The theoretical landscape Method Analysis

The theoretical landscape

English word-formation has been studied by many scholars (Adams 1973; Bauer 1983; Štekauer 2000; Plag 2003; Štekauer & Lieber 2005; Bauer et al. 2013), whose attention was primarily focused on morphological grammar. Another field of study in morphological research pertains to word creation, which “refers to all operations for the production of new lexemes which are not covered by regular word formation” (Ronneberger-Sibold 2008: 201), including “creative techniques” such as shortening or blending. For these and related creative techniques, Baldi & Dawar (2000: 963) have used the label “unconventional word-formation”, in that blending and abbreviations are out of ordinary (conventional) norms and rules, whereas Zwicky & Pullum (1987) classify them as “expressive (vs. plain) morphology”, on account of their pragmatic effect. According to Hohenhaus (1996), blending and acronyms are part of what he calls “Ad-hoc-Wortbildung” (ad hoc word-formation), that is, the creative formation of new words, often by means of direct analogy. Aronoff (1976: 20) groups together clippings, blends, and acronyms under the label “oddities”, while Bauer (1983: 232) calls them “unpredictable formations”, though emphasising that they are so common in English that “it is misleading to consider them out of the ordinary”. The present study explores the relationship between blending and the analogical process, showing how the latter affects the former, conferring regularity on blends and increasing their degree of predictability.

In a volume on lexical creativity, Lehrer (2007: 116) defines blends as “underlying compounds which are composed of one word and part of another, or parts of two (and
occasionally three) other words”, adding that each word part in a blend is called “splinter”. In this study, the term ‘splinter’ will be likewise used as synonymous with ‘blend’s part’ (cf. its broader sense in Bauer et al. 2013).
Blends have been variously studied for their contribution to neology (Kemmer 2003; Lehrer 2003), for their prosodic structure (Gries 2004, 2012), and their preferred contexts of use, namely literary texts and product names. For instance, in Richard Lederer’s book Adventures of a Verbivore®, the latter is a blend that merges verb and (h)erbivore, and, similarly, the creative name of the fruit-flavoured drink Fruitopia® originates by blending fruit and (u)topia (both in Lehrer 2007: 129–130).
While blends are similar to compounds in that they merge two or more words into one, their irregularity and only partial predictability make them out of ordinary English word- formation rules, i.e. extra-grammatical. Hence, blends exhibit a higher degree of creativity than creative (analogical) compounds. Indeed, creativity is a scalar concept (cf. Plag 1999; Bauer 2001; see Ladányi’s 2000 productivity-creativity scale) and can involve analogy. For instance, Benczes (2006: 6) labels “creative compounds” those compounds which are based on metaphorical or metonymic associations, such as scarlet-collar (worker) [2000] ‘a woman who operates an Internet pornographic site’ (Wordspy), inspired by white-collar (worker), blue- collar (worker) and similar complex words. However, the formation of the compound adjective scarlet-collar, based on the compound family X-collar (Mattiello & Dressler forth.), is more predictable and less creative than the formation of a blend such as boatel ← boa(t + ho)tel [1950] ‘a boat which functions as a hotel’ (OED3), after the unique model motel ← mo(tor + ho)tel.
Nonetheless, recent approaches to blends have conferred some regularity on blends, by claiming that morphological and prosodic factors can influence their structure. For instance, Mattiello (2013) stresses that the prototypical blending pattern retains the initial part of one source word and the end of another, as in smog ← sm(oke + f)og [1905] (OED2). Moreover, Arndt-Lappe & Plag (2013) have proved that 1) blend’s length most often conforms to the length of the longer source word, and 2) there seems to be a preference for blends to have no more than three syllables. In this paper, we assume that the regularity and partial predictability of blends are also connected with some frequently occurring splinters, such as -ercise ← (ex)ercise, found in sexercise [1942] ‘sexual activity regarded as exercise’ (OED3), dancercise [1967] ‘dancing performed as an exercise’ (OED2), and boxercise [1985] ‘a form of aerobic fitness routine incorporating exercises from boxing’ (OED3). As Lehrer (2007: 121) observes: “[w]hen a splinter becomes so common that people start using it frequently, it may lose its connection with the source word and can be considered a morpheme in its own right”.
Nowadays some of these splinters have become so common and productive in English that they deserve the label of “combining forms” (Warren 1990) or “secreted affixes” (Fradin 2000). This regularity in use is connected with the analogical process.

1.