Skip to main content

New immunity for websites which host defamatory user generated content



United Kingdom

New immunity for websites which host defamatory user generated content

This article first appeared in Entertainment Law Review, 1 June 2012.

A Rhys Griffiths is a Senior Associate and a member of the Defamation Group at Field Fisher Waterhouse LLP. Rhys acts for Claimants and Defendants and has particular expertise in online issues having acted for Amazon in McGrath v Amazon and ADVFN in Smith v ADVFN.


The recent case of Tamiz v Google (2012) is groundbreaking. It marks a sea change in the law applicable to websites which host defamatory user generated content, effectively providing complete immunity from defamation claims even after a complaint is made and the offending material drawn to the attention of the website host.

This is totally at odds with the law as it stood.  The legal position had been clear: website hosts were not liable for defamatory user generated content provided they did not in some way participate in its initial publication, for instance by vetting or editing it before publication. Thereafter, if a complaint was made, the website host had a choice to make: take down the material, in which case it would not be liable; or leave the material on the website, in which case it would assume responsibility for it and the Claimant could pursue the website host as well as the original author.

The 'intermediary' defence outlined above is provided for by the overlapping provisions of s.1 of the Defamation Act 1996 and Regulation 19 of the Electronic Commerce (EC Directive) Regulations 2002, as defined and developed by a body of case law starting with the landmark case of Godfrey v Demon Internet Limited [2001] QB 201. The use of the intermediary defence for website hosts is well known and well understood – both by lawyers and those involved in dealing with complaints about user generated content. However, this has all changed following Tamiz v Google.

The facts of the case

The case involved allegations posted about Mr Tamiz on the service provided by Google. is a platform which allows users in any part of the world to create an independent blog free of charge. Importantly, if the user does not have his/her own webpage they can use one hosted by, as was the case here.

The relevant blog concerned an article about Mr Tamiz’s resignation as a Conservative Party candidate for local elections in Thanet. This prompted a number of comments from readers of the blog, some of which made serious allegations about Mr Tamiz. For instance, one of the comments read:

"Anonymous said…

I know Mr Tamiz very well and am surprised that it has taken this long for all this to come out, Payam is a known drug dealer in thanet and has been taken to court for theft from his employers tescos in Ramsgate. His whole family are criminals his mother Mrs Sohela Tamiz has several convictions for theft and shoplifting and got sentenced at Maidstone crown court."

Mr Tamiz complained to Google about the various comments on the basis that they were defamatory. However, instead of removing them immediately in line with the intermediary defence outlined above, Google decided to forward the complaint to the author of the blog page, which in fact led the blogger voluntarily to remove the blog and the comments complained of. However, in spite of this, Mr Tamiz sued Google for the period the comments remained on the website after he had complained to Google until they were removed by the author of the blog.

Google's position

Google contended that it has no control over any of the content. It does not create, select, solicitor, vet or approve the content – this is all controlled by the blog owner. merely provides the tools for users to operate and maintain their sites. Moreover, as noted above, Google had no way of knowing whether the comments complained of were true or not, or whether they were eligible for some other defence. It cannot reasonably be expected to investigate and determine the truth or falsity of allegations made by bloggers when a complaint is made. Mr Justice Eady summarised Google's position as follows:

'[I]t may perhaps be said that the position is, according to Google Inc, rather as though it owned a wall on which various people had chosen to inscribe graffiti. It does not regard itself as being more responsible for the content of these graffiti than would the owner of such a wall.'

On this basis, Google contended that it was not a publisher of the relevant comments. In any event, even if it was, it was entitled to make use of the usual intermediary defence.

The decision

Mr Justice Eady agreed with Google. He noted that Google was merely a provider and facilitator of the means by which the defamatory comments were published. Its role was a passive one and it did not publish or authorise publication of the defamatory statements.  In adopting Google's analogy, Mr Justice Eady stated:

'It is no doubt often true that the owner of a wall which has been festooned, overnight, with defamatory graffiti could acquire scaffolding and have it all deleted with whitewash. That is not necessarily to say, however that the unfortunate owner must, unless and until this has been accomplished, be classified as a publisher.'

In addition, Mr Justice Eady held that there was no reason why this position should change following notification given that Google attempted to remain neutral on the issue and given that Google is not required to take any positive step, technically, in the process of continuing the accessibility of the offending material following a complaint. Accordingly, Mr Justice Eady held that Google should not be classified as a publisher of the comments complained of and so it can have no liability in defamation for those words.

In addition to this finding on the issue of whether Google was a publisher of the words complained of, Mr Justice Eady went on to rule that, if he was wrong about that, and Google was a publisher of the words complained of, it would have been entitled to rely on the usual intermediary defence.

The implications of the decision

This case has very far reaching implications. Google hosted the website on which the relevant words appeared and so under the law as it had been understood it would be deemed to be a publisher. Indeed, this is how the law was applied in Davison v Habeeb (2011), where HHJ Parkes QC held that it was arguable that Google was a publisher of the material hosted on the platform, at least following notification of the relevant material. In Tamiz, a complaint was made and so, following Davison v Habeeb, it would certainly be a publisher following the complaint and, in order to make use of the intermediary defence, it would have to remove the words. If not, Google would assume responsibility for the words in accordance with the usual principles.

In spite of previous case law, Mr Justice Eady decided that Google should not be categorised as a publisher of the words hosted, even after a complaint was made. The effect of this decision is to grant Google immunity from claims in defamation given the need to establish publication by Google to fix it with liability. As a result of this immunity, it matters not that Google was notified of the words complained of, and was asked to remove those comments – this would only be relevant if Google was a publisher and it sought to rely upon the intermediary defence.

The finding that Google was not a publisher of the words hosted means that websites which host user generated content can now ignore complaints about defamatory comments, whereas before they had to take down the material in order to ensure the applicability of the intermediary defence. This could lead to the continued publication of far more controversial and defamatory material than before. Indeed, controversy is often good for business given that it tends to attract users to a website. There will now be no need to remove this material upon complaint.

The other main implication of the decision is that it removes a simple and effective way for an individual to seek removal of defamatory material from a website. Under the existing law, all that an individual had to do was to bring the offending words to the attention of the website, explain why they were unlawful and ask that they be removed. Typically, a website host would accede to the request if it appeared to be a legitimate complaint so that it could have the benefit of the intermediary defence. Now, there is no legal reason for the website host to consider such takedown requests.

However, it would be brave of a website host to scrap its current takedown procedures in light of the decision in Tamiz v Google. The case is not a strong precedent: it is only a first instance decision, it is susceptible to being pigeon holed as specific to its facts (particularly given that it seems to cut across how the law was widely understood) and it might also be appealed and overruled. Indeed, as at the time of writing, an application for permission to appeal is pending in the Court of Appeal. Nevertheless, it might mark the beginning of a change in how the law of defamation is applied to websites which host user generated content and no doubt one which those websites will seek to develop and expand the next time they are in court.

Sign up to our email digest

Click to subscribe or manage your email preferences.