This article presents a detailed formal approach to concepts and concept combination. Sense generation is a competence‐level theory that attempts to respect constraints from the various cognitive sciences, and postulates “quasi‐classical” conceptual structures where attributes receive only one value (but are defeasible and so do not represent necessary and sufficient conditions on category membership) and where classification is binary (but explicitly context‐sensitive). It is also argued that any general theory of concepts must account for “privative” combinations (e.g., stone lion, fake (...) gun, apparent friend) as extreme test‐cases of representational and classificatory flexibility. The approach presented therefore provides a treatment of these combinations. The approach differentiates between the “lexical concept” (the stable information represented in a mental lexicon) which acts as a base from which the various “senses” (flexible contents associated with words and phrases in context, and used in classification) are “generated.” Generation allows nonmonotonicity, so that in different circumstances, different attributes may be defeated or modified. Classification is treated as relative to the perspective adopted, so that a classification acceptable from one perspective may be unacceptable from another, without contradiction. The result is a view that assumes bottom‐up priority in concept combination, where the range of senses generated by bottom‐up rules of combination is tempered by pragmatic‐communicative constraints on classification. An account of the representational and classification behavior of privative combinations is outlined, and the article concludes with a discussion of some of the implications of the approach. (shrink)

A uniform theory of conditionals is one which compositionally captures the behavior of both indicative and subjunctive conditionals without positing ambiguities. This paper raises new problems for the closest thing to a uniform analysis in the literature (Stalnaker, Philosophia, 5, 269–286 (1975)) and develops a new theory which solves them. I also show that this new analysis provides an improved treatment of three phenomena (the import-export equivalence, reverse Sobel-sequences and disjunctive antecedents). While these results concern central issues in the study (...) of conditionals, broader themes in the philosophy of language and formal semantics are also engaged here. This new analysis exploits a dynamic conception of meaning where the meaning of a symbol is its potential to change an agent’s mental state (or the state of a conversation) rather than being the symbol’s content (e.g. the proposition it expresses). The analysis of conditionals is also built on the idea that the contrast between subjunctive and indicative conditionals parallels a contrast between revising and consistently extending some body of information. (shrink)

“Contextual” vocabulary acquisition is the active, deliberate acquisition of a meaning for a word in a text by reasoning from textual clues and prior knowledge, including language knowledge and hypotheses developed from prior encounters with the word, but without external sources of help such as dictionaries or people. But what is “context”? Is it just the surrounding text? Does it include the reader’s background knowledge? I argue that the appropriate context for contextual vocabulary acquisition is the reader’s “internalization” of the (...) text “integrated” into the reader’s “prior” knowledge via belief revision. (shrink)

This target article presents a new computational theory of explanatory coherence that applies to the acceptance and rejection of scientific hypotheses as well as to reasoning in everyday life, The theory consists of seven principles that establish relations of local coherence between a hypothesis and other propositions. A hypothesis coheres with propositions that it explains, or that explain it, or that participate with it in explaining other propositions, or that offer analogous explanations. Propositions are incoherent with each other if they (...) are contradictory, Propositions that describe the results of observation have a degree of acceptability on their own. An explanatory hypothesis is acccpted if it coheres better overall than its competitors. The power of the seven principles is shown by their implementation in a connectionist program called ECHO, which treats hypothesis evaluation as a constraint satisfaction problem. Inputs about the explanatory relations are used to create a network of units representing propositions, while coherende and incoherence relations are encoded by excitatory and inbihitory links. ECHO provides an algorithm for smoothly integrating theory evaluation based on considerations of explanatory breadth, simplicity, and analogy. It has been applied to such important scientific cases as Lovoisier's argument for oxygen against the phlogiston theory and Darwin's argument for evolution against creationism, and also to cases of legal reasoning. The theory of explanatory coherence has implications for artificial intelligence, psychology, and philosophy. (shrink)

We translate sentence generation from TAG grammars with semantic and pragmatic information into a planning problem by encoding the contribution of each word declaratively and explicitly. This allows us to exploit the performance of off-the-shelf planners. It also opens up new perspectives on referring expression generation and the relationship between language and action.

In this article we discuss what constitutes a good choice of semantic representation, compare different approaches of constructing semantic representations for fragments of natural language, and give an overview of recent methods for employing inference engines for natural language understanding tasks.

In this paper, we address several puzzles concerning speech acts, particularly indirect speech acts. We show how a formal semantictheory of discourse interpretation can be used to define speech acts and to avoid murky issues concerning the metaphysics of action. We provide a formally precise definition of indirect speech acts, including the subclass of so-called conventionalized indirect speech acts. This analysis draws heavily on parallels between phenomena at the speech act level and the lexical level. First, we argue that, just (...) as co-predication shows that some words can behave linguistically as if they're `simultaneously' of incompatible semantic types, certain speech acts behave this way too. Secondly, as Horn and Bayer (1984) and others have suggested, both the lexicon and speech acts are subject to a principle of blocking or "preemption by synonymy": Conventionalized indirect speech acts can block their `paraphrases' from being interpreted as indirect speech acts, even if this interpretation is calculable from Gricean-style principles. We provide a formal model of this blocking, and compare it with existing accounts of lexical blocking. (shrink)

Web legal information retrieval systems need the capability to reason with the knowledge modeled by legal ontologies. Using this knowledge it is possible to represent and to make inferences about the semantic content of legal documents. In this paper a methodology for applying NLP techniques to automatically create a legal ontology is proposed. The ontology is defined in the OWL semantic web language and it is used in a logic programming framework, EVOLP+ISCO, to allow users to query the semantic content (...) of the documents. ISCO allows an easy and efficient integration of declarative, object-oriented and constraint-based programming techniques with the capability to create connections with external databases. EVOLP is a dynamic logic programming framework allowing the definition of rules for actions and events. An application of the proposed methodology to the legal web information retrieval system of the Portuguese Attorney General’s Office is described. (shrink)

In this paper we explore how compositional semantics, discourse structure, and the cognitive states of participants all contribute to pragmatic constraints on answers to questions in dialogue. We synthesise formal semantic theories on questions and answers with techniques for discourse interpretation familiar from computational linguistics, and show how this provides richer constraints on responses in dialogue than either component can achieve alone.

Situation theory has been developed over the last decade and various versions of the theory have been applied to a number of linguistic issues. However, not much work has been done in regard to its computational aspects. In this paper, we review the existing approaches towards 'computational situation theory' with considerable emphasis on our own research.

In a discussion of Sherlock Holmes’ “science of deduction” and the related “method of exclusion,” I show that Holmes’ claim that his inferences are deductive makes sense, if we consider his theoretical presuppositions. So, it is more accurate to say that he tries to reduce abduction to deduction than that he confuses them. His theoretical framework, albeit inadequate as a theory of empirical reasoning, can be seen as a basic model of classical (symbolic) AI. The main problems of this approach (...) are surveyed, and abduction is brought into play as both a better characterization of Holmes’ inferences and a better guide for building AI systems. This is a good background for raising the question of creativity because, according to Peirce, it relates to abduction in a substantial way. Is Sherlock Holmes creative? Can machines be creative? An affirmative answer to the second question might be given by a “logic of discovery” but the problem remains that inventing such a logic already presupposes creativity. (shrink)

Deictic (or pointing) gestures are traditionally known to have a simple function: to supply something as the referent of a demonstrative linguistic expression. I argue that deixis can have a more complex function. A deictic gesture can be used to _say something_ in conversation and can thereby become a full discourse move in its own right. To capture this phenomenon, which I call _rich demonstration_, I present an update semantics on which deictic gestures can indicate situations from a conversation’s context (...) and those situations coherently connect to prior discourse to generate information. (shrink)

This paper presents a novel account of focal stress and pitch contour in English dialogue. We argue that one should analyse and treat focus and pitch contour jointly, since (i) some pragmatic interpretations vary with contour (e.g., whether an utterance accepts or rejects; or whether it implicates a positive or negative answer); and (ii) there are utterances with identical prosodic focus that in the same context are infelicitous with one contour, but felicitous with another. We offer an account of two (...) distinct pitch contours that predicts the correct felicity judgements and implicatures, outclassing other models in empirical coverage or formality. Prosodic focus triggers a presupposition, where what is presupposed and how the presupposition is resolved depends on prosodic contour. If resolving the presupposition entails the proffered content, then the proffered content is uninteresting and hence the utterance is in-felicitous. Otherwise, resolving the presupposition may lead to an implicature. We regiment this account in SDRT. (shrink)

Nonindexical Context-Dependence and the Interpretation as Abduction Approach Inclusive nonindexical context-dependence occurs when the preferred interpretation of an utterance implies its lexically-derived meaning. It is argued that the corresponding processes of free or lexically mandated enrichment can be modeled as abductive inference. A form of abduction is implemented in Simple Type Theory on the basis of a notion of plausibility, which is in turn regarded a preference relation over possible worlds. Since a preordering of doxastic alternatives taken for itself only (...) amounts to a relatively vacuous ad hoc model, it needs to be combined with a rational way of learning from new evidence. Lexicographic upgrade is implemented as an example of how an agent might revise his plausibility ordering in light of new evidence. Various examples are given how this apparatus may be used to model the contextual resolution of context-dependent or semantically incomplete utterances. The described form of abduction is limited and merely serves as a proof of concept, but the idea in general has good potential as one among many ways to build a bridge between semantics and pragmatics since inclusive context-dependence is ubiquitous. (shrink)

In our 2015 book Imagination and Convention, we explore the scope and limits of linguistic knowledge in semantics and pragmatics for natural language. We draw heavily on the notion of coordination from David Lewis' book on conventions. To the extent that the account we develop is right, general principles like Grice's cooperative principle and the maxims of conversation have little to say about about interpretation. Three commentators—Anne Bezuidenhout, Laurence Horn, and Zoltan Gendler Szabo—discuss and evaluate our program in three essays (...) in this issue. What follows is our response to their challenges. (shrink)

I show how a conversational process that takes simple, intuitively meaningful steps may be understood as a sophisticated computation that derives the richly detailed, complex representations implicit in our knowledge of language. To develop the account, I argue that natural language is structured in a way that lets us formalize grammatical knowledge precisely in terms of rich primitives of interpretation. Primitives of interpretation can be correctly viewed intentionally, as explanations of our choices of linguistic actions; the model therefore fits our (...) intuitions about meaning in conversation. Nevertheless, interpretations for complex utterances can be built from these primitives by simple operations of grammatical derivation. In bridging analyses of meaning at semantic and symbol‐processing levels, this account underscores the fundamental place for computation in the cognitive science of language use. (shrink)

This paper concerns the phenomenon of pragmatic enrichment, and has a proposal for predicting the occurrence of such enrichments. The idea is that an enrichment of an expressed content c occurs as a means of strengthening the coherence between c and a salient given content c’ of the context, whether c’ is given in discourse, as sentence parts, or through perception. After enrichment, a stronger coherence relation is instantiated than before enrichment. An idea of a strength scale of types of (...) coherence relations is proposed and applied. (shrink)

Abstract -/- Quantifier domain restriction (QDR) and two versions of nominal restriction (NR) are implemented as restrictions that depend on a previously introduced interpreter and interpretation time in a two-dimensional semantic framework on the basis of simple type theory and categorial grammar. Against Stanley (2002) it is argued that a suitable version of QDR can deal with superlatives like tallest. However, it is shown that NR is needed to account for utterances when the speaker intends to convey different restrictions for (...) multiple uses of the same quantifying determiner. We argue that NR generally fares better with such examples but also observe that examples like Every sailor waves at every sailor might be pragmatically anomalous. An account of contextual domain restriction is proposed that (i) excludes these anomalous readings (but it is shown how they could be included), (ii) makes it possible to express different contextual domain restrictions as long-range dependencies on an interpreter and an interpretation time, and (iii) additionally models restrictions based on locative constructions as general mereological constraints introduced by shifting the index. (shrink)

This article describes abductions as special patterns of inference to the best explanation whose structure determines a particularly promising abductive conjecture and thus serves as an abductive search strategy. A classification of different patterns of abduction is provided which intends to be as complete as possible. An important distinction is that between selective abductions, which choose an optimal candidate from given multitude of possible explanations, and creative abductions, which introduce new theoretical models or concepts. While selective abduction has dominated the (...) literature, creative abductions are rarely discussed, although they are essential in science. The article introduces several kinds of creative abductions, such as theoretical model abduction, common cause abduction and statistical factor analysis, and illustrates them by various real case examples. It is suggested to demarcate scientifically fruitful abductions from purely speculative abductions by the criterion of causal unification. (shrink)

Abduction is or subsumes a process of inference. It entertains possible hypotheses and it chooses hypotheses for further scrutiny. There is a large literature on various aspects of non-symbolic, subconscious abduction. There is also a very active research community working on the symbolic (logical) characterisation of abduction, which typically treats it as a form of hypothetico-deductive reasoning. In this paper we start to bridge the gap between the symbolic and sub-symbolic approaches to abduction. We are interested in benefiting from developments (...) made by each community. In particular, we are interested in the ability of non-symbolic systems (neural networks) to learn from experience using efficient algorithms and to perform massively parallel computations of alternative abductive explanations. At the same time, we would like to benefit from the rigour and semantic clarity of symbolic logic. We present two approaches to dealing with abduction in neural networks. One of them uses Connectionist Modal Logic and a translation of Horn clauses into modal clauses to come up with a neural network ensemble that computes abductive explanations in a top-down fashion. The other combines neural-symbolic systems and abductive logic programming and proposes a neural architecture which performs a more systematic, bottom-up computation of alternative abductive explanations. Both approaches employ standard neural network architectures which are already known to be highly effective in practical learning applications. Differently from previous work in the area, our aim is to promote the integration of reasoning and learning in a way that the neural network provides the machinery for cognitive computation, inductive learning and hypothetical reasoning, while logic provides the rigour and explanation capability to the systems, facilitating the interaction with the outside world. Although it is left as future work to determine whether the structure of one of the proposed approaches is more amenable to learning than the other, we hope to have contributed to the development of the area by approaching it from the perspective of symbolic and sub-symbolic integration. (shrink)

Formalising deontic concepts, such as obligation, prohibition and permission, is normally carried out in a modal logic with a possible world semantics, in which some worlds are better than others. The main focus in these logics is on inferring logical consequences, for example inferring that the obligation O q is a logical consequence of the obligations O p and O. In this paper we propose a non-modal approach in which obligations are preferred ways of satisfying goals expressed in first-order logic. (...) To say that p is obligatory, but may be violated, resulting in a less than ideal situation s, means that the task is to satisfy the goal p ∨ s, and that models in which p is true are preferred to models in which s is true. Whereas, in modal logic, the preference relation between possible worlds is part of the semantics of the logic, in this non-modal approach, the preference relation between first-order models is external to the logic. Although our main focus is on satisfying goals, we also formulate a notion of logical consequence, which is comparable to the notion of logical consequence in modal deontic logic. In this formalisation, an obligation O p is a logical consequence of goals G, when p is true in all best models of G. We show how this non-modal approach to the treatment of deontic concepts deals with problems of contrary-to-duty obligations and normative conflicts, and argue that the approach is useful for many other applications, including abductive explanations, defeasible reasoning, combinatorial optimisation, and reactive systems of the production system variety. (shrink)

’s glue logic for computing logical form dynamic. This allows us to model a dialogue agent’s understanding of what the update of the semantic representation of the dialogue would be after his next contribution, including the effects of the rhetorical moves that he is contemplating performing next. This is a pre-requisite for developing a model of how agents reason about what to say next. We make the glue logic dynamic by using a dynamic public announcement logic ( pal ). We (...) extend pal with a particular variety of default reasoning suited to reasoning about discourse—this default reasoning being an essential component of inferring the pragmatic effects of one’s dialogue moves. We add to the pal language a new type of announcement, known as ceteris paribus announcement, and this is used to model how an agent anticipates the (default) pragmatic effects of his next dialogue move. Our extended pal validates certain intuitive patterns of default inference that existing pal s for practical reasoning do not. We prove that the dynamic glue logic has a pspace validity problem, and as such is no more complex than pal with multiple $${\square}$$ operators. (shrink)

We describe a methodology for learning a disambiguation model for deep pragmatic interpretations in the context of situated task-oriented dialogue. The system accumulates training examples for ambiguity resolution by tracking the fates of alternative interpretations across dialogue, including subsequent clarificatory episodes initiated by the system itself. We illustrate with a case study building maximum entropy models over abductive interpretations in a referential communication task. The resulting model correctly resolves 81% of ambiguities left unresolved by an initial handcrafted baseline. A key (...) innovation is that our method draws exclusively on a system’s own skills and experience and requires no human annotation. (shrink)

Despite the attention that various forms of ellipsis have received inthe literature, the conditions under which a representation of anutterance may serve as a suitable referent for interpreting subsequentelliptical forms remain poorly understood. This fundamental questionremains as a point of contention, particularly because there are datato support various conflicting approaches that attempt to characterizethese conditions within a single module of language processing. Weshow a previously unnoticed pattern in VP-ellipsis data with respectto the type of coherence relation extant between the antecedentand (...) elided clauses. This pattern is explained by an account of howellipsis resolution processes interact with the inference processesunderlying the establishment of these relations. The analysis alsoexplains a similar yet distinct pattern in gapping constructions whichare not accounted for by purely syntactic approaches. Finally, wediscuss event reference and compare the resulting account to thedichotomy of types of anaphora posited by HanSag. (shrink)

This study introduces an enigmatic construction in Japanese called chūshakuteki nibun-renchi ‘annotative dual-clause juxtaposition’ (ADCJ), exemplified below: Hiro wa, dare ni au no ka, resutoran o yoyakushita. top who dat meet nmlz int restaurant acc reserved Lit. ‘Hiro, (I wonder) who (he) will meet, reserved a restaurant.’ This construction is ubiquitous and yet little known even in Japanese linguistics circles. Because the matrix predicate of ADCJ cannot semantically accommodate such a component as dare ni au no ka ‘who (he) will (...) meet’ above, this paper argues that ADCJ is parenthetical, a construct that should be recognized as an essential element of verbal communication and, in turn, a determining factor in how utterances are to be formed and interpreted. This construction is dissimilar to any other type of parentheticals hitherto reported in the literature. What is so special about it is its merger of portraying two situations through abduction and expressing the entire circumstance in a single communicative unit. For example, in the above example, the parenthetical element explains why the speaker wishes to convey the matrix statement. From an interactional perspective, the primary function of ADCJ is to highlight the speaker’s intellectual and communicative involvement in the depicted scene. This style of communication, when compared with an ‘objective’ and apathetic description, is likely to induce more earnest reactions from the hearer or reader and, consequently, promote a more favorable continuation of the conversation or reading. This paper advocates a wide-ranging examination of thetical grammar (Kaltenböck et al. 2011), for which detailed analyses of constructions such as ADCJ that traditional syntactic/semantic theories cannot capture are indispensable. (shrink)

This paper examines the task of understanding dialogues in terms of the mental states of the participating agents. We present a motivating example that clarifies the challenges this problem involves and then outline a theory of dialogue interpretation based on abductive inference of these unobserved beliefs and goals, incremental construction of explanations, and reliance on domain-independent knowledge. After this, we describe UMBRA, an implementation of the theory that embodies these assumptions. We report experiments with the system that demonstrate its ability (...) to accurately infer the conversants’ mental states even when some speech acts are unavailable. We conclude by reviewing related research on dialogue and discussing avenues for future study. (shrink)

This chapter investigates the computational consequences of a broadly Gricean view of language use as intentional activity. In this view, dialogue rests on coordinated reasoning about communicative intentions. The speaker produces each utterance by formulating a suitable communicative intention. The hearer understands it by recognizing the communicative intention behind it. When this coordination is successful, interlocutors succeed in considering the same intentions— that is, the same representations of utterance meaning—as the dialogue proceeds. In this paper, I emphasize that these intentions (...) can be formalized; we can provide abstract but systematic representations that spell out what a speaker is trying to do with an utterance. Such representations describe utterances simultaneously as the product of our knowledge of grammar and as actions chosen for a reason. In particular, they must characterize the speaker’s utterance in grammatical terms, provide the links to the context that the grammar requires, and so arrive at a contribution that the speaker aims to achieve. Because I have implemented this formalism, we can regard it as a possible analysis of conversational processes at the level of computational theory. Nevertheless, this analysis leaves open what the nature of the biological computation involved in inference to intentions is, and what regularities in language use support this computation. (shrink)

Individuals usually go to art museums to enjoy artworks. Generally, in order to appreciate the art in museums, a brief summary of certain information is provided as a caption. Viewers usually read these descriptions to aid their understanding. To provide broader technical support for this activity, several researchers have proposed a protocol for art appreciation. For instance, Leder et al. proposed a stage model for aesthetic processing, which combines aspects of understanding and cognitive mastering with affective and emotional processing. We (...) have also conducted several experiments in order to determine the effect of information during art appreciation. For instance, we conducted an experiment where information about a piece of art was offered gradually and incrementally. In the experiment, the participants seemed to be able to gradually understand the artwork according to the obtained information. Our observations indicate that they tried to create stories for the artworks in order to explain the obtained information. In addition, for the abstract artworks, if they saw the title, they understood the artworks within their own explanation in the context of the title. Our research framework suggests that we can consider this observed framework as a process of abduction, where the incremental presentation of details about art helps a user form a hypothesis about the piece of art. In this paper, we will analyze artwork appreciation and understanding with this framework from the viewpoint of abduction. (shrink)