Mosaiko

Don’t Blame the Messenger

• Comments
• Email This Post
• Share on Facebook

By Cynthia Wong

Intermediary liability and protecting Internet platforms

When an Italian court held Google liable for a video uploaded by a third party to one of the Internet giant’s Web sites, it offered a stunning example of intermediary liability. Google, the platform host, was the intermediary between the content creator who made the video clip and the content consumers who viewed it. Other intermediaries include Internet service providers (ISPs), e-commerce platforms, and social media platforms like Orkut, Facebook, and YouTube. These services provide valuable open forums for user-generated content that are often free of charge and require minimal technical knowledge to use.

The open nature of these services also means that they can be used for ill as well as good. Governments wishing to censor free expression or address harmful behavior on the World Wide Web often seek to pressure, intimidate, or, more subtly, expose Internet intermediaries to legal liability for third party content. One way to prevent a citizen from posting videos of political dissent is to hold YouTube liable for material that users post.

The equities of intermediary liability often are not this simple. Sometimes even well-meaning governments are trying to circumscribe behavior that a deep social consensus in a given nation believes wrong: obscenity, defamation, hate speech, privacy violations (as in the Google Italy case) or criminal activity. One readily sees why the technique offers an attractive alternative: the intermediary often is large and easy to identify, while individual Internet users can be hard to find and at times are outside a particular government’s jurisdiction (although this was not the case in Italy).

Whatever the reason a government allows intermediary liability, the substantial harm to information flows and Internet growth that follows outweighs the perceived benefits. First, freedom of expression inevitably is limited. A social networking platform, for instance, that can be held liable for money damages when a third party posts objectionable content, will wish to screen content before posting. Intermediaries will err on the side of caution in deciding what users may post, especially when the laws defining “illegal content” are vague and overbroad or where the speech is unpopular. Far safer simply to remove disputed content than to challenge a removal demand in court. In many cases, the sheer volume and associated cost of this task will be too much for many platforms to bear and they could not offer their services at all.

Second, intermediary liability disrupts the free flow of information and services on the Internet and thus stifles creative innovation and economic development. Companies are less likely to invest in technologies that may expose them to liability. The world may never see tomorrow’s Twitters, eBays, or other startups holding the promise of lowering prices and better connecting global markets; or of new initiatives that might increase access to educational resources or in other ways spark broader, deeper, and more equitable economic development.

Approaches to Intermediary Liability

An early policy consensus emerged on the question of intermediary liability equities in the United States and the European Union. In the United States, two statutes address these concerns. Section 230 of the Communications Act generally immunizes intermediaries from a variety of claims arising from third-party content, among them negligence, defamation, and violations of civil rights laws and state criminal laws. Section 512 of the Digital Millennium Copyright Act affords online service providers a “safe harbor” from liability if they meet certain criteria, including removal of infringing material when notified by the copyright owner of its presence, known as a “notice-and-takedown” system.

The European Union similarly shields several kinds of intermediaries from a range of claims: “mere conduits” of information, “caching” services offering temporary data storage to facilitate onward transmission, and “hosting” services that quickly remove unlawful content upon becoming aware of it. Because the U.S. and E.U. policies do not generally oblige intermediaries either to monitor content on their services nor to investigate possible unlawful user activity, the policies help safeguard user privacy. If service providers faced liability, they might feel compelled to collect more information about users and keep that information longer.

The Chinese government is among those taking a very different approach. Beijing imposes responsibility for unlawful content on entities at every access point—from user to ISP, to social networking platform and Internet hosting company. If any intermediary allows users to distribute “harmful” content, or fails sufficiently to monitor or police the use of its services, it could face criminal liability, or revocation of its operating license. Further the government defines unlawful content in overbroad and vague terms. How is a blogging platform service like Blogger.com to decide which postings are “harmful” or damaging to “the interests of the nation”? China’s approach to intermediary liability is a key component of its larger system of online information control.

Addressing Potential Concerns

One objection to protecting intermediaries is the fear that genuinely harmful and offensive expression will grow online. But governments already possess many tools to address this concern while minimizing the impact on lawful expression and innovation. They might, for instance, promote and subsidize voluntary consumer use of filtering software that blocks pornography and other objectionable material. Some countries also require notice-and-takedown systems like the approach in U.S. copyright law and E.U. law to address this concern. However, notice-and-takedown systems are often easily misused to silence critics, especially where it is difficult to assess whether the challenged content is actually unlawful (as in the case of defamation). Finally, intermediaries can and do take voluntary steps to remove harmful material from their services (like spam or sexually explicit material) without government mandate, demonstrating that intermediary protections are compatible with advancing other important social goals.

Another key concern is that law enforcement officials must be able to legitimately pursue criminal wrongdoers, and victims must be able to pursue legitimate claims against those who have done them harm. An important aspect to the U.S. and E.U. approaches is that protection is afforded only to the intermediary, not to the parties that originally created or disseminated the objectionable content. Nothing under the U.S. or E.U. law prevents prosecution or claims against the original wrongdoer. One proper role for intermediaries might be to facilitate action against users (even anonymous users) in response to legitimate court orders, with procedures in place to safeguard privacy and some degree of anonymity.

Conclusion

Protecting intermediaries from liability is critical to preserving the Internet as an instrument for free expression and access to information, and thus as an engine of innovation and economic development. If liability concerns force the closure of user-generated content sites and other vital forums for social, economic, and political expression, we all will be the poorer. Governments should instead strengthen and adopt policies that protect intermediaries as key enablers of innovation, human rights, and economic development.