The Dangerous Power of Crowds (2011) | Hacker News Reader

The Dangerous Power of Crowds (2011)

newyorker.com

56 pointsdbreunig19 comments

In 1998, the Czech hockey team won the Olympic Games at Nagano and there was a huge crowd waiting for their return on the Old Town Square in Prague.

I went there too, having no experience with crowds of that size. Waves of movement started going through the crowd, sudden bursts of violent pressure that moved you 10 feet or so suddenly. I found it hard to stay upright. Of course, everyone was screaming, so no cries for help would be of any use.

I fled the situation as soon as there was an opening in the crowd. AFAIK no one was hurt that night, but I am still glad I didn't stay. The vicious power of the surges was overwhelming.

Amazing, I just read this article, then pulled up reuters and saw this... if it wasn't real news, I'd think it was targeted advertising.

https://www.reuters.com/world/us/least-8-dead-many-injured-t...

That news is more than likely why this article was posted in the first place.

Ah, you're probably right. I see now that the date on the NY piece is January 30, 2011.

Yikes.

> Scott stopped multiple times during his 75-minute performance when he spotted fans in distress near the stage, video posted online showed. He asked security to make sure they were safe. Emergency vehicles, lights and alarms flashing, cut through the audience several times.

> "We need somebody to help. Somebody passed out right here," Scott said, according to video of his set, which included a guest appearance by Canadian rapper Drake. "Security, somebody, jump in here real quick."

75 minutes? At what point do you say to yourself, "Geez, I don't normally have to stop performing and ask security to help passed out people, maybe I should stop"? I'd ask the same of the venue and promoter, too.

I mean, there were multiple ambulances in the middle of the crowd while he's performing. Geez.

If he simply stopped, it could create an even bigger problem.

For whatever reason I've found this "Human Crush" phenomena simultaneously interesting & terrifying ever since I first heard about it some years ago in regards to the Hillsborough disaster. People in a dense enough group cease to function as individual units... but instead as a large fluid mass incapable of immediate course correction.

It's stuck with me, and made me more wary and alert when in crowded scenarios to this day. I suppose there's a thought to be had that as urban centers become more populous and common we'll have to figure out better ways of preventing and policing stuff like this. (And like the tragic stuff happening at Astroworld last night.)

Be safe folks.

> Hillsborough disaster

The "We'll There's Your Problem" podcast had a interesting episode[1] about the Hillsborough disaster.

> a large fluid mass

Apparently mosh pits are an ideal gas. From [2]:

>> ... we examined the 2D speed distribution; previous observations of human pedestrian traffic and escape panic led us to expect a broad distribution not well described by simple analytic expressions. However, the measured speed distribution in mosh pits wads well fit by the equilibrium speed distribution of classical 2D gassed, otherwise known as the Maxwell-Boltzmann distribution.

[1] https://www.youtube.com/watch?v=j2BjTfwhbh8

[2] https://arxiv.org/abs/1302.1886#

Crowd management and continuous crowd monitoring is essential at any potentially high-density event.

Some cities have banned festival seating since it leads to potentially deadly crowd densities near the stage. This matches my own experience right before a popular act comes on and the crowd density immediately doubles or triples as people cram into all available space at the front.

A domino effect or pileup of just a few people can create 1000 pounds of force, making it impossible for someone to breathe. Now imagine shockwaves/quakes/pileups in a crowd of thousands of people.

Rigid metal or concrete barriers/fences that crush people rather than giving way, along with security staff who refuse to allow distressed attendees to leave, make things even more dangerous.

..and a reminder to add Elias Canetti's 1960 masterpiece "Crowds and Power" to your to-read list! He won the Nobel prize in 1981 largely because of this book.

You can read it online (as PDF too), at no charge.

Also Psychology of Crowds by Gustave Le Bon.

Ezra Kliens new book why we are divided asserts that rationality is also an emergent phenomena, individuals post hoc justify even smart ones. then I am reminded of the nietsche quote about groups being mad and indivuals rational. both positions have good supporting evidence and show we need more specificity, and need to talk about a distribution not just an average.

I suspect there are pronounced peaks accross different group sizes where rationality is more emergent, and of course it depends on lot of factors: subject matter, groups homogeneity, expertise, connectivity probably this is a game theory field of study.

I think luskin's citizens assembly research showed that a few hundred people from a culture, without any prior training, could with the help of domain experts make seemingly excellent policy decisions about difficult moral and technical issues.

Side note: "The Wisdom of Crowds" by James Surowiecki

https://www.amazon.com/Wisdom-Crowds-James-Surowiecki/dp/038...

Extraordinary Popular Delusions and the Madness of Crowds: https://web.archive.org/web/20040623130633/http://www.litrix...

"... Employees asked the police for help. According to a court filing, the police responded that dealing with this crowd was “not in their job description,” and they left. ..."

At least after the 1989 Hillsborough Disaster there was an inquiry into that police.

Article should have been titled The Dangers of Government Inaction and Lack of Accountability

I will never attend one of those jam-packed music festivals/concerts. One emergency (even a small one), and you are likely to be under a pile of bodies. No thanks.

I immediately thought about the Wall of Death from mosh pits when I was a kid.

very old --

some previous discussion: https://news.ycombinator.com/item?id=10488996

2011

Reply on news.ycombinator.com