When President Clinton signed the Communication Decency Act, online pornography—and not defamation—was the focus of debate. However, in Reno v. American Civil Liberties Union, 521 U.S. 844 (1997), the Supreme Court struck down the pornography provisions, leaving §230’s protections from liability arising from third-party content.
On the other side of the Atlantic, however, the United Kingdom took a very different approach to online defamation. While the average British Parliamentarian may have had little experience of the internet at this time (by 2000, still only approximately one third of Westminster MPs were making use of the email facilities provided to them), concerns were raised about the Internet from early stages.
It would seem logical that the pronounced divergence between US and UK law in this regard was influenced by the context in which each first appeared: whereas Congress was seeking to shield service providers and intermediaries with the hope they would be emboldened to seek out pernicious sexual content and remove it from their systems, the UK instead saw the question of liability for third party provided libelous content in a pure defamation setting.
Much before 1995, legal academics speculated that the UK would go in much the same direction as the early US cases of Cubby v. CompuServe and Stratton Oakmont v. Prodigy, with liability being determined on the basis of awareness and potential for control. In Parliament, a Draft Defamation Bill was issued for public consultation in July 1995 by then Lord Chancellor, Baron Mackay of Clashfern. Mackay’s commentary in the consultation document made much of the rise of new technology via which “[i]nformation can be disseminated in a form which gives the recipient the option to convert it to a readable form, either presented on a VDU screen, or printed on the recipient’s own printer”, before intoning, darkly: “Progress is now so rapid, that tomorrow’s technology may well make even these advances appear old fashioned.” Nonetheless, Mackay, and subsequently Parliament, felt that in principle the pre-existing common law defense of innocent dissemination (Emmens v. Pottle) could and should be set on a statutory footing, in a form which applies to Internet intermediaries. Thus, on July 4, 1996, just a few months after the CDA came into force, the Royal Assent was given to a very different legal rule.
Under §1 of the Defamation Act 1996, any person other than the author, editor or publisher—that is, any secondary distributor, including online parties—who is found to have either been actually aware, or in a position in which the court considers they should, objectively, have been aware, of a defamatory statement published via their channels, will face legal liability for publication of a defamatory statement.
The first ruling on the matter came in the form of an early hearing on whether the §1 defense would be available to an ISP which was hosting a Usenet newsgroup on which a defamatory posting had appeared. In Godfrey v. Demon, Morland J ultimately ruled that a single, defamatory posting buried in an otherwise innocuous newsgroup, one of many hundreds of thousands of pages hosted but not edited or in any way monitored by the ISP, would not be something of which the defendant should have been aware until actual notice had been received. Godfrey had complained several times about the defamation, but Demon had failed to remove it until the system automatically deleted the post in question some ten days after it was first made. This actual notice would, Morland ruled, render the defense unavailable to Demon at full trial. The ISP consequently settled the case, paying Godfrey a reported £15,000 in damages and a further £485,000 costs.
In 2002, the UK enacted domestic legislation incorporating §4 of the European Electronic Commerce Directive 2000/31/EC, which further entrenched the awareness-based approach to online service provider liability for third-party content by expanding it beyond defamation, to cover all forms of unlawful content, both civil and criminal.
By comparison, across this period in the US, §230 as applied in Zeran v. AOL went from strength to strength, both in terms of whom it protected, from traditional online hosts, to website hosts of third party reviews, even non-commercial emailing lists and reposting of bulletin board posts, and in terms of the legal claims it immunized. Not only defamation, but also the provision of false stock information, discriminatory third-party comments in breach of fair housing laws, and even the marketing of obscene photographs of a minor in a chatroom.
One of the very few exceptions to the scope of §230 remained intellectual property law—copyright in particular, which had its own, awareness-based regime under the DMCA (and on which many have commented with regards to what this may say about relative lobbying strengths viz-a-vis Congress).
Whereas in the US there have been a whole slew of cases in which various defendants have successfully applied the §230 defense, the UK equivalents under the Defamation Act 1996 and the Ecommerce Regulations 2002 have produced very few reported cases. Instead, critics of the UK’s position claim that British ISPs, rather than risk liability, simply remove material at the simplest complaint, and thus, they argue, these laws provide a clear chill upon freedom of expression.
There is an absence of conclusive academic research supporting these claims. Certainly, anecdotal evidence suggests that generally UK ISPS do make some effort to determine the legitimacy of complaints, while some, such as the UK arm of the Facebook behemoth, are notoriously arbitrary with what they consider to be perfectly acceptable pursuant to their “community standards.” Nonetheless, many in the internet industry continued to feel that intermediaries were being unfairly required to decide whether material was unlawful and should be removed, with serious consequences should they make a wrong decision. As one former Demon staffer memorably and venomously put it during an academic conference, “We have people to make these decisions for us: we call them courts.”
Nonetheless, little changed in the UK position until the launch of the Libel Reform Campaign in 2009, with its manifesto Free Speech Isn’t Free. The LRC, comprised of various interest groups led by Index on Censorship and English PEN, called for numerous reforms of English defamation law, some more desirable than others.
One early proposed reform was for internet intermediaries to be immunized from all forms of liability for third party content. This was never going to be a popular idea among many lawyers or Parliamentarians, who by and large saw that §230, rather than encouraging service providers to be proactive, simply emboldened them to shrug and do nothing in response to claims of defamation. Further, as the LRC may have realized had they given the specific matter a little further legal research ahead of their initial report, the UK’s commitments under EU law provided a framework within which any changes here would be required to operate.
In the end, the eventual Defamation Act 2013 produced what might be termed a very British compromise. Section 10 provides that secondary distributors, including internet intermediaries, cannot be subject to legal action “unless the court is satisfied that it is not reasonably practicable for an action to be brought against [those directly responsible for its publication]”.
Further, §5 provides a defense for “operators of websites”. While the term is not defined, it would seem designed to encompass anyone who runs a social media site, a blog platform, a BBS or any form of website to which individuals can post content (as distinct from traditional, passive mere-hosting).
As first published, the defense seemed to make little sense, and add nothing to the pre-existing legal position. In essence, it provides that it is a defense for the website operator to show that the disputed content was provided by a third party. The defense is lost where the website operator acts maliciously, or where the client has been unable to identify the actual source of the defamatory posting, and has notified the operator, who has failed to respond “in accordance with any provision contained in regulations.”
As ever, God (or, depending on one’s view of the provisions therein, and /or theology, the Devil) is in the details. The Defamation (Operators of Websites) Regulations 2013 (which followed some months after the Act) set out a clear notice and takedown procedure which, if followed, will shield the Operator from liability. In essence, on receipt of a clear notice which properly identifies the content complained of, the Operator has a set period during which to respond, to contact the source of the contested posting to see if they wish to defend the content, and set circumstances in which the posting must either be deleted, or may be left alone.
While in some respects (all 48 hours and five days, not counting weekends, bank or public holidays) these time limits can rapidly begin to resemble Biblical numerology, the instructions are clear, simple, and should avoid much difficulty and expensive litigation for website operators in future. In spirit, they are not dissimilar to the reposting provisions to be found in the US DMCA under §512(g), although somewhat less complicated. Also, the §5 defense makes clear that standard content moderation practices do not constitute editing for the purposes of defamatory liability. Elsewhere, I have argued that this approach should be broadened to encompass all areas of civil law with regards to third party content online.
Though I’m very happy with the ‘third way’ approach that the UK has evolved in relation to service provider liability for third party provided defamatory content as distinctly different from §230 as applied by Zeran, to date this rule applies only in one of the UK’s three legal jurisdictions, that of England and Wales. In 2013, the responsible minister in the Northern Ireland Assembly’s arcane, power-sharing government elected not to adopt the new Defamation Act into Northern Ireland law, officially stating that he would “wait and see” how effective it proved in England and Wales. There has since been much speculation about this decision, but little progress as a series of unfortunate circumstances have once more collapsed the devolved government in Northern Ireland for the time being.
In the UK’s other jurisdiction, Scotland—where, unlike Northern Ireland, defamation law has long been markedly different than that in England—the Scottish Law Commission has, following a 2016 public consultation on defamation law reform, recently issued a draft Defamation and Malicious Publications (Scotland) Bill. Significantly, §3 provides that there are to be “no proceedings against secondary publishers”; exceptions to this general rule will only be made in respect of specific categories of secondary publisher to be determined by Scottish Ministers, granted authority to do so by §4. Distributors so identified will be subject to awareness based liability in exactly the same manner as under §1 of the Defamation Act 1996. What categories of secondary publisher may be exempted from liability and which will face liability, under what ministerial regulations, remains as of yet unspecified. Scotland could, if this model is taken forward into law, end up with a hybrid system between those with a §230 style ‘get out of jail free card’ immunity, and those subjected to something akin to the new English third-way approach. With existing EU obligations in this regard seemingly on borrowed time as the UK rushes towards the Brave New Post-Brexit World, it seems that Scotland’s options will soon be wide open. In Autumn 2017, the UK’s broad approach may be one of rejecting the §230 style of immunity, but by Autumn 2037, who knows?
Gavin Sutter is Senior Lecturer in Media Law and a member of the Institute of Computer and Communications Law at the Centre for Commercial Law Studies, Queen Mary University of London. He established, manages and teaches on various LLB and LLM Media Law courses at Queen Mary, and serves as Programme Director for the LLM in Media Law, as well as writing in the field. In his spare time, he also manages and lectures on the CCLS contribution to QMUL’s Joint Programme with the Beijing University of Post and Telecommunications, teaching Beijing-based undergraduate software engineers about jurisprudence, tort, privacy and criminal law.
This essay is part of a larger collection about the impact of Zeran v. AOL curated by Eric Goldman and Jeff Kosseff.