My Web Stats


“In the law of defamation, ‘publication’ is a term of art … A defamatory writing is not published if it is read by no one but the one defamed.  Published, it is, however, as soon as read by anyone else.”
— Cardozo CJ in Ostrowe v Lee (1931) 175 NE 505 at 505.

The essentials

What is the test?
Normally when we think of something being published, it is in the issuing of a book or publication of a news story.  The test is much narrower under defamation law; something is ‘published’ simply by its communication to at least one third party.   And one third party is enough.  Indeed, claims have rested upon remarks overheard in conversation, intercepted postcards, and emails copied to a single person.

That said, ‘third party’ means someone other than the plaintiff.  So, if Sally called Louis a fraudster to his face, but no-one else heard it, Louis would not be able to establish publication.

What sorts of publications have given rise to defamation claims?
Almost any medium you can think of.  Most claims arise from written statements: news media articles, emails, press releases, blog posts, Facebook posts, tweets, etc.  But claims may arise from other written mediums too.  In our Case Reviews, we’ve covered two cases involving signs or placards: a classic English case about a man who was sued simply for pointing at a sign; and a 100-year-old New Zealand case between neighbouring boot merchants.   Oral statements also lead to many claims, normally in political or business speeches and seminars, but some have been brought from stand-up comedy performances and birthday speeches.  In an Australian case in our Case Reviews, the claim was brought in relation to remarks called out to people sitting in their car.

In fact, any medium that communicates a defamatory message can give rise to a defamation claim: there have been many cases involving cartoons, including a major Canadian proceeding in our Case Reviews.  Others have been based on statues with defamatory messages, and there was even one claim based on a message engraved into wet concrete.

A publication doesn’t even need to have words; gestures may be enough.  A drinking motion might be understood to mean: “Sally has a drinking problem”.  In our Case Reviews, we reviewed an Australian proceeding brought over gestures, which involved a teacher who was mocked by students during a school play.  Claims based on memes are becoming more commonplace, too.

Does a plaintiff need to prove that a statement was communicated to a third party?
It depends on the context.  If the statement is written in a newspaper or book, posted to a news website, or is broadcast on a national radio or TV station, it’s pretty safe to assume the Court will comfortably infer the statement was published to third parties (and defendants would look foolish denying it).  So in these cases, the plaintiff doesn’t need to lead evidence to establish publication.

  However, if a statement is scribbled on a bathroom wall, alleged to have been overheard at a dinner party, or posted on an obscure website—that is, situations where a Court cannot blindly infer that a third party actually read or heard the statement—then the plaintiff will need to prove it was received by at least one third party.  This evidence would usually be taken from the third parties who read or heard the statement, but evidence might also include CCTV footage or, in the case of social media comments, subsequent comments on a particular thread.

Can it be defamatory to say something if someone else has already said it?
Yes.  The law applies what is known as the ‘repetition rule’ (or ‘republication rule’).  Under this rule, the liability of a person who repeats a rumour or allegation is no different to the person who first made the statement.  This includes rumours or allegations that come from reputable sources like the Police or news sites.   This aspect of the law has some wriggle room when it comes to establishing defences.  But for at least the element of publication, anyone who repeats a defamatory statement is liable.

Are defamatory statements actionable in New Zealand if they originated overseas, such as foreign websites?
Yes.  Under New Zealand law, publication occurs where the statement is received.  For websites based offshore, publication will occur at the place the information is downloaded (i.e. to where the offending data is transmitted).  It is no defence for the overseas publisher to say that they didn’t explicitly intend for the material to be read in New Zealand.  If the content has been made available to be accessed (published) in New Zealand, then the publisher can be accountable for any defamation it contains.

  The same goes for books or newspapers that originated overseas.

However, a line can maybe be drawn around online content where the overseas publisher sets up a geo-blocker, or where a book was strictly unavailable for purchase here.  In this way, the publication could only occur by the transmission of the content by a third party, and the publisher might be able to negate the proposition that it was a reasonably foreseeable consequence the content would be published in New Zealand.

Can someone be liable for what another person has published?
Yes, sometimes.  There are different categories of liability.

  • If an employee publishes a defamatory statement in the actual (or even apparent) course of their duties, the employer can be vicariously liable.  They might not even have known the statement was to be made, but can still be liable for it.  This includes the Crown in relation to most (but not all) public servants.
  • If several people are involved in a publication, each of them can be directly liable.  For example, when a defamatory news article is published, the author, editor, sub-editor, publishing company, and any sources who were reasonably aware their comment would be published, can all be sued.  That said, most plaintiffs sue only the publishing company, and sometimes also the author or source.  The publishing company, and any direct employers of the journalist concerned, may also be vicariously liable, but plaintiffs don’t normally bother suing media across multiple bases of liability.
  • If someone runs a website and third parties post defamatory comments, the owner can be directly liable.
  • If someone incites another person to make a defamatory statement, then both can be directly liable.  The law is still working out the level of ‘incitement’ that is required.  We think it would be enough to prove that the first person knew, or ought to have known, that it was a reasonably foreseeable consequence that the defamatory statement would result from their actions.
  • If someone administrates content on another website, say a Facebook page, and then another person posts a defamatory comment, the administrator can be liable if the plaintiff can prove the administrator had actual knowledge about the statement and failed to remove it within a reasonable time.  This is sometimes called endorsement/adoption liability, but it is probably just as accurately described as contingent direct liability.  It would seem that ISPs and social media companies could be liable on the same basis.

One area that hasn’t been resolved in New Zealand is whether someone is liable for providing hyperlinks to defamatory material posted by another person.  Liability would likely depend on whether the person hyperlinking the material would be viewed as having endorsed or adopted the material on the hyperlinked page.  For example, if Sally provided a link on her website for readers to “see how Louis has totally exposed the Police Chief as an absolute crook”, readers could infer that Sally was effectively re-publishing Louis’s defamatory statement and she could sensibly be liable for it.  Structurally, Louis’s publication would be considered as part of Sally’s publication.  By contrast, if Sally provided the hyperlink for readers in neutral language or, better still, expressly disassociated herself from Louis’s statement, then it would be hard to say she had endorsed Louis’s statement as her own.  So, if the Police Chief was to allege that Sally had republished Louis’s defamatory content, then the Court have to at least view the hyperlinked content in light of what Sally had to say about the matter.

The issue is also quite challenging because it would seem strange that a person could be liable for footnoting references to defamatory material in an academic journal.  But then again, hyperlinked material is much more accessible.  On one hand, the law needs to keep in-step with the realities of how people access content.  But then it is problematic to apply a doctrine of liability inconsistently across different mediums.

Can Facebook, search engines and web-hosting services be sued for defamatory material published through their servers?
Yes, but it’s not straightforward.   The plaintiff can establish a claim if the material complained of:

  • went through the server of the social medium/search engine/web host,
  • and was accessed in New Zealand,

But there are some real hoops to jump through.

First, the law provides for the defence of ‘innocent dissemination’.  We cover this under other defences.  Suffice to say, the social medium, etc. would likely have a defence unless, and until, the plaintiff could show they were put fairly on notice that they were publishing defamatory material, and had failed to remove it.

Second—and a practical point—it is very difficult to proceed with a claim against these companies.  Most of the big names are based in California.  Most people can’t simply ‘ring Facebook’ to have content removed.  Filling out online complaint forms is unsatisfactory and progress is often terminally slow-going.  Add to all this, because US defamation law is much more publisher-friendly (because of legislation supporting the First Amendment) even if you were successful bringing a claim through the New Zealand courts, these companies could, practically speaking, simply ignore the result because the judgment is unlikely to be enforceable in the US.

That all said, in our experience these companies usually, albeit somewhat eventually, respond positively and remove defamatory content without the need for court proceedings.  But one does need to be very patient, which can be difficult with one’s reputation on the line.


Legal practice tips
For publishers:

  • Before the internet took hold, very few people had the means to be a publisher.  It’s true that anybody with the ability to utter a defamatory slur could find themselves facing legal action.  But the claims that actually made it to Court normally centred around publications by news media and book publishers.  Indeed, in those days one could even expect that the inflammatory letter-to-the-editor would be abridged to remove defamatory content (for obvious reasons from the newspaper’s perspective).  Nowadays, everyone has the means to publish defamatory content far and wide, particularly through social media.  This has opened up greater freedom to disseminate views and information, but has increased the risk of ordinary people being sued.  So, exercise your new-found freedoms with this in mind.
  • The ‘repetition rule’ is engaged constantly on social media, with shared content and retweets.  If one has a big following on social media, then the risks of sharing or retweeting inflammatory content should be obvious.
  • If you have posted something regrettable which is being shared and you are minded to minimise your exposure, you’re probably better served not only to delete the original posting but also expressly disassociate yourself from it, and take steps to message whoever has shared it to ask them to also remove it.  This might go at least some way towards mitigating any damages you might have to pay, even if others continue to share it.
  • Allowing third-party comments on your website always carries some risk, especially for publishers who don’t moderate their comments.  The safer strategy is to direct user comments to the publisher’s social media pages, because under current New Zealand law the plaintiff will have to establish that the publisher had actual knowledge of the defamatory material, and failed to remove it within a reasonable time, in order to establish liability.  This seems to be the strategy adopted by, for example, Newsroom.
    Newsroom comments policy
    Newsroom’s comments policy.
  • Ironically, you might be better placed legally not to moderate comments on social media at all because then it is up to the plaintiff to notify you of any defamatory ones, only at which time you’ll have to delete it if you wish to avoid liability.

For plaintiffs:

  • Intending plaintiffs should preserve evidence of publication – e.g. newspaper cuttings, screenshots, audio files, video files, etc.  For online content, screenshots should be taken of the offending content at the earliest opportunity and they should be time-stamped in some way.  Perhaps, just like how people used to take a photo with the day’s newspaper in the background to record the date, it would be wise to take screenshots that capture the date on one’s computer, or device (with screenshots saved as a time-stamped photo).
  • Screenshots of Facebook comments can be troublesome to screenshot because, in order to give the full context of a publication, you have to expand all the subthreads.  Laborious it might be, but it has to be done.  Make sure you do this at the earliest opportunity, but also keep monitoring the thread for any additional comments.
  • If the issue of publication may be disputed, ensure that you plead sufficient particulars of how the statement was published to third parties.
  • Even if third-party publication may be readily inferred, precise publication statistics (e.g. newspaper circulation or webpage hits) are valuable evidence for damages purposes.  Given this information will often be held only by the defendant, ensure that pleadings trigger discovery of these statistics.
  • As a matter of good practice, ensure Facebook threads, etc. are saved in high-resolution colour.  These look much more powerful in evidence than poor black-and-white printouts that have been scanned multiple times.
  • To establish liability of, say, a Facebook administrator, it is always worth seeing if they have responded themselves to the thread containing the defamatory posting.  If they have, the plaintiff will likely be able to establish they had actual knowledge of it.  This is primarily for comments posted after the defamatory statement, but you might also be able to establish actual knowledge if they commented on the thread before the defamatory statement, since there might be evidence they received a notification about it (this might depend on their account settings).
  • If one is looking to take on a social media or search engine behemoth, it might be valiant but normally one’s efforts and resources will be better directed to tracking down the original commenter and taking action against them.  Indeed, even those who post anonymously or under pseudonyms can often be unmasked with some digging and the right technology.

For defendants:

  • Be careful admitting third-party publication on claims based on social-media statements.  This is one of the most difficult things for plaintiffs to establish.  There is nothing wrong with putting the plaintiff to proof (unless it’s blindingly obvious that third parties have received the statement, in which case you risk costs exposure).  Further, keep in mind that the breadth of publication will be relevant to damages, so if the plaintiff can establish only a small number of recipients, the exposure to damages is unlikely to be significant.
  • If available, take your own screenshots and copies of offending social media threads.  Sometimes a plaintiff’s copy will omit—inadvertently, one would hope—full threads or subthreads, which can distort the context in which an impugned statement was made.
  • For those who may be liable for comments posted by third parties, consider issuing third-party notices or even see if an indemnity can be argued.
  • Removing online content is not an admission of liability.  But it is always safest to expressly point this out.
  • In so many cases, simply removing online content can assuage a plaintiff from issuing proceedings.  Plaintiffs are normally just as worried about their search results as the original publication.  If you’re getting scant value from keeping it online, removal might well be a more economical way to resolve a complaint.

Recommended reading

For liability under the ‘repetition rule’ – i.e. for republishing defamatory statements: see paragraph [24].

For liability of Facebook page administrators for third-party postings; also relevant more generally when considering liability for third-party publications: see paragraphs [81]-[144].

For principles of pleading causes of action based on a multiplicity of statements and authors, as well as the different bases of liability: see paragraphs [17], [21]-[24] and [35]-[65].  Or save yourself some time, and see our Case Review here.

One element down.  The next one’s easy.  Let’s move on to identification.