Thursday 24 September 2015

Hajj 'stampede'?

Today saw the worst tragedy at the Hajj in Saudi Arabia for 25 years, as 717 Muslim pilgrims were killed and nearly 900 injured in a crowd crush while on their way to the holy sites of Mecca. Information is still filtering through, but it looks increasingly like a fatal crush occurred when two large groups of pilgrims converged from different directions onto one road, known as 204 Street (see the following map for more details). As a result of this tragedy, the Saudi King has ordered a safety review of the pilgrimage. This safety review is of course welcome news, but I also worry that the coverage of this disaster still draws too easily upon outdated notions of crowd behaviour- namely the use of the term 'stampede' to describe what happened, and that the media must stop using the term to describe such incidents.

Rushing to use the term 'stampede'?
As soon as news started breaking in about this tragedy (about 10.00 GMT on 24/9/15), the UK media began using 'stampede' in their headlines (and have largely continued to do so, ever since), which rapidly launched myself and others onto Twitter to take issue with the use of the term. In a blog from 2011, John Drury looks in detail at why the term 'stampede' is problematic when describing emergency behaviour, because it implies selfish and/or animalistic 'panicked' behaviour by those affected. More recently, in a blog after the Shanghai crowd tragedy on 13/12/2014, I argued that calling such tragedies as stampedes is rarely supported by later detailed examination of events. For instance, people rarely deliberately trample over others in crowd disasters (as is implied by the term 'stampede'), with victims more likely to die of compressive asphyxia because of dangerous levels of crowd density. Furthermore, the term 'stampede'  could also serve to deflect blame away from possible crowd management failings and onto the victims themselves (e.g. "crowd 'panic' causes disasters"). This point of prematurely attributing blame was addressed by John Drury and Keith Still, in interviews for the Telegraph newspaper today, where they were both critical of the apparent attempts by some members of the Saudi authorities to blame crowd members for the tragedy before the full facts were known. For instance, soon after the disaster happened, the Saudi Health Minister was reported as already speculating that the tragedy was caused by crowd members ignoring official advice;
'If the pilgrims had followed instructions, this type of accident could have been avoided'

Conclusion:
The terrible scenes in Mina show the dangers that can exist in large crowds. However, I believe very strongly that it is possible to safely manage large numbers of people, and that tragedies such as the one seen today are not inevitable. However, when such tragedies do occur (and they are mercifully rare), I don't think it helps to describe them as 'stampedes', as it is such a loaded term, and does not accurately describe what actually happens in such incidents. Furthermore, it could also serve to unduly influence any future investigations (such as the safety review that has been announced). Therefore, I believe that this is not a simple semantic issue of language use, and if we are going to improve safety at large crowd events, using outdated  terms such as 'stampede' when things go wrong, will only get in the way of trying to create safer crowd experiences for everyone involved.

hundreds of thousands of Muslim pilgrims make their way to cast stones at a pillar symbolizing the stoning of Satan in Mina, Saudi Arabia, Thursday, Sept. 24, 2015.

Thursday 10 September 2015

Las Vegas plane fire and evacuation behaviour

The recent evacuation of British Airways flight BA2276 that suffered an engine fire as it began take-off at Las Vegas airport has caused some controversy, as illustrated in a BBC article about the incident. It focuses on the chorus of disapproval circling around social media (two examples are copied below) in response to reports that some passengers delayed their own and others' evacuation as they retrieved their hand luggage (in direct contradiction to the usual instructions for evacuation procedures). Such criticism has been countered by the Guardian journalist Jacob Steinberg who found himself on this very flight, and seems to be excusing this behaviour because people 'panicked';  
'There were even some passengers who tried to get their luggage out of the overhead lockers. I’ve subsequently seen some criticism of them on Twitter but if you weren’t there, how do you know how you would have reacted? People do odd things when they panic'.
I think this incident raises some interesting issues about how people behave in emergencies. First of all, it's worth highlighting that not all passengers behaved in this way and took their hand luggage with them, so it was not action done by the group as a whole. For instance, a British passenger interviewed said that he followed instructions to leave his luggage on the plane (and reports that others also did the same) and also that he wasn't personally inconvenienced himself by others taking out their luggage (although he saw people outside the plane who had obviously done so). Furthermore, Jacob Steinberg also points out that others on the plane voiced their disapproval when it happened, so it was by no means a generally accepted norm of behaviour;
 "There were certainly shouts for people not to do it when they opened lockers"

Secondly, while people delaying their own (and possibly others') evacuation to get their luggage out during plane fires is not particularly sensible or co-operative behaviour, I would say it's not necessarily 'panic' either. This is because the classic view of 'panic' behaviour would imply that people would rush blindly to the closest exit with complete disregard for their possessions (or other people in their way) and there doesn't seem to be any evidence that this happened during this incident. Rather than 'panicking' I would suggest that focusing on retrieving one's belongings may be an example of the dissociative behaviour sometimes displayed by individuals in emergencies to help them ignore the seriousness of the situation- something suggested in a report I wrote with John Drury looking at mass emergency behaviour (Drury & Cocking, 2007, p.8).

Finally, such behaviour is not always seen in plane evacuations, and I think it's worth drawing some historical comparisons with more serious plane fires that have ended in tragedy. For instance, the Manchester airport disaster, on August 22nd 1985, killed 55 people after one of the plane's engines caught fire on take off. The air accident report concluded that most of the fatalities occurred because of delays in evacuation, since people who were not able to evacuate the plane immediately became quickly incapacitated by the toxic smoke that filled the cabin and died from smoke inhalation. However, the delays in evacuation were largely due to passengers' difficulties in being able to see their way through the smoke and also in getting the cabin doors open, and one of the surviving cabin staff reported that passengers were not carrying any 'noticeable or unacceptable hand baggage' p.43. Therefore,perhaps the less serious nature of the incident at Las Vegas airport (mercifully, everyone survived the evacuation of flight BA2276) meant that passengers felt less of a sense of urgency to evacuate than passengers would have during the Manchester airport fire, and so they thought it wouldn't be a problem if they paused to get their luggage .


Conclusion:
I'm not saying that passengers delaying their own and others' exit during plane evacuations to retrieve their luggage isn't problematic (I'm getting on a plane tomorrow afternoon and will make sure my hand luggage stays put if we have to evacuate!), but I would resist attempts to describe such behaviour as 'panic'. This is because trying to excuse such behaviour as 'understandable in the circumstances because people can't help themselves' doesn't really help us tackle the problem. Perhaps such behaviour could be better explained in terms of a social dilemma, which is something I have looked at in previous blogs. This is because what may be perceived to be in the individual's short term interests (to evacuate with all one's possessions) may not be in the group's interests (as it could delay others evacuating). So I would say the crucial thing is to try much harder to convey the message to air travellers that it's in everyone's collective interest to act co-operatively during evacuations, and that delaying your exit to retrieve your duty free from the holdall above does not serve the collective good (and could end up risking your own life as well). Furthermore, if social norms develop where such behaviour is not considered appropriate and other passengers routinely express their disapproval to those who do it, we will hopefully see less of it in future plane evacuations. Failing that, I have seen some on-line comments suggesting that overhead lockers should be electronically locked until everyone has safely evacuated during emergencies!  
   

References:
Drury, J. & Cocking, C. (2007). The mass psychology of disasters and emergency evacuations: A research report and implications for practice. University of Sussex, Brighton
http://www.sussex.ac.uk/affiliates/panic/Disasters%20and%20emergency%20evacuations%20(2007).pdf 


Tweet by Stuart McAllister
Tweet expressing disapproval of some passengers' behaviour


Cartoon by airline pilot Chris Manno