Fair Use and Digital Rights Management: Preliminary Thoughts on the (Irreconcilable?) Tension between Them
Fred von Lohmann
Senior Intellectual Property Attorney
Electronic Frontier Foundation
Digital rights management (DRM) technologiesare aimed at increasing the kinds and/or scope of control that rights-holderscan assert over their intellectual property assets. In the wake of the DigitalMillennium Copyright Act's (DMCA) ban on the circumvention of DRM technologiesused to control copyrightable works, DRM restrictions are now backed up withthe force of law. In essence, copyright owners now have the ability to writetheir own intellectual property regime in computer code, secure in theknowledge that the DMCA will back the regime with the force of law.
It's not surprising that in light of thesedevelopments many have expressed alarm that DRM technologies may be used bycopyright owners to erode capabilities that had previously been permitted tothe public by copyright law under the "fair use" doctrine (or its cousins, suchas first sale or limited term).
What is the nature of the tension between DRM(as backed by the DMCA) and fair use? Is the tension irreconcilable? If so,which should give way?
I.Fair Use: What's It Good For?
While the "four factor test" set out in Section107 of the Copyright Actis widely known (at least in legal circles), the role of fair use in thecopyright system is somewhat less settled. Attention to three aspects of fairuse "in the real world," however, are worth considering in some detail here:
(1) a wide variety of activities depend uponfair use for legitimacy;
(2) the ambiguous, case-by-case nature of thedoctrine is central to its role in mediating between new technologies andcopyright; and
(3) fair use has served as a catalyst for innovation and a buttress forcompetition.
A.Activities that Rely on Fair Use
In any discussion of fair use, it is importantto keep in mind the full range of activities that depend upon fair use forlegitimacy. Often legal commentators are tempted to restrict their discussionof fair use to the uses that have been tested in published court decisions.This mistake can lead to a crabbed view of the doctrine.
First, consider the reach of copyright law inthe absence of fair use. The Copyright Act, by its terms, grants to ownersdominion over the reproduction, public performance and display (a term of artthat includes transmissions to the public), and distribution of a work, as wellas a monopoly on the creation of derivative works.
Given the broad scope of the Copyright Act,copyright would intrude into everyday life in innumerable ways were it not forthe fair use doctrine and other exceptions. The First Amendment, privacypriorities, and common practice would not be able to tolerate a copyright lawlacking a fair use "safety valve."
Fair use serves a crucial role in limiting thereach of what would otherwise be an intolerably expansive grant of rights tocopyright owners. Were it not for the fair use doctrine, each of the followingactivities would be infringing:
- whistling a tune while walking down the street (publicperformance)
- cutting out a New Yorker cartoon and posting it on your office door (public display)
- photocopying a newspaper article for your files(reproduction)
- quoting a line from The Simpsons in an email to a coworker (reproduction)
- reverse engineering of computer code (reproduction)
- "time-shifting" a radio or television program(reproduction)
- playing an excerpt of Roy Orbison's "Pretty Woman" in acopyright law course (public performance)
- quoting from a novel in a review (reproduction)
If they are to preserve fair use in itstraditional form, DRM technologies must leave room for these unauthorized usesof copyrighted works, as well as myriad other commonplace uses that have not beentested in court.
B.Ambiguity as a Feature, not a Bug
Many DRM vendors express frustration with theimprecision surrounding fair use. Often they exclaim, with some exasperation,"Define fair use for me, and we'll express it in the Ôbusiness rules.'"
Unfortunately, fair use cannot be defined withprecision. The statutory four factor test, for example, includes concepts like"effect on the market" and "commercial/noncommercial" — concepts that are noteasily expressed in precise "business rules." The problem is especially vexingonce you recognize that "fair use" has changed over time, and that it is vitalthat it continue to be able to evolve.
The fair use doctrine operates as a "safetyvalve" not just for free expression, but also to mediate the tension betweencopyright and new technologies. As new technologies develop, courts generallyhave the first opportunity to apply copyright law to them, with Congresslagging behind. This spares the public, technologists, and copyright ownersfrom having to apply to Congress for a legislative solution for each newtechnology that is developed. In this way, the fair use doctrine plays animportant role in preserving a space for innovation and consumerexperimentation, while leaving the final word to Congress.
Under this regime, the normal evolution of fairuse proceeds thus: a technologist or other creative person makes some use ofanother's work that she believes to be fair. If the rights-holder agrees, the use continues. If therights-holder disagrees, she can call upon the courts intervene and rule on theuse. If, however, the use is frustrated from the outset by a DRM measure (thecircumvention of which is unlawful under the DMCA), there is no opportunity forthis process to unfold. Fair use is limited to those uses that the courts havepreviously affirmed, and new uses cannot evolve.
When the problem is viewed in this light, itbecomes clear that the ambiguity of the fair use doctrine is not a bug, but acrucial feature. If DRM systems are to preserve fair use, they must somehowpreserve its ambiguity, its ability to evolve and embrace as yet unrealizeduses of copyrighted works. A consideration of technologies past, present andfuture, and their collisions with the fair use doctrine, illustrates thevirtues of ambiguity in fair use.
The Past: the VCR
In 1976, Universal City Studios and the WaltDisney Company sued Sony, seeking to have the Betamax VCR impounded as a toolof piracy. In their view, there were virtually no noninfringing uses of theVCR, since home taping of television was thought to violate the copyrightowner's reproduction right. The Supreme Court in 1984 disagreed, ruling thathome taping of television programs for later viewing ("time-shifting")constituted a fair use.
Two aspects of the Betamax case are importantfor our purposes. First, most copyright scholars at the time felt that hometaping should not constitute fair use.In particular, it was unprecedented for a court to find a use to be fair where(1) the copyist reproduced the entirety of a work and (2) did so for a purelyconsumptive, nontransformative purpose. So, had you considered the shape of thefair use doctrine in 1976, you would probably have concluded that time-shiftingwas not a fair use. The Supreme Court in 1984 evolved the doctrine in response to the new possibilities created by theVCR.
Second, if copyright owners had been able toimpose DRM technology on the VCR in 1976, the Supreme Court would have beendenied the opportunity to evolve fair use. While the state of DRM technologywas more primitive in 1976, there were already efforts by motion picturestudios to control video content. For example, during the Betamax case, themotion picture studios argued to the court that Sony should build a sensor intoevery VCR that would detect "no copy" signals that would be embedded intotelevision broadcasts, thereby enabling copyright owners to mark their moviesas "not for copying." Had this DRM solution been adopted, the Supreme Courtwould have been denied the opportunity to address time-shifting, leaving thisactivity under the exclusive control of copyright owners.
The Present: MP3 and Ditto.com
More recent examples reinforce the importanceof permitting fair use to evolve in response to new technologies. Thedevelopment of PC technology and the MP3 music format has given music fans awide variety of new capabilities. As a result, electronics manufacturers havebegun offering new categories of music products, including MP3 jukeboxes, homemusic servers, and Internet radio receivers. Each of these technologies,however, depends upon copying lawfully-obtained digital music, an activitythat, unless considered fair use (or otherwise privileged under the law), wouldinfringe copyright.
The Internet has similarly spawned innovativeservices that have called on courts to evolve fair use jurisprudence. Forexample, copyright litigation recently erupted over Ditto.com, a search enginethat catalogs photographs and other images that can be found on the Internet.Prior court decisions regarding fair use provided only ambiguous guidance. TheNinth Circuit Court of Appeals recently concluded that the reproduction ofimages involved in this activity should be considered a fair use.
Had DRM technologies blocked these unauthorizedexperiments from arising in the first place, the opportunity to further developthe fair use doctrine would never have presented themselves. The public wouldnever have had the benefit of these new technological capabilities, or wouldonly have received them if copyright owners could be persuaded to authorizethem.
3.The Future: DTV
With these earlier examples in mind, we canbetter understand the threat that DRM technologies pose to the future of fairuse. If fair use is to continue to evolve, to create space for free expression,innovation and new uses of copyrighted works, DRM technologies must somehowaccommodate the ambiguity of fair use. Unless the public has the opportunity toexperiment with new technologies, courts will not have the opportunity to testthem against the fair use doctrine. If innovators and consumers arepresumptively barred from experimenting without copyright owner authorization,fair use will become increasingly irrelevant. After all, how useful is a rightto time shift analog television in a world where all broadcasts are digital andprotected by DRM technologies?
This is not a hypothetical question. Aninter-industry group called the Broadcast Protection Discussion Group (BPDG) iscurrently drafting standards for mandatory DRM systems for digital broadcasttelevision (DTV). While these systems may attempt to preserve time-shifting asdefined by the Supreme Court in the 1984 Betamax decision, they will fail toprotect the full range of future fairuses that will be made possible by DTV. What those fair uses might be isdifficult to imagine in advance, just as time-shifting was difficult to imaginein the era before VCRs. Nevertheless, unless DRM technologies make room forthese future fair uses, fair use will have lost much of its ability to protectthe public's side of the copyright bargain.
C.Fair Use, Innovation and Competition
Fair use has repeatedly been invoked to preventcopyright owners from misusing their copyrights in order to stifle legitimatemarketplace competition. For example, courts have concluded that intermediatereproductions of software made in the course of reverse engineering can qualifyas fair uses. The fair use doctrine thus operates to limit copyright in orderto preserve competition.
The fair use doctrine also plays an importantrole by providing a reservoir of incentives to spur innovation. For example,where the public is permitted to use copyrighted works freely, a powerfulincentive arises to develop technologies and services that help the public getthe most from media content.
This incentive explains the rapid innovationsurrounding the VCR. To the extent time-shifting television was considered afair use, electronics companies saw a market for devices that would allowconsumers to get the most from time-shifting. The same incentive continues todrive the new "PVR" industry, led by TiVo and ReplayTV.
Similarly, the MP3 industry is fueled by thereservoir of incentives that arises from another activity considered to be afair use: "space-shifting." Millions of Americans have a large music library ofCDs. Technology companies have an incentive to develop devices that will helpthe music lover to get more value from the CDs she has already purchased. TheApple iPod, for example, emerged to meet that demand.
In other words, wherever an activity has beendeemed a fair use (and often even before, so long as a company is willing togamble that it will be deemed a fair use), innovation flowers as technologycompanies help the public to make the most of copyrighted works. Examplesinclude the photocopier, the audio cassette deck, the CD-RW drive, the webbrowser, among others.
Of course, this is not the only road toinnovation — innovation can also flow from activities that depend onauthorization from copyright owners (e.g., DVD). But innovation generallyproceeds most rapidly in an environment where copyright owners do not have aveto over new technologies, and where free competition between technologists isthe rule.
Some may object that this view unfairly forcescopyright owners to subsidize innovation. After all, part of the value of a VCRis rooted in the fact that it can be used to make copies of copyrighted workswithout the payment of a royalty. To put it another way, consumers would payless for a VCR if it could not record copyrighted works. While that additionalincremental value certainly operates as an incentive for innovation, why shouldthe VCR manufacturer capture it, rather than the copyright owners?
This argument neglects two importantconsequences that flow from innovation in a competitive marketplace. First,copyrighted works and the new technologies that interact with them are complementary products. For example, while the VCR may haveencouraged some piracy in the short run, in the long run it vastly increasedthe value of "back catalog" films by creating a new market for them. So, ifcopyright owners are entitled to demand the portion of the VCR's value that isrooted in its ability to copy movies, then VCR manufacturers should be entitledto that portion of the value of Warner Bros. film vault that is rooted in videosales and rentals.
Second, it is worth remembering that, generallyspeaking, the "fair use portion" of a VCR's value does not end up in thepockets of the VCR maker. For example, JVC will not be able to increase theprice of its VCR to capture the value that the recording function offers toconsumers, because Sony (or any other VCR maker) would then be able to reducethe price of competing VCRs. To put it in economic terms, in most mass-markettechnology markets, price is relentlessly driven to marginal cost. Because thefair use value is not part of the marginal cost of production, technologyvendors are unable to capture it in a competitive marketplace.
So who pockets the marginal value derived fromcopyrighted works? The public, in the form of lower technology prices andenhanced capabilities. Since copyright exists for the benefit of the public,this appears to be an appropriate result.
II.The Copyright Bargain Restruck: Trading Fair Use for DRM
In light of the purposes of fair use discussedabove, it seems unlikely that any DRM technology (at least one that will beembraced by the copyright industries for their products) will be able toaccommodate the full range of fair use.
If this is true, the question then becomeswhether some quantum of fair use shouldbe sacrificed in order to stem the disruption caused by new technologies toexisting media company business models. Or, to put it more bluntly, should thepublic be required to give up some measure of fair use in order to solve the"piracy" problem?
A.What's in it for the Public? Quelling "Pirates."
In evaluating this question — whether the publicshould give up some quantum of fair use in order to quell digitalinfringement — it may be useful to remind ourselves of the purpose of copyrightlaw.
Copyright law represents a bargain between thepublic and copyright owners. The public grants certain limited exclusive rightsto copyright owners in order to create an incentive for the production, and amarketplace for the distribution, of creative works.
In order to evaluate whether the law is servingthis end, then, it may be helpful to imagine yourself as an advisor to thepublic. In Jessica Litman's words, imagine you are a lawyer, attempting to getfor the public the best possible copyright bargain.
So what do DRM technologies (and the legalmandates that support them, like the DMCA) offer the public? Of course, DRMtechnologies are often touted as reducing copyright infringement (sometimesequated pejoratively with "piracy"). The public, however, has no directinterest in reducing infringement for its own sake. Why should the public beconcerned about infringement?
Rampant piracy, so the argument goes,undermines the incentives that make it worthwhile to create new content. If,because of infringement, copyright owners cannot make a return on theirinvestments in content creation, they will stop investing in new content (orthe continued exploitation of old content, for that matter). Taken to theextreme, rampant infringement will result in the collapse of the music, movieand publishing industries, say copyright owners.
If you were the public's lawyer, you wouldlikely treat this argument with some skepticism. While unlawful copying is (andalways has been) a problem, no one is proposing that copyright law beeliminated. Accordingly, copyright-based incentives for continued production ofcreative works will remain. Nor is there any compelling empirical evidence that"piracy" is bringing copyright industries to their knees today. Band are stillbeing signed, movies are still being "green-lighted," and writers are stillreceiving advances. In fact, 2002 promises to be for major movie studios thebest U.S. summer box office season in history. Of course, the digital futureposes serious challenges, but many of those challenges have not yetmaterialized in any economically substantial way. As the public's lawyer, youwill be hesitant to give something up in the face of these claims, especiallybecause you will remember the many other occasions where the copyrightindustries "cried wolf" — the VCR, cable television, the audio cassette, radio,the player piano.
Also, as the public's lawyer, you will rememberthat incentives in the form of an airtight monopoly may not be as necessary ascopyright owners would have you believe. In the1995 National InformationInfrastructure proceedings, for example, copyright owners claimed that the"information superhighway" would remain barren until they could be assured ofstrong copyright protection on the Internet. As it happened, the Internetbecame a vibrant place for new publishers — both noncommercial andcommercial — without the assistance of Hollywood. In fact, when media companiesfinally joined the party (not because of increased copyright protection, butrather out of commercial necessity), some found that their content simply couldnot compete with the new kinds of media being created on the web.
A tough-minded lawyer for the public will alsorecall that creative content can proliferate even in the face of widespreadinfringement. For example, the software industry has long tolerated and thrivedin a marketplace with much higher piracy rates than those faced by the musicand movie industries. The film industry in India, the world's most prolific,manages to attract continued investment notwithstanding extremely high rates oftrue commercial piracy. So, as the public's lawyer, you will be skeptical ofclaims that content will not be made in the absence of ever-stronger copyrightprotections.
B.What's in it for the Public? More Choices & Lower Prices.
Some economically-minded DRM advocates claimthat widespread adoption of DRM technologies will result in more products andlower prices for consumers. DRM systems, the argument goes, will permit copyrightowners to engage in more refined forms of price discrimination, resulting inmore products and lower prices. For example, a copyright owner who sells a DVDfor $30 today, may be willing to sell a "single-viewing" version for $5. Untilthe DRM technology exists to support such a "single-viewing" version, however,the public can only obtain the $30 DVD. In the absence of DRM, we all payhigher prices for products that include features we may not need (e.g.,unlimited repeat viewings).
While this argument seems plausible, would you,as the public's lawyer, be persuaded? Some skepticism is certainly in order — sofar, DRM systems that enable price discrimination have had only limited successin the marketplace. The DIVX format for DVDs, for example, proved unpopular.The MusicNet and pressplay music services have also been less than successfulin the marketplace. In addition, the use of DRM technologies may not benecessary to induce reduced prices. For example, DVD prices have been fallingrapidly, driven by Hollywood's desire to compete with video rental.
In addition, there is reason to questionwhether truly competitive price discrimination is likely to occur in themarketplace for copyrighted works. Many copyright industries show a high degreeof concentration, giving rise to market distortions that may swamp thedesirable aspects of price discrimination.
What Does the Public Stand to Lose?
As discussed above, DRM technologies backed bylaws like the DMCA threaten to undermine fair use in a number of ways. As thepublic's lawyer seeking to get the best deal for your client, you will need toweigh these losses against the gains detailed by DRM advocates.
An erosion of fair use in favor of DRM comeswith the following potential costs:
- a reduction in freedom of expression, to the extent DRMinterferes with review, commentary, scholarship, and parody;
- a reduction in innovation, to the extent that DRMeliminates the reservoir of incentives that spur companies to developtechnologies that interact with copyrighted works;
- a reduction in innovation, to the extent that DRMdepends on legislative mandates (whether in the form of the DMCA, a mandatefrom the Broadcast Protection Discussion Group or the pending Hollings bill)that interfere with science and technology development;
- an erosion of privacy, to the extent that DRMcompromises user anonymity;
- the "freezing" of fair use, to the extent that DRMsystems will prevent courts from evolving fair use in response to newtechnologies;
- undermining archives, libraries, and others who storeand preserve our cultural heritage, to the extent DRM systems prevent freearchiving of copyrighted content;
- lessened competition, to the extent that DRM systemsprevent companies from engaging in legitimate reverse engineering of competitors'products.
D.Is DRM the Best Dealfor the Public?
As the public's lawyer, your goal is to get thebest possible deal for the public. Do DRM technologies, backed governmentmandates, represent a good bargain for the public?
While it may be too early to draw finalconclusions, it is plain that DRM technologies, backed by laws like the DMCA,pose a serious potential threat to fair use. While technical refinements mayaddress or minimize some of the social costs that stem from an erosion of fairuse, it is unlikely that they will entirely resolve the tension.
Accordingly, the problem becomes one, not ofreconciliation, but of trade-offs. A hard-nosed negotiator for the public willask for concrete empirical evidence to support the promised benefits of DRMtechnologies, while demanding limiting principles to protect as much of thepublic's side of the bargain as possible.
Unfortunately, to date, Congress has failed totake up this challenge, leaving the public without a zealous advocate in thecopyright arena. Instead, they have delegated copyright law to copyright ownersand DRM vendors. Whether this will result in a good copyright bargain for thepublic remains to be seen.
- thepurpose and character of the use, including whether it is commercial ornoncommercial;
- thenature of the work (e.g., factual works are entitled to less protection thanpurely creative works);
- theamount and substantiality of the portion of the work used;
- theeffect of the use upon the potential market for the work.
See 17 U.S.C. 107.