vidal 2010, 2014, 2020), it is argued that mirativity is contextually inferred as a result of a combination of multiple factors that constrain interpretation. These factors include restrictions on word order (only VS), intonational patterns (either fall-rise or high rise), and kind of predicates (individual-level, gradable, negative predicates). In addition, the speaker encodes that s/he has only indirect evidence, but the context of utterance makes it clear that this is not so. Given these conditions, mirative uses are essentially rhetorical questions. From this perspective, the Spanish mirative future reproduces a consistent pattern across languages: mirative interpretations are obtained when indirect or inferential evidential markers are used in contexts where the speaker has direct experience of the facts.

This chapter sets out the initial parameters for the concept of serial verbs with some initial examples and a list of definitional properties. The classification of serial verbs into symmetrical and asymmetrical is introduced. The components of a serial verb may have to be strictly contiguous. Alternatively, other constituents may intervene between them. Serial verbs may form one grammatical word, or they may consist of several words. Various grammatical categories can be marked on each components (concordant marking) or be expressed just once per construction (single marking). Verb serialization can be productive, or it can be limited. The history of studies of serial verbs is discussed in some detail in the Appendix. Special focus is on the empirical basis of our cross-linguistic analysis based on the investigation of c.700 grammars of languages from every part of the world, and the author’s own fieldwork in Amazonia and New Guinea.

This chapter discusses how children’s conceptual representations of the mind make contact with language. It focuses on two domains: the understanding of the conditions that lead to knowledge, and the ability to attribute knowledge to oneself and others. Specifically, it asks whether language provides the representational resources necessary for representing mental states and whether cross-linguistic differences in encoding of mental states influence sensitivity to the features that distinguish the conditions that allow people to gain knowledge. Empirical findings in these domains strongly suggest that language scaffolds the development of these cognitive abilities without altering the underlying conceptual representations of mental states.

It is argued in this chapter, on the basis of evidence from grammar, discourse, and verbal art, that for Amazonian Quichua speakers, there is a cultural preference for expressing uncertainty, which is linked with animistic perspectivism. Animistic perspectivism endows nonhumans with subjectivity and implies that there is an infinite multiplicity of perspectives, thereby making a single, totalizing truth impossible. Respectable uncertainty is also apparent in the system of evidentiality, in speech reports, echo questions, and verbal art, all of which emphasize perspective over certainty. A type of certainty that Runa do value, however, and which would not be valid within a rational framework of inquiry, is that of emotional truth, involving feelings of empathy for others, including nonhumans. Emotional truth, then, provides an exception to the preference for uncertainty, and may lead people to confidently reason about ethical matters.​

This chapter is an overview of the nature of evidentials and of some existing formal analyses of them. It begins with a definition of evidentials and some examples of them, and continues to a summary of semantic analyses of evidentials. This part focues on how existing analyses handle the two types of evidentials that have been shown to exist: so‐called illocutionary and epistemic evidentials. Finally, the treatment of evidentials in formal pragmatics is considered.

This chapter provides a novel analysis of evidentials with an eye toward their pragmatic behavior. It begins with some criteria that a pragmatically adequate analysis must satisfy, and then turns to an overview of the formal system that forms the core of the analysis, a dynamic model involving sets of information states, each equipped with a plausibility relation, which are merged into a global state by a process of belief aggregation. Some principles are proposed for the mapping between evidential content and these information states. Finally, some case studies are presented.

This chapter uses the model of chapter 7 to analyze the concept of best possible grounds for a speech act, an important one in the analysis of evidentials. It is argued that best possible grounds amounts to the most reliable source for a particular assertion, as computed on the basis of initial probabilities of reliability together with the history of interaction in the sense of the first part of the book. It is also shown how the history of reliability of a particular evidence source can influence future judgements about its reliability and the overall ordering of evidence sources.