SOCIETY:

Do warmer average temperatures stimulate more crime?

From spitfires to cold shoulders, the link between temperature and human emotion is deeply embedded in our vocabulary. Arguments heat up, tempers boil over and hotheads must be calmed by cooler voices. Down to their etymological roots, the words "temperature" and "temperament" share a common ancestor: the Latin temperare, meaning "to mix in correct proportion."

But does heat really correlate with anger and aggression? On a physiological level, the answer seems to be yes. Human beings respond to hot weather in much the way language like "hothead" and "spitfire" suggests: Our heart rates, blood pressure and testosterone levels all climb with the mercury, making us more aggressive in stressful or threatening situations.

But a more important question for a world on the cusp of global warming -- a world where heat waves and absent winters have become increasingly common phenomena -- is how such a correlation might manifest for society as a whole.

Part of that manifestation might already be visible, in the form of more frequent, violent crime. A significant body of evidence exists linking rising temperatures to social disruption in the form of higher crime rates, wrote John Simister, a senior lecturer in economics at Manchester Metropolitan University, in a recent email.

"The pattern of more crime at very high temperatures [has] been known by scientists for over a hundred years," he wrote.

Simister, who has written extensively about the correlation between crime and climate, said the main question now is not whether crime rises with temperature, but why. He wonders whether temperature itself actually triggers violent crime, or whether a third, related variable might be tipping the scales.

Motor City's warm, bloody winter

That question was on the minds of Detroit law enforcement this winter, after a record warm January and February coincided with one of the bloodiest seasons in city history. Motor City saw more than 70 murders between Jan. 1 and March 20, a period also marked by uncharacteristic heat.

"I wouldn't blame it all on the weather, but this unseasonably warm weather hasn't helped at all," Detroit Police Chief Ralph Godbee told The Detroit News. "Usually during the winter, we get a bit of a break because we have a stretch of below-zero weather. ... But this year, it's almost like we went from fall straight to spring."

Most law enforcement officials take it as a given that crime rates peak during hotter summer months, and more nuanced studies have found that rates tend to rise during heat waves, regardless of the season. Other research finds that temperature affects the rates of violent crimes, such as assault, more than those of nonviolent crimes like theft, a further indication that heat-fueled aggression may be driving the trend.

But many people who work in the criminal justice field resist such direct assumptions about cause and effect. For Roger Humber, director of the Criminal Justice Department at South University in Montgomery, Ala., the link between temperature and crime comes down to a matter of opportunity.

"It's a simple but crucial fact that people tend to get out and about more when it's pleasant out," he said. "During the winter months, people can't get out. There aren't as many opportunities to interact."

Warm weather lets people mix socially, he said. And it is only a matter of probability that sometimes that mixture may prove volatile.

"More opportunity for violent crime means more violent crime," he said.

A curve or a straight line?

That theory -- that crime and temperature might be related primarily through the third variable of social interaction -- is gaining traction among many of the researchers who study crime and temperature. It is also a fairly easy hypothesis to test, since people are likely to stay inside at a certain heat threshold, decreasing the chances of interaction.

"We call this the curvilinear hypothesis," Simister wrote. "Crime rates rise as temperatures go up, but at extremely high temperatures, crime rates fall again."

Still, "It's hard to be sure," wrote Simister. "The best data on crime rates and temperatures is from the USA, but few parts of the USA get hot enough to test what happens to humans above (for example) 37 degrees Centigrade (98.6 degrees Fahrenheit), i.e. approximately blood temperature."

Yet two American psychologists, Ellen Cohn and James Rotton, both of Florida State University, found their tipping point at a much lower temperature threshold. By analyzing violent crime rates over a two-year period, classifying assaults according to time, day of the week, month and temperature, they found that violence rose as the temperature rose to about 80 degrees Fahrenheit -- and then declined.

Those results seem to suggest that, while hot temperatures may make individuals more irritable -- a universal truism -- it is likely that most people will keep that bad humor at home.