Social dimensions of melodic identity, cognition, and association

Abstract

While music researchers are interested in developing tools for automatically culling related melodies and psychologists seek a clearer understanding of how people learn, recognize, and remember melodies, musicological and ethnomusicological studies offer numerous studies of tune families. Tune families cohere, it seems, partly on the basis of cultural agreement. Melodies seem to be similar if people say they are. Are there particular musical characteristics which are favored in the formation and cohesion of family members? Music scholars have been investigating this question for a half century. As their investigations have accrued, respect for complexity of melody and the many ways in which two melodies can be the same or different has increased proportionally. This study examines five tune families and evaluates selected members of each according to a previously proposed cognitive-distance measure. Concordances and cases of musical divergence in the context of claimed relatedness are evaluated in relation to collection-formation types, melodic types, and transmission methods (printed, oral, both). Neither hand-picked members of families nor those culled by title preservation necessarily show a high degree of musical coherence. However, among the families chosen, the predominance of pieces originally associated with dancing is conspicuous. This suggests that even when contour and other features vary, lengthy rhythmic patterns may underlie the social identity of melodic similarity.