|James Gibson, Wednesday, 25 April 2007 [View as PDF]|
Editor's Note: Accidental Rights is adapted from James Gibson, Risk Aversion and Rights Accretion in Intellectual Property Law, 116 Yale L.J. 882 (2007).
Longstanding intellectual property doctrine is inadvertently expanding intellectual property entitlements and eroding public privileges. And, as my recent Yale Law Journal article points out, it’s only going to get worse.
Suppose you’re making a documentary film about the history of satire. You’d like to use a five-second clip from the television show South Park. Do you need permission from the owners of the show’s copyright? The answer depends on the notoriously vague “fair use” doctrine, which means the answer is unclear. So you get permission and pay for a license, just to be safe. Maybe you don’t need to, but why risk a lawsuit?
What happens over time, however, is that as more filmmakers in your position get licenses, the question of whether a license is needed becomes clearer. Courts essentially say, “Of course a license is needed—just look at all these filmmakers who are getting one!” The “license, don’t litigate” decisions combine to expand copyright’s reach and reduce the public’s privileges.
The same thing happens with trademark law. Suppose someone in your film is drinking a can of Pepsi. Do you need to get a license to include that brand in the film? The answer (again) is not entirely clear, and so (again) you would probably decide to get a license, or maybe even strike a product placement deal. Or perhaps you would “blur out” the mark so consumers will not recognize it. As consumers encounter more and more trademarks that are licensed or blurred out, they will come to expect that marks appear only when the mark owner consents. And when they are wrong—when someone rolls the dice and includes an unblurred, unlicensed mark—the result will be consumer confusion over whether the owner consented to the mark’s inclusion. As it turns out, such confusion is a trigger for trademark liability. As with copyright law, then, the aggregate effect of all of these licensing decisions is an expansion of trademark’s reach and a narrowing of the public’s ability to use marks without permission.
The most interesting thing about this “doctrinal feedback” phenomenon is that its expansion of intellectual property rights is inadvertent; it results from prudent licensing decisions and longstanding legal standards rather than from new laws or court rulings. The intellectual property user knows that the cost of going to court to find out whether the law requires a license is higher than the cost of the license itself. And his or her risk-averse insurers and downstream distributors know it too. Better safe than sued. So almost no one litigates, which means there is no opportunity for our formal governance mechanisms to decide what balance between licensing and the public domain best serves the public interest. The result is an expansion of rights almost by accident, with no assurance that the expansion is good for us.
Yet not everyone is so fearful of the courtroom. What happens when an intellectual property user is not risk-averse, and has no insurers or distributors whispering in its ear about the wisdom of playing it safe? Such risk-takers are few and far between, but we can find one prominent example in Google Book Search, a project whose ambitious goal is to make every book in the English language text-searchable, just as Google’s better-known Internet search engine has made the World Wide Web searchable. But to realize this goal, Google must scan the text of every book. And because scanning involves making a copy, a question arises: does Google need a license—or, rather, millions of licenses—from the publishers who own the copyrights?
At this point, it should come as no surprise that the answer is unclear and depends on the famously indeterminate fair use factors. Therefore, if Google had a “license, don’t litigate” attitude, it would either get licenses or abandon the project. But Google is no penniless documentarian, beholden to the risk-averse whims of insurers, investors, and distributors. No, Google is an intellectual property owner’s worst enemy: a risk-taking iconoclast with deep pockets, seemingly unafraid to litigate licensing issues all the way to the Supreme Court.
Perhaps Google Book Search is simply the exception that proves the rule, a circumstance so rare that it can tell us nothing new about doctrinal feedback. If the controversy ends up at the Supreme Court, then great: we will have that exceptional case that results in a formal legal proclamation—that is, an intentional rather than inadvertent shift in the law.
If the issue is resolved without an ultimate judgment on the merits, however, we may see yet another shift in licensing culture that reduces public privilege. Such a result is hardly far-fetched. Given Google’s determination and resources, it may be the publishers who experience risk aversion: they may choose to offer can’t-miss licensing terms to Google rather than face the prospect of an adverse precedent-setting ruling. Moreover, Google itself may decide that its interests are best served by making licensing deals—not because it is afraid of losing in litigation, but because it is one of the few companies with the resources to pay licensing fees to a wide array of publishers. In other words, a world in which copyright law requires licenses for this kind of project may be better for Google, because it means fewer competitors than a world in which anyone can provide a book search service under the fair use privilege. And if my argument about doctrinal feedback is correct, then once Google pays licenses to publishers, any second-comer’s chance of winning a fair use argument for the same conduct diminishes considerably.
This highlights a worrisome aspect of doctrinal feedback: it may initially occur inadvertently, but that doesn’t mean it isn’t subject to manipulation once the players—particularly the rights-holders—realize what’s going on. When copyright and trademark owners understand that more licensing leads to broader rights—at least in the long run—they might stop demanding that users pay exorbitant fees and instead make licensing more feasible and attractive, even to the penurious documentarian. After all, the doctrinal feedback phenomenon teaches us that it is usually in a rights-holder’s long-term interest to agree to license an arguably privileged use, so as to create a market whose existence can later be cited as proof that the use is not in fact privileged.
This may already have happened in certain industries. Copy shops now routinely obtain licenses for university “course packs,” perhaps because publishers used a carrot-and-stick strategy to get a critical mass of shops on board before suing the holdouts. A more recent example involves the question of whether an Internet search engine that enables users to locate images online must pay for the right to display search results in the form of low-resolution thumbnails. The first court to consider the issue said no, the second court said yes. The difference? The second plaintiff had developed a licensing market for low-resolution thumbnails—as background images on cell phone screens—before filing suit.
Unless we like the idea of defining our rights through mere happenstance—or worse yet, through purposeful industry manipulation of the public’s tendency to avoid risk—either Congress or the federal judiciary needs to recognize and remedy doctrinal feedback. Legislation and court rulings may have little to do with the feedback loop, but they represent our best hope of short-circuiting it.
But what exactly is the remedy? Changing the legal standards that fuel the feedback effect creates more problems than it solves, given how ingrained those standards are in established law and practice. In the end, I offer a less intrusive solution: subtle refinements in how the law scrutinizes licensing information (in copyright) and consumer motivation (in trademark). Copyright law needs to look to why a licensing market exists, not just whether a licensing market exists, and it should discount those markets that owe their existence to uncertainty and risk aversion. Trademark law should focus not just on whether consumers are confused by a mark’s appearance, but whether that confusion affects their purchasing decisions. If it does not, then there’s no reason to impose liability—confusion or not.
In any event, simply recognizing that doctrinal feedback exists is an important first step in deciding where private rights end and public privilege begins. Licensing regimes do not spring fully formed from the head of the god Equity, and they do not necessarily involve any conscious policy choice or reflect an optimal outcome. If the law is to be more than a mere scrivener—if policy is to be made, not found—then let’s shed some light on this previously hidden phenomenon.
Jim Gibson is the Director of the Intellectual Property Institute and an Associate Professor at the University of Richmond School of Law. He is usually averse to negative feedback, but is willing to take the risk: firstname.lastname@example.org.
Preferred Citation: James Gibson, Accidental Rights, 116 Yale L.J. Pocket Part 348 (2007), http://yalelawjournal.org/2007/04/25/gibson.html.