No, if Aereo is a "Cable System," That Doesn't Mean That Netflix is, Too

Given that the Supreme Court said that Aereo has an “overwhelming likeness” to cable systems, Aereo is now quite reasonably arguing that for copyright purposes it is one–and is therefore entitled to the statutory license that cable systems get. Under this theory, if Aereo pays a relatively small fee to the Copyright Office, it does not infringe copyright by retransmitting broadcast programming. There is caselaw in the Second Circuit that says Aereo can’t do this–a case I think was wrongly decided–but that’s neither here nor there, as the point is that Aereo arguably overrules that decision.

The statutory license in question–17 U.S.C. § 111, is pretty complicated, but it boils down to this: If you’re a cable system, you don’t negotiate for copyright rights for broadcast programming. Instead, you follow FCC rules. If you follow FCC rules, you qualify for the license. If you don’t follow FCC rules, you don’t–which means that if, for example, Time Warner Cable began retransmitting broadcast signals in violation of FCC rules, not only would it be violating those rules, but also copyright, since the license wouldn’t apply. But provided it follows FCC rules, it doesn’t need to negotiate copyright permissions for the programming it carries.

FCC rules right now say that multichannel video programming distributors, a category that includes cable systems, do have to get permission to retransmit TV stations, which means they usually have to pay. But this is not a copyright license. Often, the local stations cable companies negotiate with don’t even have any rights to the programming they carry, beyond the right to air it. They couldn’t grant copyright permission even if they wanted to, but because of the statutory license system, it’s not necessary.

This is all a very complicated system that I’m not a big fan of, but let’s leave that aside for today. Let’s get back to Aereo. Aereo is claiming it’s a cable system for copyright purposes, which means that provided it is following FCC rules, it doesn’t violate copyright law. “But wait a minute,” you might be asking, “aren’t they not following FCC rules, if they’re not getting permission from the stations themselves?” Well, maybe, maybe not.

The FCC’s retransmission rules only apply to cable systems, satellite TV companies and so on–the aforementioned “multichannel video programming distributor” (MVPDs). The statute that first codified the obligation of MVPDs to get consent is clear: “No cable system or other multichannel video programming distributor shall retransmit the signal of a broadcasting station, or any part thereof, except … with the express authority of the originating station [or some other random reasons that are not relevant here].” So if I were to retransmit a TV station without consent I might be violating any number of laws including copyright–but not this particular law. I am not an MVPD.

But here’s the catch–“cable” and “MVPD” are defined in the Communications Act, and it’s the FCC that decides what they mean in a communications context. Even if Aereo is considered to be a cable system for copyright purpose, it might not be one for FCC purposes. What might be a “cable system” for copyright law purposes might not be one for communications law purposes. This is not unusual. Words often mean different things from one law to another–especially when different laws actually provide different definitions! The Copyright Act defines “cable” like this:

A “cable system” is a facility, located in any State, Territory, Trust Territory, or Possession, that in whole or in part receives signals transmitted or programs broadcast by one or more television broadcast stations licensed by the Federal Communications Commission, and makes secondary transmissions of such signals or programs by wires, cables, microwave, or other communications channels to subscribing members of the public who pay for such service.

But the Communications Act says a cable system is

a facility, consisting of a set of closed transmission paths and associated signal generation, reception, and control equipment that is designed to provide cable service which includes video programming and which is provided to multiple subscribers within a community[.]

These definitions are different enough that it is possible for Aereo to be a cable system under one law and not another. To the extent that the Communications Act language is ambiguous, it’s up the the FCC (and not courts) to adopt a reasonable interpretation.

But even that is not the most relevant question. We need to know whether Aereo is an MVPD, because if it’s an MVDP and not a cable system, it still needs to get permission to retransmit a TV station. Well, MVPD

means a person such as, but not limited to, a cable operator, a multichannel multipoint distribution service, a direct broadcast satellite service, or a television receive-only satellite program distributor, who makes available for purchase, by subscribers or customers, multiple channels of video programming[.]

This means that the relevant question should be whether Aereo makes “channels” of programming available. I should stop hiding the ball–I think there’s a very good argument that it does, and I’ve argued (more) that systems such as Aereo (and Sky Angel) should be considered MVPDs by the FCC. (However, they are probably not “cable” for FCC purposes.) But online services are currently not considered MVPDs by the FCC, which means that if Aereo is a cable system for copyright purposes, it’s not violating any FCC rules by retransmitting broadcast signals, which means the statutory license applies. Under this construction, if Aereo pays the small fee associated with the statutory license, its service is legal.

Now, let’s assume that Aereo is considered both the be a cable system for copyright purposes, and also an MVPD (and maybe also a cable operator) for Communications Act purposes. Contrary to a common assumption, I think there would be no implications for most online video operators.

First, cable for copyright purposes. An on-demand system like Netflix does not receive signals from broadcast stations. Thus, it can’t be a cable system for copyright purposes–but Aereo can, since it does.

Second, cable for communications purposes. I don’t think that “a facility, consisting of a set of closed transmission paths and associated signal generation, reception, and control equipment” accurately describes online video services–they don’t have “closed transmission paths,” but share transmission paths with other Internet services. (Which does not mean they are somehow less “secure.”) Furthermore, most online providers offer their service nationwide, not to a “community.” (As Gus Hurwitz notes, Aereo does limit its retransmissions to their original local markets.) Finally, given the structure of the Communications Act, there’s good reason to think that if a service does’t fit the definition of MVPD, it cannot be a cable service. “Cable” is pretty clearly set out as a subcategory of MVPD. Thus, I think it’s unlikely that Netflix could be a cable system for communications act purposes, and still unlikely (though less so) that Aereo-like services could be cable systems for communications act purposes.

Third, multichannel video programming distributor. Here, while I’ve disagreed with the FCC on what a “channel” means, the Communications Act makes at least thing clear: On-demand programming is distinct from “channels.” The Act draws a distinction between “interactive on-demand programming” and “video programming prescheduled by the programming provider.” As PK argued at great length, the word “channel” in the Communications Act means what viewers think it means–a linear stream of prescheduled video programming–in other words, a “TV channel,” whether it’s broadcast (WNBC) or cable (ESPN). The term as used in the definition does not include on-demand services (and, contrary the FCC’s Media Bureau, it shouldn’t only mean “transmission path”). Normal cable systems offer multiple channels of programming, and so does Aereo. Most online video services do not. Thus, classifying Aereo as an MVPD would affect Aereo and Aereo-like services but none of the most popular online services like Netflix, Hulu, and iTunes. In other words, services like Netflix and Hulu and Amazon Instant Video cannot be considered multichannel video programming distributors because they do not offer any “channels” or programming.

I find the legal implications of Aereo #1 (the public performance theory) and Aereo #2 (the cable theory) fascinating: they’re a Lemarchand’s box that open a door to obscure dimensions of copyright law that most practitioners and judges will never see in a lifetime. But the public policy should be clear: We should promote competition and new entry into the video market from new kinds of services. Interpreting the law in ways that would allow this (which also happens to be what Congress intended) will not have dire consequences for existing online video providers or subject them to new forms of regulation. Instead, acknowledging that Aereo ought to be able to follow the same rules as incumbent cable systems will benefit consumers and undo some of the uncertainty the Supreme Court created.

The Aereo Ouroboros

Here is the simplest way I think I can explain as to why the Aereo decision is circular. It hinges on the “volitional conduct” argument, and the Court’s use of loose analogical reasoning rather than attention to technical specifics of how systems operate and the potential audience for particular transmissions of content.

I am not asserting that I have some surefire way of determining volitional conduct in all copyright cases. Nor do I deny that a single one-to-one transmission cannot under some circumstances either be, or be part, of a public performance (the statute demands this). However, I think the Court’s “similarity” test is flawed, since reasoning by analogy between very different technological systems can have unpredictable results. Different people can reasonably see different systems as more or less similar–in the Cablevision case, for instance, broadcasters argued that the service is question was “just like” a video-on-demand service that would require a license, rather than “just like” a DVR that would not.


The Supreme Court’s analysis only works if you assume that Aereo is similar to a cable system instead of assuming that it is similar to an antenna. The court provides no grounds for making one assumption rather than the other; rather, it runs the analysis while assuming that Aereo is like a cable system, and the result of this anlaysis (that Aereo is infringing) is what seems to inform the Court’s view that Aereo is similar to a cable system to begin with. That is, there is no way to decide that Aereo has an “overwhelming likeness” to a cable system unless you are already assuming that it publicly performs.

Court’s analysis with the “It’s like cable” assumption.

Step One:

Aereo is the volitional actor because it is similar to cable systems (which are volitional actors).

Thus, “An entity that engages in activities like Aereo’s performs.”

Step Two:

Whether Aereo performs “publicly” depends on some relationship of privity between the viewers and the content.

When a home viewer records TV from over the air, the viewer has “privity” with those lawful fair use copies because the viewer is the volitional actor. (The Court does not make this argument expressly but I believe this would be the correct analysis under its test.)

But an Aereo user is not the volitional actor. So the Aereo user has no privity with the copies the Aereo system makes.

Step Three:

Aereo is therefore a public performer.

Alternate reality: Court’s analysis with the “It’s like cable” assumption.

Step One:

Aereo is not a volitional actor because it is similar to other equipment providers (which are not volitional actors).

Thus, an entity that engages in activities like Aereo’s does not perform.

Step Two:

A user cannot perform publicly to himself so Aereo cannot be secondarily liable.

Also, an Aereo user, as the volitional actor, has a relationship of privity with the recorded content, which is a further reason he cannot publicly perform.

Step Three:

Aereo does perform (and therefore cannot publicly perform). Its users do not publicly perform either, so Aereo cannot be secondarily liable.


I don’t think either of these two ways of looking at Aereo are very good. Whether or not Aereo seems “similar” to cable or an antenna should not have much bearing on the Court’s analysis. That’s because many systems are “similar” to Aereo (many cloud computing services) but not similar to cable TV. Where does the “similarity” analysis end? If Aereo is not allowed because it is similar to cable TV, how do you justify allowing systems that are similar to an unlawful service such as Aereo? Copyright is unavoidably technical and to keep a distinction between “public” and “private” that is consistent and can apply to many different kinds of services I think that the Court should have attended to the technical design of the systems in question more closely. I’ve never been very happy with Cablevision’s “one copy = private performance” test but it at least provides clear guidance.

Uh oh, Aereo

The Aereo decision is bad news for consumers, since it could take away a promising new model for watching free over-the-air television. Cord-cutters still have options. They can still access other online video services–and maybe some reconfigured version of Aereo–and watch broadcast TV with an antenna. But there’s no doubt that today’s opinion (especially when considered alongside ivi and Sky Angel) is good for the network/affiliate/cable status quo.

The outcome is not surprising, given the tenor of questions at oral argument. People are always asking me for my predication as to the outcome of a case where Public Knowledge filed an amicus brief, which is a bit awkward. Of course if you asked me on the record before today I would have told you that I was optimistic that a unanimous Supreme Court would simply adopt our brief. But like most observers I concluded that the Court was looking for a way to rule against Aereo but limit the collateral damage against the tech industry as a whole. However, I thought that the Court would do a better job. Instead, the Court uses reasoning that could apply very easily to any number of online services–file hosting, cloud lockers, even VPNs–as well as services that no one has even come up with yet. It then simply declares that Aereo is different, because it looks so much like cable. Unfortunately, only the Supreme Court’s actual legal reasoning is binding on future courts–the rest of the opinion just “dicta.” Judges can quote poetry and make jokes in their decisions, too. None of that is binding either. Dicta has the same legal status as legislative history–which is to say, it’s as binding as lower courts want it to be. A lower court is bound by a higher court’s legal holdings, not its stated intent.

On to a few of the legal specifics. Remember, the question the Court had to decide was whether Aereo was engaged in a public performance, when it provided each of its subscribers with an individual antenna that separately transmits broadcast programming to just that user. If Aereo is not the one “performing”–if it was the Aereo subscriber, instead–then Aereo wins. And even if Aereo is the one doing the performing, it would win if the performance was not “to the public.”

Many of the briefs in this case were worried that a ruling that says that Aereo “publicly performs” works could be applied to other online services, as well, which until recently everyone assumed were legal. They were right to worry because much of what the Court said today could easily apply to other kinds of services. This case gets very technical very fast and will give law professors a lot to do in years to come. The easiest way to see why the Supreme Court’s attempt to limit its holding to Aereo-like services is just to comment on the actual language it uses in the attempt. I’m not confident that the various ways it tries to distinguish Aereo from other services will work very well.

Here’s the Court:

We agree that Congress, while intending the Transmit Clause to apply broadly to cable companies and their equivalents, did not intend to discourage or to control the emergence or use of different kinds of technologies. But we do not believe that our limited holding today will have that effect.

Fair enough.

For one thing, the history of cable broadcast transmissions that led to the enactment of the Transmit Clause informs our conclusion that Aereo “perform[s],” but it does not determine whether different kinds of providers in different contexts also “perform.”

Here, the Court is addressing the concept of “volitional action.” Making a copy of a work might infringe copyright–but it matters who is doing the copying. The Court (and the dissent) use the copy shop example–the customers of a copy shop, not the copy shop itself, are the ones making copies. So if there’s copyright infringement, it’s the customers, not the copy shop, that are liable. (You can get to the copy shop via some theory of secondary liability, but first you need to establish who the direct infringer is.

But the Court does not offer any clear test for how to figure out who the “actor” is in a given circumstance. Instead, it suggests that future courts will have to undertake a historical analyses of some kind. It has answered the “volition” question for Aereo but hasn’t provided any certainty for other kinds of services.

For another, an entity only transmits a performance when it communicates contemporaneously perceptible images and sounds of a work.

I don’t see how this excludes many online services. A cloud locker service that allows people to store and stream back their own music or movies “communicates contemporaneously perceptible images and sounds.” Heck, if multiple people store and play back copies of the same work at the same time, they’re even simultaneously contemporaneously perceptible images and sounds! While this language may be comforting to some kinds of online services many others appear to meet this test.

Further, we have interpreted the term ‘the public’ to apply to a group of individuals acting as ordinary members of the public who pay primarily to watch broadcast television programs, many of which are copyrighted. We have said that it does not extend to those who act as owners or possessors of the relevant product.

The Court is suggesting that a whether people count as the “public” for copyright purposes depends on the nature of their relationship to the work. This logic was in many of the briefs–notably, the Solicitor General’s. But it’s a circular argument.

When I record a TV show off the air, then that copy is lawful. You don’t need a license to tune in to a broadcast, and home recordings of broadcast content are accepted fair uses. So why is it any different if I use the antenna that Aereo provides to record a program from the airwaves?

If you assume for the sake of argument that Aereo is legal, then Aereo subscribers are in fact lawful “possessors” of the programming in question–just like the home viewer. Only if you start by assuming that Aereo is illegal does Aereo fail this test. In other words, the Court’s test is that “Illegal services are illegal. But legal ones are legal.” This is not very helpful.


The Court appears to have a reasonable policy position: That Aereo ought to be treated like other services that help users watch broadcast content. (“Aereo ought to be treated like other antenna sellers” is another reasonable policy position, of course.) But the way it got there is troubling, because it adopts reasoning that could be used against many different kinds of services, and its attempt to distinguish them from Aereo has many failings. The most likely outcome of this case is more litigation, and perhaps future trips to the Supreme Court. The Court should have ruled in favor of Aereo, and perhaps sent a stern note to Congress to write better laws. Instead, it’s issued a decision that decides very little–and that will provide ample ammunition for litigants on both sides of future copyright fights.

I will grant the Court one thing: The statute is not very clear. As the opinion states at one point, “the language of the Act does not clearly indicate when an entity 'perform[s]’ (or 'transmit[s]’) and when it merely supplies equipment that allows others to do so.” I think when copyright laws are ambiguous the tie ought to go to the public’s right to use and access content–particularly when, as here, the content is already available for free to anyone with an antenna. The Court decided, instead, that the tie ought to go to how it imagined that Congress (in 1976) would have thought about Aereo if anyone had asked it. This was probably the wrong call. Like in the Brand X case, I join Justice Scalia in dissent.