Cross-posted on the Law Theories blog.

With Aereo’s upcoming oral argument before the Supreme Court, I’ve got tens of thousands of tiny antennae on the mind. The novelty—and, of course, the absurdity—of Aereo’s service makes it one of the more interesting copyright puzzles. Most of the issues with Aereo are repetitive of the issues with Cablevision, and debate abounds over whether the number of source copies matters and whether multiple transmissions should be aggregated. I’ve come to view these issues as red herrings. But there’s another red herring that I’d rather discuss here, and that’s whether a decision against Aereo would be the death knell to the cloud. After all, performances are transmitted from Aereo’s service to members of the public just like they are from some cloud computing services. So why shouldn’t we worry?

In my last post about Aereo, I unconcernedly claimed: “This case isn’t about the future of cloud computing companies—those services are protected by the DMCA.” Why we shouldn’t worry about the cloud is certainly in large part answered by the DMCA, but I want to take it a step deeper and show that, even without the DMCA, some cloud computing companies would not incur liability when content is transmitted from their services to members of the public. The key to understanding why such cloud computing companies are not infringers is the key to understanding why Aereo is an infringer, and this turns on application of the volitional conduct test—a test codified in DMCA, but which is applicable in any context where the Transmit Clause is implicated. The volitional conduct test provides us with a way to separate the Aereo-like goats from the cloud computing sheep.

Under the Transmit Clause, it is a public performance to (1) transmit a performance of a work (2) to the public.1 There is no doubt that a performance of a work is being transmitted with Aereo’s service, so (1) is not at issue. The issue is whether this transmission is “to the public,” which is a term of art. Whether a transmission is “to the public” does not depend on the place to which it is sent, and a transmission to a private place can be “to the public” nonetheless. What makes it “to the public” is the relationship between the sender and the receiver, and this relationship, naturally, depends on the identity of each party. When the relationship between the sender and the receiver is a public one, the transmission is a public performance. And when the relationship between the sender and the receiver is a private one, the transmission is a private performance.

In Aereo,2 the Second Circuit focused on the receiver for each transmission, concluding that when there is a one-to-one relationship between the source copy and the receiver, the performance is private—even if we presume that Aereo is the sender. I think this makes no sense. If Aereo is the sender and the subscriber is the receiver, then that relationship is a public one and the transmission is a public performance. The Second Circuit’s reliance on there being a unique source copy for each receiver was misplaced, for the Transmit Clause cares only about who is transmitting to whom. It says nothing about a source copy—and in fact no source copy is even needed to publicly perform via transmission.3 The issue is simply who is directly causing the transmission to occur. Until we know who is transmitting to the receiver, that is, until we identify the sender, we can’t know whether the relationship between the sender and the receiver is a public one such that it’s “to the public.”

Without fair use or the DMCA to fall back on, the Aereo appeal squarely presents the issue of who is directly causing the transmission when a performance of a work is transmitted from Aereo’s service to the subscriber. In other words, the case boils down to who is the direct performer in this scenario. If the subscriber is the direct performer, then the performance is private since the sender and the receiver are the same party and the transmission therefore is not “to the public.” If Aereo is the direct performer, then the performance is public since the relationship between the sender and the receiver is a public one and the transmission therefore is “to the public.” And whether Aereo is the direct performer turns on the volitional conduct test: Is Aereo’s volitional conduct sufficient such that it directly causes the transmission? I think the answer is clearly “yes,” and how I get there is by utilizing the only bright-line rule under the volitional conduct test.

The genesis of the volitional conduct test is the famous Netcom decision,4 penned in 1995 by District Judge Ronald M. Whyte of the Northern District of California. Pre-DMCA, Judge Whyte was faced with deciding the liability of Netcom, an internet service provider, for infringement taking place utilizing its system. Critically, “Netcom was not itself the source of any of the infringing materials on its system,”5 and the issue was whether it could be a direct infringer of the reproduction right nonetheless. Judge Whyte held that it could not: “Although copyright is a strict liability statute, there should still be some element of volition or causation which is lacking where a defendant’s system is merely used to create a copy by a third party.”6 Since “such copies are uploaded by an infringing user,” Judge Whyte continued, Netcom’s actions were not a direct infringement of the reproduction right.7

Judge Whyte acknowledged that copyright infringement is a strict liability tort, meaning that neither knowledge nor intent to infringe need be shown. However, strictly applying that strict liability to reproductions occurring in cyberspace “would hold the entire Internet liable for activities that cannot reasonably be deterred.”8 Grounded in the policy argument that “it does not make sense to adopt a rule that could lead to the liability of countless parties whose role in the infringement is nothing more than setting up and operating a system that is necessary for the functioning of the Internet,” Judge Whyte proposed a dichotomy between a passive conduit, which provides only dumb pipes used by others to copy, and an active participant, which takes affirmative steps to bring about the copying.9 And central to Judge Whyte’s holding that Netcom was a passive conduit was the fact that it was the users who uploaded the content to Netcom’s service in the first place.

The key to understanding Netcom and its progeny is to recognize that the volitional conduct test is not used to determine whether a service provider should be held directly liable in the first place. Instead, a service provider seeks Netcom immunity because, without it, it would otherwise be a direct infringer. The volitional conduct test is perhaps best understood in terms of causation: Even though, because of strict liability, every service provider that is the cause-in-fact of the copying is thereby a direct infringer, certain service providers are nevertheless not directly liable for the copying because their actions are sufficiently remote. Perhaps the leading iteration of the volitional conduct test comes from the Fourth Circuit in CoStar:

[T]o establish direct liability under §§ 501 and 106 of the Act, something more must be shown than mere ownership of a machine used by others to make illegal copies. There must be actual infringing conduct with a nexus sufficiently close and causal to the illegal copying that one could conclude that the machine owner himself trespassed on the exclusive domain of the copyright owner.10

The volitional conduct test has also been adopted by the Second and Ninth Circuits,11 as well as by several district courts.12 In fact, Congress baked Netcom immunity under the volitional conduct test right into the DMCA.13 For example, Section 512(c) grants a qualifying service provider immunity “for infringement of copyright by reason of the storage at the direction of a user of material that resides on a system or network controlled or operated by or for the service provider.”14 The Second Circuit in YouTube and the Ninth Circuit in Shelter Capital read this language very broadly, finding that even transcoding and indexing user-submitted content to facilitate its public playback was protected by the safe harbor.15 While Section 512(c) protects a service provider for many automated acts that occur after a user uploads content to the service, the one thing it doesn’t provide immunity for is content supplied by the service provider itself.

How the volitional conduct test operates in the cloud is demonstrated in the Hotfile case, where the district court stated:

Thus, the law is clear that Hotfile and [the owner] are not liable for direct copyright infringement because they own and manage internet facilities that allow others to upload and download copyrighted material. . . . [T]he website merely allows users to upload and download copyrighted material without volitional conduct from Hotfile or [the owner]. . . . [N]othing in the complaint alleges that Hotfile or [the owner] took direct, volitional steps to violate the plaintiffs’ infringement. There are no allegations, say, that Hotfile uploaded copyrighted material. Therefore, under the great weight of authority, the plaintiffs have failed to allege direct copyright infringement.16

Because neither Hotfile nor the owner had engaged in sufficient volitional conduct, by, for example, uploading the infringing content at issue, the district court found that they could not be held directly liable for the infringement that occurred. Contrasted with Hotfile is Tasini,17 where the Supreme Court held that a database provider was directly liable for violating the public distribution right for content that it had supplied and made available as part of its publicly-accessible database. Because the database provider had supplied the infringing content at issue, this was a sufficient condition for finding that it was a direct infringer.

The import of all this is that where the service provider itself supplies the infringing content that is publicly distributed, displayed, or performed, that service provider can claim neither Netcom immunity nor DMCA immunity. Supplying the infringing content is sufficient for finding that the service provider has directly caused the infringement to occur—but it’s not necessary for such a finding. In other words, if the service provider itself supplies the infringing content, then it is a direct infringer. But sometimes the service provider can also be a direct infringer even if it did not supply the infringing content. This happens when the service provider engages in other volitional conduct “with a nexus sufficiently close and causal to the illegal copying,” to borrow the Fourth Circuit’s phrase, such that it is not only the factual cause of the copying, but the legal cause as well.

For example, in Megaupload, the district court found that the plaintiff had alleged that Kim Dotcom’s infamous website “serves as more than a passive conduit, and more than a mere ‘file storage’ company.”18 Several facts, including Megaupload’s reward program where it paid users to upload content, meant that Megaupload had possibly crossed the line from being a passive conduit to being an active participant in the infringement. Similarly, in Usenet, the district court found that the bulletin board service was “not merely a ‘passive conduit’ that facilitates the exchange of content between users who upload infringing content and users who download such content.”19 The “active measures” and “active steps” engaged in by the defendants “transformed [them] from passive providers of a space in which infringing activities happened to occur to active participants in the process of copyright infringement.”20

Like all proximate causation analysis, the volitional conduct test involves both art and science, and courts engaging in it typically juggle somewhat-vague concepts of foreseeability and temporality with normative decisions about social justice.21 As Dean William Prosser, author of Prosser on Torts and Reporter for the Restatement (Second) of Torts, put it, proximate causation is “our more or less inadequately expressed ideas of what justice demands.”22 Exactly how much volitional conduct it takes for a service provider to cross the line from passive conduit to active participant, no one can say. The analysis is, necessarily, somewhat arbitrary. The only bright-line criteria that exists under the volitional conduct test is that of who supplies the infringing content. When it’s the service provider itself, the volitional conduct on the part of the service provider is sufficient to find it directly liable.

This past week, District Judge Dale A. Kimball of the District of Utah found that the plaintiffs were likely to succeed on the merits against Aereo because it “provid[es] paying customers with retransmission of copyrighted works.” And this gets to the heart of why Aereo should lose before the Supreme Court. Aereo has crossed the line from being a passive conduit to being an active participant because it supplies the very content that is available using its service. Presumably, this is at least part of the reason why Aereo does not argue that Section 512(c) gives it DMCA immunity. Aereo cannot maintain that the content residing on its service is “by reason of the storage at the direction of a user” because the content comes from Aereo itself. The subscriber doesn’t go to Aereo to upload the content he already has to the cloud; the subscriber goes to Aereo to get the content he wants but doesn’t yet have—content which Aereo itself supplies.

The reason why Aereo, and not the subscriber, is the direct performer is simple. Because of copyright’s strict liability nature, every party that is the cause-in-fact of the copying is a direct infringer unless some other doctrine absolves that party from direct liability. The volitional conduct test uses a proximate causation analysis to determine whether a party’s actions are sufficiently remote such that it should not be held directly liable for the infringement. While the metes and bounds of the volitional conduct test are imprecise, to say the least, there is one bright-line rule: When the service provider itself supplies the very content at issue, that service provider’s actions can never be sufficiently remote to absolve it of direct liability. Because Aereo itself supplies the very content at issue—by implementing and making available its system of tens of thousands of tiny antennae that receive and retransmit performances—Aereo is the identity of the sender. And since the relationship between the sender (Aereo) and the receiver (the subscriber) is a public one, the transmission is a public performance.

The argument that it is Aereo’s subscriber, and not Aereo itself, who is the sender is untenable. The Transmit Clause was enacted to capture any service doing exactly what Aereo does, namely, retransmitting a broadcast to the public by any means possible. The petitioners in Aereo do a great job of traversing the history of the Transmit Clause in their opening brief before the Supreme Court, but the short version is this: The Supreme Court had held in a pair of cases that a community antenna television system or a cable system that captured an over-the-air broadcast and retransmitted it to the public was not engaged in a public performance.23 Congress enacted the Transmit Clause to legislatively overrule these two cases, and the words Congress chose, namely, “any device or process” used to transmit a performance, leave no doubt that its concern was with the function of the device or process utilized to make the transmission—not the underlying technical details of the device or process itself.24

Aereo’s argument boils down to it claiming that, even though the Transmit Clause on its face applies to “any device or process” that transmits a performance to the public, its particular device or process is somehow so special under-the-hood that the Transmit Clause doesn’t reach it. The problem with this argument is that the Transmit Clause means what it says, and it’s no answer to say that Aereo is simply doing for a subscriber what he could do for himself. This was the Supreme Court’s reasoning in the two cases the Transmit Clause was enacted to overrule. And even if we accept the argument that Aereo’s subscribers are really lessees of its equipment, the outcome is the same. The test is volitional conduct, and the conduct that matters is the conduct that Aereo itself has actually engaged in. If Aereo’s own actions are sufficient to find it directly liable absent any lease relationship, then its own actions are necessarily sufficient to find it directly liable even with the lease relationship. Because unless there is also some agency relationship between Aereo and its subscriber—for example, if Aereo were the subscriber’s employee—then Aereo’s own actions are only attributable to itself.

Cloud computing companies need not be worried about the outcome of the Aereo appeal because these service providers will continue to be protected by Netcom immunity under the volitional conduct test or by DMCA immunity. Assuming such service providers are protected by the DMCA, there’s an expansive amount of volitional conduct they can engage in without losing their safe harbor. How much volitional conduct they can engage in without losing Netcom immunity is less clear, but what is clear is that when a service provider itself supplies the very content at issue, there is no immunity under either the volitional conduct test or the DMCA. If Aereo loses before the Supreme Court, as I believe they should, it won’t negatively affect cloud computing companies because there is no reason for the Court to look any further than the fact that Aereo supplies the very content at issue. If anything, the Court’s decision will further cement into place the one bright-line rule that there is with the volitional conduct test, and this clarity in the law will redound to the cloud.

Special thanks to Terry Hart and Jonathan Greenfield for their valuable feedback in drafting this post.

Follow me on Twitter: @devlinhartline


  1. See 17 U.S.C.A. § 101 (West 2014) (“To perform or display a work ‘publicly’ means– (1) to perform or display it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or (2) to transmit or otherwise communicate a performance or display of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance or display receive it in the same place or in separate places and at the same time or at different times. *** To ‘transmit’ a performance or display is to communicate it by any device or process whereby images or sounds are received beyond the place from which they are sent.”). []
  2. See WNET, Thirteen v. Aereo, Inc., 712 F.3d 676 (2d Cir. 2013). []
  3. For example, live performances and television broadcasts can be captured and transmitted to the public even though the sender possesses no source copy. []
  4. See Religious Tech. Ctr. v. Netcom On-Line Commc’n Servs., Inc., 907 F.Supp. 1361 (N.D. Cal. 1995). []
  5. Id. at 1367. []
  6. Id. at 1370. []
  7. Id. at 1371. []
  8. Id. at 1372. []
  9. Id. []
  10. CoStar Grp., Inc. v. LoopNet, Inc., 373 F.3d 544, 550 (4th Cir. 2004). []
  11. See Cartoon Network LP, LLLP v. CSC Holdings, Inc., 536 F.3d 121 (2d Cir. 2008); Fox Broad. Co., Inc. v. Dish Network L.L.C., Case No. 12-cv-57048, 2014 WL 260572 (9th Cir. Jan. 24, 2014). []
  12. See, e.g., Field v. Google Inc., 412 F.Supp.2d 1106 (D. Nev. 2006); Playboy Enterprises, Inc. v. Webbworld, Inc., 991 F.Supp. 543 (N.D. Tex. 1997); Playboy Enterprises, Inc. v. Russ Hardenburgh, Inc., 982 F.Supp. 503 (N.D. Ohio 1997); Disney Enterprises, Inc. v. Hotfile Corp., 798 F.Supp.2d 1303 (S.D. Fla. 2011). []
  13. See, e.g., H.R. Rep. 105-551(I), at *11 (“As to direct infringement, liability is ruled out for passive, automatic acts engaged in through a technological process initiated by another. Thus, the bill essentially codifies the result in the leading and most thoughtful judicial decision to date: Religious Technology Center v. Netcom On-line Communications Services, Inc., 907 F. Supp. 1361 (N.D. Cal. 1995).”); ALS Scan, Inc. v. RemarQ Communities, Inc., 239 F.3d 619, 622 (4th Cir. 2001) (“[T]he ultimate conclusion on this point is controlled by Congress’ codification of the Netcom principles in Title II of the DMCA.”). []
  14. 17 U.S.C.A. § 512(c) (West 2014). []
  15. See Viacom Int’l, Inc. v. YouTube, Inc., 676 F.3d 19, 39 (2d Cir. 2012); UMG Recordings, Inc. v. Shelter Capital Partners LLC, 718 F.3d 1006, 1016 (9th Cir. 2013). []
  16. Disney Enterprises, Inc. v. Hotfile Corp., 798 F.Supp.2d 1303, 1308 (S.D. Fla. 2011) (internal citations and paragraph break omitted). []
  17. New York Times Co., Inc. v. Tasini, 533 U.S. 483 (2001). []
  18. Perfect 10, Inc. v. Megaupload Ltd., Case No. 11-cv-0191, 2011 WL 3203117, at *4 (S.D. Cal. July 27, 2011). vacated pursuant to settlement, 2011 WL 10618723 (S.D. Cal. Oct. 11, 2011). []
  19. Arista Records LLC v., Inc., 633 F.Supp.2d 124, 149 (S.D.N.Y. 2009). []
  20. Id. (internal quotations, citations, and brackets omitted). []
  21. See Mark Bartholomew & Patrick F. McArdle, Causing Infringement, 64 Vand. L. Rev. 675, 703 (2011). []
  22. Id. (internal quotations and footnote omitted). []
  23. See Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390 (1968); Teleprompter Corp. v. Columbia Broad. Sys., Inc., 415 U.S. 394 (1974). []
  24. See, e.g., H.R. Rep. No. 94-1476, at *63 (1976) (A performance occurs “either directly or by means of any device or process, including . . . any sort of transmitting apparatus, any type of electronic retrieval system, and any other techniques and systems not yet in use or even invented.”); id. at *64 (“The definition of ‘transmit’ . . . is broad enough to include all conceivable forms and combinations of wires and wireless communications media, including but by no means limited to radio and television broadcasting as we know them. Each and every method by which the images or sounds comprising a performance . . . are picked up and conveyed is a ‘transmission,’ and if the transmission reaches the public in [an]y form, the case comes within the scope of clause[] (4) . . . of section 106.”). []

The political conflicts and ambitions of [Shakespeare’s] England are known to the scholar and the specialist. But his plays will forever move men in every corner of the world. The leaders that he wrote about live far more vividly in his worlds than in the forgotten facts of their own rule.

Our civilization, too, will survive largely in the works of our creation. There is a quality in art which speaks across the gulf dividing man from man, nation from nation and century from century. …

[I]t is important to know that the opportunity we give to the arts is a measure of the quality of our civilization. It is important to be aware that artistic activity can enrich the life of our people; which is the central object of government. It is important that our material prosperity liberate and not confine the creative spirit.

—President Johnson, at the groundbreaking ceremony for the John F. Kennedy Center for the Arts on December 2, 1964.1

National Endowment for the Arts, Sunil Iyengar, on “Why The Arts Really Do Mean Business” — Check out this podcast with Copyright Alliance’s Sandra Aistars and National Endowment for the Arts Sunil Iyengar on the NEA’s recent report showing the arts and culture sectors contribute over half a trillion dollars to U.S. GDP and create nearly 2 million jobs.

‘Grindhouse’ to ‘John Carter’ – 18 ‘dream projects’ that turned into nightmares — An interesting article in its own right, though I was struck in particular by the author’s remarks that, “The best art comes from dealing with the limitations under which something is produced, and when all of those limitations are taken away, you can end up with something that is wall-to-wall indulgence.” and “The lesson to take away is that when you want to pay tribute to the guys who came before you, do it in a way like Lucas and Spielberg did with Indiana Jones. Take that energy and channel it into something of your own”

New report says how much advertising is going to piracy sites — $227 million in 2013. And, according to the report, a significant chunk of that is coming from dozens of “blue-chip advertisers”, major brands like “AT&T, Lego, and Toyota.”

‘House of Cards’ Searches on Google Turn Up Pirate Links as Top Results — Variety reports, “In a Google search for “Watch House of Cards” on Wednesday, the top two results were links to apparent pirate sites, and” That reminds me of the story about Charlie Chaplin entering a Charlie Chaplin look-alike contest and coming in third.

Utah Judge Gives TV Broadcasters an Injunction Against Aereo — Although the Supreme Court has already agreed to determine the liability of the internet streaming TV service, the Utah court heard and granted a motion to enjoin Aereo in a parallel proceeding. Said the judge, “Aereo’s retransmission of Plaintiffs’ copyrighted programs is indistinguishable from a cable company.”

Kim Dotcom raid warrants legal, Court of Appeals rules — Despite continual proclamations of innocence, Dotcom has done nothing but delay having his day in court. Here, a New Zealand court strikes down one such attempt.

Wikipedia: as accurate as Britannica? — A 2005 study has given rise to the gospel that Wikipedia is as accurate as Encyclopedia Britannica. Here, Andreas Kolbe points out the sharp limitations of that study. “Nicholas Carr put it this way: in limiting itself to topics like the ‘kinetic isotope effect’ or ‘Meliaceae’, which no one without some specialized understanding of the subject matter would even be aware of, the Nature survey played to Wikipedia’s strengths. Carr also established that the Nature ‘study’ was not actually an expert-written research article of the type that built the reputation of Nature, but a non-peer-reviewed piece of news journalism (a fact he confirmed with the piece’s author, Jim Giles).”


  1. As quoted in Copyright Law Revision Pt. 6, Supplementary Report of the Register of Copyrights on the General Revision of the U.S. Copyright Law: 1965 Revision Bill, xiv (1965). []

How to photograph the Sochi Winter Olympics, as told by a pro — Sports photojournalist Adam Pretty describes what its like shooting the 2014 games. As a Getty Images photographer, Pretty also receives some help to capture the best moments. “Getty will have laid 20km of cable to create a state-of-the-art fiber optic network that will connect all the key photo positions inside the venues directly back to the Getty Images office in the Main Press Center. With the VLAN connections on the mountains, the photographers will be able to dump the contents of their memory cards onto their laptops, and they’ll automatically be back with the editor a minute or so later.”

Proposal for Compulsory Remix License Has Foes in Steven Tyler and Attorney Dina LaPolt — A post-deadline comment for the IPTF Green Paper proceedings on copyright policy from a number of music luminaries, including Aereosmith frontman Tyler, Don Henley, Sting, Ozzy Osbourne, Britney Spears, Deadmau5, and Joe Walsh takes on the idea of a compulsory license for remixes. Could this mean that attention to current copyright review discussions is starting to pick up?

Spanish box-office loses almost half of its audience over nine years — The Spanish film industry has almost completely crumbled in the past decade. Recently, the IIPA has also recommended to the USTR that Spain be placed on the Special 301 Watch List, noting “In stark contrast to so many of its neighbors in Western Europe, Spain suffers from unrelentingly high rates of digital piracy in every sector – music, film, videogames, software, and books.”

Pandora Suit May Upend Century-Old Royalty Plan — Few realize that ASCAP and BMI, organizations that collect public performance royalties for songwriters and music publishers, are limited in what they can do by an over 70 year old consent decree with the DOJ. Is it time for a change?

The Australian Law Reform Commission officially recommends adoption of fair use — The Australian Law Reform Commission (ALRC), an independent statutory body in Australia tasked with reviewing existing law and offering recommendations for reform, has dropped a 478 page report recommending the country repeals specific exemptions for, to name a few, personal copying and educational uses, and adopts a US-style fair use provision. Around 60% of ALRC recommendations are substantially implemented, while another 30% are partially implemented.1

The Day the Internet Didn’t Fight Back — Apparently repeating “SOPA” over and over is starting to lose its effectiveness.

Silicon Valley Needs to Lose the Arrogance or Risk Destruction — From Wired, no less. “It’s all adding up to a nasty picture of Silicon Valley—of an industry that hoovers up personal data and reaps massive profits from its use, preaching a gospel of sharing but refusing to share back.”


  1. Australian Law Reform Commission Annual Report 2011-12, pg. 30. []

On January 24th, the Ninth Circuit denied a petition for rehearing en banc in Fox Broadcasting v. Dish Networks. The decision was, however, amended slightly from the original opinion, released last July.

Fox Broadcasting v. Dish Network involved satellite television provider Dish’s Hopper, a set-top device with digital video recording (DVR) and video on demand (VOD) capabilities. Relevant to this discussion, Dish began providing its Hopper customers with a service called “PrimeTime Anytime” (PTAT). PTAT, when enabled, automatically recorded all primetime (8 p.m.-11 p.m.) programming from the four broadcast networks (ABC, CBS, NBC, and Fox). The recordings are saved locally on a customer’s Hopper for a predetermined amount of time (typically 8 days), after which they are automatically deleted.

Fox sued for copyright infringement, alleging, in part, that Dish directly infringed its reproduction right when it made PTAT copies of Fox television programs. Fox sought a preliminary injunction

The District Court in Dish concluded that “at this stage of the proceedings, the Court is not satisfied that PTAT has crossed over the line that leads to direct liability. Despite Dish’s involvement in the copying process, the fact remains that the user, not Dish, must take the initial step of enabling PTAT after deciding that he or she wants to initiate the recording. The user, then, and not Dish, is the ‘most significant and important cause’ of the copy.” This, despite the fact that Dish designed, houses, and maintains the system that creates the copies. The court also did not consider the following facts indicative of causation:

  • “Dish decides which networks are available on PTAT and has defaulted the PTAT settings to record all four networks.”
  • “Dish also decides the length of time each copy is available for viewing: PTAT recordings are automatically deleted after expiration of a certain number of days, and a user may neither delete nor preserve the original PTAT copy before that time.”
  • “Dish decides when primetime recordings start and end each night, and it maintains the authority to modify those times according to the particular programs airing that night. Additionally, the user cannot stop a copy from being made during the copying process, but must wait until the recording ends before disabling the link.”

The Ninth Circuit affirmed this result, saying “the district court did not err in holding that Fox did not establish a likelihood of success on its direct infringement claim.”

The “volitional conduct” test relied upon by both courts is idiosyncratic within the larger contours of tort law, and neither court explains why it decided to extend the test so far beyond previous applications. I’ve detailed some criticisms of the opinion when the original decision was released last summer. But today I want to head overseas and look at a recent case where a court got it right.

TV Now

In 2012, an Australian court considered a very similar question. Telecom service provider Optus created “TV Now”, a service that enables customers to watch and record television broadcasts through a mobile device or personal computer. It was sued by a group of sports leagues for infringing its copyright in broadcast football games. The trial court judge held that Optus was not liable for infringement, relying in part on the US court decision in Cablevision and analogizing the service to a VCR or DVR. The decision was appealed.

The primary issue on appeal was “who ‘does’ the act of copying.” The appellate court said “the rival contenders are Optus, the subscriber, or Optus and the subscriber jointly.”

The court first considers the case that the copies were made solely by the subscriber. This approach is based on the premise that “Optus makes available to a subscriber a facility (a service) which enables the subscriber as and when he or she is so minded to use that facility to record broadcasts and later to view them. The copies that are made are the result of the subscriber’s use of the facility though the actual making of them requires Optus’ technology to function as it was designed to.” This is the position taken by the trial court judge, which said that to “make” a copy means “’to create’ by selecting what is to be recorded and by initiating a process utilising technology or equipment that records the broadcast”, a conception the court here said “robs the entirely automated copying process of any significance beyond that of being the vehicle which does the making of copies.”

Besides the analogy of Optus as a DVR operated by a subscriber, the trial court judge also saw the subscriber choosing to copy as “the last ‘volitional act’ in the sequence of acts leading to a copy being made and, for that reason, is significant in determining the identity of the maker.” The appellate court found these arguments less than compelling.

It noted too the trial court’s citation to Cablevision and the “volitional conduct concept.” However, the court viewed this concept as relevant only when there is a need to distinguish between direct and contributory liability—which is the case in U.S. jurisprudence but not under Australian law. The court added, referring to the academic criticism of the Cablevision decision, “It equally is not apparent to us why a person who designs and operates a wholly automated copying system ought as of course not be treated as a ‘maker’ of an infringing copy where the system itself is configured designedly so as to respond to a third party command to make that copy.”

The court concluded:

[W]e consider that Optus’ role in the making of a copy – ie in capturing the broadcast and then in embodying its images and sounds in the hard disk – is so pervasive that, even though entirely automated, it cannot be disregarded when the “person” who does the act of copying is to be identified. The system performs the very functions for which it was created by Optus. Even if one were to require volitional conduct proximate to the copying, Optus’ creating and keeping in constant readiness the TV Now system would satisfy that requirement. It should also be emphasised that the recording is made by reason of Optus’ system remaining “up” and available to implement the subscriber’s request at the time when its recording controllers poll the user database and receive a response indicating that a recording has been requested. What Optus actually does has –

a nexus sufficiently close and causal to the illegal copying that one could conclude that the machine owner … trespassed on the exclusive domain of the copyright owners: CoStar Group Inc v LoopNet Inc, 373 F3d 544 at 550 (4th Circ. 2004).

… Put shortly Optus is not merely making available its system to another who uses it to copy a broadcast. Rather it captures, copies, stores and makes available for reward, a programme for later viewing by another.

The real question for the court was whether Optus alone is liable or whether Optus and the subscriber are jointly liable. And here, the court indicated that the latter was most likely (since only Optus was named as a defendant, it was not strictly necessary for the court to resolve the question of the subscriber’s liability.) Said the court:

If one focussed not only upon the automated service which is held out as able to produce, and which actually produces, the copies but also on the causative agency that is responsible for the copies being made at all, the need for a more complex characterisation is suggested. The subscriber, by selecting the programme to be copied and by confirming that it is to be copied, can properly be said to be the person who instigates the copying. Yet it is Optus which effects it. Without the concerted actions of both there would be no copy made of a football match for the subscriber. Without the subscriber’s involvement, nothing would be created; without Optus’ involvement nothing would be copied. They have needed to act in concert to produce – they each have contributed to – a commonly desired outcome. The subscriber’s contributing acts were envisaged by the contractual terms and conditions. How they were to be done were indicated by the prompts given on the Optus TV Now TV guide page. The common design – the production of the selected programme for transmission to the subscriber – informed the solicitation and the taking of a subscription by the subscriber; it was immanent in the service to be provided.

The Australian court’s conclusion seems to hew more closely to general principles of tort law. Though the issue is settled, for now, in the Ninth Circuit, the Second Circuit is set to take up the issue soon in a parallel case involving the same Dish services and the other three major broadcast networks. Last October, the Southern District Court of New York denied a preliminary injunction against Dish in part because it agreed that the customer, not Dish, made the PTAT copies. That decision has been appealed, and oral arguments are set for next week.

Previously, I began looking at the legal questions involved in American Broadcasting Companies v. Aereo, currently in front of the Supreme Court. The issue is whether Aereo, by providing a service that allows paying subscribers to watch broadcast television online, is publicly performing the copyrighted programming. The Second Circuit said it is not, based on its 2008 decision in Cablevision, which held that a transmission from a unique copy of a work is merely a private performance. Aereo, relying on the Cablevision decision, designed its service to purportedly assign a unique antenna to each individual subscriber.

The Copyright Act states that to perform a work publicly means, in relevant part, “to transmit or otherwise communicate a performance … of the work … to the public, by means of any device or process, whether the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.”

The seemingly straightforward language belies some of the complexities that result, as evidenced by the Supreme Court’s agreement to hear Aereo. The approach I began in my previous post was to break the definition up into its separate components to understand it better: we have an actor (otherwise unnamed in the Act, we can refer to this as a “performer”), the action (“to transmit or otherwise communicate”), the object (“a performance of the work”), and a prepositional phrase (“to the public”).

To the public redux

I first took a look at what “to the public” means. Having had the opportunity to consider some of the feedback I’ve received on the article, I do want to make a slight modification to my definition.1

Originally, I wrote, “Any transmission from one person to another person who is not a family member or a close social acquaintance is a public performance.” The mention of family members and social acquaintances invites a level of specificity that I did not intend and distracts from the original purpose, which is to focus on the relationship between performer and audience. The legislative history of the Copyright Act says that under the second clause of the definition of “publicly”, performances during “routine meetings of business and governmental personnel” would be exempt “because they do not represent the gathering of a ‘substantial number of persons.'” I think the definition of “to the public” should similarly exempt such performances. It seems it would be more accurate to step back one level of abstraction; rather than referring to family members and social acquaintances, we should refer to “public” and “private” relationships. “Public” relationships are those that tend to be described as commercial, arms-length, or impersonal, and are strongly unidirectional, while “private” relationships tend to be described as familial, social, or collegial, and are much more reciprocal in nature.

What is a performance?

With that out of the way, we can turn to the next question in the analysis: what is a performance?

As with “to the public”, “performance” is not directly defined in the Copyright Act. But the Act does define the act of performing:

To “perform” a work means to recite, render, play, dance, or act it, either directly or by means of any device or process or, in the case of a motion picture or other audiovisual work, to show its images in any sequence or to make the sounds accompanying it audible.2

A performance right for dramatic compositions first appeared in US copyright law in 1856.3 Nondramatic musical works gained a performance right in 1897,4 and the 1909 Copyright Act, the last general revision before the current Copyright Act, provided performance rights for dramatic works, nondramatic literary works, and musical compositions. However, the current Copyright Act is the first to actually define “perform.”

Though the current Copyright Act did not become law until 1978, the language of the definition for performance is substantially the same as the definition included in the 1965 version of the bill.5 Following the introduction of that bill, the US Copyright Office released a report explaining the bill in detail. The Supplementary Report explained:

Under clause (1) of section 106(b), to ”perform” a work means ”to recite, render, play, dance, or act it.” This includes, for example, the reading aloud of a literary work, the singing or playing of music, the dancing of a choreographic work, and the acting out of a dramatic work or pantomime. A work may be performed ”either directly or by means of any device or process,” and these devices or processes would encompass sound or visual reproduction equipment of all kinds, amplifying systems, radio and television transmitting and receiving apparatus, electronic retrieval devices, and a host of other techniques, undoubtedly including some not invented yet. In the case of a motion picture, performance would mean ”to show its images or to make the sounds accompanying it audible.” It would be clear under this language that the purely aural performance of a motion picture sound track would constitute a performance of the motion picture; but, if the sounds on the soundtrack are reproduced on an authorized phonorecord, performance of the phonorecord would not be a performance of the motion picture.

The definition is relatively clear, and there are very few cases that have have been confronted with issues relating to its interpretation. One of the most relevant and thorough is US v. ASCAP.6 There, the court was tasked with considering whether a download of a music file was a performance. It looked at the “ordinary sense” of the words used in the Copyright Act’s definition of performance and concluded that performance requires “contemporaneous perceptibility.” Said the Second Circuit:

The downloads at issue in this appeal are not musical performances that are contemporaneously perceived by the listener. They are simply transfers of electronic files containing digital copies from an on-line server to a local hard drive. The downloaded songs are not performed in any perceptible manner during the transfers; the user must take some further action to play the songs after they are downloaded. Because the electronic download itself involves no recitation, rendering, or playing of the musical work encoded in the digital transmission, we hold that such a download is not a performance of that work, as defined by § 101.

In other words, “Transmittal without a performance,” said the Second Circuit, “is not a ‘public performance.’”

Transmission without performance

As support for this proposition, the court cited to Columbia Pictures Indus. v. Prof’l Real Estate Investors.7 In Columbia, the operators of a hotel resort offered guests the ability to rent movies on videocassette at the front desk, which they could watch on hotel-provided equipment in their own rooms. Plaintiffs sued, relying on the Third Circuit’s line of cases holding video store operators liable for public performance for operating private viewing booths on their premises.8 The Ninth Circuit, however, rejected this argument, holding that hotel rooms, though offered to the general public, become private spaces once they are rented.9 Thus, the operation of the equipment is not a performance in a public place.

Plaintiffs, however, also argued that the act of providing videocassettes to hotel guests implicated the Transmit Clause because the hotel was “otherwise communicat[ing]” the films. The Ninth Circuit rejected this argument as well.

A plain reading of the transmit clause indicates that its purpose is to prohibit transmissions and other forms of broadcasting from one place to another without the copyright owner’s permission. The Act provides a definition of “transmit.” “To `transmit’ a performance or display is to communicate it by any device or process whereby images and sounds are received beyond the place from which they are sent.” According to the rule of ejusdem generis, the term “otherwise communicate” should be construed consistently with the term “transmit.”  Consequently, the “otherwise communicate” phrase must relate to a “process whereby images or sounds are received beyond the place from which they are sent.”

This reading is reinforced by the rest of the transmit clause which refers to the use of transmission devices or processes and the reception by the public of the performance. Devices must refer to transmission or communication devices, such as, perhaps, wires, radio towers, communication satellites, and coaxial cable, while reception of the performance by the public describes acts, such as listening to a radio, or watching — network, cable, or closed-circuit — television “beyond the place” of origination.

In sum, when one adds up the various segments of clause (2), one must conclude that under the transmit clause a public performance at least involves sending out some sort of signal via a device or process to be received by the public at a place beyond the place from which it is sent.

Nothing that La Mancha has done has violated this common sense construction of the transmit clause. While La Mancha has indeed provided the videodisc player, television screens, guest rooms, and makes videodiscs available in the lobby, we are not persuaded that any transmission of the kind contemplated by the statute occurs. If any transmission and reception occurs, it does so entirely within the guest room; it is certainly not received beyond the place from which it is sent. We are not persuaded that the term “otherwise communicate” can be read so broadly as to include the videodisc arrangements at La Mancha.

The reasoning of the Ninth Circuit here and the Second Circuit in ASCAP seems sound.10

To perform a work includes the transmission of a contemporaneously perceptible rendition of the work and is distinguished from the delivery of a work, including electronic delivery via transmission.11

The nature of a performance

A “performance of a work”, then, follows from this definition. If “performing” means rendering a work so that it is contemporaneously perceptible, a “performance of a work” is something that is contemporaneously perceptible. It is intangible—the Copyright Act refers to the tangible objects that embody works as copies. It is conceptual, an act rather than a thing.12 A performance is not the actor on stage, nor the sound waves emanating from a speaker, nor the photons transmitted across fiber optic data lines.13

It is also, perhaps, worth pointing out that it is not a performance that is embodied in a copy but a work.14 A performance is, by definition, incapable of embodiment; indeed, one does not even need a copy to perform a work, as is the case of a singer singing from memory.

And it is a conceptual unity. Consider two members in an audience. Neither will be perceiving the exact same thing, both because they are in different locations and because they themselves have variations in their eyes and ears that shape their personal perception. But conceptually there is only one performance of the work, and it is exactly the same for each audience member for copyright purposes.

The same holds true if a performance reaches its audience via transmission rather than via sound and light waves through the air. The Central District Court of California explains why in its decision enjoining FilmOnX (then “BarryDriller”):

The definition section sets forth what constitutes a public performance of a copyrighted work, and says that transmitting a performance to the public is a public performance. It does not require a “performance” of a performance. The Second Circuit buttressed its definition with a “cf.” to Buck v. Jewell-La Salle Realty Co., 283 U.S. 191, 196 (1931), which interpreted the 1909 Copyright Act’s provision of an exclusive right to publicly perform a musical composition and held that “the reception of a radio broadcast and its translation into audible sound” is a performance. But Buck, like Cablevision and this case, was concerned with a copyright in the work that was broadcast. The Supreme Court was not concerned about the “performance of the performance” – instead, it held that using a radio to perform the copyrighted song infringed the exclusive right to perform the song (not to perform the performance of the song).

The Transmit Clause explicitly recognizes this conceptual unity, saying a work is performed publicly when it is transmitted even if “the members of the public capable of receiving the performance … receive it in the same place or in separate places and at the same time or at different times.” There could be thousands of separate transmissions, but there is still only the performance.

These may seem like inconsequential distinctions, but I think they undermine even more the Second Circuit’s notion that we can conceptually sever performances of the same work made by the same actor but from distinct copies. If the Transmit Clause spoke about the audience capable of receiving the physical transmission, as the Second Circuit says it does, then it would be possible to have multiple private performances since physical transmissions are discrete and separable. But it doesn’t—the Transmit Clause speaks about the audience capable of receiving the “performance of a work”, which remains a conceptual unity no matter how scattered the audience is.15

This interpretation of performance is, in my opinion, far more consonant with the statutory language and the ordinary use of the terms than the Second Circuit’s “unique copies” interpretation. It also means, as we’ll see in the next installment of this series, that most of the heavy lifting (at least for issues arising in the cloud computing context) occurs in the causation inquiry, i.e., “who is the performer”?


  1. I also flubbed when I referred to the three situations as being “mutually exclusive”. What I meant instead is that each situation excludes the previous. For example, a “place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered” could include a place open to the public, but that would make the previous clause redundant. Similarly, the last clause, transmission to the public, could include transmission to the public in a public place, or transmission to a place where a substantial number of persons is gathered, but that would make the previous two clauses redundant.

    I think it still follows that each situation is analyzed by exclusive means. We can presume (from the case law) that a “place open to the public” turns on the nature of the place. The “place where a substantial number of persons…” situation doesn’t turn on the nature of the place: if such a place were public, then the first situation covers it, while being a private place is insufficient to tell us if there are a substantial number of persons there. The same holds true for the last situation. The nature of the place is irrelevant (otherwise the first clause would be redundant), the size and relationship of the audience to each other is irrelevant (otherwise the second clause would be redundant). So what is there left to consider? The only thing I can come up with is the relationship between the performer and the audience. There could be others I am not aware of, but the relationship between performer and audience seems to provide a workable rule consistent with the Copyright Act’s structure and purpose. []

  2. 17 U.S.C. § 101. []
  3. Act of August 18, 1856, 34th Cong., 1st Sess., 11 Stat. 138. []
  4. Act of January 6, 1897, 44th Cong., 2d Sess., 29 Stat. 694. []
  5. H.R. 4347 and S. 1006, 89th Cong. []
  6. 627 F. 3d 64 (2nd Circuit 2010). []
  7. 866 F.2d 278, 282 (9th Cir. 1989). []
  8. Columbia Pictures Industries v. Redd Horne, 749 F.2d 154 (3d Cir.1984); Columbia Pictures Industries, Inc. v. Aveco, Inc., 800 F.2d 59 (3d Cir.1986). []
  9. In support of this conclusion, the court relied on Fourth Amendment case law holding that hotel guests have constitutional protection from unreasonable searches and seizures in hotel rooms. []
  10. And indeed, the Copyright Office’s 1965 Supplementary Report adds support.

    In general the concept of ”performance” must be distinguished sharply from the reproduction of copies on the one hand and the exhibition of copies on the other. It has been suggested that some of the internal operations of a computer, such as the scanning of a work to determine whether it contains material the user is seeking, is closely analogous to a ”performance.” We cannot agree, and for this reason we deleted from the definition of ”perform” the ambiguous term ”represent” which appeared in the 1964 bill. A computer may well ”perform” a work by running off a motion picture or playing a sound recording as part of its output, but its internal operations do not appear to us to fall within this concept. []

  11. Thus, uploading of a work to a web site, like downloading, is not a performance. []
  12. performance“, Merriam-Webster Online, “an activity (such as singing a song or acting in a play) that a person or group does to entertain an audience”. []
  13. “Very few people gather around their oscilloscopes to admire the sinusoidal waves of a television broadcast transmission. People are interested in watching the performance of the work.” Fox Television Stations v. BarryDriller, at 5 (C.D. Cali December 27, 2013). []
  14. Congress separately and outside the Copyright Office has made it a criminal offense to fix, without authorization, a live musical performance in a copy, thus underscoring this distinction. See 18 U.S.C. § 2319A. []
  15. Likewise, the Transmit Clause does not speak about “the potential audience of a particular ‘work’,” an alternative interpretation that the Second Circuit raises and dismisses. []

Too often, copyright critics speak dismissively of permission and belittle the mere individual right involved. Such rhetoric surrounded last week’s IP Subcommittee hearing on the scope of fair use, and while the panelists largely agreed that no legislative changes were needed regarding fair use, there were sharp divisions over whether the doctrine currently goes too far or not far enough.

One does not win popularity contests critiquing fair use. How can you ignore the public benefit of Google Books? How can you be against fan fiction? But the question is not whether or not there is a public benefit or creativity involved, the question is how to mediate between public and private interests. And, as fair use is on an expansive course—and increasingly becoming part of the business model of large corporations—it is worth asking how easily public interests should trump private rights.

That’s not to say there are not, as with any form of property or right, competing equities that must be considered, and balanced against a copyright owner’s interests when they come into conflict. Fair use has traditionally played this role. But lately, it has seen a steady and troubling expansion, one that more and more is comfortable with letting the common good override individual rights.1

For those who fall on the “fair use has gone too far” side, who are worried about large companies getting a “volume discount” on fair use to excuse the wholesale copying of millions of works,2 or who share the concerns of Supreme Court Justice Kennedy—who said in his concurrence in Campbell v. Acuff-Rose Music, “If we allow any weak transformation to qualify as parody, however, we weaken the protection of copyright. And under-protection of copyright disserves the goals of copyright just as much as overprotection, by reducing the financial incentive to create”—it is worth reaffirming some of the principles that fair use is established on.

A Marketplace for Ideas

Fair use is necessary, but it necessarily must be limited—especially if the primary goal of the doctrine is to complement the goal of copyright rather than subvert it. “[T]he author’s consent to a reasonable use of his copyrighted works ha[d] always been implied by the courts as a necessary incident of the constitutional policy of promoting the progress of science and the useful arts, since a prohibition of such use would inhibit subsequent writers from attempting to improve upon prior works and thus . . . frustrate the very ends sought to be attained.”3 (Note here the reference to consent.)

One of the reasons fair use analysis includes an examination of the effect of the use upon the potential market for or value of the copyrighted work is because the creation of a marketplace is a core function of copyright law.

By understanding the predicate concepts of the Copyright Clause (limited times, authors, exclusive rights, and writings) the central purpose of the Copyright Clause can be seen: it achieves a public benefit in the progress of knowledge by securing such rights. The structure of the Copyright Clause is not limited to a proclamation that this is beneficial. Rather, it assumes that promoting the progress of knowledge is advantageous and directs Congress to achieve this benefit by securing exclusive rights in intellectual property. It mandates the creation of a marketplace, wherein this unique form of property, the copyright, may be traded and protected.4

Thus, as an economic asset, copyright provides the certainty necessary for the creation of a market, one where transactions can occur peacefully and systematically, where private ordering can emerge, and where economic efficiency can be promoted. Considering the size of this cultural, scientific, and artistic market in the U.S. reached over one trillion dollars in contributions to the GDP in 2013, it seems safe to say this approach is working generally well.

But as with any market, there will be disagreements over price, or refusal to work together. These disagreements, by themselves, should not invite fair use. There are plenty of situations where licensing might be too difficult or expensive, where some modicum of public benefit might be achieved absent the barriers of prices or permission—but this is true of all markets and forms of property. Fair use should be reserved, as with any limitation on individual rights and property, for those situations where a compelling public interest exists, one that demands recognition. As famed treatise Nimmer on Copyright explains, “if the ‘progress of science and useful arts’ is promoted by granting copyright protection to authors, such progress may well be impeded if copyright protection is virtually obliterated in the name of fair use.”5

Equally important is a defense of the notion of consent, or permission. Despite consent being a foundation of civil government—the Declaration of Independence expressly notes governments only derive “their just powers from the consent of the governed”—and a moral and legal concept that pervades our institutions, some copyright skeptics reject the idea.6 What they see as the foundation instead—divine right, perhaps?—is unclear.

Nevertheless, my guess is that most of us would choose consent over coercion.7 And this provides an additional reason to make sure fair use is properly cabined.

The “right to exclude” is perhaps overstated as an essential element of property. Property in general, and copyright specifically, create a freedom to choose how to use economic assets. Legal theorist Felix Cohen puts it this way: “Private property is a relationship among human beings such that the so-called owner can exclude others from certain activities or permit others to engage in those activities and in either case secure the assistance of the law in carrying out his decision.”8 The key here is that the choice itself (so long as it does not interfere on someone else’s interests) is irrelevant.9 I’d like to point out two general reasons why this sort of structure is important to maintain in copyright.

One, it respects the dignity and autonomy of individuals:

An autonomous person is one who directs her own life, and not someone whose life is controlled by others. Of course, we can’t all do whatever we would like without running the risk of interfering with each other. Hence, from the idea of autonomy, we might derive the idea of moral rights and duties that create for each individual a sphere of autonomous action, in which each individual can direct her own life without interfering with the like freedom of others to do the same. The moral force of consent comes naturally if one accepts autonomy as a central moral value. Consent allows others to enter one’s sphere of autonomy. So long a consent is freely given, consented-to rights violations seem perfectly consistent with the idea that rights protect a sphere of individual choice.10

This respect is especially important  regarding expressive works, which can be seen as an extension or reflection of a creator’s personality.11

Two, it provides for a plurality of valid choices. Some creators may choose to partner with larger businesses, some may choose to go their own way and stick to specific formats, some may choose to disseminate freely and as widely as possible. None of these is an inherently “right” choice. All creators place different priorities on getting compensated, seeking exposure, or maintaining creative control, priorities that can easily shift from one project to the next and evolve throughout a creator’s career. None of these is an inherently “correct” choice. One of the principles of the free market is that vesting this choice at the granular level will result in the greatest advance of liberty and efficiency. A system that narrows the range of plurality is inferior to one that preserves a wide range of reasonable choices.

Exceptions to this sort of regime, including fair use, should thus be applied judiciously and conservatively.

The Inherent Public Interest in Copyright

For all the talk about fair use’s role in facilitating public interests, it is easy to forget that copyright—indeed, all property rights—inherently serve the public interest.12 So it bears repeating. The creation of a marketplace for expressive works and the protection of autonomy, both discussed above, undoubtedly advance public interests.13

The results of copyright as being in the public interest are palpable. But for the labor of an author to write about how to cultivate a certain skill, the public would not have that particular expression of knowledge. But for the labor of a filmmaker, members of the public would not see a film that may change the way they view the world or provide them a greater understanding of themselves.

The interests of the public and the author coincide, something recognized by the drafter of the Copyright Clause, James Madison, in the early days of the republic, and reiterated to this day. 14 In Eldred v. Ashcroft, the Supreme Court explicitly rejected as missing the mark the assertion that copyright law “must serve public, not private, ends.” Instead, it said, “The two ends are not mutually exclusive; copyright law serves public ends by providing individuals with an incentive to pursue private ones.”

So, yes, fair use has a long pedigree in U.S. copyright law and an important role to play. But at the same time, courts and policy makers should be mindful that the doctrine does not end up subverting the very goals it is meant to reinforce.


  1. See, for example, Princeton University Press v. Michigan Document Services, 99 F.3d 1381, 1393 (6th Cir. 1996) (J. Martin dissent), saying fair use is intended “to protect society’s vested interest in the sharing of ideas and information against pursuits of illegitimate or excessive private proprietary claims” (Emphasis added); Monge v. Maya Magazines, 688 F.3d 1164, 1184 (9th Cir. 2012) (J. Smith dissent), “The majority’s fair use analysis in this case … thwarts the public interests of copyright by allowing newsworthy public figures to control their images in the press.” []
  2. Statement of June Besek, Hearing on Fair Use (Jan. 28, 2014), “It’s as though courts are according some kind of ‘volume discount’ for fair use, where a massive taking justifies a lower level of scrutiny in a fair use determination. It becomes increasingly difficult to explain to authors and public alike a copyright regime that rigorously examines the extent of a single scholar’s partial copying, while essentially according a free pass to a for-profit enterprise’s massive takings.” []
  3. Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 US 539, 549 (1985), quoting H. Ball, Law of Copyright and Literary Property 260 (1944). []
  4. David A. Householder, The Progress of Knowledge: A Reexamination of the Fundamental Principles of American Copyright Law, 14 Loy. L.A. Ent. L. Rev. 1, 35 (1993). See also Adam Mossoff, How Copyright Drives Innovation in Scholarly Publishing, George Mason Law & Economics Research Paper No. 13-25 (2013), “This commercialization policy—that intermediaries like publishers should be rewarded for their labors in creating the legal and market mechanisms necessary to disseminate works—is essential to the American copyright system.” []
  5. 4 Melville B. Nimmer and David Nimmer, Nimmer on Copyright § 13.05[E][1] (2013). []
  6. See, for example, here, here, and here. []
  7. See Cass R. Sunstein, It’s for Your Own Good! NY Review of Books, March 7, 2013. []
  8. Felix Cohen, Dialogue on Private Property, 9 Rutgers Law Review 357, 373 (1954). It is worth noting that Cohen considered copyrights as property. []
  9. See, for example, John Stuart Mill, On Liberty (1869), “the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right.” []
  10. Consent“, Legal Theory Lexicon. []
  11. Accord Justin Hughes, The Philosophy of Intellectual Property, 77 Geo. L. J. 287 (1988). []
  12. See Thomas W. Merrill and Henry E. Smith, The Morality of Property, 48 Wm. & Mary L. Rev. 1849 (2007). []
  13. See, for example, Daniel Solove, ‘I’ve Got Nothing to Hide’ and other Misunderstandings of Privacy, 44 San Diego Law Review 745, 762 (2007), agreeing with proposition that “Society makes space for the individual because of the social benefits this space provides.” []
  14. Writing in the Federalist Papers, Madison said of the Clause, “The utility of this power will scarcely be questioned. The copyright of authors has been solemnly adjudged in Great Britain, to be a right of common law. … The public good fully coincides…with the claims of individuals.” []