Conditionals, or if-then sentences, form a crucial ingredient of everyday reasoning and argumentation, as they enable us to express our thoughts about possible states of the world. They are used in... Show moreConditionals, or if-then sentences, form a crucial ingredient of everyday reasoning and argumentation, as they enable us to express our thoughts about possible states of the world. They are used in very different ways, and the main aim of this dissertation is to investigate to what extent these different uses of conditionals are connected to one another and to their grammatical features.The first part of this dissertation presents an analysis of conditionals in terms of implicatures of 'unassertiveness' and 'connectedness'. Insights from semantics, pragmatics, cognitive linguistics, and neighbouring fields are combined. In the second part, the analysis is tested on a corpus of spoken and written Dutch discourse. To investigate the relation between the meaning and grammar of conditionals, several cluster analyses are conducted. The results show that grammatical features such as verb tense and modal marking do not, or only weakly, license generalised implicatures of unassertiveness and connectedness. This outcome sheds light on difficulties in applying general categories of conditionals to language use data, and it suggests that the fundamentals of categorising conditional constructions need revision.The dissertation shows the benefits of combining semantic and pragmatic analyses of conditionals. It provides an extensive discussion of classifications of conditionals, an overview of the grammatical features of Dutch conditionals, and it presents cluster analyses using state-of-the-art machine-learning techniques. The study should therefore be of interest to anyone concerned with the syntax, semantics, and pragmatics of conditionals, and to anyone working on Dutch grammar, corpus linguistics, theories of argumentation, and the interface between semantics and pragmatics. Show less
This dissertation is concerned with the development of grammar. Starting from a usage-based perspective, which holds that children use domain-general learning mechanisms to acquire the... Show moreThis dissertation is concerned with the development of grammar. Starting from a usage-based perspective, which holds that children use domain-general learning mechanisms to acquire the grammatical patterns of their mother tongue, Beekhuizen shows how to operationalize various concepts from this tradition in a computational model. In order to arrive at a sound set of assumptions, Beekhuizen compares and criticizes various earlier usage-based modeling approaches and scrutinizes the concepts of a usage-based theory of language acquisition from the perspective of a computational modeler. As the model should be able to produce utterances on the basis of a meaning to be expressed, as well as to interpret utterances, the availability of meaning from the situational context is studied empirically. The resulting model, the Syntagmatic-Paradigmatic Learner, simulates an increasing ability to understand utterances on the basis of a grammar of constructions, as well as to produce utterances on the basis of this grammar. Several developmental effects are simulated and the internal states of the model are carefully examined. Show less
Words may have multiple interpretations. Generally, native speakers do not perceive this as a problem, because the context provides enough clues as to what is meant. For non-native speakers and... Show moreWords may have multiple interpretations. Generally, native speakers do not perceive this as a problem, because the context provides enough clues as to what is meant. For non-native speakers and students of dead languages, however, the existence of multiple interpretations sometimes does raise problems. This suggests that the context is not the only clue native speakers use to interpret words.In this dissertation, it is studied what types of context Dutch speakers need to interpret the poly-interpretable word ergens ‘somewhere/anywhere’, modal particle. The results of this investigation were used to find out more about the Ancient Greek form που ‘somewhere, anywhere’, modal particle.This thesis shows that the study of contextual cues that allow native speakers to interpret their language provides insights that may be used in the study of dead languages. The modal interpretations of ergens and που turned out to be quite different, but the context of both words clearly showed recurring (albeit different) patterns. Knowledge of the common interpretation of words in specific contexts seems crucial for their interpretation, suggesting that it is not words themselves that carry meaning, but words-in-context. Show less
This dissertation investigates Extended Lexical Units (ELUs), elements that are bigger than just one word and which are stored in the lexicon. ELUs are an interesting problem on at least two... Show moreThis dissertation investigates Extended Lexical Units (ELUs), elements that are bigger than just one word and which are stored in the lexicon. ELUs are an interesting problem on at least two different levels: from a grammatical (or grammar-theoretical) as well as from a computational perspective. How do ELUs fit into a grammar model, and how can they be implemented in a computational system? The answers to these questions are illustrated with three case studies of rather well-investigated Dutch constructions, namely the NPN (as in dog after dog) and a related Dutch phenomenon, the NCoN construction (as in kind noch kraai, `child nor crow’), the Dutch way-construction, and the dative alternation (as in Jan geeft Piet een bal (`Jan gives Piet a ball’) vs. Jan geeft een bal aan Piet (`Jan gives a ball to Piet’)). For all three constructions, a formal Construction Grammar analysis, based on Sign-Based Construction Grammar, is presented, and additionally, it is shown how these ELUs are tackled in the semantic parser/generator Delilah. Show less