On Covers

First, definitions. A “cover” is a work that is largely derivative of a different, largely original work. Of course, nothing is original and everything is derivative, but I think this definition is functional when taken in good faith. Also a work that is largely derivative of a different, largely original work is what I’m calling a “translation”. Translations and covers are subtly but profoundly different. A translation is incredibly direct. It answers the question, “How would I play this song?”. A cover is filtered, altered, changed. It’s a reflection of the original song colored by the coverer’s experiences. It answers the question, “What does this song mean to me?”. For clarity: both of these terms are specifically for works that are published publicly. If it stays private, if it’s for you, it’s a study. Study’s are private, for learning. You can do whatever you want with that, I don’t care. I only start to care when you make something everyone else’s business. When you make something everyone else’s business, when you publish, you take on the responsibility to do it right. I believe this goes for everything, and everything includes covers.

Covers are often translations in disguise, works that add practically nothing and exist solely to mooch off of the success of their foundations. Translations are everything I hate about covers with none of the redeeming factors. They only take, they don’t give. They confuse and deceive. There are only two scenarios in which a translation has any value. The first is when it’s a live performance. Recorded translations are dumb because in the act of recording and producing, performance becomes part of composition and the composition was already decided by the original artist so what’s the point? In live, performed translations, performance has it’s own value, so you at least get that. The second scenario is when it’s a literal translation from one language to another. If a translation is the only way for a song to be heard by whatever other language’s speakers, then fine. I’d rather the original artists do it, but you take what you can get. So long as the translation is pure, or at least tries to be. The Bible doesn’t have a Book of Luther.

With that said, now I’m gonna make a little addition to my definition, get a little more specific. A “cover” is when somebody makes a derivative work that informs what an original work means to the coverer themself- and everyone knows that that’s what it is. It’s when the work is of good quality and just done well. I don’t like covers, but others do. They have value, if what I would call niche value. What I cannot abide by is what I dub the “covn’t”. This is when a cover is made in a way outside of the particular tolerances that make it… well, tolerable. Specifically, it’s when a cover is shared in the wrong channels with the wrong labels. When you publish you have to publish correctly. If you don’t, you lose something, you fail, your product is corrupted. Covn’ts are bad. I believe covers, generally, don’t succeed. I believe covn’ts can’t.

Why should the wrong channels and the wrong labels effect the quality of the product itself? How much harm could calling a product something less than accurate really do? It’s obviously not great, but surely it doesn’t change anything about the actual cover, right?

Wrong. Covers have issues with possession and categorization, troublesome but easily enough ignored. Covn’ts have issues with something far more vital, far less able to be overlooked: semantics. The most critically important area of knowledge in all of society. I say that a little humorously, but it’s no joke. Semantics are, well, everything. Language is a filter we use to process sensory inputs into reality. Language is the filter we use to communicate that reality to others. If you disagree, I ask that you find a way to let me know, maybe tell me why. I wonder how you’ll do that. Semantics are critical in any and all forms of communication, and communication is what art is. When you communicate with lies and deception, you lose clarity, which is absolutely necessary to communicate any sort of specific idea. The argument could be made that clarity is not always a value in art, and within a work I’d say that that’s true. Vagueness and misdirection are valid effects within a work. But without a work? That’s meta in such an obnoxious way. It’s not artsy, it’s illegible.

So, how do covn’ts take advantage of lies and deception?

First is the obvious: the misleading standard we’ve adopted for labeling covers (thereby making them covn’ts). Take the song, “I Will Survive” by Gloria Gaynor. It’s made and released in the 70’s and all is well. I tell somebody I like the song. I direct somebody to the song. All is well. Then, some time passes, and Cake covers the song. I love this “cover”. I think it’s better than the original. So now what? I tell somebody I like the song. They get confused, and think about the original- after all, would the original not be the definitive version by default? Well, not for me. I like Cake’s version so much more, so that has become the definitive version by virtue of quality over originality. So now to the next person I start by saying I like Cake’s “I Will Survive”. “Oh! I didn’t know Cake wrote that song!”. So I explain, no, they didn’t, it’s from this person in 70’s, the Cake version is a cover. The next time, I say I like the song “I Will Survive” by Gloria Gaynor, as arranged and performed by Cake. Third time’s the charm. “I Will Survive” by Cake is a covn’t. Cake’s arrangement and performance of “I Will Survive” by Gloria Gaynor is a cover. My frustration here is that covers aren’t contextualized this way to begin with, and they should be. I’ve personally revealed to too many people that “We’re Gonna Be Friends” is not, in fact, by Jack Johnson. That’s an issue! And it could be so easily and obviously solved.

The other way covn’ts deceive is a little less direct: the misleading practice we have in organizing covers. When we put covers in the same places as original songs, we make them covn’ts because we imply that they are the same, that they should occupy the same space. Which is false, because they shouldn’t, because they are not the same. The clearest example of this I can think of is artists releasing original albums with a covn’t or two in them. And they are covn’ts, even if they’re properly labeled. When I put a song in an original album, I’m claiming it as my song. That implication is strong, undeniable. So when there is a song there that isn’t actually mine, I’m made a liar. There are ways to release a good cover so as to keep it a cover and not a covn’t. Cover albums are good, obviously. Harder to communicate clearer than that. Themed groups, themed projects are good. That is, artists and/or projects whose purpose is making covers. Cover bands are the most obvious example of this but there are also others. B-side compilations are iffy. They are clear in stating that they aren’t representative of the artist’s typical work, that they are not “regular songs”, so to speak, but they’re not for covers. I think that if a cover is properly labeled it could reasonably be put in a B-side compilation. The key is that the space in which a cover is shared must be clearly for covers, or at least not clearly for something else. Else, the cover is automatically corrupted and becomes a covn’t.

That’s my biggest issue with covers- systemically, they are broken and deceptive. In isolation they become a study, and I’m largely apathetic to that. But when published they’re often published into a broken system in which they exist as covn’ts, which sucks. They aren’t always published poorly. Cover albums and other appropriate avenues exist. But… well, beyond this point I’m more ranting than constructing an argument. The rest of this is largely personal preference. I still think I’m right, of course, but, well. That’s just how preference goes, isn’t it.

I have issues with covers. I generally just don’t prefer them, I think they’re inefficient. I think that most anything they can do, something else can do better. There’s nothing objectively and inherently wrong with covers (assuming they’re done right, assuming they’re truly covers), but I don’t think they are… optimal, efficient, ideal. I don’t like them.

I tend to think of covers as parasitic. There existed a definitive version of a thing (and truly definitive anythings are lovely in and of themselves) and you go and try to undefinitivize it? Covers have three potential results: you make a worse song than the original and it’s effort wasted, you make a better song, or you make a song that’s not strictly better or worse but is strictly different. Making a better song is awesome, but now you have a definitive version literally and a definitive version practically- and they’re different, which sucks. If you make a song different enough from the original it makes one wonder why you bothered with a cover over an original in the first place. I dunno. None of these options are appealing to me.

I don’t know if you could tell this about me, but I like it when things are easy to organize, when they fit into their spots. People love to say the world isn’t black and white, and, well, yeah. It’s not. But the world is black and not-black. False dichotomies are a fallacy, but true dichotomies are really useful tools for analysis and decision making, and they’re pretty available if you know how to look. When you frame something into a binary decision, you gain a lot of clarity in how you make your comparisons and you can thus be a lot more granular and accurate.

Covers make this really hard. Everything is black and not-black, sure. But covers offer conflicting information about the color of an object. One source says black, one says not-black. What do you do?

The core of the issue here is that I’m trying to evaluate a song, but I can’t do that without a singular song. A good thesis needs a defensible claim, and I can’t claim what I can’t define. I need to be able to be working with one song, one definition. Covers make this difficult, and I don’t like that.

Having multiple versions of a song makes ambiguous the ownership of the song. It makes ambiguous the brand and identity of the coverer. It makes everything more ambiguous (which is more difficult), and I do not see the appeal. Write your own song! If it’s just about performance, you wanna perform somebody’s song, then fine, whatever, go for it (preferably in a live performance setting). But if there’s any amount of songwriting and/or production and arrangement involved, then why not write your own song? What did you do to the song to make it worth your time, to make it worth anyone’s time? No, really, what did you do? Different speed? Different instruments? Totally different vibe, really just using the song as inspiration? There’s such a range of possibilities here, and nobody gets to know what’s really happening. The best the audience can do is listen to the cover and the original side-by-side and play spot-the-difference. That kinda sucks, though, because when you look at the cover you’ll find elements that were changed on purpose, elements that were changed on accident, elements that stayed the same because the coverer really liked them, and elements that stayed the same just because they became the default, and it’s not possible to identify between purposeful and non purposeful decisions just by listening. The safest option becomes assuming the coverer did little intentionally and disallowing their cover to define them at all- or at least, in any significantly useful way. The riskiest option is to have a huge amount of trust in the technical abilities of the coverer and assume that all decisions were purposeful. Of course, there is every level of risk and trust in between those extremes. And of course, that scale exists for original songs as well, for any creative work ever. The difference is that covers make you question purpose, make you put yourself on that scale, twice. Once for the standard judgement of a creative work, and again to differentiate it from the original song. Covers, then, are half as good as an original song when it comes to building understanding of an artist. I’d argue they’re worse, actually, because original songs live in packs, they come in EPs and LPs and they let you build a model of what an original song from an artist means. Artists typically don’t make so many covers, so it’s far more difficult to build a model of what a cover from an artist means. Cover artists who primarily cover just a few different artists alleviate this issue, so I get to learn what a cover of an artist from them means, but if they’re not also making original songs it’s really hard to learn who they are, who the cover artist is. Covers just don’t give good information- the data they provide is murky, diluted, impure. In a best case scenario like “I Will Survive”, the original and the cover are wildly different and it’s clear what influence the cover draws from the original and what it rejects. Both the original and cover then have individual, independent value that’s clearly discernible even at a glance. But then, if you ended up writing a song that’s so unique, so individual, then why didn’t you just make it a completely new song!?

“I Will Survive” as arranged and performed by Cake is probably one of my favorite covers I’ve ever heard. It’s incredibly different from the original, and I think it’s better. I think it makes much better use of the underlying material of the song. I think it saw potential and it realized it. It’s phenomenal. And I still almost wish it didn’t exist, at least not as it is in its current form. Because now there’s an original song that is definitive by virtue of being the original and there’s a derivative song that is definitive by virtue of being the better song. And that clash sucks. It requires the audience know both definitions to truly claim to know either, creating this excessive barrier to entry that’s not explicitly stated. Implicit gatekeeping sucks even more than explicit gatekeeping. The people who like the gate are pissed because newbies will dodge the gate on accident, and the newbies are pissed because they had no way to know they were still newbies and were under the illusion that they weren’t.

Covers just make everything more confusing. When done improperly they are covn’ts and poor translations, so they misattribute credit for original songs and they flood the ecosystem with (mostly) uselessly derivative works. When done properly I still feel like they could have been something else, something better.