Perceived vowel quantity in Swedish: Native and British listeners

Abstract

In many languages, vowels are characterized by their use of contrastive phonological vowel quantity and vowel quality. In Swedish, vowels have traditionally been described as being distinct in quality as well as having a phonological distinction between short and long vowel quantities. In English, however, phonological distinctions among vowels are described as primarily qualitative. This investigation examines the perceptual use of vowel duration and the first two vowel formant frequencies in distinguishing Swedish vowel pairs by three groups of listeners: native Swedish listeners (SS), British English listeners who do not know Swedish (EE), and British listeners who know Swedish well (ES). For each of three pairs of Swedish vowels (IPAKiel[i=DC]IPAKiel-[I], IPAKiel[o=DC]-IPAKiel[O], IPAKiel[A=DC]IPAKiel-[a]), /kVt/ words were resynthesized having ten degrees of vowel duration and ten degrees of F1 and F2 adjustment. Listeners’ responses and reaction times in a rhyming task show that unlike native listeners, L2 (i.e., ES) listeners distinguish Swedish vowel quantity using duration almost exclusively, which is consistent with what they have been taught, whereas TimesEE listeners respond as expected for a comparable English vowel continuum, using both vowel duration and spectra. These results are discussed in terms of the different stategies used by the different listeners.