Dualism and superposition in the analysis of English synthetic compounds ending in -er

Abstract

This article studies the role of synthetic-compound families, both formal families and their semantic (or rather conceptual) subfamilies, in the analysis of synthetic compounds (SCs). For this purpose, four formal families of English non-Latinate synthetic compounds sharing their second base and three Latinate families have been investigated. Unlike previous approaches ranging from a purely syntactic treatment of SCs to a more lexical treatment, this study aims at providing a novel explanation for these complex formations. First, it argues that SCs have an ambiguous nature, hovering between (a) morphological suffixation of a verb/word group and (b) morphological derivation and subsequent compounding. Second, it emphasizes the importance of compound families and subfamilies in SCs' formation and interpretation. By combining a corpus-based analysis with a qualitative synchronic and diachronic investigation of seven compound families - namely X-breaker, X-holder, X-killer, X-maker, X-manager, X-producer, and X-provider - the article provides a fine-grained semantic categorization of their subfamilies. Results show that (a) a homogeneous approach to SCs is not sufficient to account for observations in corpus data and (b) there are different types of SCs reflecting different constructional types. Nearly all results from questionnaires completed by native speakers confirm, or are at least compatible with our results. From a theoretical viewpoint, these results suggest a novel addition to the definition of productivity in word-formation: i.e., the ability to create new (sub-)families. The basic ambiguity between a derivational and compositional analysis of SCs can be resolved by assuming dualism and superposition of suffixing and compounding, similar to Albert Einstein's assumption of duality and superposition of waves and particles (photons) for light. Thus, several subfamilies (clearly or predominantly) consist of derivations, others of compounds, still others show superposition of both, and at least partial superposition seems to be possible for most of the other subfamilies.