Randy Roth (Ohio State), the leading historian studying homicide rates, has a piece in Reviews in American History [available only to some readers logging on through their university libraries] that examines two items of academic folklore. In this post, I address the widespread myth that homicides were rare in the "Old West."
In recent years it has become fashionable for historians (such as Robert Dykstra and Michael Bellesiles) to claim that it was a myth that the Old West was particularly violent. Notheless, other historians, such as Clare McKanna and David Peterson Del Mar, have reported very high rates of homicide in the West in the late 19th century (compared to current rates in the US).
Who is right?
Roth carefully reviews the data and confirms the work of McKanna and Peterson Del Mar, showing it to be consistent with recent work by Kevin Mullen, John Boessenecker, and (the late, great) Eric Monkkonen, .
Because the counties in McKanna's study reflect the diversity of rural southern and central California as a whole, there is reason to believe that the homicide rate in the southern two-thirds of the state (excluding San Francisco) was between 66 and 80 per 100,000 adults per year—the 99% confidence interval for McKanna's seven counties combined. If we include San Francisco and Los Angeles counties, the interval for all of southern and central California was between 60 and 70 per 100,000 adults per year—seven times the homicide rate in the United States today (and 28.7 standard deviations away). An adult exposed to that rate for sixteen years stood a 1 in 96 chance of being murdered, and an adult exposed to that rate for 45 years would have stood a 1 in 34 chance of being murdered. We cannot make assumptions about the homicide rate in northern California, which has yet to be studied. But with McKanna's study alone, we have 29 percent of the population of southern and central California (38 percent outside San Francisco); and with the addition of Mullen's study of San Francisco and Monkkonen's of Los Angeles, we have 57 percent of the population. The claim that the area was not unusually homicidal is statistically and arithmetically impossible.
The data of Peterson Del Mar and McKanna show that there is no such thing as a "fallacy of small numbers." The laws of probability make it possible to predict the character of a large population from a sample of surprisingly modest size, as long as that sample is representative of the population as a whole. That is why national opinion polls of 1,500 or 3,000 potential voters can be so accurate, even for subgroups of the population. That is the genius of statistics.
How homicidal was the Old West? According to the best historical evidence today, the answer is: Extremely Homicidal. Thus, another bit of academic folklore bites the dust.