Psychological Causes of Violence in Sports Riots
Recently, the Los Angeles Lakers won game 7 against the Boston Celtics and there were riots in the streets of los angeles. Below is a video of some of the scene.
This scene is not unique to Los Angeles. In fact, riots appear to occur with regularity when sports teams win. There were riots in Boston when the Celtics won in 2008 and riots in Los Angeles when the Lakers won in 2009 too. This seems to counter the common sense idea that people should be happy when they win, such that they are more generous with others. Happy people tend to be generous people (though the causal relationship might run in the reverse direction), not rioters. Shouldn't the people in the losing cities be the ones who rampage out of frustration? Yet there is an astonishing correlation between rioting and winning in the Lakers-Celtics series and in sports rioting more generally.
A colleague of mine dug up this study (Bernhardt et al, 1998) to explain it to me and I think it's worth sharing. It's been replicated by others as well. Unfortunately, the article itself is protected by the wall of the academic journal system, but the basic pattern of results is illustrated below.
Fans of Winners Experience Testosterone Increases
Basically, fans of the winning team gain testosterone, which has been linked to aggressive behavior. Fans of losing teams lose testosterone, which makes sense from an evolutionary perspective. Winners are encouraged to compete more...losers cut their losses.
Does this same effect extend to politics? My gut tells me no, as politics is less primal and the results develop over months, not hours. In fact, most of the time, we know who will win before an election and so what the winners feel is relief (an idea somewhat validated by this study). This article (fully visible by the public, since it was commendably published in an open access journal) illustrates that for some individuals, there was indeed no testosterone increase among winners, but the same decrease among losers, in the 2008 presidential election.
Another interesting resource, for those interested in the consilience of multiple views on the subject, is Bill Buford's book, Among the Thugs, where he lives among chronic sports rioters, fans of English football. His explanation dovetails nicely with Bernhardt et al's research (quote thanks to this source):
I had not expected the violence to be so pleasureable....This is, if you like, the answer to the hundred-dollar question: why do young males riot every Saturday? They do it for the same reason that another generation drank too much, or smoked dope, or took hallucinogenic drugs, or behaved badly or rebelliously. Violence is their antisocial kick, their mind-altering experience, an adrenaline-induced euphoria that might be all the more powerful because it is generated by the body itself, with, I was convinced, many of the same addictive qualities that characterize synthetically produced drugs.
For more information, here is another parallel view and a link to a more general overview of the causes of violence in sports riots (unfortunately, again, full text inaccessible without a university login...hrm!...I hope someday to be in a position to publish only in open access journals).
- Ravi Iyer
July 1st, 2010 - 09:28
The testosterone thing doesn’t surprise me in the least bit. But is it even the least bit scientific to attribute the riots to testosterone using a mere correlation? In fact, I would think that such an attribution is pretty clearly disconfirmed. After all, if these increases in testosterone are a product of victory and the riots are a product of these increases then there should be riots everywhere there are victories. But there aren’t.
A -> B -> C. -C. Therefore, not -(A -> B-> C). Ehrm, I hope that is valid.
There should also be the most riots in the areas where there are the most victories. Are there? It seems pretty clear that these riots are cultural – it’s a vastly simpler explanation, and while it isn’t one that is scientific it is at worst equally bad as already falsified scientific claims.
July 1st, 2010 - 09:29
Oops.
By “Therefore, not -(A -> B -> C)” I meant “Therefore, -(A -> B -> C)”.
July 1st, 2010 - 10:29
I didn’t mean to imply that testosterone always cause riots. Of course there are other factors involved. Merely that it helps explain why rioting occurs more often in cities that win as opposed to cities that lose.