Tell Congress: Online Creators Need Net Neutrality!Learn how net neutrality affects your ability to make money online
Everyone, but especially copyright reformers, should look to more than fair use to protect legitimate uses of copyrighted works. While fair use is a vital doctrine, it should not be, and never has been, the only way to protect users from spurious claims of infringement.
When there is a contested copyright issue, many legal observers--including copyright reformers--immediately jump to a fair use analysis without first considering whether a use is non-infringing for other reasons. As a friend of fair use, I want to give it the day off once in a while. There are other ways that some uses of copyrighted works are non-infringing besides fair use. This is particularly so in the case of sampling in music.
Fair use, of course, is very important. Judges long ago realized that not every use of a copyrighted work required permission of the copyright owner, even when those uses fell within one of the "exclusive" rights the law gives copyright owners control over. Thus, the common law tradition gave rise to the fair use doctrine, which recognizes that while the rights of copyright owners to control certain uses are broad, they are not unlimited, and cannot be used to limit criticism or commentary, prevent education or transformative uses, and so on. (The test that courts apply is codified in 17 U.S.C. § 107, though the common law is still relevant.) Because of fair use, a user of a copyrighted work might be able to make a reproduction of all or some of it, without permission, even though the law grants copyright holders the right to control reproductions.
But notice what fair use is doing here. It is a limitation on the exclusive rights of the copyright holder. Those rights are listed in Title 17, Section 106 of the US Code. To summarize these (a bit inaccurately, since they vary by kind of work, but bear with me), they are the rights to reproduce the work, to prepare derivative works, to distribute the work to the public, to perform the copyrighted work publicly, and to display the copyrighted work publicly. Fair use means that there are uses that directly touch on one of these things that are nonetheless lawful. (It's not the only copyright limitation that works like this; first sale, for example, does the same thing.)
You know what fair use does not address? Uses of copyrighted works that do not touch one of the copyright holder's exclusive rights to begin with. Reading a book is not a "fair use," it is simply a use that falls outside of copyright. The Aereo case, which some have described as being about "a broad interpretation of fair use rights," is no such thing. That case hinges on whether Aereo is engaged in "public performances" (with a copyright holder's exclusive rights), or is simply enabling private performances (something copyright owners have no right to control).
Similarly, when it comes to the use of samples in music, many observers seem to think that all sampling is infringing, licensed or authorized in some way by the original copyright owner, or else a fair use. But there are other options! (PK discusses this in more depth in our recent Department of Commerce Copyright Green Paper comments.
The test in copyright law for whether a "reproduction" has occurred at all is the "substantial similarity" test. This means that a lay observer would have to look at the original, then look at the "reproduction," and say, "Yep, that's a copy." If I go into an art museum with a pencil and a sketchpad and try to make a copy of a painting, I will fail, since I suck at drawing. There will be no substantial similarity between the original work and my pathetic scribbles, which means that there could be no copyright infringement. Similarly, if you can't tell a work has been sampled in a new song--for example, because the sample is mixed in with so many other sounds it's not noticeable, or because the sample has been modified or distorted--then it makes no sense to say that a "reproduction" has taken place. The substantial similarity test is basic copyright law, but it is overlooked when it comes to sampling.
One famous early sampling dispute happened in the UK, where a member of the group MARRS mentioned on the radio that their track "Pump Up the Volume" incorporated a sample from "Roadblock" by Stock, Aitken & Waterman. This prompted copyright claims by Pete Waterman--but it seems absurd to claim that there could have been copyright infringement in a case where the original artist apparently had been unable to detect any "substantial similarity" without being told about it. It would be better if more sampling disputes were viewed with this lens: even though an existing recording may have been used as "raw material" in another artist's sonic creation, that does not mean that a "reproduction" for copyright purposes has occurred, or that the new musical work is somehow a "copy" of some other work simply because it incorporates such a sample.
There are two similar doctrines that are well-established in copyright law. The first is the de minimis doctrine. This comes from the Latin phrase de minimis non curat lex (The law does not concern itself with trifles), and it means that minor incidents of copying or other purported violations of a copyright owner's rights are not infringing. This is a broader doctrine than "substantial similarity" since it is possible that the copying might be noticeable, but still extremely minor. It is a way to prevent copyright holders from playing "gotcha!" games over insignificant activities and a way to dispose of certain trivial claims without having to step through a fair use analysis.
In one famous case, for instance, the composer of a work sued the Beastie Boys for infringing just three notes of his work. While they did have a licensing deal for the actual sound recording, as is normal practice, they saw no need to cut a separate deal to license the composition the sound recording was a recording of. And they were right not to: the court threw out the composer's claim as de minimis. Suing people over three notes is not standing up for copyright rights; it's a cash grab.
There's one final way that sampling cases can be resolved in favor of defendants without resorting to the fair use test--copyrightability. Only "original works of authorship" are eligible for copyright protection, and samples of some material--single notes, standard drum beats, common chord sequences--may not be eligible for copyright protection to begin with. When combined with other musical elements in a longer work, they might be protected--but copyright is not infinitely divisible, and samples below a certain threshold may simply be uncopyrightable. Their use, therefore, cannot be infringing.
To summarize, it is not, and never has been the case that copyright holders have the right to control all "uses" of their work--not even all "valuable" uses. This is not hard: Copyright holders are given a short list of exclusive rights, all of which are precisely defined, and subject to various limitations. If a use falls outside of the ambit of these rights, it does not infringe. It would be analytically clearer, and better for the development of a body of law that supports creative use and re-use of culture, if more copyright disputes were resolved before they even get to the fair use stage. This is particularly the case when it comes to sampling, where an unfortunate--and legally baseless--culture of "if there's a use, there must be a license" and "if there's a value, it must be paid for" has taken root, to the detriment of cultural freedom and artistic creativity. Some kinds of music are no longer made, or are made in the shadows, with limited opportunities for sample-based musicians to profit from their work. With renewed attention to other, sometimes-overlooked copyright doctrines such as the "substantial similarity" test, maybe sample-based music creation can happen more in the open, and with less paperwork and fewer lawyers.
Original image by flickr user manthatcooks.