Key findings
- Bladed weapon attacks have become the favoured methodology of terrorists in Europe because they are quick, cheap to execute and hard to prevent;
- Furthermore, such attacks fulfil the limited objectives of the most prolific terrorist actors: ample media attention, public alarm, and in many cases, ‘martyrdom’ for the attacker;
- However, their rudimentary nature means that bladed weapon attacks typically have a limited physical impact and cause minimal losses;
- Many recent bladed weapon attacks in Europe have not resulted in any fatalities, caused relatively few serious injuries, and rarely caused any damage to property;
- While bladed weapon attacks can cause significant business interruption losses, instances of this are rare, as cordons are generally limited in both extent and duration;
- While the longer-term, cumulative impact of multiple bladed weapon attacks may affect public confidence and ultimately lead to loss of attraction impacting businesses, the immediate economic impact of such attacks is normally small and unlikely to exceed policyholders’ retentions;
- Crisis management and business continuity planning and exercises are indispensable tools for minimising the impact of bladed weapon attacks on businesses.
Context
Terrorists using bladed weapons to conduct attacks is not a new phenomenon. In 2010, Stephen Timms MP was stabbed by Islamist extremist at a constituency surgery. Three years later, Lee Rigby was killed by terrorists in a combined knife and vehicle attack. In May 2016, Al Qaeda’s flagship English publication Inspire exhorted followers to conduct bladed weapon attacks. Shortly thereafter, Daesh followed suit, publishing guidance and circulating instructional videos online. The success of Daesh’s propagation of ‘low complexity’ Methodology saw a rapid increase in bladed weapon attacks in Europe from 2014. While 2017 represented a high point for bladed weapon attacks, their frequency remains well above historical averages, and Great Britain experienced three in as many months between November 2019 and February 2020. In response, Pool Re has compiled and analysed a dataset of attacks involving bladed weapons which have occurred in Europe over the past 13 months to assess their implications for the (re)insurance market and other stakeholders.
Overview
Since January 2019, there have been at least 20 incidents in Europe in which individuals have used bladed weapons to attack either random members of the public or uniformed security personnel.
Not all of these incidents were terrorist attacks.[1] However, in twelve of these instances, the authorities either confirmed that the violence was terroristic, or have not discounted a terrorist motive on the part of the attacker. For the remaining eight, the nature of the attacks led police to investigate whether the offender was motivated by extremist ideologies before discounting terrorism as a motive.
This ambiguity has arisen from the both the increasing prevalence of terrorists employing ‘low complexity’ Methodology involving the use of knives, and the emulation of this behavior by violently disturbed individuals without any links to extremist ideologies. While such incidents in the UK would not be certified as terrorist attacks (meaning terrorism policyholders would not be able to claim for any related losses), their consequences are largely indistinguishable from certified terrorist attacks using bladed weapons, hence their inclusion in this study.
Perpetrators
All but two of the bladed weapon attacks in Europe since January 2019 were carried out by individuals acting alone. While there are examples of marauding knife attacks by groups of terrorists, these are markedly less common.
The prevalence of lone actor bladed weapon attacks is unsurprising. First, lone actors frequently lack the tradecraft to plan and resource more attacks utilizing more complicated Methodology. Second, one of the main advantages of bladed weapon attacks is that minimal attack planning is required, making it easier to avoid advanced detection and interdiction by security services. The involvement of multiple assailants requires a level of coordination that, to some degree, will obviate this advantage. Finally, acquiring a bladed weapon is very easy, with few restrictions placed on buying them online or physically in stores. This is exemplified by the recent Streatham attack, where the attacker stole the knife from a high-street store.
The two attacks involving multiple attackers both involved assaults on guards within prisons, where the propinquity of co-conspirators is less likely to draw the attention of counter-terrorism authorities.
The two prison attacks were both conducted by known Islamist extremists, while at least a further three of the 20 attacks were carried out by individuals who had either been convicted of terror offences or were known to police for their extremist views. Given that both categories of individuals can expect to be subject to ongoing scrutiny from counter-terrorism authorities, the minimal planning required for bladed weapon attacks significantly increases their likelihood of success.[2]
Weapons and tactics
In all 20 of the attacks analysed the perpetrator employed one or more bladed weapon, most of which are believed to have been large chef’s knives. Notable exceptions include the two recorded attacks which occurred within prisons in which edged weapons improvised from sharp objects were used. While earlier terrorist attacks have seen the use of purpose-built melee weapons (swords, combat knives etc.), the much greater availability of kitchen knives, and the ability to legally purchase them without scrutiny, means they are more likely to be employed by terrorists and other malicious actors.
Bladed weapon attacks in Europe have previously been combined with other Methodology to increase their efficacy, most notably alongside vehicular impact. However, the frequency of vehicular impact attacks has dropped significantly since 2017, possibly partly due to the widespread implementation of hostile vehicle mitigation measures in urban areas. In all of the bladed weapon attacks in Europe since the start of January 2019, the attackers have commenced their attack on foot.
In one incident, the attacker detonated incendiary devices inside a building while conducting his attack, while in a further three attacks, the perpetrators wore imitation explosive vests. The intent to use incendiary devices in combination with bladed weapons is not unprecedented, as seen at Borough Market in 2017[3], and their employment is a relatively simple means of causing confusion and alarm, complicating the response of authorities, and potentially increasing the attack’s lethality. It also essentially guarantees some level of damage to property, which is rare in attacks where only bladed weapons are used. The use of imitation explosive vests, while also having the potential to cause additional panic, is primarily intended to ensure ‘martyrdom’ (i.e. that the attacker is killed by responding officers). These imitation vests also may be used to deter by-standers from intervening. However, the 2019 Fishmongers’ Hall attack suggests the presence of such vests will not necessarily dissuade members of the public from intervening.
Lethality
The number of non-lethal casualties caused by bladed weapon attacks is typically significantly higher than the number of fatalities. Over the 20 analysed attacks, there was an average of 2.65 injuries per attack, with only two attacks causing in excess of five casualties.
The lethality of bladed weapon attacks conducted by lone actors is likely to be lower than those involving multiple assailants. In the June 2017 London Bridge attack (which saw three terrorists attack members of the public with knives) six people died from stab wounds. Of the 20 attacks analysed, only five resulted in one or more fatality (nine deaths in total, an average of 0.45 deaths per attack). The most lethal was the December 2019 attack on the national police headquarters in Paris, in which radicalized police employee Mickael Harpon killed four colleagues. Harpon likely benefited from his ‘insider’ knowledge of the target location and familiarity with his victims. The second most lethal of the analysed attacks was the November 2019 attack in Fishmongers Hall, in which two people (who were also known to the attacker) were killed.
This suggests that bladed weapon attacks by ‘insiders’ or those known to their victims are likely to be more lethal. It could also indicate that bladed weapon attacks carried out indoors are more likely to cause higher number of casualties due to the reduced ability of potential victims to put distance between themselves and an aggressor.
Available evidence suggests that most fatalities occur early on in the attack, with attackers often exploiting the element of surprise to inflict lethal injuries. Thereafter, physical and emotional exertion reduce the ability of the attacker to cause life-threatening injuries. Furthermore, nearby members of the public are typically quickly alerted to the danger and are able to either evade or confront and subdue the attacker (as happened in at least two of the 20 analysed attacks) until the police arrive.
The majority of attacks against uniformed security personnel are quickly ended by the targeted officer or colleagues either killing or subduing the assailant. This happened in five of the eight recorded attacks against security personnel. In two other such incidents, the attacker managed to evade immediate capture but did not harm anyone else.
Target and location
More than two thirds of the analysed attacks occurred in France (nine) and Great Britain (five). This is broadly reflective of the incidence of terrorism by European country more generally, with those two countries facing the acute terror threat. However, all five of the attacks in the UK were treated as terrorist in nature, whereas prosecutors only confirmed a terrorist motive in four of the nine attacks in France. Of the 20 attacks, twelve saw the indiscriminate targeting of civilians, with the remainder involving assaults on uniformed security personnel (police officers (six), publicly employed security guards (one) and solders providing aid to the civil power (one)).
While there was considerable variation in the location of the attacks, a significant minority (nine) occurred within the centre of major cities. Such areas have traditionally been seen as at higher risk of terrorism due to the density of iconic sites, government and financial infrastructure and crowded places. Given this, the preponderance of attacks which occurred in these areas is unsurprising. However, the volume of attacks which occurred outside areas traditionally considered at high risk form terrorism was notable. There were four attacks in smaller cities and towns, but also three attacks in suburban areas of major conurbations and two attacks in small towns. Even excluding those attacks in which terrorism was later discounted as a motive, the geographical distribution of events across area types is broadly the same.
This suggests that, while major city centres are more likely to experience attacks, incidents outside such areas are not markedly rarer. This supports the theory that the most prolific terrorist actors in European countries are now less fixated on attacking iconic or high value target than previously. Indeed, in some cases they may consciously eschew targets which are likely to have robust counter-terrorism measures in favour of more remote areas with fewer counter-terrorism resources committed. It also potentially speaks to the greater spontaneity of attacks (at least those employing bladed weapons), with a minority of the analysed attacks displaying significant planning like hostile reconnaissance or the acquisition of materiel other than a knife. While these findings are not conclusive, they do suggest that terrorism risk is becoming less concentrated in major urban conurbations and capital cities.
Losses arising from bladed weapon attacks
Ascertaining to a high degree of accuracy the extent of insured losses or the wider economic impact of bladed weapon attacks is beyond the scope of this study. However, some general observations on the financial impact of bladed weapon attacks can be deduced from circumstances of those attacks. Attacks in which bladed weapons are exclusively used are considerably less likely to result in material damage than attacks involving explosives or indeed firearms or vehicular impacts. Only two of the analysed attacks were confirmed to have resulted in property damage. In one case, this was caused by the aforementioned use of incendiary devices alongside bladed weapons, and in the other the damage was caused by the discharge of a firearm by responding officers. In at least a further three incidents, there is reason to believe that some minor property damage occurred, but this cannot be confirmed.
Attacks indoors are typically more likely to result in material damage, due to the presence of soft furnishings and furniture that are more susceptible to damage during bladed weapon attacks than building exteriors and public spaces. Equally, the use of firearms by responding officers to stop attackers is also more likely to cause some material damage than where non-lethal means are used to subdue the attacker. In three-quarters of the analysed attacks, police are believed to have discharged their firearms to protect themselves or the public.
Due the relatively low incidence and level of property damage arising from bladed weapon attacks, significant losses are much more likely result from business interruption caused by the attack and subsequent clean-up and investigation. In particular, lengthy police cordons around the site of an attack can cause considerable losses to businesses within them, as demonstrated by the June 2017 London Bridge attack. There, a 10-day cordon resulted in losses of at least £1.4m to businesses that were unable to trade or access their premises. Notwithstanding the additional commercial difficulties linked to maintaining business activity remotely, meeting contractual obligations with clients and/or suppliers etc.
Analysis of the 20 bladed weapon attacks in Europe since January 2019, suggests that the cordon around the site of London Bridge attack was an outlier in both its extent and duration. One of the most significant influences on the extent and duration of cordons will be the number and geographical distribution of evidence, including the casualties and attackers. With the London Bridge attack, the involvement of three attackers employing multiple Methodology resulted in a significant number and density of casualties (8 dead, 48 injured, plus the three terrorists shot by police) across a complex area (the attackers movements were not restricted to major roads or a few premises off them). In the majority of the attacks in the dataset, casualties were confined to a relatively small area around the site where the perpetrator launched his attack. In only three instances, were there believed to be casualties more than 100m apart, and in these instances the total number of casualties (including the perpetrator) did not exceed four people. In each incident, the attacker followed a fairly linear path from the site of the initial attack, simplifying post-attack investigation and clean-up.
Detailed information on cordons is difficult to ascertain from open sources, but from available evidence it appears that the majority of attacks in the dataset, cordons were lifted (at least to such an extent to facilitate the return of traffic and access to nearby premises) within 24 hours. In some cases where the attack occurred or began indoors, it is believed that lengthier cordons were sometimes imposed on the directly affected premises, likely in part due to the absence of any consequent traffic disruptions. However, this is not possible verify except in the case of the November 2019 Fishmongers’ Hall attack, where the building was closed for six days. In the case of the October 2019 Arndale Centre attack the shopping centre itself reopened the following day, but the café where attack began may have remained closed for longer.
The extent of business interruption losses arising from post-attack cordons is beyond the scope of this study. However, the timing and location of attacks will be important determinants. Cordons which prevent trading by businesses which typically experience a high footfall will typically cause greater losses than those in primarily residential areas. Equally, if a cordon is imposed towards the end of the working day and lifted prior to the commencement of the subsequent working day (as appeared to be the case in several analysed incidents), business interruption losses are likely to be minimised. Of the 20 analysed attacks, all but two are believed to have occurred within working hours (0800-1800), while all of the attacks against crowded places occurred during working hours. In cases, where cordons were removed within 24 hours, most affected business are believed to have lost between 0.5-1.5 days of trading.
Moreover, employers typically have a duty of care to provide ongoing support to employees who have been involved in a terrorist incident. Witnessing violent action can be very distressing, and could necessitate employees taking time off over an extended period for mental health reasons, resulting in long term costs to business. Providing the necessary aftercare and support for employees could also become a significant cost to employers.
Additionally, a long-term economic impact that may affect businesses following a terrorist event stems from loss of attraction and damage to reputation. Potential customers may avoid a business that was either caught up in or located in close proximity to a terrorist event out of fear.
Forecast
The frequency of bladed weapon attacks in Europe has climbed significantly in the past decade, with large increases corresponding with the rise of Daesh and its propagation of bladed weapon attacks as an alternative to more complex attack Methodology. While the frequency of terrorist attacks declined as Daesh’s ability to inspire violence abroad diminished, the incidence of bladed weapon attacks in Europe appears to have stabilized below its 2017 peak but well above historical averages.
It is anticipated that the methodology will continue to present an attractive alternative to terrorist actors who are constrained by limited tradecraft or scrutiny from security services. While protective security measures can limit their efficacy, bladed weapon attacks require virtually no planning or training, making them difficult to prevent. Therefore, a material decline in the frequency of bladed weapon attacks is not anticipated as long as the tactic remains a viable method for causing public alarm, garnering media attention, and providing an opportunity for ‘martyrdom’ for those who employ it.
Indeed, the frequency of bladed weapon attacks could increase once more, as the large pool of terrorists convicted of historic offences are released in the coming months and years. Many of these individuals continue to harbor malicious intent and are cognizant that plots involving bladed weapons much more likely to reach fruition and achieve their often limited aims, relative to more complex Methodology.
Policy response
There are few viable legislative responses available to constrain the ability of terrorists to conduct bladed weapon attacks; the ubiquity of knives means attempts to restrict terrorists’ access to them would have a marginal effect at best. Nonetheless, there are several measures which organizations can implement to improve their resilience to bladed weapon attacks.
Forward planning and staff training are key to limiting the impact of attacks. Actions taken during the initial stages of an attack can make a big difference to the effect it has. Therefore, staff should be alert to suspicious behavior and potential dangers, and prepared to respond in the event of an attack. Potential responses include invacuating or evacuating premises, and controlling access points. Emergency response plans alone are not sufficient; staff should undergo regular training and undertake exercises to ensure they know how to act in an emergency. Appropriate equipment is also important; frontline security staff should be issued with personnel protective equipment, and the presence of trauma kits and trained first-aiders can save lives.
Effective risk management includes not only emergency responses, but also recovery planning. Businesses should consider contingencies like denial of access, loss of attraction and the key person dependencies.
[1] We have included these non-terroristic attacks within the dataset as they were largely indiscriminate attacks, rather than a targeted attack related to criminality. As such, the consequences for businesses remain largely the same, and their inclusion offers more points of reference for analysis.
[2] Whilst levels of violence in prisons are much higher than within the general public, we have included these within the dataset, as both incidents were perceived and investigated as terrorist attacks, and may have been mounted in response to calls for such attacks in extremist literature.
[3] The attackers at Borough Market had a number of petrol bombs in the back of their van but it is assessed that they did not use them as they crashed the vehicle before entering the market. If they had used them, the fatality and injury rate could have been much higher and there would likely have been significant property damage caused by fire.