Edited by Peter Petré, Hubert Cuyckens and Frauke D'hoedt
[Current Issues in Linguistic Theory 343] 2018
► pp. 227–252
Shortening processes are increasingly productive in Present-day English (PDE), generating 9–15% of the new words in the language. They are therefore a valuable source of information on the interaction of production, perception, prosodic well-formedness, morphological patterning, and sociolinguistic and pragmatic aspects of word-formation. One type of shortening, word clipping, is considered unpredictable by some scholars (Durkin 2009: 116; Don 2014: 27), while others (Plag 2003; Lappe 2007; Berg 2011) find non-random patterns in the outcome of clippings in PDE. The diachronic records of clipping have not been addressed systematically in the literature. Data collected from the online OED show that clipping in early English is restricted to fore-clipping, which peaked between 1300–1600 and then decreased quite sharply. Back-clipping was practically unattested until the end of Middle English, whereupon it rapidly became the dominant model, making fore-clipping a marginal pattern for common words. The association between prefixation and lack of stress during the period of widespread fore-clipping, and the stability of the left edge of the word in back-clippings after the 15th century require reference to both the surplus location and composition, and the residue structure. An account in terms of onset, stress, Morphological identity, and align-l shows how the relative importance of these constraints changes in tandem with independently established changes in the history of English.