Riots and Probabilities

I saw on the news the other night footage of the London rioters attacking a restaurant, a casino and food store. Finally, after one year the perpetrators were given custodial sentences. That got me thinking – what starts off a riot, and why that date and not some other date? Which series of precursory events caused the riots to occur at that particular time in those particular places? Why not one day earlier, or one week earlier, or one month later? It’s not an easy question to answer, because there are so many antecedent causes and variables. But I think I do know the answer; it’s the same sort of procedural rule that can be applied to large groups with the same general causes. I’ll explain what it is. 

There’s one thing I think you should know about bad behaviour in large groups of people - it is contingent and contagious. History certainly has some ugly stains in the shape of bad human groups who have committed bad deeds. Whether it is Nazi groups, extreme Stalinists, inner city rioters, football hooligans, religious groups that put people to death as witches or heretics, or the Taliban, I think there is a way to define their behaviour - as groups that must be considered in a different way to individual acts.  

Regarding the recent rioters in London (last year) - around the time no one knew what to expect. Was this soon going to abate, or would it escalate into a kind of Hobbesian collapse of society? If you think it was always going to abate, you should be reminded that most of the major events in history were entirely unpredictable - and because of which, not predicted - and that history has a habit of throwing up excessive behaviour far above and beyond what could have been reasonably expected beforehand.  

I said ‘bad behaviour’ is contingent and contagious, and I will give you an illustration to show how I think it works. Picture two gangs of 100 youths; one gang in North London and one in East London. Both gangs have the intention to riot en masse, but at the close of the day one group, the North Londoners, cause mayhem and are all over the news headlines, and the other group, the East Londoners, are involved in only a few minor disturbances. What was different? Well psychologists tell us that when individuals become part of a larger group (street gangs, religious terrorism outfits, football hooligan groups, etc) they are deindividuated; which means they are prone to worse behaviour because of a sense of losing their individuality within a larger body of people.  

Here’s the reason why in my illustration North Londoners rioted and East Londoners caused only a few minor disturbances. Take the 100 members in each gang – you’ll find that each member of that gang will have a threshold - a tipping point at which he will be prepared to engage in hostile and anti-social behaviour. In a gang of 100 people, each individual's tipping point will vary in intensity. If we rate them 1 to 100 on an intensity scale; out of the 100, number 1 will be the individual most likely to be anti-social, number 2 will be the second most likely, and so on, right up to number 100, who will be the least likely. Now there's no way we could ever accurately rate each of the 100, but that doesn't matter - all that matters is that they could be rated 1 to 100.  

Let’s start with the North London gang – all of which went on to be complicit in a mass 100 man riot. Number 1 starts it off by lobbing a brick, number 2 soon joins in and breaks a shop window, which prompts 3 and 4 to push two shopping trolleys into a moving car. Now we have an all-out disturbance – and number 5 is easily prone to adding fuel to these types of fire. At this early stage of the riot (the soon to be actualised riot) we find that because five men are at it, numbers 6, 7 and 8 are easily persuaded to become deindividuated and join in, which makes it easier for numbers to 9 to 13, and 17 to 20, and so on. If this contagion continues, a minor disturbance is soon a minor riot with 25 or so people. Now numbers 30 to 40 and 50 to 75 aren’t the sort of people who would easily riot – and numbers 76 to 100 are even less likely. But there are already enough people rioting to elicit deindividuation in numbers 30 to 40, so they soon join in, as do numbers 50 to 75. What was a minor riot consisting of 20 people is now a fairly major riot consisting of about 75 people – and now numbers 76 to 100 find it much easier to get involved themselves. 

Bear one thing in mind – all this could happen in a matter of a few minutes; in very quick successions 1 has become anti-social, which has caused 2 to be anti-social, which has caused 3, and so on, right up to 100. Of course number 87 doesn’t have easy proclivities for mass rioting – but he isn’t often in the company of 86 of his peers who are rioting in front of him. That’s how 100 young men from North London can soon find themselves deindividuated and in a 100 strong riot – it’s a bit like a domino effect.  

But what about the gang in East London – why was there’s only a minor disturbance that soon abated? Let’s pretend the same thing happens in East London at first; number 1 starts it off by lobbing a brick, number 2 soon joins in and breaks a shop window, which prompts 3 and 4 to push two shopping trolleys into a moving car. Now we have an all-out disturbance – but this time while number 5 is easily prone to adding fuel to these types of fire, he is not as prone as the number 5 in the North London gang. Instead of carrying out an anti-social act, instead number 5 has a sudden burst of culpability and tries to grab 3 and 4 to stop them doing any more damage, after which it is much less probable that 6, 7 and 8 are going to add fuel to an already abating fire.  Even 1 and 2, who are usually the most belligerent, are part of a group that is being calmed down after a minor disturbance.  What follows is a quick disturbance and then things soon calm down. 

That is principally the difference between the 100 strong riots in North London and the 4 strong minor disturbances in East London – it is down to fractions – those tiny details that, like the butterfly effect, have bigger consequences henceforward. I can’t prove that what I’ve said is right – but I’m pretty sure it is right – and this ought to give us some pause for thought when we consider the many extreme groups we feel so comfortable denigrating. Nazi groups, extreme Stalinists, inner city rioters, religious groups that put people to death as witches or heretics, the Taliban, and other groups of this kind are all sensitive to the same procedural rules as stated above – the occurrence of n increases the probability of n1, which increases the probability of n2, and n3, and n4, and so on. At the very least, this ought to make us think a bit more about how quick we are to reproach large groups, when perhaps we should be considering what it is like for an individual to be caught up somewhere in that spectrum of probability.