Mark's Musings

A miscellany of opinions, thoughts, rants and comments

Twitter joke trial: it’s not just about free speech


Listed in: Current Events, Internet, Politics and News

Plenty has already been said and written about the Twitter joke trial, so there isn’t really a huge amount I can add to it. But one thing did strike me as I was reading the full text of the judgment.

One of the advantages of reading what the judge has to say about something is that you tend to get the key facts presented impartially, without the usual media spin and interpolation. Picking up half way through paragraph 13, this is how Lord Judge describes the sequence of events that led to the tweet being investigated:

[The duty manager] was at home searching generally for any “tweets” which referred to Robin Hood Airport. In cross examination he said that he did not know whether the “tweet” was a joke or not, but as even a joke could cause major disruption it had to be investigated. Accordingly he referred the “tweet” to his manager, Mr Armson. Mr Armson was responsible for deciding whether any perceived threat to the airport should be graded as “credible” or “non-credible”. If “credible”, it was to be referred immediately to the Ministry of Defence, but if “noncredible”, as a matter of standard practice it was to be reported to the airport police. Mr Armson examined the appellant’s “tweet”. He regarded it as “non-credible”, not least because it featured the appellant’s name and, as he noted, the appellant was due to fly from the airport in the near future. Nevertheless in accordance with airport procedure he passed this “tweet” to the airport police. The airport police themselves took no action, presumably for exactly the same reason, but they decided to refer the matter on to the South Yorkshire police.

Up to here, this is the system working as designed. Airports are prime targets for terrorism, so it’s entirely reasonable that staff should, as a matter of course, check for anything potentially suspicious on social media. It’s also entirely understandable that even mentions deemed “non-credible” are still reported and logged, because something which appears innocuous at the time may later turn out to be significant in the light of subsequent events. So the fact that Paul Chambers’ tweet was logged and reported is to be expected. The airport authorities have also got a lot of #iamspartacus tweets in their collection now, for exactly the same reason. It was logged, it was reported, and at each stage it was deemed not worthy of action. Job done.

However, for some reason known only to themselves, South Yorkshire police decided to take action. They arrested Chambers and questioned him. And then, after taking advice to the CPS, referred the case for prosecution. He was charged, and convicted. The original conviction in the Magistrates Court was firstly appealed to the Crown Court, and then eventually to the High Court where, at the second attempt, the appeal was granted and the conviction overturned. What’s interesting here, though, is the reason given by the Crown Court for upholding the original conviction. As Lord Judge describes it:

On the basis of these facts the Crown Court was “satisfied” that the message in question was “menacing per se”. The court took the view “that an ordinary person seeing the “tweet” would see it in that way and be alarmed. The airport staff did see it and were sufficiently concerned to report it”.

If you look back at the earlier description of how the tweet came to the attention of the authorities in the first place, though, you will see that that statement by the Crown Court is simply wrong. The airport staff did not report the tweet because they were “sufficiently concerned” about it. They reported it because their standard procedures required them to report it whether they were concerned or not.

What’s happened here, therefore, is a serious failure of communication. Somehow, the airport’s policy of “report everything, however trivial” has been transmuted by the CPS and the courts into “if it has been reported then it must be important”. That should not have happened. It would not have happened had the CPS and courts understood the airport’s policy on reporting anything at all which might be a potential threat.

Of course, the High Court did grant the appeal and this was one of the points mentioned by Lord Judge when giving his reasons for granting it:

No weight appears to have been given to the lack of urgency which characterised the approach of the authorities to this problem, while the fact that those responsible for security at the airport decided to report it at all, which was treated as a significant feature, rather overlooked that this represented compliance with their duties rather than their alarmed response to the message.

To that extent, therefore, justice has been done. At least, it has in this case, and the High Court judgment has also laid down some important (and welcome) guidelines on how the law should treat free expression on the Internet. But it seems to me that there are still some significant underlying issues that need to be addressed.

It is increasingly the case that the police, security services and other law enforcement agencies take the same approach as Robin Hood airport with anything that may possibly be relevant and insist on it being logged and recorded even if those doing so have no immediate reason to be concerned. I can understand why these procedures are followed, and up to a point I don’t have any problem with it. But such a process has two significant causes for concern.

Firstly, there is the potential for a “record first, disregard later” policy to lead inexorably towards more and more monitoring and logging of material that is entirely innocuous. That represents a potential invasion of privacy. I accept that if I chat to my friend as we walk down the street, I cannot complain if we are overheard and where we go is overseen, but I still don’t expect to find that my every steps and words are being logged by a secret service agent walking three paces behind me. Equally, I know that anything I post on Twitter is visible to the whole world if they care to look, but I don’t expect it to be routinely logged by GCHQ. There needs to be a balance between deliberately searching for useful and relevant information and just slurping the web in the hope of picking up something interesting.

The other point is that the more our everyday actions and communications are logged and reported, the more opportunities there will be for the “no smoke without fire” fallacy to lead to prosecutions similar to the Twitter joke trial. “It must be important, otherwise it wouldn’t have been reported” is a very normal human reaction, but it’s also often a wrong reaction – and even more so when normal procedure is to report even the unimportant. Consider, also, the possibility for character assassination by the tabloid media when they discover that a prominent politician or sportsman has been recorded as talking about, say, drugs, sex or bribes in their emails. Whether it’s a casual, throwaway remark or a direct reference to nefarious deeds won’t make any difference – the media can simply “report the facts” and let their readers draw their own conclusions.

For all those reasons, the Twitter joke trial isn’t just about free speech on the Internet. It’s also about the increasing movement towards more and more intrusive monitoring of our everyday lives, and the ramifications for privacy and law enforcement. And the quashing of Paul Chambers’ conviction hasn’t done anything about that.

  • Simon Farnsworth

    To play the devil’s advocate here, the counterbalance is the observation from health and safety that most major incidents are preceded by many minor incidents.

    For example, a stretch of road on which there’s a fatal incident is likely to be a stretch of road where there have been minor crashes in the past, rather than one with an unblemished record. A datacentre worker who’s got a track record of pulling out too many cables, then plugging the right ones back in is more likely to be the one who takes the datacentre down than a worker who does things one cable at a time.

    Getting into criminality, the driver who routinely exceeds the speed limit in a residential area is more likely to be the driver who has a crash involving a pedestrian. A person routinely gets into fights is more likely to commit manslaughter or murder.

    Given this fact of life, surely it’s better to track minor incidents, and look for the patterns that hint that you’re about to have something major happen? That way, an incident involving an airport shut down for days and people killed by a bomb can be converted into an incident where one person is hurt by the bomb they were going to set off in the queues.

    • MarkSG

      I don’t necessarily disagree, and that’s why (as I wrote earlier) I think that the airport staff were behaving absolutely correctly by logging the tweet even though it gave them no cause for alarm. And I accept that, up to a point, the same principle applies even in aspects of life that are far less of a target for terrorism and crime than airports.

      I’m not necessarily opposing greater use of communication monitoring and pattern spotting in the way you describe. The point I’m making here is twofold: Firstly, the benefits of doing this are not cost-free; they come at at the price of losing some privacy, and it’s essential that this cost is borne in mind as a counterbalance to the otherwise natural tendency to want to intrude more and more. And, secondly (and the aspect I’m really most concerned about), the potential of the “no smoke without fire” fallacy to make criminals out of those merely careless in action or speech is a significant threat to freedom. 

      All of us have, at times, done or said something a bit dim that we later regret. But regretting them is one thing, being hauled up before the bench facing a criminal charge is another. It’s important – really important – that the criminal justice system as a whole has a firm grasp of the distinction between what is worth recording and what needs action to be taken.

      • Simon Farnsworth

        Getting away from the devil’s advocate position, I have no problem with monitoring of communications leading to sensibly targeted monitoring of a person – I do have a problem when monitoring of a person who hasn’t actually started to act leads to arrest or charges.

        Going back to the Chambers case, I could understand why the police would choose to monitor Mr Chambers’ behaviour after his tweet; had he built or bought a bomb, and taken it in the direction of Robin Hood Airport, then action would have been reasonable. Once it was clear that he wasn’t going to obtain explosives, they should have dropped it.