Google Inc v Duffy - [2017] SASCFC 130

SUPREME COURT OF SOUTH AUSTRALIA
(Full Court)
GOOGLE INC v DUFFY
[2017] SASCFC 130
Judgment of The Full Court
(The Honourable Chief Justice Kourakis, The Honourable Justice Peek and The Honourable Justice Hinton)
4 October 2017
DEFAMATION - PUBLICATION - GENERALLY - INTERNET PUBLICATIONS
DEFAMATION - PUBLICATION - GENERALLY - REPUBLICATION
DEFAMATION - PRIVILEGE - QUALIFIED PRIVILEGE - IN GENERAL
DEFAMATION - JUSTIFICATION - TRUTH - SUBSTANTIAL TRUTH AND CONTEXTUAL TRUTH
DEFAMATION - DAMAGES - GENERAL DAMAGES - ASSESSMENT - IN GENERAL
Appeal against a judgment that Google Inc (the appellant) defamed Dr Janice Duffy (the respondent) by publishing information which was the result of using the appellant’s search engine to search the respondent’s name.
The respondent paid for and received psychic readings over the internet. Those readings were false. The respondent engaged in an online conflict with the psychics. Defamatory articles about the respondent, calling her among other things a ‘psychic stalker’ and referring to her conflict with psychics were published on a website called the Ripoff Report (the Ripoff Reports). Other websites published material which derived from articles on the Ripoff Report (collectively the underlying webpages). When the appellant’s search engine was used to search the respondent’s name it reproduced content from, and hyperlinked to, the underlying webpages (the paragraphs). The Autocomplete function of the search engine also produced an alternate search term ‘Janice Duffy Psychic Stalker’.
In September 2009, the respondent requested that the appellant remove the paragraphs. The appellant did not take steps to remove the paragraphs until March 2011.
The Judge found that the appellant was a secondary publisher of some of the paragraphs from the time that the respondent had notified it that the search engine was producing the paragraphs. The Judge found that no relevant defences had been established. In a later trial, the Judge awarded the respondent $115,000 inclusive of interest in general damages but did not make an award for loss of earning capacity.
The appellant appeals the Judge’s findings as to liability. The respondent cross-appeals the quantum of damages and seeks a higher award.
Held per Kourakis CJ (Peek & Hinton JJ agreeing):
(1)     The Judge was correct to find that the appellant was a secondary publisher of the relevant paragraphs and knowledge of those search results should not be attributed to the appellant until notice is given.
(2)     The search results published by the appellant incorporated the defamatory material contained in the Ripoff Reports because it facilitated the reading of the defamatory material in a substantial and proximate way.
(3)     The fact that the defamatory imputations were contained in search results which repeated material from the underlying pages does not deny their capacity to convey a defamatory meaning.
(4)     To justify the imputation that the respondent stalked psychics the appellant had to show that the respondent persistently did, or had a strong proclivity to, make unlawful or improper posts on the internet which were calculated to shame or denigrate psychics or cause them substantial distress. On the facts the respondent stalked only one psychic on a single occasion and so the truth of the imputation was not justified.
Per Peek J (Hinton J agreeing) dismissing the appeal:
(1)     The Judge correctly delineated “the platform of facts” from which an inference of publication can be drawn and correctly found that three publications to Ms Palumbo and publications to “a substantial number of persons unknown” were established.
(2) For an individual to have an “interest in having information on some subject” sufficient to satisfy s 28(1)(a) of the Defamation Act 2005 the interest must be more than a matter of mere curiosity. A user does not, by merely entering a search term into a search engine, establish an interest so as to satisfy s 28(1)(a).
(3) It is insufficient to satisfy s 28(1)(b) that the words were published at the same time as words giving readers information about a relevant subject; there must be a “sufficient connection” between the defamatory words and the subject of interest.
(4) In relation to the three publications to Ms Palumbo, the Judge was correct in finding that the statutory defence of qualified privilege did not apply because the defendant did not prove that the defamatory matter was published in the course of giving her information on the subject of smoking. For the second and third publications, the defence did not apply for the additional reason that it was not proven that there was an interest within the meaning of s 28, rather than the searches being made from curiosity alone.
(5) Each cause of action in relation to the publications to the “substantial number of persons unknown” was made out unless the defendant proved qualified privilege in respect of each cause of action, which it failed to do. The requirements of s 28 must be applied to the facts of each publication; absent proof, it cannot be assumed that those who entered the search terms had the sort of interest in a subject required by the section, nor that the defamatory matter was published to them in the course of giving them information on that subject.
(6) In respect of each publication, having regard to the criteria provided for in s 28(3), including the actions of the defendant towards the plaintiff throughout her efforts to have the defamatory search results removed, s 28(1)(c) was not satisfied as the defendant did not prove that its “conduct in publishing the matter [was] reasonable in the circumstances”. The trial Judge correctly found as such.
Per Kourakis CJ (contra):
(1)     The Judge was correct to find that the appellant had published the search results to a substantial number of users. However, on the evidence the Judge overstated the number of people to whom the search results were published.
(2)     The onus on the defendant is to prove the circumstances that render the publication an occasion of qualified privilege.
(3)     The first class of persons to whom the paragraphs were published and who used the appellant’s search engine to search for the respondent’s name were in, or considering entering into, professional, commercial, employment or personal association with her.
(4)     The second class of persons to whom the paragraphs were published and who used the appellant’s search engine to search for the respondent’s name were interested in the controversy between the respondent and the internet psychics.
(5) The paragraphs were published to the first and second class of persons on a matter in which they had a legitimate interest in circumstances which were reasonable pursuant to s 28(1)(c) of the Defamation Act 2005 (SA).
(6)     It is not possible on the evidence to find that a relevant search was undertaken by a person who did not a member of one of those two classes.
(7)     The appellant has established the defence of statutory qualified privilege.
Per Kourakis CJ (Peek & Hinton JJ agreeing), dismissing the cross-appeal:
(1)     The quantification of damages is a discretionary exercise, the respondent has not established that the award of damages was manifestly inadequate.
(2)     The Judge was correct to find that the respondent’s failure to seek employment was independent of the appellant’s publication of the defamatory material.
(3)     The appellant’s failure to apologise and continued plea of justification is not a basis for the award of aggravated damages.
Obiter, per Kourakis CJ:
(1)     The tort of defamation requires that the defendant knows that the work contains defamatory material.
(2)     That knowledge is presumed conclusively in the case of a primary publisher.
(3)     That knowledge may be rebutted by a secondary publisher who does not know and could not reasonably have known of the presence of the material.
Obiter, per Hinton J:
(1)     The elements of the tort of defamation differ depending upon whether the defendant is either a primary or subordinate publisher.
(2)     Where the tort is committed by a primary publisher, no mental element attaches to the act of publication (other than that necessary to establish the act to be voluntary), hence the first ingredient of the tort is one of absolute liability.
(3)     Where the tort is committed by a subordinate publisher, the subordinate publisher is presumed to have published the material knowingly unless he or she can establish that they did not know, and could not reasonably have known, of the content of the material (the defence of innocent dissemination). Thus, where the publisher is a subordinate publisher, the first ingredient of the tort is one of strict liability in that the plaintiff need not prove any mental element to succeed, but a defendant can avoid liability by making out the defence of innocent dissemination.
(4) As to qualified privilege, the application of the statutory defence in this case turns on the actual interest possessed by Google search engine users searching “Dr Janice Duffy” or “Janice Duffy” between January 2010 and late 2010. The search terms used do not allow one to characterise the searcher’s interest as anything less than an interest in the respondent. That it would be so limited is highly improbable. The evidence does not support a finding on the balance of probabilities that the search engine users making the relevant searches possessed an interest falling within s 28(1)(a) of the Defamation Act 2005.
Defamation Act 2005 (SA) s 3, s 28; Defamation Act 1974 (NSW) s 20, s 22, referred to.
Phillips v Police (2016) 125 SASR 427, applied.
ACCC v Trading Post (2011) 197 FCR 498; Adam v Ward [1917] AC 309; Andreyevich v Kosovich (1947) 47 SR (NSW) 357; Austin v Mirror Newspapers Ltd [1986] 1 AC 299; Barbaro v Amalgamated Television Services Pty Ltd (1985) 1 NSWLR 30; Bashford v Information Australia (Newsletters) Pty Ltd (2004) 218 CLR 366; Blatch v Archer [2015] FCA 652; Bunt v Tillley [2006] EWHC 497 (QB); Byrne v Deane [1937] 1 KB 818; Crookes v Newton [2011] 3 SCR 269; Chappell v TCN Channel Nine P/L (1988) 14 NSWLR 153; Dow Jones & Co Inc v Gutnick (2002) 210 CLR 575; Echo Publications v Tucker (No 3) [2007] NSWCA 320; Guise v Kouvelis (1947) 74 CLR 102; Hockey v Fairfax Media Publications P/L [2015] FCA 652; Horrocks v Lowe [1975] AC 135; Howe & McColough v Lees (1910) 11 CLR 361; John Fairfax Publications Pty Ltd v Zunter [2006] NSWCA 227; Lazarus v Deutsche Lufthansa AG (1985) 1 NSWLR 188; Metropolitan International Schools Ltd (trading as SkillsTrain and/or Train2Game) v Designtechinca Corp (trading as Digital Trends) [2009] EWHC 1765 (QB); Morgan v Odhams Press [1971] 2 All ER 1156; Morosi v Mirror Newspapers Ltd [1977] 2 NSWLR 749; Morgan v John Fairfax & Sons Ltd (No 2) (1991) 23 NSWLR 374; Oriental Press Group Limited v Fevaworks Solutions Ltd (2013) 16 HKCFAR 366; Papaconstuntinos v Holmes ̀̀̀à Court (2012) 249 CLR 534; Roberts v Bass (2002) 212 CLR 1; Rogers v Nationwide News (2003) 216 CLR 327; Sands v State of South Australia [2011] SASC 146; Stephens v West Australian Newspapers Ltd (1994) 182 CLR 211; Stone v Moore (2016) 125 SASR 81; Tamiz v Google Inc [2012] EWHC 499 (QB); Tamiz v Google Inc [2013] 1 WLR 215; Telegraph Newspaper v Bedford (1934) 80 CLR 632; Thompson v Australian Capital Television Pty Ltd (1996) 186 CLR 574; Trkulja v Google Inc LLC (No 5) [2012] VSC 533; Urbanchich v Drummoyne Municipal Council & Another (1991) Aust Tort Reports 81-127; Webb v Bloch (1928) 41 CLR 331; Wright v Australian Broadcasting Commission [1977] 1 NSWLR 697, discussed.
A v Google New Zealand [2012] NZHC 2352; Aktas v Westpac Banking Corporation Limited [2010] HCA 47; Cush v Dillon (2011) 243 CLR 298; Emmens v Pottle (1885) 16 QBD 354; Emmerton v University of Sydney [1970] 2 NSWR 633; Griffith v ABC [2010] NSWCA 257; Macintosh v Dun (1908) 6 CLR 303; Oriental Press Group Ltd v Fevaworks Solutions Ltd (2013) 16 HKCFAR 366; Pritchard Englefield (A Firm) v Steinberg [2005] EWCA Civ 288; Rana v Google Australia Pty Ltd [2013] FCA 60; Restifa v Pallotta [2009] NSWSC 958; Thompson v Australian Capital Television Pty Ltd (1996) 186 CLR 574; Vizetelly v Mudie’s Select Library Ltd (1900) 2 QB 170; Wishart v Murray [2013] 3 NZLR 246, considered.

GOOGLE INC v DUFFY
[2017] SASCFC 130
Full Court: Kourakis CJ, Peek and Hinton JJ
KOURAKIS CJ.
Introduction ...................................................................................................... 2
Psychics and romantic disappointment.............................................................. 3
The Google search engine.................................................................................. 5
Google is notified................................................................................................ 7
Direct and circumstantial evidence of publication............................................. 8
The defamatory imputations............................................................................ 19
Elements of publication.................................................................................... 25
Innocent dissemination – the statutory defence............................................... 32
The trespass cases............................................................................................. 36
Publication on the World Wide Web................................................................ 38
Google is a publisher of search results and the Ripoff Reports...................... 55
Publication by inference................................................................................... 58
Capacity to convey........................................................................................... 65
Truth of the imputations................................................................................... 69
Qualified privilege............................................................................................ 74
Damages......................................................................................................... 100
Conclusion....................................................................................................... 105

PEEK J.
Introduction.................................................................................................... 106
THE MULTIPLE PUBLICATION RULE AND PROVING
INTERNET PUBLICATION ........................................................................ 109
The publications of defamatory matters to Ms Therese Palumbo................ 113
THE COMMON LAW QUALIFIED PRIVILEGE BACKGROUND...... ... 121
SECTION 28 OF THE DEFAMATION ACT 2005........................................ 125
Section 28(1)(a) & (b): The Google approach.............................................. 126
Section 28(1)(a) & (b): The requirement of an interest or apparent
interest in having information on the subject matter in question.................. 126
THE APPLICATION OF THE PRINCIPLES TO THE PRESENT
FACTS............................................................................................................ 138
SECTIONS 28(1)(c) & 28(3) – THE REASONABLENESS
REQUIREMENT............................................................................................ 146
The Google “notification procedure” and the correspondence..................... 149
THE RESPONDENT’S CROSS APPEAL..................................................... 167

HINTON J.
Publication and the Primary and Subordinate Disseminator........................ 168
Qualified Privilege.......................................................................................... 181

KOURAKIS CJ.
Introduction

This is an appeal by Google Inc (Google), the operator of an electronic search engine on the World Wide Web. Google appeals against a judgment that it defamed the respondent, Dr Duffy, by publishing to a substantial number of people information which was the result of using Google’s search engine to search her name. The search results were published in paragraphs which reproduced extracts from webpages. Those webpages, and accordingly the extracts, referred to Dr Duffy as a ‘psychic stalker’. Some extracts also accused Dr Duffy of misusing the office equipment of her employer, but the Judge found that that imputation was justified. The Judge awarded Dr Duffy $115,000.00, inclusive of interest, in general damages but did not make an award for loss of earning capacity. Dr Duffy cross-appeals against that award.
I agree with the Judge’s conclusion that Google was a secondary publisher of the paragraphs of both the search results and two of the underlying webpages from the time it was put on notice that its search engine was producing those results. I accept that the Judge was correct to find that the contents of the paragraphs, and their publication to a substantial number of persons, can be inferred from the search results printed out by Dr Duffy and from the evidence about the operation of the World Wide Web. I would affirm the Judge’s finding that the nature of the publication of search results does not negate the defamatory imputation which the paragraphs carry on their face. I also accept that the Judge correctly found that Google failed to justify the imputations that Dr Duffy stalked and harassed psychics. Dr Duffy’s posts were almost entirely robust, but not improper, criticisms of certain internet psychics. The single proved instance of improper harassment of one of the psychics did not justify the imputation because Dr Duffy was not shown to repeatedly engage in improper harassment or stalking or to have a proclivity to do so.
However, I have concluded that the Judge erred in finding that the persons to whom the paragraphs were published did not have a legitimate interest in the information conveyed by the search results. Publication through a search engine on the internet differs in an important respect from publication by the mass media in that it is published only to persons who search for the information provided. The evidence supports the conclusion that the persons to whom Dr Duffy proved, circumstantially, publication of the Google search result paragraphs were either in family, social, professional or business relationships with Dr Duffy, or were followers of the controversy between Dr Duffy and a number of internet psychics. Dr Duffy failed to prove that persons, other than those who had a legitimate reason to be informed about her on-line dispute with internet psychics, searched for her name and read the Google search results which carried the pleaded imputations. The evidence did not support a finding that the impugned paragraphs were published to a person with respect to whom Google could not show an interest greater than idle curiosity.
Furthermore Google’s conduct in publishing the paragraphs to persons with a legitimate interest in the material they contained was reasonable in that:
·Google’s internet search engine provided an important resource to the public; and
·even after it was notified of the nature of the paragraphs produced by its search engine it had no reason to conclude that the paragraphs were being published to persons with no legitimate interest in them or that the publishers of the material summarised in them were actuated by malice.
Dr Duffy’s claim should therefore be dismissed.
If the judgment on liability were affirmed, I would not have interfered with the award of damages.
My reasons follow.
Psychics and romantic disappointment

The events out of which the publication of the paragraphs arose can be briefly described. Dr Duffy is a medical researcher. In mid-2005, Dr Duffy met a man called Jon in New York and hoped that they might become romantically involved. From August 2005, Dr Duffy consulted and paid on-line psychics searching for prophetic reinforcement of her own hopes. She found it in spades. The psychics she consulted operated on a website known as Kasamba where psychics provided paid advice to clients. Some of the psychics on Kasamba went by the on-line pseudonyms of Master Z, Fruno, Powerful Visions (the psychic formerly known as Soul Connection to Dr Duffy), SunShiningUponYou (Sun). Bereft of any Delphic subtlety, they unanimously and unambiguously predicted a positive relationship.
Fruno, Powerful Visions and Master Z continued to predict a positive relationship, even after Dr Duffy reported to them that Jon had informed her by email in March 2006 that he had fallen in love with a married woman, doubted that he would come to Australia and wondered whether he even wanted to see Dr Duffy again. In December 2006 and January 2007, Sun continued to encourage Dr Duffy to believe that there was an amazing month ahead for Jon and Dr Duffy.
They were all wrong. In January 2007 Jon informed Dr Duffy that his girlfriend had divorced her husband, that he was enjoying his life and could not promise that he would ever return to Australia. When, in May 2007, Jon finally ended any hope of a relationship, Dr Duffy fell into a deep depression.
In June 2007, Dr Duffy started writing a report about her experiences with the psychics which she never published. She also complained to various psychics on Kasamba about her disappointment. On 18 July 2007, Dr Duffy posted complaints on a website called ‘the Ripoff Report’ against psychics including Master Z, Fruno and Powerful Vision. The Ripoff Report website enables persons who register with it to post a consumer complaint on the website about a supplier of goods or services. The website allows others to post support for the maker of the complaint. It also allows the supplier, or a supporter of a supplier, to respond to the complaint. The Ripoff Report website was also the source of information which was republished on the 123 People website, a website designed to find information about people on other websites and reproduce it.
On 29 July 2007, Dr Duffy created a chat group on yahoo.com for the purpose of enabling persons with complaints about Kasamba psychics to publish their experiences. The chat group was known as ‘kasambavictims’.
In September 2007, Dr Duffy communicated with Sun through the Kasamba website complaining about the previous predictions concerning Jon.
In December 2007, Dr Duffy, using the pseudonym Oswald Billet and the username ‘ozzieb’, posted messages to Sun using the Kasamba website. Billet claimed that his wife’s friend committed suicide on 17 December 2007 because of the dashing of romantic expectations which had been engendered by Sun. Under the ozzieb pseudonym Dr Duffy claimed that Sun was responsible for the death of his wife’s friend because she had been deliberately given false hope that she would be with someone who was married. Ozzieb threatened to inform the head of Kasamba, and the national media, of Sun’s wrongdoing. Sun did not respond. On 19 December 2007, again using the Oswald Billet pseudonym, Dr Duffy emailed Kasamba complaining that Sun’s wrongdoing had led to the suicide of his wife’s friend.
On 28 December 2007, Dr Duffy, under the user name ‘Friend’s Husband’ posted a report on the Ripoff Report about the suicide of his wife’s friend and complaining about Sun’s involvement in it.
On 30 December 2007, a person using the name ‘Mary anne’ posted a report on the Ripoff Report which described Dr Duffy as a psychic stalker and complained about Dr Duffy’s stalking of Kasamba’s internet psychics. On the same day, a person calling herself ‘Mary Anne’ responded to the report of ‘Friend’s Husband’ of 28 December 2007 claiming that it was a false report posted by Dr Duffy. More posts on the Ripoff Report followed, the posts complained about Dr Duffy’s stalking of psychics and referred to her as a ‘psychic stalker’. Dr Duffy became aware of the adverse reports on the Ripoff Report after being informed about them by a member of the Kasamba victims’ group.
On 23 March 2008, Dr Duffy, using a male pseudonym, engaged Master Z in an on-line chat session on the subject of finding a romantic partner. The very next day, Dr Duffy participated in another exchange with Master Z in which she accused Master Z of posting scripted readings. On the same day, Dr Duffy, using the pseudonym ‘Aussieboy’ posted a complaint about Master Z’s conduct, including the scripted responses, on the Ripoff Report website. She posted a further report using the pseudonym ‘Gretal’ on 26 March 2008. Sometime later in 2008, Dr Duffy under the username ‘Janice’, also posted complaints on the Ripoff Report about Fruno, Master Z and Powerful Visions. The evidence of these postings, in the form of hardcopy screen shots, was received by consent as Exhibit D5. The material in Exhibit D5 was said by counsel for Google to relate to the stalking and harassing imputation. The receipt of that exhibit was not conditional. Contrary to a submission put by Dr Duffy on the appeal, the Judge’s adverse rulings on amendments to Google’s defence did not limit the admissibility or use of the documents in Exhibit D5 for these purposes.
Even though at the time of the receipt of Exhibit D5 Dr Duffy had been under cross-examination over many days, Dr Duffy consented to the admission of Exhibit D5 after the Judge allowed a short adjournment for Dr Duffy to consider her position. On the appeal Dr Duffy contended that the Judge wrongly found that she had posted the accusations against Sun. That contention should be rejected. Dr Duffy admitted doing so. Moreover, the Judge was entitled to reject Dr Duffy’s testimony that the information in those posts was true and had been given to her by others whom she believed. The Judge was entitled to infer from Dr Duffy’s own testimonial admission to campaigning against the internet psychics and her inability to verify the identity of her informants, let alone their precise accusations, that she had fabricated the contents of her posts.
Dr Duffy’s on-line activities generated much internet material in which she was accused of being a psychic stalker. Google’s search engine identified, hyperlinked, and reproduced extracts of the web pages containing responses to Dr Duffy’s complaints. It was on the material reproduced by the search engine that Dr Duffy based her claims.
From herein when I refer to the trial Judge’s reasons, I refer to his Honour’s reasons on liability.[1] When I refer to the other reasons, for example the reasons on damages, I will specify.
[1] Duffy v Google Inc (2015) 125 SASR 437.
The Google search engine
The general operation of the World Wide Web and the Google search engine is described in the reasons for judgment of Blue J.[2] I cannot improve on his Honour’s exegesis and there is little reason to attempt to summarise it given the contemporary widespread familiarity with the internet.
[2] Duffy v Google Inc (2015) 125 SASR 437 at [23]-[42].
It is necessary, however, to set out the way in which Google displays its search results. The Google search results were referred to by the Judge as paragraphs. Examples of search results printed off by Dr Duffy and received into evidence appear in [11]-[13] of the Judge’s reasons.[3] Each paragraph comprises a number of elements. At the time Dr Duffy was searching, each paragraph had a unique electronic address referred to as a Uniform Resource Locator (URL) which acted as an electronic digital link (hyperlink) to the website to which it referenced (the underlying webpage). Clicking the mouse on an hyperlink accesses and displays the underlying webpage, including any written and pictorial materials contained on it. Each paragraph also has a title which is itself derived from the underlying webpage, and reproduces a selection of text from that webpage (the snippet). The title also operates as a hyperlink.
[3] Duffy v Google Inc (2015) 125 SASR 437.
The results displayed in response to a Google search are not static webpages. They are generated on each occasion that a user conducts a search and appear only on the screen of the searcher’s electronic device. They do not subsist on the World Wide Web after the search is made.
Google can, and could at all relevant times, prevent its search engine displaying search results referring to a particular webpage if it were provided with that webpage’s specific URL. By way of example, an aggrieved person could provide the URL to Google, or, the URL could be discovered by successfully searching for the webpage by reference to any information provided about that website. Blocking a URL has the effect that a search, using any search terms, will not produce a result referable to the webpage which has that particular URL. For instance, if a page were blocked at Dr Duffy’s request because it contained defamatory material, but it also contained Dr Duffy’s responses to that material describing her as a stalker then when someone later searched for a combination of search terms designed to return results relating to the criticism of internet psychics that page or URL would not be retrieved as a search result. That consequence is important because it shows that a step which would absolve Google of liability would deny access to information in which many persons may have a legitimate interest.
However, URL’s can be changed, and are, in fact, often changed by administrators of websites (web-masters). Indeed, there was evidence that the URL of the first Ripoff Report webpage changed in about June 2013.
Google cannot block search results by using the search terms themselves except for searches conducted in the nations of the European Union or on European Union domains. That technology was adopted by Google following the decision of the European Union Court of Justice in Google Spain v AEPD and Mario Costeja Gonzalez which established a ‘right to be forgotten’ for individuals.[4] The technology allows Google to block search results linked to a specified URL if the search enquiry includes a particular name. The evidence showed that there are a number of persons with the name Janice Duffy. Other than the appellant, one of those persons uses the title ‘Dr’. The name blocking technology could not be used to block the paragraphs defamatory of Dr Duffy without ‘overblocking’ search results with respect to other persons with the same name, but there was no evidence that those other persons were mentioned on the relevant webpages.
[4] Google Spain v AEPD and Mario Costeja Gonzalez (C-151/12) [2014] ECLI:EU:C:2014:317.
However, if that technology were available and deployed at the relevant time, it would have denied persons who had an interest in the specific controversy between Dr Duffy and the internet psychics, the capacity to search for the webpages by entering her name as the search term.
The issue of overblocking bears on the reasonableness element of statutory qualified privilege with which I deal in [308]-[326] below.
From 2014, Google implemented a system whereby it could prevent specified autocompletions of search queries being presented to a user when the Autocomplete facility predicted that the user was typing a person’s name in the search box. However that technology did not block an actual search of that same material using those search terms. That technology merely prevented the autocompletions from being shown to the user.
Dr Duffy printed in hardcopy the results of many of her searches. Those printouts were received into evidence. However, the actual search results displayed on the screen of Dr Duffy’s computer could not be seen or accessed by anyone else and were therefore not published to any other person. Nonetheless they established an evidential foundation for an inference, drawn by the Judge, that others who conducted a search with the same search terms would have seen a display of the same search results.
Google is notified
After Dr Duffy discovered that Google searches were generating paragraphs in which she was described as a psychic stalker she complained to Google and asked it to remove the links it was providing to the webpages where she was so described. Dr Duffy made her first request by email on 7 September 2009. Dr Duffy did so again on 8 September 2009 identifying the source webpage by name and URL. In mid-September Google requested Dr Duffy to provide more particulars of the material of which she complained. Dr Duffy provided that information no later than 23 September 2009. On 7 October 2009 Google responded (wrongly) that there was nothing it could do without the cooperation of the sites’ webmasters. Only in March 2011, after Dr Duffy instituted this action, did Google block paragraphs containing the abstracts which the Judge ultimately found were defamatory from searches made using its Australian website.
The Judge found that Dr Duffy’s email of 8 September 2009 provided sufficient information to enable Google to prevent its search engine from producing paragraphs which reproduced parts of the relevant Ripoff Reports. The Judge also found that on 22 September 2009 Dr Duffy provided sufficient information for Google to prevent the generating of paragraphs referring to the 123 People webpage. Neither finding is challenged. The Judge found that Google could reasonably have prevented the search engine from doing so by 7 October 2009. That finding is not challenged. Dr Duffy complained faintly that less time might have been allowed but any lesser period would only marginally affect the quantum of damages.
Direct and circumstantial evidence of publication
Apart from her own searches, Dr Duffy also adduced direct evidence of the publication of the results of a Google search of Dr Duffy’s name from the witness Theresa Palumbo, who was Dr Duffy’s hairdresser. Ms Palumbo’s evidence was accepted by the Judge.
A statement given by Ms Palumbo was received into evidence as exhibit P15. Ms Palumbo states that she knew that Dr Duffy was a health researcher and had researched chronic lung disease and the health effects of smoking. Ms Palumbo was aware that Dr Duffy had published papers with ‘a professor who was a leading expert in this field in Australia’. Ms Palumbo’s statement records that when Dr Duffy attended to have her hair done on 25 June 2010 they discussed Ms Palumbo’s desire to give up smoking.
The statement records that Dr Duffy suggested that Ms Palumbo search the internet for details of the professor to access his published work. The statement continues that when Ms Palumbo decided to conduct the search that night she could not remember the professor’s name. She then typed the words or some combination of the words ‘Dr Janice Duffy’ into the search engine. The search results revealed paragraphs referring to Dr Duffy as an ‘Australian psychic stalker’.
In her evidence, Ms Palumbo gave a slightly different account in answer to questions directly from Dr Duffy, who was then representing herself, omitting any reference to attempting to search for the professor’s name.
QYou have been my regular hairdresser ever since [2009].
AYes.
QI haven’t deserted you. Now you were aware of like girls talk at hairdressers and you know I chat etc. You were aware of my WorkCover issue.
AYes.
QBut you were also aware of the nature of my work at that time.
AYes.
QAnd what was that.
AYou were a researcher Health Department [sic].
QInto COPB, I mean you know lungs.
AYes.
QDid we have an occasion to have a discussion sometime in the middle of 2010.
AYes, we did because I was in the middle of trying to give up smoking.
QIt’s hard I know.
AAnd I booked an Allen Carr Seminar and I happened to tell you about it and you mentioned that you published a paper on the effects on smoking on lungs.
QDid I say, you know, give any details about the publication.
AYou said that if I wanted to look at it that I could look it up on line, so I Googled it.
QNow what did you actually Google.
AI Googled Janice Duffy.
Ms Palumbo then proceeded to testify as to the search results.
Ms Palumbo’s statement records that shortly after making that search she informed Dr Duffy of the contents of the search results. Dr Duffy told her that ‘she was already aware of the results, and that she had been trying to get them removed for some time’.
The Judge found that Ms Palumbo first searched for Dr Duffy’s name to find the article to which Dr Duffy had referred. The Judge also found that the paragraphs Ms Palumbo read related to what he described as the first and second Ripoff Report webpages because those reports referred to Dr Duffy as an ‘Australian psychic stalker’ or a ‘psychic stalker’ from Australia. The Judge also relied on the fact that the hard copy printed pages of Dr Duffy’s searches, which were conducted between January and October 2010, did not contain any reference to what was described as the third to sixth Ripoff Reports in the first few pages of search results.
The Judge therefore found that Ms Palumbo had seen at least the following paragraphs after making her Google search in June 2010:[5]
First Ripoff Report
Janice Duffy – Psychic Stalker Psychics Beware Of…
Psychics beware of psychic stalker Janice Duffy…
Cached
Second Ripoff Report
Rip-off Report Dr Janice M Duffy
Dr Janice Duffy Stop the Australian Psychic Stalker Dr. Janice M Duffy Adelaide South Australia Adelaide South Australia
Cached
[5] Duffy v Google Inc (2015) 125 SASR 437 at [268]-[269].
Ms Palumbo’s statement continues that sometime in around 2012, Dr Duffy mentioned that the Google search engine still loaded a page that contained results with derogative comments similar to those which she had come across in June 2010. Shortly after that conversation Ms Palumbo again Googled ‘Dr Janice Duffy’ and discovered similar search results. In order to make that search Ms Palumbo entered Dr Duffy’s name as the search term, however her statement does not mention any auto complete results.
However, in her evidence Ms Palumbo referred to another search she made in-between the two searches mentioned in her statement:
Q Did you at any other time between then and now do a search for my name.
AYeah, curiosity gets the better of you. You want to see if it is still there and yes, I did.
QWhen was that.
AI think I did it a few months after and of course I did it again this year, yeah. Even as I sort of start to type your name psychic stalker comes up straight away basically, yeah.
QWhere would that be.
AThat’s in the little search box, you know the Google auto – you know how you type.
QAuto complete, yes.
AIt thinks of what you want before you think of it.
Ms Palumbo testified that search results similar to those she first found were again displayed.
In cross-examination Ms Palumbo gave the following explanation for her search on the second occasion:
QThe second time you searched, was that as a result of a conversation you had with Dr Duffy.
ANo, it was more curiosity.
QAre you sure it wasn’t a case that she mentioned to you that it was still on Google and that is what prompted you to do the search.
ANo, we only talked about it a couple of times that year because she was trying to have it removed and then it was removed I think, and then just – it was just one day, you know, as you are on the computer, I just thought ‘I wonder if that’s still there’ and then I found it again; I searched for it and it was there again.

The Judge did not make any finding as to the reason for Ms Palumbo’s second search. Ms Palumbo’s interest or reason for conducting the searches in 2010 can be summarised as follows.
On the first occasion Ms Palumbo’s interest was to find the article on chronic lung disease to which Dr Duffy had contributed. For reasons on which I elaborate below, there is a sufficient connection between the publication of the debate between Dr Duffy and the internet psychics and a Google search for articles authorised by Dr Duffy such that Ms Palumbo’s enquiry would constitute an occasion of qualified privilege. The material was therefore published in the course of, and on an occasion of, qualified privilege.
Ms Palumbo’s interest on the second occasion was to ascertain whether Google had acceded to the requests of which she had been informed by her client and acquaintance Dr Duffy, to block search results with derogatory references to Dr Duffy. Ms Palumbo’s curiosity was not merely idle or that of an intermeddler. It arose from Dr Duffy confiding in her of attempts to have the results removed. For reasons on which I elaborate below, the search results with which she was provided fell within an occasion of qualified privilege constituted by her request to ascertain whether the material critical of her acquaintance was still being published.
The Judge found by inference from the search results tendered by Dr Duffy that Ms Palumbo saw the following paragraphs in late 2010 which referred to the first and second Ripoff Reports:[6]
Janice Duffy – Psychic Stalker! Psychics Beware Of…
Dr Janice Duffy is truly an embarrassment to her profession as a Senior Researcher …
Cached
Rip-off Report Dr Janice Duffy …
Dr Janice Duffy Stop the Australian Psychic Stalker Dr. Janice Duffy!! Adelaide South Australia Adelaide South Australia
Cached
The Judge also found that whilst conducting the search in 2012 Ms Palumbo saw words referring to Dr Duffy as a psychic stalker generated by Google’s autocomplete function.
[6] Duffy v Google Inc (2015) 125 SASR 437 at [274].
The Judge accepted that Ms Palumbo conducted the search in 2012 and that that that search was prompted by a conversation with Dr Duffy:[7]
[277]I find that Ms Palumbo chose to undertake the search in 2012 of her own volition and was not asked to do so by Dr Duffy, although what led her to undertake the search was a conversation with Dr Duffy during which Dr Duffy said that the Google search engine still loaded search results containing derogatory comments about her.
[7] Duffy v Google Inc (2015) 125 SASR 437 at [277].
That search too constituted an occasion of qualified privilege for the same reasons the second search did.
The Judge rejected Google’s contention that the publication to Mr Palumbo was not defamatory because Ms Palumbo did not believe the allegations to be true:[8]
[282]I reject Google’s contention. It is well established that publication is complete and the cause of action in defamation is good even if the publishee does not believe the imputation or give it any credence. Google’s proposition of law summarised in the previous paragraph was rejected by the English Court of Appeal in Dow Jones & Co Inc v Jameel. The doctrine developed and applied by the Court of Appeal in that case, namely that it may be an abuse of process to sue for defamation when the publication has been minimal and caused no significant damage to the claimant’s reputation such that the expense of an action is disproportionate to the available remedy, is inconsistent with Google’s proposition of law.
[283]I accept (without deciding) that there might not be an actionable publication if a plaintiff instigates a friend to access from a website defamatory matter solely for the purpose of the plaintiff relying on it as publication to give rise to a cause of action. However, while Ms Palumbo made her search in 2012 following and as a result of Dr Duffy telling her that the defamatory material was still on the internet, she nevertheless made that search of her own volition and it was not instigated by Dr Duffy.
(citations omitted)
[8] Duffy v Google Inc (2015) 125 SASR 437 at [282]-[283].
The Judge found that Ms Palumbo also conducted a search in April 2015 but was not satisfied that Ms Palumbo saw a paragraph on which Ms Duffy had sued and in respect of which she had notified Google. However, the Judge accepted Ms Palumbo’s evidence and found that there had been a publication in April 2015 by Google through its auto-complete function on that occasion. Ultimately he dismissed Dr Duffy’s action based on publication by the auto-complete function for different reasons.
Accordingly, the Judge found that Google published to Ms Palumbo:
1The first and second Ripoff Report webpages and paragraphs relating to them in June 2010 and late 2010; and
2The autocomplete term ‘Janice Duffy psychic stalker’ in 2012 and 2015.
To circumstantially prove a wider publication, Dr Duffy tendered screenshots of readable data generated by a Google website known as ‘Google AdWords’ (the AdWords data and site). Dr Duffy gave evidence that on 1 September 2011 she used the Keyword Tool on the AdWords website to search for data showing the number of average monthly searches made from Australia and the average monthly searches made globally for her name over the preceding 12 months by enquiring into the phrases ‘Janice Duffy’ and ‘Dr Janice Duffy’.
Dr Duffy made enquiries for both a ‘Broad Match’ search and ‘Phrase Match’ search. A Phrase Match search is conducted by either entering the phrase in inverted commas or checking the Phrase Match box and generates data on searches using the exact phrase or a close variation thereto. A Broad Match search is conducted by entering the words without using inverted commas or any other punctuation, or checking the relevant box. A Broad Match generates a search for words entered (in any order), a close variation, related searches or any other relevant variation. Enquiries for Janice Duffy using both Broad Match and Phrase Match search types showed 480 Australian monthly searches and 1,300 global monthly searches. Enquiries for ‘Dr Janice Duffy’ using the Broad Match function revealed 260 Australian searches and 320 global searches, monthly. The Phrase Match function identified 210 Australian searches monthly and 260 global searches monthly.
Dr Duffy conducted further Keyword Tool inquiries of the AdWord website. The Judge tabulated the Australian AdWord data of the number of monthly searches at [307] of his reasons. I reproduce the table below and I add to it, in chronological order, events, as found by the Judge, which may have affected the number of searches and from which the reasons for some searches may be inferred.[9]
[9] Duffy v Google (2015) 125 SASR 437 at [307].

Date Match type Janice Duffy Dr Janice Duffy Janice Duffy psychic stalker

16.2.2011 Action instituted

1.9.2011 Broad (Phrase) 480 (480) 260 (210)

9.10.2011
Dr Duffy creates a blog referring to the internet defamation of her among other things

21.11.2011 Article in The Australian referring to Dr Duffy’s action

2.3.2012 Broad 880 390 91

2.4.2012 Articles in The Age and Sydney Morning Herald about Dr Duffy’s action against Google.

April 2012 Mr Trkulja uses Google to search for Dr Duffy and Janice Duffy.

29.8.2012 Broad (Phrase) 590 (590) 260 (210)

26.10.2012 Broad 590 260 46

13.6.2013 Broad 320 140 110

I mention with respect to that table that the Broad Match search numbers for the terms ‘Janice Duffy’ and ‘Dr Janice Duffy’ necessarily include the Phrase Match for each search respectively. Moreover, the Broad Match search numbers, and probably the Phrase Match search numbers, for ‘Janice Duffy’ necessarily include the Broad Match search numbers for ‘Dr Janice Duffy’. It follows that the total search numbers are those for the Broad Match search of ‘Janice Duffy’. Of those searches the number of Phrase Match searches for ‘Dr Janice Duffy’ indicates searches by persons who knew of Dr Duffy’s academic qualification.
It is important to keep in mind that those numbers overstated the number of persons to whom search results were published because multiple searches might be undertaken by the same person, including Dr Duffy and her supporters.
A document generated by Google recording the number of searches of Dr Duffy’s name from different devices conducted between 1 August 2013 and 31 October 2014 was received into evidence by consent. That document showed that between those two dates there were searches from a total of 278 different electronic devices for ‘Janice Duffy’ and searches from a total of 196 different electronic devices for ‘Dr Janice Duffy’ using the Google Australian search engine (the Google search data). Again, it is likely that the 196 electronic devices from which searches for Dr Janice Duffy were made are included in the devices from which searches for Janice Duffy were made. The importance of the Google search data is that it excludes multiple searches by the same person. Google did not produce a readout from its data bases showing the number of searches of the Google Australian website for the name Duffy from different devices before August 2013.
The Google data was obtained by specially interrogating Google’s search engines for forensic purposes. In the ordinary course of Google’s operations such data was not produced nor kept in readable form. The Google data was based on searches from unique internet protocol addresses (IP addresses) whereas the AdWord data counted multiple searches from the same IP address. Because the Google data left aside the possibility of multiple users of the same device, and therefore the same IP address, it was a better indication of the number of persons who made the relevant searches. It is evidence that the Google search results were published to a significantly smaller number of persons in the period after Dr Duffy attracted publicity to herself by instituting her action against Google than the number of searches shown in the AdWord data might otherwise have suggested.
The Judge found that data of that kind would have been available for as far back as at least the 12 months ending in March 2011 had Google chosen to generate and retain it upon being served with the summons by Dr Duffy. I return to that finding in [199]-[209] below.
The Judge found that there was no reason to believe that the number of searches increased dramatically between October 2009 and August 2011. However, the Judge observed that in October 2011 Dr Duffy created a blog on which she referred to the use of the internet to defame her and that in November 2011, there was also media publicity about her action against Google. The increase in searches from that time is demonstrated by the table. The Judge found that between October 2009 and February 2011 there were at least 100 monthly searches for ‘Dr Janice Duffy’ and at least 200 monthly searches for ‘Janice Duffy’:[10]
[310]The earliest period for which data is now available is from September 2010 to August 2011 referred to at [306] above. That data shows large numbers of searches being conducted for both “Dr Janice Duffy” and “Janice Duffy” over that period. There is no reason to believe that the number of searches increased dramatically between October 2009 and August 2011. By contrast, in October 2011 Dr Duffy began her blog and in November 2011 there was media publicity about this action which was likely to increase searches for Dr Duffy’s name and this is borne out by the table at [307] above. I find that between October 2009 and February 2011 there were at least 100 monthly searches for “Dr Janice Duffy” and at least 200 monthly searches for “Janice Duffy”.
[10] Duffy v Google Inc (2015) 125 SASR 437 at [310].
Even though it is not clear, the Judge’s finding appears to be that there were an average of 300 searches monthly in the 12 months to, and including August 2011, 100 of which were for Dr Janice Duffy. Whichever finding was made, it was necessarily an estimate.
The Judge’s discounting of the number of searches was supported by evidence later given on the continuation of the hearing as to damages. A Google software engineer gave evidence that the Keyword Tool counted each move to a new page of results and each return to search results after visiting a hyper-linked web-page as a new search.[11] Indeed, that evidence perhaps makes the finding unduly favourable to Dr Duffy.
[11] Duffy v Google Inc (No 2) [2015] SASC 206 at [42].
The Judge found that the searches in Australia on the Google Australian website for ‘Dr Janice Duffy’ were probably undertaken by persons searching for information about the plaintiff. She was the only ‘Dr Janice Duffy’ in Australia at the time. The Judge found that the inclusion of the title ‘Dr’ strongly suggests that the searcher was not searching for another Janice Duffy. I would add that the fact that a search is made for Dr Janice Duffy rather than Janice Duffy renders it less likely that it was randomly or idly made.
It is important to observe, therefore, with respect to the Judge’s finding that:
·there is a large degree, if not a complete, overlap between the searches;
·one in two of the enquirers were sufficiently familiar with Dr Duffy to know of her academic qualifications;
·searches may have been conducted from the same device or devices by the same person;
·searches in any one month may have been conducted by some of the persons who made searches in other months.
It follows that relatively few persons measurable in the hundreds, and not thousands, read the search results.
The Judge found that searches for ‘Dr Janice Duffy’ between January and December 2010 returned the following paragraphs relating to the second and first Ripoff Report webpages respectively:[12]
Janice Duffy – Psychic Stalker Psychics Beware Of…
Dr Janice Duffy is truly an embarrassment to her profession as a Senior Researcher …
Cached
Rip-off Report Dr Janice Duffy …
Dr Janice Duffy Stop the Australian Psychic Stalker Dr. Janice Duffy!! Adelaide, South Australia Adelaide, South Australia
Cached
[12] Duffy v Google Inc (2015) 125 SASR 437 at [312].
The Judge found that it was likely that a significant proportion of persons searching for ‘Dr Janice Duffy’ read those paragraphs because the printout of the results of searches conducted in that time showed those paragraphs on the first page of results. The Judge rejected Google’s contentions that Dr Duffy’s search results were influenced by her earlier searches and might not have been typical of the paragraphs seen by other persons searching ‘Dr Janice Duffy’. Google’s appeal grounds complaining about those findings of fact are dealt with in [189]‑[193] below.
The Judge also found that:
· a significant proportion of the persons whose searches returned those paragraphs relating to the first and second Ripoff Report webpages are likely to have read the paragraphs; and
· a significant proportion of those persons who noticed the relevant paragraphs are likely to have followed the hyperlink to the Ripoff Report webpages.
The Judge found that a substantial number of persons in Australia conducted searches on the Google Australian website for the name ‘Dr Janice Duffy’ read the paragraphs relating to the first and second Ripoff Report webpages, and followed the hyperlinks to those webpages. The Judge concluded, therefore, that Google published the first and second Ripoff Report webpages and paragraphs relating to them to a substantial number of users between January and December 2010.
The Judge found that searches between January and December 2010 of the term ‘Janice Duffy’ would have produced at the least the following paragraphs relating to the first and second Ripoff Report webpages:[13]
First Ripoff Report
Janice Duffy – Psychic Stalker Psychics Beware Of…
Psychics beware of psychic stalker Janice Duffy…
Cached
Second Ripoff Report
Rip-off Report Dr Janice M Duffy Stop the Australian Psychic …[14]
Dr Janice Duffy Stop the Australian Psychic Stalker Dr. Janice Duffy!! Adelaide, South Australia Adelaide, South Australia
Cached
[13] Duffy v Google Inc (2015) 125 SASR 437 at [320]-[321].
[14] In one variation, the words ‘Stop the Australian Psychic’ are replaced by the number ‘#295925’.
The Judge, for similar reasons as those he gave for searches of the term ‘Dr Janice Duffy’, concluded that Google published the first and second Ripoff Report webpages, and paragraphs relating to them, to a substantial number of persons between January and December 2010, who had used Google to search the name ‘Janice Duffy’.
The Judge was not satisfied that a substantial number of persons who conducted searches in the terms above read paragraphs relating to the third Ripoff Report webpage because those paragraphs appeared only intermittently in searches undertaken by Dr Duffy and when they appeared they generally appeared on pages after the first two pages of search results. Dr Duffy did not sue in respect of paragraphs relating to the fourth to sixth Ripoff Report webpages and, in any event, those paragraphs appeared rarely, if at all, in searches undertaken by Dr Duffy for those names.
Searches of the name ‘Janice Duffy’ and ‘Dr Janice Duffy’ also generated paragraphs for another underlying webpage entitled Complaints Board but the Judge ultimately found that these paragraphs were not defamatory:[15]
Dr Janice Duffy Senior Researcher
Are you also a victim of the Dr Janice Duffy Senior Researcher? Submit a complaint to help other consumers to be educated and don’t let them get away…
Error! Hyperlink reference not valid./…/dr-janice-m-duffy-senior-researcher-a55917.html Cached
Dr Janice Duffy
Are you also a victim of the Dr Janice Duffy? Submit a complaint to help other consumers to be educated and don’t let them get away with it!
Error! Hyperlink reference not valid./…/dr-janice-duffy-a55921.html Cached
[15] Duffy v Google Inc (2015) 125 SASR 437 at [328], [364]-[365].
The Judge found that Google published those paragraphs to a substantial number of persons searching for the term ‘Dr Janice Duffy’ between January 2010 and October 2011.[16] Yet the Judge did not find that those paragraphs were published to persons who searched only the term ‘Janice Duffy’ because on a search of that name the Complaints Board paragraphs appeared only intermittently and when they appeared they generally appeared on pages after the first page of search results. The Judge was not satisfied that Google published any paragraphs to a substantial number of persons relating to any other Complaints Board webpages.
[16] The finding may be a typographical error which was intended to read October 2010 but is of no significance because of the low order in the search results of that paragraph.
The printout of searches conducted by Dr Duffy between January and 16 October 2010 for ‘Dr Janice Duffy’ also showed paragraphs for an underlying website known as ‘123 People’. Those paragraphs had in common the following words:[17]
Janice Duffy – Email Address, Phone Numbers everything!
Dr Janice Duffy Stop the Australian Psychic Stalker Dr Janice Duffy!…
- United States
[17] Duffy v Google Inc (2015) 125 SASR 437 at [334].
The Judge found that Google published the ‘123 People’ paragraphs to a substantial number of persons who searched ‘Dr Janice Duffy’ between January and December 2010. Paragraphs linked to another ‘123 People’ webpage appeared only intermittently and when the results appeared they were generally only found on pages after the first page of the search results. For that reason, the Judge was not satisfied that a substantial number of persons who conducted searches on the Google Australia website using either of the search terms read the paragraphs relating to the other ‘123 People’ webpage.

It is not obvious to me why the Judge limited the finding of publication to the calendar year 2010 notwithstanding his findings that Google could, acting reasonably, have blocked the offending users by October 2009 and that they were not, in fact, blocked until March 2011. However the more limited period of publication appears to reflect the period during which a finding of publication of the particular contents of the searches could confidently be made based on printouts of the searches conducted by Dr Duffy in that calendar year. For that reason I would dismiss Dr Duffy’s grounds of appeal which complain that the findings as to the specific period of publication were unduly restricted.
The Judge was satisfied that a substantial number of persons in Australia who conducted searches on the Google Australia website for ‘Janice Duffy’ between August 2011 and June 2013 read the words ‘janice duffy psychic stalker’ generated by Google’s Autocomplete function. The Judge based that finding on Ms Palumbo’s evidence and on the AdWords searches which showed that in March 2012, September 2012, and June 2013 an average of 91, 46, and 110 monthly searches respectively were made for ‘Janice Duffy psychic stalker’ over the preceding 12 months. The Judge found that it was likely that many of those searches were the result of the Autocomplete function. This was because when the words were generated by Google’s Autocomplete function, those searchers were very likely to have noticed the generation of the words in order to make the search. However the Judge found that no defamatory imputation was conveyed by the words so generated because the person undertaking the searches would have understood that the autocomplete function simply reproduced commonly made searches. I observe here that the acceptance of the auto-complete suggested term is some evidence that any searcher’s interest may have been the controversy between Dr Duffy and the psychics.
The defamatory imputations
The Judge concluded that the following paragraphs, which his Honour identified by reference to letters for the purpose of further considering the imputations found in them,[18] were published to a substantial number of persons between January and December 2010 and were defamatory. I have included beneath each paragraph the imputations which the Judge found they contained.
[18] Duffy v Google Inc (2015) 125 SASR 437 at [346]-[351].
First and Second Ripoff Report webpages (results obtained by searching for ‘Dr Janice Duffy’):
A     Janice Duffy – Psychic Stalker Psychics Beware Of…
Dr Janice Duffy is truly an embarrassment to her profession as a Senior Researcher …
Cached
1.Dr Duffy stalks clairvoyants and others who have a claim to have paranormal powers (psychics)
2.Dr Duffy is an embarrassment to her profession
B, D Rip-off Report Dr Janice Duffy …
Dr Janice Duffy Stop the Australian Psychic Stalker Dr. Janice Duffy!! Adelaide, South Australia Adelaide, South Australia
Cached
1.Dr Duffy stalks psychics
First and Second Ripoff Report webpages (results obtained by searching for ‘Janice Duffy’):
C     Janice Duffy – Psychic Stalker Psychics Beware Of…
Psychics beware of psychic stalker Janice Duffy…
Cached
1.Dr Duffy stalks psychics
E     Janice Duffy – Email Address, Phone Numbers, Everything!
Dr Janice Duffy Stop the Australian Psychic Stalker Dr Janice Duffy!…
- United States
1.Dr Duffy stalks psychics
The Judge also found that Google published the underlying webpages linked to paragraphs A and B to a substantial number of persons between January and December 2010 being the webpages at the following URLs:
H First Ripoff Report web page:
first Ripoff Report webpage);
I Second Ripoff Report web page:
http:/ deb8p.htm (the second Ripoff Report webpage).
The first Ripoff Report included the following relevant text, the comments on that text are the numbered paragraphs:
Report: Janice Duffy – Psychic Stalker! Psychics Beware of Australian Psychic Stalker!
Reported by: (Riverton Wyoming)
Janice Duffy – Psychic Stalker! Psychics Beware of Australian Psychic Stalker! Janice Duffy Stalked me on the computer for several months. She uses the name Janice and other phoney names and keeps on sending you her name over and over again. I know of her because she stalked me on the psychic website Kasamba and other websites. I am sad to say that she has stalked other psychics as well. Australia Adelaide internet.
…
Psychics must beware of a psychic stalker named Janice Duffy. … She threatens psychics to abide by her rules or else she will blackmail them by writing rip off reports about them. … She … harasses psychics over and over again. She will not stop and has a stalker like mentality. She spreads melicious [sic] lies and gossip about people in hopes to gain sympathy for her life. She cannot even work right now and has been laid off by the hospital in which she works because she cannot even function on a day to day life. … She is also someone who spreads lies and says that she is another person from Australia … She also makes up lies about people dying and committing suicide. … She … stalks psychics for more information and if she doesn’t get any, she writes fake and phoney information about them on-line. … Nobody likes to be stalked and Janice Duffy will not stop until the psychic community becomes aware of who Janice Duffy is and the fact that she has a serious stalking problem. …
…She is a PHD and should not be using her power to try to harass and talk psychics. She works for a hospital and it’s a shame that she knows the rules of the psychic websites and yet stalks psychics continuously on them. Psychics, please beware of this woman. You have been warned. If you wish to file a harassment lawsuit against her for stalking, please write to me and I will give you her real mailing address so that you can call her local police station and have her stopped for stalking. She harasses continuously. …
Mary anne
Riverton, Wyoming
USA
This report was posted on ‘Ripoff Report on 12/30/2007 4:00:21 PM and is a permanent record located here: Ripoff Report has an exclusive license to this report. It may not be copied without the written permission of Ripoff Report.
#1 Story sounds similar to my friend’s experience with Janice
Katie – Smalltown (USA)
… Dr Janice Duffy is truly an embarrassment to her profession as a Senior Researcher in Adelaide, Australia.
…
#3 Dr. Janice Duffy uses her government work email address to email anti Kasamba psychic clients.
Katie –Smalltown (USA)
…I can’t believe she used her government email for personal purposes and it proves how stupid she is. This will be reported to her superior for breaking the law. …
…
The Judge found that the first Ripoff Report made the following defamatory imputations:[19]
[19] Duffy v Google Inc (2015) 125 SASR 437 at [368].
1.    Dr Duffy stalks psychics;[20]
2.    Dr Duffy harasses psychics by persistently and obsessively pursuing them;[21]
3.    Dr Duffy misused her government work email address by sending emails for non-work or other wrongful purposes;[22]
4.    Dr Duffy spreads malicious lies and gossip about other people with a view to gaining sympathy for herself;[23]
5.    Dr Duffy spreads lies about people dying;[24]
6.    Dr Duffy spreads lies about people committing suicide;[25]
7.    Dr Duffy is an embarrassment to her profession; [26]
8.    Dr Duffy is unable to function in day-to-day life;[27]
9.    Dr Duffy has been laid off by the hospital where she works because she cannot function in day-to-day life.[28]
[20] ‘Janice Duffy – Psychic Stalker…she stalked me on the psychic website Kasamba and other websites. I am sad to say that she has stalked other psychics as well…she has a serious stalking problem…stalks psychics continuously…please write to me and I will give you her real mailing address and you can call her local police station and have her stopped from stalking…In most countries stalking is against the law’.
[21] ‘harasses psychics over and over again. She will not stop … has harassed psychics on an ongoing basis … She harasses continuously’.
[22] ‘I can’t believe she used her government email for personal purposes and it proves stupid she is. This will be reported to her superiors for breaking the law.’
[23] ‘she spreads malicious lies and gossip about people and hopes to gain sympathy for her life.’
[24] ‘she also makes up lies about people dying and committing suicide.’
[25] ‘she also makes up lies about people dying and committing suicide.’
[26] ‘truly an embarrassment to her profession’.
[27] ‘she cannot even function on a day-to-day life’
[28] ‘she … has been laid off by the hospital in which she works because she cannot even function on a day to day life’
The relevant parts of the second Ripoff Report read:
Report: Dr Janice Duffy
Reported by: (Riverton Wyoming)
Dr. Janice Duffy Stop the Australian Psychic Stalker Dr. Janice Duffy!! Adelaide, South Australia Adelaide, South Australia
…
Dr Janice Duffy P.H.D has been stalking psychics for a long time now and she must be stopped. Her harassing emails have caused many psychics to go into hiding because of her blackmailing and forcing psychics to respond to her emails. She is from Adelaide, South Australia and has made numerous rip off reports and has lied about many psychics on-line. … If you don’t do what she says, then she is known to use blackmail and says that she will write rip off reports about you… Dr. Janice Duffy has tried to blackmail numerous psychics begging them for free updates and email updates and if you don’t respond to her, she starts creating rip off reports with many lies to try to gain sympathy from people who will respond to her posts. … she has also written fake and deceptive rip off reports about psychic websites. …After the first few psychic readings, she tends to write emails to psychics for updates and when she doesn’t hear back from you fast enough, she begins to send you hate mail which usually says that she is going to ruin your name and reputation on-line as a psychic. …
Mary anne
Riverton Wyoming
U.S.A
This report was posted on Ripoff Report on 12/31/2007 1:54:36 PM and is a permanent record located here: Ripoff Report has an exclusive license to this report. It may not be copied without the written permission of Ripoff Report.
…
#2 Australian Dr. Janice Duffy senior researcher stalks Kasamba psychic advisors and uses her Dr. title to allow people to believe that she is being ripp
Mary Anne – Riverton (U.S.A.)
Australian Dr. Janice Duffy senior researcher stalks Kasamba psychic advisors and uses her Dr. title to allow people to believe that she is being ripped off by psychics. She is responsible for most of the false rip off reports created about Kasamba … Stop stalking psychics doctor Janice Duffy ...
Dr. Janice Duffy is responsible for the numerous fake rip off reports about psychics … she wants to stalk psychics until she gets her way. … She is a very sick woman and wants to stalk and destroy Kasamba and their psychics… She won’t stop until she is locked up or sued. … She …manipulates and threatens people to give her what she wants. …
As you can see, she is very equipped to stalk psychics and does this on a regular basis. … She makes up lies and spends her free time stalking … her obsession cannot end. Stalking is a mental illness and unless she gets treatment, she cannot stop herself. …
… She is using blackmail and is currently writing and telling others what she is going to be doing. Your name may or may not be on her list. … Dr. Janice Duffy … will make it her priority to stalk you. Stalking is a serious illness and she cannot stop until she is medicated or put in jail. … She … stalks daily. … She has been stalking for a long time now … Please also note that many people in her anti psychic groups feel that she has email hacking software or knows and uses hackers. It is rumoured in some of her anti psychic groups that she tries to get your personal email address and then tries to hack into it. … What is even more bizarre is that Dr. Janice Duffy has used her government email address to email anti Kasamba clients. This can be tracked by the governments webmaster. We will be forwarding this report to her superiors as a way to gain court evidence that she did indeed write and receive emails from her anti psychic kasamba group members with her government email address; therefore breaking government rules. You cannot use a company’s email address for personal purposes. …
…
The Judge found that the second Ripoff Report made the following defamatory imputations:[29]
[29] Duffy v Google Inc (2015) 125 SASR 437 at [371].
1.    Dr Duffy stalks psychics;[30]
2.    Dr Duffy harasses psychics by persistently and obsessively pursuing them;[31]
3.    Dr Duffy misused her government work email address by sending emails for non-work or other wrongful purposes;[32]
4     Dr Duffy without their permission, fraudulently and/or maliciously accesses other peoples emails, stored electronic materials and/or electronic memberships; [33]
5.    Dr Duffy made unfair, inaccurate and damaging reports about psychics;[34]
6.    Dr Duffy has disseminated lies about psychics;[35]
7.    Dr Duffy spreads malicious lies and gossip about other people with a view to gaining sympathy for herself; [36]
8.    Dr Duffy spreads lies about people dying;[37]
9.    Dr Duffy has engaged in criminal conduct;[38]
10.    Dr Duffy threatens and manipulates people to further her own ends.[39]
[30] ‘Dr Janice Duffy PHD has been stalking psychics for a long time now and she must be stopped….now this stalker is being exposed to stop her… she is very equipped to stalker psychics and does this on a regular basis … stalking is a mental illness and unless she gets treatment, she cannot stop herself … stalking is a serious illness and she cannot stop until she is medicated or put in jail … She is from Australia and stalks daily … She has been stalking for a long time ago now.’.
[31] ‘Her harassing emails have caused many psychics to go into hiding… She likes to go to psychic websites and harass psychics … She has a crazy obsession with psychics and simply won’t stop harassing them’.
[32] ‘What is even more bizarre is that Dr. Janice Duffy has used her government email address to email anti Kasamba clients. This can be tracked by the governments webmaster. We will be forwarding this report to her superiors as a way to gain court evidence that she did indeed write and received emails from her anti psychic kasamba group members with her government email address; therefore breaking government rules. You cannot use a company’s email address for personal purposes’.
[33] ‘Please also note that many people in her anti-psychic groups feel that she has email hacking software or knows and uses hackers. It is rumoured in some of her anti-psychic groups that she tries to get your personal email address and then tries to hack into it.’
[34] ‘she starts creating Ripoff reports with many lies… She has also written fake and deceptive rip-off reports about psychic websites …She is responsible for most of the false rip-off reports created about Kasamba …’.
[35] ‘she… has lied about many psychics on-line… she starts creating rip-off reports with many lies … still making false claims against the psychics … She makes up lies … ‘
[36] ‘she starts creating a rip-off reports with many lies to try to gain sympathy from people who will respond to her posts’.
[37] ‘she is making up stories about death’.
[38] ‘Janice is committing a crime.’
[39] ‘she is a doctor and knows how to manipulate innocent people… She knows how to manipulate …She manipulates and threatens people to give her what she wants. ‘
It is important to note here that the author of the Ripoff Reports, who was in effect replying to Dr Duffy’s posting, refers to her PhD qualifications and alleges that Dr Duffy was inappropriately using her academic title to add weight to her criticisms of the psychics.
Elements of publication
The trial Judge usefully identified that the tort of defamation has the following elements:[40]
1.    the defendant participates in publication to a third party of a body of work;
2.    the body of work contains a passage alleged to be defamatory;
3.    the passage conveys an imputation;
4.    the imputation is about the plaintiff;
5.    the imputation is damaging to the plaintiff’s reputation.
(citations omitted)
[40] Duffy v Google Inc (2015) 125 SASR 437 at [158].
Publication is a multilateral act which occurs when words and or images created and disseminated by one or more persons are comprehended by another or others.[41] It is for that reason that the focus on participation in the first of the Judge’s elements is both useful and important. Publication of the printed word has always been a process rather than a single act. The process can be complex and involve many actors. As shall be seen the nature of one’s participation may critically affect the participant’s ultimate liability.
[41] Dow Jones & Co Inc v Gutnick (2002) 210 CLR 575 at 600, [26] per Gleeson CJ, McHugh, Gummow and Hayne JJ.
The participants in the publication may play primary or secondary roles. Authors and editors are primary participants. The principal of an enterprise which creates or procures and then disseminates words or images is a primary participant in the process of publication and is commonly referred to as the publisher. Historically printers have also been regarded as primary publishers but that rule has been doubted in modern times. Both publishers and printers are generally liable as primary participants even though they print and publish the work through their servants and agents.[42]
[42] Emmens v Pottle (1885) 16 QBD 354 at 357 – 358 per Lord Escher MR.
In question in Webb v Bloch[43] was the responsibility of the whole committee of an association for the publication of a defamatory circular published at the instigation of one of its members. The other members of the committee confirmed the member’s instructions to a solicitor to publish what was ultimately found to be the defamatory libel. Isaacs J in that case discussed the meaning of publication by reference to both the civil law concepts of agency and the criminal law concepts of accessorial liability:[44]
The meaning of ‘publication’ is well described in Folkard on Slander and Libel, 5^th ed. (1891), at p. 439, in these words: ‘The term published is the proper and technical term to be used in the case of libel, without reference to the precise degree in which the defendant has been instrumental to such publication; since, if he has intentionally lent his assistance to its existence for the purpose of being published, his instrumentality is evidence to show a publication by him.’ In Starkie on the Law of Slander and Libel, 1^st ed. (1830), vol. II., at p. 29, it is said: ‘The declaration generally avers, that the defendant published and caused to be published; but the latter words seem to be perfectly unnecessary either in a civil or criminal proceeding; in civil proceedings, the principal is to all purposes identified with the agent employed by him to do any specific act’ &c. In Parkes v Prescott Giffard Q.C. quotes from the second edition of Starkie: ‘All who are in any degree accessory to the publication of a libel, and by any means whatever conduce to the publication, are to be considered as principals in the act of publication: thus if one suggest illegal matter in order that another may write or print it, and that a third may publish it, all are equally amenable for the act of publication when it has been so effected.’ In R v Paine it is held: ‘If one repeat and another write a libel, and a third approve what is wrote, they are all makers of it; for all persons who concur, and show their assent or approbation to do an unlawful act, are guilty: so that murdering a man’s reputation by a scandalous libel may be compared to murdering his person; for if several are assisting and encouraging a man in the act, though the stroke was given by one, yet all are guilty of homicide.’ A little later in R v Drake, that law was reaffirmed. In The Queen v Cooper Lord Denman CJ said: ‘If a man request another generally to write a libel, he must be answerable for any libel written in pursuance of his request: he contributes to a misdemeanour and is therefore responsible as a principal.’ … In relation to Webb, it is Norman who was the ‘real author,’ the master mind, and the defendants, for their own independent objects, no doubt, were the real ‘intermediate agents’ to disseminate the libel. They cannot employ the master mind for the very purpose, accept its suggestions, approve and disseminate its production, and then disclaim its malice. S Pearson & Son Ltd v Dublin Corporation is rightly considered by text-writers (as Fraser on Libel and Slander, 6^th ed., at p. 269; Gatley on Libel and Slander, at p. 409; Spencer Bower on Actionable Defamation, 2^nd ed., at p. 265) an authority for the position that principal and agent inter se are principals in relation to the person defamed. It answers the contention of the respondent that, however this might be the case had the defendants not reserved to themselves the final right of approval, the reservation and exercise of that right made a difference. In the case cited Lord Loreburn LC said: ‘The principal and the agent are one, and it does not signify which of them made the incriminated statement or which of them possessed the guilty knowledge.’
(emphasis in original; citations omitted)
[43] (1928) 41 CLR 331.
[44] Webb v Bloch (1928) 41 CLR 331 at 363 – 365.

The statutory defence is contained in s 28 of the Defamation Act 2005 (SA). It provides:
28—Defence of qualified privilege for provision of certain information
(1)There is a defence of qualified privilege for the publication of defamatory matter to a person (the “recipient”) if the defendant proves that—
(a)     the recipient has an interest or apparent interest in having information on some subject; and
(b) the matter is published to the recipient in the course of giving to the recipient information on that subject; and
(c)     the conduct of the defendant in publishing that matter is reasonable in the circumstances.
(2)For the purposes of subsection (1), a recipient has an apparent interest in having information on some subject if, and only if, at the time of the publication in question, the defendant believes on reasonable grounds that the recipient has that interest.
(3) In determining for the purposes of subsection (1) whether the conduct of the defendant in publishing matter about a person is reasonable in the circumstances, a court may take into account—
(a)     the extent to which the matter published is of public interest; and
(b)     the extent to which the matter published relates to the performance of the public functions or activities of the person; and
(c)     the seriousness of any defamatory imputation carried by the matter published; and
(d)     the extent to which the matter published distinguishes between suspicions, allegations and proven facts; and
(e)     whether it was in the public interest in the circumstances for the matter published to be published expeditiously; and
(f)     the nature of the business environment in which the defendant operates; and
(g)     the sources of the information in the matter published and the integrity of those sources; and
(h)     whether the matter published contained the substance of the person’s side of the story and, if not, whether a reasonable attempt was made by the defendant to obtain and publish a response from the person; and
(i)    any other steps taken to verify the information in the matter published; and
(j)      any other circumstances that the court considers relevant.
(4)For the avoidance of doubt, a defence of qualified privilege under subsection (1) is defeated if the plaintiff proves that the publication of the defamatory matter was actuated by malice.
(5) However, a defence of qualified privilege under subsection (1) is not defeated merely because the defamatory matter was published for reward.
It may be observed; first, from the chapeau, that the burden of establishing the defence lies on the publisher. Second, the requirements of s 28(1) are cumulative. Third, the publisher must prove that the publishee possesses the requisite interest or apparent interest in a subject. Fourth, the distinction between an interest and an apparent interest is important. The former requires proof of an actual interest possessed by the publishee in the subject, whilst the latter requires proof of the publisher’s belief that the publishee has an interest in the subject and that such belief was based on reasonable grounds. The latter cannot be defeated by proof that the publishee did not in fact possess such interest provided that the publisher’s belief to the contrary was based on reasonable grounds. Importantly, to establish possession on the part of the publisher of a belief of an apparent interest based on reasonable grounds will require that the publisher, or someone who may be said to be part of the controlling mind of the publisher, give evidence of a belief actually held and the grounds on which such belief was held. Whether or not those grounds are reasonable will be for the trier of fact to determine in all the circumstances. It should also be made clear that the apparent interest is in information on the subject, not the defamatory matter – they may be the same thing, they may not. Fifth, the defamatory matter is published by the publisher as a consequence of giving the publishee information on the subject in which he or she has an interest or apparent interest. The defamatory matter may then comprise the information or be a component of the information. Sixth, if the publication is shown to be actuated by malice the fact that the defamatory matter was published in the course of giving information to the publishee on a subject in which he or she had an interest or apparent interest will not defeat a claim. This is because the publisher’s motive is to harm the claimant’s reputation which is not in keeping with the rationale underpinning the defence, namely, the publication and discussion of matters of public interest and importance. Seventh, to publish the information in return for payment does not defeat the statutory defence of qualified privilege. To receive payment in return for the publication of the information is also not to act contrary to the rationale underpinning the defence. Eighth, the act of publishing the information to the publishee must be reasonable in the circumstances. This is an objective assessment made upon determining the circumstances in which the publication was made. The factors referred to in s 28(3), and any other relevant circumstance, must be grounded in the evidence. Ninth, it has been said that s 28(1)(c) & (3) reflect the criteria articulated by Lord Nicholls of Birkenhead in Reynolds v Times Newspapers Ltd & ors.[369] The weighing exercise contemplated is intended to confine interference with the freedom of expression and discussion on matters of public interest to only what is necessary in the circumstances.[370]
[369] [2001] 2 AC 127 at 205.
[370] Reynolds v Times Newspapers Ltd & Ors [2001] 2 AC 127 at 204-205 (Lord Nicholls of Birkenhead); Defamation Act 2005 (SA), s 3.
I return to s 28(1). It mirrors s 22(1) of the Defamation Act 1974 (NSW).[371] In New South Wales it has long been held that s 22(1) was intended to widen the scope of the common law defence of qualified privilege.[372] Thus, an interest for the purposes of s 28(1)(a) is to be understood in its “broadest popular sense”.[373] The difficulties that beset newspapers as discussed in Morosi v Mirror Newspapers Ltd were largely overcome by the introduction of the statutory defence and the expansion of the interest accounted for.[374] Importantly the common law requirement that there be community of interest between publisher and publishee is not required by the statutory defence. In Stone v Moore Doyle J, with whom Kourakis CJ and Stanley J agreed, said:[375]
The intention of the statutory defence of qualified privilege is to broaden the nature of the interest which will found a defence relative to the common law. The focus is on the interest of the recipient rather than a reciprocity or community of interests. Again, the concept of interest is not to be narrowly or technically construed. It is used in the broader, popular sense of a matter of substance in which the relevant audience might legitimately have an interest in knowing, as long as that interest goes beyond being a matter of curiosity or prurient interest. The interest or apparent interest may be direct or indirect, but it must nevertheless be definite or tangible, and not vague or insubstantial.
(footnotes omitted.)
[371] Parliament of South Australia, Hansard, Legislative Council, 14 September 2005 at 2541.
[372] Austin v Mirror Newspapers Ltd [1986] 1 AC 299 at 311-312; Morosi v Mirror Newspapers Ltd (1977) 2 NSWLR 749 at 797 (The Court).
[373] Barbaro v Amalgamated Television Services Pty Ltd [1985] 1 NSWLR 30 at 40 (Hunt J).
[374] (1977) 2 NSWLR 749 at 797.
[375] (2016) 125 SASR 81 at [114].
Despite the breadth to be afforded the type of interest that will fall within the ambit of s 28(1), interests that are prurient or amount to no more than curiosity have been held to be excluded.[376] It could be said that it is unnecessary to confine the notion of what is an interest for the purposes of s 28(1)(a) in view of the weighing exercise required by s 28(1)(c). However, Stone v Moore and the New South Wales authorities that precede it on this question are binding. Neither party submitted to the contrary nor did either contend that the authorities that I have referred to were plainly wrong. So construing s 28(1) is not inconsistent with the stated objects of the Defamation Act 2005 and the balance to be struck between protecting reputations on the one hand, and ensuring that the law of defamation does not place unreasonable limits on the publication and discussion of matters of public interest and importance on the other.[377] I do not, with respect, consider that to limit the type of interest as indicated in Stone v Moore is to construe s 28(1) in a manner reflecting the common law defence. The expansion effected by the statutory defence is made plain in Austin v Mirror Newspapers Ltd and the rejection of the narrow construction proffered in that case by the plaintiff.[378] Thus to construe s 22(1) in the manner that Hunt J did in Barbaro v Amalgamated Television Services Pty Ltd, which underpins Doyle J’s conclusion in Stone v Moore, does not necessarily undermine the expansion intended by the statutory defence.
[376] See also Barbaro v Amalgamated Television Services Pty Ltd (1985) 1 NSWLR 30 at 40 (Hunt J); Austin v Mirror Newspapers Ltd [1986] 1 AC 299 at 311-312.
[377] Defamation Act 2005 (SA) s 3.
[378] Austin v Mirror Newspapers Ltd [1986] 1 AC 299 at 311.
The appellant did not call anyone to give evidence to the effect that, post-notification by the respondent, the appellant possessed a belief based on reasonable grounds that Google search engine users searching “Dr Janice Duffy” or “Janice Duffy” had an interest in having the information provided in the relevant paragraphs and underlying web pages that was not prurient and amounted to more than curiosity. Thus the application of the statutory defence in this case turns on the actual interest possessed by Google search engine users searching “Dr Janice Duffy” or “Janice Duffy” between January 2010 and late 2010.
I agree with the Judge that the appellant published the content of the paragraphs and underlying web pages to anyone who chose to use the Google search engine and the relevant search terms.[379] With the exception of Ms Palumbo, there was no direct evidence of the interest that those people who used the Google search engine to search “Dr Janice Duffy” and/or “Janice Duffy” had in making such searches during the relevant period (January 2010 – December 2010). The only evidence as to the searchers’ interests is that which could be inferred from the fact that they deliberately entered those search terms.
[379] Duffy v Google Inc (2015) 125 SASR 437 at [400].
In determining what may be inferred two things must be borne firmly in mind. First, as Sir Frederick Jordan said in Carr v Baker:[380]
The existence of a fact may be inferred from other facts when those facts make it reasonably probable that it exists; if they go no further than to show that it is possible that it may exist, then its existence does not go beyond mere conjecture. Conjecture may range from the barely possible to the quite possible. Inferences of probability may range from a faint probability - a mere scintilla of probability such as would not warrant a finding in a civil action: Hiddle v. National Fire & Marine Insurance Co. of N.Z. - to such practical certainty as would justify a conviction in a criminal prosecution.
…
It is well established that if there is no piece of evidence which, taken at its highest, is more than equally consistent with the existence and with the non-existence of a fact, it cannot be treated as established: Cofield v. Waterloo Case Co. Ltd.
(citations omitted.)
[380] Carr v Baker (1936) 36 SR (NSW) 301 at 306-307.
Second, whether or not an inference may be drawn from objective facts will depend in no small part upon a judge’s assessment of the common course of human affairs.
To the fact of the search terms used may be added the time period in which they were used and in relation to which the appellant was found to be liable – January 2010-December 2010. To this may be added that the evidence established that the last post on the Ripoff webpage was dated January 2009 and that as at that date the controversy between the respondent and the various psychics had ended, at least to the extent that no further postings or similar posts on the web were made until the respondent began her blog in October 2011. Further, this matter and the underlying controversy did not attract media attention until November 2011. Thus, the at least 100 monthly searches for “Dr Janice Duffy” and at least 200 monthly searches for “Janice Duffy” found by the trial Judge to have been made between October 2009 and February 2011[381] were made during a period where it cannot be inferred that they were prompted by any media attention. Further, the publications for which Google was held liable occurred during a period commencing twelve months after the last post was made on the Web in the controversy with the psychics. In those circumstances, it is unlikely that the relevant searches were prompted by a desire to follow that controversy. It had ended, and ended sometime ago. Absent the media attention that these proceedings subsequently attracted, or commencement of the respondent’s blog and any attention it attracted, what would have triggered the at least 100 monthly searches for “Dr Janice Duffy” and the at least 200 monthly searches for “Janice Duffy” made in the period January 2010 to December 2010?
[381] Duffy v Google Inc (2015) 125 SASR 437 at [310].
Clearly it may be inferred that the search engine user who used the search terms “Dr Janice Duffy” and/or “Janice Duffy” had an interest in information about the respondent. That is, the subject for the purposes of s 28(1) is “Dr Janice Duffy” and “Janice Duffy”. The ranking algorithm strengthens such inference.
The search terms used do not allow one to characterise the searcher’s interest as anything less than an interest in the respondent. That said, it is highly improbable that the search engine users did not have a narrower interest in the respondent – i.e. an interest in the respondent for a particular purpose or reason. With respect to the 100 monthly searches for “Dr Janice Duffy” the searcher used the respondent’s title, suggesting a greater knowledge of the subject than the user of the “Janice Duffy” search term, and a narrower interest.
Under s 28(1)(b) the information provided on the subject must be germane to the interest. It follows that in the circumstances of this case the appellant must establish that the Google search engine users’ interest in information on the respondent included an interest in obtaining information on her involvement with the relevant psychics. That is because it was information of her involvement with the relevant psychics that was conveyed by the paragraphs and underlying web pages and which included the defamatory material. In my view the evidence does not allow one to so conclude.
I accept that it is possible that some of those who searched “Dr Janice Duffy” and “Janice Duffy” might have fallen within a class of persons who used internet psychics and/or who had an interest in internet psychics and their integrity. However, there is no evidence that allows one to determine how likely this possibility is. On the assumption that the respondent was unknown to these people (there being no evidence to suggest otherwise), why would they search her name in the period January 2010 – December 2010? To follow the controversy they must have been alerted to it. As I have mentioned the controversy entered an hiatus in January 2009 with the last posting on the sixth Rip-off Report. By the time of the relevant publications the controversy had long been in abeyance. To hypothesise that this class of Google search engine users obtained the respondent’s name in some way that then led to them using the specific search terms during this period without any obvious prompt or motivation is speculative. In my view, the evidence does not support a conclusion on the balance of probabilities that a number of those to whom the paragraphs and underlying web pages were published in response to the search terms “Dr Janice Duffy” and “Janice Duffy” as found by the trial Judge fell within a class of persons who used internet psychics and/or had an interest in internet psychics and their integrity and thus had a relevant interest for the purposes of s 28(1)(a).
I do not think that so concluding undermines the conclusion arrived at by the trial Judge as to the number of searches undertaken. That conclusion was arrived at via a reasoning path that was not dependent in any way upon the identity of the searcher. Here the inability to draw the suggested inference is merely a product of the evidence.
As a matter of probability it is highly likely that a number of the search engine users who used the relevant search terms were either in, or considering entering into, a professional, commercial, employment or personal association with the respondent. It is possible that a number of these people had an interest in knowing about the respondent and her relationship with psychics, but not all, and what proportion cannot be discerned. I do not accept that an interest in a person arising from a contemplated or continuing professional, commercial, employment or personal association necessarily carries with it an interest in all information that may be obtained about such person. Just because they contemplate or share a relationship does not mean that the information is germane to the fabric of the contemplated or shared relationship.
Peek J has referred to the appellant’s employment and academic history. It is entirely possible that a user of the Google search engine who entered the search terms “Dr Janice Duffy” or “Janice Duffy” during the relevant period was interested in work that she had done or been involved in and upon which her involvement with psychics had no bearing. Ms Palumbo falls into this class, at least in relation to the search she conducted in and around June 2010. It cannot be inferred that members of this class had an interest in the paragraphs published during the relevant period to those Google search engine users who used the relevant search terms. If, for example, a person used the Google search engine to search for information on smoking to which the respondent contributed, they might be concerned to understand the authoritative nature of the publication, the robustness of the authors’ research, and the qualifications and experience of the authors. I cannot see how such person could have a relevant interest in the fact that the respondent consulted psychics and subsequently became embroiled in a controversy with those psychics during which she behaved, at times, appallingly.
In my view the evidence does not establish the requisite interest possessed by those who used the Google search engine and searched “Dr Janice Duffy” or “Janice Duffy” during the relevant period.
Turning to the interest possessed by those who clicked on the hyperlink and were transported to the underlying web pages, in light of my conclusion that the evidence does not establish the requisite interest possessed by those Google search engine users who used the relevant search terms, curiosity as the then motivating reason for proceeding to the web pages cannot be eliminated. Again Ms Palumbo is an example. As the majority said in Stone v Moore, the interest must be beyond curiosity. As to what other interest may have existed, there is no evidence.
In my view, the Judge was right to reject the statutory defence of qualified privilege on the basis that the appellant had failed to prove the relevant interest for the purposes of s 28(1).
If I am wrong, I would, in any event, agree with the Judge for the reasons he gives that the appellant’s conduct after notification cannot be characterised as reasonable. I also agree generally with Peek J’s reasons on this issue.

The appellant contends that making a search engine operator liable for defamatory material published in search results and linked web pages will result in over-blocking. That is, not being the author of the web page and having no means of evaluating the truth or otherwise of any defamatory material, the search engine operator will block the web page URL rather than run the risk of a defamation action. That places the search engine operator in the position of censor which is antithetical to freedom of expression and should be avoided. To accede to the appellant’s submission would, however, result in the conferral of immunity upon search engine operators from suit for defamation.
I readily accept the great benefits that the internet and search engine operators bring to society and the world generally. But the positives carry with them negatives. The damage to reputation that may occur consequent upon a defamatory posting that is available to all internet users viewed many times over and disseminated uncontrollably may be catastrophic.
The argument must be kept in proportion. If there are over sixty trillion web pages in existence and greater than 100 billion searches each month conducted on the Google search engine, I very much doubt that liability of a search engine operator for the publication of defamatory material post-notification and after the passing of a reasonable time to take action will have a chilling effect on the freedom of expression. If I am wrong, it is in my view for the legislature to intervene.
Bearing in mind the period of time that lapsed post-notification and prior to action being taken, and the appellant’s conduct in that post-notification period, I do not think it can be said that it was reasonable to publish the relevant paragraphs and underlying web pages. The publications do not concern matters of public interest generally. To say that the appellant’s only option was to remove the offending URL seems extreme, particularly in the light of the experience in the European Union and the availability of search query blocking. Once notified of the search chain the appellant was on notice that the paragraphs and underlying web pages were being published to persons with no relevant interest in the subject matter. I appreciate that blocking the particular URL does not remove the webpage from the Web with the consequence that it may resurface via a different search route or after an alteration to the URL. But publication by way of a different search route is a fresh act of publication in relation to which the innocent dissemination defence may apply afresh.
Arguably proving publication in the way that the respondent did could result in unfairness to a defendant contemplating running the defence of qualified privilege at common law or under the Defamation Act 2005. In this case the fact of publication was proven and no complaint was made at trial of any unfairness. In some ways the postulated unfairness arises from the nature of the appellant’s undertaking. I would not be prepared to hold without hearing full argument that the antidote is to require proof by the plaintiff of the actual identity of those to whom defamatory material is published. In my view the innuendo cases provide no suitable analogy. To require proof of actual identity may well result in unfairness to people defamed by others using the internet; how is the identity of the author of the defamatory material posted on a web page to be proved?
I would dismiss the appeal.

Date	Match type	Janice Duffy	Dr Janice Duffy	Janice Duffy psychic stalker
16.2.2011	Action instituted
1.9.2011	Broad (Phrase)	480 (480)	260 (210)
9.10.2011	Dr Duffy creates a blog referring to the internet defamation of her among other things
21.11.2011	Article in The Australian referring to Dr Duffy’s action
2.3.2012	Broad	880	390	91
2.4.2012	Articles in The Age and Sydney Morning Herald about Dr Duffy’s action against Google.
April 2012	Mr Trkulja uses Google to search for Dr Duffy and Janice Duffy.
29.8.2012	Broad (Phrase)	590 (590)	260 (210)
26.10.2012	Broad	590	260	46
13.6.2013	Broad	320	140	110