|
Rhyme occurs when two or more words are phonologically identical from the final stressed vowel onward. However, there are several types of so-called imperfect rhymes in which vocalic and/or consonantal segments are allowed to differ. Some of these types frequently replace perfect rhymes in verse-final positions in the German poetic tradition, which suggests that they are licensed by genre conventions. Thus far, however, there is little empirical investigation into whether or not specific subtypes of imperfect rhymes are actually perceived as rhyming. Using a speeded rhyme judgment task, this study examines (a) how imperfect rhymes are perceived by comparing them to perfect rhymes and nonrhymes and (b) systematic differences between subtypes of imperfect rhyme. Specifically, we studied three subtypes in which the relevant segments differed in terms of vowel roundedness, voicing, and vowel quantity. If participants did not answer the speeded rhyme judgment task within 750 ms, the trial timed out and the next one appeared. We analyzed both rhyme judgments (yes/no) and time-outs (valid/timed-out), the former indexing rhyme acceptability, the latter indexing uncertainty. Our results indicate that imperfect rhymes are less acceptable than perfect rhymes and elicit greater uncertainty in the rhyme judgment task than both perfect rhymes and nonrhymes. Metered verse context increases the acceptability of imperfect rhymes, whereas perfect and nonrhymes were equally acceptable in word pairs and couplets. Furthermore, our results corroborate the notion that the degree of phonological similarity plays a crucial role in the perception of words as rhyming. However, our study does not lend support to the idea that frequent occurrence of specific imperfect rhymes in the poetic canon makes them more acceptable as rhymes.
|
