Will internet service providers (ISPs) get caught in the iiNet (decision) for online defamation?

Take-away points

• An internet service provider (ISP) that is put on notice of potentially defamatory (or otherwise unlawful) content on its website/online platform should heed the notice and take steps to take down the offending content if there are means to do so.

• An ISP may have difficulty successfully relying on the defence of innocent dissemination if the allegedly defamatory matter is hosted on the ISP’s website/online platform in circumstances where the ISP has some ability to control (and take down) the content complained of.

The iiNet decision

The recent High Court decision in the much publicised iiNet case¹ will impact on the liability protection of ISPs in cases involving copyright infringements. But what impact the decision may have on liability exposure for ISPs notified of the uploading and posting of offensive, menacing, harassing or defamatory content on their platform will be interesting to watch.

It seems the position following the iiNet decision remains unchanged in relation to an ISP’s liability for innocent dissemination of defamatory material (both at common law and pursuant to section 32 of the uniform defamation legislation²) in Australia. That is, once an ISP has knowledge of the presence of such material, and they are in a position to be able to take down offending material from their websites/online platforms, then it is contended an ISP will be under a positive obligation to take positive steps to take down the offending content.

The decision in iiNet

In iiNet³, the High Court, in dismissing an appeal from a decision of the Federal Court below, held that iiNet had not authorised the infringement of the appellants’ copyright in various films and television programs by its customers because iiNet had no direct power or technology to actually prevent its customers from infringing copyright in the appellants’ films. The Court determined that the extent of iiNet’s ability to prevent its customers from infringing copyright was limited to an indirect power to terminate the contractual relationship iiNet had with its customers through the use (or in this case, misuse) of its user terms and conditions.

iiNet will likely impact on the capacity for an ISP to be held liable for copyright infringements. The decision is important because it was held that notwithstanding that iiNet was formally notified of its customers’ potential breaches of copyright, it was not held liable because it did not have the power or control (or the technology) to enable it to stop the peer-to-peer file sharing complained of. The only control iiNet could exercise in the circumstances was to suspend users or to terminate them for breaching iiNet’s user terms and conditions.

So, can the decision in the iiNet case be applied broadly in the defamation context to see a reduction in an ISP’s exposure to liability for content published by a third party via the ISP’s website or online platform? The answer is probably “yes”, though to a limited extent, and subject to certain conditions which are set out below.

ISPs and the defence of innocent dissemination

With the increase in use of social media networking sites such as Facebook, Twitter, MySpace and other personal blogs and online forums, the ability for individuals to publish and upload defamatory material on the internet is easier than ever.

Offensive posts made in such online environments have the capacity to “go viral” at the click of a button, and can have a disastrous impact on a person’s reputation or cause irreparable harm to a company’s business and trading reputation. Aside from being defamatory, content uploaded to the internet may also be offensive, menacing, harassing and used to intimidate its natural person victims. Such content may well be uploaded in breach of federal law.⁴ However, what liabilities do ISPs have to remove such content once they have been alerted to the presence of this offending online content?

ISPs and social media sites typically require users to agree to various terms and conditions in order for the user to be permitted to upload or view content. Often this includes the user agreeing not to publish material that is defamatory or would otherwise be deemed unlawful. Notwithstanding users agreeing to such terms and conditions, frequent breaches occur and defamatory, derogatory, offensive and threatening material continues to be uploaded by users posting under what they (mistakenly) believe is a cloak of anonymity.

Although this content can be uploaded quickly, removing the content and identifying those responsible can be problematic. In circumstances where the author of the defamatory and offensive content cannot be identified, or they have no means to satisfy a judgment, whether an ISP can be held liable for the infringing acts of its customers or website users is an issue that has not really been tested in the Australian courts.

The common law defence of innocent dissemination has largely been codified in the uniform defamation legislation.⁵ This defence is intended to protect those who unwillingly and unknowingly publish defamatory matter without negligence on their own part. For example, newsagents and paperboys who are arguably involved in the “distribution” of defamatory material will generally be able to rely on this defence because they typically play only a small “innocent” part in the process of publication of the defamatory material.

Web hosts and ISPs are usually conduits by and through which others publish material. However, the removal of defamatory or offensive material from these sites can be difficult for several reasons. First, there is often a lack of understanding from site administrators that they are under an obligation (if they have power and control over the content posted to their sites) to remove offensive and defamatory material drawn to their attention. Alternatively, many hold the (mistaken) belief that they will be protected by the defence of innocent dissemination.

Secondly, some blogs, internet forums and social media networks are small operations, run with small budgets, and so are not in a position to monitor all of the content uploaded to its website. With administrators of these websites often located overseas and dealing with a high volume of complaints, the response time from receipt of a complaint can take between minutes and weeks. Injury to reputation increases the longer a publication remains accessible online.

Lastly, users often upload defamatory content using false identities or “pseudonyms”, making the road to identification of the rogue uploader longer. In order to pursue defamation proceedings, the culprit principally responsible for uploading the offensive material needs to be identified. The task of identification can take longer, and be costlier, when the content is published by an “anonymous” user.

To effectively rely on this defence, an ISP must establish three things:

• it was not the primary distributor of the matter;

• it did not have knowledge (actual or constructive) of the publication; and

• its lack of knowledge was not due to any negligence on its own behalf.⁶

This issue has not been tested to any great extent in Australia, however there is a line of English authorities which have dealt with this issue.⁷ In the past, the English authorities have held that an ISP that has knowledge of defamatory content uploaded to its websites or servers but does take steps to remove the infringing content will not be able to rely on the defence of innocent dissemination.⁸ However, whether an ISP might be held liable for publication, notwithstanding the ISP has been put on notice of the offensive material, will turn on the factual circumstances of each case.

ISPs have been referred to as “mere conduits” of defamatory material.⁹ A distinction has been drawn between ISPs as publishers and hosts of defamatory content on one hand, and ISPs as mere conduits or facilitators of defamatory content on the other. Accordingly, an ISP may be able to rely on the defence of innocent dissemination where it did not host the defamatory material and did not have the ability to exercise power or control to amend or modify the content of the defamatory material. This can be compared to the decision in iiNet.

Recent English cases

Recently the English courts have adopted an approach in relation to the liability of ISPs that can be contrasted to the reasoning of the High Court in iiNet. In the matter of Tamiz v Google Inc & Anor¹⁰, Justice Eady held that notwithstanding that Google Inc. was put on notice of Mr Tamiz’s complaint, it could not be held liable as a publisher of defamatory material in the circumstances of that case.

In Tamiz, the complainant notified Google Inc. of various offensive comments which were uploaded in response to an original article posted of and concerning him on the website Blogger.com. It was discussed that Blogger.com is an internet service that enables users who do not have their own web address to use the Blogger.com service to upload their blogs. Google Inc. argued that it did not have control over the content posted to Blogger.com. It argued that it was a “neutral” service provider and was not a publisher of the allegedly offensive material.¹¹

In similar factual circumstances to iiNet, Mr Tamiz notified Google Inc. of his complaints by first using the “Report Abuse” mechanism on the Blogger.com site. Mr Tamiz then sent a formal letter of complaint to Google Inc (albeit some two months later). Google Inc. responded by informing Mr Tamiz that it would not be removing the content complained of, but requested permission to forward his complaint to the blog’s author. Three days after the complaint was forwarded, the author removed the page.

The issue at hand was whether Google Inc. could be deemed to be a “publisher” of the material it hosted, and whether it had any liability from the point in time at which it was notified of the alleged offensive comments on its hosted webpage. Google Inc. argued that although it was required to notify the author of the blog once a complaint was made, it was not required to remove the offending material from its sites because it was not in the position (practically speaking) to investigate the legality of each and every complaint it receives.¹²

Justice Eady held that in the circumstances of this case, Google Inc.’s role as a platform provider was a passive role, not to be confused with that of a publisher or one who authorises publication.¹³ In contrast to the position in iiNet, Justice Eady noted that although Google Inc. had the technical capability of removing the offensive material, it nevertheless could not be held to be an author or authoriser of the publication.¹⁴ Further, Justice Eady noted there was also no suggestion that Google Inc. was aware of, or had any knowledge of, the offensive material before Mr Tamiz first alerted it to the existence of the offending content.¹⁵

To sue or not to sue?

The answer to the question of whether or not proceedings can be commenced successfully against an ISP or web host appears in respect of defamatory material uploaded to a site it hosts remains unclear. The iiNet decision, if its principles in respect of copyright issues can be extended by analogy to the liability of ISPs for defamatory uploaded content, might have some impact on how this position is determined in Australia. What is apparent from the iiNet decision is that the liability of an ISP might depend on the degree to which it has knowledge of the offensive material, its capacity to exert control over the content uploaded to its sites, and its technical ability to actually remove offensive content from its sites.

In iiNet, the fact that iiNet did not have the technology to prevent the infringements of copyright from occurring was key. If extended to instances where an ISP is put on notice of defamatory content uploaded to its site, the decision in iiNet might mean that an ISP could not rely on the defence of innocent dissemination if it has knowledge of the offensive material and has the technical ability and control to remove the material complained of but does not take active steps to actually remove it.

As noted, the position in England is split in this regard. While there is authority to support the proposition that an ISP that has or had knowledge of the offensive material which was hosted on their website, and which did not take steps to actively remove it, could place the ISP in the chain of liability for failing to take active steps to remove the content from its websites, the recent decision in Tamiz moves away from this position.

If followed in Australia, Tamiz might mean that ISPs can escape liability for content published on their website (notwithstanding that they have the means to remove the content) on the basis that an ISP which hosts material and creates a mere “platform” for users to post their material is not deemed to have authorised the content of the material uploaded by its users. Whether this position is adopted in Australia remains to be seen, but the ability of ISPs to successfully rely on a defence of innocent dissemination remains a murky area.

Until the position is clear, ISPs which have the ability to remove offensive content from their website once notified of the presence of this material should continue to take steps to take down this material once they are put on notice of its presence or else potentially risk being unable to rely on the defence of innocent dissemination either under the Defamation Act or at common law.

For more information, please contact:

¹ See “The iiNet appeal – a win for creative industries”, Sabiene Heindal, Internet Law Bulletin (2011) Vol 14 No. 1, pages 6-9; and “The age of online copyright infringement”, Kimberley Evans, Internet Law Bulletin 2010 Vol 13 No.s 7 and 8, page 132- 136.