Update: Here are a few thoughts which are partly a summary and partly what I think is really interesting about this case. Be sure to read the post itself, though, there is a lot to unpack in this case.
The innocuous looking case of H v W which was handed down in the South Gauteng High Court on 30 January 2013 is anything but. Judge Willis’ 30 page judgment recognises the harm a Facebook post can do to a person’s reputation and throws the weight of the Court behind the person defamed (and who can afford the legal fees).
What the judgment says
Judge Willis’ judgment begins with some background information (the names of the parties were redacted to prevent further harm). W (the “respondent”) published an open letter to H (the “applicant”) on Facebook “for public consumption” which included the following paragraph:
I wonder too what happened to the person who I counted as a best friend for 15 years, and how this behaviour is justified. Remember I see the broken hearted faces of your girls every day. Should we blame the alcohol, the drugs, the church, or are they more reasons to not have to take responsibility for the
consequences of your own behaviour? But mostly I wonder whether, when you look in the mirror in your drunken testosterone haze, do you still see a man?
Judge Willis continues and explains that –
The applicant is an insurance broker who is separated from his wife. The respondent had been a close friend of the applicant. This friendship extends back from the time before the applicant married his wife. In terms of a Deed of Trust, the applicant and his wife had jointly appointed the respondent to be the guardian of their three minor children in the event that both the applicant and his wife died or became incapacitated before their children attained their majority. The applicant had provided the respondent with guidance in starting her current business venture. The respondent had lent the applicant money to tide him over certain financial difficulties.
The applicant and his estranged wife are engaged in a divorce action. The applicant’s estranged wife is presently residing with the respondent. The applicant’s wife left him to stay with the respondent on 14 January, 2012. The applicant pays for the children’s medical aid, extra mural classes, stationery and a full time tutor to assist them. The three minor children born of the marriage between the applicant and his estranged wife are Z, born in 1997, M, born in 1999 and C, born in 2001. These minor children have been residing with the applicant for the last few months. The two minor daughters are both ‘friends’ on Facebook with the respondent. A ‘friend on Facebook’ is a ‘term of art’ to which I shall later refer. The applicant and the respondent were friends on Facebook but, consequent upon the applicant’s wife leaving him and moving into the home of the respondent, the applicant has ‘defriended’ the respondent.
The applicant complains that the posting in question publishes information which portrays him as:
(i) A father who does not provide financially for his family;
(ii) A father who would rather go out drinking than caring for his family;
(iii) A person who has a problem with drugs and alcohol.
The applicant’s attorney, in her letter dated 28 February 2012 addressed to the respondent, referred to the possibility of a claim for damages. The respondent claims that she posted the posting not to defame the applicant but in order for the applicant to reflect on his life and on the road he had chosen.
The law the Court relied on
The lawyers involved in the matter conducted what appears to be fairly substantial research on the law on defamation online and with reference to Facebook. Judge Willis relied fairly heavily on two academic articles by –
- South African Professor of Private Law at the University of South Africa, Anneliese Roos (her article is titled “Privacy in the Facebook Era: A South African Legal Perspective” and was published in the 2012 South African Law Journal); and
- Associate Professor of Law at the New York Law School, James Grimmelmann (his article is titled “Saving Facebook” and can be found in the 2009 Iowa Law Review).
The Court quoted extensively from Roos’ article, largely as a means to understand Facebook and its mechanics. Although reliance on Roos’ article and her explanation of how Facebook works may not be integral to the judgment, it is unfortunate. Roos appears to have a fairly limited understanding of how Facebook and its privacy settings work and this gives the impression that activity on Facebook is unavoidably more public than it need be. One example of this is Roos’ explanation of Facebook’s primary privacy options:
In Facebook there are three privacy settings to choose from for information other than the user’s name, profile picture, gender and networks. Users can choose to make other information visible to “everyone”, or to “friends of friends”, or to “friends only”. It is important to recognise that a “friend” on Facebook is someone you have listed as a “contact” – such persons are not necessarily friends in real life.
Users can control their visibility to some extent by using the privacy settings to make personal information available only to friends, and of course by limiting the number of friends or contacts they add to their networks. In general, however, most users give out an extraordinary amount of information.
The last part about people sharing a lot of information is true, of course, but what Roos missed is that users can also share fairly selectively using Facebook lists which stand apart from the default “Public”, “Friends”, “Friends except acquaintances” and “Only me”. To be fair to Roos, though, her article was published in 2012 and Facebook’s privacy controls have changed since then and, in many respect, have become somewhat more secure in some respects, and have eroded protections like users’ privacy by obscurity. That said, she made a few good remarks about what I see as privacy’s contextual nature:
In the context of SNSs, one could argue that subscribing to an SNS and completing your profile information is similar to appearing in a public place. The Internet is a very public place, and Facebook clearly warns subscribers that their privacy cannot be guaranteed. However, in my opinion the privacy settings that one choose, should also be taken into account when considering whether a person has really chosen to disclose his or her information to an indeterminate number of persons. If you chose to reveal your personal information to “Friends Only” and if you limited the number of friends that you added, it could be argued in my opinion that you did NOT choose to reveal your information to an indeterminate number of persons. You have, in fact, revealed your personal information to a limited number of people. If one of your Facebook Friends then further discloses personal information that was provided by you in these circumstances, I would argue that you should have a delictual claim for infringement of privacy.
Roos made another important point to bear in mind when considering whether a person’s right to privacy has been infringed is similar to a defamation analysis (and which Judge Willis touched on in his judgment):
It should also be remembered that the wrongfulness of an infringement of privacy is negated by the presence of a ground of justification. Neethling identifies the following traditional grounds of justification as relevant to the right to privacy: necessity, private defence, consent to injury, and performance in a statutory or official capacity. Another ground of justification which is relevant to privacy is the protection of legitimate interests, including the public interest.
In the context of SNSs, consent is a particularly relevant ground of justification. Whenever the user discloses personal information on his webpage, he or she consents to the publication of that private information. However, in order to be valid the consent must meet certain criteria.
She then went on to discuss the parameters of the consent required to justify a privacy infringement. Essentially, consent must be informed although in light of the Protection of Personal Information Bill, we can adopt the Protection of Personal Information Bill’s defintion of “consent” as –
any voluntary, specific and informed expression of will in terms of which permission is given for the processing of personal information
Unfortunately Roos’ understanding of Facebook’s privacy controls and how it works, generally, is limited as expressed by her article and Judge Willis has accepted her explanation as a complete and accurate exposition of privacy on Facebook. This will likely, and unfortunately, colour future Court judgments regarding Facebook privacy issues. One inaccurate statement is Judge Willis’ understanding that –
Accordingly, although one can control one’s own Facebook profile but there is no method, within the Facebook system itself, by which one can control what other people place on their profiles about
oneself and who can look at that.
This is not really correct. Users have the ability to exercise a fair amount of control over whether other users can post their personal information and tag them (Roos seems to have largely misconstrued how tags can be restricted and controlled by the person being tagged).
After exploring Twitter briefly, Judge Willis turned to established case law in South Africa including authority for the proposition Roos expressed that a privacy infringement can be justified in a similar way that defamation can be justified and a more recent Supreme Court of Appeal judgment in the 2004 Mthembi-Mahanyele v Mail & Guardian case which, according to Judge Willis –
affirmed the principle that the test for determining whether the words in respect of which there is a
complaint have a defamatory meaning is whether a reasonable person of ordinary intelligence might reasonably understand the words concerned to convey a meaning defamatory of the litigant concerned
The Court, in the Mthembi-Mahanyele case set out the test for defamation as follows (and cited a 1993 case in the then-Appellate Division of Argus Printing and Publishing Co Ltd v Esselen’s Estate) –
The test for determining whether words published are defamatory is to ask whether a ‘reasonable person of ordinary intelligence might reasonably understand the words . . . to convey a meaning defamatory of the plaintiff. . . . The test is an objective one. In the absence of an innuendo, the reasonable person of ordinary intelligence is taken to understand the words alleged to be defamatory in their natural and ordinary meaning. In determining this natural and ordinary meaning the Court must take account not only of what the words expressly say, but also of what they imply’
Referencing one of the justifications for (or defences to) defamation, namely that the defamatory material be true and to the public benefit or in the public interest, Judge Willis drew an important distinction that is worth bearing in mind –
A distinction must always be kept between what ‘is interesting to the public’ as opposed to ‘what it is in the public interest to make known’. The courts do not pander to prurience.
The Court moved on to explore another justification, fair comment. In order to qualify as “fair comment” –
the comment “must be based on facts expressly stated or clearly indicated and admitted or proved to be true”
The person relying on this justification must prove that the comment is, indeed, fair comment and “malice or improper motive” will defeat this justification or defence, regardless of its demonstrably factual nature. In this particular case, the Court found that W acted maliciously and she was unable to prevail with this defence.
The Court’s finding
Flowing from the Court’s finding that W’s Facebook post was defamatory and unlawful, Judge Willis considered a procedural issue. H instituted proceedings on motion, which means that H used an expedited approach relative to the more conventional trial action and sought a court order requiring W to remove the post and stop posting about H (the very simplied and incomplete version). Normally an applicant instituting motion proceedings would be asked to explain to he Court whether there is “an absence of similar protection by any other ordinary remedy”. W’s legal team pointed out (as respondents frequently do) that H could have sued her for damages (monetary losses) through a trial action.
Judge Willis considered this argument and took a pretty important step in developing the common law that governs these sorts of proceedings. He said the following:
It is in respect of the remedy where infringements of privacy take place in the social media that the common law needs to develop. The social media form a subset of the electronic media but are not coextensive with it: the social media are all part of the electronic media but not all the electronic media are social media. The electronic media were, almost certainly, beyond the imagination of the court when Setlogelo v Setlogelo was decided in 1914. Not only can items be posted and travel on the electronic media at a click on a computer in a moment, in an instant, at the twinkling of an eye, but also they can, with similar facility, be removed therefrom. This can also be done at minimal cost. The situation is qualitatively different from the scenario where newspapers have been or are about printed in hardcopy and distributed. The law has to take into account changing realities not only technologically but also socially or else it will lose credibility in the eyes of the people. Without credibility, law loses legitimacy. If law loses legitimacy, it loses acceptance. If it loses acceptance, it loses obedience. It is imperative that the courts respond appropriately to changing times, acting cautiously and with wisdom.
He went on to explain that –
[t]he historical reluctance of the courts to interdict publication in the media has its roots only in the issues relating to technology and economics that arise from ‘stopping the press’ but also a concern about the social consequences of stopping the free flow of news and information. This concern about the ‘chilling effect’ of court orders on freedom of expression has been manifested in the case of National Media Limited v Bogoshi recently decided in the SCA.
This is an interesting perspective. Just as social media accelerates the rate at which people can share (and remove) their thoughts, it also addresses the courts’ concerns about ordering that defamatory materials be removed from these social services because the costs involved are minimal. Ironically, though, litigation remains a pretty costly exercise so this must also factor into litigants’ cost analysis. For the courts, though, it would seem that the social Web removes an obstacle to judicial activism when it comes to defamation. Of course it isn’t just about the economics involved, courts will continue to assess the relative value of one party’s right to freedom of expression when weighed against the other other’s right to dignity, for example. As Judge Willis points out –
resolving the tensions between every human being’s constitutionally enshrined rights both to freedom of expression and to dignitas is all about balance. In the case of Le Roux v Dey Freedom of Expression Institute and Another as amici curiae) the Constitutional Court emphasized the need to take into account the context in which a publication occurs.
W’s advocate attempted to argue that H could take the matter up with Facebook itself as a violation of its terms and conditions but Judge Willis wasn’t convinced. He didn’t feel there was anything before him to “assure me that Facebook would comply with such a request”. He further noted James Grimmelmann’s argument that “it is better for the courts to focus on users rather than Facebook itself if intrusions on privacy are effectively to be curbed”. Essentially, “if one wants to stop wrongdoing, it is best to act against the wrongdoers themselves”.
Media reports about this case referenced the orders H sought against W which included a prohibition on W “posting any information pertaining to the applicant on Facebook or any other social media”. Judge Willis refused to grant this order as he said he has “no way of knowing for certain that there will be no cicumstances in the future that may justify publication about the applicant”. I think he made the correct decision there and this means that these complaints will have to be dealt with on a case by case basis with careful consideration of each case’s merits.
The Court also refused to grant orders placing W under arrest if she fails to comply with the order and that the Sheriff of the court be ordered to remove the offending post. Instead W was ordered to remove the posts and pay H’s costs. As a practical matter, Judge Willis made the following point:
Those who make postings about others on the social media would be well advised to remove such postings immediately upon the request of an offended party. It will seldom be worth contesting one’s obligation to do so. After all, the social media is about building friendships around the world, rather than offending fellow human beings. Affirming bonds of affinity is what being ‘social’ is all about.
A few more interesting points
Judge Willis went on to deal with public figures and the prospect of having to deal with a flurry of relatively minor complaints after this judgment.
With respect to public figures, he pointed out that while they enjoy a right to privacy, “[t]here is legitimate public interest in the affairs of public figures” and this means that they may not enjoy the same degree of protection as citizens not in the public spotlight when it comes to defamation online. As Judge Willis put it –
Trenchant commentaries on the performances of politicians as politicians, entertainers as entertainers, musicians as musicians, artists as artists, writers as writers, poets as poets, sports stars as sports stars will generally pass legal muster, even if posted in the social media. When it comes to freedom of expression in South Africa, there are oceans in which to swim and upon which to sail as freely as the wind blows.
When it comes to the prospect of being inundated by defamation claims (my words), the Court referenced the National Media Limited v Jooste case and Judge of Appeal Harms’ comments that –
the question of whether private facts are worthy of protection is determined by reference to ‘ordinary or reasonable sensibilities and not to hypersensitivities’.
Similarly, Grimmelmann has referenced the legal maxim de minimis non curat lex which Judge Willis translated as “the law is not concerned with trivia”. Ultimately, the cost of litigating, even when it comes to motion court proceedings of this nature, will stem the flow of applications to court. Litigation remains a costly exercise and this cost may not always be warranted, regardless of how hurtful the comments may be. This is especially so in the context of motion court proceedings where courts tend not to grant damages as relief because of how these motion court proceedings work.
Lastly, the judgment didn’t deal with another complication: the infamous Streisand Effect. This phenomenon kicks in regardless of your entitlement to protect legitimate rights. It has the ability to effectively negate the practical value of a court order such as the one Judge Willis granted and should always be carefully considered and weighed up against the risks in any matter. There are going to be cases where it is better to walk away and focus on damage control and other cases where it is worth risking the phenomenon and going to court. A critical factor will be whether the legal advice you take is adequately informed about not only the law but the social Web dynamics to form a better risk assessment before you pull the trigger.
This case is an important one. It advances the law dealing with social Web issues and, if it is upheld or applied by other (and higher) courts, it will change the tone for how courts will deal with issues like online defamation and privacy.