Well, this is interesting. Ordinarily, Public Knowledge is an organization that sows a lot of confusion—and sometimes outright falsehoods—about copyright law. As a rule, I group them among the “digital rights” activists who tend to promote their opinion of what the fair use doctrine should be rather than a more realistic description of what it is. So, it’s interesting that in response to allegations of copyright infringement against people they don’t like, the organization has produced a fairly sober and reasonable analysis of the fair use defense. For the record, I’m no fan of the video in question either.
After FCC Chairman Ajit Pai reversed the 2015 Open Internet Order, a video featuring Pai, apparently produced by the Daily Caller, went viral. It shows Pai demonstrating various things people will still be able to do after the hugely unpopular reversal of what is generally called “net neutrality.” As readers know, I’ve tried to cut through some of the rhetoric on the neutrality issue without fully defending Pai, or certainly the telcos; but I also think the video was a pretty dumb PR move. Its content may be factual, but its tone and style were begging for ridicule; its association with the odious Daily Caller makes it fatally dismissible; and its use of about ten seconds of the song “Harlem Shake” drew a DMCA takedown and threat of litigation by artist DJ Baauer.
Faced with the dichotomy of hating both Pai’s policy and copyright law, Public Knowledge uncharacteristically published a fairly rational explanation, written by Meredith Filak Rose, of the fair use doctrine along with an analysis of the use of “Harlem Shake” in this video. Her assessment predicts that the video makes a fair use of the song, though not without sprinkling a bit of anti-copyright messaging along the way.
Rose does a good job of explaining the legal meaning of “parody,” which is one of the more widely misunderstood, colloquial terms invoked to assert a fair use. I agree with her that the use of “Harlem Shake” in the video does not meet the standard of parody and am glad to see even an anti-copyright organization make an effort to clarify this point. But from there, I think her fair use analysis is a bit all over the place, straining to make a case for the kind of use Pai and colleagues made, while clearly hating the video itself.
I’ll dispense with the fairly simple analysis under the second fair use factor, which generally considers whether the original work is expressive or factual in nature. Use of the former tends to weigh against a finding of fair use, while use of the latter tends to weigh toward a finding of fair use. Clearly “Harlem Shake” is expressive, and so Rose is correct to expect that this disfavors a finding of fair use under this prong of the four-factor test.
Under the first factor analysis, I agree with Rose that the use of “Harlem Shake” is not parodic and that it is also non-commercial; but I disagree when she concludes that it is “at least mildly transformative,” thus favoring a finding of fair use. In fact, it is contradictory to conclude that the use of the song is not parody but is transformative. Because this still begs the question as to what has been transformed.
Originally, transformativeness described the creation of a truly new expression, which could not possibly exist without making use of the exact work in dispute. In the video, Pai concludes his list of “post neutrality” activities by saying “You can still do the ‘Harlem Shake’,” and then he dances to the track (and yeah, it’s cringy) with staff members of the Daily Caller.
This use is no more transformative than if the track were synched to the introduction of a promo video made by Pfizer. The use does not produce a new work that directly parodies, comments upon, criticizes, or builds upon the original song itself. In fact, in a subtle way, Pai is telling people, “Don’t worry, you can still randomly infringe various copyrighted works after I reverse the 2015 Order.” (By the way, transformativeness is a principle that has indeed been applied rather schizophrenically in the courts, but it is organizations like Public Knowledge that continue to promote ever-broadening interpretations of the term.)
The intended communication in the Pai video could have been made by using any number of pop songs. The purpose of the video as a whole was not a work of commentary upon art and culture but was instead a work of PR with the intent to promote a position on public policy. In a fair use assessment, this use should be seen as equivalent to precedent examples in which a creative work is used for general comment or satire (i.e. to comment upon something external to the original work), which are not typically held to be fair uses.
In this context, Rose errs when she considers whether or not it was necessary to use “Harlem Shake” per se in this video. First she improperly places the question under the third factor—the “amount and substantiality of the use”— rather than the first. Because the need (or not) to use a specific work underlies the purpose and character of the use, the question belongs under the first factor analysis. Next, Rose rather bollixes up the principle in her effort to defend “meme culture” by confusing creative necessity with the much broader notion of personal taste. She writes…
“But judges can also ask whether you could have achieved the same purpose without using the work at all, or using another work. These are borderline existential questions when it comes to memes: did I have to use a copyrighted image of a screaming porg in my Twitter avatar to convey my distaste for the current political climate, or could I have used something else to equal effect?”
Despite the anti-copyright crowd’s love of memes, the test under fair use as to whether a specific work is necessary to create an expression is not so broad as considering the whimsical taste of the user to make some generalized statement. Under fair use, the resulting new work should, in some way, be unimaginable without making use of the protected work in dispute. For instance, the seminal case (1994) in which 2 Live Crew parodied the song “Oh, Pretty Woman,” was one in which the new expression relies substantially on a cultural familiarity with the sounds copied from the original. It is understood that the new work is, in part, spoofing the spirit of the original song.
Rose’s reference to her screaming porg avatar errs as an example because she is in no way commenting upon the original photograph. She’s simply wearing it like a button on a lapel to express her personal state of mind about the current political climate. If she used the same image in a blog post about the political climate, the photographer could take action, and she would almost certainly lose in a fair use defense.
Finally, while Rose’s explanation of the fourth factor is correct, part of her application with regard to the Pai video is odd. The fourth factor considers the effect of the use on the potential market for the original work. I certainly agree with her that nobody would ever consider the Pai video to be a substitute for listening to “Harlem Shake,” and I also agree that Pai’s use under this factor favors a finding of fair use. But then, Rose meanders into other legal territory—namely, moral rights—when she writes, “There is a tenuous (but extant) argument that Pai’s association with the song actually devalues the song.”
Regardless of one’s opinion about moral rights for artists—Rose calls it dangerous—this consideration is not generally applied in a fair use analysis. In fact it is essential to note that fair use will protect commentary upon a work, even if the commentary itself may harm the market value of the original work. Correctly applied, the fourth factor only considers whether the new use may act as a substitute for the existing or potential market of the protected work, including a market that the original author may choose to exploit at some point in the future. So, whether one buys into the argument that Pai’s use of “Harlem Shake” could create a negative association with the song, this is not legitimately a question under the fourth fair use factor.
On instinct, my prediction would be that that the Pai video would fail on the first and second factor analyses, prevail on the fourth factor, and could go either way on the third factor. Although the video makes very limited use of “Harlem Shake,” the third factor considers whether the portion used represents the “heart of the work.” One could imagine DJ Baauer making a case that the campy communication intended can only have worked by making use of the “heart” of the song such that it would be instantly recognizable in context. This would disfavor a finding of fair use, if the use also failed under the first factor analysis.
One way or another, I don’t doubt the infringement claim associated with this particular video caused a bit of dissonance over at Public Knowledge. And they should be commended for at least trying to discuss fair use as a nuanced principle, rather than the natural right of all netizens. That said, I think they missed a few key distinctions in an effort to make clear they still really don’t like copyright.
