Covers and Remakes

“Nothing Compares 2 U,” in Sinead O’Connor’s version, once grabbed the number 2 spot on a list of the Best Cover Songs Ever. It’s funny, therefore, that one theory of covers disqualifies it from even competing. A recording of a song, Theodore Gracyk says, is a remake, if the song has been recorded before; and “Nothing Compares 2 U” had. But more is needed for it to be a cover. For that status, the artist must have intended listeners to compare the new version with some particular recording of the song; and also have intended that the audience recognize (try not to get dizzy) that the artist intended this. It follows that a cover of a song cannot stand on its own as a vehicle of meaning. Fully appreciating a cover involves laying it against the original, and sussing out the significance of every departure and of every moment of faithfulness. I am old enough to testify that Sinead O’Connor’s first eager listeners did no such thing. Neither—almost certainly—were they intended to do so, by the singer or her producers.

On Gracyk’s view, cover song is a recent piece of cultural technology. No one could have made one before the mid-20th century, any more than one could have ridden a steamboat in ancient Rome—the difference in abstractness of the relevant technologies notwithstanding. Of course to call the cover song a piece of cultural technology is not to deny that innovations in material technologies were required for its development. You cannot cover a song unless a decent segment of your audience is familiar with the version you are covering—a prerequisite only met with the invention of the turntable, the radio, and in a crowning achievement, the eight-track tape player. O brave new world.

Also required is a musical culture that treats particular recordings of songs as the primary bearers of musical interest. That’s why the cover song is “a thing” in the world of rock and pop, but not in jazz or classical music, even though those forms of music may also be broadcast to the masses over the airwaves, or pumped one by one into bluetooth headphones. Tons of jazz musicians have recorded “My Funny Valentine,” but no recording of it is “canonical,” in the way that The Rolling Stones’ recording of “Satisfaction” is canonical. And only if something is canonical can, say, Cat Power’s recording of the song be a cover, rather than just another version.

Gracyk says cover songs are “saturated allusions”—they “intentionally parallel and thus reference another [performance],” where “every aspect of the performance is to be treated as referencing all aspects of the earlier recording at parallel points.” But not all saturated allusions are covers. The cover (in the other sense) of The Clash’s London Calling intentionally parallels and references the cover of Elvis Presley’s Elvis Presley. And every aspect of similarity and difference is significant. If the slightly-bubbly lettering on Elvis Presley is a bit fun (and now looks expressive of the relative innocence of the 1950’s), the same lettering style on London Calling is ironic—especially in the context of the different cover photos. The London Calling photo does not just, as Q magazine said, “capture the ultimate rock 'n' roll moment—total loss of control,” it comments on how that loss of control has changed since 1956, from Elvis’s joyous and unbuttoned sexual energy, to something darker and more violent. For all this, the cover of London Calling is not a cover of the cover of Elvis Presley, because the two do not stand to each other as do two recordings of the same song.

Being the first to record a song is no guarantee of canonicality—of standing-out as a recording enough to be the target of cover versions. Kris Kristofferson wrote and then made the first recording of “Me and Bobby McGee,” but it is Janis Joplin’s remake of his song that gets covered. A fascinatingly-complex case is Leonard Cohen’s “Hallelujah.” It’s been remade countless times, but the important remakes were by John Cale and then Jeff Buckley, the last of which is the one all new recordings are covers of. Does Cale’s recording satisfy Gracyk’s conditions to be a cover of Cohen’s? Does Buckley’s—modeled more on Cale’s than Cohen’s—satisfy the conditions to be a cover of either?

Assuming the answer to any of these questions is yes, Buckley’s recording is not just a song about estranged love (if that’s what “Hallelujah” about); it also has layers of meaning generated by comparison to, and layers of meaning as commentary on, the way John Cale or Leonard Cohen performed the song. That’s a lot of meaning—and so “the alleged simplicity and shallowness of popular music...is challenged by the existence of covers.”

Can the polarity of covering be reversed? If a cover is a remake intended to be compared to a known predecessor, then an anti-cover would be a remake intended to be heard in cultivated ignorance of an earlier version. Indeed, in its most extreme form, an anti-cover would be a remake of a song intended to destroy all knowledge of some previous version. Unlike covering, anti-covering has never become a common practice, maybe because the cultural conditions under which it could thrive have never been in place. But as in so many other things, Taylor Swift is here the trailblazer: the “Taylor’s version” versions of her pre-Lover albums are remakes meant to drive the originals both from the marketplace and from our entire musical consciousness. The anti-hero indeed.