Le principal objectif de cette étude est de situer les constructions verbales du wolof dans une perspective typologique. Il s'agit tout d'abord de proposer une description synthétique du système de prédication verbale du wolof dans une perspective typologique, en nous appuyant sur les travaux de référence concernant la conjugaison du wolof. Contrairement à ce que l'on observe dans un très grand nombre de langues, en wolof la majeure partie des catégories grammaticales liées au verbe est exprimée par des constructions périphrastiques. L'analyse typologique de ces constructions périphrastiques nous sert de base empirique pour proposer une nouvelle approche de la notion d'auxiliaire. Nous considérons que, dans une perspective typologique, l'auxiliaire ne doit pas être défini comme une catégorie lexicale spécifique, ni comme une étape dans un chemin de grammaticalisation, mais plutôt comme un élément prédicatif autonome ayant une fonction spécifique. Par ailleurs, nous proposons une analyse constructionnelle de l'organisation du système de prédication verbale du wolof. Nous considérons que les constructions verbales du wolof ne forment pas un ensemble non structuré d'entités indépendantes, mais plutôt un système extrêmement structuré (un réseau de constructions). En outre, nous montrons que certaines idiosyncrasies apparentes dans le paradigme de conjugaison du wolof peuvent s'expliquer à la lumière de la diachronie. Enfin, nous proposons une analyse comparative des constructions verbales des langues atlantiques afin de déterminer ce qui, dans la conjugaison du wolof, est issu du proto-atlantique. / This thesis is a study of Wolof verbal constructions in a typological perspective. Based on available descriptions of Wolof verbal conjugation, I first provide a summary of the system of verbal predication in the light of the typological literature. Contrary to what is observed in many languages, most Wolof verbal categories are expressed periphrastically. The typological analysis of these periphrastic constructions provides us with the empirical basis to propose a new approach to the notion of “auxiliary”. I argue that auxiliaries should not be cross-linguistically defined as items belonging to a specific lexical class or as items on a grammaticalisation path but rather as autonomous predicative elements with a specific function. In addition, I propose a constructional analysis of the organisation of the verbal predication system of Wolof. The entirety of Wolof verbal constructions is not assumed to form an unstructured set of independent entities, but it is instead taken to constitute a highly structured system (a network of constructions). Furthermore, some apparent idiosyncrasies in the conjugation paradigm of Wolof can be explained from a diachronic point of view. Finally, I provide a comparative analysis of verbal constructions in Atlantic languages in order to determine which elements of the Wolof conjugation are inherited from Proto-Atlantic.

京都大学 / 0048 / 新制・課程博士 / 博士(人間・環境学) / 甲第22900号 / 人博第969号 / 新制||人||229(附属図書館) / 2020||人博||969(吉田南総合図書館) / 京都大学大学院人間・環境学研究科共生人間学専攻 / (主査)教授 谷口 一美, 教授 藤田 耕司, 准教授 守田 貴弘, 教授 山梨 正明 / 学位規則第4条第1項該当 / Doctor of Human and Environmental Studies / Kyoto University / DFAM

Different from Generative Grammar which sees grammar as a formal system of how words are put together to form sentences, Construction Grammar suggests that grammar is more than just rules and surface forms; instead, grammar includes many form-and-meaning pairings which are called constructions. For years, Construction Grammarians have been investigating constructions with various approaches, including corpus-linguistics, pedagogical, second language acquisition and so on, yet there is still room for exploration. The present paper aims to further investigate the [V the Ntaboo-word out of]-construction (Hoeksema & Napoli, 2008; Haïk, 2012; Perek, 2016; Hoffmann, 2020) (e.g., I kick the hell out of him.) and propose a new umbrella construction, “X the Y out of Z” (XTYOFZ) construction, for it. Another aim is to examine the usage and comprehension of the XTYOFZ construction by English as a Second Language (ESL) and English as Native Language (ENL) speakers. The usage context, syntactic and semantic characteristics of the XTYOFZ construction were examined through corpus linguistic methodology. Furthermore, processing and understanding of the construction by ESL and ENL speakers were tested via an online timed Lexical Decision Task as well as an online follow-up survey consisting of questions on English acquisition and usage, and a short comprehension task on the XTYOFZ construction. Corpus data shows that in general, the combination of non-motion action verbs (e.g., scare, beat) as X and taboo terms (e.g., shit, hell) as Y was the most common. Also, it was found that the construction occurs mostly in non-academic contexts such as websites and TV/movies. On the other hand, results from the Lexical Decision Task show that ESL speakers access constructional meaning slightly more slowly than ENL speakers. The follow-up survey also reflects that ESL speakers seem to have a harder time to produce and comprehend the construction compared to ENL speakers. By investigating the features of a relatively less-discussed construction and its usage by ESL speakers, this study hopes to increase the knowledge base of Construction Grammar and ESL construction comprehension and usage, particularly on the constructions that are mainly used in more casual settings.

Entgegnung auf Hubert Haider. 2018. Grammatiktheorien im Vintage-Look – Viel Ideologie, wenig Ertrag. In: Angelika Wöll-stein u.a. (Hg.). Grammatiktheorie und Empirie in der germanistischen Linguistik (Germanistische Sprachwissenschaft um 2020, 1). Berlin, Boston: de Gruyter. 47-92. DOI: https://doi.org/10.1515/9783110490992-003. Wer unter dem Titel 'Grammatiktheorien im Vintage-Look – Viel Ideologie, wenig Ertrag' einen fast 50seitigen Aufsatz publiziert, hat offenbar ein Anliegen. Als interessierter Leser erhofft man sich, dass sich Haider dezidiert mit den von ihm kritisierten Paradigmen des 'Minimalist Program' und der 'Construction Grammar' auf Basis einer sachlichen Diskussion aktueller Forschung auseinandersetzt. Um es, aus Sicht der Konstruktionsgrammatik, für die ich hier sprechen möchte, kurz zu machen: Diese Erwartung wird bitter enttäuscht.

There is a need to have an automated system that can read family history books or other historical texts and extract as many genealogy facts as possible from them. Embley and others have applied traditional information extraction techniques to this problem in a system called OntoES with a reasonable amount of success. In parallel much linguistic theory has been developed in the past decades, and Lonsdale and others have built computational embodiments of some of these theories using Soar. In this thesis we introduce a system called OntoSoar which combines the Link Grammar Parser using a grammar customized for family history texts with an innovative semantic analyzer inspired by construction grammars to extract genealogical facts from family history books and use them to populate a conceptual model compatible with OntoES with facts derived from the text. The system produces good results on the texts tested so far, and shows promise of being able to do even better with further development.

Abstract: This dissertation deals with nominal constructions in nicht- ([Nicht-S]) in German from a synchronic perspective within the framework of construction grammar. On the basis of a corpus of [Nicht-S] drawn from the DWDS database, it is shown that the [Nicht-S] pattern is a morphological construction in which nicht- is a prefix. [Nicht-S] can have three different interpretations, which are called ontological, classifying and normative, depending on the base noun and pragmatic information provided in the context. An analysis is proposed whereby they are represented as instantiations of a more general construction.

In recent years, Neural Language Models (NLMs) have consistently demonstrated increasing linguistic abilities. However, the extent to which such networks can actually learn grammar remains an object of investigation, and experimental results are often inconclusive. Notably, the mainstream evaluation framework in which NLMs are tested seems largely based on Generative Grammar and nativist principles, and a shared constructionist approach on the matter has not yet emerged: this is at odds with the fact that usage-based theories are actually better suited to inspect the behaviour of such models. The main contribution of this thesis is the introduction of CALaMo, a novel framework for evaluating Neural Language Models’ linguistic abilities, using a constructionist approach. We especially aim at formalizing the relationship between the computational modelling phase and the underlying linguistic theory, thus allowing a more refined and informed discussion of settings and results. We focus on two specific areas that, we believe, are currently not easily tractable within the mainstream evaluation framework. The first scenario deals with language acquisition from child-directed data. Our main experimental result shows how it is possible to follow schematization paths during the acquisition process of the model, and how this relates to core hypotheses in constructionist theories. The second scenario deconstructs the mainstream view of the Neural Model as an average idealized speaker by proposing a way to simulate and analyze a population of artificial individuals. We show how the amount of “shared linguistic knowledge” across speakers is highly dependent on the specific linguistic background of each individual. Overall, we believe our framework opens the path for future discussion on the role of computational modelling in usage-based linguistic theory and vice versa, and provides a new formal methodology to both fields of study.

The work in this thesis presents and evaluates a number of strategies by which English-learning children might discover the major open-class parts-of-speech in English (nouns, verbs and adjectives) on the basis of purely distributional information. Previous work has shown that parts-of-speech can be readily induced from the distributional patterns in which words occur. The research reported in this thesis extends and improves on this previous work in two major ways, related to the constructional status of the utterance contexts used for distributional analysis, and to the way in which previous studies have dealt with categorial ambiguity. Previous studies that have induced parts-of-speech from word distributions have done so on the basis of fixed “windows” of words that occur before and after the word in focus. These contexts are often not constructions of the language in question, and hence have dubious status as elements of linguistic knowledge. A great deal of recent evidence (e.g. Lieven, Pine & Baldwin, 1997; Tomasello, 1992) has suggested that children’s early language may be organized around a number of lexically-specific constructional frames with slots, such as “a X”, “you X it”, “draw X on X”. The work presented here investigates the possibility that constructions such as these may be a more appropriate domain for the distributional induction of parts-of-speech. This would open up the possibility of a treatment of part-of-speech induction that is more closely integrated with the acquisition of syntax. Three strategies to discover lexically-specific frames in the speech input to children are presented. Two of these strategies are based on the interplay between more and less frequent words in English utterances: the more frequent words, which are typically function words or light verbs, are taken to provide the schematic “backbone” of an utterance. The third strategy is based around pairs of words in which the occurrence of one word is highly predictable from that of the other, but not vice versa; from these basic slot-filler relationships, larger frames are assembled. These techniques were implemented computationally and applied to a corpus of child-directed speech. Each technique yielded a large set of lexically-specific frames, many of which could plausibly be regarded as constructions. In a comparison with a manual analysis of the same corpus by Cameron-Faulkner, Lieven and Tomasello (2003), it is shown that most of the constructional frames identified in the manual analysis were also produced by the automatic techniques. After the identification of potential constructional frames, parts-of-speech were formed from the patterns of co-occurrence of words in particular constructions, by means of hierarchical clustering. The resulting clusters produced are shown to be quite similar to the major English parts-of-speech of nouns, verbs and adjectives. Each individual word token was assigned a part-of-speech on the basis of its constructional context. This categorization was evaluated empirically against the part-of-speech assigned to the word in question in the original corpus. The resulting categorization is shown to be, to a great extent, in agreement with the manual categorization. These strategies deal with the categorial ambiguity of words, by allowing the frame context to determine part-of-speech. However, many of the frames produced were themselves ambiguous cues to part-of-speech. For this reason, strategies are presented to deal with both word and context ambiguity. Three such strategies are proposed. One considers membership of a part-of-speech to be a matter of degree for both word and contextual frame. A second strategy attempts to discretely assign multiple parts-of-speech to words and constructions in a way that imposes internal consistency in the corpus. The third strategy attempts to assign only the minimally-required multiple categories to words and constructions so as to provide a parsimonious description of the data. Each of these techniques was implemented and applied to each of the three frame discovery techniques, thereby providing category information about both the frame and the word. The subsequent assignment of parts-of-speech was done by combining word and frame information, and is shown to be far more accurate than the categorization based on frames alone. This approach can be regarded as addressing certain objections against the distributional method that have been raised by Pinker (1979, 1984, 1987). Lastly, a framework for extending this research is outlined that allows semantic information to be incorporated into the process of category induction.

<p>The expression /on the one hand…on the other hand/ (OH1 OH2) is a fixed linguistic pattern which is used to emphasize the comparison between two possibly complex propositions (henceforth X and Y). The static syntactic form of the pattern and the specific semantic comparison it evokes are strong indicators that it is a construction of the type discussed in the analytical method Construction Grammar (henceforth CxG). Thus, the aim of this essay is to argue that the pattern OH1X OH2Y is a CxG construction with specific syntactic and semantic constraints, and at the same time to give a descriptive account of the features of the construction. The British National Corpus was used to get examples containing the pattern. The syntactic and semantic features of these examples were analysed and the results were compared with how the traditional descriptive grammarians account for the pattern.</p>

Dans le cadre des approches constructionistes, cette thèse étudie les verbes labiles de l’anglais, qui peuvent manifester des configurations syntaxiques variées sans changer de forme. L’étude de la complémentation de ces verbes montre que leur catégorisation en familles sémantiques est pertinente mais pas suffisante pour expliquer leur comportement. La thèse défend une approche syncrétique de la complémentation du verbe qui rend compte de son importante productivité et de ses limites parfois arbitraires. Une analyse montre que les verbes ont tous une configuration syntaxique par défaut, qui n’est pas signifiante et qui permet simplement au verbe d’exprimer ses arguments de façon non marquée, en accord avec certains principes de cohérence conceptuelle. À l’inverse, lorsque la complémentation du verbe a un apport sémantique identifiable, l’existence de schémas de complémentation pleinement signifiants est postulée. Il s’agit d’ensembles de compléments dont le sens est distinct de celui du verbe auquel ils sont associés et se retrouve de façon régulière avec des verbes de catégories diverses. Il est démontré que les schémas de complémentation doivent être considérés comme des unités linguistiques de plein droit de l’anglais. Cela implique qu’en synchronie, ces schémas sont emmagasinés par les locuteurs plutôt que le résultat d’un processus d’analogie avec des constructions existantes. Leur statut d’unité linguistique permet d’étudier leur sémantisme de la même façon que des unités lexicales plus classiques. S’ils sont en majorité polysémiques, certains schémas ont des emplois difficiles à relier sémantiquement et doivent donc être considérés comme des homonymes. / Within a constructionist framework, this thesis studies English labile verbs, which can enter into various syntactic configurations without changing form. A study of their complementation shows that categorizing them into semantic families is relevant but not sufficient to explain their behavior. The thesis defends a syncretic approach to verb complementation to that accounts for its important productivity and its sometimes arbitrary limits. It is shown that all verbs have a default syntactic configuration, which is not meaningful and which simply allows the verb to express its arguments in an unmarked way, in accordance with certain principles of conceptual coherence. Conversely, when the complementation of the verb has an identifiable semantic contribution, the existence of fully meaningful patterns of complementation is posited. These are defined as sets of complements, whose meaning is distinct from that of the verb with which they are associated and is found regularly with verbs of diverse categories. It is shown that patterns of complementation should be considered fully-fledged English linguistic units. This implies that synchronically, these patterns are mentally stored by speakers rather than the result of a process of analogy with existing constructions. Their status as linguistic units makes it possible to study their meaning in the same way as more classical lexical units. Although most of them are polysemic, some patterns of complementation exhibit uses that are difficult to link semantically and must therefore be viewed as homonyms.