“It is as if some titanic aberrant genius, uncompromisingly inimical to man, had devoted all the ingenuity of Hell to the making of them. They show grotesqueries of ugliness that, in retrospect, become almost diabolical.” – H.L. Mencken, Libido for the Ugly (1927)
A paper published in August by Kal Raustiala of UCLA Law and Christopher Jon Sprigman of NYU Law proposes a new, generalized rationale for limiting copyright protections: “data-driven authorship.” Titled The Second Digital Disruption: Data, Algorithms and Authorship in the 21st Century, the central thesis hinges on the assumption that because authors of creative works will soon able to use Big Data to predict a greater likelihood of market success for a given work, this reduces the investment risk in producing that work, which in turn recommends limiting copyright protections. From the paper …
“The next digital disruption is going to reach deeper. It will re-order how creative work is produced, and not simply how it is promoted and sold. It will transform our notions of authorship. It will raise fundamental questions about the nature and value of human creativity. And, perhaps less consequentially for the world at large — but of central importance to lawyers — it may shift how we think about the value and utility of, and even the moral justification for, intellectual property rules.”
Raustiala and Sprigman see the inevitable adoption of “data-driven authorship” as a predicate for a philosophical shift in the way society perceives the author and that this perceptual change ought to yield a revision of the underlying moral rationale for copyright. Rather than viewing the author as “Promethean” (i.e. as an individual bestowing her works upon society), the paper proposes that “data-driven authorship” proposes a concept of the author as “Panoptian” (i.e. as a “watcher” who synthesizes creative expression in a collaborative process with her audience whose proclivities for various types of content can be understood and even predicted through data).*
Betraying the Purposes of Creative Expression
If I had to pick just one flaw in this paper (and it’s hard to choose one for a fairly short rebuttal), it would be the naive bias, typical among copyright skeptics, that misrepresents the relationship between creators and their audiences by presuming that the goal of creative expression (art) is necessarily to give people what they want. While it is certainly true that most creators hope for market success and that big-money investors in works like blockbuster movies have always endeavored to predict, if possible, the likelihood of producing a hit, this business narrative does not fully represent the creative impulses, talents, or labors of individual authors, who must remain the central figures of copyright law, if copyright law is to mean anything at all.
If every creator were overly motivated to second-guess what his audience wants, then both creators and audiences would be left with the homogenous output that would eventually result from such a tedious feedback loop. All works of note (i.e. the works we love) in every medium have generally earned their place in the anthology of “greatness” by virtue of being distinctive, experimental, or ground-breaking—works produced by artists who did not ask what we expected of them, but instead gave us works we did not know we wanted until we had them. And this does not even account for art that meaningfully provokes because it is controversial, unconventional, or uncomfortable.
It is particularly odd that the authors of this paper so eagerly anticipate the prospect of creators seeking merely to satisfy audience taste, when it is precisely that endeavor, taken to an extreme, which produces safe, formulaic, repetitive fare that many viewers cite as pabulum that only deserves to be accessed via piracy. In other words, the authors’ predicate for proposing a shift in copyright theory is to embrace more corporatization of creative works; and this directly contradicts their fellow copyright skeptics, who claim that a central flaw with copyright’s status quo is that it overly favors corporate rightholders at the expense of individual authors.
As is often seems true of these papers, the authors present “data-driven authorship” as yet another rationale to address a problem that nobody has actually proven to be a problem—namely the scope of copyright protection as it stands. In concert with many in academia, Raustiala and Sprigman proceed from a commonly-held (though not universally accepted) assumption that copyright can only be seen as a necessary evil—a devil’s bargain society makes with authors to extract work from them. While no one will deny that a key rationale for copyright is incentive, I maintain that the view espoused by this and other papers—stated as though it were axiomatic—is ahistorical, cynical, and a needlessly mercenary way to view the relationship between creators and their audiences.
To support their thesis, Raustiala and Sprigman devote considerable attention to—indeed almost fawn over—the apparent adaptation to online piracy by the “porn industry,” asserting this narrative as one that ought to be instructive to mainstream (non-porn) authors of works. They write …
“We use the adult entertainment industry as our primary lens to examine the second digital disruption, but as we will show, the techniques so effectively deployed there are increasingly apparent in the music, film, and television industries as well.”
To be sure, anybody who can cite the innovation called “dildonics” and the 1841 copyright skepticism of Thomas Macaulay in the same paper deserves an AttaBoy; but just as so many modern skeptics tend to ignore the broader context of Macaulay’s oft-quoted speech,** the authors of this paper gloss over significant differences between pornography and just about every other form of copyrightable expression. They further commit the fallacy of referring to the successful adaptation of the “porn industry” while minimizing the fact that said adaptation came at the cost of consolidation whereby one Big Data firm called MindGeek now controls nearly the entire online market. So, if there really is a lesson non-porn creators can take from porn creators, it might be Don’t Let This Happen to You.
The Porn Narrative is a Cautionary Tale to Creative Authors
Raustiala and Sprigman explain that pornography adapted to piracy—rather than attempt to litigate or effect policy changes—in part because of MindGeek’s ability to leverage massive amounts of data about viewer behaviors. The paper then compares the MindGeek model to the ways in which Netflix currently uses its data about viewer habits and then implies that future non-porn motion pictures, episodic series, and all other media can learn the art of “data-driven authorship” from their porn-making cousins.
Of course, the MindGeek narrative in this paper simply expands on a familiar theme in which access to the content itself is understood to be a loss leader that can attract tens of millions of viewers, a percentage of which can be relied upon to spend money in other ways. For instance, while most visitors to porn sites will only ever view a handful of clips at a time for free, a valuable number of visitors will be amenable to spending money on things like “Camming” (private web-cam interactions), “Customs” (custom-made pornographic scenes), or sex-related merchandise (you get the idea).
Although the paper’s authors acknowledge that pornography is fundamentally “utilitarian,” they do not seriously consider how this fact alone generally disqualifies the industry from having much to teach authors of non-pornographic works. Without wading into a discussion about the line separating erotic art from pornography, suffice to say, most porn—and particularly the porn to which the paper refers—has nothing in common with most creative expression, either for the creators or the audiences. The paper states …
“MindGeek is at the forefront of this new approach. The fine-grained data that MindGeek collects from its billions of views each month puts it in a position to analyze and then leverage consumers’ revealed preferences regarding a wide range of adult content. We already see MindGeek using what it learns about viewing patterns to better categorize and organize content. Most strikingly, MindGeek increasingly uses its data to shape the content that its production companies create.”
Sprigman and Raustiala may find these observations “striking,” but it is hardly more than obvious to understand that MindGeek is able to learn through data collection that particular tastes in porn may be trending and that these data can then inform even the producers of scenes to make quick, responsive decisions about camera angles, casting, wardrobe, scenarios, etc. in order capitalize on those trends. It is equally unremarkable to recognize that relatively short, pornographic clips can be served up like free snacks in order to entice some viewers to spend money in ways ancillary to the pornographic videos themselves. Sex sells. Nobody needs academic research to support that premise.
But the MindGeek “lesson” is useless to many a non-porn filmmaker, who asks 90 minutes of our attention for a story he wants to tell in his own distinctive way. His motives for producing (and ours for watching) are exactly the opposite of the MindGeek model. As a creator, he is not interested in satisfying audience taste—let alone reenacting some fleeting, primordial fantasy. Instead, it is essential that, to a great extent, the author doesn’t give a damn what the audience wants to see. And it is only when a creator surprises us the most, that we bestow upon him the compliment artist.
On the other hand, as Raustiala and Sprigman correctly observe, major producers of non-pornographic works are not only leveraging Big Data in their business models, but they cite testimony from TimeWarner’s proposed merger with AT&T, which asserts that without access to data, the content producer will be at a disadvantage relative to companies like Netflix and Amazon. The paper’s authors are not wrong to describe the use of data in the market, but I suspect they may be over-emphasizing the usefulness of that data and, more importantly, jumping the gun by presuming that the power of said data to inoculate against risk is a sound rationale for limiting the role of copyright.
In Part II of this post, I’ll respond to this proposed revision of copyright as a rational byproduct of the presumed value of “data-driven authorship.”
* “Panoptian” comes from the Greek myth of Argus of Panoptes. See Part II for further comment.
** For instance, Macaulay’s speech was in the context of a five-year debate that he ultimately lost; his dire predictions about copyright terms did not come to pass; and he was advocating the views of the publishing industry over those of authors.
Photo by kentoh
