Argument Structure and lexical semantics

WordNets mostly deal with lexicalized expressions and lexical-semantic relations among them. Concepts are represented by sets of synonyms (synsets), which constitute the edges of the network. Each synset includes the lexicalized expressions that correspond to ...

In Russian, the prepositions o and про cover partially overlapping semantic areas. Both prepositions can express Topic and Content with speech and mental predicates. When expressing Content, the differences between the prepositions are mitigated, whereas when expressing Topic, the differences become more noticeable. The preposition о more often introduces general topics («говорить о политике» –‘to talk about politics’), while про tends to introduce more specific topics («говорить про Машу» –‘to talk about Masha’). Because of this, о frequently accompanies noun phrases with a generic status and is used with imperfective verbs, while про introduces noun phrases with a specific referential status and is used with perfective verbs.

Their key non-overlapping usages are as follows. Про, but not о, can
metonymically reference a situation (e.g., «Он рассказал про Машу» –‘He
told a story about Masha,’ meaning ‘what happened to Masha’).
Conversely, о, but not про, is used in contexts where it conveys the
meaning of indicating or providing evidence (e.g., «Симптомы говорят о
воспалении» –‘The symptoms indicate inflammation’). Additionally, о
governs Stimulus valency in emotive predicates (e.g., «скучать о» – ‘to
long for’) and emotive behavior predicates (e.g., «заботиться о» – ‘to take
care of’).

Die Sprecher können dadurch auf einer bipolaren Achse ("Tradition" / "Modernität", vgl. HUCK, 1998, 225 sqq.) eingestuft werden, je nach den eingesetzten Sprachmitteln ihres Repertoires. Die meisten Mundartsprecher im Elsass benutzen auch regelmäßig und oft ziemlich intensiv Französisch, können (oder benutzen) auch z.T. Deutsch, sodass keine wirklichen Grenzen zwischen den sprachlichen Ressourcen entstehen (die Stil-Fragen mit einbegriffen). Das ganze muss als ein Kontinuum von sprachlichen Ressourcen angesehen werden, die der Sprecher je nach Bedarf, Situation, Ansprechpartner, Intention, etc. einsetzt.

This chapter reviews the fundamental classifications of nouns in Romance languages, as well as the criteria by which they are established. The main cross-linguistic differences to which these divisions give rise are also analysed, together with the fundamental controversies that are usually recognised in each of them. The main distinctions addressed are 'common-proper', 'count-mass', and 'individual-collective', but other less traditional distinctions among noun classes (such as 'argumental/non-argumental' or 'eventive/non-eventive') are also considered. The chapter pays special attention to the syntactic consequences of lexical distinctions. It concludes with an overview of class switching processes, in which lexical recategorisations are distinguished from syntactic ones.

We present a dependency annotation scheme for Finnish which aims at respecting the multilayered nature of language. We first tackle the annotation of surfacesyntactic structures (SSyntS) as inspired by the Meaning-Text framework. Exclusively syntactic criteria are used when defining the surface-syntactic relations tagset. Our annotation scheme allows for a direct mapping between surface-syntax and a more semantics-oriented representation, in particular predicate-argument structures. It has been applied to a corpus of Finnish, composed of 2,025 sentences related to weather conditions.

espanolEn este trabajo, argumentamos que en los actuales treebanks que aplican el formalismo de las Dependencias Universales, la anotacion de los reflexivos espanoles es un problema sin resolver, que afecta claramente a la precision y consistencia de los parsers actuales. Evaluamos diferentes propuestas para afinar las diferentes categorias y discutimos los problemas pendientes. Creemos que la solucion para estos problemas se puede encontrar en una anotacion en multiples niveles, combinando la anotacion de la relacion de dependencia y de las caracteristicas (features) de los tokens, en lugar de ampliar el numero de categorias en un solo nivel de anotacion. Aplicamos la propuesta a la version espanola del treebank UD AnCora (v2.5) y proporcionamos una tabla de conversion categorizada que se puede ejecutar mediante un script Python. EnglishIn this paper, we argue that in current Universal Dependencies treebanks, the annotation of Spanish reflexives is an unsolved problem, which clearl...

We present a dependency annotation scheme for Finnish which aims at respecting the multilayered nature of language. We first tackle the annotation of surfacesyntactic structures (SSyntS) as inspired by the Meaning-Text framework. Exclusively syntactic criteria are used when defining the surface-syntactic relations tagset. Our annotation scheme allows for a direct mapping between surface-syntax and a more semantics-oriented representation, in particular predicate-argument structures. It has been applied to a corpus of Finnish, composed of 2,025 sentences related to weather conditions.

This paper examines argument alternation constructions in the Bantu language, Kiwoso. The data demonstrates that alternation is constrained by selectional properties of a root and its combination with different functional heads, which is language specific. The findings establish that external arguments of anticausative, passive and middle alternations are distinct in terms of syntax and semantics. The data indicates that external arguments appear as DPs in causative alternates but in anticausative, passive and middle sentences they surface as PPs, albeit with different interpretations. In passives, the PPs are introduced by a na-phrase and express event participants, while in anticausatives and middles they are introduced by a kophrase which denotes event modifiers. It is also established that anticausative, passive and middle verb constructions are related in that their sole subject argument is the object argument of their causative (transitive) variants.

The concept of centre of attention (CA) is used in Distributed Grammar focusing on its relevance for the syntax of human languages. Based on psychological evidence, this notion captures what is common between subject, topic and theme in an integrated system of concepts rather than as a disparate collection of them. We define respectively subject as the main CA of a base utterance, topic as the main CA of an extended utterance (containing both Old and New meta-informative status) and theme as the main (composite) CA of a text/discourse. When choosing an entity or an element of the semantic situation as the CA of the utterance, a speaker creates a common ground on which it becomes possible to communicate with the addressee. 1. The problem of Subject in Logic and Grammar: Subject or Argument? In ancient logic, a proposition was considered to be composed of two terms: the subject and the predicate (SP). In Aristotle‟s metaphysics, this two-fold definition of a proposition led to the ont...

We develop a formal account of the syntactic and semantic properties of Spanish adnominal PP modifiers headed by sin ('without') where the preposition selects for a singular or plural bare noun as complement—henceforth sin-PPs—,(1). Specifically, we investigate: a) the syntactic-semantic properties of pseudo-incorporation, and b) the interaction between countability, degree quantification and negation in these structures.(1) un cinturón sin hebilla (lit.'a belt without clasp'), un hombre sin corbata (lit.'a man without tie')--una ...

QU'EST-CE AU JUSTE QUE LA PRÉDICATION ? "All that I regret is that neither Russell nor Frege recognised a naming role for common nouns like 'man' or 'philosopher'. It would be highly desirable if someone succeeded in formulating a logical theory which satisfied this Aristotelian requirement but did not blur the distinction between names and predicables. To my mind such a theory would be an integral restoration of that was lost in Aristotle's Fall: it would be like Paradise Regained." (Peter Thomas Geach 1971 and 1979) "… one important function of language is to serve as an attention-directing system."

Dans ce livre, nous aborderons un ensemble des théories syntaxiques modernes  qui ont visé à décrire et à analyser la construction et l’organisation des signes linguistiques à l’intérieur de la phrase. Ces théories sont nées du structuralisme saussurien qui voit la langue comme un système doté d'une structure décomposable. Pour Saussure la langue est un système au sein duquel chaque signe linguistique n'est défini que par ses relations qu'il entretient avec les autres, relations d'équivalence ou d'opposition. Cet ensemble de relations forment ce qu’on appelle une structure.

Cette étude montre que le système de la langue française contient une large panoplie de moyens linguistiques, combinables entre eux, qui permettent de construire la mise en relief du contenu d'un énoncé publicitaire. Outre la segmentation, la présence du pronom personnel indéfini « on », des déictiques ainsi que l'ajout de certains types d'adverbes, notamment intensifs et focalisateurs jouent un rôle syntactico-pragmatique de premier rang. De même, la présence des marqueurs totalisants comme tout(e) et tous renforce l'effet d'intensification.

Where does a verb's frame come from? The obvious answer is the verb itself, and this is the answer that syntacticians have traditionally provided, whether they describe predicator-argument relations as syntactic sisterhood relations or as lexical properties (the predicator's combinatoric potential, or valence). Thus, Haegeman, in her introduction to Government and Binding theory, states, "the thematic structure of a predicate, encoded in the theta grid, will determine the minimal components of the sentence" (Haegeman 1994: 55). Similarly, Bresnan, in her introduction to Lexical Functional Grammar (LFG), states, "[o]n the semantic side, argument structure represents the core participants and events (states, processes) designated by a single predicator. […] On the syntactic side, argument structure represents the minimal information needed to characterize the syntactic dependents of an argument-taking head" (Bresnan 2001: 304). In lexicalist theories like LFG, whenever the arguments of a verb can have more than one set of syntactic realizations, each distinct realization pattern corresponds to a different mapping from semantic roles to grammatical functions, as expressed in a unique lexical entry, and lexical entries, or classes of lexical entries, are related by lexical rules (Neidle 1994). The drive to streamline lexical entries by removing predictable properties has led theorists to develop more general, putatively universal, mapping principles, as well as principles for deriving

This paper uncovers how metaphor forces argument overtnessacross languages and parts of speech. It addresses the relationship between semantically unsaturated terms, functors, and the argument terms that complete them. When the component terms' default senses clash semantically, a metaphor arises. In such cases, the argument must be overt, in contrast to literal uses. It is possible to say Everyone was waiting at the hotel. Finally, Kim arrived. By contrast, people do not use arrived metaphorically without a goal argument: Everything had been pointing to that conclusion all along. *Finally, Kim arrived. What they say is Finally, Kim arrived at it. We illustrate the phenomenon with powerful and diverse evidence: three corpus studies (Indo-Aryan languages, British English, Vera'a) and a sentence-completion experiment with around 250 native speakers of English. Both the corpus studies and the experiment show no or almost no exceptions to metaphor-driven argument overtness. The strength of the effect contrasts with a complete lack of speaker awareness. We propose that metaphor-driven argument overtnessas well as the lack of speaker consciousnessis a universal phenomenon that can be accounted for in terms of human language processing.

Modalität, Verarbeitung visuell-gestisch aural-oral Artikulatoren grobmotorisch manuell (Hände), nicht-manuell (Mund, Gesicht, Körper) feinmotorisch Vokaltrakt Dimension der Verarbeitung vertikal horizontal Serialität in Perzeption und Produktion Periphere zeitliche Auflösung gering 25-30 ms hoch 2 ms Simultaneität in Perzeption und Produktion Geschwindigkeit der Übertragung hoch 230 km/s gering 331 m/s

Este artículo examina la teoría temática, un módulo en el marco del modelo gramatical de Principios y Parámetros. Por esta razón, se considera necesario considerar los niveles de representación, el léxico, las categorías léxicas y funcionales, la teoría temática, el criterio- Ө, los papeles o roles temáticos y la estructura o red argumental de los predicados. También menciona algunas ventajas y dificultades de considerar roles temáticos que planteó Bosque (2009), y comenta la propuesta restrictiva de Guillermo Lorenzo González (1992).

All that I regret is that neither Russell nor Frege recognised a naming role for common nouns like 'man' or 'philosopher'. It would be highly desirable if someone succeeded in formulating a logical theory which satisfied this Aristotelian requirement but did not blur the distinction between names and predicables. To my mind such a theory would be an integral restoration of that was lost in Aristotle's Fall: it would be like Paradise Regained." (Peter Thomas Geach 1971 and 1979) "… one important function of language is to serve as an attention-directing system." (Marlene Taube-Schiff and Norman Segalowitz, 2005

This dissertation is about the nature of Japanese complex predicates and the notion 'word'. The main theme of this work is that the notion 'word' must be relativized to at least three different senses: morphological, grammatical, and semantic (cf. Bresnan & Mchombo 1995). The word is an atomic unit in grammar (Di Sciullo & Williams 1987), but this atomicity can be defined in different ways. The notion 'word' can be defined in terms of its morphological integrity: the word is a unit whose parts cannot be separated from the rest. It can also be defined in terms of grammatical-functional properties: in the case

We explore the annotation of reflexives in Universal Dependencies (UD) 2.2 treebanks (Nivre et al., 2018), with a stronger focus on Slavic languages. We have tried to find out if the current guidelines are transparent and clear enough for the annotators to follow them successfully. We point out a number of inconsistencies in the current annotation across languages, and propose improvements—sometimes of the guidelines, but mostly of the annotation. The goal of the paper is to contribute to more consistent annotation of reflexives in future releases of UD, which, in turn, will enable broader cross-linguistic studies of this phenomenon.

In this paper, I argue in favor of the view that a generative lexicon, as advanced in lexicalist theories, might not be needed to derive verbal's argument structure. I instead support the hypothesis that a verb's meaning emerges as a result of the syntactic structure in which it is merged and that the role of lexical items/roots reduces to their idiosyncratic encyclopedic content. These two assumptions are executed via adopting the proposal that splits the traditional VP structure into two main functional heads, namely VoiceP and vP, and via endorsing the architectural assumptions advanced in the framework of Distributed Morphology. The main empirical support for this claim comes from verbs that appear in syntactic structures that are not in consonance with their semantic-conceptual content, Arabic varieties that lost their vocalic melodies that would otherwise encode thematic roles, and spray-load alternation. This paper concludes by exploring language-particular processes whose non-applicability goes beyond morphology, the analysis of which supports the role of the Encyclopedia.

We propose a general model for joint inference in correlated natural language processing tasks when fully annotated training data is not available, and apply this model to the dual tasks of word sense disambiguation and verb subcategorization frame determination. The model uses the EM algorithm to simultaneously complete partially annotated training sets and learn a generative probabilistic model over multiple annotations. When applied to the word sense and verb subcategorization frame determination tasks, the model learns sharp joint probability distributions which correspond to linguistic intuitions about the correlations of the variables. Use of the joint model leads to error reductions over competitive independent models on these tasks.

We explore the annotation of reflexives in Universal Dependencies (UD) 2.2 treebanks (Nivre et al., 2018), with a stronger focus on Slavic languages. We have tried to find out if the current guidelines are transparent and clear enough for the annotators to follow them successfully. We point out a number of inconsistencies in the current annotation across languages, and propose improvements—sometimes of the guidelines, but mostly of the annotation. The goal of the paper is to contribute to more consistent annotation of reflexives in future releases of UD, which, in turn, will enable broader cross-linguistic studies of this phenomenon.

Part-of-speech (POS) tagging is usually considered a solved problem in NLP, with state-of-the-art precision usually above 97% precision for most languages. However, these figures are often based on small and/or coarse POS tag sets. With highly granular tag sets, the number of ambiguous forms may increase significantly, as thiner distinctions may apply to the same word form. Portuguese is a highly inflected language, and several state-of-the-art POS taggers are already available. Nevertheless, some POS categories, which are key to later NLP tasks, remain a resilient POS-tagging problem, with suboptimal results. This is due to their level of ambiguity, particularly when a fully-fledged POS tag set is used, and also for their high frequency in texts. In Portuguese, clitic pronouns are one such categories. Correct disambiguation of personal pronouns is particularly relevant in this language, not only because they constitute a very frequent category in texts, but especially because of th...

Part-of-speech (POS) tagging is usually considered a solved problem in NLP, with state-of-the-art precision usually above 97% precision for most languages. However, these figures are often based on small and/or coarse POS tag sets. With highly granular tag sets, the number of ambiguous forms may increase significantly, as thiner distinctions may apply to the same word form. Portuguese is a highly inflected language, and several state-of-the-art POS taggers are already available. Nevertheless, some POS categories, which are key to later NLP tasks, remain a resilient POS-tagging problem, with suboptimal results. This is due to their level of ambiguity, particularly when a fully-fledged POS tag set is used, and also for their high frequency in texts. In Portuguese, clitic pronouns are one such categories. Correct disambiguation of personal pronouns is particularly relevant in this language, not only because they constitute a very frequent category in texts, but especially because of th...

This paper proposes a formal semantic approach to Slavic verbal prefixes. A hypothesis is put forward according to which a verbal prefix imposes a relation between two degrees on a scale. One is a degree associated with the event denoted by the verbal predicate. The other one is the standard of comparison, contributed by a linguistic expression that appears in the sentence, or by the context. The scale to which the prefix applies is most typically contributed by the linguistic environment in which the prefix appears. The paper demonstrates how the proposed analysis applies to two specific morphemes in Russian: the prefixes pod- and do-.

Phraseology in a narrow sense studies stable combinations, choosing the lexical meaning of a word as the main criterion of analysis. Phraseology in a broad sense is engaged in the study of phraseological units, taking into account their types, structure, semiotic functions.

All that I regret is that neither Russell nor Frege recognised a naming role for common nouns like 'man' or 'philosopher'. It would be highly desirable if someone succeeded in formulating a logical theory which satisfied this Aristotelian requirement but did not blur the distinction between names and predicables. To my mind such a theory would be an integral restoration of that was lost in Aristotle's Fall: it would be like Paradise Regained." (Peter Thomas Geach 1971 and 1979) "… one important function of language is to serve as an attention-directing system." (Marlene Taube-Schiff and Norman Segalowitz, 2005

The supportive attitude of my parents, Joke en Zach, my sister Nynke, and my friends, who kept me in touch with normal life, needs a special word of thanks. My family has always stimulated my interest in language and research, and, what is more important, they have always been there for me when I needed them. The same is true for my friend Jeroen de Boer, who definitely made finishing this thesis much easier. Thanks a lot! Even though you were in London (both 'floated' and 'at a distance') during the past year, I look back at this period with much pleasure.

All that I regret is that neither Russell nor Frege recognised a naming role for common nouns like 'man' or 'philosopher'. It would be highly desirable if someone succeeded in formulating a logical theory which satisfied this Aristotelian requirement but did not blur the distinction between names and predicables. To my mind such a theory would be an integral restoration of that was lost in Aristotle's Fall: it would be like Paradise Regained." (Peter Thomas Geach 1971 and 1979) "… one important function of language is to serve as an attention-directing system." (Marlene Taube-Schiff and Norman Segalowitz, 2005

Although neither theoretical nor computational linguists did provide sufficiently careful insight into the problem of semantic roles, recently some progress is being achieved in robotics (study of the simulation of human interaction), and mostly in multi-agent systems. Taking advantage of this motivation and applying it to the study of languages, I distinguish between various abstract ontological levels. Instead of using such concepts as agentive, objective, experiencer, etc., on the highest (generic) ontological level, I postulate generalised agents which are defined by the following ontological features, among others: (1) features of control (autonomy): goal and feedback, (2) features of emotion (character): desire and intention, (3) epistemic features (reason): belief and cognition, (4) communication features (language faculty): verbal and visual.
In accordance with such ontological concepts, natural and artificial entities are obviously suited to fulfil the semantic roles of agents and figures respectively in the widest sense of these terms. I further propose to distinguish between three classes of generic ontological roles, namely ACTIVE, MEDIAN or PASSIVE. Here are examples of generic roles: (1) active role (Initiator, Causer, Enabler, Benefactor, Executor, Stimulant, Source, Instigator etc.), (2) passive role (Terminator, Affect, Enabled, Beneficient, Executed, Experiencer, Goal, etc.) and (3) median role (Mediator, Instrument, Benefit, Motor, Means etc.). Figures can play quasi-active (Q-active) roles.

Valency slot filling is a semantic glue, which brings together the meanings of words. As regards the position of an argument in the dependency structure with respect to its predicate, there exist three types of valency filling: active (canonical), passive, and discontinuous. Of these, the first type is studied much better than the other two. As a rule, canonical actants are unambiguously marked in the syntactic structure, and each actant corresponds to a unique syntactic position. Linguistic information on which syntactic function an actant might have (subject, direct or indirect object), what its morphological form should be and which prepositions or conjunctions it requires, can be given in the lexicon in the form of government patterns, subcategorization frames, or similar data structures. We concentrate on non-canonical cases of valency filling in Russian, which are characteristic of nonverbal parts of speech, such as adverbs, adjectives, and particles, in the first place. They are more difficult to handle than canonical ones, because the position of the actant in the tree is governed by more complicated rules. A valency may be filled by expressions occupying different syntactic positions, and a syntactic position may accept expressions filling different valencies of the same word. We show how these phenomena can be processed in a semantic analyzer.

We propose a general model for joint inference in correlated natural language processing tasks when fully annotated training data is not available, and apply this model to the dual tasks of word sense disambiguation and verb subcategorization frame determination. The model uses the EM algorithm to simultaneously complete partially annotated training sets and learn a generative probabilistic model over multiple annotations. When applied to the word sense and verb subcategorization frame determination tasks, the model learns sharp joint probability distributions which correspond to linguistic intuitions about the correlations of the variables. Use of the joint model leads to error reductions over competitive independent models on these tasks.

The theoretical framework of this approach is the structural theory of language, coming from the European tradition and based upon the study of various languages. Despite the distinction usually made between functional and formal linguistics, the formalisation of structural (mostly ...

Where does a verb's frame come from? The obvious answer is the verb itself, and this is the answer that syntacticians have traditionally provided, whether they describe predicator-argument relations as syntactic sisterhood relations or as lexical properties (the predicator's combinatoric potential, or valence). Thus, Haegeman, in her introduction to Government and Binding theory, states, "the thematic structure of a predicate, encoded in the theta grid, will determine the minimal components of the sentence" (Haegeman 1994: 55). Similarly, Bresnan, in her introduction to Lexical Functional Grammar (LFG), states, "[o]n the semantic side, argument structure represents the core participants and events (states, processes) designated by a single predicator. […] On the syntactic side, argument structure represents the minimal information needed to characterize the syntactic dependents of an argument-taking head" (Bresnan 2001: 304). In lexicalist theories like LFG, whenever the arguments of a verb can have more than one set of syntactic realizations, each distinct realization pattern corresponds to a different mapping from semantic roles to grammatical functions, as expressed in a unique lexical entry, and lexical entries, or classes of lexical entries, are related by lexical rules (Neidle 1994). The drive to streamline lexical entries by removing predictable properties has led theorists to develop more general, putatively universal, mapping principles, as well as principles for deriving

1.2. Theoretical background – Grimshaw: • complex event nominals : event-reading means “existence of an internal semantic analysis of the event provided by the even t structures” (Grimshaw 1990: 59) correlates with the presence of argument structure • simple event nominals: event-reading without an internal semantic analysis lack of argument structure • result nominals: result-reading, lack of argument structure “resu lt nominals name the output of a process or an element associated with the process” (Grimshaw 1990:49)

La tesis se ubica entre los estudios teoricos del campo linguistico generativo siguiendo las pautas proporcionadas por sucesivos trabajos de Chomsky cuya mision ha sido delinear un programa de perspectiva minimalista en la concepcion de la Facultad del Lenguaje. Procuramos mostrar, en primer lugar, que los procesos tratados se someten plenamente al Principio de Economia, por lo que se da un paso firme en la reduccion de la complejidad computacional. En segundo lugar, dados los mecanismos de la teoria, fundamentalmente los que se aplican en la Derivacion por Fase (Chomsky, 1999), se quiere consolidar la definicion de estos mecanismos del Sistema Computacional construyendo un aparato unico y simple. Y, en tercer lugar, tratamos de hacer una aportacion, aunque solo sea parcial, en relacion con un problema candente, como es el diseno de los instrumentos utilizados en la construccion de oraciones: la necesidad de una descripcion completa de la arquitectura interna de rasgos de los elemen...