Until about a year ago, as the New York Times recently reported, Todd Rutherford had a successful business working with writers to help them market their self-published books on the Internet. Rutherford’s previous career had involved more traditional publicity efforts — talking up his clients’ work in the hope that a reviewer at a newspaper or a blog would take notice. But eventually he realized that it made more sense to “cut out the middleman and write the review himself.” And so GettingBookReviews.com was born, a business that, depending on how much the author was willing to pay, would write one, twenty, or even fifty online reviews singing a book’s praises. “Before he knew it,” the Times reported, “he was taking in $28,000 a month” and had to hire freelancers to keep up with the demand. Rutherford may have been particularly up-front about the nature of his business practices, but he was by no means an outlier; one estimate is that about one-third of online reviews purporting to be by actual consumers are marketing schemes rather than genuine reviews.
Rutherford did not, apparently, assert a proprietary interest in his company’s reviews; in any event, the service later foundered when Google, and then Amazon, took notice. But one might suggest that the inherently creative nature of the reviews – at least one of the freelancers admitted that she hadn’t actually read the books she “reviewed” – would put them squarely at the heart of copyright law’s scope of protectability. If this is the case, does that suggest something troubling about ratings? Or about copyright law more generally?
A recent article by James Grimmelmann seems to pose a simple question: Are ratings copyrightable? But what makes this short piece especially thought-provoking is the way in which Prof. Grimmelmann uses this question as a way of interrogating various fundamental doctrines of copyright law: the idea/expression dichotomy, the originality and creativity requirements, and the nature of fact versus opinion among them.
Prof. Grimmelmann begins by providing an overview of the case law considering the copyrightability of ratings. Because individual ratings are too short to be copyrightable, post-1976 Copyright Act courts analyze ratings systems as potentially copyrightable compilations. In some cases, courts treat ratings as statements of fact, capable of being proved true or false (such as the ranking of a particular CEO as receiving the fifth highest income in the country). In other cases, courts treat ratings as an opinion, entirely the product of their creator (such as the typical restaurant or product review). And in a third category of cases, courts treat ratings as “self-fulfilling prophecies, which remake the world in their own image” (854). Akin to John Searle’s social fact, a “self-fulfilling prophecy” takes on factual status because the relevant community treats it as such. A reporter’s prediction that a new comedy will be the highest grossing film of the upcoming weekend may itself cause audiences to flock to see the film, thus “proving” the reporter correct. The use of such ratings by others cannot be condemned as infringement; rather, it “reflect[s] the plaintiff’s influence” (864) in determining the price, value, or quality of the subject of the rating. Like a trademark that eventually becomes the generic name for a good, a self-fulfilling prophecy may have started life as the product of creative effort but eventually becomes a basic tool of communication for the public.
As Prof. Grimmelmann explains, however, these categories are not particularly rigid, as various ratings and rankings incorporate more than one of these features. The U.S. News & World Report rankings of U.S. law schools are based in part on data such as acceptance rates and LSAT scores (which are only as accurate as the information the schools themselves supply) but also on more subjective factors such as the reputation scores accorded by a sample of law school faculty and administrators. A restaurant critic’s award of one to four stars to a local restaurant may be seen by readers as subjective opinion, but those readers no doubt assume that the rating is based on the reviewer’s own experience at the restaurant and not on a grudge she holds against the manager. And, of course, even the most fact-driven rating derives from choices about what types of data to include in the analysis, as Malcolm Gladwell has demonstrated, and some courts that have found ratings to be copyrightable typically focus on such choices as providing the necessary modicum of creativity.
This kind of evaluation matters because copyrightability depends, in part, on whether the work at issue represents facts (which should be available to all) or creative expression. The problem with focusing on the process used to produce the rankings as the source of creativity, as Prof. Grimmelmann explains, is that copyright law is ultimately concerned with works of authorship. As he perceptively notes, a photographer who takes a photo with the lens cap on may have engaged in a variety of creative decisions up to that point regarding lighting, framing, positioning, and so forth, but her error means that she has failed to create a copyrightable work. And to the extent that copyright law is about economic or other incentives (a conventional wisdom that has been increasingly challenged of late), Prof. Grimmelmann notes that the idea/expression distinction may get things exactly backward. If the creators of ratings are told that they can protect their work only if they are opinions, then the ratings will become increasingly less fact-bound and, presumably, less useful to users.
But is utility copyright’s proper focus? If copyright law should set aside notions of aesthetics, as Bleistein suggests (however impossible or inadvisable the task), why should it care about the Todd Rutherfords of the world? Here, Prof. Grimmelmann concludes with a gentle encouragement to reconsider first principles. Perhaps we should evaluate not only whether ratings (or any such effort) are copyrightable but also what is gained or lost by such a decision. Perhaps copyright law would benefit from more direct consideration of the social utility of various works, taking its cues from doctrines (such as defamation law) that more directly address the value of various forms of speech. Or perhaps (to invoke one of my own hobbyhorses) copyrightability is not the driving force here at all; validating reputational or attributional concerns may provide creators with all the incentive they need while increasing the availability of the underlying work.
Prof. Grimmelmann does not presume to answer these questions in the space of his short article. But his recognition that a deceptively simple question —are ratings copyrightable? — can give rise to many further, and interrelated, avenues of inquiry makes the article well worth reading.