“The only relevant rule is that in order to be actionable the defamatory words must be understood to be published of and concerning the plaintiff.”
— Lord Atkin in Knupffer v London Express Newspapers Ltd [1944] 1 All ER 495 (HL) at 497.
The essentials
What is the test?
If the recipient—sometimes described as the ‘publishee’—would reasonably understand a statement to be about a particular person, that person will have been identified. It is an objective test. This means the Court will assess the issue from the perspective of an ordinary reasonable person.
What if no names are used?
A plaintiff can often establish identification even if they aren’t named explicitly. They only need to show that the statement reasonably identified them to readers (or even some readers). So, a plaintiff might be identified by reference to their initials, nickname or a satirical reference. In some cases, a reference to a person’s employment position may be enough. In others, simply the surrounding facts and circumstances might lead readers to identify the plaintiff.
What if the publisher doesn’t intend to refer to the plaintiff?
This is strictly irrelevant. The publisher doesn’t even need to have the plaintiff in mind, for they to be identified. This comes back to the fact that it is an objective test. There have been many cases involving completely accidental identification.
That said, a defendant’s lack of intention to identify the plaintiff might be relevant to other issues, such as damages and certain defences.
What if only a few people can identify the plaintiff?
Sometimes, a plaintiff will sue on the basis that, although a particular statement did not identify them to readers generally, they were identified by a certain audience who, unlike other readers, were in possession of certain facts that allowed them to identify the plaintiff.
For example, a newspaper article might set out certain misconduct committed by an unnamed person. The general reader might have no idea who this person is. As a result, the plaintiff will not have been identified to the general reader. But on the other hand, the article might give sufficient details that, perhaps, co-workers of the plaintiff might reasonably infer that it is the plaintiff to whom the newspaper was referring. The plaintiff will have a legitimate claim in respect of his or her identification by these people, but at trial they will need to establish the facts that allowed them to be identified in this way.
What about statements directed at groups of people?
These statements can be defamatory, but only if the words published would themselves reasonably lead people acquainted with the plaintiff to the conclusion they were one of the people referred to. So, if one wrote: “All police are corrupt”, police officers at large would not be in a position to say they were identified. On the other hand, if one wrote: “All police in Louisville are corrupt”, then, assuming Louisville is a small enough place that the statement could only be referring to a handful of police officers, each of those individuals would likely be able to establish identification.
As a rough guide, it will be difficult for someone in a group of over 20 people to bring a claim – that is, it is unlikely readers could associate a criticism of such group with a particular individual.
That said, readers might understand a statement criticising a particular group to identify those in leadership positions. And further, sometimes those in a group of more than 20 people might be identifiable. For example, “Those brainwashed parishioners of Sally’s congregation are out in force tonight, spewing her dishonest propaganda.”