Cross-posted on the Copyhype 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 hotfile.com 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 for his valuable feedback in drafting this post.
Follow me on Twitter: @devlinhartline
© 2014 Devlin Hartline. Licensed under the Law Theories Public License 1.0.
- 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.”). ↩
- See WNET, Thirteen v. Aereo, Inc., 712 F.3d 676 (2d Cir. 2013). ↩
- For example, live performances and television broadcasts can be captured and transmitted to the public even though the sender possesses no source copy. ↩
- See Religious Tech. Ctr. v. Netcom On-Line Commc’n Servs., Inc., 907 F.Supp. 1361 (N.D. Cal. 1995). ↩
- Id. at 1367. ↩
- Id. at 1370. ↩
- Id. at 1371. ↩
- Id. at 1372. ↩
- Id. ↩
- CoStar Grp., Inc. v. LoopNet, Inc., 373 F.3d 544, 550 (4th Cir. 2004). ↩
- 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). ↩
- 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). ↩
- 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.”). ↩
- 17 U.S.C.A. § 512(c) (West 2014). ↩
- 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). ↩
- Disney Enterprises, Inc. v. Hotfile Corp., 798 F.Supp.2d 1303, 1308 (S.D. Fla. 2011) (internal citations and paragraph break omitted). ↩
- New York Times Co., Inc. v. Tasini, 533 U.S. 483 (2001). ↩
- 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). ↩
- Arista Records LLC v. Usenet.com, Inc., 633 F.Supp.2d 124, 149 (S.D.N.Y. 2009). ↩
- Id. (internal quotations, citations, and brackets omitted). ↩
- See Mark Bartholomew & Patrick F. McArdle, Causing Infringement, 64 Vand. L. Rev. 675, 703 (2011). ↩
- Id. (internal quotations and footnote omitted). ↩
- 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). ↩
- 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.”). ↩