Material is often published online to promote causes, further campaigns, air grievances and raise concerns about the conduct of others. When this is done responsibly it performs an invaluable public function. The availability of free-to-use platforms to express views and to engage in public debate is among the most important benefits of the internet in general and social media in particular.

Yet the ease of access to online publication platforms, which may be used anonymously, also carries risks of abuse. The promotion of particular causes or the ventilation of grievances may include the targeting and abuse of individuals, sometimes with tragic consequences. Reflecting the prevalence of this problem, a range of new words have been invented to describe such abuse: cyberbullying, trolling (the sending of menacing or upsetting messages), doxing (making personal information available) and cyberstalking.

Much of this conduct is now covered by the criminal law. The civil law also provides potential remedies – injunctions and damages – to protect the rights of individuals under Article 8 of the European Convention on Human Rights to “psychological integrity”. The primary framework for the protection of these rights is provided by the Protection from Harassment Act 1997 (“PHA”) which prohibits a “course of conduct” which a person knows or ought to know amounts to harassment of another (s.1(1)). A “course of conduct” must involve conduct on at least two occasions, although the ongoing publication of material online may be sufficient to satisfy this requirement.  Harassment is both a crime and a civil wrong.

The nature of harassment

The PHA does not provide a comprehensive definition of “harassment”, referring only to conduct which alarms a person or causes the person distress (s.7(2)). A number of points can be derived from the civil and criminal case law on the meaning of harassment under the PHA:

Online harassment

The PHA has been applied to various types of online publication including remarks on internet forums (Cray v Hancock [2005] All ER (D) 66 (Nov)); an internet-based campaign and spoof websites (Petros v Chaudhari [2004] All ER (D) 173 (Mar)); and the dissemination of naked images of the claimant (AMP v Persons Unknown ([2011] EWHC 3454 (TCC)).

The publication on a website of the name of an individual in the knowledge that such publication will inevitably come to their attention on more than one occasion and will on each occasion cause her/him alarm or distress may constitute harassment (Law Society v Kordowski [2014] EMLR 2 at [61] and [75]).

Harassment by “journalism”

Although journalism is now very broadly defined (see for example Kordowski at [99] in which journalism was held to entail the “communication of information or ideas to the public at large”), many acts of online harassment (such as abusive messaging or posts on a private Facebook page) may not fall into even the broadest categories of journalism. This is because they may not involve disclosing information to the public at large (see by analogy: University of Oxford v Webb [2006] EWHC 2490 (QB) at [72] (Irwin J)).

Where online publications amount to journalism, a claimant will nevertheless face difficulties in succeeding with a claim for harassment. In Thomas v News Group Newspapers [2001] EWCA Civ 1233) Lord Phillips MR held that:

“before press publications are capable of constituting harassment, they must be attended by some exceptional circumstance which justifies sanctions and the restriction on the freedom of expression that they involve. It is also common ground that such circumstances will be rare” [35].

This observation was built upon by Tugendhat J in Trimingham v Associated Newspapers [2012] EWHC 1296 (QB):

“It would be a serious interference with freedom of expression if those wishing to express their own views could be silenced by, or threatened with, claims for harassment based on subjective claims by individuals that they feel offended or insulted” [267].

Nevertheless, in the recent case of  it was held that the contents of a series of articles concerning the claimant were capable of constituting actionable harassment in that some of them were defamatory and they contained various inaccuracies.  The court declined to grant summary judgment to the defendant and directed that the matter should go to trial.


Given the growing problem of online abuse, a claim for harassment is a potentially valuable tool for curtailing abuse. Significantly, claims can be brought in circumstances in which libel or privacy claims would be difficult or impossible. Online publications amounting to a course of conduct may be actionable regardless of whether the information in a publication is private, true, a statement of fact or opinion, or already in the public domain.

It nevertheless remains difficult for a claimant to establish that online publications constitute harassment. The threshold for harassment is a high one. A defendant’s point of departure will often be that the impugned conduct was not, when objectively judged, oppressive and unacceptable in the sense that it would justify criminal liability. Further, the utility of a claim in harassment may be limited by the fact that few harassers harass in their own in their name; it is often very difficult to identify the individual(s) behind online harassment (as to which see chapter 3 of Online Publications Claims: A Practical Guide).

The PHA also provides important protections for would-be defendants through its three defences (s.1(3)). Most relevantly, a publisher can defend her/himself on the basis that the conduct complained of was, subjectively assessed (and subject only to a rationality test), pursued for the purpose of preventing or detecting crime; or the pursuit of the course of conduct was reasonable in the circumstances.

Aidan Wills is a member of Matrix Chambers and an author of Online Publication Claims: a Practical Guide. 

This is an edited extract from Chapter 4 of the book which will be published on 9 November 2017.