Expressive semitones: Music students’ perceptual preferences for melodic intonation on the violin

Abstract

Studies of performance intonation and musicians’ own statements suggest that classical string instrumentalists often deviate in their intonation from equal temperament for expressive purposes. However, it is not clear to what extent corresponding perceptual preferences for intonational deviations might rely on listeners’ instrumental expertise or such contextual aspects as the metrical placement of tones. We investigated higher-education music students’ perceptual preferences for melodic intonation of local leading tones in unaccompanied classical violin performances. Recordings of 12 excerpts were manipulated in the size of ascending semitones (110, 90, or 70 cents) leading to tones that were more stable in the tonal context. Groups of violin students and music education students listened to pairs of excerpts differing only in the size of semitones and chose the intonation variant that they preferred. In the comparison between 90- and 110-cent semitones, violin students showed a stronger group consensus for preferring 90 cents. However, greater instrumental expertise did not result in a stronger overall preference for the sharpest 70-cent variant. Instead, the violin students showed an expertise-related connection between intonation preference and meter, which was not observed for the music education students. In particular, the violin students more often preferred 70-cent intonation (i.e., the sharpest leading tones) in metrically unaccented than in accented positions. In effect, this is to prefer an expressive intonation that colors the music while not challenging the harmonic structure at metrically salient tones. It is argued that understanding expressive intonation in musical performance requires consideration of the metrical context.