Playing Canute with Defamation Law

November 1, 2003

Once again the issue of defamation and the Internet grabs our headlines as Internet Service Providers and others try to hold back the tide of rumour and speculation. In response to a frenzy of online speculation as to the identities of a number of Premier League football stars alleged to have been involved in the rape of a girl at a London hotel, a number of football-related Web sites, discussion boards and chat rooms have been withdrawn by their owners.

Not only have Web site and discussion forum owners faced the threat of legal action for the appearance of such content but, as is often the case, people have looked to ISPs to take responsibility both for the content that passes through their networks and the content that is hosted on their servers – material which, until brought to their notice, they know nothing about.

In English law, any disparaging comment made by one person about another, which is communicated or “published” to a third party may well be a defamatory statement which can give rise to an action for either libel (ie a comment in permanent form such as newspaper article or online) or slander (ie a comment made in conversation).

The Potential Defendants

The proper person to sue as the defendant will be the person who published the defamatory words or caused them to be published. Potential defendants would be as follows:

(i) the author of the words complained of

(ii) any editor of the Web site

(iii) the employer of the author and editor (if the works were published in the course of the author’s/editor’s employment)

(iv) the ISP upon whose servers the Web site appears if, and only if, they are fixed with notice of the potentially defamatory content and refuse to suspend access to it.


There are certain defences that can be relied upon in an action for defamation which will include the following.

(a) The person who allegedly defamed the complainant was neither the author nor the “publisher” of the statement.

(b) The complainant consented to the publication.

(c) The statement complained of does not refer to the complainant.

(d) The statement complained of is not, in fact, defamatory of the complainant.

(e) The allegations made, even if defamatory, are true (the defence of justification).

(f) The statement was published without malice to a person where the public interest in publishing the statement outweighs an individual’s right to protect his reputation (the defence of qualified privilege).

(g) The statement is a matter of comment (ie opinion based on true facts that a fair minded man could honesty hold) on a matter of public interest (the defence of fair comment).

Section 1(1) of the Defamation Act 1996 introduces a codified defence to a claim for defamation for those that are not viewed as the authors, editors or publishers of the defamatory comments by virtue of s 1(3). It is generally accepted that an ISP is able to avail itself of this defence to the extent that it did not know, or had no reason to believe, that its servers contained a defamatory statement and that it took reasonable care in relation to its publication as it does not fall within the definition of an author, editor or publisher.

What Is The Current Process?

The usual course of action for the person who has been defamed (as was well publicised in the case of Godfrey-v-Demon) is to complain to the ISP, thus fixing it with notice. Once the ISP has knowledge of the defamatory material it can no longer claim the benefit of the s 1(1) defence even if it has good grounds for believing (either through inquiries with the author or otherwise) that there is a defence of justification, fair comment or privilege that would apply to the statement.

In the event that the ISP refuses to suspend access to the Web site concerned, the person defamed may (as in the case of Mr Godfrey) be entitled to commence an action against the ISP for libel. In any action the claimant may claim damages for loss of reputation and for financial losses and an injunction restraining further publication.

The issue of taking “reasonable care in relation to the publication” of the defamatory content is a potential difficulty for ISPs. Particularly the wording of s 1 (5)(c) which provides that in determining whether reasonable steps have been taken regard shall be had to “the previous conduct or character of the author”. The Directive on Electronic Commerce (from which the Electronic Communications (EC Directive) Regulations derive) declared (in Article 15) that there was no positive obligation to moderate or monitor placed on ISPs. However, this provision has not been transposed into the Regulations. Does this mean that, once an ISP is put on notice of the fact that an individual has posted defamatory content, it must take steps to prevent that individual from doing so again in the future? What steps would be considered to be reasonable? The wording of s 1 (5) appears to impose on ISPs a positive requirement to monitor which is at odds with the provisions of the Electronic Communications Regulations.

These Regulations set out three different defences for ISPs – the first is the Mere Conduit defence (mentioned below) the second is the Hosting defence (also dealt with below) and the final one is the Caching defence which falls outside the scope of this article.

The mere conduit defence provides that, in principle, an ISP has no liability in damages or criminal sanction for transmissions via their network provided that (a) they did not initiate the transmission; (b) they were not involved in the selection of the recipients; and (c) there is no selection or modification of the information being transmitted. This defence would apply where the ISP is merely providing a conduit through which the information is sent.

In cases where an ISP is hosting Web sites that contain defamatory content, the principle again is that they have no liability in damages or criminal sanction for the storage of information if (a) they have no “actual knowledge” of any unlawful activity or reason to believe this to be in existence; and (b) on receiving actual knowledge they act expeditiously to remove the content to which the complaint relates. This defence will apply only where the user is not acting under any direction of the ISP in question.

Sources Of Concern

Unfortunately the Regulations do not clarify what actual knowledge means. However, determining factors will be (a) where the notice was received; (b) whether the full name and address of the sender was provided; (c) whether any information to assist in the location of the material being complained of was provided; and (d) whether there were any details of the unlawful activity.

A further source of concern is s 1(3)(e) of the Defamation Act 1996 which adds a potential complication for ISPs seeking to rely on the s 1(1) defence. It provides that ISPs will be deemed to not be either editors or publishers only where they are merely “involved as the operator of or provider of access to a communications system by means of which the statement is transmitted, or made available, by a person over whom he has no effective control”.

The issue here is that most ISPs do have terms of use governing their customers’ use of the service and many of those clauses prevent their customers from using their services to distribute or knowingly receive defamatory content. It is this contractual clause that could be argued to bring the ISPs within the definition of a publisher where it is their own customer making the defamatory comment. The argument could be that they do have “effective control” over their customers. This argument would not appear to apply in relation to those that are not their customers.

Where Does This Leave ISPs?

ISPs are seen as easy, tactical targets with deeper pockets than the authors of the Web sites who in many cases are inaccessible due to distance or simply anonymous. There is currently no “notice and takedown regime” that provides ISPs with immunity from liability where they act in good faith in response to a complaint about hosted material. In the absence of this statutory protection, ISPs are required to act as both judge and jury when making a decision about the legality of the hosted content.

Given that defamation is such a subjective issue, any decision to be taken by the ISP involves the weighing up of risk. The ISP may doubt the validity of the complainant’s case or may consider the information to be legal and may therefore choose to retain the information on the hosted Web site, thus risking action from the complainant. The flip side of the coin is that, should the ISP take action to remove or disable access to the content and the material subsequently turns out to be legal, the ISP runs the risk of legal action by the content owner. Given these issues to be balanced, it is difficult for the ISP to act expeditiously to remove the material as the Regulations specify.

A related issue is that there is no guidance on what constitutes a notice and who can put an ISP on notice. In the USA for instance the Digital Millennium Copyright Act prescribes precisely the information needed in order for a notice to be valid and requires the notice to be accompanied by a signed declaration of truth.

UK ISPs therefore have the additional burden of having to weigh up whether a notice is a valid one or whether it is vexatious. The ISPs are usually taking a look at complaints in a vacuum, based on the limited information that the complainant provides. ISPs are not in a position to ascertain all the facts pertaining to the case before action is taken. This has the effect of most ISPs erring on the side of caution and removing the material as soon as a complaint is received, regardless of whether the information is illegal or not; complainants often play to this practice as a tactical move to prevent comments that may in fact not be defamatory from appearing on a Web site.

There continues to be a general misunderstanding of the status of an ISP. They are considered by many as publishers (despite the provisions in the Regulations outlined above) – as evidenced by the injunction to suppress the new identity and whereabouts of Jamie Bulger’s killers amongst others which would have required pro-active monitoring of postings by ISPs. But there also appears to be a misunderstanding about the automated nature of the Internet. For instance, when a message is posted to a newsgroup, it is then automatically replicated and distributed to servers all over the world. The ISP has no control over this dissemination of the information and is unable to monitor all traffic passing through its systems due to the sheer volume involved.

ISPs have long been calling for a reform in the law granting statutory immunity when allegations of defamation are dealt with in good faith by disabling access to Web site content.

The Need For Reform

The Law Commission in its Preliminary Investigation into Defamation and the Internet (Scoping Study No 2, published December 2002) acknowledged that there “is a strong case for reviewing the liability of ISPs for material over which they have no editorial control but which they store on behalf of others. Although freedom of expression may legitimately be restricted in order to protect the reputation of others, it is important to ask whether this goal can be achieved without the regular removal of material which deals with matters of public interest and which may be untrue”. They also mentioned a couple of possible reforms to address the current dilemma facing ISPs such as following the US example of safe harbour provisions or an extension of the s 1(1) defence under the Defamation Act 1996.

Jessica Hendrie-Liaño is a partner at the Commercial law firm Beachcroft Wansbroughs and is also the Chair of the Internet Services Providers Association. She can be contacted at