Last Friday’s murder of 20 young children and 6 adults at Sandy Hook Elementary School in Newtown, Connecticut, was extraordinarily disturbing.
Such things should not happen.
Moreover, they need not happen. Societies have existed in which they did not. For example, the United States between the early 1900s and 1960.
Here is data of murderous rampages – mass homicides striking indiscriminantly at strangers – compiled by Peter Turchin (source):
Murderous rampages as compiled by Peter Turchin
Prior to 1960 there were few murderous rampages in the United States; in fact, between 1910 and 1940 Turchin’s data set records not a single rampage with 4 fatalities or more. But since 1960 the number of murderous rampages per capita has grown more than ten-fold.
It behooves us to consider what may have changed in the last fifty years to make rampages more frequent.
Rampages May Indicate a Ubiquitous Health Problem
Murderous rampages constitute only a small number of events – a few rampages per year today, up from a few per decade in the 1950s – and therefore occupy the extreme of the mentally ill “tail” of the population distribution.
However, as students of statistics know, slight changes in standard deviation can have huge effects in the tail of a distribution. If mental health is only slightly impaired in most people but becomes more variable in the population, the number of mentally ill killers may increase dramatically.
If the whole spectrum of violence-prone mental states, from irritation and anger on the mild end to murderous rampages on the extreme tail, has become more prevalent, then we should see an increase in all measures of violence.
There is evidence that this has occurred. Here is a long-term time series from Manuel Eisner (from his “Long-Term Historical Trends in Violent Crime”) of homicide rates in Europe:
In most countries for which we have data, homicide rates bottomed around 1950 and have doubled or tripled since.
The millennium-long decrease in homicide rates up to 1950 is thought to be due to evolutionary selection, as cooperative people experienced greater reproductive success. But the increase in homicide rates since 1950 is not likely to have genetic roots. It is more than likely due to an environmental cause, such as diet or lifestyle.
Mental Illness May Reflect Impaired General Health
The brain is part of the body and generally the brain’s health cannot be impaired until bodily health has been compromised. So whatever environmental changes are increasing the incidence of violence should also be observably impairing bodily health.
Indeed, we know that some indexes of bodily health have worsened over this same time frame. Obesity and diabetes for instance; here is an image from our most recent post (“The Rise—and Fall?—of American Health”):
The obesity and diabetes epidemics
In that post
we argued that the healthfulness of the American environment
improved dramatically from 1860 to 1950, but that there has been little to no improvement in the healthfulness of the environment since 1950, and that there is some evidence for a recent deterioration in the healthfulness of our environment.
Omega-6 Fats as a Dietary Cause of Violence
One major dietary change that may have contributed to rising rates of violence has been the shift toward omega-6 rich seed oils, such as soybean oil and corn oil, in foodstuffs.
Joseph Hibbeln and collaborators had the idea of comparing omega-6 consumption to rates of violence. They found, across countries and over a period of decades, that omega-6 consumption was correlated to homicide rates:
Dietary omega-6 intake and homicide rates
American consumption of omega-6 fats has doubled per capita since 1960, and tripled since 1920.
Plausible mechanisms, beyond the scope of this post, connect high omega-6 fat intake to impaired mood and to an increased proclivity to violence. Even more plausible mechanisms, discussed in our book, connect omega-6 fats to the obesity and diabetes epidemics.
If omega-6 fat consumption does indeed promote violence, then we would expect to find that killers eat unusually high amounts of omega-6 fat. It’s likely that Adam Lanza, the Newtown killer, fits that description. According to Salon, Lanza was a vegan (because he “didn’t want to hurt animals”). Vegan diets, insofar as they utilize modern industrial foods, tend to be rich in the vegetable seed oils which are the primary source of omega-6 fats.
Other dietary and environmental trends, too, are likely to have increased rates of violence. To name three:
- Water treatment has decreased dietary intake of lithium below our evolutionary norm; and there is a notable association between low intakes of dietary lithium and high rates of suicide, homicide, and crime. (Schrauzer GN. Lithium: occurrence, dietary intakes, nutritional essentiality. J Am Coll Nutr. 2002 Feb;21(1):14-21. http://pmid.us/11838882.)
- Whatever factors are increasing rates of diabetes will also increase incidence of hypoglycemia, which has been linked to aggression. (Benton D. Hypoglycemia and aggression: a review. Int J Neurosci. 1988 Aug;41(3-4):163-8. http://pmid.us/3053477. Roy A et al. Monoamines, glucose metabolism, aggression towards self and others. Int J Neurosci. 1988 Aug;41(3-4):261-4. http://pmid.us/2460415.)
- Some psychiatric drugs have been linked to violence, and mass killers have often been reported to be taking precisely those drugs. (Moore TJ, Glenmullen J, Furberg CD (2010) Prescription Drugs Associated with Reports of Violence Towards Others. PLoS ONE 5(12): e15337. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0015337.)
As Samuel Johnson said, we live in a world bursting with sin and sorrow. But we should not willfully add to that sorrow. Perhaps it is time to consider returning to a diet and lifestyle that better accords with our evolutionary heritage – with the way we were meant to live. It just might save the lives of children, and improve the health of millions.