Why it's dangerous to be a witch in a recession.
Why did people murder suspected witches in Renaissance Europe? And why do they still do so today in sub-Saharan Africa? As someone whose main source of information about witch trials is Monty Python and the Holy Grail, I was fascinated to learn that witch-burning has its own grim economics.
Clearly, some of the fervor for murdering women—typically elderly widows—had cultural and religious origins. In the early medieval period, the Catholic Church dismissed the idea that witches had supernatural powers, and some church documents argued that it was heresy to believe in witchcraft. Without church support, it's easy to see why witch trials were not popular.
Yet when the trial and execution of suspected witches surged in the mid-16th century and throughout the 17th, it was a cross-cultural phenomenon. Trials took place in many countries and were conducted by both Protestants and Catholics, and in both secular and religious courts. Perhaps a million women were killed across Europe after being accused of witchcraft, and most of them died during this period. Why?
Historian Wolfgang Behringer has one possible explanation: Temperatures dropped sharply around the time that the trials gained in popularity. The "little ice age," in which average temperatures fell by about 1 degree Celsius, was enough to freeze the Thames River on many occasions.
Emily Oster, an economist at the University of Chicago, has tried to gather systematic data on the link between witch trials and the weather. The results look striking: Between 1520 and 1770, colder decades go hand-in-hand with more trials. The link may be simply that witches were often blamed for bad weather. Or there may be a less direct link: People tend to lash out in tough times. There is some evidence, for instance, that lynching was more common in the American South when land prices and cotton prices were depressed.
Such deaths are, sadly, not a historical footnote. In Meatu, Tanzania, half of all reported murders are "witch killings." Such murders have been documented elsewhere in Africa, Bolivia, and rural India. The difference between the historical executions and modern attacks are that a Tanzanian "witch" typically dies at the hands of her own family. The machete is the weapon of choice.
Edward Miguel, an economist at the University of California, Berkeley, and co-author of Economic Gangsters, a book about the economics of crime, corruption, and war, has studied the Tanzanian situation. He argues that there is a direct economic motive for the attacks. Tough times in a Tanzanian household may well result in starvation, and the elderly—especially women—are at risk of being sacrificed to free resources. As evidence, Miguel points out that victims of witch attacks in Meatu district—almost all old women—tend to be from the poorest households. The murders are much more common during years of drought or flood.
If the problem truly is an economic one, the solution might be, too. One possibility is to give the elderly generous pensions. Witch killings all but stopped in South Africa's North Province after such a pension scheme was introduced in the early 1990s. Unfortunately, such pensions are probably too expensive for Tanzania.
A grass-roots alternative has emerged in another Tanzanian district, Ulanga, where traditional healers "cure" elderly women of witchcraft by shaving their bodies and smearing their pates with "anti-witchcraft paste." Miguel does not think it's a coincidence that the healers also provide the women with food and shelter during famines, in expectation of payments from their families in better times. Spiritual ceremony meets social insurance: It's a solution, of sorts.
Tim Harford is a Financial Times columnist. His latest book, The Logic of Life, will be published in paperback on Feb. 10.