Abstract
This thesis investigates the interaction between perceived rhythm and underlying reference structures in the experience of rhythm in music. The point of departure is that both music performers’ and perceivers’ body motions are relevant to the study of rhythm. Accordingly, in music genres with an intimate relationship to dance, here referred to as music–dance, rhythm needs to be understood in relation to the corresponding dance. The studies included in this thesis investigate rhythm patterns in sound and body motion in two music–dance styles, Brazilian samba and Norwegian telespringar, based on motion capture and sound recordings of professional musicians and dancers.
Both samba and telespringar consist of complex rhythm patterns. Samba is often characterized by so-called systematic microtiming at the sixteenth-note level. This was confirmed in our sound analysis of the samba groove, showing a medium–medium–medium–long duration pattern at the sixteenth-note level. In addition, motion analysis of the percussionist’s heel tapping and the dancer’s steps revealed motion patterns in synchrony with this rhythm pattern.
Telespringar, on the other hand, is often described as featuring a so-called asymmetrical triple meter—that is, the three beats in a measure are of uneven duration. According to previous studies, both the fiddler’s foot stamping and the dancers’ vertical body motions are related to this underlying meter. This relation was confirmed in the motion analysis of a fiddler’s foot stamping, which revealed a very stable long–medium–short duration pattern at the beat level. The dancers’ vertical motion patterns, however, deviated from theories suggesting that the turning points in the dancers’ vertical motion curves correspond to the meter. The thesis therefore suggests an alternative interpretation with regard to the dancers’ vertical motion curves—that it is the shape of the dancers’ vertical motion that corresponds to the underlying beat duration, rather than the turning points that correspond to the underlying beat positions.
The main conclusion is that the underlying sixteenth-note level in samba and the underlying beat level in telespringar should not be understood as deviations from an isochronous pulse of some sort. Instead, they should be understood as inherently (and necessarily) non-isochronous, in and of themselves.