By Dr Dermot Barr

The UK national threat level was raised to ‘Severe’, the second highest level, on the 3rd November 2020 after a series of terrorist attacks in France and Austria. This level means an attack is thought to be ‘highly likely’. This blogpost discusses the psychology around this decision and its potential impacts. I will also discuss the relevance of threats and threat-level to a new ESRC-funded research project into crowd responses to perceptions of hostile threat, often problematically known as ‘stampedes’. Before we discuss the research project, let’s review the incidents leading up to this decision by the Joint Terrorism Analysis Centre (JTAC).

With roots that can be traced back to a Danish depiction of the Prophet Muhammad in 2005, through the 2015 Charlie Hebdo attack, the latest attacks in France in September and October 2020 began in response to depictions of the Prophet Muhammad in France. The pictures were republished again by Charlie Hebdo on 1 September 2020, a day before a trial connected to the 2015 attack began. Depictions of the Prophet are forbidden within Islam and seen as blasphemous by some. On 25th September two people were injured with a meat clever in an attack outside the Charlie Hebdo offices. Statements by the French and Turkish premiers arguably added to rising tension. On 16th October 2020, a school teacher, Samuel Paty, was beheaded and his attacker shot dead by police, ostensibly in response to the teacher showing the pictures to his high-school students in a class on freedom of speech. After Vienna also witnessed an Islamist Terror attack on 2nd November 2020, the concern among British security services was that violence could spread to the UK.

This fear of spreading violence is not necessarily unfounded. The depictions of the prophet and the attacks motivated collective action, both from those who found depictions of the prophet to be sacrilegious and those expressing solidarity with Samuel Paty. Within weeks, there were reprisal attacks on Mosques in Bordeaux and Montélimar and further Islamist attacks in Nice and Vienna. Research has shown the importance of shared social identities in influencing the ‘spread’ of behaviour from one geographical area to another. People who share a sense of injustice can define themselves as similar to those involved in an initial incident. This shared definition can provide the normative motivation to act in opposition to perceived common grievances in ways that embody that shared social identity. Shared social identities can be the basis of a process of empowerment where people believe that there is support for their actions and those actions are therefore legitimate. The influence of social identity processes in legitimizing and empowering social identity consistent behaviour can be seen through the increase in hate crime after events such as the Brexit vote, the global resonance of the Black Lives Matter movement or the spread of rioting in England in 2011. The events in France and Austria have the potential to show how a shared identity could act as motivation for further action to oppose perceived injustice. This is true for both Islamist terrorist sympathizers and for Islamophobes.

While the concern motivating the Joint Terrorism Analysis Centre’s (JTAC) decision to increase the terror threat level is not unfounded, it is not clear what impact this increase will have. Raising the threat level undoubtedly sends a message that ‘we’ (the UK) are under attack from ‘them’. The impact is therefore likely to be experienced differently by different groups in society depending on how they define themselves and are defined by others in terms of the UK ‘us’ and ‘them’. The effect is likely to be felt most by ‘suspect communities’. People perceived as belonging to ‘suspect communities’ will face increased suspicion, potentially fueling a narrative of persecution. In the US official discourse has been argued to legitimize the ‘war on terror’ through clash of civilizations rhetoric. These discourses may simultaneously alienate communities and legitimize and empower people to act against ‘suspect communities’.

The ‘threat level’ system has been criticized by some commentators as ‘absurd abstractions of no help to anyone except the security lobby raising cash through fear’. Indeed, the threat level is primarily a tool for security services to assess and respond to the risk of terrorism. However, as we have seen from the Covid-19 crisis, messaging from the government about different threats is integral to public perceptions of, and responses to, threats. When raising the terror threat level the Home Secretary Priti Patel asked the public to be ‘alert but not alarmed’, much like Theresa May did 10 years before her. Raising the threat level is a piece of government messaging that appears more abstract than concrete. What does the public do with that message?

MI5 state that while a direct public response is unnecessary, vigilance is encouraged, especially ‘given the current national threat’. The consequences, if not the desired goal, of publicly raising of the national threat level may be increased vigilance.

Indeed my work in the ESRC ‘stampedes’ project has found potential evidence of increased vigilance in an increase in evacuations due to false alarms raised as a result of misperceived threats posed by things like unattended bags, exploding battery packs or bearded black men carrying umbrellas. This suggests that people increased their levels of vigilance around spotting potential signs of terrorism that correspond to particular notions of terrorism.

These incidents are represented in graph 1 below.

Some of these false alarm incidents were slow evacuations like when a fire alarm goes off and people reluctantly walk outside once they realise it’s not a drill. Others were much more urgent affairs where people ran for their lives. Perhaps the most famous of these ‘urgent’ crowd response incidents in recent years was on Black Friday 2017, the busiest shopping day of the year, on Oxford Street, London. This incident saw hundreds of people running from what they thought was a terrorist attack, thoughts which turned out to be unfounded.

Graph 2 below illustrates the pattern for both urgent (Orange) and non-urgent (Blue) evacuations.

Both urgent and non-urgent evacuations follow a distinct pattern that peaks in 2017, with a second peak in 2019. 2017 was the most deadly year for terrorist attacks in Britain in over a decade. This makes it difficult to disentangle the effect of messaging from the very real increase in frequency of deadly attacks. However, it is notable that the increase in both urgent and non-urgent evacuations correlates with an increase in the threat level through which government messaging to the public encouraged continued vigilance. Furthermore, while the terror threat level may be abstract, government advice if caught in a terror attack is the much more direct behavioural instruction to ‘Run, Hide, Tell’.

Graph 3 below illustrates the correlation.

With the benefit of hindsight, running from an ambiguous noise that turns out to be a false alarm may seem like an irrational overreaction from the public. You are very unlikely to be caught up in a terrorist incident and extremely unlikely to be injured or die as a result of one. However, a climate of heightened vigilance reflecting an increase in deadly attacks, increasingly visible counter-terrorism policing operations, and sensationalist media discourse meant an attack on Oxford Street was plausible. Furthermore, much like members of the public involved in the ‘stampede’, emergency services acted as if they were dealing with a real attack. Armed police and paramedics rushed to the scene, shops and roads were closed, and shoppers faced a range of instructions to avoid the area or shelter in shops. In this context, a calculation that the possible cost of not acting and being caught in an attack came to outweigh the risk of acting and looking foolish. As previously mentioned the government’s advice if caught in a terror attack is to ‘Run, Hide, Tell’. Rather than over-reacting, arguably the public followed the government’s advice.

Moreover, 2017 did see an increase in both the frequency and severity of terror attacks. There was an objectively higher risk of being subjected to a terror attack that year. The risk remained extremely small for an average member of the public but it was greater. Evaluation of this risk may also have been altered by the sensationalist media reports that reported false alarms as if they were near misses rather than misperceptions. Despite this, people were ridiculed for ‘panicking’ (especially Olly Murs).

Evidence gathered as part of our ‘false alarms and ‘stampedes’ project problematizes the pathologizing of the public as prone to ‘panic’, noting that in most emergency situations there is a tendency for the public to underestimate risk while in certain other contexts (high actual incidence, high relevance, high cost) people may be more likely to perceive threat, even if there isn’t one. Additionally, the emergency services are rarely criticized as panicking even when they make similar risk assessments as the public. The project is also working with the emergency services to analyze their role in these situations. There is a particular focus on developing effective different messaging for authorities during crowded events where a hostile threat is perceived. This messaging is likely to be much more direct and concrete than the abstract threat level can be.

Clearly, correlation itself does not imply causation. The effect of increasing the threat level is difficult if not impossible to disconnect from the increasing attacks it reflects. The increase in misperceptions of hostile threats and the increase in urgent public responses to these misperceived threats could be related to the increase in frequency, magnitude or relevance of real attacks without any reference to the national threat level. Neither is increased vigilance alone enough to explain this shared social behaviour. The role of shared understanding of the stimuli and environments that signal threat indicates the inability of individual vigilance to provide a full explanation of our collective (mis)perceptions and responses. Yet it does appear to be a necessary component in a full explanation. Despite these limitations, I hope to have pointed to some of the areas where psychology can help understand collective fear responses to perceived terrorist threats.

Dr Dermot Barr is a postdoctoral research fellow working on the ESRC funded ‘Stampedes’ project with Prof John Drury, and he is part of the Crowds and Identities Group. This post was originally published on 3rd December 2020 in the blog of the Crowds and Identities Research Group.

Find out more about our research on Social and Applied Psychology.