Revisiting The Common Law Liability Of Online Intermediaries Before Section 230
from the nuts-and-bolts dept
On February 8, 1996, President Clinton signed into law the Telecommunication Act of 1996. Title V of that act was called the Communications Decency Act, and Section 509 of the CDA was a set of provisions originally introduced by Congressmen Chris Cox and Ron Wyden as the Internet Freedom & Family Empowerment Act. Those provisions were then codified at Section 230 of title 47 of the United States Code. They are now commonly referred to as simply “Section 230.”
Section 230 prohibits a “provider or user” of an “interactive computer service” from being “treated as the publisher or speaker” of content “provided by another information content provider.” 47 U.S.C. § 230(c)(1). The courts construed Section 230 as providing broad federal statutory immunity to the providers of online services and platforms from any legal liability for unlawful or tortious content posted on their systems by their users.
When it enacted Section 230, Congress specified a few important exceptions to the scope of this statutory immunity. It did not apply to liability for federal crimes or infringing intellectual property rights. And in 2018, President Trump signed into law an additional exception, making Section 230’s liability protections inapplicable to user content related to sex trafficking or the promotion of prostitution.
Nevertheless, critics have voiced concerns that Section 230 prevents the government from providing effective legal remedies for what those critics claim are abuses by users of online platforms. Earlier this year, legislation to modify Section 230 was introduced in Congress, and President Trump has, at times, suggested the repeal of Section 230 in its entirety.
As critics, politicians, and legal commentators continue to debate the future of Section 230 and its possible repeal, there has arisen a renewed interest in what the potential legal liability of online intermediaries was for the content posted by their users under the common law, before Section 230 was enacted. Thirty years ago, as a relatively young lawyer representing CompuServe, I embarked on a journey to explore that largely uncharted terrain.
In the pre-Section 230 world, every operator of an online service had two fundamental questions for their lawyers: (1) what is my liability for stuff my users post on my system that I don’t know about?; and (2) what is my liability for the stuff I know about and decide not to remove (and how much time do I have to make that decision)?
The answer to the first question was not difficult to map. In 1990, CompuServe was sued by Cubby, Inc. for an allegedly defamatory article posted on a CompuServe forum by one of its contributors. The article was online only for a day, and CompuServe became aware of its contents only after it had been removed, when it was served with Cubby’s libel lawsuit. Since there was no dispute that CompuServe was unaware of the contents of the article when it was available online in its forum, we argued to the federal district court in New York that CompuServe was no different from any ordinary library, bookstore, or newsstand, which, under both the law of libel and the First Amendment, are not subject to civil or criminal liability for the materials they disseminate to the public if they have no knowledge of the material’s content at the time they disseminate it. The court agreed and entered summary judgment for CompuServe, finding that CompuServe had not “published” the alleged libel, which a plaintiff must prove in order to impose liability on a defendant under the common law of libel.
Four years later, a state trial court in New York reached a different conclusion in a libel lawsuit brought by Stratton Oakmont against one of CompuServe’s competitors, Prodigy Services Co., based on an allegedly defamatory statement made in one of Prodigy’s online bulletin boards. In that case, the plaintiff argued that Prodigy was different because, unlike CompuServe, Prodigy had marketed itself as using software and real-time monitors to remove material from its service that it felt were inappropriate for a “family-friendly” online service. The trial court agreed and entered a preliminary ruling that, even though there was no evidence that Prodigy was ever actually aware of the alleged libel when it was available on its service, Prodigy should nevertheless be deemed the “publisher” of the statement, because, in the court’s view, “Prodigy has uniquely arrogated to itself the role of determining what is proper for its members to post and read on its bulletin boards.”
The Stratton Oakmont v. Prodigy ruling was as dubious as it was controversial and confusing in the months after it was issued. CompuServe’s general counsel, Kent Stuckey, asked me to address it in the chapter I was writing on defamation for his new legal treatise, Internet and Online Law. Tasked with this scholarly mission in the midst of one of the digital revolution’s most heated legal controversies, I undertook to collect, organize and analyze every reported defamation case and law review commentary in this country that I could find that might bear on the two questions every online service faced: when are we liable for user content we don’t know about and when are we liable for the user content we know about but decide not to remove?
With respect to the first question, the answer dictated by the case law for other types of defendants who disseminate defamatory statements by others was fairly clear. As I wrote in my chapter, “[t]wo common principles can be derived from these cases. First, a person is subject to liability as a ‘publisher’ only if he communicates a defamatory statement to another. Second, a person communicates that statement to another if, but only if, he is aware of its content at the time he disseminates it.” Hamilton, “Defamation,” printed as Chapter 2 in Stuckey, Internet & Online Law (Law Journal-Seminars Press 1996), at 2-31 (footnotes omitted).
I concluded that the trial court had erred in Stratton Oakmont because it failed to address what the term “publish” means in the common law of libel—to “communicate” a statement to a third party. When an intermediary disseminates material with no knowledge of its content, it does not “communicate” the material it distributes, and therefore does not “publish” it, at least as that term is used in the law of libel. Thus, whether the intermediary asserts the right of “editorial control” over the content provided by others, and the degree of such control the intermediary claims to exercise, are immaterial to the precise legal question at issue: did the defendant “communicate” the statement to another? I wrote:
While it is true that a publisher’s “choice of material to go into a newspaper” constitutes “the exercise of editorial control and judgment” by that publisher, his “increased liability” for defamation arises from the knowledge of content that he inherently acquires as a result of exercising that judgment to include the material in the newspaper; it does not arise from the mere fact that he has a right to make that judgment. All distributors, like primary publishers, exercise the very same right to determine what material they will disseminate and what material they will not. . . . Indeed, the liability standard applied to a distributor presumes that he has such a right to refuse distribution and requires him to exercise it whenever he knows or has reason to know that a particular publication contains unlawful or tortious content. His efforts to exercise that right, therefore, cannot create the very same general duty to inspect content that is prohibited by that common law standard (and by the First Amendment).
Id. at 2-62 (footnotes omitted, quoting Stratton Oakmont, 23 Media L. Rep. (BNA) 1794, 1796 (N.Y. Sup. Ct. May 25, 1995).
With respect to the second question online intermediaries had for their lawyers—when are we liable for stuff posted by users we decide not to remove—the answer dictated by the common law was anything but firmly established and settled. In the pre-digital world, the economics of communicating with the public made it far more practical for aggrieved plaintiffs to sue only the producers of such content rather than those who merely distributed it. The truth is that in the pre-Internet history of the common law of libel, entities in the business of distributing the printed content of others were rarely sued, and even then only as an afterthought to defeat diversity of citizenship and thereby prevent the defendants in a state court action from removing the lawsuit to the federal courts. And in only two of those rare cases was the distributor defendant alleged to have actual knowledge of the defamatory content it was selling to the public; in both cases, the distributor defendant was eventually dismissed before the case ever went to trial.
Thus, as I noted in my chapter, I did not find a single reported case of defamation liability actually being imposed on an entity in the business of distributing to the public printed content produced by others. That meant that, prior to the enactment of Section 230, when a lawyer advised his intermediary client as to when he might be held liable for deciding not to remove users’ content, the lawyer could refer only to dicta by courts and speculation by commentators as to how courts might apply the law in that circumstance.
And there certainly was no consensus in such speculation. As I noted in my chapter, Professor Keeton observed in 1984 that “[i]t would be rather ridiculous, under most circumstances, to expect a bookseller or a library to withhold distribution of a good book because of a belief that a derogatory statement contained in the book was both false and defamatory of the plaintiff.” Prosser and Keeton on the Law of Torts, § 113, at 811 (5th ed. 1984). Indeed, do we really expect Kroger to make decisions whether to pull an issue of the National Enquirer from the shelves in every one of its grocery stores across the country because the CFO’s spouse told her at breakfast that he read in that week’s issue that a celebrity claimed one of his critics was a “liar”?
As I observed in my chapter, it is also noteworthy that in 1992, the National Conference of Commissioners on Uniform State Laws considered, but did not adopt, a standard that would immunize from republisher liability any “library, archive, or similar information retrieval or transmission service” that provides access “to information originally published by others,” if it is not “reasonably understood to assert in the normal course of its business the truthfulness of” such information or if it “takes reasonable steps to inform users” that it makes no such assertion. See Perritt, “Tort Liability, the First Amendment, and Equal Access to Electronic Networks,” 5 Harv. J.L. & Tech. 65, 108 (1992).
In 1996, Section 230 was enacted into law at the same time my research and analysis of the applicable common law standards was published as Chapter 2, “Defamation,” in Kent Stuckey’s treatise, Internet & Online Law (Law Journal-Seminars Press 1996). I then added a section to the chapter discussing Section 230 and continued to update it for a few years to discuss the initial cases applying it. Eventually, however, it became apparent that, in light of the courts’ construction of Section 230, an extensive discussion of the pre-Section 230 case law with respect to the liability of online intermediaries for user content was no longer needed, and my chapter in the treatise was replaced with one that focuses instead on the cases applying Section 230.
Kent Stuckey’s treatise is still being updated and is available for purchase from Law Journal Press. The chapter I wrote, however, which details all of the reported case law and commentary I found that might bear on the potential liability of online intermediaries for defamation under the common law at that time, before Section 230 was enacted, has not been made available to the public for many years. In light of the renewed interest in this topic as part of the current debates about Section 230’s future, that chapter is being made available online, with permission, here (pdf link, also embedded below).
Robert W. Hamilton is Of Counsel at Jones Day. He has more than 36 years of experience in state, federal, and bankruptcy court litigation and in First Amendment and media cases. Bob represented CompuServe in Cubby v. CompuServe in 1990-1991. The views and opinions set forth herein are the personal views or opinions of the author; they do not necessarily reflect views or opinions of the law firm with which he is associated.
- See Spence v. Flynt, 647 F. Supp. 1266, 1273-1274 (D. Wyo. 1986) (convenience store continued to sell magazine for two days after plaintiff’s attorney notified store employee of its defamatory content); Janklow v. Viking Press, 378 N.W.2d 875, 876 S.D. 1985) (plaintiff sued bookstore owners and operators who “willfully refused to remove the book from the shelves” of their stores “even though he had notified them of its libelous nature”). In the Spence case, the convenience store was dismissed from the litigation prior to trial, at which liability was imposed on the original publisher for describing an attorney as a “vermin-infested turd dispenser.” See Spence v. Flynt, 816 P.2d 771 (Wyo. 1991). In the Janklow case, the claims against the bookstores eventually were dismissed on summary judgment for the same reason that judgment was entered for the producers of the book (speech was “opinion” protected by 1st Amendment). See: In the Spirit of Crazy Horse.
- I found two reported cases in which a court imposed liability on a property owner for defamatory graffiti on an interior wall in his building, and three extremely old cases in which a property owner was held responsible for refusing to remove a defamatory statement displayed on his property. In none of those cases was the defendant in the business of distributing to the public printed speech produced by others.