And behind the hiss and crackle you can hear a rendition of the Ave Maria that sounds extraordinary to modern ears. There are specific features you might pick out; for instance, Moreschi glides rapidly on to some of the notes from an octave or more below, in a way that no modern singer would. But it is the sound of his voice, what you might call the tonal ideal embodied in it, that seems most bizarre: it has an acute, almost painful focus, as if it were a kind of sublimated primal scream. Of course we don’t know whether the recording exemplified Moreschi at his best; he may well have been nervous, for early recording processes were very intrusive. And we don’t know how far the way Moreschi sung the Ave Maria is typical of how it was sung elsewhere. Maybe the recording sounded as strange to Moreschi’s contemporaries as it does to us. But then again, maybe it did not. And that is the point: we don’t know, and what is more there is no way we ever can know. The conclusion is obvious: if we don’t really know what music sounded like at the turn of the twentieth century, how can we possibly know what medieval music sounded like? The honest answer is that we can’t.