Pollution causes people to commit violent crimes --homicide, aggravated
assault, sexual assault and robbery --according to new research by
Roger D. Masters and co-workers at Dartmouth College. Sociologists
have known for a long time that violent crimes occur more in some
places than in others. Some U.S. counties have only 100 violent crimes
per 100,000 people per year; other counties have rates of violent crime
that are 30 times as high. The question is why some places have high
crime rates and others don't. Masters says pollution is part of the
Masters has developed what he calls the neurotoxicity hypothesis of
violent crime. According to this hypothesis, toxic pollutants --
specifically the toxic metals lead and manganese --cause learning
disabilities, an increase in aggressive behavior, and --most
importantly --loss of control over impulsive behavior. These traits
combine with poverty, social stress, alcohol and drug abuse, individual
character, and other social factors to produce individuals who commit
Masters argues that, to be taken seriously, such a hypothesis must pass
five tests. He then demonstrates how the neurotoxicity hypothesis meets
all five, as follows:
1) It must be shown that individuals who engage in criminal behavior
are more likely to have absorbed toxic chemicals than a comparable
control population. Masters cites studies showing that low-level
poisoning by lead, and by manganese, is associated with learning
disabilities and attention deficit disorder, which are themselves
associated with deviant behavior. (We reviewed some of this evidence
for lead in REHW #529). Masters cites seven other studies showing that
violent prisoners have significantly elevated levels of lead,
manganese, cadmium, mercury or other toxic metals, compared to
prisoners who are not violent.
2) If it is valid, the neurotoxicity hypothesis must be able to predict
future violent behavior of young people exposed to toxins. Masters
cites two prospective studies (and suggests we need more) showing that
lead uptake at age 7 is associated with juvenile delinquency and/or
increased aggression in teenage and early adult years. (See also REHW
#529.) The largest study, of 1000 black children in Philadelphia,
showed that both lead levels, and anemia, were predictors of the number
of juvenile offenses, the seriousness of juvenile offenses, and the
number of adult offenses, for males.
3) Is there a biological basis for believing that lead, manganese and
other toxic metals could cause a person to lose control over impulsive
and aggressive behavior? Here Masters cites a wealth of studies showing
how lead and manganese cause changes in the development of the brain,
and in the functioning of neurotransmitters in the brain.
Different pollutants harm the brain differently. Lead in the brain
damages glia, a kind of cell associated with inhibition and
detoxification. Manganese has the effect of lowering levels of
serotonin and dopamine, which are neurotransmitters associated with
impulse control and planning. Masters notes that low levels of
serotonin in the brain are known to cause mood disturbances, poor
impulse control, and increases in aggressive behavior --effects that
are increasingly treated with Prozac.
Masters emphasizes that children who are raised from birth on infant
formula and who are not breast fed will absorb five times as much
manganese as breast-fed infants. Calcium deficiency increases the
absorption of manganese. A combination of manganese toxicity and
calcium deficiency adds up to "reverse Prozac," Masters says.
Masters says toxic metals affect individuals in complex ways. For
example, because lead diminishes a person's normal ability to detoxify
poisons, lead may heighten the effects of alcohol and drugs.
4) For the neurotoxicity hypothesis to hold up, individuals must
receive doses of toxic metals sufficient to be associated with violent
behavior. Masters argues that, despite recent significant decreases in
lead in the environment (because leaded gasoline and lead paint have
been banned in the U.S.), in neighborhoods where automobile traffic has
historically been high, and in towns where industries have released
large quantities of toxic metals for years, many local soils still
contain toxic quantities of lead, cadmium, and manganese sufficient to
poison children who play in the dirt. He also argues that aging water
delivery systems very likely contribute lead and manganese because lead
pipes and even iron pipes contain these toxins.
Masters argues that (a) children absorb up to 50% of the lead they
ingest (compared to 8% for adults); (b) even low exposures in the womb
and in early childhood can have permanent effects on intelligence and
behavior; (c) current lead levels are known to have direct effects on
neurotransmitters that are known to affect cognition and to influence
impulse control; and (d) the highest levels of lead uptake are reported
in precisely the demographic groups most likely to commit violent
crimes (inner city minority youths).
Masters emphasizes the importance of studies showing a synergistic
effect (multiplier effect) between toxic metals and poor diet. For
example, it has been thoroughly documented that uptake of lead is
greatly increased among individuals who have a diet low in calcium,
zinc, and essential vitamins. Similarly, as noted above, calcium
deficiency greatly increases one's absorption of manganese. Thus,
Masters argues, amounts of lead and manganese that wouldn't harm a
well-nourished individual may poison undernourished children.
Masters cites federal studies of nutrition to make the point that black
teenage males consume, on average, only about 65% as much calcium as
whites. The calcium needs of pregnant or breast-feeding women are
higher than average, which creates a particular problem for minority
women. And non-Hispanic black women get only 467 milligrams of calcium
per day (mg/d), compared to 642 mg/d for white women, government
Because of increased manganese absorption by babies who drink infant
formula and who are not breast fed, Masters considers infant formula
toxic. He emphasizes that poor mothers tend not to breast-feed their
babies. By 1986-87, 73 percent of infants born to mothers with more
than 12 years of education were breastfed compared with 49 percent of
infants born to mothers with 12 years of education, and 31 percent of
mothers with less than 12 years of education. Furthermore, white
infants are more than three times as likely to be breast fed as black
infants. "The effects of manganese toxicity associated with infant
formula are thus greatest for the poor, for ethnic minorities, and for
those with little education," Masters says.
Masters cites studies showing that alcohol increases the uptake of
toxic metals, at least in laboratory animals, and probably has a
similar effect on humans.
5) If the neurotoxicity hypothesis is valid, then measures of
environmental pollution should correlate with higher rates of violent
To test his hypothesis, Masters acquired data from the FBI for violent
crimes in all counties of the U.S. He correlated this with data on
industrial releases of lead and manganese into the environment of each
county, using data from the U.S. Environmental Protection Agency's TRI
[toxic release inventory] database. He also examined other variables
for each county --population size, population density, housing built
before 1950, number of police officers per person, number of school
dropouts and high school dropouts, educational achievement,
unemployment rate, race and ethnicity [white, black, hispanic], persons
below the poverty level, number of people on welfare, infant deaths per
1000 live births, all alcohol-related causes of death, and all causes
of death with explicit mention of alcohol.
The EPA's recorded releases of toxic metals are not predicted by these
demographic or socio-economic variables. In fact, less that 5% of the
variance of reported releases of lead is accounted for by 19 socio-
economic factors (many of them listed in the previous paragraph).
Masters split all U.S. counties into six groups --those with and
without industrial lead releases; those with and without industrial
manganese releases; and those with higher-than-average or lower-than-
average rates of alcohol-related deaths. After controlling for all the
conventional measures of social deterioration (poverty, school
dropouts, etc.), Masters found that counties having all three measures
of neurotoxicity --lead, manganese, and high alcohol --have rates of
violent crime three times the national average.
In other words, environmental pollution and alcohol have a strong
effect on violent crimes, completely independent of any of the
"traditional" predictors of violent crime (poverty, poor education,
As Masters says, neurotoxicity is only one of many factors contributing
to violence, but he believes it may be especially important in
explaining why violent crime rates differ so widely between geographic
areas and by ethnic group. Masters says that traditional sociological
approaches to crime cannot explain why the availability of handguns or
drugs triggers violent behavior in only a small proportion of the
population, a proportion that varies greatly from place to place. Part
of the explanation may be the way the physical environment affects
brain chemistry and behavior, Masters says.
"The presence of pollution is as big a factor as poverty," Masters said
recently in an interview in NEW SCIENTIST magazine. "It's the
breakdown of the inhibition mechanism that's the key to violent
behavior," he says. When our brain chemistry is altered by exposure to
toxins, we lose the natural restraint that holds our violent tendencies
in check, Masters believes.
Former U.S. Surgeon General C. Everett Koop has said, "Regarding
violence in our society as purely a sociologic matter, or one of law
enforcement, has led to an unmitigated failure. It is time to test
further whether violence can be amenable to medical/public health
For decades, researchers have focused on the human health consequences
of toxic metals --mainly asking, do they cause cancer? This new
research seems to be telling us that we should also be looking at the
way these pollutants are affecting human BEHAVIOR.
--Peter Montague (National Writers Union, UAW Local 1981/AFL-CIO)
 Roger D. Masters, Brian Hone, and Anil Doshi, "Environmental
Pollution, Neurotoxicity, and Criminal Violence," in J. Rose, editor,
ENVIRONMENTAL TOXICOLOGY (In press. [London and New York: Gordon and
Breach Publishers, 1997]).
The particular crimes are defined as follows:
Murder and nonnegligent manslaughter: the willful (nonnegligent)
killing of one human being by another.
Forcible rape: carnal knowledge of a female forcibly and against her
will. Assaults or attempts to commit rape by force are also included;
however, statutory rape (without force) and other sex offenses are
Robbery: taking or attempting to take anything of value from the care,
custody, or control of a person or persons by force or by threat of
force or violence and/or by putting the victim in fear/
Aggravated assault: unlawful attack by one person upon another for the
purpose of inflicting severe or aggravated bodily injury.
 Alison Motluck, "Pollution may lead to a life of crime," NEW
SCIENTIST Vol 154, No. 2084 (May 31, 1997), pg. 4.
 C.E. Koop and G.D. Lundberg, "Violence in America: A Public Health
Emergency," JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION," Vol. 267, No.
22 (1992), pgs. 3075-3076.
Descriptor terms: crime; lead; manganese; homicide; aggravated assault;
sexual assault; robbery; toxic heavy metals; human behavior; roger d.
masters; dartmouth college; cadmium; mercury; violence; aggression;
glia; serotonin; dopamine; prozac; infant formula; breast feeding;
nutrition; calcium deficiency; poverty;