The relation between functionally similar forms is often described in terms of competition. This leads to the expectation that over time only one form can survive (substitution) or each form must find its unique niche in functional space (differentiation). However, competition cannot easily explain what causes functional overlap or how form-function mappings will be reorganized. It is argued here that the changes which competing forms undergo are steered by various analogical forces. As a result of analogy, competing forms often show attraction, becoming functionally more (instead of less) alike. Attraction can maintain and increase functional overlap in language. At the same time, competing forms are analogically anchored to a broader constructional network. Cases of differentiation typically follow from the relations in that network. Evidence is drawn from the literature and from three corpus-based case studies, addressing attraction and differentiation in English aspectual constructions, English secondary predicate constructions, and in a pair of Dutch degree modifiers. Evidence is provided of a phenomenon competition-based accounts could not predict (attraction), and a solution is offered for one they could not very well explain (differentiation). More generally it is shown that the development of competing forms must be understood against their broader grammatical context.

This article offers a cognitive perspective on the evolution of the semantics of English nominal gerunds (NG) (I regret the signing of the contract) and verbal gerunds (VG) (I regret signing the contract). While the formal differences between NGs and VGs are well documented, their semantics remains largely unexplored territory. The perspective that is taken here is centered on the linguistic notion of reference and various aspects of the conceptualization involved in it. As they formally hover between more nominal and more clauselike internal properties, gerunds form an interesting test case for the cognitive perspective on referentiality. Our corpus analysis describes how the situations that NGs and VGs refer to are conceptualized as deictic expressions grounded in the speech event in Present-day English, and how this has changed since the Early Modern period. It is shown that only a multi-layered model of referentiality can account for the subtle differences found between NGs and VGs: while no fundamental shifts are found with regard to the traditional referential subtypes (specific, non-specific, generic), NGs and VGs do turn out to differ in their choice for either nominal or clausal grounding mechanisms, in their status as existentially stable or flexible entities and in the mental spaces in which they situate the events that they conceptualize.

This paper illustrates how different methodological approaches can be combined to reveal complex patterns of constructional variation and change in the diachronic development of English ing-nominals. More specifically, we argue that approaching the data from a schema-based (rather than morphemebased) perspective shows that nominal gerunds in English, from the 16 th to the 19 th century, have undergone a semantic drift towards more "nouny" construal variants. This hypothesis is supported not only by raw frequency counts, but also by association measures and by a detailed analysis of hapax legomena.

This paper provides a detailed comparison of the referential behaviour of noun phrases and nominal and verbal gerunds from Middle to Late Modern English. It will be shown that in earlier stages of English, nominal and verbal gerunds to a large extent resemble prototypical noun phrases in their referential functioning, but also exhibit ‘non-nominal’ uses that depend on clausal rather than nominal grounding strategies. It is argued that the study of (diachronic changes in) the semantic and functional behaviour of nominal and verbal gerunds in Middle and Modern English should take into account that these are functionally hybrid constructions, showing referential traits of both prototypical noun phrases and clauses. This functional hybridity, then, was gradually sorted out, with nominal gerunds specializing to nominal reference and verbal gerunds continuing to adhere to the functional apparatus associated with subordinate clauses.

AbstractThis study investigates the extent to which there is individuality in how structural variation is conditioned over time. Earlier research already classified the diachronically unstable gerund variation as involving a high fraction of mixed-usage speakers throughout the change, whereby the proportion of the conservative variant versus the progressive variant as observable in the linguistic output of individual language users superficially resembles the mean proportion as observable at the population level. However, this study sets out to show that there can still be heterogeneity within such a centralized population in terms of how each individual conditions the observed variation. A random forest and conditional inference tree analysis of over 14,000 gerunds uttered by nineteen seventeenth-century authors is presented to show that, while the most important language-internal factors conditioning the gerund variation are adopted by (and shared between) all authors, we can still attest inter-individual variation (i) at lower levels of variable importance, and (ii) in the breadth of the range of contexts individual authors employ to condition the attested variation.