Volume 51, Issue 4 - Thoughts on Typologies: Different Perspectives on Verb Classifications

Standard models of aspectual classes focus on event decompositional or featural distinctions. However, such classifications often over- or under-generate, and do not necessarily capture the temporal properties aspectual classifications are based on. I develop a predictive model of aspectual classes based on theories of scalar change, taking two independently motivated properties of scales as key: (a) how specific the predicate is about the theme's final state on the scale and (b) the scale's mereological complexity. The resultant classification accommodates the standard Vendler dynamic classes, plus additional classes that have proved difficult for prior approaches, and makes novel predictions about the relation of aspectual class and argument realization.

The verb lexicon can be classified in different and complementary ways; each approach faces challenges and limits. We focus on two large-scale lexical resources, WordNet (Miller et al. 1990; Fellbaum 1998) and FrameNet (Baker et al. 2003). WordNet is a semantic network where lexical meaning is represented in terms of relations among word forms. In contrast to WordNet's paradigmatically organized entries, FrameNet's Lexical Units are embedded in corpus-derived contexts, which serve as a basis for semantic distinctions. The classificatory perspective that each resource contributes to the analysis of the verb lexicon is illustrated with specific examples. Constructing verb typologies, even when based on attested data, requires the lexicographer's introspection and judgments, and no two classifications are completely alike. In a case-based approach to compare and harmonize WordNet and FrameNet, we ask human annotators to select the context-appropriate senses from each resources that best fit tokens in a corpus. A typology of alignments is proposed.

In my paper, two variants of verb classification in Adyghe, a language of the West Caucasian family, are discussed. The first variant is a purely morphological classification based on the choice of person cross-referencing prefixes. The second one is a derivational classification which builds on the morphological mechanisms of reciprocalization and reflexivization. The main research question which lies behind my study is whether verbs derived by means of a derivational marker, for instance, a possibilitive one, behave in the course of other valency-changing operations differently from nonderived verbs. I will show that verb classification in Adyghe has some typologically peculiar properties, the main one being that the derivational classification distinguishes more specific classes than the purely morphological one. In other words, the fact that a verb is derived and how it is derived is crucial for its behavior. The language-specific properties of Adyghe are also typologically relevant. They show that derived verbs and derivational mechanisms are of particular relevance in verb classification and should be given more attention in linguistic work on verb classification than is currently done.

The paper investigates the processing correlates of core verb features used to generate verb typologies. The aim was to contrast the effects of verb internal structure (as in semantic/conceptual features e.g., Jackendoff [1991]; Levin and Rappaport Hovav [2005]) with argument structure and argument realization in sentence processing. To this end, we designed a self-paced reading task with internal structure and argument realization as variables in verb processing. Results showed that absence of a prototypical Agent or mismatch between thematic hierarchies and argument realization yields longer processing times possibly due to some form of thematic reanalysis or as a reaction to a “surprise” effect by the human processor while detecting a non-expected thematic assignment. No effects of verb internal structure were found. We take this result as an indication that argument structure and argument realization play an important role in verb recognition during sentence processing. We further propose that this reflects their prominent role in verb representation and we suggest that this finding could give additional support to verb classifications based on verb argument structure features compared to those based on internal structure.

This study deals with a particular group of predicates called “predicates/verbs of relevance” or “predicates/verbs of indifference” in the literature. Its purpose is to investigate to what extent verbs of this particular group present common structural properties. It therefore seeks to establish the structural manifestations of synonymy. These structural manifestations are not to be found in argument-function mapping à la Levin (1993), but rather in polarity, decategorialization and sentence structure. Corpus data reveal that syntax, semantics and pragmatics interact in particular ways in the field of relevance. This interaction appears to be grounded in pragmatic constraints arising from the principle of relevance (Sperber and Wilson 1986). The basic idea is that, as relevance is presupposed in human communication and need not be expressed, verbs of relevance are more likely to be used with negative than with positive polarity. Used with positive polarity, they tend to occur in sentence forms that present them as strongly presupposed. Used with negative polarity, they are more likely to occur in the focal area of the sentence. As statements about relevance express speaker's points of view, relevance verbs are also markers of intersubjectivity and are therefore subject to grammaticalization phenomena, such as the omission of prepositions.

The paper demonstrates how verb and noun classes can be used as a common interface in contrastive Construction Grammar. It presents an innovative approach to the contrastive analysis of constructional spaces (sets of constructions covering a certain semantic domain). We compare English and Dutch analytic causatives by using the statistical technique of multiple correspondence analysis applied to data from large monolingual corpora. The method allows us to explore the common conceptual space of the constructions, in particular the salient semantic dimensions and causation types, which emerge on the basis of co-occurring semantic classes of the nominal and verbal slot fillers in constructional exemplars. The formal patterns of the constructions at different levels of specificity are projected onto this space. Our analyses show that an average Dutch analytic causative refers to more indirect and abstract causation with fewer animate than its English counterpart. We have also found that the languages “cut” the common conceptual space in unique ways, although the semantic areas of many English and Dutch constructions overlap substantially. Nevertheless, the form-meaning mapping in the two languages displays commonalities. Both English and Dutch constructions with prepositionally marked or implicit causees are strongly associated with animate causees. We have also observed a correlation between the directness of causation and the crosslinguistic hierarchy of affectedness marking proposed by Kemmer and Verhagen (1994).