Extracting fictional truth from unreliable sources

A fictional text is commonly viewed as constituting an invitation to play a certain game of make-believe, with the individual sentences written by the author providing the propositions we are to imagine and/or accept as true within the fiction. However, we can’t always take the text at face value. What narratologists call ‘unreliable narrators’ may present a confused or misleading picture of the fictional world. Meanwhile, there has been a debate in philosophy about ‘imaginative resistance’ in which we resist imagining (or even accepting as true in the fiction) what’s explicitly stated in the text. But if we can’t take the text’s word for it, how do we determine what’s true in a fiction? The chapter proposes an account of fiction interpretation in a dynamic setting (a version of Discourse Representation Theory (DRT) with a mechanism for opening, updating, and closing temporary ‘workspaces’) and combines this framework with belief revision logic.

Sharing real and fictional reference

This chapter extends the framework of MSDRT (Mental State Discourse Representation Theory) to the problem of reference in fiction, and to the role and function of fictional names. Central to the investigation is the notion of an Entity Representation (ER), a central feature of MSDRT and used previously in the communication-theoretic analysis of the pragmatics and semantics of non-fictional names in Kamp (2015). As argued in that paper, the use of proper names within a speech community leads to networks of connected ERs in the mental states of their users. These networks provide the names with a kind of intersubjective identity. In this respect, fictional names resemble non-fictional names—those that refer to real entities, that exist in the actual world in which we live. This chapter proposes an analysis of fictional names and fictional reference that capitalizes on this resemblance.

Alguns adjuntos temporais com especial relevância para a tradução inglês-português: o caso de since-desde e expressões afins

This paper addresses the semantic analysis of polyvalent temporal adjuncts headed by (mainly) English since and Portuguese desde, and problems in translation from English to Portuguese. Four semantic values of the single operator since are considered (the second and fourth of which are not normally considered autonomously in the English literature) – durative location, derived durative location (in association with adjunct-triggered Aktionsart shift), simple inclusive location and temporal circumscription of quantification. Furthermore, the typically monovalent phrase ever since and the bivalent phrase long since are also taken into account. The fact the Portuguese desde – contrary to English since – is not normally associated with simple inclusive locations is the source of many translation problems. Other interesting grammatical issues, involving long since, are also addressed. The translation data is obtained from the website linguee.com (where six different types of problems are found), and the semantic analysis is made with the logic of the Discourse Representation Theory, elaborating on my previous work, Móia (2000).

Universal Discourse Representation Structure Parsing

We consider the task of cross-lingual semantic parsing in the style of Discourse Representation Theory (DRT) where knowledge from annotated corpora in a resource-rich language is transferred via bitext to guide learning in other languages. We introduce Universal Discourse Representation Theory (UDRT), a variant of DRT that explicitly anchors semantic representations to tokens in the linguistic input. We develop a semantic parsing framework based on the Transformer architecture and employ it to obtain semantic resources in multiple languages following two learning schemes. The Many-to-One approach translates non-English text to English, and then runs a relatively accurate English parser on the translated text, while the One-to-Many approach translates gold standard English to non-English text and trains multiple parsers (one per language) on the translations. Experimental results on the Parallel Meaning Bank show that our proposal outperforms strong baselines by a wide margin and can be used to construct (silver-standard) meaning banks for 99 languages.

Clinical Decision Support Systems Question and Answering

A question answering system to help clinical practitioners in a cardiovascular healthcare environment to interface clinical decision support systems can be built by using an extended discourse representation structure, CIDERS, and an ontology framework, Ontology for General Clinical Practice. CIDERS is an extension of the well-known DRT (discourse representation theory) structures, intending to go beyond single text representation to embrace the general clinical history of a given patient represented in an ontology. The Ontology for General Clinical Practice improves the currently available state-of-the-art ontologies for medical science and for the cardiovascular specialty. The chapter shows the scientific and philosophical reasons of its present dual structure with a deeply expressive (SHOIN) terminological base (TBox) and a highly computable (EL++) assertions knowledge base (ABox). To be able to use the current reasoning techniques and methodologies, the authors made a thorough inventory of biomedical ontologies currently available in OWL2 format.

Presupposition and Teaching Chinese as a Foreign Language

In cooperative human communication, the speaker has to mark the connection between his utterance and therefore the given information, because the hearer interprets the utterance regarding the data that has already been obtained. Languages adopt various devices to mark the connection between the utterance and also the context. This paper investigates semantic and pragmatic presupposition in Discourse Representation Theory (DRT) within the Chinese language and enhances the pragmatic perspective of presupposition in DRT.

The Cyclical Fluctuation of U.S. Public Higher Education Institutions Establishment Impact by the Kandretieff Cyclical Wave

In cooperative human communication, the speaker has to mark the connection between his utterance and therefore the given information, because the hearer interprets the utterance regarding the data that has already been obtained. Languages adopt various devices to mark the connection between the utterance and also the context. This paper investigates semantic and pragmatic presupposition in Discourse Representation Theory (DRT) within the Chinese language and enhances the pragmatic perspective of presupposition in DRT.

Varieties of Logical Form

The paper reviews some conceptions of logical form in the light of Andrea Iacona’s book Logical Form. I distinguish the following: logical form as schematization of natural language, provided by, for example, Aristotle’s syllogistic; the relevance to logical form of formal languages like those used by Frege and Russell to express and prove mathematical theorems; Russell’s mid-period conception of logical form as the structural cement binding propositions; the conceptions of logical form discussed by Iacona; and logical form regarded as an empirical hypothesis about the psychology of language processing, as in the Discourse Representation Theory tradition. Whereas neither schematization, nor the use of special languages for mathematics, raise general methodological or empirical difficulties, other conceptions of logical form raise at least apparent problems.

Modelling Structures for Situated Discourse

This paper describes a corpus of situated multiparty chats developed for the STAC project (Strategic Conversation, ERC grant n. 269427). and annotated for discourse structure in the style of Segmented Discourse Representation Theory (SDRT; Asher & Lascarides,2003). The STAC corpus is not only a rich source of data on strategic conversation, but also the first corpus that we are aware of that provides discourse structures for multiparty dialogues situated within a virtual environment. The corpus was annotated in two stages: we initially annotated the chat moves only, but later decided to annotate interactions between the chat moves and non-linguistic events from the virtual environment. This two-step procedure has allowed us quantify various ways in which adding information from the nonlinguistic context affects dialogue structure. In this paper, we look at how annotations based only on linguistic information were preserved once the nonlinguistic context was factored in. We explain that while the preservation of relation instances is relatively high when we move from one corpus to the other, there is little preservation of higher order structures that capture "the main point" of a dialogue and distinguish it from peripheral information.