Skip to main content

The Manchester Arena Inquiry has now concluded. The closure notice from the Inquiry Chairman is available here.

Volume 1: Security for the Arena
Volume 1: Security for the Arena (large format)

Training

Key findings

SMG

  • SMG should have ensured that its CCTV operators had SIA CCTV monitoring licences.
  • SMG provided inadequate training to its CCTV operators in relation to counter-terrorism.
  • SMG’s senior staff had undertaken adequate general training in counter-terrorism.
  • SMG did not provide adequate CCTV-specific counter-terrorism training for its senior staff, despite requests from both people who should have received it. This was a significant failure.

Showsec

  • The content of Showsec’s online training for its stewards in relation to counter-terrorism was capable of improvement but was adequate.
  • Showsec should have followed up this online training with practical, person-to-person training, which checked that the online training had been understood and built confidence around the reporting of concerns.
  • Showsec should have made counter-terrorism refresher training compulsory for stewards.
  • Showsec’s procedures for ensuring that counter-terrorism training had been understood by its stewards were inadequate.
  • Had Showsec taken robust steps to ensure that all of its staff had completed their training diligently, provided a practical opportunity to demonstrate their knowledge and build their confidence, it is possible that either or both Mohammed Agha and Kyle Lawler would have acted more decisively to escalate Christopher Wild’s concerns regarding SA.
  • The role occupied by Mohammed Agha on 22nd May 2017 required specific training. Mohammed Agha did not receive adequate training in that role. This reduced the prospect of him responding appropriately and robustly to Christopher Wild’s concerns.
  • There was confusion about the functionality of the radios issued to some Showsec staff.
  • Showsec staff should not have been used to monitor SMG’s CCTV during events: they lacked the necessary SIA licence, training and experience.
  • Showsec took adequate steps to ensure that its senior operational staff had sufficient counter-terrorism training.

SMG and Showsec

  • While it reflects well on both SMG and Showsec that they undertook joint training exercises, they did not do so in relation to counter-terrorism.

SMG

CCTV operators’ counter-terrorism training

SMG required its CCTV operators to have an SIA door supervisor’s licence.427 However, SMG’s CCTV operators did not undertake the CCTV component of the SIA licence.428 The SMG employees should have undertaken such training.

I cannot say that the absence of SIA CCTV training for the SMG CCTV operators made any material contribution to events on 22nd May 2017. But it reveals a lack of care on the part of SMG towards an important aspect of its counter-terrorism strategy.

SMG provided in-house training for its CCTV operators across a number of areas.429 That training did not include any significant CCTV counter-terrorism element, such as in relation to the identification of hostile reconnaissance or activity which may be an immediate precursor to a terrorist attack.430 Given the importance of CCTV as a counter-terrorism measure, this was a substantial deficiency.

Paul Johnson, the SMG Security Supervisor, stated that he asked SMG to provide CCTV-specific training more than once. He stated that “a couple of control room operators” asked him for such training. He was told by SMG that because it was not a legal requirement, CCTV-specific training was not necessary.431 As a result, Control Room Operators did not receive SIA CCTV training.

I accept that Michael Edwards, who was the Control Room Operator in Whisky Control on 22nd May 2017, did have a general understanding of what hostile reconnaissance was. I also accept that he understood that his role involved looking out for what he regarded as suspicious.432 Despite this, I was struck by the fact that in 11 years of acting as Control Room Operator, Michael Edwards had never once suspected that he was observing anyone engaging in hostile reconnaissance.433 This was a product of the lack of adequate training and the shortcomings in the CCTV system. I address both of these later in Part 6.

SMG senior staff’s counter-terrorism training

James Allen, Miriam Stone and Paul Johnson had all received Project Griffin training.434 Like Project Argus, Project Griffin was a training module provided by NaCTSO. While Project Argus was “focused more at a strategic level”, Project Griffin was “focused at more of a ground level in terms of those people delivering security”.435 Miriam Stone and Paul Johnson also attended Project Argus training.436 James Allen undertook the Showsec stewards’ counter-terrorism module.437 Paul Johnson had undertaken the SIA door supervisor training and had held an SIA licence for this activity.438 All of these represent good practice by SMG.

Where SMG should have done better for their senior Arena staff was in relation to CCTV. Both Miriam Stone and Paul Johnson’s responsibilities included the viewing of CCTV during events. Both Miriam Stone and Paul Johnson requested external CCTV training. Both were told that this training was not a legal requirement. As a result, both were told that they could not have that training.439

I address the confusion caused by SMG’s corporate structure and SIA licence in Parts 2 and 3. Setting the legal position aside, it was obviously highly desirable that both Miriam Stone and Paul Johnson received proper training in the use of CCTV from a counter-terrorism perspective. They were the most senior SMG personnel in Sierra and Whisky Control. It was a significant failure in SMG’s training regime for its senior staff that no CCTV-specific counter-terrorism component was provided. That this failure occurred, despite the fact that both relevant people were requesting this training, makes this failure all the more serious.

Showsec

Online counter-terrorism training for stewards

Showsec had its own bespoke online e-learning platform which it had developed in collaboration with Derby University from 2011.440 There are approximately 60 to 70 online courses available on the platform.441 Showsec required all staff to undertake six modules from the online training before they started in their role as a steward.442 One of these was a module devoted to counter-terrorism.443 There were additional modules which could be undertaken as a steward’s duties required once they had commenced work, including one specific to the Manchester Arena.444

The counter-terrorism module was entitled “Counter-terrorism at Events”.445 It consisted of a series of webpages with information relevant to the identification of hostile reconnaissance, the UK threat level and a link to the MI5 website.446 It provided examples of terrorist attacks which had taken place in other parts of the world.447

The training also provided links to NaCTSO-prepared training videos: Operation Fairway and Eyes Wide Open.448 Operation Fairway was described in the training text as containing “very important and relevant” information to the module.449 One of the key measures of the Eyes Wide Open videos was having “the confidence to report suspicious incidents.”450 The inclusion of these two videos was an example of good practice by Showsec. Showsec anticipated that the whole module would take approximately 45 minutes to complete.451

Guidance was provided at the start of the module which stated the following to those undertaking the training:

“Your vigilance is essential to ensuring protective measures are kept. Stewards/SIA staff will know their work areas very well and therefore maybe better identifiers of risk than their supervisors or line manager. It is therefore essential workers alert any unusual behaviour or items out of place to their supervisor. Workers should be confident speaking to their supervisors if they believe there is suspicious activity on the premises and should understand the importance of reporting these (even if it is a false alarm).”452

This was an appropriate way for the training to begin as it emphasised the twin key messages to staff of the need for vigilance and having confidence raising matters with their supervisors. There was insufficient evidence for me to determine whether or not Showsec’s training generally was better than that provided by other similar organisations at the time. However, I do regard this as an example of good practice within the online counter-terrorism training.

The training included reference to “patrolling” as being one of the measures that might be taken to “help reduce the chance of a terrorist attack”.453 This is a further example of Showsec correctly identifying the importance of patrolling as a counter-terrorism measure.454 The inclusion of patrolling in the counter-terrorism module makes it all the more inexplicable that Showsec did not put this into practice in the City Room.

The Security Experts were critical of the content of the online training in a number of respects.455 While I agree that the content was capable of being improved, it was nevertheless adequate. The target audience was Showsec stewards who, without a licence, were not permitted to carry out active profiling of audience members.456

The fact that the content of the online material was adequate is not the end of the matter. Those undertaking the training may be doing so in their free time,457 and will be using their own equipment to access the content.458 While it was not unreasonable for Showsec to require its staff to do this initial training in their own time459 and with their own equipment, it gives rise to a number of obvious risks. There is an obvious and significant risk that the person doing the training would not be doing so in an optimal environment. They may be doing it late at night, others may be present while they are doing it, they may be distracted, they may be doing it piecemeal. There is a risk that the person may skip through the content without paying any real attention. There is a risk that they will be accessing it on a device with a small screen which may make absorbing the information difficult.

There were a number of measures available to Showsec to improve the prospect that the important content of the online training was being understood and assimilated. The webpages could have been displayed for a minimum period before permitting a trainee to move on.460 This would prevent the trainee skipping forwards. A knowledge test at the end of the training would provide an opportunity to check that the trainee understood the content. In fact, in December 2016 Showsec did introduce a knowledge test to the counter-terrorism module.461 While this important improvement was in place several months prior to the Ariana Grande concert, it was of no benefit to those working for Showsec in the City Room that night. This was because all of them had undertaken the counter-terrorism module prior to December 2016462 and there had been no requirement to retake it once the knowledge check was implemented.463

A further check that Showsec should have implemented was the monitoring of the duration of the time it took trainees to complete each module. Showsec commissioned a retrieval of the time spent by Mohammed Agha on his training for the purposes of collecting evidence for the Inquiry. Following this, the Inquiry requested the same information in relation to a number of other key Showsec employees.464

This data is not the product of a system which was designed and intended to monitor compliance. It is a re-creation from information which is now available.465 For this reason there are limitations to this data.466 However, I am satisfied that in the case of four of the 10 people for whom there is data available, they moved through the counter-terrorism module at too great a speed to properly understand its content.467

In the case of Mohammed Agha, he accepted that if he had attended to his studies properly, he would have been better informed as to how to carry out his duties on 22nd May 2017.468 He also stated he had not watched any videos as part of his training.469 This supports the conclusion that he did not carry out the online training diligently.

Showsec should have foreseen these risks and taken robust steps to ensure that its staff had taken on board, in a meaningful way, the critically important information contained within the online training. Such information was capable of making a large number of people much safer through the actions of Showsec staff.

Once the online training had been completed, it remained available to staff to return to it of their own volition.470 Showsec also had a facility to encourage its staff to undertake refresher training in the counter-terrorism module using an e-shot.471 This facility was used on 14 November 2015 immediately following the terrorist attack in Paris.472 It was good practice on Showsec’s part to recognise the need for refresher training in light of the situation in France. However, Showsec did not make this refresher training compulsory.473 Showsec should have done so. I do not accept that the reasons given for not making it compulsory provide an adequate justification given the importance of this information.474 I found it particularly concerning that even after the Attack, Showsec did not immediately mandate refresher training in counter-terrorism for all its staff.475

In addition to the counter-terrorism module, Showsec’s portfolio of e-learning topics included one specific to the Arena. This served as an introduction to the layout, systems and terminology specific to the Arena. It was a customer service-focused document.476 It did not contain any information which focuses the attention of those undertaking that training on matters relevant to counter-terrorism when working at the Arena.

It did, however, envisage that supervisors would have “thoroughly checked their working area of any issues ahead of egress” before completing a pre-egress report.477 This was an appropriate and necessary procedure. It would have been improved by making clear what the supervisor might be on the lookout for. Within its own terms, though, it was an adequate description of what needed to occur. The City Room was one of the working areas for Showsec. The mezzanine was part of the City Room. The natural meaning of this sentence is that the City Room, including the mezzanine, would be thoroughly checked before egress. In practice, Showsec did not interpret this part of its own training as requiring a thorough check of the mezzanine for the potential threat from a terrorist. As a result, the check was confined to the direct route from the Arena concourse doors to the Fifty Pence staircase, and from the doors to the raised walkway to the Arena concourse doors.478

Other counter-terrorism training for stewards

Showsec’s training incorporated a period in the classroom of at least half a day.479 I accept that this included some element of counter-terrorism training. It is unclear from the evidence I heard exactly what it contained.480 However, I heard no evidence that the classroom work on counter-terrorism involved a structured, formal and robust check that all of the online training had been understood. Nor did I hear evidence that there was any other testing of the knowledge of the individuals who participated in it.481 At no point has it been submitted on Showsec’s behalf that this training did contain this element.

Whichever way it was done, a person-to-person interaction would have provided a valuable check that the message of the online training had been received and understood. The online training conveyed the message that staff should have confidence reporting concerns adequately. For this to have any effect on attitude, a person would need to complete this training diligently.

It is my view that the message was sufficiently important that it needed to be repeated in a one-to-one setting to check it had been understood and to instil the necessary confidence in each and every member of staff. Everyone is different. Showsec employs stewards who can be as young as 16-years old. While age is not a reliable indicator of confidence, a significant age gap can act to make some people feel less empowered to speak up. Counter-terrorism is too important an activity to leave to chance. It was incumbent on Showsec to take adequate steps to make sure that each and every member of staff understood the importance of being vigilant and speaking up.

Briefings

Showsec had in place a system for reinforcing the counter-terrorism aspect of the training through supervisors’ briefings482 and steward briefings.483 The need for vigilance formed part of both levels of briefing.

David Middleton, who gave the briefing to the stewards on 22nd May 2017, stated that he included an instruction that he needed to know if there was anything they were not happy with.484 However, I was concerned that his answer to the question as to whether he was an approachable supervisor was that he was “a strict supervisor” and that he was “not there to make friends with staff”.485 By May 2017, David Middleton had worked for Showsec for 21 years.486 He had been a supervisor for 16 years and a senior supervisor for five years.487

Mohammed Agha did not ask any questions of David Middleton, despite not receiving any specific instruction for the role. This was a role he was undertaking for the first time that night. Kyle Lawler, while accepting that he had real difficulty recreating his mental state from the night,488 stated that he had hesitated reporting SA because of his concern about what the reaction might be.489 He also said that his concern was not focused on David Middleton but more generally that he might be accused of racism.490

The culture Showsec should have been instilling in all its staff was that they should be receptive and approachable when it came to reports of potential terrorist threats.

In any event, while it was good practice to include in the briefing both the need for vigilance and the need to report, it was necessary to do more than this and address this on an individual and personal level. As Assistant Chief Constable (ACC) Sean O’Callaghan recognised in the context of BTP briefings, there is a risk that people stop paying attention when the same message is repeated every time.491

The briefings were not an adequate substitute for the diligent completion of the online training. A subsequent person-to-person check that the importance of that training had been understood and that each and every person had the confidence they needed should have occurred.

Conclusion: stewards’ counter-terrorism training

Showsec should have had in place a system which focused on each individual. This should have checked that they had understood their online learning and provided an opportunity for the member of staff to demonstrate that understanding. This might have been done through a short period of mentorship,492 through interactive classroom-based learning which set out to assess this in a structured way, or through on-site training as part of a practical scenario.

This training should have dealt not only with what staff were looking for, but also how they should react when confronted with a potential terrorist situation. The training needed to instil in an individual the necessary confidence to report potential terrorist activity.

Two Showsec employees, Mohammed Agha and Kyle Lawler, were provided with information arising from Christopher Wild’s concerns about SA on 22nd May 2017. This information should have prompted immediate action on their parts. Both had been present at the briefing which included the need to be vigilant and to raise concerns. But neither had completed the online training in circumstances which ensured that they absorbed and understood it adequately.

Neither Mohammed Agha nor Kyle Lawler reacted as robustly and effectively as they should have. This was because, when presented with the information, neither of them had at the forefront of their mind the very high degree of importance of doing so. Had Showsec taken robust steps to ensure that all of its staff had completed their training diligently and provided a practical opportunity to demonstrate their knowledge and build their confidence, it is possible that either or both Mohammed Agha and Kyle Lawler would have acted more decisively to escalate Christopher Wild’s concerns. I address the potential effect of such action in greater detail in Part 1.

In reaching these conclusions I have not lost sight of the fact that both Mohammed Agha and Kyle Lawler had, in addition to their Showsec training, an SIA door supervisor’s licence. The SIA door supervisor licence training included the active profiling of the crowd. This went beyond the level of vigilance expected of the steward role.493 It included some basic counter-terrorism training.494 However, the SIA training was not specific to working at the Arena. Nor was it specific to the hierarchy, relationships and personalities of those others who were working on 22nd May 2017.

The duty was on Showsec to ensure that all of its staff, whether SIA licenced or not, were adequately trained. This meant ensuring a minimum of two things. First, that all staff possessed the level of counter-terrorism understanding required of the role they were given, so that they could immediately recognise when they needed to act. Second, that all staff felt sufficiently confident and empowered in the environment they were working in, and with the people they were working with, to report potential terrorist threats to the right person with appropriate speed and clarity.

While the training that Showsec offered may have been successful for some, if not many of its staff, Showsec was under an obligation to ensure that all staff had taken it on board and knew what to do. Showsec’s methods for ensuring this were inadequate.

Training for the Grey Doors role

As set out in Part 2, Mohammed Agha was allocated to an area of the City Room known as the Grey Doors. This was an SIA licence-holder position. The Grey Doors provide access to the station platforms via the platform overbridge and are positioned between the two staircases to the mezzanine area. The location of the Grey Doors within the City Room appears in Figures 14 and 15.

Figure 14: Plan showing location of the Grey Doors in the City Room. A plan showing how the Grey Doors provide access from the City Room to the station platforms via the platform bridge. Also shown is the raised walkway leading from the City Room down to the Victoria Station Concourse via steps or lift.

Figure 14: Plan showing location of the Grey Doors in the City Room495

Figure 15: Photograph showing location of the Grey Doors in the City Room. The focus of the image are four doors beneath a raised area with railings on top of it. The doors are marked as “Grey Doors”. Either side of the Grey Doors are a flight of stairs that lead up to the mezzanine.

Figure 15: Photograph showing location of the Grey Doors in the City Room496

Mohammed Agha had never previously been deployed to this position.497 He was not briefed by any supervisor as to exactly what his role was but was told that he should not leave the Grey Doors unless they were covered by another member of staff or in the event of an emergency.498 He received a brief explanation as to what his role involved from another steward. This explanation was to the same effect as the supervisor’s briefing.499 He received no written instructions in relation to this role, despite the requirements of SMG (UK)’s premises licence.500 This was an obligation on SMG, not Showsec. However, the idea which gives rise to the licence condition is a good one. I make clear, though, that I am not criticising Showsec for a failure to provide Mohammed Agha with written instructions in relation to his role on the Grey Doors. This was just one of a number of ways which was open to Showsec to make sure he knew what he was doing.

The position on the Grey Doors was unlike other positions by fire doors: they were key operated doors, which automatically unlocked in the event of a fire alarm, the function of which was to prevent access to the platform side of the barriers on the station concourse.501 As such, the Grey Doors did not present the same security risk as other fire doors. Consequently, that position required a specific briefing for a person undertaking it for the first time. That briefing should have made clear that there was greater scope for Mohammed Agha to move away from them than was the case for other fire doors.

Mohammed Agha received no specific training on how, when in the Grey Doors role, he should have escalated any concern raised with him by a member of the public. I heard no evidence to suggest that this lack of detailed instruction was a one-off oversight on Showsec’s part. Rather there was a failure to appreciate the need for such training. The training would have provided an opportunity to discuss with the person what their role involved. In turn, this is likely to have built the confidence of that person in knowing how to respond to a variety of situations they may be presented with, including a report of potential terrorist activity.

Mohammed Agha did not receive adequate training in his role in front of the Grey Doors. His training was capable of leading him to believe that it was more important to remain where he was than in fact was the case.502 In turn, this reduced the prospect of him responding appropriately and robustly to Christopher Wild’s concerns.

Training for Showsec radio holders

Some members of Showsec staff were issued with radios. On 22nd May 2017, Kyle Lawler was one such person.503 The radios were supplied by SMG.504 In addition to the standard talk function, the radios had another button on the top.505

Thomas Bailey, one of the two Showsec Heads of Security, understood this to be an “emergency button” which, when pressed, would cause the base station in Sierra Control to beep against the assigned call sign.506 He stated that this had only occurred once in his experience, when the button had been pressed by accident.507 He stated he did not know if the radios had an “override” function.508 Thomas Bailey recognised, however, that his understanding of this function was limited. He stated that he had never seen any supervisor provide training in its use when telling staff how to use the radios.509 He accepted that if the function was used infrequently staff would lose familiarity with it510 and that staff should have received regular training in its use.511 He also accepted he should have had a better understanding of it, but asserted that there were differences between handsets.512

Paul Johnson, SMG Security Supervisor, expressed the view that he did not think Showsec knew anything about this button and that he did not think Showsec staff had been told about the button or trained in its use.513 He expressed surprise at Thomas Bailey’s evidence.514 Paul Johnson said that the button was for his staff when patrolling alone.515 Paul Johnson’s account of this button’s function was that when depressed it would broadcast for 10 seconds “right over the channel it’s on”.516

Jordan Beak described a process in which a user “cut in” to a conversation to say they had a “priority message”. This, he said should cause people to stop their broadcast.517 If that did not work, he stated the radios had a red button which when pressed “cuts everybody else and communicates only your radio.”518

David Middleton described pressing “a button” in order to say that there was a “priority message”.519 It is unclear from his account520 whether he was describing the first part of Jordan Beak’s explanation, which does not involve an override function of others’ radios, or the use of a button the sole purpose of which was to interrupt the broadcasts of others.

Kyle Lawler described his training in the use of the radio in this way: “the first time I was given a radio, I was shown basically how to send a message and to change the channel and the volume. That was pretty much it”.521 He stated that he was not aware of an “override” button in May 2017.522 He stated he had seen a button on the top of the radio which he said he did not know the purpose of.523 He confirmed that in his statement he had described it as a “panic button”, but that he did not know what would happen if he pressed it.524 He stated he did not think it worked.525 He indicated that he had not given any thought to the purpose of the button.526 He also stated that he could not recall anybody ever using the “panic button” before.527 The use of the phrase panic button is consistent with Thomas Bailey’s apparent misunderstanding of the button’s function. Kyle Lawler’s description of his training is also consistent with Thomas Bailey’s.528 Kyle Lawler’s account, that he was not aware of the functionality of this emergency button, is consistent with Paul Johnson’s belief that its use did not form part of Showsec’s training. It appears that Jordan Beak did understand what the button did.

I accept Kyle Lawler’s evidence that he did not have a clear understanding of the full functionality of the radio he had during the concert. He was not aware of the purpose of the override button, what would happen if he pressed it or its ability to deliver urgent messages immediately. This was because he was not properly trained by Showsec in its use. This did not, in the event, make any difference to events on 22nd May 2017. That is because Kyle Lawler did not take immediate, decisive and robust action in response to learning of Christopher Wild’s concerns because he did not appreciate the importance of doing so. However, Kyle Lawler’s deficient understanding of the radios did reveal substantial weakness in Showsec’s training.

Showsec staff monitoring SMG’s CCTV

SMG and Showsec had an agreement that led to Showsec staff providing relief cover to SMG employees in Whisky Control. When this occurred, Showsec staff monitored the CCTV during events.529 They did this despite not having an SIA CCTV licence.530 They should have had such a qualification. The confusion caused by SMG’s corporate structure, which I have set out in Parts 2 and 3, did not apply when Showsec were undertaking such work.

An explanation for this may be Thomas Bailey’s belief that those Showsec employees would be undertaking tasks other than CCTV monitoring when in Whisky Control.531 Such a misunderstanding does not, however, absolve SMG of the responsibility of ensuring that those who monitored its CCTV system were suitably qualified. Further, and quite aside from the lack of compliance with the SIA regime, I am not satisfied that a Showsec employee would have the necessary experience and training to undertake the work of an SMG employee whose role was to monitor the CCTV.

The regulatory non-compliance and lack of training did not directly affect events of 22nd May 2017, as there were no Showsec employees in Whisky Control. It does reveal a further example of a lack of care on SMG’s part in relation to CCTV. This would not have occurred if SMG had approached CCTV as being an important counter-terrorism measure.

Counter-terrorism training for Showsec senior operational staff

Those in the Head of Security role held an SIA CCTV monitoring licence.532 The supervisors all held SIA door supervisor licences.533 Showsec was entitled to rely upon the SIA training as providing some information to those members of staff whose role included the active profiling of event-goers and potential threats to them.

Showsec’s senior operational staff Thomas Rigby, Thomas Bailey, David Middleton and Daniel Perry had all undertaken NaCTSO Project Griffin.534 In addition, Thomas Bailey and Thomas Rigby, who both undertook the most senior operational role of Head of Security, had completed NaCTSO Project Argus training.

In my view, Showsec had taken adequate steps to ensure that its senior operational staff had sufficient counter-terrorism training.

SMG and Showsec

Part of Showsec’s agreement with SMG was to provide “senior management, supervisors and key staff” with “venue specific desktop safety exercises including but not limited to fire, crowd control, terrorist attack and show cancellations.”535 This was a responsible thing for the two organisations to agree to do. As a result, Showsec and SMG organised for members of staff to participate in tabletop exercises at the Arena.536 Thomas Bailey and Miriam Stone created these exercises.537

One of those exercises, which was run on 17th December 2014, involved a scenario of a terrorist attack in the City Room.538 Although it was characterised at some points in the evidence as a counter-terrorism exercise,539 it was not focused on detecting or preventing a terrorist attack.540 The focus of this training was on what should occur after an attack had begun.541 As a result, although it reflects well on both SMG and Showsec that they undertook joint training exercises, they did not do so in relation to counter-terrorism. This exercise scenario provided an opportunity for SMG and Showsec to think carefully about all the challenges presented by the City Room in the context of a terrorist attack.