Why Arguing That Google Is In The Best Position To Stop Infringement Is Wrong

from the it's-not-and-it-makes-no-sense dept

Wed, Apr 20th 2011 9:44am — Mike Masnick

While we did cover Google/YouTube's own filing in Viacom's appeal of YouTube's victory at the district court concerning its liability for infringing videos on the site, we did not cover the various "friend of the court" (amici) briefs filed by a variety of folks in support of YouTube's argument. If you'd like to see them all, Michael Barclay has put them all together in a single page, including his own (excellent) analysis of the key points raised in the briefs. Eric Goldman also does a nice job discussing many of the briefs and the key issues raised by them. If you want to understand what people are saying, read those two pages, and you should be in great shape (though, if you were reading both of their blogs already -- as you should -- you'd already be ahead of the game).

However, I did want to focus in on one particular brief, by law professor Michael Carrier, author of an excellent book on innovation (and how the law interferes with innovation), addressing the issue of "least cost avoidance," when it comes to stopping copyright infringement (pdf and embedded below).

It's one of the key arguments often made by content industry folks in arguing that tech companies and ISPs should be responsible for somehow policing their systems for infringement: that those service providers are in the best position to stop the infringement. For many years, we've always explained why this didn't make sense from a pure liability standpoint. These service providers aren't the ones doing the infringement and thus -- fundamentally -- it makes little sense to pin the liability on a party that didn't perform the law-breaking action. The response from the content industry is often to point to the concept of the "least-cost-avoider principle," which shows up in various aspects of the law, in which courts will sometimes shift liability around to put it on those in the best position to stop the law breaking at the lowest cost. And, in fact, this was found in some of the arguments supporting Viacom.

Carrier's amicus brief does a wonderful job debunking the argument. He does brush over a key point, that I wished got more attention: that it's incredibly unlikely that the service providers really are the least-cost-avoiders in these circumstances, since they have no clue what's really infringing and what's not (as demonstrated by Viacom's own confusions over staff members uploading promotional clips to YouTube...). However, for the purpose of making his point, he goes even further, suggesting that even if it's true that YouTube would be the least-cost-avoider, such a least-cost-avoider makes no sense in the context of copyright and innovation.

The brief points out that least-cost-avoider rules are used in tort law, in particular involving accidents -- which are situations in which no beneficial externalities are created as a result:

In these settings, lowering the costs of preventing accidents makes sense. Accidents do not offer any benefits for society. And the only downside to requiring actors to prevent accidents is that their costs will increase.

More important, requiring parties to take measures to reduce accidents will not have detrimental effects on third parties. No third parties will suffer collateral consequences if a driver is forced to slow down to the speed limit. Or if a barge fortifies its hull to prevent oil spills. Or if a manufacturer reduces the use of asbestos in its products. In short, the harms from the application of the least-costavoider principle in tort law are observable and do not threaten adverse unanticipated effects across other sectors of the economy.

But that's not the case in copyright law. In copyright law, infringing on copyright may certainly hurt some parties, but it also creates positive externalities and can lead to certain innovations. In fact, entire books have been written on how infringement is often a driver of innovation. Thus, taking the "least-cost-avoider" principle (even assuming it's true, which remains in question) could potentially cause serious harm:

The least-cost-avoider standard would result in innovative technologies becoming less useful and more cumbersome. Application of such a standard would have made some of today's leading technologies just a shadow of the invaluable innovation they ultimately became. For example, courts could have required photocopier manufacturers to modify their copiers to prevent the copying, absent a copyright owner�s approval, of "any document displaying a ©."

The brief goes on to note that, to copyright holders, pretty much every disruptive innovation first appears as some evil form of infringement that must be stopped. Applying this kind of principle would likely have stifled all sorts of beneficial technologies before they could have gotten off the ground.

It is understandable--if short-sighted--to not recognize the benefits of disruption. But it is not appropriate to reengineer the law to block disruptive innovation. The Constitution promotes the "Progress of Science and useful Arts." It does not guarantee that copyright owners are entitled to protect their existing business models against the onslaught of innovation.

The disruptive innovation unleashed by new technologies reveals the difference between tort law and copyright law in the application of the least-cost-avoider standard. There is no legitimate fear in tort law about eliminating valuable activity from society by overdeterring fast drivers or brittle boat hulls filled with gasoline. In contrast, there is monumental concern with stifling disruptive innovations by saddling them with the burdens of copyright owners.

In other words, even if we assume that service providers are the lowest cost providers of stopping infringement, that doesn't mean that's the best way to "Promote the Progress of Science and useful Arts." There's a lot more in the brief, but wanted to post this as a key piece of the puzzle in responding to claims that it makes sense to put the burden on service providers based on such a principle. While I tend to disagree that service providers are the least cost avoiders in the first place, even if they are, the potential net harm on innovation they can cause should be a huge concern in shifting liability.

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: copyright, least cost avoider, liability, third party liability, youtube
Companies: google

20 Comments

If you liked this post, you may also be interested in...

Reader Comments

Subscribe: RSS

View by: Time | Thread

Anonymous Coward, 20 Apr 2011 @ 9:59am

Major labels have done pretty much all they could to obfuscate who owns a copyright. Just look at the trouble that the RIAA radar site has with determining what is and is not a major label protected copyright.

One version by an artist is and another isn't. Major labels did not clear the air by awarding labels to their better selling artists making it look like the major labels were no longer the representative holding the copyright.

Just this one issue alone makes it near impossible to determine what is to be removed and what is ok. With digital files, there are no fences around the IP. Nothing in the metadata that can not be changed to say whatever someone wishes, including the name on the file. While the labels may wish it or not, there is no copyright notice on the digital file that can not be removed, making it even more difficult to identify.

Allowing the labels to provide and update a block list results in those same labels putting up ticklers to try and advertise on the cheap. While doing that they leave the users of such data out in the blue without a clue if it is or is not infringing. This was precisely the case with ICE ceasing domains and using the excuse that a song here and there were infringing after being provided by the labels themselves.

We have tools in place to deal with infringement. Let them use the tools they themselves got put into law to deal with it.
[ link to this | view in thread ]
Chris Rhodes (profile), 20 Apr 2011 @ 10:00am

Mangled Paragraphs
Or if a barge asbestos in its products. In short, the harms from the application of the least-costavoider principle in tort law are observable and do not threaten adverse unanticipated effects across other sectors of the economy fortifies its hull to prevent oil spills. Or if a manufacturer reduces the use of

Should be:

Or if a barge fortifies its hull to prevent oil spills. Or if a manufacturer reduces the use of asbestos in its products. In short, the harms from the application of the least-costavoider principle in tort law are observable and do not threaten adverse unanticipated effects across other sectors of the economy.

I think.
[ link to this | view in thread ]
jhn, 20 Apr 2011 @ 10:16am

There's no reason to get rid of the least-cost-avoider principle as applied to copyright. You just have to account for *all* costs, including the cost to a service provider of having to reduce the usefulness of its service.

Additionally, if the cost to X in preventing a harm to Y is greater than the harm to Y in the first place, and if you find that X is the least cost avoider, then the correct response is to not avoid the cost at all. To do so would be a deadweight loss to society. Similarly, if the negative externalities of avoiding a cost are greater than the cost, the cost should not be avoided. (And of course, you have to take into account positive externalities on both sides of the ledger as well, to determine whether a cost should be avoided.)
[ link to this | view in thread ]
Brendan (profile), 20 Apr 2011 @ 10:53am

Re: Mangled Paragraphs
Just came by to post the exact same correct. I agree with you.
[ link to this | view in thread ]
Anonymous Coward, 20 Apr 2011 @ 11:47am

Frankly, I find the amicus far from compelling, and have to wonder why it was even submitted in the first place.

The question before the appeals court is how should 512(c)(1)(A)(ii) be interpreted? This is what is referred to in some circles as the "Red Flag" provision.

I could find nothing in the amicus that shed any light on how this question should be resolved. If fact, in some regards the amicus is misleading in suggesting that the case involves a rights holder shedding all costs of compliance onto the backs of service providers, which is simply not the case.

Takedown procedures remain intact, and the burden (both time and cost) of such procedures rest almost entirely on rights holders, with the burden on service providers being quite minimal and easily handled using a virtually automated system.

Here, however, we are dealing with the "Red Flag" provision and asking an entirely different question. At what point is apparent infringment so ubiquitous that a service provider should be called upon to shoulder a greater share of the burden?

I do not profess to know the answer, but in my view the district court opinion places all significant responsibility and associated costs on rights holders, rendering the "Red Flag" provision virtually meaningless.

In the non-internet world hosts are held accountable in certain limited instances. Is it a fair solution in the internet world that a different set of rules should apply?
[ link to this | view in thread ]
FUDbuster (profile), 20 Apr 2011 @ 12:01pm

Re:
I agree. Arguing as a general matter that we shouldn't shift the burden onto service providers does nothing to help the court interpret the DMCA as it applies in this case. It's a policy argument that's only indirectly on point, and it's better suited to make that argument to Congress, not the courts.
[ link to this | view in thread ]
Hephaestus (profile), 20 Apr 2011 @ 12:04pm

Re:
"Major labels have done pretty much all they could to obfuscate who owns a copyright ... "

There is only a single solution to this problem. Force all rights holders to register and pay a registration fee for all copyrighted material. If its not register with in a certain amount of time, have it fall into the public domain.
[ link to this | view in thread ]
Joel Coehoorn, 20 Apr 2011 @ 12:09pm

Normally I agree pretty well with what you're saying, but I think there's a subtle flaw here. You make two points:

1. That "least-cost-avoider" shouldn't apply in the first place.
2. Even if it does, Google is not the least cost.

I'm okay with point #1. I think the analysis there is spot on. It's #2 I take slight issue with. The valid part of the argument is that Google is in a poor position to make judgments about what is infringing. But just because Google's position is bad, that doesn't necessarily mean that it's worse than anyone else's. Google may still be left holding the least-cost stick even though it's position is bad, if it turns out that it's much worse for everyone else.

Hopefully this doesn't matter. If the courts take the correct view on point #1, then the flaw in point #2 becomes moot. But if after all is said and done courts decide that this an appropriate least-cost case, it is very likely that Google (and other similar operators) will come out holding the bag on this.
[ link to this | view in thread ]
Mike Masnick (profile), 20 Apr 2011 @ 12:26pm

Re: Mangled Paragraphs
Oops. Not sure how that happened. Fixed.
[ link to this | view in thread ]
Anonymous Coward, 20 Apr 2011 @ 12:29pm

Re:
It's pretty hard to see how many other people are worse off at stopping infringement than service providers like Google. A basic blacklist would be cheap but cost Google (or other provider) and a significant number of other people in lost views and make it impossible to find a lot of legitimate content, while a "foolproof" fingerprint program would be overly costly to implement and use up a ton of resources just to run. Rights holders need only pay some number (probably relatively high, but need not be terribly so) an hourly wage to look at possible links to potentially infringing content to see if it's actually infringing. A job as simple as looking at links can't be worth more than $10-20/hour and most of the hard work doesn't require 24/7 computer clusters churning away.
[ link to this | view in thread ]
Mike Masnick (profile), 20 Apr 2011 @ 12:32pm

Re:
Frankly, I find the amicus far from compelling, and have to wonder why it was even submitted in the first place.

Given your general support of all things copyright, I don't find this surprising.

The question before the appeals court is how should 512(c)(1)(A)(ii) be interpreted? This is what is referred to in some circles as the "Red Flag" provision.

Indeed. And that's been discussed elsewhere.

I could find nothing in the amicus that shed any light on how this question should be resolved. If fact, in some regards the amicus is misleading in suggesting that the case involves a rights holder shedding all costs of compliance onto the backs of service providers, which is simply not the case.

First of all, the key point was responding to other amici briefs which focused on the least-cost avoider issue, as part of the argument why a broad interpretation of the red flag provision (contrary to Congress' own statements on the issue) makes the most sense.

Carrier is seeking to counter that argument.

Separately, he does go include an analysis of how the DMCA is structured such that this interpretation doesn't make sense.

However, my point in discussing this was to highlight that larger argument, concerning the reliance by you and your friends on "least cost avoider" principles, not to comment on the Viacom case specifically.

Takedown procedures remain intact, and the burden (both time and cost) of such procedures rest almost entirely on rights holders, with the burden on service providers being quite minimal and easily handled using a virtually automated system.

Ignoring the burden on free speech and users, of course.

I do not profess to know the answer, but in my view the district court opinion places all significant responsibility and associated costs on rights holders, rendering the "Red Flag" provision virtually meaningless.

This is simply not true and you know it. There are red flag provisions if they are aware of specific cases of actual infringement. Those remain intact. The problem is that Viacom has presented no evidence that Google was aware of any such specific instances.

In the non-internet world hosts are held accountable in certain limited instances. Is it a fair solution in the internet world that a different set of rules should apply

Hilarious. So you claim that this brief does not apply... and then you conclude by making the very argument the brief debunks.
[ link to this | view in thread ]
Anonymous Coward, 20 Apr 2011 @ 1:17pm

Re: Re:
It's an amicus brief. They're supposed to talk about policy, which can be a tiebreaker when the law is close.
[ link to this | view in thread ]
FUDbuster (profile), 20 Apr 2011 @ 1:53pm

Re: Re: Re:
I don't see how the court would need to look at these broader public policy arguments to answer the questions before it. Sure, you can make the arguments, but I don't how they would actually guide the court in this case. Congress already established the policy with regard to red flag knowledge. I'll be really surprised if the court's analysis includes a policy argument. I expect it will be rooted on statutory construction, the congressional record, and caselaw.
[ link to this | view in thread ]
pixelm1 (profile), 20 Apr 2011 @ 3:42pm

This is a totally bogus argument - it essentially argues against copyright law, not least cost avoider. (By the way, there is lots of collateral damage from stuff like rules against asbestos - but congress has made the judgment that the benefit to public safety outweighs other benefits, like low cost fire prevention). The fact that you're protecting someone else is why you need rules of liability law at all. Otherwise, you're just protecting yourself and you don't need the law to do the cost benefit. On the other hand, driving carefully saves pedestrians who didn't hurt anyone and it's lower cost to ask drivers to be careful than to protect every pedestrian with concrete suits. Copyright owners don't benefit from infringement - they are innocent bystanders getting hurt by it. In contrast, service providers do benefit. In some cases, you might be right that they aren't least cost avoiders, but in some cases they are. And that's the right inquiry to make in each case, not the argument that even if they can avoid infringement, they shouldn't.
[ link to this | view in thread ]
Anonymous Coward, 20 Apr 2011 @ 3:54pm

Re: Re:
"There are red flag provisions if they are aware of specific cases of actual infringement."

In which case they would have actual knowledge. Thing is, this is what is covered by 512(c)(1)(A)(i).

While somewhat inapt, I would term the "Red Flag" provision akin to constructive knowledge.
[ link to this | view in thread ]
Mike Masnick (profile), 20 Apr 2011 @ 5:00pm

Re:
This is a totally bogus argument

I don't see how it's bogus at all.

it essentially argues against copyright law, not least cost avoider

Again, I don't see that at all. It's talking about the proper placement of liability. It says nothing against copyright.

The fact that you're protecting someone else is why you need rules of liability law at all.

No, that's simply untrue. Liability laws are so that you place *proper* liability, not the least cost liability.

Copyright owners don't benefit from infringement - they are innocent bystanders getting hurt by it.

Again, that's simply untrue. Many, many copyright owners do benefit from infringement. After all, why did Viacom upload so many videos to YouTube itself, pretending they were infringing.

Do you have an answer for that?

In some cases, you might be right that they aren't least cost avoiders, but in some cases they are. And that's the right inquiry to make in each case, not the argument that even if they can avoid infringement, they shouldn't.

That's not the argument that's being made here at all.
[ link to this | view in thread ]
Michael Barclay (profile), 20 Apr 2011 @ 9:00pm

What prompted Prof. Carrier's brief
Additional response to Anonymous Coward: One of the briefs supporting Viacom was by Stuart Brotman et al., and argued that tort law's "least cost provider" principle should be applied to YouTube. Brotman's brief is here:
http://mbarclay.suekayton.com/IPDuckDocs/2010-12-10_Brief_of_Amici_Curiae_Stuart_Brotman_et_a l.pdf
Professor Carrier shows why Brotman's analysis should not apply in DMCA-type situations. YouTube's brief, and many of the amicus briefs, discussed other issues in the case.
[ link to this | view in thread ]
Anonymous Coward, 20 Apr 2011 @ 11:56pm

Re: What prompted Prof. Carrier's brief
"Least cost" aside, Mr. Carrier makes a perfectly rational argument that Congress struck a balance with the DMCA. Where he goes astray, in my view, is that he then argues for a construction of the DMCA's "Red Flag" provision that would emasculate and render it mere surplusage within the statutory scheme.
[ link to this | view in thread ]
Michael, 21 Apr 2011 @ 3:57am

Barge Pirates!
"Or if a barge fortifies its hull to prevent oil spills"

This would clearly ruin the oil-spill clean-up industry and should be stopped at all costs!

What about the poor corn farmers that supply canned corn to the families of the guys sent out to clean up oil spills? Is nobody considering how they will be impacted?
[ link to this | view in thread ]
FUDbuster (profile), 21 Apr 2011 @ 7:12am

Re: Re: Re:
While somewhat inapt, I would term the "Red Flag" provision akin to constructive knowledge.

I think that's a good analogy. Red flag or apparent knowledge isn't about what you know, it's about what you should know. It's where the law imputes you with the knowledge that you should have known--that you would have known had you not turned a blind eye. The DMCA clearly delineates two levels of knowledge: actual and apparent. Apparent knowledge is more than general knowledge, but less than actual knowledge. It's willful blindness.
[ link to this | view in thread ]