The International Forum for Responsible Media Blog

Prosecuting social media: the DPP’s interim guidelines – Alex Bailin QC and Edward Craven

TwitterThe ubiquity of social media presents new challenges for law enforcement and free speech. The Leveson Report described the internet as an ‘ethical vacuum’ where ‘bloggers and others may, if they choose, act with impunity’. But this picture of an unregulated domain of free expression is at odds with recent prosecutions for messages posted on Twitter, Facebook and other social media websites.

Amid growing disquiet about the use of the criminal law against users of social media, the Director of Public Prosecutions, Keir Starmer QC, has published Interim guidelines on prosecuting cases involving communications sent via social media.

The guidelines set out, for the first time, the approach prosecutors should take when deciding whether to prosecute individuals for offences committed on social media. They represent an important development in the relationship between cyberspace, free expression and the law. But while they are certainly a step in the right direction, they do not remove the uncertainty and chilling effect caused by legislation that criminalises “grossly offensive” communications.

Background to the interim guidelines

Under section 1 of the Malicious Communications Act 1988 it is an offence to send an electronic communication which is indecent, grossly offensive, threatening or false with the intention of causing distress or anxiety to the recipient. Section 127 of the Communications Act 2003 similarly criminalises the sending of a message which is ‘grossly offensive’ or of an ‘indecent, obscene or menacing character’ via a ‘public electronic communications network’.

These provisions have been used in a number of social media prosecutions In October, Matthew Woods was sentenced to 12 weeks’ imprisonment for posting offensive ‘jokes’ and sexually explicit comments about the missing five year-old girl April Jones. The same month, Azhar Ahmed received 240 hours community service for posting a message on Facebook about the deaths of six British soldiers in Afghanistan which said: ‘All soldiers should die and go to hell’.

The most high profile example of a social media prosecution is Paul Chambers, author of the infamous ‘airport bomb tweet’ (‘Crap! Robin Hood airport is closed. You’ve got a week and a bit to get your shit together otherwise I’m blowing the airport sky high!’). Chambers was convicted of sending a menacing communication under s127 – a ruling that unleashed a storm of protest about how the criminal law was being used to stifle free speech. His conviction was ultimately overturned by the High Court in July.

Communications posted on social media may also give rise to a number of other offences. Messages targeting a particular individual, for example, may constitute a course of harassment under the Protection from Harassment Act 1997. A threat of violence may amount to assault (which can occur by words alone) or a public order offence.  Naming the victim/complainant of a sexual crime is illegal under the Sexual Offences (Amendment) Act 1992, and the deliberate publication of information in breach of a court order is a contempt of court.

The DPP’s decision to launch a consultation was prompted by the arrest of a man for posting homophobic comments on Twitter about the Olympic divers Tom Daley and Peter Waterfield. Although the DPP took no action in that case, the arrest drew significant criticism and highlighted concerns about the heavy-handed enforcement of the criminal law.

The DPP’s interim guidelines

The DPP’s interim guidelines do not replace the Code for Crown Prosecutors but supplement it. Any prosecution must therefore pass the test of evidential sufficiency (i.e. a realistic prospect of conviction) and the CPS must also consider that a prosecution is in the public interest.

In social media cases prosecutors must now make an initial assessment of the content of the communication and any course of conduct that it forms part of. The guidelines identify three categories of case that will be ‘prosecuted robustly’:

(1)  Credible threats of violence or damage to property;

(2)   Messages which specifically target an individual or group of individuals and which may constitute harassment or stalking;

(3)  Communications which breach a court order

By contrast, communications which are ‘grossly offensive, indecent, obscene or false’, but which do not fall within any of those categories will be subject to a ‘high threshold’ (see below).

The guidelines draw a sharp distinction between threats that are credible and threats that are not:

As a general rule, threats which are not credible should not be prosecuted unless they form part of a campaign of harassment specifically targeting an individual within the meaning of the Protection from Harassment Act 1997.

Communications that amount to a course of conduct within the meaning of the Protection from Harassment Act will be considered under the CPS Legal Guidance on Stalking and Harassment. Likewise, cases that potentially breach a court order must be considered under the CPS Legal Guidance on Contempt of Court and Reporting Restrictions.

Communications which are grossly offensive, indecent, obscene or false

Given the scale and ubiquity of social media, s1 of the MCA 1988 and s127 of the CA 2003 create the potential for a very large number of prosecutions. The guidelines acknowledge this possibility and prosecutors are therefore instructed to exercise ‘considerable caution’ before bringing charges under either provision. There is ‘a high threshold that must be met before any proceedings are brought and in many cases a prosecution is unlikely to be required in the public interest.

The guidelines emphasise the unique context of social media communications:

‘prosecutors should have regard to the fact that the context in which interactive social media dialogue takes place is quite different to the context in which other communications take place. Access is ubiquitous and instantaneous. Communications intended for a few may reach millions.’

Drawing on comments made by the Lord Chief Justice in the Chambers case, the guidelines indicate that a prosecution should only be brought under s1 or s127 where the communication is ‘more than’:

  • Offensive, shocking or disturbing; or
  • Satirical, iconoclastic or rude comment; or
  • The expression of unpopular or unfashionable opinion about serious or trivial matters, or banter or humour, even if distasteful to some or painful to those subjected to it.

The public interest test

The guidelines list various factors that must be considered in deciding whether a prosecution is in the public interest. A prosecution is unlikely to be necessary and proportionate where:

(1)   The individual has taken swift action to remove the communication or expressed genuine remorse;

(2)   swift and effective action has been taken by others, for example, service providers, to remove the communication or block access to it;

(3)   communication was not intended or obviously likely to reach a wide audience, particularly where the intended audience did not include the victim or target of the communication;

(4)  content of the communication did not obviously go beyond what could conceivably be tolerable or acceptable in an open and diverse society which upholds and respects freedom of expression.

On the other hand, where a particular victim is targeted and there is clear evidence of an intention to cause distress or anxiety, prosecutors should ‘carefully weigh’ the effect on the victim. A prosecution under s1 of the MCA 1988 may be in the public interest, particularly if the offence is repeated.

The age and maturity of suspects should also be given ‘significant weight, particularly if they are under the age of 18’. Children and young persons may not fully appreciate the potential harm and seriousness of their communications. Prosecuting a minor is therefore ‘rarely likely to be in the public interest’.

Public order offences

The guidelines caution that ‘particular care’ should be taken when using public order offences to prosecute social media cases, since public order legislation is primarily concerned with words or actions carried out in the presence/hearing of the accused. Prosecutors are reminded of the decision in Redmond-Bate v DPP, where Sedley LJ said the mere fact that words are irritating, contentious, unwelcome and provocative is insufficient to justify invoking the Public Order Act, unless they also tend to provoke violence.


The DPP’s guidelines are a welcome attempt to bring coherence and predictability to the prosecution of offences involving social media. But while the content of the guidance is largely uncontroversial, it is unlikely to provide the clarity and foreseeability needed in this area.

The decision to prioritise credible threats of violence and harassment is sensible and uncontentious. However the guidance on grossly offensive, indecent, and obscene communications is unlikely to help users of social media gauge when they are at realistic risk of prosecution. Crucially, it contains no explanation about what qualifies as ‘more than’ offensive, shocking or disturbing speech, rude or iconoclastic comment, the expression of unpopular opinion etc.

Part of the difficulty stems from the inherent vagueness of the underlying offence – gross offensiveness, indecency and obscenity are inherently subjective concepts. Yet this is precisely why clear guidance is needed. Defining something negatively, by reference to what it is not, does little to clarify matters.

The DPP’s public interest criteria are sensible but limited. A number of other factors could be included. For example:

  • Whether the communication was intended to be made in jest – This was one of the DPP’s reasons for not prosecuting Daniel Thomas, author of the offending tweets about Tom Daley and Peter Waterson. Generally speaking, an ill thought-out joke is less likely to warrant prosecution than a deliberate attempt to cause gross offence.
  • Whether the communication is a first time offence – A prosecution is more likely to be in the public interest where a person has previously sent other messages of a similar nature.  Conversely, first-time offenders may be better dealt with by a warning.
  • The existence of concurrent civil proceedings – If a civil claim has already been started against the suspect – for example proceedings for libel or harassment – then a prosecution may not be necessary, since the communication is likely to be examined and sanctioned by a court anyway.

Some commentators have objected to the very idea of ever bringing prosecutions for messages posted on the internet. However there are plainly good reasons why cyberspace is not immune to the reach of the criminal law. The ability to communicate instantly with an audience of millions can cause significant harm as well as good. A message posted on Facebook or a threatening tweet can be equally menacing as a poison pen letter, a silent phone call or an intimidating email (which it is well established can fall within the proper ambit of the criminal law). Electronic witch-hunts can accumulate momentum at a frightening pace, with devastating consequences for victims.

The guidelines feed into a wider debate about the use of the criminal law to restrict offensive and inflammatory speech. The European Court of Human Rights has repeatedly emphasised that Article 10 protects speech which shocks, offends and disturbs. However some public order prosecutions are hard to reconcile with this principle. In Abdul v DPP [2011] EWHC 247 (Admin), for example, the High Court upheld convictions under s5 of the Public Order Act 1986 against a number of protestors who shouted ‘burn in hell’, ‘rapists’, ‘baby killers’ and ‘terrorists’ at soldiers returning from Iraq and Afghanistan. In CPS v Choudhury a man was convicted under s5 for chanting ‘British soldiers burn in hell’ and burning poppies outside an Armistice Day memorial service; whereas a teenager who posted a picture of a burning poppy on Facebook escaped prosecution.  And in Hammond v DPP [2004] EWHC 69 (Admin) a man who held a placard with the words ‘Stop Immorality’, ‘Stop Homosexuality’ and ‘Stop Lesbianism” was convicted of a public order offence after the High Court found that his words were capable of being ‘insulting’.

There is also an alarming lack of consistency in sentencing. A sentence of 12 weeks imprisonment for posting tasteless comments about an abducted child, or eight weeks imprisonment for making racist tweets about a stricken footballer, is hard to square with the penalties of £624 imposed on nine individuals who named and denigrated a rape victim on Twitter. And it is difficult to understand why wearing a T-shirt mocking the deaths of two police officers attracts a four-month prison sentence, while mocking the deaths of six soldiers on Facebook receives a community order.

There are important general questions about the direction of the law in this area. Parliament has been accused of enacting criminal laws that stifle free speech. The Racial and Religious Hatred Act 2006, which made it an offence to deliberately stir up religious hatred with threatening language, was widely criticized as a draconian restriction on free speech even though it is subject to an express ‘freedom of expression’ defence  (‘Nothing in this Part shall be read or given effect in a way which prohibits or restricts discussion, criticism or expressions of antipathy, dislike, ridicule, insult or abuse of particular religions or the beliefs or practices of their adherents, or of any other belief system or the beliefs or practices of its adherents, or proselytising or urging adherents of a different religion or belief system to cease practising their religion or belief system.’).

There have also been moves in the opposite direction. In 2009 Parliament abolished the offences of criminal libel and blasphemy. More recently, the House of Lords voted to amend s5 of the Public Order Act to remove ‘insulting’ words and behaviour from the scope of the offence, while moves are afoot in the Crime and Court Bill to abolish the arcane offence of ‘scandalizing’ (i.e. insulting) the judiciary.


The DPP is in a difficult position: providing useful prosecutorial guidance without rewriting the law is no easy task. The guidelines illustrate the difficulty of applying old laws to new media. While the guidelines contain a good deal of common sense, users of social media still face considerable uncertainty when it comes to predicting which offensive communications will fall on the wrong side of the law.  The prosecutions and non-prosecutions to come will determine the contours of the law.  In the meantime, all those who use social media would be wise to tread carefully.

Alex Bailin QC and Edward Craven are barristers at Matrix Chambers.


  1. Póló

    Thank you for this post which is very useful (even though I am not in UK).

    As you say, there is no reason why social media should be immune from the rules of civilisation. On the other hand, as much social media is effectively a gateway to alternative media, we don’t want to give those in authority any excuse to shackle it more than is absolutely necessary.

    I have two twitter accounts, four blogs and an extensive website, and I treat everyting I post as a “publication” and attempt to show due regard to others and to the law. This does not mean that I do not slag the Pope or the Roman Catholic Church or my own political “leaders”. Quite the reverse. But I am conscious of the wide reach of what I am launching into cyberspace and treat it as if it were a book or a newpaper article or comment.

    I have no time or sympathy for the rash of vindictiveness, bullying, and the like which is currently sweeping through the social media. Action against this sort of thing is quite in order. Perhaps the use of social media would be a useful module in school civics courses.

  2. Lawrence van Rijn

    i was reading through this and questions arose.
    In article 1. Does this apply when it is send to a mass, and only a few find it offensive? does the 5% rule count? also, considering this: “(1) The individual has taken swift action to remove the communication or expressed genuine remorse;”
    The person could bully, then remove it, stating ‘accidentally published, item removed, our apologies’. The damage would already have been done.

    My other issue is the protection of minors in this. My issue is that cyber bullying starts younger and younger. Their advocated voice of remorse seems to be less and less genuine.
    Giving them a free pass should not be an option any more. There is the question what would be appropriate in that case, however, clear punishment should be handed out (of course depending on the age).
    Just a few thoughts that came to mind.

Leave a Reply

© 2023 Inforrm's Blog

Theme by Anders NorénUp ↑

%d bloggers like this: