I see taxonomy as kind of a tool, Sandy. I was wading through Bronson's scale analysis when out visiting Sam Hinton in La Jolla. I found it fascinating. I think that Alan's Cantrometrics is a useful tool as well as Charlie Seeger's "mellotron" (his notational graphic machine to measure microtones. Regarding the accompaniment of traditionally unaccompanied songs, I think that it can be handled tastefully but not necessarilly as an element of that tradition. Bringing a guitar or banjo into it may change it somewhat from the original tradition of the music. Then it becomes a re-interpretation of the original which is OK by me. Alan Lomax had a lot of problems with it though when it strayed too far from the musical traditions, enough so that he was vocal about it. An aesthetic in the re-interpretations is set up sometimes. An example of a violation of this is the lack of understanding that would cause some coffee-house "folkie" performer to play Waltzing Matilda in three-quarter time. (As a waltz). I think that maybe Richard-Dyer Bennett's interpretation of John Henry might be questionable but his rendering of early English and Scottish ballads in the traditon of the Elizabethan Troubadour (ala Campion or Dowland) I believe is interpretively on the mark. Dyer-Bennett is a classically trained singer and guitarist so he would interject this eclectic approach into his music. But he would never refer to himself as a folk-singer or attempt to convey that he was a part of the traditions of the songs he sang. Now what happens when a singer or musician attempts to imitate the primary source folk singer?