Polyvoracious: The Sony-Betamax Rule Lives!

So Why Didn’t it Apply to Cox Communications?

One thing that had been gnawing at me about Rightscorp’s¹Not really Rightscorp, since it didn’t own the copyrights, but it was surely calling the shots.$25,000,000²Not as impressive as it sounds. victory over Cox Communications: why didn’t the Sony-Betamax rule apply to Cox Communications’ plain-vanilla ISP service? The answer, I realize now, is obvious, but only after you’ve stepped back for a moment. A recent decision where the rule did apply helps to explain why Cox Communications wasn’t eligible, although plain-vanilla services are generally going to be covered by Sony-Betamax. It’s the same reason why Cox Communications managed to lose its DMCA safe-harbor protection: its colossally horrible implementation of repeat-infringer policies.

The Common Sony-Betamax’s Natural Habitat

As typically explained, the Sony-Betamax rule is that providers of a service are not liable for contributory copyright liability merely for providing the service, if the service has “substantial non-infringing uses.”³“Pursuant to the Sony-Betamax rule, a defendant who distributes a product that materially contributes to copyright infringement will not be liable for contributory infringement if the product is ”widely used for legitimate, unobjectionable purposes“ or is ”merely…capable of substantial noninfringing use.” Arista Records v. Lime Group (S.D.N.Y. 2011) (emphasis in original). A recent decision out of the Southern District of New York shows the rule in action.

In BWP Media USA v. Polyvore, the owner of copyrights in photographs sued Polyvore, the operator of a website that “allows users to create free online accounts to upload, create and share photographs and other images.” But that’s not all Polyvore does. Polyvore also “offers a online tool called the ‘Clipper’ that allows users to ‘clip’ images from other webpages and collect these images on Polyvore’s platform.” The use of the Clipper causes at least one copy of the photograph to be copied onto Polyvore’s server.

Stand back a bit, and you begin to see the problem. Photo by Bit Boy, of a 2006 installation by Banksy titled “Barely Legal,” licensed under Creative Commons 2.0 License. (This is meant to be a statement about global poverty. I guess the joke is that it’s not that obvious. Also, that’s a real elephant.

The rights holders brought claims for direct infringement, contributory infringement and vicarious infringement. On summary judgment, the court dismissed all claims. The claim for contributory infringement was dismissed based on the Sony-Betamax rule. The court defined the rule as: “[A] defendant who distributes a product that materially contributes to copyright infringement will not be liable for contributory infringement if the product also is ‘widely used for legitimate, unobjectionable purposes’ or is ‘merely … capable of substantial noninfringing use.’” (quoting Sony-Betamax; emphasis in original).
To the court, this was not a difficult issue. In a two-paragraph analysis, it concluded:

Here, even assuming that Polyvore had knowledge of the infringing activity and materially contributed to copyright infringement, the Sony-Betamax rule shields Polyvore from liability. As noted above, the Clipper tool allows users to clip images from anywhere online, copyrighted or not. These features make Polyvore’s system, at the very least, “capable of substantial noninfringing use.”

Let’s compare this “Clipper” tool to plain-vanilla ISP services, like Cox Communications’. The Clipper tool allows you to make a copy of works that are probably, but not necessarily, under copyright, and probably, but not necessarily, unauthorized. An ISP lets you have access to the Internet, where you can read news, play games, send emails, make phone calls, make posts on Facebook, download software that you’ve paid for, and download copyrighted material without permission. Which do you think has more “substantial non-infringing uses”?

Yet Polyvore’s Clipper is OK, but Cox Communications’ ISP is not? Sure, Cox Communications “implemented” a deceptive joke of a “repeat infringer” policy, which is how it lost its DMCA safe-harbor protection in the first place. But you don’t need any safe harbors if you aren’t liable in the first place. And aren’t ISPs far more valuable than “clipper” tools?⁴The short answer: yes. That’s why Congress granted ISPs their own flavor of DMCA safe harbor protection, and made it so it was dead easy to qualify for. Cox managed to find a way to disqualify itself.

A Habitat the Common Sony-Betamax Can’t Survive In

A common mistake in legal analysis is to get caught up in the exciting details, and forget about the fundamentals. The Sony-Betamax rule was propounded by none other than the U.S. Supreme Court—twice. And Supreme Court copyright decisions are rare enough! And it goes to the heart of the balance between intellectual-property protection and innovation, something everybody understands, even if they don’t agree on how the balance should be struck. It’s easy to focus exclusively on it. But the Sony-Betamax rule fits into a wider framework; it’s just the shiniest, prettiest part of the framework. It pays to start your legal analyses at the beginning, instead of jumping to the exciting end.

To start at the beginning: Contributory copyright infringement has three components: (1) direct infringement by someone else; (2) knowledge of the infringement; and (3) material contribution to the infringement. Traditionally in law, there are two kinds of “knowledge”: actual knowledge, and “constructive” knowledge. This is usually expressed as: “knew or should have known.” “Constructive” knowledge is stuff you “should have known,” even if it can’t be proved you actually knew it.

When parties try to prove knowledge, they’re nearly always trying to prove constructive knowledge. That’s because actual knowledge is nearly always impossible to prove. Defendants are pretty unlikely to testify, “Oh, yeah, sure, I knew it.” Proof of actual knowledge requires a smoking-gun level of proof, which is rare, and when it’s available, it typically leads right to settlement. So, it’s easy to overlook the effect of actual knowledge.

In the context of contributory copyright infringement, actual knowledge is even harder to prove. Consider: What if you told Sony that your neighbor has been regularly using his Sony Betamax to pirate TV shows for re-sale, and further, that he plans to record The Equalizer (the 1980’s TV series, natch) tomorrow evening (and not just for “time shifting”)? Sony couldn’t do anything about it. Thus, actual knowledge requires proof that the defendant knew about the actual act of infringement, and that the defendant could do something about it. In Sony’s case, that proof was impossible. In the case of, say, a file-sharing service, it would be possible but still quite difficult. You’d need to prove that the service knew about actual subscribers who were regularly infringing copyright content, and the service had the ability to stop the infringement (such as by kicking the subscriber off the network) but failed to exercise that ability. Good luck with that.

The type of knowledge at issue in Sony-Betamax was constructive knowledge. At issue was the VCR devices that Sony manufactured and placed in the stream of commerce. The rights holders were able to prove that some users were able to use the VCRs for infringing purposes, and that Sony knew it. This was not actual knowledge, however. It was constructive knowledge because Sony didn’t know exactly who was mis-using the VCRs (and neither, really, did the rights holders), only that such mis-use occurred. Even if Sony knew, there wasn’t much Sony could do to stop it. Once the direct infringer bought the device, it was out of Sony’s hands. I don’t think Sony could even has sued the users on any theory (though the rights-holders could have, in theory).

The Sony-Betamax rule was, therefore, an exception to constructive, not actual, knowledge. If you have constructive knowledge that your product is being used (or will be used) in a manner that infringes copyright, that constructive knowledge won’t be imputed to you if your product has substantial non-infringing uses. There is a policy reason behind this rule. An injunction would stop law-abiding folks from enjoying new technology, while damages would effectively be a tax on those folks (assuming the damages are passed along to the consumers).

Later, the Supreme Court revisited the rule in Grokster. That case involved a file-sharing service that, theoretically, could have non-infringing uses, but, practically, was used almost exclusively to infringe. What was more, the defendants not only knew this, but encouraged and promoted infringing uses. Unauthorized sharing of copyrighted contented was the main selling point of the service. The Ninth Circuit, following Sony-Betamax, held that the defendants were not liable for contributory copyright infringement because the files that got shared could—and sometimes were—non-infringing. The Supreme Court reversed, noting that the Sony-Betamax rule shouldn’t be applied so broadly. It then held that there was another way of imputing constructive knowledge to the defendants: inducement, i.e., encouraging third parties to use a product to infringe. And the Sony-Betamax rule did not apply to this type of imputing constructive knowledge. This meant that Sony-Betamax was not an exception to constructive knowledge generally, but only to a particular flavor of constructive knowledge.⁵And who knows? Maybe there are other flavors of constructive knowledge.

Thus, the Sony-Betamax rule applies to Polyvore’s “Clipper” tool because the copyright holders’ case was based on imputing constructive knowledge to Polyvore that the “Clipper” tool was used for infringement. It did not apply to Cox Communications because Rightscorp’s case was based on Cox Communications’ actual knowledge of infringement. The same proof that wiped away Cox Communications’ DMCA safe harbor protection also proved its actual knowledge of specific acts of infringement. This proof not only identified regular infringers, but it showed that Cox Communications had the ability to stop the infringement but didn’t.
The rights holders in Polyvore, by contrast, barely took any discovery in advance of Polyvore’s motion for summary judgment.⁶They focused on proving direct infringement, but that claim failed because they couldn’t prove volitional conduct. Unable to prove actual knowledge, they were sunk.

Safe from the Predations of the Common Sony-Betamax, Rightsholders Nevertheless Find Find Little to Eat in the Habitat.

While proof going to Cox Communications’ failure to reasonably implement a repeat-infringer policy, and proof going to its actual knowledge of infringement, had common sources, they are not coextensive. This is to say, just because you’ve proven actual knowledge doesn’t mean you’ve eliminated an ISP’s DMCA safe harbor. It should be much harder to prove an ISP failed to reasonably implement a repeat-infringer policy than it is to prove actual knowledge. Cox Communications, however, made it easy by not only implementing a deceptive repeat-infringer policy, but by fulsomely documenting it.

But notice an important limitation when proving contributory infringement with actual knowledge: liability extends only to those acts of infringement that you had actual knowledge of at a time you could do something about it. When rights holders went after Sony in the 1980’s, they were hoping to enjoin all future sales of the Betamax, as well as collect damages for all past sales. When rights holders went after Grokster, they were seeking to shut it down (and get whatever damages they could). But Rightscorp’s damages were limited to acts of infringement by Cox subscribers whose infringing activities Cox actually knew about.⁷This still leaves the mystery of how the jury calculated damages, since the evidence about the actual acts of infringement was kind of sketchy. That wouldn’t be a reason to deny liability, though, unless the damages were completely incalculable. Once liability is established, we don’t let wrongdoers off the hook just because damages are hard to prove with certainty and precision. And an injunction—shutting down Cox’s entire network—was never an option, since that would have punished the vast majority of Cox’s subscribers, who were generally law-abiding.

This goes some ways to explaining (1) why Cox was liable for contributory copyright infringement, but (2) the damages were kind of pathetic. This also shows why the policy reason behind the Sony-Betamax rule—to encourage innovation—doesn’t apply where actual knowledge if proven. Since contributory infringement based on actual knowledge doesn’t threaten the device or service as a whole, there’s little threat to innovation. And to the extent there is, Congress made extra sure by implementing a special safe harbor for ISPs. And it goes some ways to explaining why the Sony-Betamax rule is often described in legal opinions has having broader application than it does. Since proof of actual knowledge is so rare, the broad description is accurate over 90% of the time.

Thanks for reading!

Rick Sanders

Rick is currently General Counsel for Software Freedom Conservancy. Previously, he has been practicing law as an intellectual-property litigator since 2000.