Response to the consultation by the Ministry of Justice on the Draft Defamation Bill

Response to the consultation by the Ministry of Justice on the Draft Defamation Bill

Gavin Sutter [1] [2]

Cite as: Sutter, G., ‘Response to the consultation by the Ministry of Justice on the Draft Defamation Bill’, European Journal for Law and Technology, Vol. 3, No. 1, 2012


This is a collaborative submission from a group of academics based in the UK with expertise in information technology law and related areas. This response deals with selected questions which fall under the remit of BILETA's stated interests. The preparation of this response has been funded by the Information Technology Think Tank, which is supported by the Arts and Humanities Research Council and led by the SCRIPT/AHRC Centre for Research in Intellectual Property and Technology, University of Edinburgh.

This response has been prepared by Gavin Sutter, Lecturer in Media Law and a member of the Institute of Computer & Communications Law at the Centre for Commercial Law Studies, Queen Mary, University of London. In addition, this response is submitted by the following individuals: Mr Mohammad Alramahi, University of Hull, Dr. Abdulrahman Alzaagy, University of Wales, Dr Subhajit Basu, University of Leeds, Mr Michael Bromby, Glasgow Caledonian University, Dr Abbe Brown, University of Edinburgh, Dr Bela Bonita Chatterjee, Lancaster University, Dr Catherine Easton, Manchester Metropolitan University, Mr Felipe Romero Moreno, Oxford Brookes University

Clause 6: A Single Publication Rule

Question 16
Do you agree with the inclusion of a clause in the Bill providing for a single publication rule?

Very much so. The multiple publication rule is, quite simply, outdated. It was rejected as such by US courts in the 1940s. In Gregoire v GP Putnam & Sons, the court considered that it was simply ‘not suited to modern mass publishing’. [3] The core problem with the multiple publication rule is that it effectively emasculates the limitation period. The purpose of the limitation period is to ensure that a case is brought within a reasonable time, allowing the defendant a reasonable opportunity to mount a proper defence. This is, however, frustrated by the rule in Duke of Brunswick v Hamer, as it is possible for a new limitation period to be triggered, and thus action may be taken, many years after publication was first made. [4] The only real limitation, then, becomes the lifespan of the potential claimant. In 1981, the Neill Committee wrote:

‘Memories fade. Journalists and their sources scatter and become, not infrequently, untraceable. Notes and other records are retained only for short periods, not least because of limitations in storage.’ [5]

It might be that in a case where there has been a particularly long gap initial publication and that in respect of which an action in libel has been brought, the multiple-publication rule amounts to a violation of the Article 6 right of the European Convention on Human Rights (as enshrined in English law by the Human Rights Act 1998), the so-called ‘right to a fair trial’.

Applying the rule to online archives for the first time, the Court of Appeal in Loutchansky v Times rejected the defendants’ submission that the application of the Duke of Brunswick rule violates Article 10 by being likely to discourage the provision of online archives. [6] The European Court of Human Rights ruling that the application of the multiple publication rule to internet archives in this manner did not constitute a breach of the defendants’ Article 10 right. [7] Notably, however, the Court did indicate that had there been a longer period of time elapsed between date of first publication and the date of publication giving rise to an action under the multiple publication rule, it would have been prepared to reach a different conclusion:

‘While an aggrieved applicant must be afforded a real opportunity to vindicate his right to reputation, libel proceedings brought against a newspaper after a significant lapse of time may well, in the absence of exceptional circumstances, give rise to a disproportionate interference with press freedom under Article 10.’ [8]

With such a clear indication that the Strasbourg court might in future be prepared to find the multiple publication rule in violation of the free expression right, it would seem only sensible for this to be preempted by the introduction of a statutory single publication rule. A further argument against the multiple publication rule relates to its financial implications. It has been argued by proponents of the multiple publication rule that damages will be minimal in relation to a case brought many years later, but this fails to take account of the cost of defending a libel action, which will only increase over time. While the above issues apply to publications made both online and offline, it should be noted that they are, in practice, of particular significance online. Online archives are typically much more easily, and therefore much more commonly, accessed by the general public, raising the risk of liability. Further, as a general rule, while most offline archives available to the public are created and maintained by third parties to whom an awareness-based, distributor defence is available, online archives are very frequently created and maintained by the publisher, who faces strict liability.

Question 17

Do you have any views on the substance of the draft clause?

In particular, do you consider that the provision for the rule to apply to publications to the public (including a section of the public) would lead to any problems arising because of particular situations falling outside its scope?

It is possible that the initial publication might be made to a closed group in such circumstances that a potential claimant, despite being clearly identified and obviously defamed, cannot be reasonable expected to be aware of the publication. It is conceivable that such an article could then be made available to a much wider group after the initial twelve month period of publication has been exhausted. In such a case, or any other comparable circumstance, the court would still be able to exercise its discretion under Section 32A of the Limitation Act 1980 to allow the action to go ahead if satisfied than it is in the interests of justice so to do.

Do you agree that the single publication rule should not apply where the manner of the subsequent publication of the material is materially different from the manner of the first publication? If not, what other test would be appropriate?

We would agree in principle, as otherwise any person who once publishes a defamatory allegation without suit being filed against him within the initial twelve month period would be able to republish the same allegation in any form and remain immune from liability. The factors in the Draft Bill’s Clause 6(5), which a court may consider in determining whether a ‘materially different’ publication has taken place, are right to emphasise the context of a publication. The nature of the internet is such that while a defamatory article remains the same, the website in which it is located may change significantly on a daily, or even hourly, basis. It is important that such factors as, for instance, an increase in prominence given to an article which leads to a vast increase in circulation of the piece are taken into account, while equally it would be unfair for unrelated changes elsewhere in the website to be interpreted as amounting to the defamatory article being published in a ‘materially different’ form. Were the latter to be done, it would produce the same effect as the multiple publication rule, with new limitation periods constantly triggered. It would be impossible to provide an exhaustive set of criteria in statute that would allow for every situation, and thus it is submitted that the non-exhaustive list of factors of which a court may take account, as provided in the draft Defamation Bill, are a suitable approach. Perhaps a Clause 6(5)(c) might be added, thus:

‘The fact of change to the content of a website other than to the article in question shall not alone be considered to amount to a materially different publication having been made.’

Question 18
Do you consider that any specific provision is needed in addition to the court’s discretion under Section 32A of the Limitation Act 1980 to allow a claim to proceed outside the limitation period of one year from the date of first publication?

Section 32A alone is sufficient; any duplication within the Defamation Bill would be superfluous.

Clause 7: Libel Tourism

Question 19
Do you agree that the proposed provisions on libel tourism should be included in the draft bill?

It is submitted that, despite much hysteria among the proponents of radical reform to English libel law, a convincing case for so-called libel tourism as a significant problem has yet to be made. Indeed, research undertaken by legal publishers Sweet & Maxwell found that of eighty-three libel suits before the English courts in 2009-2010, only three were brought by claimants domiciled outside the UK. [9] The term ‘libel tourism’ is commonly used in a manner which implies that a claimant who is domiciled outside the UK has no business bringing a libel action in England. It s difficult to give any credit to such an argument in the contemporary world where an individual may well travel regularly for business or pleasure and can indeed have a reputation of value to them in England despite being domiciled elsewhere. This was recognized by the House of Lords in Berezovsky v Michaels, for instance. [10]

The decision of the court in Harrods v Dow Jones, to follow the reasoning of the Australian Court in Gutnick v Dow Jones that the place of publication of an online article is that at which it is downloaded in an intelligible form and read by a third party has often been cited as opening the floodgates to libel tourism as by this standard anything uploaded to the internet anywhere is published in England. [11] This interpretation, however, fails to take account of the fact that the courts have also emphasized that mere publication alone is not enough: a successful claimant must also still satisfy the court that he or she has a sufficient connection to the jurisdiction before a case may be brought. [12] The courts are strict in the application of both requirements: in Jameel v Dow Jones, the claimant’s case was rejected as, on the basis of evidence before the court, while publication had taken place and despite there being no de minimus standard in English defamation law, publication in England was so limited that the court would ultimately be wasting its time. [13] The courts have been very clear in this area. We are not implacably opposed to placing the existing common law position on a statutory footing, however, if this is to be done it must first be understood that the true scale of this ‘problem’ is far more limited than many on the reform lobby would have one believe.

As to the US SPEECH Act, a federal statute which forbids the enforcement within the US of ‘foreign’ (impliedly, ‘British’) libel judgments which would violate the First Amendment, this particular legislative measure can be dismissed as mere posturing. The SPEECH Act does not actually change anything: no US court would have agreed to enforce such a judgment anyway. It is submitted that in any case the enforceability of otherwise of a judgment of an English court in a foreign state should not be a concern for Parliament. Rather, the Defamation Bill should be concerned with making necessary improvements for the law of defamation within the UK, including protecting the right of any potential claimant who demonstrably has a reputation worthy of protection within the jurisdiction, and which has been threatened by a publication made available in the UK via whatever medium. In respect of the ‘chilling effect’ referred to in paragraph 80 of the consultation paper, it is submitted that the true chilling effect is the cost of defending an action in defamation, rather than the fear of being sued itself.

Question 20
Do you have any views on the substance of the draft clause?

It is of great import that no court be encouraged to make a decision based upon circulation numbers alone. 500,000 readers in England might be but a drop in the ocean where total worldwide readership runs to ten million or more, however this still represents a very significant publication into the UK. By introducing the consideration of other states in which the contested article has been published, there is a danger that it will become a numbers game. We would prefer to impose the following, three-stage test:

  • The court must be satisfied that the claimant has a sufficient connection to the jurisdiction such that he or she has a reputation to protect.
  • The court must be satisfied that there has been a significant level of publication in England such that damages likely to be recovered are proportionate to the likely cost of the action.
  • The court must be satisfied that the claimant has a reasonable chance of success.

Responsibility for publication on the Internet

Question 23
Do you consider that it would be appropriate to change the law to provide greater protection against liability to internet service providers and other secondary publishers?

We would agree that some level of clarification is necessary, with such changes being made as are necessary to address concerns that intermediaries are prone to unquestioning acquiescence to any and all requests for takedown of material alleged to be libelous. It is certainly desirable to enable intermediaries to be able to make a genuine ‘best judgement’ with respect to complaints received that they are making a defamatory article available, as opposed to the threat of legitimate material being deleted for fear of a lawsuit.

N.B.: there are two key issues which must not be forgotten here:

  • Again, the true reason many intermediaries fear a lawsuit is the sheer cost of defending such an action. This is the reason why they take material down on receipt of a complaint rather than risk court action, and we firmly believe that were it to be significantly cheaper to defend an action (perhaps in the context of a libel tribunal), we would see much less reluctance by intermediaries to consider the material at hand before automatically agreeing to remove it.
  • The UK’s obligations under Chapter 4 of the Electronic Commerce Directive (2000/31/EC) are such that only minor tweaking of the law with regards to online intermediaries is actually an option.

Question 24
If so, would any of the approaches discussed above provide a suitable alternative? If so, how would the interests of people who are defamed on the internet be protected? Do you have any alternative suggestions?

The first approach suggested, removing liability altogether, is simply not viable due to existing commitments under European Union law. In any case, it is not a desirable option. This is the approach taken in the US under Section 230 of the Communications Decency Act . The first thing to note about the US position is that it has come about by accident. The Communication Decency Act originally was enacted to introduce new offences relating to making pornography available to minors. As the material in question was merely ‘indecent’ (rather than obscene) and therefore protected speech under the First Amendment, these offences were held by the Supreme Court to be unconstitutional, and struck out. [14] The immunity from civil liability set out in Section 230 was originally designed to encourage intermediaries to actively police material which they hosted without fear of occasioning strict liability as editors of any libellous content which might be found on their servers. In practice, all this provision seems to have achieved is to permit intermediaries [15] (including owners of websites [16] and even those who have simply quoted a previous posting on an internet bulletin board [17]) to ignore defamatory material with impunity even where fully aware both of the existence of the material and its defamatory nature. This is not an option which the UK should choose, even were it free so to do.

The second approach, that which is taken in the US under the Digital Millennium Copyright Act, merits consideration. Removing the service provider from the picture and thus discouraging any potential for material to be taken down as soon as a complaint is received could address any threat to freedom of expression occasioned by the perceived ‘chilling effect’ of potential libel proceedings against the intermediary. Where the content provider wishes to dispute the claim of defamation in court, this would also be in tune with the general reluctance of the English courts to issue pre-trial injunctions in libel cases save in circumstances where it is so blindingly obvious that the article in question is defamatory that a reasonable defence cannot possibly be mounted. [18] This approach may be unsuitable to apply in cases in which all the evidence available to the intermediary cannot lead to the identification of the person responsible for uploading the defamatory article. Clearly, a requisite condition for such an approach would have to be that the service provider identify the source of the material, or at least provide such information as is available to them which could lead to such identification. This could, as stated in paragraph 114, have cost implications for potential claimants, and this would have to be addressed. This is one instance where a specialist defamation tribunal would be well placed to address the situation. Requiring a court order to declare the information unlawful before mandatory takedown of the contested item would be an easy fit with the existing system of injunctions, but again this could have cost implications. This would be the simplest alternative to introduce, and could work well in conjunction with the establishment of a low-cost, libel tribunal.

Introducing differing levels of liability for larger commercial and smaller, non-commercial providers seems unwise. It would, in the first instance, cause additional complexity as it would require clear divisions to be set between the two, particularly if the overall size of the operation is to be considered (as will be the case with the enactment of the provisions of the Digital Economy Act in relation to ISPs). Secondly, this would inevitably lead to claims that larger intermediaries were being ‘punished’ for their success by facing additional liability risks. Thirdly, to immunize smaller scale providers in this manner would ignore the fact that there may well be situations in which it is entirely reasonable for an online, third party distributor of whatever scale to be held accountable for the information which they have made available. Also, while it may seem, prima facie, that a small-scale blog or BBS provider might be more likely to be able to identify the source of an alleged defamatory article posted to their system, there is no guarantee that such information as they hold will indeed allow the source to be identified, leaving some claimants potentially without significant remedy.

Question 25
Have any practical problems been experienced because of difficulties in interpreting how the existing law in section 1 of the 1996 Act and the E-Commerce Directive applies in relation to internet publications?

The chief difficulty with Section 1 appears to be distinguishing between what may or may not be permissible without the service provider losing their qualified immunity. In Section 1 of the 1996 Act, in order to qualify for the defence the service provider must satisfy the court both that ‘he did not know, and had no reason to believe, that what he did caused or contributed to the publication of a defamatory statement’ and that ‘he took reasonable care in relation to its publication’. It has long remained unclear what steps exactly constitute ‘reasonable care’ by a third party distributor as opposed to editing the material which would therefore deprive the service provider of the defence, or even to what extent such ‘reasonable care’ might give rise to a situation in which sufficient constructive knowledge is present to occasion liability. The Ecommerce Directive / Regulations have not clarified this further. Indeed, intermediaries are overall best advised to carry out no investigation whatever into the content they make available, and to remove content as soon as any complaint comes to light should they wish to avoid liability. The system becomes further confused when one considers the use of systems such as British Telecom’s Cleanfeed, a filtering software standard which blocks the availability of websites identified by the Internet Watch Foundation as featuring child sexual abuse images. Most of the major UK service providers voluntarily use this system, however the courts have yet to rule on whether this form of filtering, although mechanical rather than human, constitutes editing. Various courts across Europe have even come to very different conclusions as to whether eBay, which offers dispute resolution procedures, a ‘feedback system’, and so on, is sufficiently non-interventionist to qualify for the protections offered in the E-commerce Directive.

Question 26
Do you consider that clause 9 of Lord Lester’s Bill (at Annex C) is helpful in clarifying the law in this area? If so, are there any aspects in which an alternative approach or terminology would be preferable, and if so, what?

Overall, we would not suggest this approach be adopted. It creates additional terminology in an area which is already confusing. Also, given the broad ambit of the Ecommerce Regulations which apply to all forms of unlawful content, including defamation, it would seem more appropriate to amend the law at that point rather than to create one system for defamation with significantly different terminology and another for other forms of unlawful content. The category of ‘facilitator’ seems especially limited. Those who transmit are already immunized in relation to the material they transmit. The position of hosts is unclear, as there will only be a very narrow category of players who have no capacity to control material which they store. If such a provision is to be introduced, it would have to be clarified whether this is intended to cover all hosts who do not exercise editorial responsibility, or only those who do not have the ability to edit material uploaded to their servers.

The notice procedure set out in Clause 9 of the Lester Bill would certainly be more helpful in determining whether actual notice of a defamatory posting has been received by the intermediary than are the existing provisions under Regulation 22 of the E-commerce Regulations, which merely state that a court may consider certain factors in deciding whether this has happened. Lord Lester’s provision of a notice period is very welcome, as it has never been fully clear just exactly what a court would accept as ‘acting expeditiously’. The only context in which this has been clarified relates to terrorism: Section 3 of the Terrorism Act provides that, in relation to the presence of material which encourages terrorism and the dissemination of terrorist publications, a service provided notified of such material is expected to remove it within ‘2 working days’. Fourteen days is a rather generous period during which a service provider would have ample time to consult legal advisors rather than rushing to a snap decision.

Question 27
If Lord Lester’s approach is not suitable, what alternative provisions would be appropriate, and how could these avoid the difficulties identified above?

We remain in favour of the system of qualified liability as set out in the Ecommerce Regulations. There are situations in which it would be entirely reasonable for a service provider to be held liable. The existing position does, however, require to be tweaked in order to address concerns about intermediaries simply automatically complying with takedown requests rather than risking liability. In order to address this, we would support either an amended system providing for a ‘reposting provision’ in the manner of the US Digital Millennium Copyright Act, or a requirement that a party seeking removal of a defamatory posting first obtain a court order. Parliament would have to ensure that any such amendments are compatible with the UK’s obligations under the Ecommerce Directive.

To this we would also add a statutory notice requirement, as set out in the Lester Bill, in order to clarify what constitutes ‘actual notice’. This will also require clear guidance to be developed for intermediaries in relation to constructive knowledge and how this may also occasion civil liability even if the notice received from the complainant is not sufficient under a statutory notice scheme. Involving the courts in this process will be expensive and time-consuming, not least for private individuals without the cash reserves and legal support of celebrities or corporations. It is submitted that a fast-track, libel tribunal approach could benefit all here concerned. A much earlier decision on the natural and ordinary meaning of an article for defamation purposes could be incorporated into such a tribunal.

Question 28
Have any difficulties arisen from the present voluntary notice and takedown arrangements?

As already noted several times above, the ongoing problem is that online intermediaries in receipt of a complaint of unlawful content, including defamation, are placed in a position where they must make a judgement call as to whether to remove the material or not, with takedown being the most likely option in response to the risk of liability. Arguably the strongest cause of such reluctance to question complaints (which may well result in much legitimate material being removed; in the libel context it is particularly possible that a well-funded claimant could effectively use the threat of legal action to stifle even very valid criticism) is the sheer cost of defending a libel action. The impact of costs must be considered alongside any changes to the status quo of the substantive law.

Question 29
Would a statutory notice and takedown procedure be beneficial? If so, what are the key issues which would need to be addressed? In particular, what information should the claimant be required to provide and what notice period would be appropriate?

A clear, statutory notice system would (as noted above, Question 26) help to clarify exactly what constitutes ‘actual knowledge’ for the purposes of the Ecommerce Regulations. This would be best developed in that context, however, rather than separately for defamation. It must, of course, be borne in mind that for the purposes of civil liability (under both Section 1 and The Ecommerce Regulationsas presently constituted), it is enough that a court consider that a provider should have been aware of facts or circumstances which indicated the presence of unlawful material. Unless Parliament were to decide to require actual knowledge only for civil liability for defamatory articles on the part of hosts (which would, as noted above, go against the UK’s obligations in the Ecommerce Directive), it would need to be clarified for service providers exactly how strictly the notice requirements apply as it might be argued that even if not fully met there could still be enough information received to approach a constructive knowledge standard.

[1] See This response is broadly in favour of limited and cautious reform of the law of defamation in England and Wales. In line with BILETA’s remit, this submission concerns itself with those issues specifically relating to the impact of information technology upon defamation law. Specific questions addressed are 16 to 18 (‘Clause 6: A Single Publication Rule’), 19 to 20 (‘Clause 7: Libel Tourism’) and 23 to 29 (‘Responsibility for publication on the Internet’). This response has been approved by the Executive of BILETA (the British and Irish Law, Education and Technology Association and is therefore submitted on behalf of BILETA.

[2] Gavin Sutter is Lecturer in Media Law and a member of the Institute of Computer & Communications Law at the Centre for Commercial Law Studies, Queen Mary, University of London. He established, manages and teaches on the LLB and LLM Media Law courses at Queen Mary, as well as writing in the field. His key areas of interest are in the regulation of content in both traditional and new media, with a particular emphasis on the issues of defamation, obscenity, and intermediary liability. A member of the executive committee of BILETA (British and Irish Law, Education & Technology Association) since 2002, he is currently Vice Chair of the organisation. He also sits on the advisory board of Communications Law journal. Gavin’s publications include Media Law & Practice (Goldberg, Sutter & Walden, Oxford University Press, 2009).

[3] (1948) 298 NY 119

[4] [1849] 14 QB 185

[5] Supreme Court Procedure Committee Report on Practice and Procedure in Defamation ,, July 1991 at p81 para.VIII 2.

[6] [2002] EMLR 14

[7] Times Newspapers Limited (Nos 1 and 2) v United Kingdom (Applications Nos 3002/03 and 23676/03) March 10 2009

[8] Times Newspapers Limited, para 48

[10] [2000] 1 WLR 1004

[11] [2002] HCA 56

[12] Don King v Lennox Lewis [2004] EWCA Civ. 1329]

[13] [2005] EWCA Civ 75

[14] Reno v. American Civil Liberties Union, 521 U.S. 844 (1997)

[15] E.g. Zeran v AOL 129 F 3d 327 (4th Cir. 1997)

[16] Schneider v Case No 46791-3-I, 31 P 3d 37 (Washington Court of Appeal, September 17, 2001)

[17] Barret v Rosenthal 40 Cal.4th 33, 146 P.3d 510, 51 Cal.Rptr.3d 55 (Cal. Sup. Ct., November 20, 2006) Available online at

[18] See Goldberg D, Sutter G & Walden I [2009] Media Law, OUP, Oxford at pp423-424