March 26, 2017

Debunking a Denier: Part 2 (Race and Crime)

This is the second in a three part response to a blog on race realism at Debunking Denialism. You can see the first part here.

The Basic Facts

Another query Karlsson response to is this:

“Maybe it’s no big deal to you, but if you visit America, where blacks, 13% of the population, commit 53% of all crimes, than maybe you will start to take racialism more seriously.”

Karlsson goes on to give various theories about Black crime rates, but before we look at that let’s get the facts straight. Black people in America do not commit 53% of the crime. They are over-represented in crime, especially in violent crime, but not to the extent that this fellow imagines:

Black Arrest Rates 1995-2014White Arrest Rates 1995-2014

Asian Arrest Rates 1995-2014

As can be seen, Asians are especially under-represented in crime. We see the same thing when we compare African, European, and East Asian nations to each other:

Crime and GDP by race

Rushton and Whitney (2002)

Racial differences within nations aren’t just found in America, either. Ellis, Beaver, and Wright (2009) reviewed 113 studies  conducted in Europe, Norther America, the Middle East, and Asia, which looked at whether or not Blacks commit more crime than Whites within various countries. All 113 found they did. Similarly, all 17 studies looking at crime differences between East Asians and Whites found that East Asians commit less crime than Whites.

Race and Cime around the world part 1Race and crime around the world part 2Race and crime around the world part 3

(Recidivism is not a crime rate and I am not counting the self report based studies.)

The relationship between race and crime at the regional level is also well established. Pratt and Cullen (2005) meta-analyzed 162 studies which looked at whether regions with a greater proportion of Black people had higher crime rates. Not only did they find that this was true, out of the 17 variables they were able to find at least 15 studies on “percent black” was the one most consistently associated with crime.

There is significant evidence that Black’s having trouble following the rules starts early in life. For instance:

  • A 2014 report issued by the department of education found that black preschoolers have an above average rate of suspension. This was especially pronounced among repeat offenders: though Blacks make up about 18% of preschools they make up almost 50% of preschoolers that have been suspended more than once (The Associated Press 2014).
  • Vega 2014 reported on a document from the US Department of Education which showed that in 2011 and 2012 Black females accounted for 12% of elementary school suspensions whereas white girls only accounted for 2% of school suspensions. (Most people suspended were boys.)
  • Lewin 2012 reported on a Department of Education report which analysed data from over 72,000 schools and found that Blacks made up 18% of the students population, but 35% of those to who had been suspended once, 45% of those who had been suspended more than once, and 39% of those who were expelled.
  • Skiba et al 2002 found that Black middle schoolers were more likely than White middle schoolers to be suspended even after controlling for differences in socio-economic status.

And finally, it is worth noting that Blacks have been over-represented among criminals for as long as we have been keeping records.

Black crime 1926-2013

(The data came from the DOJ’s Prisoner’s reports and this census document. Clearly, various years are missing, but there is no reason to think that the inclusion of these years would change the over-all picture)


So, now that we’ve seen that Blacks commit more crime than Whites here, and around the world, and have for a century, and begin to as children, the obvious question is “why?”. Well, Karlsson has an answer:

“The general response is that those figures do not control for confounders, such as income, unemployment, broken families etc. that we know from other studies have stronger effects… There are many articles on Debunking Denialism that goes into additional detail, such as White Genocide, Eurabia and Other White Supremacist Nonsense, with references to the scientific literature.”

I will look at said “scientific literature” in a moment. But first, allow me to make the counter case:

Land, McCall, and Cohen (1990) collected data on the homicide rates of cities, standard metropolitan statistical areas (SMSAs), and states for the years 1960, 1970, and 1980. In each year they included all 50 states and every city and SMSA included in the census. They then looked at how well the following 11 variables predicted crime variation between these areas: population size, population density, percent black, percentage aged between 15 and 29, percent divorced, percent of kids without two parents, median family income, the poverty rate, income inequality, the unemployment rate, and whether or not the city/SMSA/State was in the south. All of these variables were entered into a single regression model, meaning that the estimated effect size for each variable held all other 10 variables constant. This analysis thus produced 9 total models explaining crime variation in cities, SMSAs, and states, across 3 decades. Across these 9 models, race was a better predictor of homicide than unemployment, poverty, and median income, in 7, or 78%, cases, and a better predictor than income inequality in 8, or 89%, cases. Thus, over 3 decades of very large data sets, race was pretty consistently a better predictor of homicide rates than economic variables were. And in all cases, % black continued to predict crime rates after all these other variables were controlled for.

Another relevant analysis was carried out by owner Ron Unz. Unz (2013) looked at how well median income, population density, poverty, and % black, correlated with the crime rates of large American cities between 2006 and 2011. He found that the size of the black population was a substantially better predictor than any of the other variables tested.

Unz 1.PNG

Similarly, the New Century Foundation’s report “The Color of Crime” analyzed the violent crime rates of the 50 U.S. states and D.C for the year 2005. The analysis found that state violent crime rates correlated at .81 with the percentage of the population that was Black or Hispanic, .37 with the state’s percentage of high-school drop outs, .36 with the states poverty rate, and .35 with the state’s unemployment rate.

Amren 1Amren 2Amren 3Amren 4

Templer and Rushton (2011) significantly replicated the New Century Foundation’s results. They analyzed crime variation across the 50 U.S. states and found that the percent of the population that was black was a stronger correlate than average income for murder rates (.84 v -.40), robbery rates (.77 v .06) and assault rates (.54 vs -.23) The paper did find that income was a stronger predictor than black population size for rape rates (-.16 v -.22), but neither of these correlates were statistically significant or large.

Kposowa, Breault, and Harrison (1995) analyzed crime variation across 2,078 U.S counties and found that the percent of the population that is black was a stronger explanatory variable than poverty, income inequality, and unemployment, for explaining variation in both property and violent crime. Moreover, % black continued to predict crime after all these variables were controlled for.

Next, we have Rushton and Templer (2009) which looked at national variation in crime. They found that skin color, a proxy for race, was more strongly correlated than national income with homicide (.25 vs .17), rape (.24 vs .10), and serious assault (.20 vs .09).

Finally, we have Zaw and Darity (2016), which compared the likelihood of Blacks, Hispanics, and Whites becoming incarcerated at some point between 1985 and 2012 given what their net worth was in 1985.

Incarceration by net worth decile.PNG

Chart from Ehrenfreund (2016) data from Zaw and Darity (2016)

As can be seen, the results did not support the poverty explanation.

Tim Wise’s Lies

So, now let’s look at the  “scientific literature” that Karlsson sites to support his narrative:

“Nazis Can’t Do Math: Reflections on Racism, Crime and the Illiteracy of Right-Wing Statistical Analysis”

This is not scientific literature. It is an hysterical article written by professional anti-racist  activist Tim Wise. But whatever, let’s see what Mr. Wise has to say.

“According to the research by actual criminologists (which is to say, not by racist internet trolls), socioeconomic variables explain the difference between white and black violence rates, and where economic conditions are comparable between whites and blacks, there are no significant racial crime differences. In fact, the correlation between economic variables and crime are remarkably consistent from one society to the next. Evidence gathered from more than thirty countries has found that race and ethnicity have far less to do with crime than these environmental factors.”

Wise’s first citation is an analysis of census tracts within a single city, so we shouldn’t put much weight behind it. That being said, Wise lied about its results:

Wise 2

Wise 1

Krivo and Peterson (1996)

As can be seen, this citation actually contradicts Wise’s narrative on violent crime, though it does lend some support to this explanation for property crime. (Which, as we have already seen, Blacks are far less over-represented in to begin with.)

His second link is broke, so we’ll never be able to see how Wise lied about that citation.

He goes on:

A 1990 meta-analysis of twenty-one different studies on homicide, covering thirty years of research found much the same thing: among all the factors positively correlated with higher homicide rates, two of the most significant were unemployment rates and community resource deprivation.”

Amusingly, this is Land, McCall, and Cohen (1990), which I have already cited as showing that race is a stronger predictor of crime than economic and sociological variables and that it continues to predict crime after these things are controlled for. Wise’s lie is one of omission; unemployment and deprivation were predictors of crime looked at in this paper too, he just neglected to tell his audience that race was found to be an even better one. Moving on…

A comprehensive analysis of homicide and robbery data, which looked at the importance of such things as race, poverty, family disruption and unemployment in determining crime rates in these categories, found that black male joblessness explained black family disruption, which in turn was highly related to black murder and robbery rates, particularly for youth.”

This analysis in turn tells us:

“Sampson (1987) disaggregated the 1980 rates of homicide and robbery by race, poverty, family disruption, joblessness, and other factors. The analysis showed that black male joblessness predicted variation in rates of black family disruption, which was significantly related to rates of black murder and robbery, particularly by juveniles.”

So now that we’ve (I’ve) followed Karlsson’s link to his other blog, and then to the Tim Wise article, and then to this report, let’s go find Sampson (1987)!

Sampson’s paper shows that the more male’s are jobless the more likely it is that the family will fall apart, and this in turn predicts a higher crime rate. This is true for both Blacks and Whites and in no way implies that Black crime rates are simply caused by differences in male employment rates. In fact, as already mentioned, % Black continues to predict crime in an area after controlling for unemployment (Land, McCall, and Cohen, 1990; Kposowa, Breault, and Harrison 1995)..

Wise also talks about how Blacks are poor, which is true, but irrelevant, and also cites a book I can’t read, and an article which he names but does not link to and which I can’t find anywhere. Wise is well known for lying about sources, as we’ve already seen, so I assume that these citations are bull crap too.

In totality then, the evidence clearly shows that these simple economic and sociological narratives cannot account for racial crime differences. Wise’s citations are misleading and weak, mine are of better quality and quantity, and they actually say what I claim they say.

Moreover, even if Blacks and Whites who were equally poor were equally criminal this would not ultimately tell us why Blacks are more criminal than Whites. After all, if this difference in poverty is, itself, caused by genetic differences then racial crime gaps would still be caused by genes. Further still, there is always the possibility that the same personality traits (low IQ, low self control, aggression, ect.) cause people to be both poor and criminal.

Eighteen Guesses

Later on in his article, Karlsson makes the following statement:

“Criminology have isolated several important factors that are strongly related to crime in the sense that it increases the risk of committing crime. In general, it is a complex interaction between biology, psychology and environment, just like any other behavior. The classic textbook by Bernard, Snipes and Gerould (2010) lists 18 major factors that appear especially important, but there are probably many, many more.”

He then lists the following 18 factors which supposedly influence crime:

“1. A history of early childhood problem behaviors and of being subjected to poor parental child-rearing techniques, such as harsh and inconsistent discipline; school failure and the failure to learn higher cognitive skills such as moral reasoning, empathy, and problem solving.

2. Certain neurotransmitter imbalances such as low serotonin, certain hormone imbalances such as high testosterone, central nervous system deficiencies such as frontal or temporal lobe dysfunction, and autonomic nervous system variations such as unusual reactions to anxiety.

3. Ingesting alcohol, a variety of illegal drugs, and some toxins such as lead; head injuries; and complications during subject’s pregnancy or both.

4. Personality characteristics such as impulsivity; insensitivity, a physical and nonverbal orientation, and a tendency to take risks.

5. Thinking patterns that focus on trouble, toughness, smartness, excitement, fate, and autonomy, and a tendency to think in terms of short-term rather than long-term consequences.

6. Association with others who engage in and approve of criminal behavior.

7. Weak attachments to other people, less involvement in conventional activities, less to lose from committing crime, and weak beliefs in the moral validity of the law.

8. A perception that there is less risk of punishment for engaging in criminal behavior.

9. Masculinity as a gender role.

10. Economic modernization and development is associated with a rise in property crime rates. Property crime rates tends to increase until the society is highly developed and then hold steady at a high level. The process that results in this pattern of crime involve changes in routine activities and in criminal activities, which eventually are balanced by the increasing effectiveness of countermeasures.

11. Economic inequality is associated with a rise in rates of violence. Such violence may involve feelings of frustration and relative deprivation.

12. Cultures that emphasize the goal of material success at the expense of adherence to legitimate means are associated with high rates of utilitarian crime; an unequal distribution of legitimate means to achieve material success is associated with an inverse distribution of utilitarian crime; and in situations without legitimate means to economic success, the development of illegitimate means is associated with increased utilitarian crime, while the lack of such development is with increase violent crime. In these situations, the inability to achieve status by conventional criteria is associated with status inversion and higher rates of non-utilitarian criminal behavior. The processes involved in these structural patterns involve either frustration or th simple tendency to engage in self-interested behavior.

13. Neighborhoods with high unemployment, frequent residential mobility, and family disruption tend to have high crime rates. The process involve neighborhood anonymity that results in social disorganization.

14. Media dissemination of techniques and rationalizations that are favorable to law violation are associated with increased rates of law violation. The process involves direct learning of techniques and rationalizations and indirect learning of the consequences that criminal behavior have for others.

15. Joblessness and racism can generate an inner-city code of the street that promulgates normative violence in a variety of situations. The process includes feelings of hopelessness and alienation among inner-city residents and the generation of an oppositional subculture as a means of maintaining self-respect.

16. Increases in the objective certainty of punishments are associated with reductions in crime rates, but increases in the objective severity of punishments seems to be associated either with no change or with increases in crime rates. In addition, crackdowns on certain types of crimes are associated with short-term reduction in the rates of those crimes that may extend beyond the life of the crackdown policy itself.

17. Society that stigmatize deviants have higher crime rates than do those that reintegrate deviants. The process involves blocked legitimate opportunity and the formation of subcultures.

18. Societies in which some people control others have higher crime rates than do societies in which people control and are controlled by others in approximate equal amounts. The process involves people’s natural tendency to expand their control.”

This is obviously not a serious way to argue. He provides no evidence that any of these factors have anything to do with racial differences in crime. More other, racial differences in any of these factors could be due to genetics.

Anyway, none of these things are mutually exclusive with a partial genetics explanation. So here is some evidence that genes have something to do racial differences in criminality.

Biological Evidence

First of all, the basic model of racial crime differences fits a genetic model well. The gaps start early in life, are found all over the world, and have been found for several generations. That isn’t proof. But it is suggestive.

IQ and Crime

In part one of my response, I laid out several piece of evidence suggesting that the races differ in IQ for genetic reasons. If this is true, then genetic IQ differences may explain part of why Blacks commit more crime.

Low IQ is consistently associated with higher levels of criminality, and if you hold IQ constant the majority of the Black/White crime gap disappears:


Herrnstein and Murray (1994)



Meta-analyzes show that testosterone is correlated with aggression among humans and non human animals (Book, Starzyk, and Quinsey, 2001). Women who suffer from a disease known as congenital adrenal hyperplasia are exposed to abnormally high amounts of testosterone and are abnormally aggressive. And artificially increasing the amount of testosterone in a person’s blood has been shown to lead to increases in their level of aggression (Burnham 2007) (Kouri et al. 1995). In fact, a study of Rhehus monkeys found that injecting female fetuses with testosterone caused them to behave just as aggressively as young males (Book, Starzyk, and Quinsey, 2001). Thus, testosterone seems to cause aggression.

Testosterone is also known to differ racially. Richard et al (2014) meta-analyzed data from 14 separate studies and found clear evidence that Blacks have higher levels of free floating testosterone in their blood than Whites do.

Further still, multiple studies have found that Blacks are more likely than Whites to carry a genotype which produces an androgen receptor which will make them more sensitive than average to the effects of testosterone (Irvine et al 1995Wang et al. 2013, Bennet et al. 2002, and Shibalev et al. 2013).

And, as reviewed in Ellis, Beaver, and Wright 2013, this high sensitivity genotype has been associated with both aggressive personality traits and crime.

Thus, testosterone is likely one causal factor in the high crime rates of Blacks.


One gene which plays a role in Black’s high crime rate is the monoamine oxidase -a gene (MAO-A). This gene produces an enzyme by the same name. The enzyme MAO-A breaks down a class of neurotransmitter called mono-amines in the brain. These neurotransmitters include ones which are well known to effect behavior such as dopamine and serotonin. Some versions of the MAO-A gene lead to lower levels of MAO-A the enzyme and, therefore, more mono-amine activity in the brain.

The molecular difference between different versions of the MAO-A gene lies in the number of times a certain section for its promoter region is repeated. Three repeat alleles lead to lower than average levels of the enzyme MAO-A. The two repeat allele leads to even lower levels of MAO-A.

Beginning in the 1990’s, researchers began to suspect that variation in MAO-A levels might be related to traits like aggression and criminal behavior. This suspicious arose following a study on the particular family which carried a rare mutation that lead to extremely low levels of the enzyme MAO-A. This family had rapists, arsonists, and other violent criminals (Brunner et al. 1993).

Later, geneticists raised rates with an artificial version of the MAO-A gene which lead to low MAO-A production. It was found to cause the rats to act far more aggressive than is typical (Cases et al. 1995, Scott et al. 2008Vishnivetskaya et al. 2007, and Mejia 2002).

The first direct link between common MAO-A alleles and violence in humans came from Caspi et al 2002. This study found that people with either the 2R or 3R version of the MAO-A gene tended to be more aggressive than average, but only if they had difficult upbringings. Based on these results,  it was hypothesized that low activity MAO-A genes might cause people to have stronger reactions to negative experiences than people high high activity versions of the gene. Since then, a meta-analyses has confirmed that this interaction between MAO-A genotype and childhood trauma has been well replicated in male subjects (Bryd and Manuch 2013).


Source: Caspi et al. 2002

Since then, researchers have found that low activity MAO-A also predict higher levels of anti-social personality even among those who have not experienced child abuse (Ficks and Waldman 2014). These versions of the gene have also been linked to high levels of credit card debt and obesity (Camarena et al. 2004Neve and Flower 2009Need et al. 2005, and Feummeler 2008). Low activity versions of the gene have also been shown to significantly increase the risk that someone will end up in prison (Roux 2014Vaughn et al. 2009Beaver et al. 2009, and Beaver et al. 2013). Given this, it seems plausible that low activity versions of the MAO-A gene leads to increased impulsivity, likely due to increased dompanie activity, which in turn leads to an elevated risk for criminal activity, aggression, obesity, and credit card debt.

Racial differences in MAO-A genotypes are well established. Below you can see a review of the literature in the United States. The vast majority of the research shows that Blacks are more likely than Whites to varry both versions of the low repeat allele.




In summary, Blacks have higher levels of low repeat MAOA alleles than Whites do, and this in turn causes them to have higher crime rates, and that this explains part, though not all, of why Blacks have higher crime rates than Whites.

Skin color

Skin color may also be involved in racial differences in crime. That may sound insane, but here me out:

Variation in the color of people’s skin is largely due to variation in the concentration of melanin in their skin. The more melanin a person has the darker their skin will tend to be. Melanin is produced by a type of cell called a melanocyte. Melanocytes produce melanin in response to a class of hormone, sometimes called melanotropin, binding to a class of receptor called melanocortin receptors. When more melanotropin is introduced, via the pituitary gland, into the blood stream, more binding with melanocortin receptors occurs which in turn causes melanocytes to produce more melanin which in turn causes a darkening of the skin (Emerson).

There is some evidence that melanotropin causes aggression (Reviewed in Rushton and Templer 2012).

First, there are studies of skin color variation in humans. Studies which utilize estimates of the the average skin color of over 100 countries find that darker skin predicts higher crime rates. A similar correlation has been found in non-human animals. In fact, Animal research has shown that dark skin predicts more aggression in over 200 species.
The evidence that melanotropin causes aggression is not limited to correlations, either. Experiments which breed animals to become less aggressive, from mice to fox, tend to end up with lighter skinned animals. Even more impressively, injecting animals with melanotropin has been shown to cause them to become more aggressive.

Now, let’s say that this is all true. We don’t know how these hormones cause aggression. It is obviously an indirect effect, but we do not know if it is an effect of the hormones themselves or the chemicals they produce. Because of this, it is hard to say for sure whether or not this is a factor in racial crime differences. But it very well could be.

 Moving Forward

There are many biological variables which impact crime that we don’t know about. There are also many we do know about  in general but not in terms of population differenes. But what we do know suggests that genes and biology probably play a role in racial crime differences. Perhaps more importantly, there is absolutely no reason to think the opposite is true, that genes play no role in racial crime disparities.

And that’s part 2 on this response! On to part 3…

Facebook Comments
  • Mengerian Knight

    Yet another outstanding debunking of anti-racist pseudo-science. Keep up the great work!

  • B.B.

    Aedon Shevirah Cassiel who blogs at Zombie Meditations has been doing an excellent job exposing Tim Wise’s pseudo-scientific sophistry.

    • Àedhàn Cassiel

      Thanks for the shout–out, B.B.!

  • The Last American

    First paragraph under “Skin color”: should read “hear me out” not “here me out”.

    Thanks for the write-up, keep ’em coming!

  • This chart from the FBI crime stats is wrongly labelled, it says “White American representation in arrests”, it should be “White+Jewish+Hispanic+Arab+Turkish+North African+Sudanese American share of arrests”:

    • Well for instance, Rushton’s data covers Caucasians.

      Moreover, on fbi table 43 if you scroll right you see a ‘Hispanic’ category. I would guess that by subtracting that from the white stats you’d get something closer to reality. When arrested, they out race white, ethnicity hispanic. I had a pal growing up who got arrested who was dark ‘Hispanic’. He was PR and DR. He got put race black, ethnicity ‘Hispanic’. So there is differentiation there.

      How many Jews, Arabs and Turks get taken in to have any noticeable effect on the data? As far as I know, there are about 1 million Arabs on America. Do you think those who are of age have any effect on the crime rate?