There’s a surprisingly strong link between climate change and violence

Earlier this year, when a study came out suggesting global warming will increase the rates of violent crimes in the United States -- producing "an additional 22,ooo murders, 180,000 cases of rape," and many other crime increases by the year 2099 -- it drew widespread criticism. "This ... is what people who are losing the argument look like," noted the conservative publication National Review.

One study may seem easy to dismiss. But the combined results of 56 of them? Not so much.

In a new working paper released by the National Bureau of Economic Research, Stanford researcher Marshall Burke and two colleagues present a meta-analysis -- an evaluation and statistical synthesis of a large body of studies -- of the existing research examining the relationship between climate change and violence and conflict. After throwing out a host of studies they consider flawed, Burke and his colleagues still had 56 left, many of them quite recent.

Wonkbook newsletter

Your daily policy cheat sheet from Wonkblog.

Climate variables considered in these papers included temperature increases as well as drought and rainfall changes. Conflict was analyzed in terms of clashes between individuals (like fistfights) and fights between groups (like wars). After taking it all in, the authors found compelling evidence of a link between changes in temperature and increases in conflict, noting that "deviations from moderate temperatures and precipitation patterns systematically increase the risk of conflict, often substantially, with average effects that are highly statistically significant." Bottom line: In an ever warming world, expect more wars, civil unrest, and strife, and also more violent crime in general.

More specifically, for a degree Celsius of temperature increase (1.8 degrees Fahrenheit), Burke says there could be a 20 percent increase in civil conflict in Africa. The impact of warming varies by region, however; some places are more sensitive to small heat increases than others. In the United States, the estimate would be lower: For 1 degree Celsius of warming, he'd expect about a 1 percent increase in interpersonal conflicts, a category that includes crimes like assault and robbery but also road rage and fights at baseball games.

It is important to emphasize the strength of this conclusion: It emerges from multiple studies, the vast majority of which point to the same takeaway conclusion. For example, across 19 studies, Burke and his colleagues found 24 separate estimates of the relationship between temperature changes and various kinds of conflict outcomes, and in every case, that relationship was positive. "The probability of getting 24 positive values if there was in fact no relationship between temperature and conflict ... is less than 1 in 100 million," Burke said in an e-mail. "It's like flipping a coin 24 times and getting heads each time."

All of this begs the question: Why do hotter temperatures produce more violence?


The relationship certainly sounds intuitive, and was speculated about long before it could be scientifically studied. Burke notes that in Shakespeare's Romeo and Juliet, Benvolio tells his friend Mercutio that they should get out of the heat, seeming to suggest that hot temperatures produce violent conflicts, which is not a good idea given the ongoing feud between Montagues and Capulets:

I pray thee, good Mercutio, let's retire:
The day is hot, the Capulets abroad,
And, if we meet, we shall not scape a brawl;
For now, these hot days, is the mad blood stirring.

Since the Bard's day, psychological research has cast more light on the matter. For instance, one of the studies cited in the new meta-analysis is a 2011 paper published in Psychological Science (discussed in more depth here) showing a relationship between hot temperature days and the number of retaliatory beanballs thrown by Major League Baseball pitchers.

The figure below shows their results in more detail:


Credit: Richard Larrick et al, 2011, Temper, temperature, and temptation: Heat-related retaliation in baseball. Psychological Science, 22, 423-428. Reprinted with permission.

The study's lead author Richard Larrick, a professor at Duke's Fuqua School of Business, explained in an e-mail the psychological research linking heat with shows of aggression.

"Researchers in social psychology have studied the relationship between temperature and aggression for many decades," Larrick said. That includes studies looking at links between a day's temperature and people engaging in real world behaviors ranging from honking horns to committing violent crimes. "Research in the laboratory," Larrick continues, "allowed for tightly controlled tests to show that changes in temperature directly lead to more aggression." Such research has shown, he notes, that "heat changes the way people feel and think, increasing anger and making thoughts of aggression increase."


It is important to underscore that the temperature-violence relationship is not deterministic. In their meta-analysis, Burke and his colleagues liken the situation to "the rise in car accident rates during rainy days" -- the rain ups the risk of accidents overall, but each accident is still contingent on the individual situation and choices (and mistakes) of the drivers involved.

Similarly, warmer temperatures seem to shift the overall background risk for violent conflict -- but whether someone commits a violent act remains dependent upon the specific circumstances and the individual.

So the relationship is statistical -- but that doesn't make it any less real. "We believe there is overwhelming evidence of a strong relationship between changes in temperature in particular, and various types of human conflicts," says Burke.








Story highlights

  • Study: By 2050, risk of intergroup conflict will rise 50%

  • Higher temperatures are associated with aggression

  • Economic and psychological factors may play into this pattern

It's a hot day; sweat soaks your forehead, and you just want to get back to your air-conditioned home, so of course you're going to be angry if the car in front of you doesn't move when the traffic light turns green. Honk! Honk!

When the temperature rises, so does aggression -- and that can lead to large-scale consequences, considering that climate change is turning up the heat over the entire planet.

A new study in the journal Science shows that shifts in climate historically have been associated with violent conflicts, among both individuals and groups, and that current warming patterns could significantly increase the abundance of human conflict by midcentury.

Researchers' meta-analysis of 60 studies suggest that, consistent with links between conflict and climate shifts in the past, the risk of intergroup conflict around much of the planet would be amplified by 50% in 2050.

"It does change how we think about the value of avoiding climate change," said Solomon Hsiang, lead study author and researcher at the University of California, Berkeley. "It makes us think that avoiding climate change is actually something we should be willing to invest more in."

Future populations will hopefully also be able to figure out ways to adapt to these climate events, such that the predicted increase in violence will become less likely. Large-scale changes in technologies or political institutions may also alter the risk of violence in the future, Hsiang said.

A brief history of climate change and violence

Research on the topic of climate change and violence has been exploding onto the scene in recent years, study authors noted, with 78% of the studies included in this analysis having been published since 2009. Hsiang and colleagues incorporated research on civilizations dating back as far as 10,000 B.C. and across all major world regions.

Studies both in laboratory settings and of "natural" human situations have found a connection between heat and violence (PDF), the study said. Higher temperatures have been linked to both innocuous hostile behaviors, such as horn-honking while driving, and more serious behaviors such as domestic violence within households, assault and rape. Police officers are also more likely to use force at higher temperatures, studies have found.

Conflict is also associated with extreme rainfall, particularly in societies dependent on agriculture. Higher rates of personal violence are found in low-income settings, where agriculture income suffers from extremely wet or dry conditions.




Is climate change causing disasters? 05:03




Obama plans 'assault' on climate change 02:06




Director James Cameron on climate change 04:20

The Mayan civilization appears to have collapsed during long periods of drought, Hsiang said. The same global climate event seems to have brought down the Tang dynasty in China -- in fact, according to Hsiang, most Chinese dynasties collapsed during dry spells.

The climate conditions that contributed to the decline of civilizations in both Central America and China look like an El Nino event, Hsiang said.

That directly relates to Hsiang's 2011 study in the


journal Nature 

suggesting that the El Nino Southern Oscillation played a role in 21% of civil conflicts from 1950 to 2004. Evidence points to poorer countries of the tropics, such as Sudan and Rwanda, as being most affected in this way, whereas Australia did not have El Nino-related conflicts.

That's not to say that El Nino, a large movement of water in the Pacific Ocean that occurs about every four to five years, can be said to cause any given civil conflict. In fact, it's impossible to prove that any particular war or act of violence is the direct result of climate change. There are certainly other factors that cause and insight violence of all kinds.

But Hsiang's new study reinforces the idea that dramatic climate shifts increase the odds of violent conflict, as if loading a deck of cards in favor of this gruesome outcome.

Similarly, no individual hurricane or hot day in our time is definitively "caused" by human-induced climate change, but it does make extreme events such as last year's Superstorm Sandy more likely, said J. Marshall Shepherd, president of the American Meteorological Society and professor at the University of Georgia.

How much more violence, and why?

When Hsiang and colleagues put together evidence from 190 researchers, they determined that intergroup conflict increases 14% and interpersonal violence increases 4% for each standard deviation in climate variables. For instance, in terms of temperature, one standard deviation is about halfway between what you would expect for an average year and a really hot year.

"If your local level of violence increased by 10 to 15%, you would probably notice that," Hsiang said.

Poorer populations have been shown in other studies to be more vulnerable to shifts in climate generally, consistent with findings that such shifts raise the risk of violence among them in a more substantial way. It is possible, however, that the definitions of conflict outcomes or measurement errors could have influenced these results, the study authors noted.

As discomforting the analysis may be, Thomas Homer-Dixon, political scientist at the Balsillie School of International Affairs in Waterloo, Ontario, sees Hsiang's new study as strong overall. The conclusions are dramatic, he said, but the methods to compare various studies are sound.

CO2 levels hit new peak at key observatory

What exactly explains these patterns concerning climate and violence? More research needs to be done on mechanisms, but scientists have some ideas based on available evidence.

Shifts in temperature and rainfall influence economic productivity and food prices, which may influence discontent and therefore riots. Climate changes can force the displacement of population, or urbanization, which may lead to clashes over resources.

There's also the mental and physical responses to rising temperatures, which need further scientific investigation. "For instance, climatic events may alter individuals' ability to reason and correctly interpret events, possibly leading to conflicts triggered by misunderstandings," the authors wrote.

Implications of climate change and violence

Research on this topic is making waves. A 2011 report, in which Shepherd participated, found that new national security challenges will arise for the U.S. Navy, Marine Corps and Coast Guard if even moderate trends in climate change continue.

"Military, security, and intelligence analysts are now engaged with the issue, and there's a great deal of thought now being given to the long-term security implications of climate change," Homer-Dixon wrote in an e-mail. "There's very little skepticism within these circles about the reality of climate change, nor about the potential risk climate change poses to national and international security."

The scientific community agrees that human activity -- namely burning coal, oil and natural gas -- has been driving a rapid rise of carbon dioxide levels in the atmosphere. This is analogous to turning the dial up on an electric blanket, says Jim Butler, a senior scientist at NOAA; it takes a little while to warm up, and even if we stop emitting carbon dioxide, temperatures would still rise for a decade or two.

Indirectly, analyses such as Hsiang's can influence discussions about climate policy in general.

"If the world that these studies project will arrive in mid-century looks pretty unpleasant, then one obvious response is to be more serious about reducing greenhouse gas emissions," Homer-Dixon said.

It remains to be seen whether this research will indeed get policy makers hot and bothered enough to enact change.

Follow Elizabeth Landau on Twitter and Google+