Is it likely that music-makers in the past were very much different from those today? Of course music-making (as opposed to passively consuming music) has become restricted to a much smaller proportion of society, but that still spans all social classes. Entirely empirically, it seems to me that far more people perform songs which have been written by others than are composers themselves. Whilst composition can be learned, up to a point, it takes a certain spark which not all singers and musicians have, and I would say most do not. We only have to look around us to see that whilst we may all be capable of turning out the occasional song or tune, only a handful are doing so on a regular and consistent basis.

I have no evidence for the actual ratio, but to say that maybe 5%-10% of all singers and musicians are composers seems about right to me. I don't believe that essential spark would have exhibited itself any differently amongst previous generations. This suggests to me that the majority of singers, across all classes, have always been reliant on a small number of composers to supply them with songs.

Looked at this way, the claim that most folk songs were written by outsiders is less surprising. Otherwise we would have to believe either that the working classes were somehow isolated from other musical influences and could rely for songs only on what could be created from within their own communities, or else that they were somehow disproportionately gifted when it came to composing songs. Roud shows how the first was not the case, and the second seems unlikely simply on normal statistical distribution grounds.

I don't think such a ratio would be thought surprising for the music played by the middle and upper classes. Although there would have been some who composed for themselves, most of what they performed would have been by professional composers. Why should the sources of folk music be different? What characterises folk song is not where it came from but what then happened to it.