Monthly Archives: April 2016

Calling for a Nazi / Social Justice Warrior Alliance

Imagine a world where the following paragraph was true:

White people are just 2% of the population of South Africa.

And yet, a whopping 31% of South African media companies are owned by white people; 38% are founded by white people; 45% of their presidents are white people; and 47% of their chairmen are white people. 26% of all the reporters, editors, and executives of the major print and broadcast media are white people. 75% of the senior administrators of the best South African colleges are white people, and from 11 to 27 percent of students admitted to those colleges are white. 139 of the top 400 richest people in South Africa are white. Of the top 100 political campaign funders, at least 42 of them are white. 15 out of 30 executives at the major think tanks that determine policy are white. To top it all, 8 of 11 senior advisers to President Zuma are white.

The corollary of these statements is that Blacks are around 98% of the population, and yet make up only 69% of media company owners, 62% of their founders, and 55% of their presidents … Only 25% of senior administrators at the best colleges are black; and only 3 of 11 Presidential advisers are black.

What would leftists’ response to this situation be?

The answer to that question is beyond doubt: they’d be outraged.

And it wouldn’t matter in the slightest that whites were a minority of the South African population—that would just make their domination of the country’s most important offices worse.

In the United States, we have a group calling itself the ‘Reflective Democracy Campaign’ which finds that white men are 31% of the population—but 66% of those who run for political office, and 65% of those elected. Once these figures are produced, no further investigation is required before leftists start asking why it is that “in the year 2015, there are roughly double the number of white men in elected office as there ought to be[?]”  Another campaign strives to draw awareness to the fact that white men make up 79% of elected prosecutors.

Or to give another example, when Spike Lee thought black winners at the Oscars were underrepresented compared to white winners, he called for a boycottIt turns out he was wrong: a USC study found that blacks, who are about 13% of the U.S. population, comprise 12.5% of actors in the top 100 films from 2007; 23 of 192 Oscar nominations (12%), and 9 out of 68 academy awards since 2000 (13.2%)—close to perfect statistical representation. But the mere idea that whites might be overrepresented in the Oscars compared to blacks was all it took to set off a loud and persistent conversation, with many people instantly prepared to believe that whites are overrepresented and that this is a problem in need of urgent address.

So in the case of the Oscars, the over–representation of whites compared to blacks was exactly zero. And in the case of the Reflective Democracy Campaign’s argument, whites are overrepresented amongst political candidates at just 1.4 times their population rate (whites are 63% of the population, and a combined 89% of Republican and Democratic candidates), and amongst elected prosecutors at 1.25 times their population rate.

So we can absolutely rest assured that if our opening paragraphs were true, liberals would be outraged to find whites overrepresented at 5–36 times their rate of the population rather than a mere 1.2.

So what makes liberals different from white supremacists—besides their target?

Everything stated in the opening paragraph of this post is, in fact trueabout Jews. 

Jews are just 2% of the United States population. And yet, they make up 18 out of 24 senior administrators of Ivy League colleges (75%), 8 out of 11 senior advisors to President Obama (72%), 8 out of 20 Senate Committee chairmen (40%), 33 out of 51 senior executives of the major Wall Street banks, trade exchanges, and regulatory agencies (64%), 23 out of 40 senior executives of the major Wall Street mutual funds, private equity funds, hedge funds, and brokerages (57%), 41 out of 65 senior executives of the major newspapers and news magazines (63%), 43 out of 67 senior executives of the major television and radio news networks (64%), 15 out of 30 senior executives of the major think tanks (50%).[1]

New students admitted to Harvard University? 25% Jewish. Yale? 27% Jewish. Cornell? 23% Jewish.

And when Jewish organizations reflect on Jewish representation in Ivy League colleges, they do so not to worry about whether Jews are pushing non–Jews out through their own overrepresentation, but to analyze the puzzle that “Thirteen percent of Princeton’s undergraduate student body is Jewish, the lowest percentage of any Ivy League university besides Dartmouth, which comes in at 11 percent.” Yet, both of these are still more than 4 and 5 times the Jewish percentage of the population.

The media? If we’re looking at the CEOs of media companies, then they’re 31% of the total. If we’re looking at founders, then they’re 38%. If we’re looking at presidents, then they’re 45%. If we’re looking at chairmen, then they’re 47%. If we’re talking about the directors and writers, then Jews represent “26 percent of the reporters, editors, and executives of the major print and broadcast media, 59 percent of the directors, writ­ers, and producers of the 50 top-grossing motion pictures from 1965 to 1982, and 58 percent of directors, writers, and producers in two or more primetime television series”.

These numbers range from over 12 to over 22 times the Jewish percentage of the population.

Banking? Of the five Federal Reserve board governors (Daniel K. Tarullo, Jerome H. Powell, Lael Brainard1, Stanley Fischer2, Janet L. Yellen3), three are Jewish. Of the nine executive officers of Goldman Sachs (Edith W. Cooper, Gregory K. Palm, John F. W. Rogers, Alan M. Cohen1, Harvey M. Schwartz2, Mark Schwartz3, Gary D. Cohn4, Lloyd C. Blankfein5, Michael S. Sherwood6), six are Jewish. Of the ten operating committee members of JP Morgan Chase (John L. Donnelly, Gordon A. Smith, Jamie Dimon, Mary Callahan Erdoes, Matthew E. Zames1, Daniel E. Pinto2, Douglas B. Petno3, Marianne Lake4, Stacey Friedman5, Ashley Bacon6), six are Jewish. Combining just these three major banks, 62% are Jewish—almost 30 times the Jewish population rate.

“ … the Jews run everything? Well, we do. The Jews run all the banks? Well, we do. The Jews run the media? Well, we do … It’s a fact; this is not in debate. It’s a statistical fact … Jews run most of the banks; Jews completely dominate the media; Jews are vastly disproportionately represented in all of these professions. That’s just a fact. It’s not anti-Semitic to point out statistics … It’s not anti-Semitic to point out that these things are true.” — Milo Yiannopoulos, The Rubin Report, March 2016

So how can leftists, who immediately take any statistical over–representation of whites in anything at all as a major social problem that needs to be changed—even at just 1.1 or 1.4 times the white population rate—condemn white supremacists for being worried about statistical over–representations several times larger than that? Indeed, how are the racialist left and white supremacists anything but two different sides of the same coin?

Amusingly enough, a large percentage of my audience will probably suspect me immediately of having gone full Nazi just because I went through the effort to pinpoint exactly how overrepresented Jews are at all. Now, that suspicion may be fair—but if so, why is it that going through the effort to pinpoint how overrepresented whites are in various fields or professions is not seen as bigotry in just exactly the same way?

As a matter of fact, the ‘Reflective Democracy Campaign’ itself has apparently failed to notice that it is not “whites” who are overrepresented within the legal profession—it’s Jews, who in fact make up 26% of the nation’s law professors, and 30% of Supreme Court law clerks. In Jews and the New American Scene, Seymour Lipset and Earl Raab point out that Jews make up “40 percent of partners in the leading law firms in New York and Washington.” So Jews are overrepresented in the legal profession at 13 or more times their population rate.

And if you subtract the 26% of lawyers who are Jewish from the 79% of prosecutors the RDC calls “white”, that leaves only 53% of prosecutors who are non–Jewish whites, compared to about 61% of the U.S. population that is non–Jewish white. So it turns out that ‘whites’ are not overrepresented at all—they’re under–represented at about 0.86 times the population rate. But what would happen to the RDC’s left–wing credentials if it were to openly admit this and call explicitly for a reduction of the Jewish percentage of elected prosecutors?

Indeed, what would happen to their public image in general once this was known?

Suddenly, they’d go from being a respectable campaign calling attention to a real social issue to being classed with Nazis and white supremacists—the lowest of the low—just because the demographic their numbers targeted happened to turn out to be Jews instead of whites. But why is it that this kind of campaign is valid just so long as it targets whites, and racist bigotry the moment it hits any other demographic?

Why are Jews statistically overrepresented? There are essentially two possibilities:

  1.  Jews could be acquiring positions of power and then using them to grant favors to other Jews—say, Jews could take over the senior administrative positions in Ivy League colleges (where they indeed compose about 75% of the total), and then they could favor admitting Jews as new students over others.
  2. Perhaps Jews are simply more intelligent, or industrious, or intellectual, or otherwise have temperaments more conducive to these arenas—and so they acquire their status in these positions through legitimate success.

The first of these options is the white supremacist answer: Jews aren’t any more intelligent than the rest of us; they’re just more nepotistic, networking with other Jews to take over the world. In order to avoid sounding like bigots, then, we’re supposed to give the second answer: Jews are simply more intelligent or more industrious or more intellectual, or simply have temperaments more conducive to these arenas.

But if we’re talking about whites instead of Jews, then suddenly the first option is exactly what social justice warriors demand that we say: ‘whites aren’t any more intelligent than anyone else; they’re just more nepotistic’! Meanwhile, the second option is suddenly the one that is now inexcusably, irredeemably racist: if you claim that whites are simply more intelligent or more industrious or more intellectual, you’re a bigot.

What the ‘politically correct’ view requires us to say about Jews is exactly what it calls bigotry if we say it about whites. And what it requires us to say about whites is exactly what it calls bigotry if we say it about Jews. The disproportionate success of whites is purely the result of unjust ‘privilege’, and you’re a bigot if you think it has anything to do with greater merit. But the disproportionate success of Jews is the result of greater merit, and you’re a bigot if you try to diminish that by attributing it to ‘privilege’, much less want it to change!

The egregiousness of the naked double standard here is overwhelming. As far as resolving it, it would seem we have exactly two possible options: either we grant the argument in both cases, and encourage the social justice warriors and white supremacists to join forces against their new common foe—or else we deny it in both cases.

So which is it?


The ‘Poverty’ of Sociology

It’s obvious that there is, in general, a geographical correlation between poverty and crime. What I mean by that is that if we look at a map of the United States (or the world—but this post will focus on the United States) at any given point in time, in places where we see lots of poverty, we will also see lots of crime. 

This much is beyond serious question.

What is under–appreciated, however, is just how complicated it is to actually explain why. The obviousness of the geographical correlation between poverty and crime has led many to assume that it must be just as obvious that poverty “causes” crime. On the other hand, many social conservatives have argued that poverty and crime correlate with each other only because divorce and out–of–wedlock birth produce them both: according to this argument, single–parent families produce poverty because they earn less income than two–parent families do; and they produce crime because boys raised by single mothers have no models of masculinity to learn from and emulate, and therefore become more likely to attempt to express their masculinity through violence and affiliation with gangs.

Disentangling cause and effect in these relationships is more difficult than the proponents of either theory often assume—for even correlations that seem obvious at first glance can turn out to have causes that no one even considered. As we will see, both the “poverty causes crime” advocates and the “single parenthood causes both poverty and crime” advocates are, for the most part, wrong (though each is also about 1% correct).

A cautionary tale

By the mid–1990’s, hormone replacement therapy had become one of the most widely prescribed medications for women in North America. Books were published touting the benefits of synthetic hormones injections with titles like “Feminine Forever!” Several large studies (Stampfer 1991) found that even after controlling for other risk factors like age, “estrogen use is associated with a reduction in the incidence of coronary heart disease as well as in mortality from cardiovascular disease”. Another meta–analysis (Grady 1992) found a 35% reduction in heart disease amongst those using synthetic estrogen and concluded that “hormone therapy should probably be recommended for women … with coronary heart disease or at high risk for coronary heart disease.”

Yet, by the late 1990’s and early 2000’s, this consensus had fallen apart completely. Not only did it turn out not to be the case that hormone replacement therapy was beneficial for women with or at risk of heart disease (Rossouw et al. 2002), in many cases it actually turned out to increase the risk of heart disease (Hulley et al. 1998).

What happened?

Was the earlier research falsified? No.

The correlation between use of estrogen and lower heart disease risk found by earlier research did, in fact, exist.

It just didn’t get there because the use of estrogen causes a reduction of heart disease risk. It simply turned out to be the case that, on average, the women who were trying hormone replacement therapy were women of higher socioeconomic status who also tended to keep healthier diet, lifestyle, and exercise habits. Thus, the use of estrogen was increasing the risk of heart disease all along, despite the fact that it was true that women trying hormone replacement therapy did have lower heart disease rates on average than those who weren’t.

What we have here is an excellent example of a “hidden variable” explanation for a correlation. The original assumption behind the correlation between hormone replacement therapy (HRT) and lowered heart disease risk (–CHD risk) was that HRT caused –CHD risk. And this false assumption likely contributed to some uncertain number of unnecessary deaths. The real answer turned out to be that some other, previously unidentified factor (socioeconomic status increasing likelihood of both continued use of HRT and better lifestyle habits), was causing both.


To use a more commonplace example of a faulty inferences of correlation from causation, it is obviously true that we find fire burning only in the presence of oxygen. Wherever we see fire, then, we are bound to find oxygen. But this doesn’t make oxygen “the cause” of fire burning—indeed, since there are so many places where we can find oxygen but no fire, it is obvious that something else must be “the cause”.

Similarly, we are bound to find more human trafficking in places where there are more women who are vulnerable to being captured and exploited. But if anyone were to suggest that contact with a vulnerable woman is the literal “cause” of a man’s decision to kidnap and traffic her into sex slavery, the very same liberals who ask us to excuse crime while addressing its “root causes” would condemn this as “victim blaming” of the most horrendous and disgusting form. Yet, this seems like an arbitrary attempt to have one’s cake and eat it too—for the sake of consistency, we must either accept that human beings have (at least some degree of) free will, or else we must deny that and grant that all human behavior is the deterministic result of external circumstances across the board.

When analyzing these kinds of questions, we shouldn’t lose sight of just how banal much violent crime is.

Most aggressive crime just doesn’t look anything like a struggling family pocketing a loaf of bread after spending the rest of the grocery budget to feed their children. It looks like 39–year–old Ronald McNeil murdering a 19–year–old female college freshman, because of a fight with a different party attendant over the rules of beer pong. It looks like public gang rapes outside of rap concerts. It looks like randomly setting a 19–year–old girl on the way to get dinner on fire. But let’s not spend too much time on anecdotes before we move on to data.

Crime and poverty: does one cause the other?

First of all, it is absolutely, undeniably true that crime does help to cause poverty.

“A high crime rate will drive businesses out of a neighborhood. This eliminates both availability of products and services and a source of jobs. Further, those who do stay find it necessary to charge higher prices to offset losses due to thievery and higher costs of both security measures and insurance premiums—if insurance is available at all.

Property values are driven down by a smaller demand because of the greater difficulty potential purchasers have in obtaining mortgage loans.

The loss of productive activity by those who live by preying on others reduces the output of the area in which they live. Thus, crime injures economically both direct victims and others in the crime-ridden neighborhood.”

A more recent study calculated only the direct losses of victims; the price spent on police, prisons, and lawyers; and the opportunity costs for the perpetrator himself. It found that the average cost of each act of robbery is around $42,000; of each act of assault, more than $100,000; and of each act of murder, almost $9,000,000.

These estimates come without looking at the damage done to a community’s economy through crime’s impacts on third parties other than the perpetrator and his victim (the flight of businesses and thus opportunities away from high–crime areas, the raised price of insurance, the loss of property values, and so forth), and so they undoubtedly underestimate the true amount of damage caused by crime.

What about poverty causing crime? It is true that poverty and crime correlate geographically: in locations where we find more poverty, we are also going to find more crime. But it turns out that poverty and crime do not correlate very well historically: when poverty rises, we do not see concurrent rises in crime.

Both before and after the Great Depression, the relationship between poverty and crime actually appears to have inverted: “Most evidence suggests that the crime rate rose after World War I and the 1920s and that crime rates dropped as the nation sank into the Depression and continued to decline into the 1940s.” Eli Lehrer adds extra detail: “Crime rates fell about one third between 1934 and 1938 while the nation was struggling to emerge from the Great Depression and weathering another severe economic downturn in 1937 and 1938. Surely, if the economic theory held, crime should have been soaring.”

And as he continues, he explains that this same inverted relationship was also found during several other recessions over the past century, as well: “Crime rates rose every year between 1955 and 1972, even as the U.S. economy surged, with only a brief, mild recession in the early 1960s. By the time criminals took a breather in the early 1970s, crime rates had increased over 140 percent. Murder rates had risen about 70 percent, rapes more than doubled, and auto theft nearly tripled. … Crime rates fell in nearly all categories between 1982 and 1984, even though … wages fell for low-income workers during the same period. Likewise … wages rose for low-income workers between 1988 and 1990, despite being a period of higher crime rates. In fact, some of the worst years for crime increases were in the late 1950s, as hourly wages surged ahead. Between 1957 and 1958, for example, per–capita income increased about 8 percent while crime rose nearly 15 percent.”

Patrick F. Fagan adds: “What is true of the general population is also true of black Americans. For example, between 1950 and 1974 black income in Philadelphia almost doubled, and homicides more than doubled.” Similarly, poverty rates between different ethnic groups fail to explain their different crime rates today: in the 2006 American Community Survey, 21.5% of Hispanics lived in poor households and 37.2% of Hispanic men age 18–24 had not completed high school in 2005—compared with 25.3% of blacks and 26.3% of black men: in other words, 3.8% fewer Hispanics lived in poor households, but 10.9% more Hispanic men failed to graduate high school. If poverty were causing violent crime, then we would expect the violent crime rate to be similar amongst blacks and Hispanics. But that isn’t what we find; instead, the Hispanic crime rate is only slightly higher than the white crime rate, both of which are far lower than the black crime rate—even after controlling for age groups to account for the different proportions of young adult males (who commit the vast majority of crime) in each ethnic group.


from — for a lot more detailed charts and graphs on this topic, Random Critical Analysis has done plenty of heavy work in the Nov. 2015 post, “Racial differences in homicide rates are poorly explained by economics.”


And what holds historically about the association between poverty and crime continues to hold into the present day, with the discovery that the “Great Recession” of 2007–2009 came with a reduction in crime, too.

Writing in the Wall Street Journal, James Q. Wilson explained: “As the national unemployment rate doubled from around 5% to nearly 10%, the property-crime rate, far from spiking, fell significantly. For 2009, the Federal Bureau of Investigation reported an 8% drop in the nationwide robbery rate and a 17% reduction in the auto-theft rate from the previous year. Big-city reports show the same thing. Between 2008 and 2010, New York City experienced a 4% decline in the robbery rate and a 10% fall in the burglary rate. Boston, Chicago and Los Angeles witnessed similar declines. … In 2008, … even as crime was falling, only about half of men aged 16 to 24 (who are disproportionately likely to commit crimes) were in the labor force, down from over two-thirds in 1988, and a comparable decline took place among African-American men (who are also disproportionately likely to commit crimes).”

Heather MacDonald supplies additional data: “[B]y the end of 2009, the purported association between economic hardship and crime was in shambles. According to the FBI’s Uniform Crime Reports, homicide dropped 10% nationwide in the first six months of 2009; violent crime dropped 4.4% and property crime dropped 6.1%. Car thefts are down nearly 19%. The crime plunge is sharpest in many areas that have been hit the hardest by the housing collapse. Unemployment in California is 12.3%, but homicides in Los Angeles County, the Los Angeles Times reported recently, dropped 25% over the course of 2009. Car thefts there are down nearly 20%.”

Okay, so what if all of these hard statistical measures of the economy are too crude to capture what really matters for someone’s likelihood to commit a crime—how they perceive the economy as doing, regardless of the facts? Well, that brings us back to James Q. Wilson: “the University of Michigan’s Consumer Sentiment Index offers another way to assess the link between the economy and crime. This measure rests on thousands of interviews asking people how their financial situations have changed over the last year, how they think the economy will do during the next year, and about their plans for buying durable goods. The index measures the way people feel, rather than the objective conditions they face. It has proved to be a very good predictor of stock-market behavior and, for a while, of the crime rate, which tended to climb when people lost confidence. When the index collapsed in 2009 and 2010, the stock market predictably went down with it—but this time, the crime rate went down, too.”

Steven D. Levitt’s Understanding Why Crime Fell in the 1990s: Four Factors that Explain the Decline and Six that Do Not summarizes the research: “Empirical estimates of the impact of macroeconomic variables on crime have been generally consistent across studies: Freeman (1995) surveys earlier research, and more recent studies include Machin and Meghir (2000), Gould, Weinberg and Mustard (1997), Donohue and Levitt (2001) and Raphael and Winter-Ebmer (2001). Controlling for other factors, almost all of these studies report a statistically signiŽficant but substantively small relationship between unemployment rates and property crime. A typical estimate would be that a one percentage point increase in the unemployment rate is associated with a one percent increase in property crime.”

He concludes: “Based on these estimates, the observed 2 percentage point decline in the U.S unemployment rate between 1991 and 2001 can explain an estimated 2 percent decline in property crime (out of an observed drop of almost 30 percent)….” But yet again, even here, the direction of causation isn’t clear. Levitt misspeaks when he says the conclusion is warranted by this evidence that the decline in the unemployment rate “can explain” the decline in property crime; because the word “explain” invokes causation, and what this data shows us still isn’t causation yet.

How do we know it’s the unemployment rate that “explains” the decline in property crime? How do we know it isn’t the decline in property crime that explains the decline in the unemployment rate? If someone decides not to commit a home robbery, he obviously has a much better chance of finding a job in the near future than if he does. And in all likelihood, a business in a town with fewer property crimes is making more sales and therefore able to employ more people; more people are considering starting businesses; and more established businesses are considering moving in. At the very least, this effect must contribute to the correlation; and that means that a 1% decline in the unemployment rate must cause somewhat less than a 1% decline in the property crime rate.

So even with property crime, fluctuations in the economy don’t explain much at all. But as Levitt continues the summary, he explains: “Violent crime does not vary systematically with the unemployment rate.” What if the unemployment rate isn’t the right measurement of the economy? “Studies that have used other measures of macroeconomic performance like wages of low-income workers come to similar conclusions (Machin and Meghir, 2000; 170 Journal of Economic Perspectives Gould, Weinberg and Mustard, 1997).” Now, more astute readers may wonder why, if the hypothesis in the last paragraph about property crimes causing unemployment were plausible, violent crimes wouldn’t have the same effect. A possible answer is that generally speaking, far more of the kinds of people who would contemplate committing property crimes are potentially employable to begin with, whereas comparatively far more of the kinds of people who would contemplate committing violent rapes or murders already exhibit a demeanor or engage in other behaviors that make them less employable anyway.

Yet another point that demolishes the left–wing narrative: white–collar crime.

Isn’t it the left telling us that it’s the rich who are causing all of the real problems in the world in the first place? Aren’t they the ones telling us that it’s the rich white men running the world who are destroying the environment, lying to the public, committing embezzlement and collusion and fraud, fighting for policies that hurt the poor, and invading foreign countries to kill thousands of innocent people for no reason other than selfish gain?

Doesn’t that, in and of itself, contradict the notion that poverty “causes” crime?

Doesn’t that, in and of itself, prove that even liberals don’t actually believe that raising everyone’s economic welfare is all it takes to put an end to anti–social behavior and make people be nice to each other?

White–collar crime is interesting because of the way that it exposes the contradictory hole in the center of this set of beliefs, but it is also interesting for another reason, as well: it shows, once again, that whatever makes different demographics commit crimes at different rates, poverty isn’t a good explanation—because the disparities in crime that exist on the street actually turns out to exist in corporate offices as well.

Obviously, white people do commit the majority of white–collar crimes; and the harm that can come from these acts shouldn’t be understated. The savings and loans scandal of the 1980s was almost exclusively committed by white people, and cost U.S. taxpayers over $470 billion—more than all the conventional bank robberies in U.S. history combined. White people have “disproportionately” achieved positions of economic power and influence, and the damage that people can do in these positions substantially outweighs what any number of street criminals are capable of. However, what the data reveals is that white people in these positions are nonetheless proportionally underrepresented amongst white–collar criminals—in other words, whites are a percentage larger than their share of the population of those in corporate positions, but they still commit a percentage less than their share of the “corporate population” of all percentage of white–collar crimes. While ~99% of anti–trust and security fraud offenses are committed by whites because they’re effectively the only ones in positions to, non–whites are nonetheless found to be overrepresented in all the other corporate crimes they are in positions to commit.

These findings led Hirschi and Gottfredson to conclude in The Causes of White–Collar Crime that “When opportunity is taken into account, demographic differences in white collar crime are the same as demographic differences in ordinary crime.” But they weren’t, of course, referring solely to race: men are disproportionately likely to commit white–collar crimes relative to women, even once opportunity is taken into account, as well. In fact, men were found to be even more disproportionately overrepresented in white collar crime than they are in street crime. Likewise, the commission of white–collar crimes peaks around age 20, and falls in half by around the age of 40, and once again, this exactly fits the pattern of all other crimes. Whatever it is that causes men to commit more crimes than women, whatever it is that causes the young to commit more crimes than the old, and whatever it is that causes some ethnic groups to commit more crimes than others, it doesn’t look like poverty can be the explanation.

In 2014 came the final nail in the coffin to the “poverty causes crime” thesis. A Swedish study conducted by Amir Sariaslan was published which—for the first time—tested directly whether growing up in poverty directly contributes to crime, or whether there are other factors about the kinds of families which tend to end up poor which also cause them to breed crime. What made Sariaslan’s study uniquely insightful was the decision to take families which rose out of poverty, and compare the lives of children born and raised within those families before their rise from poverty with the lives of children born and raised within those same families after their rise from poverty.

The conclusion his research came to? “There were no associations between childhood family income and subsequent violent criminality and substance misuse once we had adjusted for unobserved familial risk factors.” Sariaslan’s study, in other words, had proven that growing up in poverty is not what creates one’s adult likelihood of committing violent crimes. Children who grow up in previously–poor families have exactly the same likelihood of committing crimes as children who actually grow up poor. The only conclusion we can soundly come to is that something else about poor families other than poverty itself must explain why their children go on to commit crimes.

Many conservatives think the root of social dysfunction is a lack of monogamy. 

Criminologist Anthony Walsh writes in Race and Crime: A Biosocial Analysis, for example:

“If racism were the culprit behind the difference in poverty rates, we would expect black families, regardless of their household composition, to be worse off than white families, regardless of their household composition. But this is not what we observe. The U.S. Census Bureau’s (McKinnon & Humes, 2000) breakdown of family types by race and income showed that non-Hispanic white single-parent households were more than twice as likely as black two-parent households to have an annual income of less than $25,000 (46% versus 20.8%). To state it in reverse, a black two-parent family is less than half as likely to be poor as a white single parent family. These figures constitute powerful evidence against the thesis that black poverty is the result of white racism, as well as powerful evidence that high rates of single-parenting is a major cause of family poverty for all racial/ethnic groups. The prevalence of single-parent families is so high in the black community that: “[A] majority of black children are now virtually assured of growing up in poverty, in large part because of their family status” (Ellwood & Crane, 1990:81).”

However, a study by Sara McLanahan found that “The dropout risk is 37 percent for those with never-married mothers and 31 percent for those with divorced parents, in contrast with the 13 percent risk of those from families with no disruption. Significantly, the risk for children who lost a parent to death is 15 percent—virtually the same as that for children from intact homes. Clearly, children of a widowed mother enjoy economic and other advantages over their peers from households headed by divorced or never-married parents.”

Emphasis mine. What are these “other” advantages?

The only truly plausible candidate for an answer is genes.

Commenting on these findings, Razib Khan (graduate student in genomics at UC Davis) writes:

“The null hypothesis which the media and the public intellectual complex sell us is that destabilized households lead to late life destabilization in individuals. What this misses is that destabilized individuals lead to destabilized households, and destabilized individuals also produce other destabilized individuals. In other words, one reason that kids whose parents didn’t stay together and are messed up is because they have the same crappy dispositions as their parents. They share genes with their parents.

This isn’t to deny that all things equal being in an intact nuclear family is preferable to being raised by a single parent. Ask anyone who grew up in a situation where they lost one of their parents to cancer or some such thing. But naive assumptions that simply increasing the marriage rate will reverse social dysfunction are going to be dashed against the reality that putting together explosive impulsive people under the same roof is not going to turn them into Ward and June Cleaver.

If behavioral genetics or the idea of heritability is new to you, one of the best introductions to the basics can be found in Brian Boutwell’s article at Quillette, “Why parenting may not matter and why most social science research is probably wrong”; as well as the follow–up, “Heritability, and Why Parents (But Not Parenting) Matter”. The twin studies, adoption studies, and family studies that these conclusions are based on have been challenged for years, and they have stood up to all of these challenges remarkably well. One problem with any attempt to critique their validity is the odd fact that they all tend to converge on the same exact estimates of how heritable various traits are: if all of them are flawed in different ways, how is it that they all consistently land on the same results? It’s like when young earth creationists critique the validity of carbon dating—do you really think it’s just sheer coincidence that carbon dating and helioseismic dating converge on exactly the same estimates for the Earth’s age? I’ll be addressing more general background on twin, adoption, and family studies as well as the critiques that have been made of them in the future. For now, I’m going to take their validity for granted and simply discuss what the research has shown.

In men, studies find that anywhere between 40% to 60% of the likelihood of divorce is the result of “genetic factors affecting personality.” More generally, a person’s “sociosexual orientation” is clearly found to be very highly heritable. Individuals are classified on this scale as either “sociosexually restricted” or “sociosexually unrestricted”. An ordinary person might simply call them “chaste” or “promiscuous”: so–called “unrestricted” individuals are more likely to engage in sex earlier in relationships, engage in sex with more than one partner at a time, seek sex for its own sake, and engage in it in relationships involving less love, dependency, and commitment.

Twin, adoption, and family studies are able to separate the role of heredity, “shared environment” (which essentially means “parenting”), and “non–shared environment” (which essentially means everything else) in the development of various behavioral and personality traits. The conservative argument about monogamy is severely damaged not just by the fact that divorce and sociosexuality have such a large genetic component, but by the fact that all indications so far reveal almost zero effect on these traits from one’s parenting, even once the influence of genes is taken out of the picture: what’s left over after genes are accounted for falls almost entirely into “non–shared environment”—a category which roughly means “we don’t know what it is, but it isn’t genes or parenting.”

As one of the studies quoted in the last paragraph states in its conclusion, “Consistent with genetic theory, familial resemblance [in sociosexuality] appeared primarily due to additive genetic rather than shared environmental factors.”

Shared environmental factors: that means parenting.

Another study compared children who experience their biological parents’ divorce with children who experience their adoptive parents’ divorce, and found that “adopted children who experienced their (adoptive) parents’ divorces exhibited elevated levels of behavioral problems and substance use compared with adoptees whose parents did not separate, but there were no differences on achievement and social competence.” While some behavioral problems (but not others) do result from experiencing one’s adoptive parents’ divorce, it isn’t the experience of divorce (or growing up in a single parent family) that molds a child’s core personality. The illusion that this is so happens because in most cases, a child both undergoes the experience of divorce and inherits his genes from the divorcing parents. But this illusion becomes untangled when adoptive children experience their adoptive parents’ divorce—some short term behavioral problems result, but not others; and most importantly, these behavioral changes do not appear to last the same way that they do in ordinary cases where a child undergoes a biological parents’ divorce.

Yet another study found that once the criminal behavior of single parents was actually controlled for, the association between single parent families and crime disappeared entirely. So the offspring of single parents are more criminal because their parents tend to be criminal.  And clearly, if being raised by one criminal parent produces poor outcomes for children, then being raised by two of them can’t be much better.

So the evidence suggests that the correlation between poverty and crime is taken care of by “unobserved familial risk factors”—but it also establishes very clearly that, in general, the individuals within families are similar in the ways that they are in large part because of their shared genes, and very specifically not because of their shared upbringing. And it proves this in the specific cases even of divorce and sociosexuality. Thus, poverty and crime can’t correlate with each other because each is the causal result of broken homes. Poverty, crime, and out–of–wedlock birth therefore must correlate with each other to the extent that they do because all three are the result of other things that tend to cause all three. But the only causes consistently found so far are genetic—and most of what isn’t genetic, as far as we’re able to tell, is simply random (again, for more, see Jayman’s Blog).

Of course, the correlation between single–parent families and crime actually has become weak, though it may have appeared stronger when the theory first originated. While it’s true that both crime and single parenthood rose together from around 1960 to 1990, this relationship decoupled during the massive crime decline of the 1990s—when crime fell tremendously even as single parenthood continued its decades–long gradual rise.



Now, many liberal commentators (like biological anthropologist Greg Laden) were too quick to pick up on the above chart as proof that there is no relationship between single parenthood and violent crime. To see why more evidence is needed before we can reach that conclusion, picture a chart with an x–axis titled “how long my stopped faucet has been running” that starts at 12:00pm and ends at 1:00pm, and a y–axis titled “how much water is spraying out towards my floor”—measured by quantifying the amount of water actually landing on my floor. My faucet stays on for the full duration of the whole hour, but around 12:30pm the relationship stops being linear because, suddenly, the amount of water on my floor decreases. Does this chart refute the notion that, all else equal, keeping my stopped up faucet running increases the amount of water spraying towards my floor?

Of course not. If the relationship decouples, we can’t immediately assume that keeping the faucet on was never increasing the amount of water spraying towards my floor. Maybe what happened is that around 12:30pm, I became more diligent at stopping it on its way towards my floor before it actually got there—say, because I put down buckets and I mopped up the floor with towels. If changes in how we tackle violent crime once it is already in existence took place during the 90’s, then perhaps we just became more efficient at fighting the crime that single parenthood was helping create. And in fact, something like this did happen: in 1972, only 158 out of 100,000 people were in prison or jail; by 1991, that doubled to about 311 out of 100,000 people.

So perhaps what this chart proves is only that the criminal justice system, by becoming more aggressive, also became much more effective at reducing crime that was being produced by, amongst other things, single parent families. Liberal commentators like Greg Laden are being profoundly dishonest when they wag their fingers at the other aisle without first considering the possibility. Unlike the relationship between poverty and crime, there is at least a long stretch of time during which the two variables rise together. And unlike the relationship between poverty and crime, we don’t repeatedly see shifts in which poverty goes down (or up) and yet crime goes up (or down).

However, the best controlled analysis shows that, just like the relationship between poverty and crime, there is only a tiny relationship left over once other factors have been controlled for. A 2009 meta–analysis of previous meta–analyses which looked at individuals actually found that less than 1% of the population’s variation in criminality could be explained by family structure (although studies which looked at different world regions did find much higher geographic correlations between single parenthood in crime—in other words, in places where we find lots of single parents, we’ll also find lots of crime. The fact that we find a strong ‘geographic’ correlation combined with a poor ‘historical’ correlation supports the claim that what correlation does exist exists only because of some other “hidden variables” that tend to come together, but don’t come together necessarily).

Single–parenthood is proposed as an explanation of criminality, first and foremost, because rates of both single–parenthood and criminal behavior are higher in black populations. Yet, it is clear now that changes in single–parenthood rates do not actually correlate well with changes in rates of crime. So it turns out that high rates of single parenthood in black communities can’t explain why crime rates tend to be higher in these areas, either.

Surprisingly, of the poorest ten counties in the United States, none contain black majorities. Most of these are either Indian reservations (like Ziebach County, South Dakota, which is ~72% Native American with ~62% of the population in poverty), or Appalachian states with large white majorities (like Owsley County, Kentucky, ~99% white with an annual median household income under $22,000). Despite the poverty rates across this second group of poor counties, however: “There’s a great deal of drug use, welfare fraud, and the like, but the overall crime rate throughout Appalachia is about two-thirds the national average, and the rate of violent crime is half the national average, according to the National Criminal Justice Reference Service.”

However, the population density of Owsley County is just 24 people per square mile.

In contrast, Chicago has a population density of over 11,000 people per square mile.

In his “Reflections on the Politics of Crime”, Tim Wise emphasizes a few citations which suggest that “concentrated poverty” (high population density in poor neighborhoods) is the real key to the link between poverty and violence: most poor whites live in places that are less poor overall than the places most poor blacks live (in other words, they live closer to more wealthier people); most poor blacks live close to many other poor blacks.

But why should living closer to other poor people increase the likelihood of a poor person committing a violent crime? On the face of it, this seems like a rather ad hoc attempt at explanation: had we found that living in richer areas increases violent crime amongst the poor, it would have seemed just as natural to suppose that living in proximity to richer people both increases the relative indignity of being poor while surrounded by wealth, and increases the opportunities those poor persons have to commit crimes ‘worth’ committing. 

In Crime & Human NatureWilson and Hernnstein discuss a community that had high poverty and high population density and faced large amounts of racial discrimination, without concurrent high crime rates: “During the 1960s, one neighborhood in San Francisco had the lowest income, the highest unemployment rate, the highest proportion of families with incomes under $4,000 per year, the least educational attainment, the highest tuberculosis rate, and the highest proportion of substandard housing of any area of the city. That neighborhood was called Chinatown. Yet in 1965, there were only five persons of Chinese ancestry committed to prison in the entire state of California.

The low rates of crime among Orientals living in the United States was once a frequent topic of social science investigation. The theme of many of the reports that emerged was that crime rates were low not in spite of ghetto life but because of it. Though Orientals were the object of racist opinion and legislation, they were thought to have low crime rates because they lived in cohesive, isolated communities. The Chinese were for many years denied access to the public schools of California, not allowed to testify against whites in trials, and made the object of discriminatory taxation. The Japanese faced not only these barriers but in addition were “relocated” from their homes during World War II and sent to camps in the desert on the suspicion that some of them might have become spies or saboteurs.

There was crime enough in the nineteenth– and early–twentieth–century Oriental communities of California, but not in proportion to the Oriental fraction of the whole population. The arrest rate of Chinese and Japanese was higher in San Francisco than in any other California city during the 1920s, but even so Orientals were underrepresented by a factor of two, the Japanese more so than the Chinese. … What is striking is that the argument used by social scientists to explain low crime rates among Orientals—namely, being separate from the larger society—has been the same as the argument used to explain high rates among blacks. The experience of the Chinese and Japanese suggests that social isolation, substandard living conditions, and general poverty are not invariably associated with high rates of crime among racially distinct groups.”

So Tim Wise’s explanation really is deeply ad hoc and therefore fails, as well. Why would concentrated poverty lead to higher crime rates amongst blacks, but not amongst Asians? The answer must lie somewhere else.

In any case, there’s a detour worth taking here. One of Wise’s key citations is an essay by Johnson and Chanhatasilpa in Darnell Hawkins’ 2003 anthology, Violent Crime: Assessing Race and Ethnic Differences, and the mechanism by which they propose that concentrated poverty leads to crime is interesting. Pay close attention.

They open with a summary of previous research: “A community that shows collective and reciprocal willingness to combat crime and disorder (“you watch my back and I’ll watch yours”) will be far less likely than its spatial counterparts to experience crime …. social networks are the foundation of informal controls because they facilitate collective action through networks of friendship and kinship ties….” They introduce and define the term “community control” as “the capacity of communities to wield social control”, and they state the hypothesis that “structural disadvantages [such as concentrated poverty] increase homicide rates in communities through their deleterious impact on community control….” So how do they measure “community control”? They create their measurement out of three different things: “(1) the percentage of owner occupied housing units; (2) the rate of residential stability; and (3) the percentage of children living in husband–wife households.” (p.96)

Hold on just a second. The irony here is actually hilarious.

Tim Wise is on the record as attacking the notion that out–of–wedlock births play any part in social dysfunction in the black community because, as he explains, the actual birth rate amongst unmarried black women has fallen—it’s just fallen faster amongst married black women. And that means the percentage of births out–of–wedlock has risen, even though the actual number of births out–of–wedlock hasn’t. He’s right, but here is why that is still actually an idiotic objection: if the black community is becoming increasingly dysfunctional, what that means is that there are a greater percentage of dysfunctional individuals within the black community than there were before. And if it were true that single–parent families produced dysfunction, then a higher percentage of births to single parents absolutely would explain why the black community today is more dysfunctional, whether the absolute numbers fell or not. A smaller but more dysfunctional black community would still be a more dysfunctional black community.

The claim that out–of–wedlock birth is responsible for crime is, as we’ve seen, generally (though not completely) false. Tim Wise may object to it on the basis of an absurd fallacy that can be dispensed with in a single paragraph; but he does object to it—and yet his key citation for the claim that “concentrated poverty” is the real cause of crime actually argues that it does so, in large part … by increasing the percentage of out–of–wedlock births.  Did he not read far enough to notice that, or did he decide not to mention it to his audience on purpose?

Well, that strikes down one out of three measurements Johnson and Chanhatasilpa used in their essay—and Tim Wise would even presumably agree with me that the correlation between out–of–wedlock birth and crime is insufficient to prove that the former is cause of the latter. Further, we have overwhelmingly good reasons from elsewhere (twin studies, adoption studies, comparison of the children of divorced parents with children whose adoptive parents divorce) to conclude that it isn’t. It should be clear enough that a correlation between residential stability or home ownership and crime raises just exactly the same kinds of issues. Criminals are likely to be bad residents, and not only are bad residents far more likely to get themselves thrown out of their apartments, but non–criminals are likely to want to move away from them as well. Both of these effects would contribute to low rates of “residential stability”. What evidence can they provide that the effect of residential instability causing crime is stronger than the effect of crime causing residential instability? So far as I can tell, they have none.

And that brings us back to the research of Amir Sariaslan. The 2014 study previously mentioned controlled for familial confounding in the association between childhood income levels and adult criminality and substance abuse, and found that the association disappeared completely. A 2013 study conducted by Sariaslan and a similar team did the same thing for neighborhood deprivation and young adult criminality and substance abuse. Once again, the team found that when they “adjusted for unobserved familial confounders, the effect was no longer present…. Similar results were observed for substance misuse. … the adverse effect of neighbourhood deprivation on adolescent violent criminality and substance misuse in Sweden was not consistent with a causal inference. Instead, our findings highlight the need to control for familial confounding in multilevel studies of criminality and substance misuse.”

In other words, criminal behavior runs in families. And the association between poverty in childhood or in neighborhoods and crime disappears completely once this is controlled for. The vast majority of research on these questions in social science has simply ignored this and failed to control for familial confounding entirely.

So is the problem with criminal families genes or parenting? The real answer, at last.

Just as we described earlier that twin studies, adoption studies, and family studies all support the idea that the risk of divorce, and promiscuity in general, are heavily influenced by genes but influenced almost not at all by parenting, so the same thing goes for criminality. Biological children of criminals adopted into non–criminal adoptive homes have approximately the risk of becoming criminals as children born to criminal parents in general do, rather than the risk of becoming criminals that children raised by non–criminal parents in general do. And when we calculate how much more likely an identical twin is to have a criminal status similar to their twins’ and we compare that to the likelihood that a fraternal twin will have a criminal status similar to their twins’, not only do we get numbers that line up with exactly what we would expect if there were a genetic component at play, we get estimates of the heritability of criminal tendencies that lines up exactly with what was already being found by the adoption studies. And so on.

The truly important point here is this: we know that violent and criminal behavior are heritable, regardless of how extensive our knowledge is of what the particular genes are or how they make their contribution to criminality. We know this by the same means we know that everything else we know to be heritable is heritable: by studying whether adopted children become more like their adoptive or biological parents as they grow into adults, by measuring how much more similar identical twins are on given traits than fraternal twins, by measuring how similar identical twins who were raised apart are compared to random members of the population, and so on.

The twin studies find that: “Genetic factors, but not the common environment, significantly influenced whether subjects were ever arrested after age 15, whether subjects were arrested more than once after age 15, and later criminal behaviour. The common environment, but not genetic factors, significantly influenced early criminal behaviour. The environment shared by the twins has an important influence on criminality while the twins are in that environment, but the shared environmental influence does not persist after the individual has left that environment.” What this means is that while being raised by criminal parents might make a child more likely to commit a criminal action as a very young teen, it has zero impact on a child’s likelihood of becoming a criminal as a (young) adult. Meanwhile, exemplary adoption studies find that adoptive children with criminal biological mothers have a 50% chance of later criminal behavior, compared to just 5% for the adopted children of non–criminals. Again, some of the best introductions can be found at Quillette: How Criminologists Who Study Biology Are Shunned By Their Field, and Criminology’s Wonderland: Why (Almost) Everything You Know About Crime is Wrong.

It’s not necessarily clear just what is being inherited when criminal tendencies are passed on, and we’re far from any complete knowledge of the range of genes involved. However, science is increasingly closing in on the answers.

We have identified a variety of genes that influence biological features which we know to play a role in criminal behavior. We also know that some of these genes are present in different ethnic groups in almost exactly the proportions at which these populations are represented in violent crime (higher in blacks, and lower in Asians).

In 1993, we discovered a condition now known as Brunner syndrome. Brunner syndrome was first identified in a single Dutch family, all of whom were found to react to perceived provocation with extreme aggression; 5 were arsonists, 5 had been convicted of rape and/or murder. It turned out that all 14 males originally studied had a mutation that caused the complete eradication of an enzyme called MAOA, which is responsible for breaking down neurotransmitters inside of the brain, including dopamine and adrenaline. Other research soon confirmed that you could even knock this same gene out in mice and produce the similar kinds of aggression.

While Brunner’s syndrome is incredibly rare, with just three families across the world now known to contain victims of the disease, the rest of the human population has genes coding for either low, medium, or high levels of MAOA activity (either the 2–repeat, 3–repeat, or 4–repeat alleles, respectively). [Note: the established convention is to use the term “MAOA–L” to refer to either the 2–repeat or the 3–repeat genes, but by grouping the “low” and “medium” activities together, this convention underscores just how significant the difference between all three really is.]

Early research found that people with low–activity MAOA genes were more violent if they had difficult upbringings—but as the research continued, it confirmed that people with low–activity MAOA genes were indeed significantly more violent regardless of their childhood experiences. Other research, in fact, continued linking the same gene to things like credit card debt and even obesity—all behaviors which revolve around impulsiveness.

The 2–repeat version of the gene was found to double the risk of violent deliquency in young adulthood compared to the other two variants. And guess what? The 2R allele is found in “5.5% of Black men, 0.1% of Caucasian men, and 0.00067% of Asian men”—which just so happens to correspond eerily to ethnic rates of violent crime. And lest anyone worry that low activity MAOA genes merely correlates with violence because black Americans are more violent and also just coincidentally happen to have more of them, other research has looked at black Americans with and without low activity genes and still found substantially more violence in 2R carriers.

(For rebuttal of common criticisms of MAOA studies, see the archives of The Unsilenced Science).

Similarly, the potential “triggers” for someone with low–activity MAOA genes turning violent (particularly carriers of the 3–repeat, which is somewhat less associated with violence on its own) expanded to include testosterone—and testosterone levels differ by race as well. A 1986 study found that the “twofold difference in prostate cancer risk” between black and white men could be explained by the “15% higher testosterone level” found in Black men.

But circulating levels of testosterone are not the only variable of interest. Many other factors, including enzyme activity and hormone exposure in utero, influence the impact of circulating hormones as well—and on these measures, too, we find generally consistent patterns in which Black subjects have the most androgenic hormone profile while East Asian subjects have least, with White subjects somewhere inbetween. A 1992 study found that “white and black men had significantly higher values of 3 alpha, 17 beta androstanediol glucuronide (31% and 25% higher, respectively) and androsterone glucuronide (50% and 41% higher, respectively) than Japanese subjects”—these being enzymes that convert testosterone into the more physiologically active hormone DHT.

Even further, It Is Not Just About Testosterone tells us that: “Vasopressin synthesis and the aromatization into estradiol both serve to facilitate testosterone’s effects.” So, guess what? “Vasopressin secretion in normotensive black and white men and women on normal and low sodium diets” found that “24-h urinary excretion of vasopressin was significantly (P<0·05) higher in men than in women and higher (P<0·05) in black than in white subjects.” And other studies confirm that Black children are exposed to higher hormone levels in utero—this one found “higher testosterone [and] ratio of testosterone to SHBG … in African–American compared to white female neonates”.

This last study is very significant.

We know that hormone exposure in the womb has drastic impacts on future behavior: Girls with congenital adrenal hyperplasia, a condition that only briefly spikes the level of hormones a developing girl is exposed to, have significantly more masculine behavioral traits despite the fact that there is no evidence that parents treat them any differently, or that there is anything different about them or the way they are “socialized” other than excess prenatal male hormone exposure. As found in a 2003 study of “Prenatal androgens and gender-typed behavior”, girls with CAH “were more interested in masculine toys and less interested in feminine toys and were more likely to report having male playmates and to wish for masculine careers. Parents of girls with CAH rated their daughters’ behaviors as more boylike than did parents of unaffected girls. A relation was found between disease severity and behavior indicating that more severely affected CAH girls were more interested in masculine toys and careers. No parental influence could be demonstrated on play behavior, nor did the comparison of parents’ ratings of wished for behavior versus perceived behavior in their daughters indicate an effect of parental expectations. The results are interpreted as supporting a biological contribution to differences in play behavior between girls with and without CAH.”

There is no reason to think that if out–of–wedlock birth and violent crime were to correlate due to genes, this would have to be because both behaviors are influenced by the same genes. It could be that the separate genes which contribute separately to each behavior just happen to correlate as well, with people who carry the first set of genes often carrying the second. However, it is at least plausible that the factors briefly identified here (testosterone and MAOA) actually could play a common role in producing both violent behavior and out–of–wedlock birth.

To my knowledge, outside the finding that persons with Brunner’s syndrome can be prone to hypersexuality, MAOA has never been studied in relation to sociosexuality directly. However, it seems fairly safe to infer that the kind of impulsivity which would lead a person to rack up credit card debt, or eat their way to obesity, or commit impulsive violent crimes, would also leave them prone to impregnate someone they haven’t married or end up divorced. And as far as testosterone, the studies on that one are clear: “people’s orientations toward sexual relationships, in combination with their relationship status, are associated with individual differences in testosterone.” More specifically, in “chaste” individuals with a restricted sociosexual orientation, testosterone rises when single but falls after acquiring a partner—but this doesn’t happen for those with an “unrestricted” orientation: as this study describes it, “partnered men who reported greater desire for uncommitted sexual activity had testosterone levels that were comparable to those of single men; partnered women who reported more frequent uncommitted sexual behavior had testosterone levels that were comparable to those of single women.”

Beyond that, we know that psychopathy both has a biological basis (psychopaths have a lower physiological response to their environments; in other words, it takes more to stimulate them) and is heritable, and we know that psychopaths “are twenty to twenty-five times more likely than non-psychopaths to be in prison, four to eight times more likely to violently recidivate compared to non-psychopaths, and are resistant to most forms of treatment” with “93% of adult male psychopaths in the United States in prison, jail, parole, or probation.”

We also know psychopaths are more likely to seek casual sex and avoid relationships—inevitably producing greater out–of–wedlock birth rates. The evidence, then, that unstable childhood environments produces criminals is weak—while unstable environments may raise the risk of criminal behavior during childhood, that effect only barely lasts into adulthood, if at all. In contrast, the evidence that there are genes which predispose a person to commit violent crimes, produce children out of wedlock, and divorce, and that these are passed on genetically to the children produced by these relationships regardless of their upbringing, is very strong. The facts aren’t particularly favorable to religious social conservatives, liberals, or mens’ rights activists: poverty doesn’t seem to be the primary cause of crime, but neither is single–parenthood (for religious social conservatives) or a lack of fathers (for MRAs).

Could the rate of psychopathy differ by race as well? I don’t know, but I was able to find some small indication that it might: judgment and the ability to discern smells are both localized to the frontal lobes of the brain, and research has linked poor sense of smell to psychopathy and aggression. Meanwhile, other research finds that men on average have a worse sense of smell than women—and blacks on average have a worse sense of smell than whites. (Update: See Razib Khan’s discussion of Lynn’s 2002 paper ‘Racial and ethnic differences in psychopathic personality’ and Skeem’s 2004 critical meta–analysis ‘Are there ethnic differences in levels of psychopathy?’).

Another study, this time in Finns, found that in addition to MAOA–L, a mutation of another gene known as CDH13 was heavily linked to extreme violent crime—and the effect of combining the two genes was more than additive. Meanwhile, a study in white and Hispanic Americans linked CHD13 to “a younger age of sexual debut”.

To reiterate, we don’t fully understand what is being inherited when criminality is inherited. But we know that criminality is highly heritable, no matter how well we do or do not understand the mechanisms of that heritability, through the converging results of years of twin, adoption, and family studies which all produce the same conclusion. The truth of this knowledge does not depend on the relevance of MAOA, testosterone, or psychopathy genes in particular, although I happen to think that very strong cases can be made for all of them. Likewise, we don’t fully understand what it is that is being inherited when promiscuous tendencies are being inherited, but we know that promiscuity is highly heritable all the same. But even the cursory evidence that behavioral genetics has produced so far at this point in time suggests several known mechanisms that just might not only be the culprits, but might even explain why some behaviors (like out–of–wedlock birth) tend to correlate with others (like violent criminality).

Divorce and out–of–wedlock birth may produce behavioral problems, but for the most part sociosexual behavior in parents and children correlates because of genes, not experiences; and the behavioral problems that result from divorce and out–of–wedlock birth per se appear not to last beyond childhood. There is perhaps a tiny impact of poverty on property crime—but none on violent crime. Genes are not deterministic, but the strongest verifiable impact by far out of all measurable impacts is that of genetic heredity on behavior.

A few disclaimers would, in an ideal world, be able to go without saying: the vast majority of men are neither violent criminals nor psychopaths; likewise, the vast majority of black people are neither violent criminals nor psychopaths. Nowhere in any of this reasoning should license be taken for blanket prejudice against all men, or against all blacks. The baseline rate of risk matters. Even if a man (or black) is 10x more likely to murder you on the street than a woman (or white), if your actual risk of being murdered by a man (or black) is 0.0001% and your actual risk of being murdered on the street by a woman (or white) is 0.00001%, then this hardly justifies viewing all men (or blacks) with suspicion and giving all women (or whites) a free pass. It is a minority of all people who are violence–prone. It is a minority of all men who are violence–prone; and it is a minority of all blacks who are violence–prone. But the minority of men who are violence–prone appears to be larger than the minority of women who are violence–prone, and the minority of blacks who are violence–prone also appears to be larger than the minority of whites, which in turn appears to be larger than the minority of Asians, who are violence–prone. And though no one explanation reveals everything, the strongest explanation of all explanations we do have is hereditarian. If stating these facts makes me racist, then it apparently also makes me twice as sexist—against myself—because the gap between men and women in violent crime is even larger than the gap between blacks and whites.

Is there anything that we can do with this sort of information? Much of the resistance that will form against an explanations of undesirable social phenomena that gives genes a larger role than environment, I believe, comes from the impression that if genes are responsible, then there’s nothing we can do about it—it seems to be a recipe for resignation. And even if eugenics could, in theory, improve human outcomes and behavior, few of us would want to come anywhere near trusting the State with the amount of power it would take to attempt it (I certainly wouldn’t).

Fortunately, it isn’t true. I’ll be discussing the various ways in which the ordinary policies we already contemplate can be evaluated for their “eugenic” and “dysgenic” impacts in the future—and how the policies that would turn out to be most beneficial in light of this analysis fit neatly into neither “conservative” nor “liberal” boxes. For example, if IQ or conscientiousness are heritable traits, then establishing maternity leave—a stereotypical “liberal” ideal—may help to encourage women with higher IQs and conscientiousness to have more children, rather than forego having more children for the sake of their careers—thus helping increase the IQ and conscientiousness of the general population, and no specter of violent Nazi concentration camps need be feared. Some warrior gene researchers have suggested another idea they think that their evidence warrants: preventing former violent criminals from purchasing alcohol, because the association between MAOA–L and violence is often mediated by alcohol

Whether that proposal would actually be effective or not, it’s an excellent example of the kind of idea we can start to try to think about, if the case laid out here is true. For another example, we can use our knowledge of the relationship between criminal behavior and genes to base sentencing lengths around verifiable statistics on the risk of re–offense. If anyone is afraid that an idea like this could be prone to abuse, they should remember that the current system already is rife with abuse: judges give around 65% favorable sentences soon after either of their two daily food breaks, but after each break and before the next that number falls to nearly 0%. Alternatives to a system that condemns a person because of how recently a judge ate lunch can hardly get much worse.

Just as identifying that the environment is the cause of some phenomena can allow us to start figuring out interventions which help reduce the impact of that environmental cause, so identifying genes as the cause of some phenomena can allow us to start targeting interventions there, too. When it comes right down to it, the fear that acknowledging biological roots to human behavior must end in violent dystopia is simply bizarre, as soon as we consider the fact that so many of the worst massacres of the 20th century were committed by blank slatists who believed just exactly the contrary—that human nature could be reformed through social control to their will.

As Christopher Szabo at Intellectual Takeout asks, “Why are we so understanding towards the crimes of communism?” Including the death toll from famines, many of which were in fact engineered intentionally, the death rate from Maoist communism was about 1.92 million killed per year (across 38 years for a total of around 77,000,000). In contrast, the death rate from Hitler’s Reich was about 1.75 million killed per year (across 12 years for a total of around 21 million). And as Mao literally wrote, “In class society there is only human nature of a class character; there is no human nature above classes.” So if acknowledging a biological basis to human behavior is supposed to be discredited because it evokes the massacres of the Nazis, why shouldn’t denying it be discredited because it evokes the massacres of the Communists? I don’t seriously believe that sociologists who promote social constructionism should be tarred by association with Communist genocides, but unless you want to admit that that whole line of reasoning is bullshit, turnabout is fair play.




A Footnote to “Is the War on Drugs Racist?”

In that essay, I quoted Frank Zimring’s position on the impact of the war on drugs on violent crime as so: He also argues (pp.90–99), correlating hospitalizations and deaths from overdose with changes in the known street price, that overall use of cocaine appears to have remained relatively constant across the period of time in which New York City’s crime drop took place. Yet, he notes (pp.91–92) that “The peak rates of drug–involved homicide occurred in 1987 and 1988”—the same year that 70% of arrestees were found to test positive for cocaine—“and the drop in the volume of such killings is steady and steep from 1993 to 2005. … The volume of drug–involved homicides in 2005 is only 5% of the number in 1990.” Meanwhile, whereas 70% of arrestees in the late 1980s tested positive for cocaine, by 1991 (see table 2 on page 14) this number hit a low of 62%—and in 1998 it had fallen all the way to 47.1%. By 2012 (see figure 3.7 on page 45) this number fell even further to 25%.

What happened here? Why would drug use amongst arrestees fall if drug use as a whole remained constant? Zimring has an important answer: “If I’m a drug seller in a public drug market and you’re a drug seller in a public market, we’re both going to want to go to the corner where most of the customers are. But that means that we are going to have conflict about who gets the corner. And when you have conflict and you’re in the drug business, you’re generally armed and violence happens. … Policing … [helped drive] drug trade from public to private space. … [this] reduced the risk of conflict and violence associated with contests over drug turf. The preventive impact [of these policies] on lethal violence seems substantially greater than its impact on drug use. … [And] once the police had eliminated public drug markets in the late 1990s, the manpower devoted to a special narcotics unit [whose funding had increased by 137% between 1990 and 1999] dropped quite substantially [and yet the policies’ impacts on homicide rates remained].”

However, Zimring is clearly incorrect that the drug war reduced drug–involved homicides without reducing drug use as a whole—the drug war reduced drug use, too.

Quoting James Q. Wilson in the Wall Street Journal in 2011: “Another shift that has probably helped to bring down crime is the decrease in heavy cocaine use in many states. … Between 1992 and 2009, the number of admissions for cocaine or crack use fell by nearly two-thirds. In 1999, 9.8% of 12th-grade students said that they had tried cocaine; by 2010, that figure had fallen to 5.5%.

What we really need to know, though, is not how many people tried coke but how many are heavy users. Casual users who regard coke as a party drug are probably less likely to commit serious crimes than heavy users who may resort to theft and violence to feed their craving. But a study by Jonathan Caulkins at Carnegie Mellon University found that the total demand for cocaine dropped between 1988 and 2010, with a sharp decline among both light and heavy users. … Drug use among blacks has changed even more dramatically than it has among the population as a whole. As Mr. Latzer points out—and his argument is confirmed by a study by Bruce D. Johnson, Andrew Golub and Eloise Dunlap—among 13,000 people arrested in Manhattan between 1987 and 1997, a disproportionate number of whom were black, those born between 1948 and 1969 were heavily involved with crack cocaine, but those born after 1969 used very little crack and instead smoked marijuana.

The reason was simple: The younger African-Americans had known many people who used crack and other hard drugs and wound up in prisons, hospitals and morgues. The risks of using marijuana were far less serious. This shift in drug use, if the New York City experience is borne out in other locations, can help to explain the fall in black inner-city crime rates after the early 1990s.”

Thus, because “drug use among blacks has changed even more dramatically than it has among the population as a whole”, if the black:white ratio of those in prison for drug use is larger than the black:white ratio of drug users in the general population, this may be because much of the disproportionately black number of users of cocaine have already been arrested—to the benefit of the black population as a whole.

Similarly“In a recent article in the American Sociological Review, my colleagues and I [Gary LaFree] found that a proxy measure of crack cocaine had a greater impact on big city crime than more common measures like unemployment.”

A 1994 study by Eric Baumer found that “… arrestee cocaine use has a positive and significant effect on city robbery rates, net of other predictors. The effect of arrestee cocaine use on homicide is more modest … [but] cocaine use elevates city violent crime rates beyond levels expected on the basis of known sociodemographic determinants.” And a 1997 Justice Department study found that “there was a very strong statistical correlation between changes in crack use in the criminal population and homicide rates … In five of the six study communities, … homicide rates track quite closely with cocaine use levels among the adult male arrestee population.”


On “Privilege” in Prison Sentencing: A Crash at the Intersection

Throughout this series, I’ve shown interest in taking popular claims of “institutional” racism against blacks, and exploring evidence which reveals that contrary to common impressions, little to no racism is involved in the institution in question at all. In Are African–Americans Disproportionately Victimized By Police?, I provided an original analysis by asking exactly how many violent crimes white and black suspects have to commit before one white or black suspect interacting with police is shot—and there I found that, per crime and therefore per encounter with police, whites are in fact almost two times more likely to be shot by police than blacks. It turns out that black suspects only appear to be more likely to be shot by police because they are more likely to interact with police to begin with because they commit more violent crimes. Control for that, and the policing bias against blacks not only disappears; it actually does in fact reverse and count against whites. (And I’d like to note that even if you already knew that blacks commit more crimes than whites, it wasn’t a pre–given that these numbers would be sufficient to produce such a reversal—it could have been the case that part of the gap in police shootings was explained by interaction rates, and part by racist bias. But in fact, there is absolutely nothing left over for racist bias to account for once those numbers are ran; and both the raw numbers corrected for violent crime rates and the best experimental study data in fact reveal a bias against whites–or, in other words, in favor of blacks.)

Similarly, in Violence Against Women and Violence Against Truth, I exposed the absurdity of a mainstream feminist publication which claimed that the fact that rates of assault and murder have fallen relatively more for men than they have for women over the past few decades is evidence that “deep structural gender inequities … marginalize women” without making any acknowledgment of the fact that across recent history 3 out of 4 victims of violent crime have been men to begin with. If anyone is being “marginalized” here, it is men: even after these disproportionate reductions in victimization by violent crime of men and women which Ms. Magazine bemoans as women failing to have “anywhere near parity rights to physical freedom and security”, in 2013, for every 2198 men victimized by violent crime, there were only 2097 women. The fact that even mainstream feminist publications can get away with a claim like this shows just how dangerously distorted so–called “feminist” reasoning can be—and how little other self–proclaimed feminists are doing to correct this kind of reasoning when it appears within their midst.

So, I might have created the impression that my aim throughout these posts is simply to debunk all claims of racism or sexism whatsoever. In this post, I’d like to address an issue that gives me the opportunity to demonstrate that this is not the case—but continue demonstrating the solidity of my actual thesis: prison sentencing length.

The worldview of the left–wing social justice warrior could well be summarized by a speech given by the novelist John Scalzi in which he claimed that if life were a role–playing game, “white male” would quite simply be the easiest difficulty setting, whereas “minority female” would be equivalent to setting the difficulty to “hardcore.” According to this scheme, one person can still have more success on the higher difficulties than another person has on a lower one, but there is a linear and one–dimensional increase in penalty regardless of one’s individual skills and talents as one moves away from “white” and “male.”

The model I’d like to propose, in contrast to this, is that setting one’s race and gender in life is less like choosing a difficulty setting in a video game than it is like choosing a class: Mage, Archer, or Warrior?

Each character type will have different advantages and disadvantages against other character types in different contexts—and especially because each class is born in a different kind of environment in the first place, these differences can’t really be measured in a way that allows them to be ranked against each other on any single–dimensional scale. In one environment with high elevation weakening the strength of the mage’s connection to the earth and thus his power but with steep mountains making travel more difficult yet allowing the archer to perch himself atop a steep hill, the warrior may win against the mage and lose to the archer. But in another environment with level plains and a low elevation, the warrior may win against the archer who can’t find high ground on the plains, but lose to the mage whose connection to the earth is at its peak.

Which class is “better”? There really is no objective way to say.

Looking at things this way need not require us to dismiss all claims of discrimination against women and minorities—but it also allows us to reject the ridiculous view that white males are simply sitting snidely together atop a superficially measured social pyramid of unjust privilege. The question is then not whether there are ever any penalties for being non–white or female in any contexts or circumstances, but rather how they compare in scale to the penalties which can also exist for being white or male in others—and my claim is that once we perform this kind of analysis more rigorously than mainstream sociology traditionally has, these advantages and penalties are roughly similar: in other words, no one comes out as the obvious winner of the “Oppression Olympics”.

We have already seen that the popular view that whites have a profound advantage over blacks when facing police, in fact, has it exactly backwards: once the fact that black suspects simply do in fact commit more violent crime is taken into account, it turns out that any given black suspect facing police has an advantage over any given white—the white suspect is actually almost twice as likely to end up shot. Similarly, in part 3 of the “Is Dylann Roof ‘White Like Me?’” series I discussed the fact that not only are men more likely to be raped in prison than women are, but white men are much more likely to be raped than non–whites—and almost always by black men, according to mainstream sociological research stretching back for decades. Like it or not, these are “institutional” forms of suffering which give people facing police or spending time in prison penalties both for being white and/or male.

But the analogy of choosing a class in a role–playing game also allows me to clarify how my view differs from that of many conservatives, and of so–called mens’ rights activists, who share my interest in this same set of facts: in sum, these perspectives often simply invert John Scalzi’s argument, without rejecting its fundamentals. According to their schemes, being white or male work just the same way that being non–white or female do in the left–wing social justice warrior’s scheme: they give one disadvantages, plain and simple. My class analogy should make it clear that I consider this view misguided and wrong for exactly the same reasons that I consider the left–wing view misguided and wrong: no demographic is an unequivocal victim in modern society; the advantages and penalties faced in each case are simply different. So, this post will start with me granting the existence of a case of “institutional” racism.

This brings me back to the matter of sentencing lengths. Once the cops have already shown up, and once we’ve already arrived at a conviction, how long are the sentences that different people tend to receive for similar crimes? Is there a penalty (or privilege) for the convicted criminal based on his (or her) race?

From the looks of it, there is. A 2013 study conducted by the U.S. Sentencing Commission found that blacks’ sentences were about 15%  longer than whites’. Much of the early research which found much higher disparities than this failed to take account of the fact that black defendants, on average, have longer rap sheets—which clearly factors in to judicial decisions. A 2012 study conducted by Sonja Starr which controlled meticulously for previous record found a gap of about 10%, to the disadvantage of blacks. Another 2013 study conducted by Beaver, et al. found “no evidence of racial discrimination in criminal justice processing” once matching defendants for self–reported lifetime violence and IQ—so perhaps what often happens is that defendants of lower intelligence behave differently in the courtroom, and judges pass sentences in response to these behavioral differences, not race itself.

But what often happens needn’t be what always happens, and it would be hasty to dismiss the entire literature on the sentencing gap out of hand without a deeper investigation. In a review of the evidence compiled for the time between 1980 and 2000, Tushar Kansal of The Sentencing Project writes that “32 state-level studies contained 95 estimates—meaning 95 different ways in which these studies sought to determine whether sentencing decisions were biased—of the direct effect of race on sentence severity[…, and] 43.2% [of these] indicated harsher sentences for blacks… 8 studies of the federal system contained 22 estimates of the direct relationship between race and sentence severity[…, and] over two-thirds (68.2%) [of these] indicated harsher sentences for blacks ….” In other words, for every ~9 state or federal studies which fail to find a racial disparity against blacks, there are 11 studies which do. The most plausible explanation is that rather than it being a coincidence that 55% of all studies find a racial disparity against blacks, with 45% mostly finding null results, there is a racial disparity—in some but not all times and places. Kansal concludes with an admission that “Despite the findings of the cited studies in the area of direct racial discrimination, a number of factors indicate that the presence of direct discrimination is not uniform and extensive. Some of the state-level studies found no evidence of direct racial discrimination, and many of those that did find evidence of direct discrimination concluded that it exercised relatively modest effects, increasing the likelihood of a minority being sentenced to prison by only a few percentage points. ”

For obtaining an estimate of how large the typical racial sentencing gap is, then, Sonja Starr’s 2013 finding of a 10% gap between blacks and whites represents the middle ground—less than the U.S. Sentencing Commission’s 15%; but more than the few percentage point, null, or reverse (that is, favoring blacks and disfavoring whites—of which there were 6) findings reported by 44.3% of the Sentencing Project review’s collection of studies.

Now, what happens when we ask the same question about gender?

Intersectional theory” is the term for the attempt within sociology to account for the fact that race and gender contribute interacting effects to one’s social disadvantages, rather than investigating each factor in isolation. The consensus is, of course, that being non–white and being non–male are always interacting disadvantages—so that in reality, as one Tumblr author who describes herself as a “genderfluid femme Black mixed bitch” writes, what intersectionality was really always about is “exposing the ways Black women are caught up in multiple systems of oppression … it is meant to help Black women understand their experiences in a white supremacist patriarchal culture [e.g., which is set up to privilege whites and men and therefore especially white men and punish non–whites and non–males and therefore especially minority women] like the U.S.”

The real findings, however, take the intersectional logic and turn that expectation flat on its head.

From 2001–2006, three studies ([1], [2], [3]) all found a gender sentencing gap of about 10%.

But guess who that gap favored?


Already, the gender sentencing gap favoring women is at least as large as the racial sentencing gap—and therefore just as important.  The second of these studies, published by Max Schanzenbach, notes something interesting: “The findings regarding gender in the case of serious offenses are quite striking: the greater the proportion of female judges in a district, the lower the gender disparity for that district … These results are hard to square with the suggestion that unobserved accomplice status or blameworthiness is behind the gender disparity. [However,] appointing more black judges to the bench is unlikely to reduce sentencing disparities for black offenders who commit serious crimes.” In other words, the three takeaways here are: (1) the theory that men organize society to create distinctly male privileges appears in the case of prison sentencing to be absolute bullshit, because when men are given power to hand out sentences, they privilege women, not men; (2) it is unlikely that this sentencing gap is a result of other unobserved variables differentiating the crimes committed by women and men, because male judges give female offenders privileges in sentencing that female judges don’t; and (3) this does not appear to be the case for black/white judges handing out sentences to black/white offenders—black and white judges do not appear to sentence black or white defendants very differently, which at least suggests that most black and white judges are usually responding to the severity of crimes committed when they sentence black and white defendants differently.

By these measures, again, the gender gap in sentencing favoring women is exactly as large as the racial gap in sentencing favoring whites. However, in the Booker vs. United States ruling passed down by the Supreme Court in 2005, it was decided that sentencing guidelines requiring that men and women who committed the same crimes and held similar criminal records be given equally long sentences were to be considered recommendations, rather than requirements—and the evidence shows that the gender sentencing gap grew as a result of this decision. A 2007 study by Supriya Sarnikar  found an even larger figure than the aforementioned three: “We find that women receive prison sentences that average a little over 2 years less than those awarded to men. Even after controlling for circumstances such as the severity of the offense and past criminal history, women receive more lenient sentences. Approximately 9.5 months of the female advantage cannot be explained by gender differences in individual circumstances. In other words if women faced the same sentencing structure as men, women would on average receive 15.4 months less prison time than men rather than 24.9 months less prison time.”

But importantly, Sarnikar makes this note in her conclusion: “[O]ur data permit us to examine only the end stage of the criminal justice system. A more comprehensive treatment would take account of the fact that before arriving at the judge for sentencing, a defendant must also pass through a jury or possible plea bargain with a prosecutor.”

And that brings us back to Sonja Starr.

In 2012, Starr decided to present an analysis of the gender sentencing gap which accounted for decisions made in the earlier stages of the criminal justice system, just as she had done recently to arrive at her 10% figure for the racial sentencing gap. Her study incorporated data collected from the U.S. Marshals’ Service (USMS), the Executive Office of U.S. Attorneys (EOUSA), the Administrative Office of the U.S. Courts (AOUSC), and the U.S. Sentencing Commission (USSC) spanning from 2001–2009 (containing periods both before and after the equal sentencing for equal crime guidelines were ruled recommended rather than mandatory) and controlled extensively for the severity of crimes committed and the previous criminal records of offenders.

The result? On average, men receive a sentence that is 60% longer than the one a woman would receive for the same crime—and Starr notes that even this number is, in fact, an underestimate, because the average male prison sentence is driven downwards by the presence of men receiving relatively short sentences in cases where women would receive no sentence at all. But already, this is six times as large as the  racial sentencing gap of 10% which Sonja Starr had recently found by applying exactly the same methodology to race.

To be specific, the gender gap varies by race: among white offenders, men receive sentences 51% longer than womens’; among black offenders, men receive sentences 74% longer than womens’. But whereas this drags black mens’ sentences down compared to white mens’, it also pulls black womens’ back up—with the end result that black women have the largest advantage in prison sentencing length of all! So much for intersectionality as a method for “exposing the ways Black women are caught up in multiple systems of oppression … in a white supremacist patriarchal culture.” How long before we hear social justice warriors asking black women to check their prison sentencing privilege?

To help visualize this, I’ve created a very oversimplified chart to represent the rough amount of “privilege” each group receives in prison sentencing by starting everyone at 500 points, creating a racial gap of 50 points (10% of 500; so add 25 points if white, and substract 25 points if black), a white gender gap of 255 points (51% of 500; so add 127 points if white female, and subtract 127 points if white male), and a black gender gap of 370 points (74% of 500; so add 185 points if black female, and subtract 185 points if black male). Black men get 290 points; white men get 398; white women get 652; and black women get 660. To translate that into years, if four people all committed identical crimes, then a black man would spend 7 years and 7 months in jail, a white man would spend 6 years and 8 months in jail, a white woman would spend 4 years and 7 months in jail, and a black woman would spend 4 years and 6 months in jail. Lest you think I’ve calculated this incorrectly and arrived at the finding that black women receive the most “privilege” in sentencing by mistake, note that on p.16 of her study (under section 3.6. Race-Gender Interactions), Sonja Starr notes that “among women, the race gap [is] reversed in sign.”

ChartGo (2)

What this means, even given the fact that the gender gap is larger amongst black offenders than it is amongst whites, is that black men in the criminal justice system share more in  oppression with white men on account of being male than they share in oppression with black women thanks to their race—and white men share more in oppression with black men on account of being male than they share in privilege with white women thanks to their race. If we want to improve the situation of black men in prison, then we should want to make their sentences more like black womens’ rather than more like white mens’.

To describe the implications still yet a few other ways, white men and white women are not united in shared privileges—white women and black women are; whereas it is white men and black men who, relatively speaking, are united in shared “oppression”. Rather than finding white men sitting atop the social pyramid, we find black women there instead. And even for black men, clearly the most disadvantaged group of all, addressing the sentencing length penalty for being male should be a much higher priority—about six times higher—than addressing the sentencing length penalty for being black. So in this case, whose efforts do black men need most? Supposedly anti–racist “intersectional” feminists would have it that the racial disparity should be everyone’s priority—but if mens’ rights activists were to instead successfully eliminate the gender sentencing gap first, this would benefit actual living and breathing black men much more than eliminating the racial sentencing gap.

Even where so–called “intersectional feminists” have something right (that the effects of race, gender, and other variables interact), they’ve still failed to bring us the whole picture. While nominally professing concern for the fact that “the patriarchy hurts men too”, it seems as though they would often rather fight wage gaps that don’t exist (except as the result of voluntary career choices and preferences) and narcissistically find ways to make men becoming somewhat less of a majority of the victims of violence a womens’ problem than work to bring to the public’s attention that the relatively little–known gender gap disfavoring men in prison sentencing is six times larger than the well known racial gap disfavoring blacks.

Perhaps that’s because that evidence so drastically undermines the “white male is the easiest difficulty setting; black female is ‘hardcore’” model of American social life by, at least in this instance, turning it all the way around on its head. So, not only are individual white men who commit a crime more likely to be shot by police than individual black men who commit a crime—and not only are white men in jail more likely to be raped (most likely by black men)—but even at the level of sentencing, white men are almost as disadvantaged in sentencing compared to both black and white women as black men are. And even here where black men are the most disadvantaged of all, and that is an “institutional” problem that should be addressed, they would still be more helped by an effort to reduce the gender penalty which also hits white men than they would by an effort to reduce the racial penalty, which only trivially hits black women (who are still despite that penalty by far the most privileged anyway).

One conclusion discoveries like these has brought me to is that it should now be perfectly reasonable to have college classes, think tanks, or student groups dedicated to examining the problems faced by whites, or men—and I stand by that even though I think much that is, or would be, produced by such groups is bunk. Why? First, because a great deal of what is produced by womens’ studies and minority studies is already bunk—but at least a variety of different focus groups could criticize each other well enough to exercise checks and balances against each others’ ideological excesses and help bring awareness of flawed arguments made by the other side to our attention.

But second, because just as there are valid issues faced by women and minorities even amidst the trash often produced by “feminist” and “anti–racist” groups and publications, so there are valid issues faced by men and whites amidst all the trash that “mens’ rights” or white–focused groups would inevitably also produce. Men (or whites) are not “the” oppressed segment of society without exception today any more than women (or non–whites) are. But the types of “oppression” faced by each of these groups are not massively different in kind—and were it not for the fact that feminist and anti–racist organizations alone are allowed to dominate the narrative airways and spread perspectives which are often flawed in ways that few will take the time to deeply examine and of which few people will ever even hear well–informed critiques, the suggestion that groups with opposing perspectives have a valid place would not sound so absurd. The reason we don’t know or hear about the facts which might go some way to validate their existence is because they aren’t here to give them to us. “Intersectional” anti–racist feminists can’t be trusted to do it sufficiently; and that goes for black men, too, whose need for the kind of analysis of the gender sentencing gap that feminists aren’t doing here is even more grave than white mens’ need for it is.