A Fresh Look at the Concept of Tune Family

Abstract

The metaphor is not particularly new. In mid-nineteenth century, scholars like George Petrie and William Forde were grouping together Irish tunes which they considered be related of one (Petrie 1855:5). Later scholars such as Francis O'Neill and Phillips Barry were quite aware of these relationships, but Samuel Bayard was first propose an actual tune-family theory. Bayard sought to identify specific melodies in as many of their forms as possible (Bayard 1950:4), study which he believed would be a worthy partner of tune-indexing efforts; logical precursor and accompanier of melodic-variation studies, analyses of musical style, inquiries into migrations and history of songs, and studies in relations of melodies and words (ibid.). Thus he anticipated three basic applications for tune family theory: cataloging/ indexing, tune histories, and analyses (of variation, style, and text-tune integration). Although Bayard laid much important groundwork, he did not provide sufficient models for these potential applications. He recognized many of basic tune contours in Anglo-American repertoire, but did not set up kind of systematic structure for comparison of variants that is necessary for thorough cataloging study. He flirted with historical studies, sprinkling his works with informed conjecture about origin and development of certain tunes, but constantly pulled himself back by insisting (correctly) that the time when any folk-tune is recorded has reference real age of variant (ibid.:5). In addition, Bayard understood that no traditional melody ever lives or is recreated in vacuum-it is always liable influence from other melodies co-existent with it (ibid.:31) and that favorite scales and formulae used habitually among Irish singers, for example, would be those singers' ordinary way of making music; and what would be more natural than that these formulae and these modal characters should slip into versions of airs learned from some

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 36 citations exist upstream.