This paper asks why information should ever be expressed vaguely, re-assessing some previously proposed answers to this question and suggesting some new ones. Particular attention is paid to the benefits that vague expressions can have in situations where agreement over the meaning of an expression cannot be taken for granted. A distinction between two different versions of the above-mentioned question is advocated. The first asks why human languages contain vague expressions, the second question asks when and why a speaker should (...) choose a vague expression when communicating with a hearer. While the former question is purely theoretical, the latter has practical implications for the computational generation of utterances in Natural Language Generation (NLG). (shrink)

This informal position paper brings together some recent developments in formal semantics and pragmatics to argue that the discipline of Game Theory is well placed to become the theoretical backbone of Natural Language Generation. To demonstrate some of the strengths and weaknesses of the Game-Theoretical approach, we focus on the utility of vague expressions. More specifically, we ask what light Game Theory can shed on the question when an NLG system should generate vague language.

Our lives are full of inexactitude. We say a person is tall or an action is just without the precision of measurement on a dial. In this engaging account, Kees van Deemter explores vagueness, cutting across areas such as language, mathematical logic, and computing. He considers why vagueness is inherent, and why it is important in how we function.

This article offers a survey of computational research on referring expressions generation (REG). It introduces the REG problem and describes early work in this area, discussing what basic assumptions lie behind it, and showing how its remit has widened in recent years. We discuss computational frameworks underlying REG, and demonstrate a recent trend that seeks to link up REG algorithms with well-established Knowledge Representation traditions. Considerable attention is given to recent efforts at evaluating REG algorithms and the lessons that they (...) allow us to learn. The article concludes with a discussion of the way forward in REG, focussing on references in larger and more realistic settings. (shrink)

This article introduces the topic ‘‘Production of Referring Expressions: Bridging the Gap between Computational and Empirical Approaches to Reference’’ of the journal Topics in Cognitive Science. We argue that computational and psycholinguistic approaches to reference production can benefit from closer interaction, and that this is likely to result in the construction of algorithms that differ markedly from the ones currently known in the computational literature. We focus particularly on determinism, the feature of existing algorithms that is perhaps most clearly at (...) odds with psycholinguistic results, discussing how future algorithms might include non-determinism, and how new psycholinguistic experiments could inform the development of such algorithms. (shrink)

Two questions dominate theoretical research on vagueness. The first is of a logical-semantic nature: What formal models offer the best understanding of vagueness? Many answers to this question have been proposed (e.g. [1], [2] for an overview), but none of these has found general acceptance so far. The second question is of a pragmatic nature and asks Why is language vague? This question has been asked forcefully by the economist Barton Lipman, who has shown that some seemingly plausible answers resist (...) analysis in terms of classical Game Theory [3], [4]. While a number of tentative answers to this question have been suggested (for a survey, see [6], [7]), Lipman’s question is still partly unresolved, particularly with respect to situations where there is no conflict between the speaker and the hearer (cf. [8]). (shrink)

A substantial amount of recent work in natural language generation has focused on the generation of ‘‘one-shot’’ referring expressions whose only aim is to identify a target referent. Dale and Reiter's Incremental Algorithm (IA) is often thought to be the best algorithm for maximizing the similarity to referring expressions produced by people. We test this hypothesis by eliciting referring expressions from human subjects and computing the similarity between the expressions elicited and the ones generated by algorithms. It turns out that (...) the success of the IA depends substantially on the ‘‘preference order’’ (PO) employed by the IA, particularly in complex domains. While some POs cause the IA to produce referring expressions that are very similar to expressions produced by human subjects, others cause the IA to perform worse than its main competitors; moreover, it turns out to be difficult to predict the success of a PO on the basis of existing psycholinguistic findings or frequencies in corpora. We also examine the computational complexity of the algorithms in question and argue that there are no compelling reasons for preferring the IA over some of its main competitors on these grounds. We conclude that future research on the generation of referring expressions should explore alternatives to the IA, focusing on algorithms, inspired by the Greedy Algorithm, which do not work with a fixed PO. (shrink)

Much of everyday language is vague, even in situations where vagueness could have been avoided. Yet the benefits of vagueness for hearers and readers are proving to be elusive. We discuss a range of earlier controlled experiments with human participants, and we report on a new series of experiments that we ourselves have conducted in recent years. These experiments, which focus on vague expressions that are part of referential noun phrases, aim to separate the utility of vagueness from the utility (...) of other factors that tend to co-occur with vagueness. After presenting the evidence, we argue that it supports a view where the benefits that vague terms exert are due to other influences, and not to vagueness itself. (shrink)

A fundamental part of the process of referring to an entity is to categorise it (for instance, as the woman). Where multiple categorisations exist, this implicitly involves the adoption of a conceptual perspective. A challenge for the automatic Generation of Referring Expressions is to identify a set of referents coherently, adopting the same conceptual perspective. We describe and evaluate an algorithm to achieve this. The design of the algorithm is motivated by the results of psycholinguistic experiments.

(Near-final version.) Accepted for publication in Artificial Intelligence Journal.

This discussion paper describes a sequence of experiments with human subjects aimed at finding out how an nlg system should choose between the different forms of a gradable adjective. This case study highlights some general questions that one faces when trying to base nlg systems on empirical evidence: one question is what task to set a subject so as to obtain the most useful information about that subject, another question has to do with differences between subjects.

8.6 Issues Raised by the Algorithms Proposed.

This informal position paper brings together some recent developments in formal semantics and pragmatics to argue that the discipline of Game Theory is well placed to become the theoretical backbone of Natural Language Generation. To demonstrate some of the strengths and weaknesses of the Game-Theoretical approach, we focus on the utility of vague expressions. More specifically, we ask what light Game Theory can shed on the question when an NLG system should generate vague language.

This book introduces the concept of information sharing as an area of cognitive science, defining it as the process by which speakers depend on "given" information to convey "new" information—an idea crucial to language engineering. Where previous work in information sharing was often fragmented between different disciplines, this volume brings together theoretical and applied work, and joins computational contributions with papers based on analyses of language corpora and on psycholinguistic experimentation.

This article explores the role of surface ambiguities in referring expressions, and how the risk of such ambiguities should be taken into account by an algorithm that generates referring expressions, if these expressions are to be optimally effective for a hearer. We focus on the ambiguities that arise when adjectives occur in coordinated structures. The central idea is to use statistical information about lexical co-occurrence to estimate which interpretation of a phrase is most likely for human readers, and to avoid (...) generating phrases where misunderstandings are likely. Various aspects of the problem were explored in three experiments in which responses by human participants provided evidence about which reading was most likely for certain phrases, which phrases were deemed most suitable for particular referents, and the speed at which various phrases were read. We found a preference for ‘‘clear’’ expressions to ‘‘unclear’’ ones, but if several of the expressions are ‘‘clear,’’ then brief expressions are preferred over non-brief ones even though the brief ones are syntactically ambiguous and the non-brief ones are not; the notion of clarity was made precise using Kilgarriff's Word Sketches. We outline an implemented algorithm that generates noun phrases conforming to our hypotheses. (shrink)

A substantial amount of recent work in natural language generation has focused on the generation of ‘‘one‐shot’’ referring expressions whose only aim is to identify a target referent. Dale and Reiter's Incremental Algorithm (IA) is often thought to be the best algorithm for maximizing the similarity to referring expressions produced by people. We test this hypothesis by eliciting referring expressions from human subjects and computing the similarity between the expressions elicited and the ones generated by algorithms. It turns out that (...) the success of the IA depends substantially on the ‘‘preference order’’ (PO) employed by the IA, particularly in complex domains. While some POs cause the IA to produce referring expressions that are very similar to expressions produced by human subjects, others cause the IA to perform worse than its main competitors; moreover, it turns out to be difficult to predict the success of a PO on the basis of existing psycholinguistic findings or frequencies in corpora. We also examine the computational complexity of the algorithms in question and argue that there are no compelling reasons for preferring the IA over some of its main competitors on these grounds. We conclude that future research on the generation of referring expressions should explore alternatives to the IA, focusing on algorithms, inspired by the Greedy Algorithm, which do not work with a fixed PO. (shrink)

This response discusses the experiment reported in Krahmer et al.’s Letter to the Editor of Cognitive Science. We observe that their results do not tell us whether the Incremental Algorithm is better or worse than its competitors, and we speculate about implications for reference in complex domains, and for learning from ‘‘normal” (i.e., non-semantically-balanced) corpora.

This article explores the role of surface ambiguities in referring expressions, and how the risk of such ambiguities should be taken into account by an algorithm that generates referring expressions, if these expressions are to be optimally effective for a hearer. We focus on the ambiguities that arise when adjectives occur in coordinated structures. The central idea is to use statistical information about lexical co‐occurrence to estimate which interpretation of a phrase is most likely for human readers, and to avoid (...) generating phrases where misunderstandings are likely. Various aspects of the problem were explored in three experiments in which responses by human participants provided evidence about which reading was most likely for certain phrases, which phrases were deemed most suitable for particular referents, and the speed at which various phrases were read. We found a preference for ‘‘clear’’ expressions to ‘‘unclear’’ ones, but if several of the expressions are ‘‘clear,’’ then brief expressions are preferred over non‐brief ones even though the brief ones are syntactically ambiguous and the non‐brief ones are not; the notion of clarity was made precise using Kilgarriff's Word Sketches. We outline an implemented algorithm that generates noun phrases conforming to our hypotheses. (shrink)

When producing a description of a target referent in a visual context, speakers need to choose a set of properties that distinguish it from its distractors. Computational models of language production/generation usually model this as a search process and predict that the time taken will increase both with the number of distractors in a scene and with the number of properties required to distinguish the target. These predictions are reminiscent of classic findings in visual search; however, unlike models of reference (...) production, visual search models also predict that search can become very efficient under certain conditions, something that reference production models do not consider. This paper investigates the predictions of these models empirically. In two experiments, we show that the time taken to plan a referring expression—as reflected by speech onset latencies—is influenced by distractor set size and by the number of properties required, but this crucially depends on the discriminability of the properties under consideration. We discuss the implications for current models of reference production and recent work on the role of salience in visual search. (shrink)

This paper asks why information should ever be expressed vaguely, re-assessing some previously proposed answers to this question and suggesting some new ones. Particular attention is paid to the benefits that vague expressions can have in situations where agreement over the meaning of an expression cannot be taken for granted. A distinction between two different versions of the above-mentioned question is advocated. The first asks why human languages contain vague expressions, the second question asks when and why a speaker should (...) choose a vague expression when communicating with a hearer. While the former question is purely theoretical, the latter has practical implications for the computational generation of utterances in Natural Language Generation (NLG). (shrink)

This article proposes the use of temporal logic for an analysis of instrumentality inspired by the work of G.H. von Wright. The first part of the article contains the philosophical foundations. We discuss von Wright’s general theory of agency and his account of instrumentality. Moreover, we propose several refinements to this framework via rigorous definitions of the core notions involved. In the second part, we develop a logical system called Temporal Logic of Action and Expectations (TLAE). The logic is inspired (...) by a fragment of propositional dynamic logic based on indeterministic time. The system is proven to be weakly complete relative to its given semantics. We then employ TLAE to formalise and analyse the instrumentality relations defined in the first part of the paper. Last, we point out philosophical implications and possible extensions of our work. (shrink)

A fundamental part of the process of referring to an entity is to categorise it (for instance, as the woman). Where multiple categorisations exist, this implicitly involves the adoption of a conceptual perspective. A challenge for the automatic Generation of Referring Expressions is to identify a set of referents coherently, adopting the same conceptual perspective. We describe and evaluate an algorithm to achieve this. The design of the algorithm is motivated by the results of psycholinguistic experiments.

STIT logic is a prominent framework for the analysis of multi-agent choice-making. In the available deontic extensions of STIT, the principle of Ought-implies-Can (OiC) fulfills a central role. However, in the philosophical literature a variety of alternative OiC interpretations have been proposed and discussed. This paper provides a modular framework for deontic STIT that accounts for a multitude of OiC readings. In particular, we discuss, compare, and formalize ten such readings. We provide sound and complete sequent-style calculi for all of (...) the various STIT logics accommodating these OiC principles. We formally analyze the resulting logics and discuss how the different OiC principles are logically related. In particular, we propose an endorsement principle describing which OiC readings logically commit one to other OiC readings. (shrink)

This work provides proof-search algorithms and automated counter-model extraction for a class of STIT logics. With this, we answer an open problem concerning syntactic decision procedures and cut-free calculi for STIT logics. A new class of cut-free complete labelled sequent calculi G3LdmL^m_n, for multi-agent STIT with at most n-many choices, is introduced. We refine the calculi G3LdmL^m_n through the use of propagation rules and demonstrate the admissibility of their structural rules, resulting in auxiliary calculi Ldm^m_nL. In the single-agent case, we (...) show that the refined calculi Ldm^m_nL derive theorems within a restricted class of (forestlike) sequents, allowing us to provide proof-search algorithms that decide single-agent STIT logics. We prove that the proof-search algorithms are correct and terminate. (shrink)

Nearly one million people take their own lives each year world-wide - however, contrary to popular belief, suicide can be prevented. While suicide is commonly thought to be an understandable reaction to severe stress, it is actually an abnormal reaction to regular situations. Something more than unbearable stress is needed to explain suicide, and neuroscience shows what this is, how it is caused and how it can be treated. Professor Kees van Heeringen describes findings from neuroscientific research on suicide, (...) using various approaches from population genetics to brain imaging. Compelling evidence is reviewed that shows how and why genetic characteristics or early traumatic experiences may lead to a specific predisposition that makes people vulnerable to triggering life events. Neuroscientific studies are yielding results that provide insight into how the risk of suicide may develop; ultimately demonstrating how suicide can be prevented. (shrink)

We present cut-free labelled sequent calculi for a central formalism in logics of agency: STIT logics with temporal operators. These include sequent systems for Ldm , Tstit and Xstit. All calculi presented possess essential structural properties such as contraction- and cut-admissibility. The labelled calculi G3Ldm and G3Tstit are shown sound and complete relative to irreflexive temporal frames. Additionally, we extend current results by showing that also Xstit can be characterized through relational frames, omitting the use of BT+AC frames.

The two great modern naturalists, Linnaeus and Darwin, expressed their intuition about how best to visualize patterns of affinities, that is, morphological similarities and divergences between taxa. Linnaeus suggested that “all plants show affinities on all sides, like a territory on a geographical map,” while Darwin thought that it was virtually impossible to understand the affinities between living and extinct species without a genealogical tree. Genealogical trees follow the diachronic, evolving logic of a timeline, whereas maps depict a synchronous pattern (...) of extant taxa. Although the two seem unrelated, various naturalists made attempts to combine them. Surprisingly, these resulted in three-dimensional images that, in order to be observed, had to be projected on paper. The naturalists Max Fürbringer and Richard Bowdler Sharpe were aware of this fact, but even Darwin himself twice combined the basic intuitions underlying the two modes of representation to produce three-dimensional images. This article is a brief history of the efforts to merge genealogical trees and map-like cross sections of affinities into one three-dimensional image. (shrink)

The sole diagram in On the Origin of Species is generally considered to be merely an illustration of Darwin’s ideas, but such an interpretation ignores the fact that Darwin himself expressly stated that the diagram helped him to discover and express his ideas. This article demonstrates that developing the so-called “tree diagram” substantially aided Darwin’s heuristics. This demonstration is based on an interpretation of the diagram and of 17 sketches found in Darwin’s scientific papers. The key to this interpretation is (...) the meaning that Darwin assigned to the graphic elements he used to construct the preliminary sketches and the diagram. I argue that each of the sketches contributed to the shaping of Darwin’s ideas and that, in their succession, each added new elements that ultimately resulted in the fully developed published diagram. (shrink)

In this work we answer a long standing request for temporal embeddings of deontic STIT logics by introducing the multi-agent STIT logic TDS . The logic is based upon atemporal utilitarian STIT logic. Yet, the logic presented here will be neutral: instead of committing ourselves to utilitarian theories, we prove the logic TDS sound and complete with respect to relational frames not employing any utilitarian function. We demonstrate how these neutral frames can be transformed into utilitarian temporal frames, while preserving (...) validity. Last, we discuss problems that arise from employing binary utility functions in a temporal setting. (shrink)

This paper examines the conceptual and terminological overlap between theories and models of practical deliberation developed within the fields of Practical Reasoning (PR) and Practical Argumentation (PA). It carefully delineates the volitional, epistemic, normative, and social commitments invoked and explicates various rationales for attributing the label ‘practical’ to instances of reasoning and argumentation. Based on these analyses, the paper develops a new approach to practical deliberation called the Stakeholder Commitment Approach (SCA). By distinguishing between ‘problem holder’ and ‘problem solver’, and (...) specifying the distributions of attributable commitments among the stakeholders, the SCA introduces an extension and refinement of the grounds for assigning the label ‘practical’ that brings PR and PA closer together. (shrink)

Centered around the analysis of the prescriptive portion of the Vedas, the Sanskrit philosophical school of Mīmāṃsā provides a treasure trove of normative investigations. We focus on the leading Mīmāṃsā authors Prabhākara, Kumārila and Maṇḍana, and discuss three modal logics that formalize their deontic theories. In the first part of this paper, we use logic to analyze, compare and clarify the various solutions to the _śyena_ controversy, a two-thousand-year-old problem arising from seemingly conflicting commands in the Vedas. In the second (...) part, the formalized Mīmāṃsā theories are analyzed and employed to provide alternative perspectives on well-known paradoxes from the contemporary field of deontic logic. Thus, we go from logic to Mīmāṃsā and back again. (shrink)

We formally introduce a novel, yet ubiquitous, category of norms: norms of instrumentality. Norms of this category describe which actions are obligatory, or prohibited, as instruments for certain purposes. We propose the Logic of Agency and Norms (LAN) that enables reasoning about actions, instrumentality, and normative principles in a multi-agent setting. Leveraging LAN , we formalize norms of instrumentality and compare them to two prevalent norm categories: norms to be and norms to do. Last, we pose principles relating the three (...) categories and evaluate their validity vis-à-vis notions of deliberative acting. On a technical note, the logic will be shown decidable via the finite model property. (shrink)

Modern democratic states are increasingly adopting new information and communication technologies to enhance the efficiency and quality of public administration, public policy and services. However, there is substantial variation in the extent to which countries are successful in pursuing such public digitalization. This paper zooms in on the role of social trust as a possible account for the observed empirical pattern in the range and scope of public digitalization across countries. Our argument is that high social trust makes it easier (...) to digitalize the public sector to the benefit of overall society. We develop a simple game-theoretical model that specifies this positive externality from a high-trust culture and provide empirical evidence that a higher level of social trust is associated with a higher level of public digitalization. Of course, high-trust countries cannot do without the rule of law and some institutionalized mechanisms of control and accountability. However, from our argument, it follows that one future challenge could be that control crowds out trust, so that the balance is no longer optimal. If such overcontrol occurs, public digitalization win–win situations between citizens and government can be undermined, even in high-trust countries. (shrink)

We present a logic of agency called LAE whose language includes propositional constants for actions and expectations. The logic is based on Von Wright’s theory of agency in general and his analysis of instrumentality in particular. An axiomatization of the logic, including an independence of agents axiom, is provided and soundness and completeness are shown with respect to its intended class of frames. The framework of LAE will allow us to formally define a manifold of concepts involved in agency theories, (...) including Von Wright’s four elementary forms of action, the notion of forbearance and notions of instrumentality that make reference to an agent’s expectations. (shrink)