Richard, I don't really want to sidetrack this thread, but I just want to point out that definition of "folk" above is just that - one definition, which has to be seen in the context of the times it was put together. While I agree with most of the elements of Cecil's definition, I venture to suggest that the requirement for anonymity specifically be waived in these days of widespread literacy and multimedia record-keeping. In Cecil's days things were very different, and a song could become "anonymous" within a few years, as it spread further afield and out of reach of a possibly illiterate originator. I wonder if Cecil simply took anonymity as an indicator that the song had been "sifted" enought through the folk process. If so, the metric was unfortunate and applicable only to his own era.

Take John Connoly's "Fiddlers Green" as just one example. It has been going round from singer to group to instrumentalist and it already exists in several versions, none of them John's own. Words have been added in some places. Parodies have been written. It is often mstaken for an "Irish traditional" song (Irish? for crying outloud!). Is it a "folk" song?

According to Cecil's full definition, the answer has to be "no", simply because we know that John wrote the original version. I think the definition here needs updating.