Operators of WebsiteMuch has been written on the mental gymnastics required by the Defamation (Operators of Websites) Regulations 2013 (“the Regulations”) issued under section 5 of the Defamation Act 2013 (“the Act”). The policy background to the legislation was stated in the following terms in the Explanatory Memorandum for the Regulations:

“The current law can lead to website operators automatically removing material on receipt of a complaint to avoid the risk of being sued. This chills free speech, as it means that material which is not genuinely defamatory may often be removed from circulation. It also means that, where defamatory material is posted, action is usually taken against the operator rather than against the person who was actually responsible for posting the statement.’

The Regulations therefore, as explained by then Justice minister Lord McNally, aim to ‘defend free speech while giving those who are defamed a reasonable opportunity for redress’ and constitute ‘an attempt to ensure that the complainant and the poster are brought face to face, as it were, as easily as possible’.

It is suggested in this article, however, that the Regulations contain a flaw that might encourage some website operators to manage their comments section in a way which is both detrimental to the interests of responsible free speech in a democratic society and contrary to the spirit of the legislation as a whole.

The new Section 5 defence

Under section 5 of the Act, website operators will have a new defence to defamation claims brought in respect of user-generated content hosted on their websites. Website operators remain free to rely on pre-existing defences to defamation, such as section 1 of the Defamation Act 1996 or Regulation 19 of the EC Directive Regulations 2002, which already provides strong protection to a website operator whose involvement is limited to the storage of information provided by others.

The way that the new defence works is that by section 5(2) of the Act, it is a defence to an action for defamation brought against the operator of a website for the operator ‘to show that it was not the operator who posted the statement on the website’. That defence will be defeated, however, if the three conditions set under section 5(3) are satisfied, namely:

(a)   it was not possible for the claimant to identify the person who posted the statement,

(b)   the claimant gave the operator a notice of complaint in relation to the statement, and

(c)    the operator failed to respond to the notice of complaint in accordance with any provision contained in the regulations.

Thus at the heart of the new defence is the requirement for website operators to follow the new procedure set out in the Regulations made under section 5. In brief, the Regulations require a claimant, unable to identify the author of an allegedly defamatory statement online, to give the website operator a notice of complaint in relation to the statement, following which the operator must respond in accordance with paragraphs 2 to 9 of the Schedule to the Regulations.

The ‘paragraph 3 flaw’

However, it is only where the operator is able to contact the poster that the complex chain of actions outlined in paragraph 2 of the Schedule (namely notifying the poster of the complaint and notifying the complainant of the poster’s response, all within specific time limits) is set into motion. Where the operator has no means of contacting the poster as per paragraph 3, paragraph 2 does not apply. Under paragraph 3(2), an operator ‘is not taken as having a means of contacting the poster unless the means available to the operator include private electronic communication (for example electronic mail or other means of private electronic messaging)’. In other words, a website provider in possession of other details enabling contact with the poster– such as the poster’s name, residential address, occupation, date of birth, and so forth – would, somewhat surprisingly, not be deemed as having a means of contacting the poster under the Regulations (although in such cases a Norwich Pharmacal order may be pursued by the complainant to obtain this information).

Under paragraph 3(1), all the operator must do to keep the section 5 defence alive is (a) remove the statement complained of from the location on the website specified in the notice of complaint, and (b) inform the complainant that the statement has been removed – both within 48 hours of receiving the notice of complaint. In such cases, none of the hypothetical scenarios that envisage various responses from the poster set out in paragraphs 5-9 will apply. Furthermore, for the purposes of the Regulations, where action is to be taken by the operator within 48 hours of any point in time, any period of time falling on a non-business day in England and Wales is to be disregarded. An operator is not required therefore to monitor receipt of notices of complaint over the weekend. Practically, the operator need only ensure that an effective system for handling notices of complaint is in place, for example, by setting up and clearly publicising a designated email address for such notices, as suggested by the FAQs to the Regulations. It follows therefore that where a website operator runs a comment section requiring no registration at all – i.e. it gives the poster no option to leave any details permitting electronic communication – the complicated elements of the Regulations fall away. There is no need to notify the poster; no need to notify the complainant of the poster’s response; no decision as to whether or not to allow the statement complained of to remain posted and no assessment of whether a reasonable website operator would consider the name and address provided by the poster to be obviously false. The only thing such an operator will be required to do upon receipt of a valid notice of complaint is to remove the comment within 48 hours and notify the complainant that he has done so. Viewed from the prism of paragraph 3, the Regulations cease to appear unduly onerous to website operators.

The risks of such a flaw  

A close reading of the Regulations suggests that website providers can secure the benefit of the section 5 defence without the burden of the paragraph 2 procedural hurdlers so long as they themselves do not know the identity of those posting on their websites. The Regulations would, therefore, appear to be encouraging a situation whereby website operators increasingly permit entirely anonymous posting (requiring no email address or information affording other means of electronic communications). This scenario raises a number of problems:

  1.  The request by a website operator for the mandatory provision of some identifying information can act as a disincentive to the posting of abusive online comments. The abusive comments on websites such as Ask.fm, the Latvian-based social networking site enabling its 52 million registered users to post and answer questions anonymously are, unfortunately, but one illustration of the misuse that can come with complete anonymity. When considering the approach to take to anonymity under the new legislation, the Joint Parliamentary Committee’s Report on the Draft Defamation Bill stated its intention to ‘promote a cultural shift towards a general recognition that unidentified postings are not to be treated as true, reliable or trustworthy’. This result will tend to be frustrated by the paragraph 3 flaw. By discouraging website operators from recording contact information from their posters, the Regulations might precipitate instead a corresponding increase in online abuse. Such comments are unlikely to be the type of speech Lord McNally had in mind when describing the increased protection of freedom of expression that the Act was hoping to ensure.
  2. Further, increased reliance on paragraph 3 places freedom of speech in jeopardy. Although anonymity might increase the likelihood of abusive comments being made, it does not follow that all anonymous comments will be abusive. Where a website operator is able to contact the poster, the Regulations provide for some form of dialogue and enable the poster to communicate its wish for the statement to remain online. Conversely, a website operator unable to contact the poster but wanting to rely on the section 5 defence must remove all material which is allegedly defamatory within 48 hours of being put on notice. If reliance on paragraph 3 does indeed increase, the automatic take-down of material which is not in fact defamatory may increase as well. The chilling effect of such a result would be in direct conflict with the Act’s mission statement of striking a fair balance between freedom of expression and the protection of reputation.
  3. Upon receipt of a notice of complaint, a website operator allowing only entirely anonymous posting will know immediately that it has no means of contacting the poster for the purposes of the section 5 defence and could remove the allegedly defamatory post without delay. Yet the Regulations effectively grant such a website operator a 48-hour ‘immunity period’ (or longer if the notice of complaint is received on the weekend) during which it may continue to display the post and, in some cases, do so for the purpose of enticing further publicity. This immunity is not novel. It appears, in fact, to be significantly less generous than that afforded under section 1 of the Defamation Act 1996 – in Tamiz v Google Inc [2013] EWCA Civ 68, Eady J held that taking five weeks to remove a statement was not unreasonable for the purposes of the section 1 defence – but arguably a little more generous than that provided under Regulation 19 of the EC Regulations 2002: this requires providers of web hosting services to act ‘expeditiously’ to remove the unlawful content once they have been put on notice.
  4. Complaints relating to anonymous posts where the website has no means of identifying the poster will be dealt with solely by the website operator. The poster is out of the picture. Such a situation, if a frequent occurrence due to the allure of paragraph 3 as submitted in this article, would defeat entirely the procedure introduced by section 5 of the Act, having as its very purpose the increased resolution of disputes between the complainant and the poster where possible (see Lord McNally addressing his peers on the Regulations).
  5. The Regulations, premised on a notice-and-take-down model, encourage passive behaviour by website operators up to the point of receipt of a notice of complaint. This is particularly the case if the paragraph 3 flaw results in an increasing number of website operators desisting from requesting identifying information from their posters. The compatibility of the Regulations with the views expressed in the recent European Court of Human Rights case of Delfi AS v Estonia (no.64569/09) (now referred to the Grand Chamber) constitutes the subject of an article in itself. Nevertheless, it is interesting to note that the passive or reactive – as opposed to a more preventive – behaviour, which website operators may safely adopt under the Regulations, runs counter to the tenor of the judgment in Delfi.


Notwithstanding the issues raised in this article, the number of website operators that will deliberately elect to pursue the section 5 defence and take advantage of the paragraph 3 flaw may be overstated. Website operators may not wish to run the risk of tarnishing their reputation if they are seen to facilitate the increased publication of abusive comments online. Furthermore, the perceived complexity of the procedural hoops which an operator must usually jump through in order to avail itself of the section 5 defence may suffice to deter operators from electing to engage with the Regulations in the first place. They may instead choose to rely on existing defences under the E-Commerce Directive/Regulations. It is submitted however that of those who do, some will be quick to appreciate the potential ease with which the defence can be made out under paragraph 3 if the identity of the poster is anonymous to the complainant and the website operator alike. A rise in abusive posts, automatic removal of material which may not be genuinely defamatory, and actions taken against the operator rather than the poster represent somewhat uncomfortably both the potential consequences of the Regulations and the very mischief these were drafted to address.

Alexia Bedat is a barrister, currently working as a paralegal at 5RB. 

This post originally appeared as an article on 5RB’s website and is reproduced with permission and thanks