Misunderstanding now cleared up, 'Guest of 24'. I don't think there's any disagreement on the issue of selection in the sense of the body of material that never took hold on the public imagination. The archives are full of unsung broadsides, for one thing.

I felt that Gibb was quite reasonable in the flexibility of his personal position, and simply pointing out that a class described as 'Ethnomusicological' might attract criticism if it ran counter to the tenets of that discipline. I found the reasons for the evolutionary concept having been rejected in the 1950s very instructive, but of course that doesn't mean that we folksong people necessarily have to reject it as well.

So, to return to the original question:

Lighter: "A process of gradual change occurring in a system, institution, subject, artefact, product, etc., esp. from a simpler to a more complex or advanced state. Also: a gradual and natural development as opposed to a sudden or instigated change (often in contrast with 'revolution')."

The second of those is more likely to be applicable in the case of folk songs. Sharp and others of his era found modal tunes so aesthetically pleasing that they must have arrived at that state as the result of improvement through evolution. At the same time they were well aware that in the real world the old songs were dying out in the face of competition from the music hall and elsewhere. The asteroid had struck, and they were struggling to survive in the altered environment.

There are several problems with Sharp's analogy of the pebble being polished to perfection. First, the songs he collected were in many cases insufficiently old for gradual evolution to have occurred across many generations. Second, even he was well capable of identifying numerous variants they found in the field as 'degenerate' or 'corrupt'. As for demonstrating the process at all, it's difficult to find much hard evidence of change in tunes over the previous hundred years, simply because most of the collecting went on between the late 19th and early 20th century. I can think of a few examples of changes occurring between one singer's rendition and that of the person who learned it from them, but demonstrating the process over additional links in the chain would be very difficult with the data we have.

As it happens, the topic of change in folksong tunes came up at the Songs in Tradition and Print conference in Sheffield yesterday. You can certainly identify examples of clearly related, but interestingly different (in respect of mode, for example) tunes in the old collections, the degree of difference sometimes reflecting their geographical separation. What is not at all certain is whether the changes that did occur were voluntary or involuntary, whether individual singers were even aware of, say, the distinction between a sharp or flat seventh, and whether the collectors were able to notate accurately the sometimes ambiguous intervals they were hearing.