The American homicide rate crept up until 1933, nose-dived in the 1930s and 1940s, remained low in the 1950s, and then was launched skyward in 1962, bouncing around in the stratosphere in the 1970s and 1980s before returning to earth starting in 1992. The upsurge in the 1960s was shared with every other Western democracy, and I'll return to it in the next section. But why did the United States start the century with homicide rates so much higher than England's, and never close the gap? Could it be a counterexample to the generalization that countries with good governments and good economies enjoy a civilizing process that pushes their rate of violence downward? And if so, what is unusual about the United States? In newspaper commentaries one often reads pseudo-explanations like this: "Why is America more violent? It's our cultural predisposition to violence." How can we find our way out of this logical circle? It's not just that America is gun-happy. Even if you subtract all the killings with firearms and count only the ones with rope, knives, lead pipes, wrenches, candlesticks, and so on, Americans commit murders at a higher rate than Europeans.

The Black Panther spokesman may have mixed up his fruits, but he did express a statistically valid generalization about the United States. Among Western democracies, the United States leaps out of the homicide statistics. Instead of clustering with kindred peoples like Britain, the Netherlands, and Germany, it hangs out with toughs like Albania and Uruguay, close to the median rate for the entire world. Not only has the homicide rate for the United States not wafted down to the levels enjoyed by every European and Commonwealth democracy, but it showed no overall decline during the 20th century, as we see in figure 3-10. (For the 20th-century graphs, I will use a linear rather than a logarithmic scale.)

Violence is as American as cherry pie.-- H. Rap Brown

Sources: Graph from Monkkonen, 2001, pp. 171, 185-88; see also Zahn & McCall, 1999, p. 12. Note that Monkkonen's U. S. data differ slightly from the FBI Uniform Crime Reports data plotted in figure 3-18 and cited in this chapter.

FIGURE 3-10: Homicide rates in the United States and England, 1900-2000

Europeans have always thought America is uncivilized, but that is only partly true. A key to understanding American homicide is to remember that the United States was originally a plural noun, as in these United States. When it comes to violence, the United States is not a country; it's three countries. Figure 3-11 is a map that plots the 2007 homicide rates for the fifty states, using the same shading scheme as the world map in figure 3-9.

Source: Data from U. S. Federal Bureau of Investigation, 2007, table4. Crime in the United States by Region, Geographical Division, and State, 2006-7

FIGURE 3-11: Geography of homicide in the United States, 2007

A second contrast is less visible on the map. Louisiana's homicide rate is higher than those of the other southern states, and the District of Columbia (a barely visible black speck) is off the scale at 30.8, in the range of the most dangerous Central American and southern African countries. These jurisdictions are outliers mainly because they have a high proportion of African Americans. The current black-white difference in homicide rates within the United States is stark. Between 1976 and 2005 the average homicide rate for white Americans was 4.8, while the average rate for black Americans was 36.9. It's not just that blacks get arrested and convicted more often, which would suggest that the race gap might be an artifact of racial profiling. The same gap appears in anonymous surveys in which victims identify the race of their attackers, and in surveys in which people of both races recount their own history of violent offenses. By the way, though the southern states have a higher percentage of African Americans than the northern states, the North-South difference is not a by-product of the white-black difference. Southern whites are more violent than northern whites, and southern blacks are more violent than northern blacks.

The shading shows that some of the United States are not so different from Europe after all. They include the aptly named New England states, and a band of northern states stretching toward the Pacific (Minnesota, Iowa, the Dakotas, Montana, and the Pacific Northwest states), together with Utah. The band reflects not a common climate, since Oregon's is nothing like Vermont's, but rather the historical routes of migration, which tended to go from east to west. This ribbon of peaceable states, with homicide rates of less than 3 per 100,000 per year, sits at the top of a gradient of increasing homicide from north to south. At the southern end we find states like Arizona (7.4) and Alabama (8.9), which compare unfavorably to Uruguay (5.3), Jordan (6.9), and Grenada (4.9). We also find Louisiana (14.2), whose rate is close to that of Papua New Guinea (15.2).

So while northern Americans and white Americans are somewhat more violent than Western Europeans (whose median homicide rate is 1.4), the gap between them is far smaller than it is for the country as a whole. And a little digging shows that the United States did undergo a state-driven civilizing process, though different regions underwent it at different times and to different degrees. Digging is necessary because for a long time the United States was a backwards country when it came to keeping track of homicide. Most homicides are prosecuted by individual states, not by the federal government, and good nationwide statistics weren't compiled until the 1930s. Also, until recently "the United States" was a moving target. The lower forty-eight were not fully assembled until 1912, and many states were periodically infused with a shot of immigrants who changed the demographic profile until they coalesced in the melting pot. For these reasons, historians of American violence have had to make do with shorter time series from smaller jurisdictions. In American Homicide Randolph Roth has recently assembled an enormous number of small-scale datasets for the three centuries of American history before the national statistics were compiled. Though most of the trends are roller coasters rather than toboggan runs, they do show how different parts of the country became civilized as the anarchy of the frontier gave way -- in part -- to state control.

Sources: Data for England: Eisner, 2003. Data for New England: 1630-37, Roth, 2001, p. 55; 1650-1800: Roth, 2001, p. 56; 1914: Roth, 2009, p. 388. Roth's estimates have been multiplied by 0.65 to convert the rate from per-adults to per-people; see Roth, 2009, p. 495. Data representing a range of years are plotted at the midpoint of the range.

FIGURE 3-12: Homicide rates in England, 1300-1925, and New England, 1630-1914

Figure 3-12 superimposes Roth's data from New England on Eisner's compilation of homicide rates from England. The sky-high point for colonial New England represents Roth's Elias-friendly observation that "the era of frontier violence, during which the homicide rate stood at over 100 per 100,000 adults per year, ended in 1637 when English colonists and their Native American allies established their hegemony over New England." After this consolidation of state control, the curves for old England and New England coincide uncannily.

The rest of the Northeast also saw a plunge from triple-digit and high-double-digit homicide rates to the single digits typical of the world's countries today. The Dutch colony of New Netherland, with settlements from Connecticut to Delaware, saw a sharp decline in its early decades, from 68 to 15 per 100,000 (figure 3-13). But when the data resume in the 19th century, we start to see the United States diverging from the two mother countries. Though the more rural and ethnically homogeneous parts of New England (Vermont and New Hampshire) continue to hover in the peaceful basement beneath 1 in 100,000, the city of Boston became more violent in the middle of the 19th century, overlapping cities in former New Netherland such as New York and Philadelphia.

Sources: Data from Roth, 2009, whites only. New England: pp. 38, 62. New Netherland: pp. 38, 50. New York: p. 185. New Hampshire and Vermont: p. 184. Philadelphia: p. 185. Data representing a range of years are plotted at the midpoint of the range. Estimates have been multiplied by 0.65 to convert the rate from per-adults to per-people; see Roth, 2009, p. 495. Estimates for "unrelated adults" have been multiplied by 1.1 to make them approximately commensurable with estimates for all adults.

FIGURE 3-13: Homicide rates in the northeastern United States, 1636-1900

The zigzags for the northeastern cities show two twists in the American version of the Civilizing Process. The middling altitude of these lines along the homicide scale, down from the ceiling but hovering well above the floor, suggests that the consolidation of a frontier under government control can bring the annual homicide rate down by an order of magnitude or so, from around 100 per 100,000 to around 10. But unlike what happened in Europe, where the momentum continued all the way down to the neighborhood of 1, in America the rate usually got stuck in the 5-to-15 range, where we find it today. Roth suggests that once an effective government has pacified the populace from the 100 to the 10 range, additional reductions depend on the degree to which people accept the legitimacy of the government, its laws, and the social order. Eisner, recall, made a similar observation about the Civilizing Process in Europe.

The other twist on the American version of the Civilizing Process is that in many of Roth's mini-datasets, violence increased in the middle decades of the 19th century. The buildup and aftermath of the Civil War disrupted the social balance in many parts of the country, and the northeastern cities saw a wave of immigration from Ireland, which (as we have seen) lagged behind England in its homicide decline. Irish Americans in the 19th century, like African Americans in the 20th, were more pugnacious than their neighbors, in large part because they and the police did not take each other seriously. But in the second half of the 19th century police forces in American cities expanded, became more professional, and began to serve the criminal justice system rather than administering their own justice on the streets with their nightsticks. In major northern cities well into the 20th century, homicide rates for white Americans declined.

Sources: New York 1797-1845: Roth, 2009, p. 195. New York 1856-85: Average of Roth, 2009, p. 195, and Gurr, 1989a, p. 39. New York 1905-53: Gurr, 1989a, p. 39. Philadelphia: 1842-94: Roth, 2009, p. 195. Philadelphia 1907-28: Lane, 1989, p. 72 (15-year averages). Philadelphia, 1950s: Gurr, 1989a, pp. 38-39. Roth's estimates have been multiplied by 0.65 to convert the rate from per-adults to per-people; see Roth, 2009, p. 495. His estimates for Philadelphia were, in addition, multiplied by 1.1 and 1.5 to compensate, respectively, for unrelated versus all victims and indictments versus homicides (Roth, 2009, p. 492). Data representing a range of years are plotted at the midpoint of the range.

FIGURE 3-14: Homicide rates among blacks and whites in New York and Philadelphia, 1797-1952

But the second half of the 19th century also saw a fateful change. The graphs I have shown so far plot the rates for American whites. 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. In the northeastern cities, in New England, in the Midwest, and in Virginia, blacks and whites killed at similar rates throughout the first half of the 19th century. Then a gap opened up, and it widened even further in the 20th century, when homicides among African Americans skyrocketed, going from three times the white rate in New York in the 1850s to almost thirteen times the white rate a century later. A probe into the causes, including economic and residential segregation, could fill another book. But one of them, as we have seen, is that communities of lower-income African Americans were effectively stateless, relying on a culture of honor (sometimes called "the code of the streets") to defend their interests rather calling in the law.

The first successful English settlements in America were in New England and Virginia, and a comparison of figure 3-13 and figure 3-15 might make you think that in their first century the two colonies underwent similar civilizing processes. Until, that is, you read the numbers on the vertical axis. They show that the graph for the Northeast runs from 0.1 to 100, while the graph for the Southeast runs from 1 to 1,000, ten times higher. Unlike the black-white gap, the North-South gap has deep roots in American history. The Chesapeake colonies of Maryland and Virginia started out more violent than New England, and though they descended into the moderate range (between 1 and 10 homicides per 100,000 people per year) and stayed there for most of the 19th century, other parts of the settled South bounced around in the low 10-to-100 range, such as the Georgia plantation counties shown on the graph. Many remote and mountainous regions, such as the Georgia backcountry and Tennessee-Kentucky border, continued to float in the uncivilized 100s, some of them well into the 19th century.

Why has the South had such a long history of violence? The most sweeping answer is that the civilizing mission of government never penetrated the American South as deeply as it had the Northeast, to say nothing of Europe. The historian Pieter Spierenburg has provocatively suggested that "democracy came too early" to America. In Europe, first the state disarmed the people and claimed a monopoly on violence, then the people took over the apparatus of the state. In America, the people took over the state before it had forced them to lay down their arms -- which, as the Second Amendment famously affirms, they reserve the right to keep and bear. In other words Americans, and especially Americans in the South and West, never fully signed on to a social contract that would vest the government with a monopoly on the legitimate use of force. In much of American history, legitimate force was also wielded by posses, vigilantes, lynch mobs, company police, detective agencies, and Pinkertons, and even more often kept as a prerogative of the individual.

FIGURE 3-15: Homicide rates in the southeastern United States, 1620-1900

Sources: Data from Roth, 2009, whites only. Virginia (Chesapeake): pp. 39, 84. Virginia (Chesapeake and Shenandoah): p. 201. Georgia: p. 162. Tennessee-Kentucky: pp. 336-37. Zero value for Virginia, 1838, plotted as 1 since the log of 0 is undefined. Estimates have been multiplied by 0.65 to convert the rate from per-adults to per-people; see Roth, 2009, p. 495.

This power sharing, historians have noted, has always been sacred in the South. As Eric Monkkonen puts it, in the 19th century "the South had a deliberately weak state, eschewing things such as penitentiaries in favor of local, personal violence." Homicides were treated lightly if the killing was deemed "reasonable," and "most killings… in the rural South were reasonable, in the sense that the victim had not done everything possible to escape from the killer, that the killing resulted from a personal dispute, or because the killer and victim were the kinds of people who kill each other."

The South's reliance on self-help justice has long been a part of its mythology. It was instilled early in life, such as in the maternal advice given to the young Andrew Jackson (the dueling president who claimed to rattle with bullets when he walked): "Never… sue anyone for slander or assault or battery; always settle those cases yourself." It was flaunted by pugnacious icons of the mountainous South like Daniel Boone and Davy Crockett, the "King of the Wild Frontier." It fueled the war between the prototypical feuding families, the Hatfields and McCoys of the Kentucky-West Virginia backcountry. And it not only swelled the homicide statistics for as long as they have been recorded, but has left its mark on the southern psyche today.

Self-help justice depends on the credibility of one's prowess and resolve, and to this day the American South is marked by an obsession with credible deterrence, otherwise known as a culture of honor. The essence of a culture of honor is that it does not sanction predatory or instrumental violence, but only retaliation after an insult or other mistreatment. The psychologists Richard Nisbett and Dov Cohen have shown that this mindset continues to pervade southern laws, politics, and attitudes. Southerners do not outkill northerners in homicides carried out during robberies, they found, only in those sparked by quarrels. In surveys, southerners do not endorse the use of violence in the abstract, but only to protect home and family. The laws of the southern states sanction this morality. They give a person wide latitude to kill in defense of self or property, put fewer restrictions on gun purchases, allow corporal punishment ("paddling") in schools, and specify the death penalty for murder, which their judicial systems are happy to carry out. Southern men and women are more likely to serve in the military, to study at military academies, and to take hawkish positions on foreign policy.

There is one thing I must explain, because I feel I must be honest and I want no misunderstandings. I have been convicted of a felony, namely manslaughter. You will probably want an explanation for this before you send me an application, so I will provide it. I got into a fight with someone who was having an affair with my fiancée. I lived in a small town, and one night this person confronted me in front of my friends at the bar. He told everyone that he and my fiancée were sleeping together. He laughed at me to my face and asked me to step outside if I was man enough. I was young and didn't want to back down from a challenge in front of everyone. As we went into the alley, he started to attack me. He knocked me down, and he picked up a bottle. I could have run away and the judge said I should have, but my pride wouldn't let me. Instead I picked up a pipe that was laying in the alley and hit him with it. I didn't mean to kill him, but he died a few hours later at the hospital. I realize that what I did was wrong.

In a series of ingenious experiments, Nisbett and Cohen also showed that honor looms large in the behavior of individual southerners. In one study, they sent fake letters inquiring about jobs to companies all over the country. Half of them contained the following confession:

The other half contained a similar paragraph in which the applicant confessed to a felony conviction for grand theft auto, which, he said, he had foolishly committed to help support his wife and young children. In response to the letter confessing to the honor killing, companies based in the South and West were more likely than those in the North to send the letter-writer a job application, and their replies were warmer in tone. For example, the owner of one southern store apologized that she had no jobs available at the time and added:

No such warmth came from companies based in the North, nor from any company when the letter confessed to auto theft. Indeed, northern companies were more forgiving of the auto theft than the honor killing; the southern and western companies were more forgiving of the honor killing than the auto theft.

As for your problem of the past, anyone could probably be in the situation you were in. It was just an unfortunate incident that shouldn't be held against you. Your honesty shows that you are sincere… I wish you the best of luck for your future. You have a positive attitude and a willingness to work. Those are the qualities that businesses look for in an employee. Once you get settled, if you are near here, please stop in and see us.

Nisbett and Cohen also captured the southern culture of honor in the lab. Their subjects were not bubbas from the bayous but affluent students at the University of Michigan who had lived in the South for at least six years. Students were recruited for a psychology experiment on "limited response time conditions on certain facets of human judgment" (a bit of gobbledygook to hide the real purpose of the study). In the hallway on their way to the lab, the students had to pass by an accomplice of the experimenter who was filing papers in a cabinet. In half of the cases, when the student brushed past the accomplice, he slammed the drawer shut and muttered, "Asshole." Then the experimenter (who was kept in the dark as to whether the student had been insulted) welcomed the student into the lab, observed his demeanor, gave him a questionnaire, and drew a blood sample. The students from the northern states, they found, laughed off the insult and behaved no differently from the control group who had entered without incident. But the insulted students from the southern states walked in fuming. They reported lower self-esteem in a questionnaire, and their blood samples showed elevated levels of testosterone and of cortisol, a stress hormone. They behaved more dominantly toward the experimenter and shook his hand more firmly, and when approaching another accomplice in the narrow hallway on their way out, they refused to step aside to let him pass.

Is there an exogenous cause that might explain why the South rather than the North developed a culture of honor? Certainly the brutality needed to maintain a slave economy might have been a factor, but the most violent parts of the South were backcountry regions that never depended on plantation slavery (see figure 3-15). Nisbett and Cohen were influenced by David Hackett Fisher's Albion's Seed, a history of the British colonization of the United States, and zeroed in on the origins of the first colonists from different parts of Europe. The northern states were settled by Puritan, Quaker, Dutch, and German farmers, but the interior South was largely settled by Scots-Irish, many of them sheepherders, who hailed from the mountainous periphery of the British Isles beyond the reach of the central government. Herding, Nisbett and Cohen suggest, may have been an exogenous cause of the culture of honor. Not only does a herder's wealth lie in stealable physical assets, but those assets have feet and can be led away in an eyeblink, far more easily than land can be stolen out from under a farmer. Herders all over the world cultivate a hair trigger for violent retaliation. Nisbett and Cohen suggest that the Scots-Irish brought their culture of honor with them and kept it alive when they took up herding in the South's mountainous frontier. Though contemporary southerners are no longer shepherds, cultural mores can persist long after the ecological circumstances that gave rise to them are gone, and to this day southerners behave as if they have to be tough enough to deter livestock rustlers.

So it's sufficient to assume that settlers from the remote parts of Britain ended up in the remote parts of the South, and that both regions were lawless for a long time, fostering a culture of honor. We still have to explain why their culture of honor is so self-sustaining. After all, a functioning criminal justice system has been in place in southern states for some time now. Perhaps honor has staying power because the first man who dares to abjure it would be heaped with contempt for cowardice and treated as an easy mark.

The herding hypothesis requires that people cling to an occupational strategy for centuries after it has become dysfunctional, but the more general theory of a culture of honor does not depend on that assumption. People often take up herding in mountainous areas because it's hard to grow crops on mountains, and mountainous areas are often anarchic because they are the hardest regions for a state to conquer, pacify, and administer. The immediate trigger for self-help justice, then, is anarchy, not herding itself. Recall that the ranchers of Shasta County have herded cattle for more than a century, yet when one of them suffers a minor loss of cattle or property, he is expected to "lump it," not lash out with violence to defend his honor. Also, a recent study that compared southern counties in their rates of violence and their suitability for herding found no correlation when other variables were controlled.

There was the mahogany landscape, the florid-faced, blue-eyed roughriders, the prim pretty schoolteacher arriving in Roaring Gulch, the rearing horse, the spectacular stampede, the pistol thrust through the shivered windowpane, the stupendous fist fight, the crashing mountain of dusty old-fashioned furniture, the table used as a weapon, the timely somersault, the pinned hand still groping for the dropped bowie knife, the grunt, the sweet crash of fist against chin, the kick in the belly, the flying tackle; and immediately after a plethora of pain that would have hospitalized a Hercules, nothing to show but the rather becoming bruise on the bronzed cheek of the warmed-up hero embracing his gorgeous frontier bride.

The American West, even more than the American South, was a zone of anarchy until well into the 20th century. The cliché of Hollywood westerns that "the nearest sheriff is ninety miles away" was the reality in millions of square miles of territory, and the result was the other cliché of Hollywood westerns, ever-present violence. Nabokov's Humbert Humbert, drinking in American popular culture during his cross-country escape with Lolita, savors the "ox-stunning fisticuffs' of the cowboy movies:

In Violent Land, the historian David Courtwright shows that the Hollywood horse operas were accurate in the levels of violence they depicted, if not in their romanticized image of cowboys. The life of a cowboy alternated between dangerous, backbreaking work and payday binges of drinking, gambling, whoring, and brawling. "For the cowboy to become a symbol of the American experience required an act of moral surgery. The cowboy as mounted protector and risk-taker was remembered. The cowboy as dismounted drunk sleeping it off on the manure pile behind the saloon was forgotten."

In the American Wild West, annual homicide rates were fifty to several hundred times higher than those of eastern cities and midwestern farming regions: 50 per 100,000 in Abilene, Kansas, 100 in Dodge City, 229 in Fort Griffin, Texas, and 1,500 in Wichita. The causes were right out of Hobbes. The criminal justice system was underfunded, inept, and often corrupt. "In 1877," notes Courtwright, "some five thousand men were on the wanted list in Texas alone, not a very encouraging sign of efficiency in law enforcement." Self-help justice was the only way to deter horse thieves, cattle rustlers, highwaymen, and other brigands. The guarantor of its deterrent threat was a reputation for resolve that had to be defended at all costs, epitomized by the epitaph on a Colorado grave marker: "He Called Bill Smith a Liar." One eyewitness described the casus belli of a fight that broke out during a card game in the caboose of a cattle train. One man remarked, "I don't like to play cards with a dirty deck." A cowboy from a rival company thought he said "dirty neck," and when the gunsmoke cleared, one man was dead and three wounded.

All and everybody, this is my claim, fifty feet on the gulch, cordin to Clear Creek District Law, backed up by shotgun amendments… Any person found trespassing on this claim will be persecuted to the full extent of the law. This is no monkey tale butt I will assert my rites at the pint of the sicks shirter if leagally necessary so taik head and good warning.

Courtwright cites an average annual homicide rate at the time of 83 per 100,000 and points to "an abundance of other evidence that Gold Rush California was a brutal and unforgiving place. Camp Names were mimetic: Gouge Eye, Murderers Bar, Cut-Throat Gulch, Graveyard Flat. There was a Hangtown, a Helltown, a Whiskeytown, and a Gomorrah, though, interestingly, no Sodom." Mining boom towns elsewhere in the West also had annual homicide rates in the upper gallery: 87 per 100,000 in Aurora, Nevada; 105 in Leadville, Colorado; 116 in Bodie, California; and a whopping 24,000 (almost one in four) in Benton, Wyoming.

It wasn't just cowboy country that developed in Hobbesian anarchy; so did parts of the West settled by miners, railroad workers, loggers, and itinerant laborers. Here is an assertion of property rights found attached to a post during the California Gold Rush of 1849:

In figure 3-16 I've plotted the trajectory of western violence, using snapshots of annual homicide rates that Roth provides for a given region at two or more times. The curve for California shows a rise around the 1849 Gold Rush, but then, together with that of the southwestern states, it shows the signature of the Civilizing Process: a greater-than-tenfold decline in homicide rates, from the range of 100 to 200 per 100,000 people to the range of 5 to 15 (though, as in the South, the rates did not continue to fall into the 1s and 2s of Europe and New England). I've included the decline for the California ranching counties, like those studied by Ellickson, to show how their current norm-governed coexistence came only after a long period of lawless violence.

FIGURE 3-16: Homicide rates in the southwestern United States and California, 1830-1914

Sources: Data from Roth, 2009, whites only. California (estimates): pp. 183, 360, 404. California ranching counties: p. 355. Southwest, 1850 (estimate): p. 354. Southwest, 1914 (Arizona, Nevada, and New Mexico): p. 404. Estimates have been multiplied by 0.65 to convert the rate from per-adults to per-people; see Roth, 2009, p. 495.

Anarchy was not the only cause of the mayhem in the Wild West and other violent zones in expanding America such as laborers' camps, hobo villages, and Chinatowns (as in, "Forget it, Jake; it's Chinatown"). Courtwright shows that the wildness was exacerbated by a combination of demography and evolutionary psychology. These regions were peopled by young, single men who had fled impoverished farms and urban ghettos to seek their fortune in the harsh frontier. The one great universal in the study of violence is that most of it is committed by fifteen-to-thirty-year-old men. Not only are males the more competitive sex in most mammalian species, but with Homo sapiens a man's position in the pecking order is secured by reputation, an investment with a lifelong payout that must be started early in adulthood.

So at least five of the major regions of the United States -- the Northeast, the middle Atlantic states, the coastal South, California, and the Southwest -- underwent civilizing processes, but at different times and to different extents. The decline of violence in the American West lagged that in the East by two centuries and spanned the famous 1890 announcement of the closing of the American frontier, which symbolically marked the end of anarchy in the United States.

The violence of men, though, is modulated by a slider: they can allocate their energy along a continuum from competing with other men for access to women to wooing the women themselves and investing in their children, a continuum that biologists sometimes call "cads versus dads." In a social ecosystem populated mainly by men, the optimal allocation for an individual man is at the "cad" end, because attaining alpha status is necessary to beat away the competition and a prerequisite to getting within wooing distance of the scarce women. Also favoring cads is a milieu in which women are more plentiful but some of the men can monopolize them. In these settings it can pay to gamble with one's life because, as Daly and Wilson have noted, "any creature that is recognizably on track toward complete reproductive failure must somehow expend effort, often at risk of death, to try to improve its present life trajectory." The ecosystem that selects for the "dad" setting is one with an equal number of men and women and monogamous matchups between them. In those circumstances, violent competition offers the men no reproductive advantages, but it does threaten them with a big disadvantage: a man cannot support his children if he is dead.

The West was eventually tamed not just by flinty-eyed marshals and hanging judges but by an influx of women. The Hollywood westerns' "prim pretty schoolteacher[s] arriving in Roaring Gulch" captures a historical reality. Nature abhors a lopsided sex ratio, and women in eastern cities and farms eventually flowed westward along the sexual concentration gradient. Widows, spinsters, and young single women sought their fortunes in the marriage market, encouraged by the lonely men themselves and by municipal and commercial officials who became increasingly exasperated by the degeneracy of their western hellholes. As the women arrived, they used their bargaining position to transform the West into an environment better suited to their interests. They insisted that the men abandon their brawling and boozing for marriage and family life, encouraged the building of schools and churches, and shut down saloons, brothels, gambling dens, and other rivals for the men's attention. Churches, with their coed membership, Sunday morning discipline, and glorification of norms on temperance, added institutional muscle to the women's civilizing offensive. Today we guffaw at the Women's Christian Temperance Union (with its ax-wielding tavern terrorist Carrie Nation) and at the Salvation Army, whose anthem, according to the satire, includes the lines "We never eat cookies 'cause cookies have yeast / And one little bite turns a man to a beast." But the early feminists of the temperance movement were responding to the very real catastrophe of alcohol-fueled bloodbaths in maledominated enclaves.

Another biological contribution to frontier violence was neurobiological rather than sociobiological, namely the ubiquity of liquor. Alcohol interferes with synaptic transmission throughout the cerebrum, especially in the prefrontal cortex (see figure 8-3), the region responsible for self-control. An inebriated brain is less inhibited sexually, verbally, and physically, giving us idioms like beer goggles, roaring drunk, and Dutch courage. Many studies have shown that people with a tendency toward violence are more likely to act on it when they are under the influence of alcohol.

The idea that young men are civilized by women and marriage may seem as corny as Kansas in August, but it has become a commonplace of modern criminology. A famous study that tracked a thousand low-income Boston teenagers for forty-five years discovered that two factors predicted whether a delinquent would go on to avoid a life of crime: getting a stable job, and marrying a woman he cared about and supporting her and her children. The effect of marriage was substantial: three-quarters of the bachelors, but only a third of the husbands, went on to commit more crimes. This difference alone cannot tell us whether marriage keeps men away from crime or career criminals are less likely to get married, but the sociologists Robert Sampson, John Laub, and Christopher Wimer have shown that marriage really does seem to be a pacifying cause. When they held constant all the factors that typically push men into marriage, they found that actually getting married made a man less likely to commit crimes immediately thereafter. The causal pathway has been pithily explained by Johnny Cash: Because you're mine, I walk the line.

An appreciation of the Civilizing Process in the American West and rural South helps to make sense of the American political landscape today. Many northern and coastal intellectuals are puzzled by the culture of their red state compatriots, with their embrace of guns, capital punishment, small government, evangelical Christianity, "family values," and sexual propriety. Their opposite numbers are just as baffled by the blue staters' timidity toward criminals and foreign enemies, their trust in government, their intellectualized secularism, and their tolerance of licentiousness. This so-called culture war, I suspect, is the product of a history in which white America took two different paths to civilization. The North is an extension of Europe and continued the court- and commerce-driven Civilizing Process that had been gathering momentum since the Middle Ages. The South and West preserved the culture of honor that sprang up in the anarchic parts of the growing country, balanced by their own civilizing forces of churches, families, and temperance.