I am perhaps in danger of projecting my own predilections outwards here I've never bothered to learn those Colin and Phoebe type songs.

But one frustrating aspect of Roud's book is that I feel he doesn't involve himself in the repercussions of some of his findings. If a wassail song or May song or a seasonal songs relating to winter mendicant traditions was probably written by an urban broadsheet writer, that to me gives rise to all sorts of questions. Did much that we take for granted about the content of those traditions never actually exist in practice? Were broadsheet writers actively intervening/shaping the content of those traditions? Given that such songs are a significant part of what many folk singers today would regard as canonical, it seems odd to me that this wouldn't be explored in a large social history of folk music.

Another omission that I find odd in the book is that, given Roud's scholarship, he's uniquely placed to provide informative demonstrations of the folk process at work: while he mentions the fact that, just because working people did not write the songs, they liked them enough to learn them and shape them, it seems bizarre that he doesn't present any examples of how transformative (or not) this was.

I say bizarre because this is pretty much the key element of folk song. I had generally adumbrated broadsides as flowery, laborious and over-written, as compared to a folk-processed poetic, streamlined economy in a folk song as I have learned it. I mentioned the Streams of Lovely Nancy earlier because it was the most dramatic example I could think of of the folk process at work: a song that common sense suggests probably wasn't first written the way it has come down to most of us. But there are much more lucid examples I can think of off the top of my head: Six Dukes Went A Fishing for example, or the version of 'Brisk Young Sailor' collected in the Vaughn Williams 'Bushes and Briars' book. Or the genuinely weird song 'The Pelican' (collected by Gardiner, I think).

It seems odd to me that someone writing such a mammoth project, entitled 'Folk Song in England' wouldn't want to discuss the folk process more, and provide examples from his considerable research showing it at work. Those conclusions might be "actually, songs don't change that much"; or they might be "it's interesting to note what the song loses in unnecessary detail from this broadside of 1860 to the version collected by Cecil Sharp in..." There's none of that (so far as I've read) in Roud's book. Which is one of many reasons I'm finding it quite a frustrating read.

Another thing that's just occurred to me is there's not much discussion of the Child ballads, which are a pretty canonical part of folk singing today. But I guess Roud would point me to the bit in his introduction where he states that the book is about what folk music was, rather than what it is. I'm increasingly feeling that Roud's own priorities about what's important to discuss, to expand on, to go into detail on, or to include or exclude, are very different from my own interests in folk music. I think I was expecting a very different book to this one.