Friday, December 2, 2011

Blacks with Bullets Embedded in Bone

nce upon a time, a people now known as African Americans left plantations of slavery to live in relative peace among white people. Unfortunately, the mean, villainous whites schemed systems of organized crime that allowed them to enrich themselves into tranquil opulence, while they excluded the African Americans from their illegal get-rich connivances. Amidst their impoverishment, the African Americans experienced a psychological transformation that forced many of them to become violent criminals. The End.

This story emerges from the works cited by Steven Pinker in his bestselling, and possibly Pulitzer Prize-winning book, The Better Angels of Our Nature. Pinker writes,

Figure 3-14 shows the rates for two cities in which black-on-black and white-on-white homicides can be distinguished. The graph reveals that the racial disparity in American homicide has not always been with us.

Here is the figure:

He cites Eric Monkkonen’s Murder in New York City, which says,

Essentially, crime was more profitable for whites, whereas even in crime, discrimination blocked the more lucrative opportunities for African Americans and encouraged more violent, destructive offenses.

Pinker’s characterization of homicide data is another example of his distortions on matters pertaining to race and violence. See my previous post on how Pinker flubbed his summary of the decades of research on the best understood violence gene, including his citation of a copy-and-paste error. Much of his data for New York City in the above graph, including all years of equal homicide rates, comes from American History by Randolph Roth, who plotted a similar graph shown below and made clear that the data plotted actually do not represent known intraracial homicide assailants, but just homicide victims.

In fact, Roth further qualified the data for Philadelphia:

It may prove that blacks accounted for a greater proportion of homicide perpetrators than victims in Philadelphia, but determining that will require more data, including a tally of blacks and whites killed in riots. Blacks committed homicide there at over twice the rate whites did, 1839-1846, according to indictment records (6.5 per 100,000 adults per year versus 2.7), because they were more likely to commit both intraracial and interracial homicide. It may be that Philadelphia’s antiblack riots, the disfranchisement of blacks in Pennsylvania, and increased competition with whites for unskilled jobs were beginning to take their toll.

Roger Lane also questioned Philadelphia’s homicide victimhood data because coroners’ records “proved unreliable.” Coroners frequently failed to indicate that a death was a homicide. Ted Robert Gurr thought this problem might exist in other American cities of that era, but no one had bothered to check as Lane had done in Philadelphia.

Roth did provide actual estimates of New York City homicides, which doubled the very small racial gap in victimization. For Philadelphia, Pinker was not clear whether and for what years he made use of Roth’s data. Roth estimated that, from 1839 to 1846, Philadelphia’s African-American rate of homicide victimization was 28 percent higher than that of whites.

Monkkonen provided some context for the period of purported New York City homicide racial equality. First, he advised caution about the quality of the data. “Even though we can outline this demographic history of black New Yorkers, we cannot yet establish precise population measures… Thus, any construction of rates with at-risk denominators must be understood as estimates.” How unfortunate that this important historic moment of violence equality was so poorly documented.

Generalizing New York City to the rest of America has other complicating factors. Monkkonen offered estimates that suggest that racial homicide equality resulted from enormous white ethnic inequality. “My best estimate for the decade centered on 1860,” he wrote, “sets the Irish homicide rate at 37.5 per 100,000 adult males, the German rate at 15.7, and the black rate at 32.” About half of New York City’s population was immigrant in 1860. It is ironic that Pinker is allowing the violent tendencies of New York Irish-American immigrants to characterize the behavior of American whites in general because English Americans took to calling them “white Negroes.”

Lastly, the age demographics of these groups differed. In 1850, Irish immigrant males in New York City were 19 percent more likely to be in the potentially more violent age range of 16 to 45 than African-American men, and, despite being half as homicidal, German immigrant men were 73 percent more likely than African-American men to be in that age range. Monkkonen does not entirely attribute the eventual growth of a large racial homicide gap to poverty because New York City experienced a massive growth of young black men, such that in 1940 only 19 percent of African-American New York City men in this age range were born there.

Another recently published study, which Pinker did not consider, about this period described a qualitative distinction of racial patterns of violence. Last year, Carlina de la Cova published the first large-scale analysis of its kind, reviewing 651 male cadavers “of low socioeconomic status born between 1825 and 1877” from Cleveland, St. Louis, and Washington DC. Though these remains are apparently not of homicide victims, they do offer evidence of racial differences in violence from that era, but only among the poor, as none owned property, and most were transient. Thus, the study should bias towards exaggerating white delinquency with a sample more unrepresentative of whites than African Americans.

Although fully 92 percent of all cadavers had skeletal fractures, in every category of fracture, whites had more. The smallest differences were for injuries that were most plausibly from serious fights. 50 percent of whites had cranial fractures, compared to 48 percent of African Americans. 44.8 percent of whites had nasal fractures, compared to 42.3 percent of African Americans. Differences for vertebrae fracture rates were much greater, at 72.7 percent for whites and 35.8 percent for African Americans. De la Cova considered osteoporosis as a possible explanation. She noted that African Americans have less osteoporosis, greater bone mass, denser cortical bone, and lower bone resorption. Indeed, black people have greater bone mineral density from infancy, and hormonal levels contribute to such racial differences, even in childhood,. (Recall that the media accused research oncologist Kathy Albain of “racial medical profiling” for discovering that racial disparities in cancer mortality remain only in cancers related to sex hormones after accounting for zip code.) Of course, skin color affects the pathways of the hormone vitamin D. De la Cova checked for an influence of osteoporosis and age of death, but that negative result does not negate the fact that black people have stronger bones even in comparisons of whites and blacks without osteoporosis. This seems to extend to tooth loss, which de la Cova also utilized in her attempt to support differing rates of fighting. Even in the present-era National Health and Nutrition Examination Survey-III, in which whites were better educated, white people still had a higher rate of tooth loss than African Americans and Mexican Americans. In fact, having less education was significantly associated with tooth loss in white people, but not African Americans and Mexican Americans.

Even so, de la Cova points to historical accounts of popular and “ritualized” pugilism among poor white men to settle disputes and protect their “honor,” which logically might play less of a role in the violent behavior of a systematically oppressed minority.

On the other hand, black men comprised all eight of the gunshot wounds, most having a bullet embedded in a bone. Though this was a small number of the cadavers, historical accounts and African-American newspaper advertisements also reflected a gun-oriented African-American culture. Perhaps their early association with guns made fistfights an untenable means of resolving conflicts.

The US Census Bureau did not start tracking crime until 1880, and despite being a minority, African Americans were receiving 54 percent of unlawfully concealed weapons charges and 75 percent of exhibiting a deadly weapon charges and were committing 83 percent of shootings, 85 percent of stabbings, 48 percent of manslaughter homicides, and 43 percent of murders.

Perhaps the best argument for an era of relative homicide equality would be the well-documented, enormous widening of the racial homicide gap in the years that followed. Gurr provided an illustrative graph of this for Washington DC.

Here is Jeffrey Adler’s homicide graph for Chicago when African Americans were a tiny minority of the population.

Societal changes can influence patterns of violence, and these years certainly qualified as profoundly transformative for African Americans. However, given that African Americans have higher allele frequencies than whites of all or (debatably) nearly all identified violence genes, anyone who advertises a distant past of black peaceableness deserves an inspection of their data. I suggest this should even be so for a beloved Harvard celebrity professor.

Adler, Jeffrey S. 2006. First in Violence, Deepest in Dirt: Homicide in Chicago, 1875-1920. Cambridge, Mass.: Harvard University Press.

De La Cova C (2010). Cultural patterns of trauma among 19th-century-born males in cadaver collections. American Anthropologist, 112 (4), 589-606 PMID: 21132946

Gurr, T. (1981). Historical Trends in Violent Crime: A Critical Review of the Evidence Crime and Justice, 3 DOI: 10.1086/449082

Hui SL, Dimeglio LA, Longcope C, Peacock M, McClintock R, Perkins AJ, & Johnston CC Jr (2003). Difference in bone mass between black and white American children: attributable to body build, sex hormone levels, or bone turnover? The Journal of clinical endocrinology and metabolism, 88 (2), 642-9 PMID: 12574194

Jimenez M, Dietrich T, Shih MC, Li Y, & Joshipura KJ (2009). Racial/ethnic variations in associations between socioeconomic factors and tooth loss. Community dentistry and oral epidemiology, 37 (3), 267-75 PMID: 19302573

Lane, Roger. 1979. Violent Death in the City: Suicide, Accident, and Murder in Nineteenth-Century Philadelphia. Cambridge, Mass.: Harvard University Press.

Monkkonen, Eric H. 2001. Murder in New York City. Berkeley: University of California Press.

Pinker, Steven. 2011. The Better Angels of Our Nature: Why Violence Has Declined. New York, New York: Viking Penguin.

Roth, Randolph. 2009. American Homicide. Cambridge, Mass.: Harvard University Press.


hailtoyou said...

From the entry:
"The US Census Bureau did not start tracking crime until 1880, and despite being a minority, African Americans were receiving 54 percent of unlawfully concealed weapons charges and 75 percent of exhibiting a deadly weapon charges and were committing 83 percent of shootings, 85 percent of stabbings, 48 percent of manslaughter homicides, and 43 percent of murders."

Blacks were 13.1% of the USA in 1880, according to [See here, opens a .xls].

The above-cited crime patterns by race seem to have been remarkably constant from 1880 on through the present day.

hailtoyou said...

A possible explanation for why Black crime may have been 'aritificially' low in NYC in the mid-1800s (perhaps applicable to similar places as well) :

First of all, we see children of Black soldiers do much better on standardized tests than non-military Blacks' children do. Why? As Steve Sailer likes to point out, the military swears by IQ testing (or a close proxy for it). The bottom fifth (is it?) of the scorers are flatly not admitted into the U.S. military at all, and and the second-to-bottom quintile is seldom admitted. This leaves the pool of Blacks admitted to the military decently-smart compared to the USA's Black average. In other words, military Blacks were selected for intelligence.

Something similar may have happened in NYC in the 1800s, with the proverbial gate shut to the lower elements of the Black population:

I found an odd pattern in the census data, as I was revieweing it following your wonderful post here. [Data from Census, opens a .xls].

NYC Population By Race and Year

NYC Whites
1820: 113,000
1870: 929,000
=3.03 doublings

NYC Blacks
1820: 11,000
1870: 13,000
=0.18 doublings

Phenomenal White growth (over three doublings in 50 years) and essentially no net-Black-growth. (1820-1870 White growth even beats the Ellis-Island boom, 1880-1930, when NYC's population saw only 2.44 doublings, by my calcuation. There would have been some natural constraints on growth by that point, i.e. lack of space in NYC, so maybe this is an unfair comparison). What explains such heavy Black emigration out of NYC in the decades straddling the U.S. Civil War? (TFRs being high in those days, Blacks would have had to emigrate in considerable numbers to maintain zero net growth in NYC). Possibly, social pressure from the Whites forced-away the lower elements of the Black population of NYC that had been there earlier. (2,400 Black slaves lived in NYC in 1790, along with 1,100 free Blacks). In those days, such a thing was possible. I don't know. I'd like to hear a better explanation for this.

Notice that according to Pinker's graph, Black crime really took off in the 1880s -- which would be more clear if he had used a more honest scale -- exactly when Black population growth began in NYC.

For comparison, a similar but weaker trend in population growth by race is seen in Pinker's other example city (where evidence for the phenomenon Pinker points to -- a posited long-lost equality in Black and White crime rates -- is also, as it happens, weaker; that is to say, the racial disparity in crime in Philadelphia was greater throughout the post-Civil-War era than it was in NYC):

Philadelphia Blacks
1820: 8,000
1870: 22,000
=1.4 doublings

Philadelphia Whites
1820: 56,000
1870: 652,000
=3.5 doublings

nooffensebut said...

Excellent analysis. Eric Monkkonen attempted to address the issue you raise about African-American emigration. Here is the full quotation:

"The city’s African American population had a very different history from its immigrants—one of declension in size from 16,000 in 1840, to about 12,000 in 1860, to perhaps less than 10,000 in 1865. By midcentury, less than 2 percent of the city’s population was black. This demographic trajectory, so different for the rest of the city’s people, tells us much about the city’s inhospitality to African Americans. Freedom, yes; friendship, no.

Traditionally, African American men had worked on the city’s docks, in its shipyards, and in various service occupations. The new immigrants competed vigorously, and sometimes violently, for these jobs. This story culminated tragically in 1863 in the New York City Draft Riots. No matter how one interprets these riots, they always contain a base element of racism. Angered by the efforts to draft poor and immigrant workers, the city’s white immigrant men quickly turned their initial political protest into a race war against African Americans. A large but never reliably established number of black residents died—many as victims of vicious public lynchings. Not too surprisingly, after the riots the city’s black population decreased even faster. Novelist Peter Quinn brilliantly depicts the African American exodus. Even though we can outline this demographic history of black New Yorkers, we cannot yet establish precise population measures, and we can expect that census enumerations were highly inaccurate. Thus, any construction of rates with at-risk denominators must be understood as estimates."

sykes.1 said...

Could we stop calling them "African Americans." I object to the "American" part.

Marmalade said...

I've seen various analyses of data. In rural Louisiana after the Civil War and before Jim Crow, blacks had a lower homicide rate than whites, both interracial and intraracial. In many of the big cities during the late 1800s and early 1900s, many ethnic whites (Italians, Irish, etc) committed higher rates of homicide than blacks.

Much of the homicide rates seem correlated with lead exposure. There have been a number of major violent crime spikes over the past century.

The first correlated to increased lead use in rural areas starting in the 1800s, which showed up in the crime data around 1900. Following that, there was the increase of industrialization and urbanization that led to a massive spike of violence. The third spike is the most well known. It followed after lead being added to gasoline and ended following leaded gasoline having been regulated.

There tends to be higher rates of violence with populations with higher lead toxicity rates. Many whites left urban areas for suburbia where lead toxicity was low, and their earlier high violent crime rates dropped. We don't have lead toxicity measurements for the earliest populations, but we do have more recent data. The lower classes have higher rates than the upper classes. Minorities have higher rates than whites. Poor minorities have higher rates than poor whites. Blacks have higher rates than Hispanics.

A large part of this has to do with more bypasses and toxic dumps having been placed in poor black communities. It's unsurprising that populations surrounded by more lead have higher rates of lead exposure and hence toxicity. It's also unsurprising that these populations show all the predictable known results of lead toxicity: stunted brain development and lower average IQ, impulse control problems and ADHD, aggressive behavior, etc; there are also many health conditions as well. I don't know why anyone pretends to be shocked that people with poisoned brains have social and behavioral problems.

Two of the comments mentioned that black crime began going up in the 1880s. Besides that being when widespread lead usage began, that is also when Jim Crow began. Also, it was around the time when blacks began to experience increasing levels of racial oppression: chain gang re-enslavement, sundown towns, redlining, lynchings, bombings, arson, etc.

It doesn't take a genius to see the obvious connections.