Volume 23, Issue 1

Dutch manner of motion verbs play a prominent role in the literature on unaccusativity. As these verbs can take both hebben ‘have’ and zijn ‘be’ as their perfective auxiliaries, they are considered to show both unergative and unaccusative behavior. The general consensus is that these verbs normally take hebben , yet occur with zijn if they are ‘telicized’ by an endpoint, and that the auxiliaries are diagnostics for the syntactic status of prepositional phrases (PPs). The paper presents attested data that reveal that this generalization is untenable: there are examples that take the opposite auxiliary from what the generalization predicts. To account for the full set of data, the paper takes a cognitive-grammar perspective, arguing that auxiliary choice, telicity and syntactic status of PPs are independent issues requiring their own explanations. Auxiliary choice is analyzed in terms of alternate construals of a motion event: with hebben as a type of act and with zijn as a change of location. In this manner, the paper adds to a growing body of literature that questions the usefulness of the coarse unergative–unaccusative distinction, advocating a ‘local analysis’ instead.

This article provides an analysis—within the framework of Radical Construction Grammar—of how become developed into a copula ‘become’ out of an original sense ‘arrive’, and wax , originally ‘grow’, also came to be used as a copula ‘become’. Importantly, it explains why these verbs successfully became fully productive copulas in a very short period of time. It is argued that this happened after a pre-copular stage had reached a cognitive threshold value. The occurrence of this threshold is related to the fact that the copular constructions featuring become and wax were not the end result of a single diachronic lineage of constructions (i.e. one construction developed out of another one, one at a time). Instead, the copularization of these verbs was the result of an interaction between lineages of constructions, belonging to two groups: (i) constructions involving become or wax , which gradually changed and interacted with each other; (ii) constructions involving already existing copulas, notably weorđan ‘become’, which provided a generally productive analog upon which the newly emerging copulas could graft. Generally, the article calls attention to the importance of multiple source constructions and thresholds in understanding grammaticalization processes and productivity.

This article presents empirical evidence of the high referential specificity of sound-symbolic words, based on a FrameNet-aided analysis of collocational data of Japanese mimetics. The definition of mimetics, particularly their semantic definition, has been crosslinguistically the most challenging problem in the literature, and different researchers have used different adjectives (most notably, “vivid,” since Doke 1935) to describe their semantic peculiarity. The present study approaches this longstanding issue from a frame-semantic point of view combined with a quantitative method. It was found that mimetic manner adverbials generally form a frame-semantically restricted range of verbal/nominal collocations than non-mimetic ones. Each mimetic can thus be considered to evoke a highly specific frame, which elaborates the general frame evoked by its typical host predicate and contains a highly limited set of frame elements, which correlate and constrain one another. This conclusion serves as a unified account of previously reported phenomena concerning mimetics, including the lack of hyponymy, the one-mimetic-per-clause restriction, and unparaphrasability. This study can be also viewed as a methodological proposal for the measurement of frame specificity, which supplements bottom-up linguistic tests.

In this study, we test a number of predictions concerning children's knowledge of the transitive Subject-Verb-Object (SVO) construction between two and three years on one child (Thomas) for whom we have densely collected data. The data show that the earliest SVO utterances reflect earlier use of those same verbs, and that verbs acquired before 2;7 show an earlier move towards adult-like levels of use in the SVO construction and in object argument complexity than later acquired verbs. There is not a close relation with the input in the types of subject and object referents used, nor a close adherence to Preferred Argument Structure (PAS) before 2;7, but both early and late acquired verbs show a simultaneous move towards PAS patterns in selection of referent type at 2;9. The event semantics underpinning early transitive utterances do not straightforwardly fit prototype (high or inalienable) notions of transitivity, but rather may reflect sensitivity to animacy and intentionality in a way that mirrors the input. We conclude that children's knowledge of the transitive construction continues to undergo significant development between 2;0 and 3;0, reflecting the gradual abstraction and integration of the SVO and VO constructions, verb semantics, discourse pragmatics, and the interactions between these factors. These factors are considered in the context of a prototype for the transitive construction.

This paper investigates the semantic bases of class membership in the noun class system of Gújjolaay Eegimaa (Eegimaa henceforth), a Niger-Congo and Atlantic language of the BAK group spoken in Southern Senegal. The question of whether semantic principles underlie the overt classification of nouns in Niger-Congo languages is a controversial one. There is a common perception of Niger-Congo noun class systems as being mainly semantically arbitrary. The goal of the present paper is to show that physical properties and culture-specific factors are central principles of semantic categorisation in the Eegimaa noun class system. I argue that the Eegimaa overt grammatical classification of nouns into classes is a semantic categorisation system whereby categories are structured according to prototypicality, family resemblance, metaphorical and metonymic extensions and chaining processes, as argued within the framework of Cognitive Linguistics. I show that the categorisation of entities in the Eegimaa nominal classification system productively makes use of physical properties such as shape as well as using culture-specific, less productive parameters for the semantic categorisation of entities denoted by nouns. The analysis proposed here also shows that the cases of multiple morphosyntactic classifications of nouns reflect multiple conceptual categorisation strategies. A detailed examination of the formal and semantic instances of multiple classification reveals the existence of conceptual correlations between the physical properties and the culture-specific semantic parameters of categorisation used in the Eegimaa noun class system.

The paper demonstrates that dependency-based syntax is in a strong position to produce principled and economical accounts of the syntax of constructs. The difficulty that constituency-based syntax has in this regard is that very many constructs fail to qualify as constituents. The point is evident with the box diagrams and attribute value matrices (AVMs) that some construction grammars (CxGs) use to formalize constructions; these schemata often represent fragments rather than constituents. In dependency-based syntax in contrast, constructions are catenae, whereby a catena is a chain of words linked together by dependencies. The catena is a novel but well-defined unit of syntax associated with dependency grammar (DG). The constructs of CxGs are more amenable to analyses in terms of the catenae of dependency-based syntax than to analyses in terms of the constituents of constituency-based syntax.