Defamation and the Internet

August 18, 2006

The recent fuss about Gina Ford’s complaints to the Mumsnet Web site has turned the spotlight onto how the law deals with the problems of libel on the Internet.

The ‘Mumsnet’ dispute started in January 2006 and hinged on some critical postings on the Web site’s discussion board.  One allegation to which Ms Ford takes vehement exception was that she ‘straps babies to rockets and fires them into south Lebanon’.  Judging by statements from both sides (available on the site), the initial complaint required the entire site to be closed down and damages paid.  The offending comments were removed, and Ms Ford’s demands were then reduced, but at the time of writing the dispute is still bogged down and it is unclear if an actual libel action will result.

Why has the case proved so difficult to sort out?  One reason arguably is that the English law of defamation at the moment fails to reflect the realities of the Internet. The prevailing view among judges seems to be that the Internet is just the latest example of mass communication.  Individual judges disagree, but have yet to change the way the law works.

The clearest expression of this controversy can be found in a recent case (Gutnick v Dow Jones [2002] HCA 56 – an Australian High Court case).  The view of the four majority judges was that:

‘the problem of widely disseminated communications is much older than the Internet and the World Wide Web.  The law has had to grapple with such cases ever since newspapers and magazines came to be distributed to large numbers of people over wide geographical areas…’

But the dissenting judge said this:

‘The Internet will require almost every concept and rule in the field … to be reconsidered in the light of this unique medium of instant worldwide communication.’

In England, the majority ‘unexceptionalist’ approach to the Internet still holds sway.  The law covering the Internet is to a great extent the same as that governing libel on television or in newspapers.  In contrast to other jurisdictions (notably the US), our system has only limited special exemptions to reflect the specific problems facing those running or hosting Web sites.

The general law of libel

A claimant has to show that a defamatory statement about him or her (or it, in the case of a company) has been published by the defendant to at least one third party.

The claimant does not have to prove that damage was actually caused by the defamatory statement: damage is presumed.

The defamatory statement is ‘published’ – ie the libel occurs – where the material is received, ie read by the third party.  Hence, in an Internet context, the wrong takes place where the material is accessed – not where the statement is typed or where it is uploaded to a server.  So, when a defamatory statement is transmitted on an ISP’s servers, the ISP ‘publishes’ the statement wherever it is accessed anywhere in the world.  This is subject to the recent decision in Bunt v Tilley and others [2006] EWHC 407 (QB) concerning purely passive Internet intermediaries (see below).

Furthermore, each time the statement is ‘published’ – ie each and every time it is accessed –  a separate and distinct cause of action is generated. 

And the right to sue runs for a year (the ‘limitation period’) from each separate publication. 

Whereas a newspaper will normally be safe from action a year after a particular day’s issue hits the streets, a Web site is permanently at risk as long as the material remains on its site – because each hit will be a new publication.

All the usual libel defences will apply – justification (truth) being the most potent.  One defence is of particular assistance in an Internet context: ‘innocent dissemination’.


Aspects of the law which assist Internet providers, hosts etc


Innocent dissemination

Back in the days when the web was just something made by a spider, newsagents and libraries had this defence of ‘innocent dissemination’’.  Those peripherally involved in the publication of a libel would not be liable if they could show that they were ignorant that they were publishing defamatory material, and that their ignorance was not the result of negligence. 

This defence was embodied into statute in the Defamation Act 1996, s 1, with the intention that it would give protection to the likes of ISPs.  In rather tortuous terms, the s 1 defence protects a libel defendant who can prove that:

(a)        he was not the author, editor or publisher of the statement complained of;

(b)        he took reasonable care in relation to its publication, and

(c)        he did not know, and had no reason to believe, that what he did caused or
contributed to the publication of a defamatory statement.’

‘Publisher’ under s 1(2) means ‘a commercial publisher, that is a person whose business is issuing material to the public, or a section of the public, who issues material containing the statement in the course of that business’.  But under s 1(3)(e): ‘a person shall not be considered the author, editor or publisher of the statement if he is only 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 key thing here is that an ISP (or other Internet intermediary with no knowledge of the defamatory material complained of) will lose the protection of s 1 if it is given notice of the defamatory material and does not delete that material.  This happened in one of the few English authorities in this area, Godfrey v Demon Internet [2001] QB 2001. The court held that the defendant ISP did come under s 1(3)(e).  But, because it had not deleted the offending post on its news server after it had received a complaint from Dr Godfrey, it fell foul of s 1(1)(b) and (c) – it did know about the defamatory posting, and so could not have the protection of the s 1 defence.

To retain the protection of the innocent dissemination defence, ‘reasonable care’ must have been exercised.  This begs the question of whether those actively involved in running a Web site with message board and similar facilities are expected to monitor each and every posting.  Would such monitoring be an essential element of ‘reasonable care’?  There has yet to be a decision on this, but practicality would suggest not (see further below, in relation to Mumsnet).

Abuse of Process


There are encouraging signs that the courts will not allow claimants to get by on a mere technicality. A 2005 case in the Court of Appeal (Yousef Jameel v Dow Jones & Co Inc [2005] EWCA Civ 75) suggests that, where there has been really trivial publication in England (eg just a few hits on a Web site), the courts will not allow a libel action to proceed even though the requirements are technically met by the claimant.  In that case there were five publications.  The Court of Appeal found that ‘It would be an abuse of process to continue to commit the resources of the English court, including substantial judge and possibly jury time, to an action where so little is now seen to be at stake … The game will not merely not have been worth the candle, it will not have been worth the wick.’


E-Commerce Regulations


Additional assistance for ISPs comes from Europe in Directive 2000/31/EC, (the ‘E-commerce Directive’).  The UK implemented this in the Electronic Commerce (EC Directive) Regulations 2002 (SI 2002/2013).

ISPs are exempted from defamation claims (in fact civil liability generally, and criminal liability to a limited degree) in certain circumstances. There are three different levels of protection:

·                     reg 17 gives complete immunity from liability to ISPs who act as ‘mere conduits’

·                     reg 18 gives qualified immunity to ISPs which undertake ‘caching’ activities (broadly speaking, they must have behaved in accordance with industry standards and must have removed material immediately on receipt of a complaint)

·                     reg 19 gives a measure of protection to ISPs which act as ‘hosts’ – obviously a key risk area – again dependent on prompt removal of material in response to notification.

As such, it might seem that the E-Commerce Regulations effectively overlay the protection of s 1 of the Defamation Act 1996.   It is open to question how much additional protection in practical terms they actually give in the defamation area.

In Bunt v Tilley and others [2006] EWHC 407 (QB), three ISPs were sued which the judge found had only acted as ‘conduits’, and had not been on notice of the alleged libels.  It was held that they were protected both by the relevant E-Commerce Regulation and by the innocent dissemination defence.  More generally, the judge stated that ‘an ISP which performs no more than a passive role in facilitating postings on the Internet cannot be deemed to be a publisher at common law.’ 

The judge confirmed that the Godfrey case was right to say that the s 1 defence would be removed if the ISP had notice of the complaint, but suggested that rather more detailed knowledge might be needed on the part of the ISP if it is to be deprived of the protection of the E-Commerce Regulations.  It may therefore be that the latter do provide rather stronger protection for ISPs than s 1.


Protection not available under English law.


It is worth noting that the English courts have specifically rejected a number of principles which apply in other countries and which protect those running and hosting Web sites. 

The most significant is that we do not accept the ‘single publication rule’.  In most of the states of the USA, each Internet dissemination (and each edition of a newspaper or broadcast of a television or radio programme) is regarded as a single publication, giving rise to only one action for damages, which can cover harm caused to the claimant in all jurisdictions throughout the world.  Hence, judgment for or against the claimant bars any other action between the same parties in all jurisdictions.  

Another difference from our rule is that in the US, the place of publication is where the publisher actually takes the steps necessary to publish the words. English judges have gone out of their way to rule out this approach in this country.



We have seen that receipt by an ISP or other Internet intermediary of notice of a libel complaint is likely to remove key protections which English law currently provides, and that it is also unclear what the legal position is in relation to monitoring and the ‘reasonable care’ requirement of s 1. 

Mumsnet have vigorously asserted the total impracticality of being expected to monitor each and every entry on their message board: they are a ‘shoestring’ operation, they say, run by a few individuals, and simply could not monitor the ‘1000 new discussions’ and ‘10,000 posts’ made on Mumsnet each day.   Many Web sites are in a similar position.

The result is that the best advice to those involved in any way with the provision of message board facilities and the like is that monitoring of each and every message is probably not required, but that any complaint about content must be responded to swiftly and the queried material removed.

The fact that, as the law stands, any complaint (well-founded or not) ought to lead to deletion is of course open to criticism.  Many argue it means that the law must currently be wrong, as it leads to a grotesque erosion of freedom of speech.  The essence of discussion boards (and the Internet generally) is to promote freedom of debate.  For every critical remark there will be contrary opinions, such that the overall ‘meaning’ of what one reads on the board will simply be that there is a debate, not that one side is right or wrong. 

It appears that Mumsnet did nonetheless respond by removing the postings complained of by Gina Ford.  Why is the dispute not at an end?  Not being in full possession of all the facts, it is impossible to be certain, but if it is the case that payment of damages and reimbursement of legal costs are also being sought, that will inevitably lead to complications.

The prominence given in the press to this particular dispute is likely to deter people from seeking to take steps about objectionable material about them on Web sites, which in theory ought to be reasonably straightforward as things stand.  Perhaps that is why such publicity has been generated in the media.

Speaking as a lawyer who has acted for individuals who have had problems with Web sites, my experience is that it is not always a good idea for someone with a complaint to risk prolonging matters and bringing the libel to a wider audience by asking for damages, costs and the rest, and risk forcing an impecunious Web site to dig in.  Because of the state of the law as described above, swift, clear notification to the Web site and the ISP plus a simple, firm request for the removal of the material complained about will usually do the trick. 


Nick Armstrong is a Partner at Charles Russell. Nick has specialised for 20 years in all aspects of media litigation and dispute resolution, particularly defamation, confidentiality and privacy, copyright, contempt and obscenity. He regularly advises prominent political and sporting figures, business people and cinema, TV and music performers.