As Michael Goodyear notes in Common Law Notice-and-Takedown, immunity for online platforms is very much on the legislative agenda after many years of relative stability. Politicians from the left and right in the United States have reacted for ostensibly different reasons to the surfeit of misinformation online by suggesting that existing statutory regimes that provide (conditional) immunity for platforms need revisiting. But elimination or modification of specific statutory safe harbors for platforms will not of itself render platforms liable for the sins of their wrongdoing customers.
Instead, that will turn on whether a platform’s conduct falls within existing causes of action, most likely (but perhaps not exclusively) under theories of secondary liability. Indeed, even complete elimination of statutory safe harbors will not signal the end of notice and takedown systems. The roots of notice and takedown in the copyright context can be found in Judge Whyte’s 1995 Netcom decision, which to some extent was implemented three years later in greater detail by the Digital Millennium Copyright Act (DMCA).
Moreover, notice and takedown systems operate to insulate platforms for liability for trademark infringements perpetrated by users, even though trademark law operates both outside Section 230 of the Communications Decency Act and without any regime-specific parallel to the DMCA. The common law standard for secondary trademark liability, most authoritatively articulated by the Second Circuit in Tiffany v eBay, encourages such systems, but without the specifics that one finds in the detail of the copyright statute.
In Common Law Notice-and-Takedown, Professor Goodyear explores the notice and takedown practices that have developed in the trademark space by sampling the takedown policies of forty-five large platforms of a variety of types (such as social media, e-commerce, print on demand). In the European Union, the European Commission facilitated a Memorandum of Understanding (MoU) between major online platforms and trademark owners that sketched out best notice and takedown practices and has in turn generated periodic reports that have shown some light on evolving norms.
Lacking a similar government intervention in the United States, academic work opening a window into how trademarks play into platform practices, such as found in Jeanne Fromer and Mark McKenna’s recent paper on Amazon, will be invaluable. Goodyear’s article is an important contribution in that regard, though he candidly acknowledges both that he studied only publicly available material and that transparency reports by several platforms offer more quantitative material that is also worth mining (although some reports contain numbers at only a very general level).
Goodyear reports that platforms have taken advantage of the latitude afforded by the common law-grounded system and have wide variation in the information that they require of takedown requests beyond those made under the DMCA. The most surprising feature, as a conceptual matter, might be the need for a trademark registration despite the use-based nature of U.S. trademark rights. But this may only highlight the fact that commercial norms are as likely to be driven as much by pragmatic assessment of what procedures work best at scale as by a searching theoretical inquiry into the merits of any particular legal claim, and the common law grounding of the system allows such latitude in ways that a detailed statutory scheme might not.
Platforms have adopted an a la carte approach to whether to mimic other features of the DMCA scheme (such as repeat infringer policies and counter notice mechanisms, the latter of which Goodyear reports has been adopted by a sizeable number of platforms despite the lack of statutory immunity that the DMCA confers for those operating such a feature). Again, this should not be a surprise given the different level of granularity in the common law of trademarks and the DMCA. But it raises a longstanding question whether efficiency and certainty concerns should counsel for the adoption of a common set of immunity conditions across all forms of legal claims, elevating those concerns over the different formal legal conditions for liability in different areas.
Despite the room for innovation in trademark policies, Goodyear’s study confirms the strong influence of the DMCA (at least in setting a baseline and providing a suite of possible features). The pragmatic impulse towards replication of mechanisms that bring immunity in adjacent areas such as copyright is understandable, but the durability of such an approach might depend upon any number of moving variables. These include not only changes in the adjacent regime, but in parallel regimes in other major jurisdictions in which the platforms operate. The influence of the DMCA is arguably amplified by the fact that the counterpart EU system of safe harbors in the E-Commerce Directive is broadly similar and does apply beyond copyright. Indeed, courts in Europe move without much thought between case law involving immunity from different legal claims.
But within the last few years, the EU has reformed its core notice and takedown approach (first with the Directive on Copyright and Related Rights in the Digital Single Market and then with the Digital Services Act). It is not clear whether a platform’s route to efficient convergence of practices will in the near future be driven by rules in adjacent legal regimes or adjacent jurisdictions. It may be that the notional flexibilities of the common law system of notice and takedown that Goodyear nicely illuminates will effectively be over-ridden by details imposed by efficiency-seeking cross-border actors.
This is an invaluable article, and it should contribute to discussion of what the landscape might (and should) look like if notice and takedown operates according to more general principles rather than detailed statutory conditions. It also should stimulate follow-on work by others, particularly if accompanied by interrogations of data regarding the implementation of the different policies that Goodyear highlights.
