A big question in the field of loan phonology, is that of the extent to which the perception of the speakers of the borrowing language affects the adaptation itself. Past researchers have had varying thoughts about this, ranging from the idea that perception is of utmost importance (Peperkamp, 2003), to the thought that is of very little significance (Uffmann, 2006). One phenomenon in this field that offers us useful insight into such issues is that of variable adaptation, namely when one word or sound is adapted in multiple ways into one language, and this is what I am interested in.
I investigate variable adaptation of the English FACE diphthong into German, which does not contain this “phoneme”. The diphthong is adapted sometimes to /ɛ:/ and sometimes to /e:/, and the source of the variability is not lexical, as German dictionaries list both pronunciations as possible options for German speakers. Jax (2011) suggested that the variation may be due to the internal timing of the diphthong’s formant trajectories, and the way this is perceived by German listeners.
Another important aspect within this field that I address is that of the methodology used in studying loan phonology. Generally, in past studies, the methods used have differed a lot based on whether researchers have been approaching the issue from a phonological of a phonetic viewpoint. For instance, the level of bilingualism that the subjects who are tested possess varies greatly. By taking methods generally used to measure the perception of an L2, such as those used by Boersma & Escudero (2002, 2004) and Iverson & Evans (2007), I am able to control for more variables, enabling me to argue that loan word adaptation can neither be treated as an entirely phonetic nor an entirely phonological process.
