Baltic International Yearbook of Cognition, Logic and Communication

A Logician's Sidelong Glance at Irony

Reinhard Kahle — Sat, 10 Nov 2018 13:41:49 PST

In "Irony as Expression (of a Sense of the Absurd)" Mitchell Green is presenting an interesting account of communicative irony where ``we express a sense of a situation's absurdity (wackiness, goofiness, etc.).'' In this line of argument, he is questioning the adequateness of irony as meaning-inversion and irony as conversational implicature. In this note, we would like to take the idea of absurdity a little bit further, considering it in its logical sense. As a consequence we can offer a possibility to defend, at least partially, irony as meaning-inversion and conversational implicature.

Irony as Expression (of a Sense of the Absurd)

Mitchell Green — Fri, 11 May 2018 19:09:48 PDT

Situational irony is, first, explained as a severe violation of one or more established, non-moral norms; such violation constitutes that situation’s absurdity. The classical “inversion” theory of communicative irony associated with Cicero and Quintilian, as well as its refinement in terms of the notion of conversational implicature (Grice 1989), are then shown to be inadequate.The echoic (Sperber (1984), Wilson (2006), Wilson & Sperber (2012)) and pretence (Currie 2010) theories are also shown to fail to account for the broad range of communicative irony, although they each contain valuable insights. Further, both theories hold that ironic speakers express attitudes but do not explain how they do so. On the basis of prior work by Green conceptualizing the notion of expression as signaling and showing a psychological state, we defend a view of communicative irony as expressing a sense of a situation’s absurdity. The view generalizes beyond absurdity to encompass expression of a sense of situations’ silliness, wackiness, or goofiness, and accommodates milder forms of irony such as we find in meiosis.

Mass/Count Variation: A Mereological, Two-Dimensional Semantics

Peter R. Sutton et al. — Thu, 08 Dec 2016 03:23:12 PST

We argue that two types of context are central to grounding the semantics for the mass/count distinction. We combine and develop the accounts of Rothstein (2010) and Landman (2011), which emphasize (non-)overlap at a context. We also adopt some parts of Chierchia’s (2010) account which uses precisifying contexts. We unite these strands in a two-dimensional semantics that covers a wide range of the puzzling variation data in mass/count lexicalization. Most importantly, it predicts where we should expect to find such variation for some classes of nouns but not for others, and also explains why.

The Semantic Role of Classifiers in Japanese

Yasutada Sudo — Thu, 08 Dec 2016 03:23:04 PST

In obligatory classifier languages like Japanese, numerals cannot directly modify nouns without the help of a classifier. It is standardly considered that this is because nouns in obligatory classifier languages have ‘uncountable denotations’, unlike in non-classifier languages like English, and the function of classifiers is to turn such uncountable denotations into something countable (Chierchia 1998a,b, Krifka 2008, among many others). Contrary to this view, it is argued that what makes Japanese an obligatory classifier language is not the semantics of nouns but the semantics of numerals. Specifically, evidence is presented that numerals in Japanese cannot function as predicates on their own, which is taken as evidence that the extensions of numerals in Japanese are exclusively singular terms. It is then proposed that the semantic function of classifiers is to turn such singular terms into modifiers/predicates. It is furthermore claimed that the singular terms denoted by numerals are abstract entities (cf. Rothstein 2013, Scontras 2014a,b), and proposed that the reason why they cannot have modifier/predicate uses in obligatory classifier languages like Japanese is because the presence of classifiers in the lexicon blocks the use of a type-shifting operator that turns singular terms denoted by numerals into predicates (cf. Chierchia 1998a,b).

Classifiers and Plurality: evidence from a deictic classifier language

Filomena Sandalo et al. — Thu, 08 Dec 2016 03:22:56 PST

This paper investigates the semantic contribution of plural morphology and its interaction with classifiers in Kadiwéu. We show that Kadiwéu, a Waikurúan language spoken in South America, is a classifier language similar to Chinese but classifiers are an obligatory ingredient of all determiner-like elements, such as quantifiers, numerals, and wh-words for arguments. What all elements with classifiers have in common is that they contribute an atomized/individualized interpretation of the NP. Furthermore, this paper revisits the relationship between classifiers and number marking and challenges the common assumption that classifiers and plurals are mutually exclusive.

Counting and Measuring: a theoretical and crosslinguistic account

Susan Rothstein — Thu, 08 Dec 2016 03:22:46 PST

In this paper, I show that expressions like two glasses of wine are ambiguous between counting and measuring interpretations, and that each interpretation is associated with a different semantic representation. In each interpretation, glasses has a different function. In the counting interpretation, glasses is a relational noun, while in the measure interpretation, glasses is a measure head analogous to litre. This difference leads to a number of grammatical contrasts which can be explained by differences in the grammatical structure. I discuss whether these differences are only semantic or also expressed in the syntactic representation. The assumption that syntax directly reflects semantic interpretation leads to assigning counting NPs and measuring NPs two different syntactic structures: counting NPs are right-branching with two modifying glasses of wine, while in measure expressions the numeral and the measure head form a measure predicate two glasses which modifies the N. I show that in Modern Hebrew and Mandarin counting structures and measuring structures clearly do have different syntactic structures, reflecting the semantic differences between counting and measuring. While the evidence in the case of English is less strong, the assumption that syntax directly reflects compositional syntactic structure results in the same basic syntactic contrasts in English as well.

Container Constructions in Yudja: locatives, individuation and measure

Suzi Lima — Thu, 08 Dec 2016 03:22:37 PST

The possible interpretations of container phrases (such as ‘cups of sugar’) has been long debated in the formal semantics literature because container phrases can be associated with a variety of possible readings that go from individuation to measure. In this paper we explore the interpretation of container phrases in Yudja (Tupi stock, Brazil), a language where container phrases are optional in construction with numerals and are morphosyntactically identical to locative phrases. Based on experimental studies with Yudja children and adults we intend to show that these expressions are ambiguous in at least three ways (locative, individuation and measure) and that a locative reading might emerge even in scenarios where the verb and the context favor a measure interpretation. Furthermore, this paper provides evidence that there is no hidden container phrase when numerals are combined with notional mass nouns and that, supporting Partee and Borschev (2012), the results of the studies show that, indeed, the individuation reading is more “primitive”, i.e. it precedes measuring in language acquisition.

Iceberg Semantics For Count Nouns And Mass Nouns: Classifiers, measures and portions

Fred Landman — Thu, 08 Dec 2016 03:22:28 PST

The background for this paper is the framework of Boolean semantics for mass and count nouns, and singular and plural count nouns, as developed from the work of Godehard Link in Link 1983 (see e.g. the expositions in Landman 1991, 2010).

Link-style Boolean semantics for nouns (here called Mountain semantics) analyzes the oppositions mass-count and singular-plural in terms of the notion of atomicity: counting is in terms of singular objects, which are taken to be atoms. Consequently, Link bases his semantics on two separate Boolean domains: a non-atomic mass domain and an atomic count domain. Singular count nouns are interpreted as sets of atoms, and semantic plurality is closure under sum, so plural objects are sums of atoms.

In this, sorted setup portions - like two portions of soup - are a puzzle: they are mass stuff - soup -, but count - two. But in order to be count they must be atoms. But they are not, because they are just soup. Mountain semantics can deal with portions, but at a cost.

In the first part of this paper I outline Iceberg semantics, an alternative to Mountain semantics within the general framework of Boolean semantics.

Iceberg semantics specifies a compositional mechanism which associates with the standard denotation of any noun phrase (here called the body) a base set, a set that generates the body under the sum operation ⊔. For count nouns, the base is the set in terms of which the members of the body are counted and to which distribution takes place. In Iceberg semantics, what allows counting to be correct is the requirement on the interpretations of count nouns that the base of their interpretation is (contextually) disjoint.

Already at this level we see two salient properties of Iceberg semantics:

-Atoms and atomicity play no role in the theory, so we can assume an unsorted interpretation domain for mass nouns and count nouns. In Iceberg semantics, mass and count can be seen as different perspectives on the same stuff (different bases for the same body). This means that we can do away with the extreme body-sorting and body-gridding that atomicity entails.

With this we allow a simpler and more elegant analysis of mass-count interactions. For instance, portions can just be 'mass' stuff, evaluated relative to a count base.

-The mass-count distinction is formulated in terms of disjointness of the base. Iceberg semantics associates bases not just with the interpretations of lexical nouns, but with NPs in general and with DPs. This means that Iceberg semantics provides a compositional semantic theory of the mass-count distinction, and hence it provides a framework in which the mass-count nature of complex NPs and of DPs can be fruitfully studied.

It is the analysis of complex NPs and their mass-count properties that is the focus of the second part of this paper. There I develop an analysis of English and Dutch pseudo- partitives, in particular, measure phrases like three liters of wine and classifier phrases like three glasses of wine. We will study measure interpretations and classifier interpretations of measures and classifiers, and different types of classifier interpretations: container interpretations, contents interpretations, and - indeed - portion interpretations. Rothstein 2011 argues that classifier interpretations (including portion interpretations) of pseudo partitives pattern with count nouns, but that measure interpretations pattern with mass nouns. I will show that this distinction follows from the very basic architecture of Iceberg semantics.

Functional Unit Classifiers in (Non)-Classifier Russian

Keren Khrizman — Thu, 08 Dec 2016 03:22:20 PST

It has often been argued that functional individuating classifiers and plural count nouns ought to be in complementary distribution (e.g. Borer 2005, Chierchia 2010). This apparently works neatly for Chinese and English. Russian, however, is an interesting case. On the one hand it has count nouns which can be directly modified by numerals. On the other hand it has three classifiers, štuka ‘item’, čelovek ‘person’ and golova ‘head’, which optionally occur in numeral constructions with plural nouns and look very much like functional individuating classifiers (cf. Sussex 1976, Yadroff 1999). I show that a closer look at the data reveals that apparently count constructions using these optional classifiers have properties of measure constructions such as five liters of water. Based on that I argue that these classifiers are not individuating classifiers but are measure words which measure mass denotations in terms of natural units in the sense of Krifka (1989, 1995).

Crime Investigations: The Countability Profile of a Delinquent Noun

Scott Grimm — Thu, 08 Dec 2016 03:22:12 PST

This paper aims to broaden our understanding of countability beyond what is found with concrete nouns, providing a one-word case study of the countable and non-countable uses of the noun crime. I show that the behavior of crime runs counter to a variety of expectations inherited from the literature on countability: its countable use cannot be directly grounded in atomic acts or events, nor is its non- countable use simply equivalent to a plural individual composed of individual crimes, as one might expect on analogy with certain analyses of furniture. Additionally, while crime has a use as a bare plural, that use does not refer to a kind. A quantitative study supports these conclusions. Altogether, crime demonstrates a novel noun type with respect to its nominal semantics and countability behavior, which is also an indication of the large empirical terrain that awaits exploration for eventive and abstract nouns.

The Semantics of Motion Verbs in Russian

Maria Gepner — Thu, 08 Dec 2016 03:22:03 PST

Within the group of imperfective motion verbs in Russian there exists a further subdivision into determinate and indeterminate verbs. Traditionally the distinction is said to lie in the direction of motion the verbs encode: motion in one direction or in different directions. In this paper I am going to argue that this distinction is not enough. I will claim that determinate verbs encode singular eventualities and indeterminate verbs are pluractional. Thus in the normal case, imperfective verbs are plural predicates which include singular and plural events in their denotations, in the case of motion verbs, imperfective denotations are subdivided into a singular and a pluractional predicate.

Bare Nouns in Brazilian Portuguese: An experimental study on grinding

Kayron Beviláqua et al. — Thu, 08 Dec 2016 03:21:53 PST

Much literature has explored the interpretation of the bare singular (BS) in Brazilian Portuguese. Pires de Oliveira and Rothstein (2011) claim that BS nouns are mass because they denote kinds and argue that this explains why only the BS in Brazilian Portuguese can have a non-cardinal interpretation. In this paper, based on an experimental task with Brazilian Portuguese adult speakers, we explore one of their predictions, namely that the ‘volume interpretation’ of the BS cannot be explained as a case of Grinding. Our results show that Grinding and Volume readings of a BS noun are not equivalent (in favor of their hypothesis). We also show that a volume interpretation of a noun is never preferred when a cardinal interpretation is available, but that this can be explained by other lexical and pragmatic factors. We conclude by suggesting that Rothstein’s (in press) distinction between counting and measuring accounts for the fact that non-cardinal readings are not grinding.

Editors' Introduction

Susan Rothstein et al. — Thu, 08 Dec 2016 03:21:43 PST

This volume contains a selection of the papers presented at the 11^th International Symposium on Cognition, Logic and Communication which took place in Riga, at the University of Latvia on December 10-11 2015. The choice of topic reflected a growing understanding in the community of linguists and cognitive scientists that fundamental grammatical features of language, in particular the mass/count distinction, use of number words, and plurality, reflect our grasp of non-linguistic numerical operations, in particular individuation and measurement.

Describing Images using a Multilayer Framework based on Qualitative Spatial Models

Tao Wang et al. — Thu, 10 Dec 2015 14:55:22 PST

To date most research in image processing has been based on quantitative representations of image features using pixel values, however, humans often use abstract and semantic knowledge to describe and analyze images. To enhance cognitive adequacy and tractability, we here present a multilayer framework based on qualitative spatial models. The layout features of segmented images are defined by qualitative spatial models which we introduce, and represented as a set of qualitative spatial constraints. Assigned different semantic and context knowledge, the image segments and the qualitative spatial constraints are interpreted from different perspectives. Finally, the knowledge layer of the framework enables us to describe the image in a natural way by integrating the domain-specified semantic constraints and the spatial constraints.

On the Polysemy Of the Lithuanian Už. A Cognitive Perspective

Inesa Šeškauskienė et al. — Thu, 10 Dec 2015 14:55:20 PST

Adhering to the principle of motivated polysemy, this paper sets out to demonstrate how the principle works in interpreting numerous senses of the Lithuanian preposition už ‘behind, beyond’. The present investigation relies on the cognitive linguistic framework employed, first of all, by Lakoff (1987), Langacker (1987), Talmy (2000), Tyler and Evans (2003), and Tyler (2012), who mainly worked on English, and such linguists as Tabakowska (2003, 2010) and Shakhova and Tyler (2010), who attempted to investigate inflecting languages, such as Polish and Russian. Based on such semantic principles as types of Figure and Ground, their relationship (geometric, functional, etc.), contextual clues and pattern of usage, etc., the present paper demonstrates that the polysemy of už used with two cases, Genitive and Accusative, is not an array of arbitrary senses, but rather a motivated network. It posits a central sense of už based on Figure located in the back region of Ground. All other senses, namely, those of function, control, obstacle, sequential location, hiding and covering, boundary or border, spatial distance, temporal distance, quality distance, replacement, retribution and remuneration, and benefactive, are directly or indirectly derived from the central sense.

Language, Culture and Spatial Cognition: Bringing anthropology to the table

Norbert Ross et al. — Thu, 10 Dec 2015 14:55:19 PST

Languages vary in their semantic partitioning of the world. This has led to speculation that language might shape basic cognitive processes. Spatial cognition has been an area of research in which linguistic relativity – the effect of language on thought – has both been proposed and rejected. Prior studies have been inconclusive, lacking experimental rigor or appropriate research design. Lacking detailed ethnographic knowledge as well as failing to pay attention to intralanguage variations, these studies often fall short of defining an appropriate concept of language, culture, and cognition. Our study constitutes the first research exploring (1) individuals speaking different languages yet living (for generations) in the same immediate environment and (2) systematic intralanguage variation. Results show that language does not shape spatial cognition and plays at best the secondary role of foregrounding alternative possibilities for encoding spatial arrangements.

Antonymy In Space And Other Strictly Ordered Domains

Jessica Rett — Thu, 10 Dec 2015 14:55:17 PST

Natural language references different types of entities. Some of these entities (e.g. degrees, locations, times) are strictly ordered with respect to one another; others (e.g. individuals, possible worlds) are not. The empirical goal of this paper is to show that some linguistically encoded relations across these domains (e.g. under, slower than) display a polar asymmetry, while others do not. The theoretical goal of this paper is to argue that this asymmetry – and its restriction to only certain relations – is due to intrinsic properties of strictly ordered domains, coupled with a bias in how language users perceive these domains.

The Lay of the Land: Sensing and Representing Topography

Nora S. Newcombe et al. — Thu, 10 Dec 2015 14:55:16 PST

Navigating, and studying spatial navigation, is difficult enough in two dimensions when maps and terrains are flat. Here we consider the capacity for human spatial navigation on sloped terrains, and how sloping terrain is depicted in 2D map representations, called topographic maps. First, we discuss research on how simple slopes are encoded and used for reorientation, and to learn spatial configurations. Next, we describe how slope is represented in topographic maps, and present an assessment (the Topographic Map Assessment), which can be administered to measure topographic map comprehension. Finally, we describe several approaches our lab has taken with the aim of improving topographic map comprehension, including gesture and analogy. The current research reveals a rich and complex picture of topographic map understanding, which likely involves perceptual expertise, strong spatial skills, and inferential logic.

A Description Of Space Relations In An NLP Model: The ABBYY Compreno Approach

Aleksey Leontyev et al. — Thu, 10 Dec 2015 14:55:15 PST

The current paper is devoted to a formal analysis of the space category and, especially, to questions bound with the presentation of space relations in a formal NLP model. The aim is to demonstrate how linguistic and cognitive problems relating to spatial categorization, definition of spatial entities, and the expression of different locative senses in natural languages can be solved in an artificial intelligence system. We offer a description of the locative groups in the ABBYY Compreno formalism – an integral NLP framework applied for machine translation, semantic search, fact extraction, and other tasks based on the semantic analysis of texts. The model is based on a universal semantic hierarchy of the thesaurus type and includes a description of all possible semantic and syntactic links every word can attach. In this work we define the set of semantic locative relations between words, suggest different tools for their syntactic presentation, give formal restrictions for the word classes that can denote spaces, and show different strategies of dealing with locative prepositions, especially as far as the problem of their machine translation is concerned.

Aspects of Space

Marcus Kracht — Thu, 10 Dec 2015 14:55:13 PST

It is argued that spatial expressions come together with an encoding of the space called "aspect", which changes as we climb up the syntactic tree. The changing nature of aspect is necessary in order to simplify the meanings of elements. What appears to be a rather peculiar property of an element will be perfectly natural once we acknowledge that the elements compute on the space viewed in a particular way. Coordinates are always rooted in the landmark, for example. Thus, for the purpose of the distinction between static and dynamic it is not the "absolute" motion of the figure that counts, but the motion relative to the landmark.