Defteros v Google LLC: Liability for search results and third-party publications
Defteros v Google LLC  VSC 219
Criminal law solicitor George Defteros has been awarded $40,000 in damages arising from Google’s publication, in its search results, of a hyperlink to an article carrying the imputation that Mr Defteros ‘crossed the line from professional lawyer for, to confidant and friend of, criminal elements’: Defteros v Google LLC  VSC 219 (Judgment). The two proceedings brought by Mr Defteros against Google were otherwise unsuccessful (including in respect of other pleaded imputations which were held not to have been conveyed).
The case arose from coverage of the so-called Melbourne gangland wars by The Age in 2004 and 2005. In particular, Mr Defteros’ claim concerned that newspaper’s coverage of the charges against him and Mario Condello for conspiracy to murder Carl Williams and others. To that end, The Age had published two articles in 2004 entitled ‘Underworld loses valued friend at court’ (the Underworld Article) and ‘Gangland’s price of peace’ (the Gangland Article). Each remained available on The Age website following print publication.
Additionally, a Wikipedia entry entitled ‘Melbourne gangland killings’ was created (the Wikipedia Article). That article outlined the charges against Mr Defteros, referenced the Underworld Article and included a hyperlink to it.
The charges against Mr Defteros were ultimately withdrawn.
In 2016, Mr Defteros initiated the first proceeding against Google when a search of his name returned a snippet of the Underworld Article and a hyperlink to the online publication of it (the Web Matter). In 2017, Mr Defteros initiated the second proceeding in relation to four additional matters, being:
- a composite image of photographs of Mr Defteros, Mr Condello, Mr Williams and George Williams;
- the Gangland Article, which was accessible via the above;
- images of Mr Defteros accompanied by the text “George Defteros. Carlton Crew lawyer. Recently got his license back. And got thru the ordeal alive! Fuckin lawyers!”; and
- the Wikipedia Article together with the Google results page that contained the link to the Wikipedia Article.
Google’s liability as publisher and no defence of innocent dissemination
Richards J was satisfied that each of the matters was published by Google upon Google being given notice of the particular search results: ,  and .
While the search function performed by Google was entirely automated and “is not capable of evaluating the meanings conveyed by the words and images on a webpage, including whether they are true, false, or defamatory”(), Google was able to manually alter the search results to remove particular webpages that would otherwise appear in response to a query: . Consistently with the earlier decisions of Trkulja v Google (No 5), Google Inc v Trkulja and Google Inc v Duffy (Duffy), her Honour rejected Google’s arguments that its search engine was a ‘passive tool’ () or that it lacked the requisite intent to publish. Her Honour found that the search engine was ”designed by humans who work for Google to operate in the way that it does, and in such a way that identified objectionable content can be removed, by human intervention, from the search results that Google displays to a user”: .
Her Honour accepted that notice to Google would be sufficient if it identified the alleged defamatory material. The higher threshold propounded by Google (against the findings in Duffy) – that it could only be liable if ”it knew or ought reasonably have known that the content in question was defamatory of the plaintiff” – was rejected: -.
Consistently with the above reasoning, Richards J also rejected Google’s defence of innocent dissemination. Once a reasonable time had elapsed since Google had been notified that it was publishing matter that was later found to be defamatory, the defence fell away. ‘Defamatory’ in this context did not mean ‘actionable’: -. A reasonable time in this case was 7 days: . Although that finding was based on evidence from Google’s employees about its capabilities to consider and respond to a complaint, it is likely to become the de facto standard in future cases.
Publication via hyperlink
Richards J concluded that “Google publishes a webpage that is reached by clicking on a hyperlink within a search result, because its provision of a hyperlinked search result is instrumental to the communication of the content of the webpage to the user”: .
Richards J rejected Google’s submission that its liability as a publisher was contingent on the information abstracted in a particular search result, declining to draw a distinction between search results which incorporated the defamatory material and those which were “content neutral”: . In making that finding, her Honour concluded that doing so would improperly elevate public policy considerations at the expense of the established common law principles of publication: . Concerns that subjecting hyperlinks to the traditional rules of publication would have the effect of seriously restricting the flow of information and free expression possibly underestimated the ‘vitality and resilience of the internet’ (), but more importantly were matters for the legislature.
No community of interest in publishing the matters
Google’s common law defence of qualified privilege failed. It submitted that its operation was for the “common convenience and welfare of society”: . It also emphasised the potential ‘chilling effect’ which could take place if individuals were unable to use search engines to locate information on particular subjects: . Her Honour concluded that even if such a proposition was accepted it did not go to establishing the relationship between Google and its users necessary to make out the defence: .
Contrary to Google having a ‘duty’ to respond to searches, it simply provided a service to users and the evidence did not establish that this was done as “a matter of legal, social, or moral duty”: . Moreover, the fact an individual used the Google search function was insufficient to establish such a relationship, noting that the interaction was automated: .
Statutory defence of qualified privilege
Google’s statutory defence of qualified privilege was partially successful.
For the purposes of engaging the defence under s 30(1) of the Defamation Act 2005 (Vic), Richards J was satisfied that at least some of the individuals who viewed both the Web Matter and the Wikipedia Article had an interest or apparent interest in having information on some subject: , . However, consistently with Duffy, her Honour was not prepared to hold that the mere use of the search engine was sufficient to establish the requisite interest: .
The question of whether Google’s conduct in publishing the Web Matter and the Wikipedia Article was reasonable focused on Google’s ‘Reputable Source Defamation Push Back Policy’ (the Policy) and its guidelines for assessing the credibility of a website as a reputable source (the Guidelines).
The Policy identified The Age as a reputable news source being ‘one of Australia’s leading broadsheets’: . In the circumstances, Richards J found that it was reasonable for Google not to remove the Underworld Article and direct Mr Defteros to raise his concerns with The Age itself. In reaching this conclusion, her Honour noted the nature of Google’s business environment and stated at  that:
Although its reach is vast, Google does not control the internet, and can only close one of many gateways to content that may be objectionable. It is reasonable for Google to rely on sources it knows to be reputable to determine what content should be published on their websites, and what should be removed.
In respect of the Wikipedia Article however, her Honour noted that contrary to its Guidelines, Google had not performed checks on the Wikipedia Article. Accordingly, her Honour was not satisfied that Google could classify Wikipedia as a reputable source, and its conduct in publishing the Wikipedia Article was not reasonable: -.
Triviality and consent to publication
Google was able to establish the defence of triviality in respect of the Wikipedia Article. Significantly, her Honour concluded that a large proportion of individuals to whom the Wikipedia Article had been published were assisting Mr Defteros in the prosecution of these very proceedings: . Her Honour also found that Mr Defteros had consented to publication of the matters insofar as it was made to employees who had been instructed to undertake particular Google searches: .
Insofar as other readers of the Wikipedia Article were concerned, she did not consider that they were looking for information about Mr Defteros: . Moreover, the references to Mr Defteros in the Wikipedia article were passing and “not especially eye-catching”: .
The triviality defence failed in respect of the Underworld Article. Her Honour was not satisfied that there “was no real possibility that [Mr Defteros] would sustain any harm at the time of each publication” of that Article, despite her acceptance that its publication did not cause Mr Defteros substantial harm: .
The case is an important step in the growing jurisprudence regarding search engines, and liability for third-party publications online generally. Whilst the parties have each now appealed, the judgment presently stands to encourage caution amongst those who provide hyperlinks to external content, and the serious consideration of complaints in line with well-thought out policies.
 Trkuljia v Google Inc LLC (No 5)  VSC 533, especially - (Beach J).
 Google Inc v Trkuljia (2016) 342 ALR 504,  VSCA 333, especially  and  (Ashley, Ferguson and McLeish JJA) (later reversed in Trkulja v Google LLC (2017) 263 CLR 149, but on different grounds).
 Google Inc v Duffy (2017) 129 SASR 304;  SASCFC 130 (Kourakis CJ, Peek and Hinton JJ).
 For example, as expressed in Crookes v Newton  3 SCR 269;  3 SCR 269, , ,  (Abella J).