Andrew Kania poses what he calls the striking cover paradox. The idea is that there could be a series of covers, each making small changes to the one before it, so that the final product sounds nothing at all like the original.
Here’s how he poses it:
If rock songs are so thin that they admit of “thickenings” as varied as Elvis’s and the Pet Shop Boys’s versions of “Always on My Mind,” we can imagine a chain of tracks, A through Z, where B is a cover of A, differing in some significant properties (such as the harmony and instrumentation of the Elvis/Pet Shop Boys example); C is a cover of B, differing from it as significantly, though perhaps along different dimensions; D is a cover of C; and so on, until we reach Z, a track that, though it is a cover of a cover of… a cover of “Don’t Be Cruel,” sounds for all the world like “Pop Goes the Weasel.”1
Kania’s preferred resolution to the paradox is to deny that A through Z is actually a series of covers. Suppose Q is the first one in the series that is sufficiently different that it is no longer an instance of the same song as A. Kania says that Q is not really a cover after all.
Kania gets into this problem because he assumes that the target of a cover is a song rather than an earlier version. In the series he imagines, B is not a cover of A but instead a cover that’s the same song as A; C is not a cover of B but instead a cover that’s same song as B (which is the same song as A); and so on. What makes them covers is that the artist who plays B learns the song from listening to A, the artists who plays C learns it by listening to B, and so on.
On Kania’s view, each version of the song is expected to serve as a paradigmatic instance of the song that you could use to learn it from— and a version is a cover when it takes a previous version in this way. He offers this definition: “A cover version is a track (successfully) intended to manifest the same song as some other track.”2
There’s a lot wrong with this, but what struck me recently is how it parallels Nelson Goodman’s conception of musical scores. A score for Goodman is the formal structure of the work. So he presumes not only that the score must allow one to perform the work, but also that a performance of the work must allow one to write down the score.
This leads Goodman to the much-ridiculed conclusion that an instance of a work must be perfect in every detail. One missed note makes it not a flawed performance of the work but not properly a performance of the work at all. Here’s his reasoning:
If we allow the least deviation, all assurance of work-preservation and score preservation is lost; for by a series of one-note errors of omission, addition, and modification, we can go all the way from Beethoven’s Fifth Symphony to Three Blind Mice. Thus while a score may leave unspecified many features of a performance, and allow for considerable variation in others within prescribed limits, full compliance with the specifications given is categorically required.3
Note that Goodman’s absurd conclusion does not follow from a view which requires approximate compliance to the original score. Suppose, with mock precision, that we allow a performance to have up to 3 wrong notes. Different performances might have different wrong notes, but no performance will have more than 3. The accumulation of errors only occurs on Goodman’s view because he insists that one could transcribe a new score for the same work from that first performance with 3 wrong notes. A performance of that new score could have 3 wrong notes in different places. And so on, until all the notes were wrong.
One can avoid all that by denying Goodman’s assumption that the work is fully recoverable from a performance. If the first performance has 3 wrong notes, then a faithful transcription will inherit those errors and be just as wrong as if one made a copy of the original score and mistranscribed 3 notes.
Back to Kania: I think he falls into the trap of thinking too much like Goodman.
If one wants to cover Elvis’ version of “Don’t Be Cruel”, one listens to Elvis. Version Q in his imagined series might be a different song that A, but it is still a cover of P if it’s a response to P in the appropriate way. Maybe the divergence shows that even P wasn’t the same song as A anymore.
In thinking about actual cases, it’s often not clear whether two versions are the same song or not. Some covers seem to be almost but not quite the same.