Deverbal zero-nominalization and verb classes: Insights from a database

We investigate deverbal zero-derived nominals in English (e.g., to walk > a walk) from the perspective of the lexical semantics of their base verbs and the interpretations they may receive (e.g., event, result state, product, agent). By acknowledging that, in the absence of an overt affix, the meaning of zero-nominals is highly dependent on that of the base, the ultimate goal of this study is to identify possible meaning regularities that these nominals may display in relation to the different semantic verb classes. We report on a newly created database of 1,000 zero-derived nominals, which have been collected for various semantic verb classes. We test previous generalizations made in the literature in comparison with suffix-based nominals and in relation to the ontological type of the base verb. While these generalizations may intuitively hold, we find intriguing challenges that bring zero-derived nominals closer to suffix-based nominals than previously claimed.

Reconstructing the ditransitive construction for Proto-Germanic: Gothic, Old English and Old Norse-Icelandic

The semantic range of ditransitive verbs in Modern English has been at the center of linguistic attention ever since the pioneering work of Pinker (1989. Learnability and cognition: The acquisition of argument structure. Cambridge, Mass: MIT Press). At the same time, historical research on how the semantics of the ditransitive construction has changed over time has seriously lagged behind. In order to address this issue for the Germanic languages, the Indo-European subbranch to which Modern English belongs, we systematically investigate the narrowly defined semantic verb classes occurring in the ditransitive construction in Gothic, Old English and Old Norse-Icelandic. On the basis of data handed down from Proto-Germanic and documented in the oldest layers of the three Germanic subbranches, East, West and North Germanic, respectively, we show that the constructional range of the ditransitive construction was considerably broader in the earlier historical stages than now; several subclasses of verbs that could instantiate the ditransitive in early Germanic are infelicitous in the ditransitive construction in, for instance, Modern English. Taking the oldest surviving evidence from Germanic as point of departure, we reconstruct the ditransitive construction for an earlier proto-stage, using the formalism of Construction Grammar and incorporating narrowly defined semantic verb classes and higher level conceptual domains. We thus reconstruct the internal structure of the ditransitive construction in Proto-Germanic, including different levels of schematicity.

Statistical observations on hierarchies of transitivity

In this paper we first test whether there is statistical support for a transitivity hierarchy viewed as an implicational hierarchy. To that end we construct data-driven transitivity hierarchies of two-place verb meanings based on the Valency Patterns Leipzig (ValPaL) database using Guttman scaling. We look at how well the hierarchies conform to strict scalarity (one-dimensionality) and, through matrix randomization, test whether their strengths are significant. We then go on to construct slightly different hierarchies based on simple counts of instances of two-participant coding frames for a given verb meaning across languages, rather than through the Guttman scaling procedure, which yields less resolution and is not designed for missing data. Finally, we assess whether the members of the hierarchies fall into semantic verb classes. The concluding section summarizes the results.

Reconstructing constructional semantics

As the historical linguistic community is well aware, reconstructing semantics is a notoriously difficult undertaking. Such reconstruction has so far mostly been carried out on lexical items, like words and morphemes, and has not been conducted for larger and more complex linguistic units, which intuitively seems to be a more intricate task, especially given the lack of methodological criteria and guidelines within the field. This follows directly from the fact that most current theoretical frameworks are not construction-based, that is, they do not assume that constructions are form-meaning correspondences. In order to meet this challenge, we present an attempt at reconstructing constructional semantics, and more precisely the semantics of the Dative Subject Construction for an earlier stage of Indo-European. For this purpose we employ lexical semantic verb classes in combination with the semantic map model (Barðdal 2007, Barðdal, Kristoffersen & Sveen 2011), showing how incredibly stable semantic fields may remain across long time spans, and how reconstructing such semantic fields may be accomplished.

Semantic Constraints on Light Verb Constructions in Modern Irish

The paper focuses on the semantics of Light Verb Constructions (LVCs) in Modern Irish. Structures made up of a light verb (déan ‘do’, tabhair ‘give’, faigh ‘get’, bain ‘take, extract’) and a verbal noun (VN) complement are investigated. LVCs are argued to have a telicising effect which results from the interaction of the aktionsart of the VN complement and syntax. Particular light verbs show systematic behavior in their ability to combine with VNs derived from certain semantic verb classes (verbs of movement, emission of sound, social interaction etc.) in order to present the situation from different angles by giving prominence to certain participants (Agent, Patient, Experiencer). The choice of a specific light verb may also lead to a subtle semantic modification such as volitionality.

Experiments on the Automatic Induction of German Semantic Verb Classes

This article presents clustering experiments on German verbs: A statistical grammar model for German serves as the source for a distributional verb description at the lexical syntax-semantics interface, and the unsupervised clustering algorithm k-means uses the empirical verb properties to perform an automatic induction of verb classes. Various evaluation measures are applied to compare the clustering results to gold standard German semantic verb classes under different criteria. The primary goals of the experiments are (1) to empirically utilize and investigate the well-established relationship between verb meaning and verb behavior within a cluster analysis and (2) to investigate the required technical parameters of a cluster analysis with respect to this specific linguistic task. The clustering methodology is developed on a small-scale verb set and then applied to a larger-scale verb set including 883 German verbs.