Jim, I think you are being a bit dogmatic there.

Steve Roud doesn't assert for no reason that we can't categorise folk song in terms of its origin: he makes a lengthy and well argued case for this position. There is, he says, not enough evidence to support the idea that there is a large corpus of songs which originated within the singing tradition rather than in the music halls or pleasure gardens. That's not a political judgement, it's an empirical one.

He clearly feels motivated to defend the early folk song collectors from what he sees as the unfair and anachronistic criticisms levelled by people like David Harker. Not having read Harker I can't say whether Roud's presentation of his ideas is fair, but I would agree that we owe people like Baring-Gould and Sharp an immense debt, and that their work should not be written off just because it doesn't meet the ideals we would have if we were doing the same thing today. I don't think Roud comes across as right-wing in any way. Indeed at one point he makes a slightly waspish comment about right-wing thought historically not being intellectual.

As you say, the book is huge already. I can totally see why he didn't feel the need to extend it further by adding detailed discussion of individual songs, though I agree a discography would be nice.

I don't really understand your accusation that he deals with folk songs as "a commercial product manufactured for the entertainment of the people". Where there is evidence that folk songs originated in other musical contexts, he says so; and his discussion of the other musical milieus that were current from the 16th century onwards is fascinating and often eye-opening.

If I have a criticism it's that Roud is clearly not a musicologist, and so the chapters on the music by Julia Bishop feel a bit 'bolted on' rather than fully integrated into the book. There's no discussion for instance of the forms of music notation that have been available through the years, or how widespread musical literacy was.