This study examines composite predicates (CPs) in the history of American English and uses an exemplar-based model to explain changes in the frequency of verb–noun pairings over time. Two different types of verb-nominal CPs are considered, including those like take a look, in which a light verb occurs with an abstract nominal object, and others like lose sight, with a more lexically specific verb. Using a corpus of texts written between 1820 and 2009, I track the frequency of different CPs and analyze several families of semantically related nouns that occur with the same verb (e.g. take a look, peak, etc.). Representative families are analyzed to determine the presence of highly frequent verb–noun pairings, or exemplars, that separate themselves over time. The success of exemplars is evaluated according to several factors that may shape patterns of use, including the relative size of noun families, the frequency band of tokens of each family and the distribution of tokens across types within a family. Results indicate that the two types of CPs differ with respect to the evolution and success of exemplary verb–noun pairings and indicate that frequency bands play a role while the size of the noun family and their distributional patterns do not.

Pseudocoordination with posture verbs in Mainland Scandinavian (e.g. the Norwegian Han sitter og arbeider ‘he sits and works’) is traditionally considered a grammaticalized progressive construction. The posture verb is said to have a bleached meaning, and to have the grammatical status of an auxiliary or a light verb. In recent years, some researchers have expressed doubt about this view. In this article, I argue that the traditional arguments for grammaticalization do not hold. However, I also give new evidence for early grammaticalization. Posture verbs can to some extent be used as light verbs in sentences such as Kebab må sittes og nytes ‘kebab must sit.pass and enjoy.pass’, which have never been discussed in connection with grammaticalization. Finally, I argue that pseudocoordination with posture verbs should not be seen as progressive, but rather as a locational (or situative) construction.

This study provides an empirical analysis of light verb constructions in Old Swedish. These constructions contain a semantically light verb, such as giva ‘give’ or göra ‘make’, that may be paired with an abstract nominal object, as in giva radh ‘give advice’ or giva hiälp ‘give help’. Using a corpus of nine Old Swedish texts written in the 13th, 14th, and 15th centuries, I track the frequency of light verb constructions and analyze the range of transitive light verb + object pairings. I consider the effects of time, genre, and the type of modification to the nominal object in the quantitative analysis. The results contribute to ongoing discussions in crosslinguistic, diachronic research on the reasons for the increase in frequency of light verb constructions as well as the possibility that this construction exhibits characteristics of grammaticalization or lexicalization.

This study employs corpus semantic techniques to examine the semantics of light verbs and light verb constructions (LVCs) in Singapore English, Hong Kong English and British English via their respective components in the International Corpus of English (ICE; Greenbaum 1996). The study investigates onomasiological variation (see Geeraerts et al.1994) by identifying selection preferences in natural use between light verb constructions and their related verb alternatives. In addition, identity evidence is forwarded as a valuable corpus semantic tool, in which instances of naturally occurring language data resemble classic identity tests for polysemy. Via a close reading and manual semantic analysis of thousands of instances of light make, take, give and their semantic alternatives, this study finds remarkable consistency across the three varieties of World Englishes (WEs) in onomasiological preferences, even in extremely nuanced features of LVCs. Onomasiological evidence and identity evidence also suggest the new finding that the three light verbs and their constructions exhibit degrees of lightness, and that these degrees of lightness are extremely consistent across regional varieties.

In this article, the author demonstrates that verbal compound constructions involving an ideophone and a light verb represent a widespread syntactic device in the world's languages. The author provides evidence that phono-symbolic morphemes cannot be treated as ‘bare’ direct objects in such constructions. Ideophones appearing in the light verb-adjacent position form a semantic unit with the verbal predicate, despite the fact that in some languages they can be syntacticized as (bare) nouns and appear in argumental position. Specifically, ideophones in complex predicates are part of the verbal domain with which they ‘blend’ (yielding a single predicate) through the mechanism of conflation, along the lines of Hale and Keyser (1993, 2002), and building on Ramchand (2008).