Collateral censorship and the limits of intermediary immunity.

AuthorWu, Felix T.

The law often limits the liability of an intermediary for the speech it carries. And rightly so, because imposing liability on intermediaries can induce them to filter out questionable content and this "collateral censorship" risks suppressing much lawful, even highly beneficial, speech. The "collateral censorship" rationale has its limits, though, and correspondingly, so should the applicability of intermediary immunity. The worry with collateral censorship is not just that intermediaries censor, but that they censor more than an original speaker would in the face of potential liability. Increased censorship, in turn, is the product of applying liability targeted at original speakers to entities whose interests diverge from original speakers. Where the "intermediary" has the interests of an original speaker, and so should be regarded as one, or where the form of liability already takes into account the intermediary's interests, collateral censorship is not the problem, and immunity is not the right response. This understanding should, in particular, inform the interpretation of [section] 230 of the Communications Decency Act of 1996, a federal statute that broadly immunizes Internet intermediaries from speech torts and many other forms of liability. In cases involving the republication of e-mails, questionnaires, member screening, and contract claims, among others, courts have begun to explore the limits of this immunity, but have done so haphazardly and inconsistently, having largely lost sight of the underlying rationale for immunity. Focusing on the conditions that generate problematic collateral censorship provides a principled basis upon which to define the limits of intermediary immunity generally, and [section] 230 in particular.

INTRODUCTION

Suppose I find that someone has posted a defamatory comment about me on an online message board. Is the operator of the message board liable? Or suppose someone e-mails me a bit of juicy gossip, and I repeat the gossip on my blog. Am I liable? Or suppose I pay my Internet service provider to scan my incoming e-mails for viruses, but my ISP fails to do so, and an e-mail virus infects my computer. Is my ISP liable?

Under a federal statute, [section] 230 of the Communications Decency Act of 1996, (1) the answer to the first question is definitively no. (2) Indeed, as that statute has been interpreted by the courts, a message board operator is not liable for a defamatory comment posted by a third party even if the target of the comment asks the operator to remove the posting and the operator refuses. (3) Some commentators have criticized this result as unduly allowing message board operators and other Internet intermediaries to ignore even patently unlawful speech without facing any consequences, to the detriment of those harmed by the speech. (4) The result, however, appears firmly entrenched in the courts, which have repeatedly defended it on both statutory and policy grounds. (5)

The second and third questions, and others like them, have received far less academic attention. Existing commentary has largely focused on critiquing intermediary immunity generally and suggesting alternatives to [section] 230, (6) rather than on defining the types of factual and legal settings in which such immunity is appropriately applied. Given the courts' interpretation that [section] 230 eliminates even notice-based liability, defining the applicability of that immunity becomes all the more important. Such issues of applicability have arisen with increasing frequency, and courts have struggled to find a coherent framework to address them. The answers they have given so far have been largely inconsistent and unprincipled. (7) This Article provides a framework for understanding both how to conceptualize the limits of intermediary immunity as a policy matter and how courts should interpret [section] 230.

In order to define the bounds of intermediary immunity, and answer the second and third questions, we need to return to the first question and determine more precisely why intermediary immunity might be an appropriate response in the situation of a message board operator's liability for defamatory content. A prime rationale for immunity in that context is concern over what has been called "collateral censorship." (8) Collateral censorship occurs when a (private) intermediary suppresses the speech of others in order to avoid liability that otherwise might be imposed on it as a result of that speech. (9) This is a problem because some of the suppressed speech might in fact be lawful, even socially desirable. (10) For example, imposing defamation liability on a message board operator for carrying defamatory content may well induce it to block a wide array of potentially defamatory content, including some which is in fact true or mere opinion, or otherwise not actionable. The result is that the accurate report of corporate malfeasance is swept away along with the nasty, fabricated rumor.

Intermediary immunity is a response to the problem of collateral censorship. To avoid giving intermediaries an incentive to block lawful content, they are immunized from claims even as to the unlawful content that they carry. If the content of a message cannot provide a basis for suing the intermediary, then the intermediary no longer has a legal incentive to suppress that message based on its content. Whether immunity is a necessary response to the threat of collateral censorship is, of course, contested. (11) Collateral censorship does, however, provide the major justification for immunity, particularly in its broadest forms, (12) so that the absence of problematic collateral censorship marks situations in which even those who generally support immunity ought to reject it.

We therefore need to understand what makes collateral censorship a problem. In particular, the problem cannot be simply that the threat of liability results in the suppression of speech, for that is true whenever there is liability for speech. People regularly engage in self-censorship under fear of liability, but if that is the crux of the problem, then the appropriate solution would be to change the substantive liability itself.

The unique harm of collateral censorship, as opposed to serf-censorship, lies in the incentives that intermediaries have to suppress more speech than would be withheld by original speakers. This additional suppression occurs because intermediaries have different incentives to carry particular content than original speakers have to create it in the first place. (13) Those incentives diverge both because original speakers obtain benefits from the speech not realized by intermediaries and because intermediaries face liability risks not borne by original speakers. When the same law is then applied to intermediaries and original speakers alike, despite the divergence of incentives, then the deterrent effect on intermediaries will be excessive. (14)

The problem of collateral censorship is thus one of applying a law crafted for original speakers to entities whose interests diverge from those of original speakers. We can therefore identify two types of situations in which collateral censorship is not the problem and intermediary immunity is not the appropriate response. The first is when the putative intermediary is actually acting not as an intermediary at all, but rather as an original speaker. When an entity faces the incentives of an original speaker, we should treat it as such and subject it to the liability faced by original speakers generally. This straightforward proposition has been lost on courts interpreting [section] 230, which have focused inappropriately on who "made up" the content, rather than who is speaking it, in separating speakers from intermediaries. If I blog about a juicy rumor, I am the speaker, and I should be subject to liability, even if the rumor started elsewhere.

The second situation in which collateral censorship is not the problem is when the form of liability is one specifically directed to the intermediary. When the liability being imposed on the intermediary is not one that could be sensibly imposed on the original speaker, then we need not worry about the potential divergence of incentives. Any such divergence becomes irrelevant to the appropriateness of the liability. To be sure, there may continue to be other substantive rules, common law or constitutional, that govern the appropriateness of the intermediary liability. But it is inappropriate to use immunity to short-circuit the analysis. If my ISP promises to scan my e-mail, it should be held liable for that promise, even if third-party activity is the basis for damages, because a promise is only sensibly enforced against the promisor.

Part I of this Article describes in more detail how intermediary immunity can be justified as a solution to the problem of collateral censorship. Part II traces the development of intermediary immunity, first in the common law and then in [section] 230, showing that collateral censorship has been a prime rationale for intermediary immunity, particularly in its broadest forms. Part III describes issues the courts have faced in determining whether intermediary immunity is appropriately applied in particular factual settings, and describes why the courts' approaches thus far have been unsatisfactory. Part IV uses the theory of collateral censorship to derive two limits on the appropriate scope of intermediary immunity. Part IV.A shows that immunity is inappropriate when the intermediary is really an original speaker, because it is situated as such. Part IV.B shows that immunity is inappropriate when the form of liability is one specifically directed at intermediaries. A brief conclusion follows.

  1. THE PROBLEM OF COLLATERAL CENSORSHIP

    The Internet promised us "cheap speech," (15) and it delivered. Measured in sheer volume, there is surely more content on Facebook today than has ever...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT