Pieter Spierenburg

AHR Forum

Democracy Came Too Early: A Tentative Explanation for the Problem of American Homicide

Eric Monkkonen was a long-time friend. On scholarly issues we often agreed but sometimes disagreed, and I regret that he will not have the opportunity to react to the argument presented here. His article forms a good starting point for me to present my ideas about the United States' experience with violence—a subject of great interest to me.¹ There is no need to take issue with the empirical data that Monkkonen discusses, which few if any historians would contest.² When it comes to interpretation, context, and explanation, he and I converge on some points but diverge on others. I share with him a firm belief that history provides an indispensable key for understanding manifestations of violence in any present-day society. Indeed, if there has ever been a problem with which historians are able to help out sociologists and criminologists, it is probably this one. Secondly—and here some scholars voice dissenting opinions—I agree that quantitative research into murder trends constitutes a necessary base for further analysis. Of all past crimes, homicide is the only one that allows for reliable estimates of its actual incidence, and any discussion of context, culture, and state institutions loses much of its meaning when the bare rates for the area or period under study are unknown.

I am convinced, with Monkkonen, that murder is more than just an index to itself. He is cautious in calling the supposition that societies with a high incidence of homicide also witness a high level of assaults no more than a reasonable guess. This supposition can of course be tested, certainly for contemporary societies. In 1998, for example, the FBI counted 16,910 homicides, as against 93,100 rapes, 446,630 robberies, and 974,400 cases that the police classified as aggravated assault.³ This constitutes an aggravated assault rate—to confine myself to this crime—for the United States of approximately 345 incidents per 100,000 persons. A Dutch study found that the number of people visiting the emergency room at the Groningen academic hospital after being assaulted increased from 210 in 1970 to 524 in 1991.⁴ The corresponding rates per 100,000 are 37.5 and 94. While it is an open question to what extent this group of patients matched that of Americans who suffered aggravated assault, the research method used is likely to capture more cases than police figures do. During the period under study, homicide rates for the Netherlands as a whole were seven to sixteen times lower than for the United States. Surprisingly, perhaps, a similar count of wounded patients can be made for early modern Europe. In some places, notably in Italian cities, it was customary to register surgeons' treatments of injuries, and occasionally these registers have been preserved. In his study of crime in sixteenth-century Rome, Peter Blastenbrei presents data on murder as well as on nonlethal violence. Homicide rates fluctuated between 31 and 74.5 per 100,000, with an average of 47.3, while the annual rate of persons visiting a surgeon after a fight was about 200 per 100,000.⁵ Admittedly, this is lower than the rate of aggravated assault in the United States, whereas we would have expected it to be higher given Rome's elevated murder rate. The actual figure of assaults in the Papal City, however, may have been higher than the number reported, because Blastenbrei emphasizes that registration was probably incomplete, and moreover, Roman men were reluctant to admit to a surgeon that an injury had been caused by assault.

Divergence is implied in Monkkonen's casual treatment of the issue of personal honor. In my view, honor—or respect or reputation—constitutes a crucial intermediate factor in many societies between macro structures, on the one hand, and individual aggressive inclinations, on the other. I refer to a specific type of male honor here, because honor is both gendered and subject to change. Secondly, I disagree with the statement that Norbert Elias's theory of civilizing processes is unsatisfactory when applied to the United States. To the contrary, this theory will lead me toward a tentative explanation for America's elevated rates of homicide. Incidentally, this means that I also reject the criticism of Elias's work by Barbara Rosenwein in this journal.⁶ Before outlining my application of Elias's theory and my views about honor, I will deal with a final point of divergence: Although I agree that we need to simplify in order to facilitate large comparisons, I have some doubts about the way in which Monkkonen organizes his data.

In particular, I am bothered by Monkkonen's use of a dividing line around 1850. This is not because of what comes after. Indeed, at that time the United States, like Europe, was witnessing industrialization, the rise of consumer capitalism, urbanization, extension of the vote, and the establishment of police forces. I have no problem, moreover, with one national society's being compared with a group of nations, since the initial problem was the United States' uniqueness among economically advanced democratic countries.⁷ What bothers me is the disparity with respect to the pre-1850 period: there is a lot more history in the European than in the American case! Significantly, Monkkonen refrains from giving a rough characterization of the era before his dividing line, and the evidence he discusses for the United States is basically confined to the first half of the nineteenth century. By 1800, Europe had undergone an evolution of more than half a millennium, in which homicide rates had declined in successive stages. As early as the seventeenth century, France had a centralized monarchy, and England and the Netherlands boasted urbanized economies. It was a long and twisted road from the thirteenth-century communes to the 1848 revolutions. Of course, Monkkonen knew all this, but apart from citing medieval homicide rates, he did not wonder about its implications. What, for example, did "America" stand for in the early seventeenth century? If it stood for anything, it was a plethora of native communities, of which we now know a good deal but whose homicide rates we can never hope to reconstruct. By contrast, I would contend that, with embryonic state structures emerging, European homicide rates can serve as a meaningful basis for the study of violence from about 1300 onward.

Hence, the problem of where to draw the line, which at first glance seems to be strictly a methodological issue, has repercussions for historical theory. In a programmatic article about the very long term, the Dutch sociologist Johan Goudsblom distinguishes three world-historical phases in the monopolization of physical force.⁸ In his first stage, adult males monopolized force by excluding women and children from using it. The beginnings of this stage probably coincided with the differentiation between hunting as a male activity and gathering as a female one. During the second phase, physical force was monopolized by a more limited group of arms-bearing warriors, who excluded, among others, peasants and priests from using it. Goudsblom calls the social figurations associated with this stage "military-agrarian societies." Much later still, relatively autonomous warrior elites increasingly had to yield to larger organizations. Force was monopolized within the framework of the institutions that have come to be called states. All specialists in violence were either incorporated into the state or eliminated. Although we can observe this ongoing process at an early date in several regions of the world, by the early sixteenth century it was independently gaining momentum in northwestern Europe and Japan. Practices such as feuding, exercising private justice, and keeping armed retainers were increasingly banned. The monopolization referred to always remained relative, since assaults and homicides, thereafter called crimes, imply a breach of the state's monopoly. For the present discussion, the most relevant conclusion from this scheme is that a neat distinction between interpersonal (or private) violence and state violence can be made only from the beginning of the third stage (large-scale collective violence in periods of revolution or unrest can be considered an intermediate category). And for a quantitative study of homicide to be meaningful, we absolutely need to be able to make that distinction.

It might be debated when, exactly, parts of North America entered the third phase, but surely this was not yet the case during most of the colonial period—neither in, say, Massachusetts nor in the Iroquois federation. With respect to the period before 1850, this means that Monkkonen's concentration on the first half of the nineteenth century is both right and wrong. It is right to the extent that colonial murder rates by themselves have a limited meaning. The homicide rate among European American adults in eighteenth-century New England may have been low, but the killing of Indians, in largely unknown numbers, contributed to the region's overall violence.⁹ The concentration on the first half of the nineteenth century is wrong to the extent that it takes insufficient account of differences between the United States and Europe in the pace of macro-level change. The North American continent moved into the modern world almost overnight, without going through the extended development that characterized European societies. The implications for violence in the United States will be clear in a moment. It should be added that regions such as the Far West entered the modern world well after 1850.¹⁰

The West and, even more so, the South also were honor-prone societies. For the South, this hardly needs to be pointed out, after the pioneering studies by Bertram Wyatt-Brown and the works of Edward Ayers and Kenneth Greenberg.¹¹ For the West, we must read between the lines. "Homicide in the American West ... usually resulted from some minor disputes with an acquaintance." All triggers of quarrels, we hear, were "minor issues" as well. When the reason for a fight seems minor or trivial to a modern investigator, this usually indicates that honor played its part. An unblemished outward appearance matters in an honor culture. Clare McKanna recognizes this explicitly when referring to the black population of Omaha. "They [blacks from the South] brought with them an exaggerated sense of honor that would not tolerate a careless comment, a jostle on the street or a derogatory gesture. Such behavior could bring a quick violent response."¹² This finding echoes the consensus among historians that the black population of the South had adopted the whites' sense of honor by the end of the nineteenth century. Some carried it with them as they moved westward, but most likely this cultural code arrived in the West by other routes as well. In Europe, by the first half of the nineteenth century, the traditional code, according to which an honorable man must exhibit virility and courage and avenge insults with physical attack, was confined to marginal regions such as the Mediterranean and the Balkans. And transformations of honor constitute a crucial intermediate factor between the development of interpersonal violence and state formation processes.

The nature of state formation processes is the primary focus of my hypothesis about American violence. This issue, as I see it, is actually the connecting thread that runs through the concluding part of Monkkonen's essay. His proposed explanations, however, constitute a disappointingly traditional set of historical factors. He merely juxtaposes four themes in U.S. history, without demonstrating their interconnectedness. One of the important lessons that Norbert Elias taught historians is that this type of traditional factor analysis is insufficient. Instead, we should try to uncover the dynamics of the entire "figuration," of which factors such as those mentioned by Monkkonen may be part. Elias offers tools for such an exercise in historical reconstruction. His theory is not a fixed set of learned pronouncements, but a working program in which statements are continually open to criticism. He always called for new research, which could lead to revisions of his and others' theories. In the case of the United States, the revision takes the following form: Even when a monopoly of force—in its outward manifestation, the federal government reacts when the country is attacked—has been established for a considerable time, strong countertendencies may operate against the internal effectuation of this monopoly. These countertendencies lead to a stagnation in the spread of more "civilized" standards of behavior in some areas of social life—in the case of the United States, the area of conflict and aggression.

In order to comprehend the historical development that led to this peculiar situation, we have to take account of an element definitely present in Elias's original exposition of his theory: the sociogenesis of military and taxation monopolies.¹³ In Europe their formation was a gradual process; it took several centuries before such a dual monopoly was established in a stable form. Its emergence, which Elias analyzed for France in particular, meant that in the long run, centripetal forces prevailed over centrifugal ones. There had always been tendencies in French society that favored centralized rule, but in the Middle Ages the countervailing tendencies were much stronger. These centrifugal forces varied in nature. In an early phase they included the absence of durable chains of communication, which, for example, made it impossible for a king who had beaten a rival duke in an outlying region to rule the duchy effectively. Centrifugal forces also included rebellious groups, notably high nobles and independent-minded cities. When the royal monopoly was still weak and unstable, such rebellious groups were bent on encroaching upon or even destroying it. Dukes and counts wished to be autonomous rulers, with the king as no more than their peer. In French history, Elias identified the Fronde (1648–1652), which opposed the nobility and the Paris parlement to the regent Mazarin, as the last revolt in which an attempted destruction of the king's monopoly, rather than some other goal, was the dominant factor.

Whether the Fronde was indeed the last revolt of that type in French history is a question of lesser relevance. It is more important to identify the other goal. After the Fronde—and at a certain point in processes involving the monopolization of force anywhere—the nature of opposition and rebellion shifted. Political struggles increasingly assumed the character of aiming at co-possessing the monopoly rather than destroying it. When a military and taxation monopoly is co-possessed, this means that it is no longer the king's sole property. Its very existence is taken for granted, but representatives of the Third Estate—or similar groups in other countries—wish to co-determine, together with the king and his advisors, the ends to which it is used. In due course, the former subjects entirely deprive the king of his monopoly, which itself continues to exist. That is of course what the French Revolution was primarily about, even though a king temporarily returned afterward. As Alexis de Tocqueville observed, the state came out of the Revolutionary period much stronger. In the nineteenth and twentieth centuries, in France and other European nations, democratization meant that citizens, who were accustomed to being disarmed (at least in normal, nonrevolutionary circumstances), struggled to co-decide the ends to which the monopoly of force would be employed.

The distinction between struggles aimed at the destruction of a central monopoly and those aimed at its co-possession is relevant in several ways. For one thing, it points to the limited value of concepts such as "ascending and descending themes of government." These terms belong to political philosophy, and they inadequately describe actual realities. In particular, an "ascending" political philosophy would encompass both medieval feudalism and the French Revolution. Hence this concept fails to differentiate between anti-autocratic tendencies in a society where centrifugal forces are dominant and in a society where centripetal forces have taken the upper hand. A similar confusion is inherent to many of the present debates about the European constitution. Its democratic content is measured by looking both at the prerogatives of the European parliament and at the authority that remains with the member states.

More important for my purposes, the distinction just outlined forms a clue to what happened on the American continent. It lends great significance to the fact that the territory which became the United States lacked the gradual development characteristic of France and other European nations. Across the Atlantic, there was no phase of centralization before democratization set in. One might say that democracy came to America too early. That its revolution replaced a king with a president hardly mattered. "Too early" is a rhetorical statement, not a value judgment. It means that the inhabitants did not have sufficient time to become accustomed to being disarmed. As a consequence, the idea remained alive that the very existence of a monopoly of force was undesirable. And it remained alive in an increasingly democratic form: not of regional elites carving out their private principality, but of common people claiming the right of self-defense. In the period before, during, and after the War of Independence, the majority of the population were accustomed to relying on themselves, their families, and the assistance of neighbors for protection. Local elites and, increasingly, common people equated democracy with the right of armed protection of their own property and interests.

It would be wrong to assume that the transition from struggles aimed at destruction to struggles aimed at co-possession failed to take place at all in the United States. The best that one can say is that the majority of the population wanted it both ways. Citizens and the politicians who represented them accepted the reality of governmental institutions, but at the same time they cherished an ethic of self-help. There is an obvious relationship here with Monkkonen's factor of federalism, but my hypothesis situates federalism within a more encompassing historical-sociological theory. Illustrations of the tension between the ethic of self-help and the need for governmental control can be found in Michael Bellesiles's Arming America. The scandal about this author's gun count should not make us discard his entire research effort. Indeed, claims for self-defense can be made with various kinds of weapons in hand, or even without them. The first is implied in the Pennsylvania Constitution of 1776: "The people have a right to bear arms for the defense [of] themselves and the State; and as standing armies in time of peace are dangerous to liberty, they ought not to be kept up."¹⁴ Bellesiles emphasizes that citizens were expected to use their weapons for the defense of Pennsylvania, but I think it significant that self-defense was explicitly mentioned as well. In his chapters about the period from approximately 1780 to 1830, he concentrates on militias, and it would be worthwhile to supplement this with research into statements regarding the defense of one's home, family, and community. Today, the idea that individuals cannot and should not rely on state institutions to protect their homes is alive and well. Members of the Michigan Militia say so explicitly in Bowling for Columbine.¹⁵

Neither was slavery, another of Monkkonen's four themes, an independent factor. As a centrifugal polity of an undemocratic kind, the slave system formed part of the entire "figuration" that came out of the American Revolution. In the antebellum South, elite patriarchs ruled like petty princes on their plantations, exercising sovereignty over their families and slaves. State governments were reluctant to interfere with the power of masters.¹⁶ Although the masters did not wage war on each other, the situation is reminiscent of the European Middle Ages. An illustration from Harriet Beecher Stowe's Dred shows "the slave-owners' greatest fear": a black man with a gun.¹⁷ The fact that this image represented anxiety rather than reality reinforces my point that the plantation was a little principality. It could well be compared to a medieval duchy (with the state government resembling the suzerain king): the duke kept armed retainers, but he would never have allowed the peasants in his domains to bear arms. Slave patrols, although supervised by courts or city governments, operated with a large measure of independence—and violence.¹⁸ After emancipation, planters continued to rule their plantations as semi-autonomous domains for a long time. Before and after the Civil War, the jury system (the nucleus of Monkkonen's factor of "tolerance") further blocked an effective monopolization of force. Whites were almost never convicted of homicide (whether on blacks or on other whites); they routinely got away with self-defense or similar excuses. In an unpacified society, those excuses had a measure of reality. As Ayers explains, you had to shoot your enemy first, because otherwise he would shoot you at the next encounter.¹⁹

Thus, a kind of political-cultural stalemate existed. Even though monopolization of force within the territory of the United States progressed to some extent, in comparison to the European situation it was a slow and only partial monopolization. This is further demonstrated by the frequent resurgence of vigilante committees. Vigilantism has been defined as "extralegal coercion by a group of private individuals seeking to maintain the existing distribution of power."²⁰ Coercion may consist of threats only, but often it amounts to actual violence. Although the vigilantes themselves are not necessarily members of an elite, they do defend the established order. In terms of my theory, they should be seen as a group of citizens exercising collective self-defense and private justice in defiance of the state's claim to a monopoly of force: "People were convinced of a right to shortcut government and overrule officials."²¹ Vigilantism was especially characteristic of the Old West, but the range of its history and geography was wider. The Carolina Regulators of 1767–1771 are usually considered the first vigilantes. Within an urban context, the phenomenon continued well into the twentieth century. In Tampa, for example, vigilantism began in the 1850s, became an organized form of community justice in the 1880s, when Tampa grew into a town, and continued to claim victims until the 1940s.²² A thin line differentiated vigilante committees from lynch mobs, especially when the latter acted against "real" offenders rather than alleged ones whose principal offense had been to challenge white supremacy. This form of "rough justice," too, was exercised not so much in the absence of state institutions as in defiance of them, and that persisted until the early twentieth century.²³

Armed "detective agencies," conspicuous throughout the United States around 1900, constituted still another example of competition with the state's monopoly of force. Whereas modern private security companies are subordinate to the police, these earlier agencies operated more autonomously. Especially notorious were the episodes in which industrial employers called upon them during a labor conflict. The 1892 battle between the steelworkers of Homestead (near Pittsburgh) and the Pinkerton agency, which resulted in the deaths of seven Pinkertons and nine strikers, is well-known.²⁴ The suppression of labor unrest by private agencies equally occurred under frontier conditions. The Colorado Fuel and Iron Company hired gunmen from the Baldwin-Felts agency to police their property. During the coal miners' strike of 1913–1914, the sheriff of Las Animas County deputized 348 private detectives on the spot.²⁵ The fact that they were formally made state agents should not be confused with state control here. Labor unrest was also quelled by vigilantes sometimes, or by U.S. troops, but the use of private armed corps in this period seems a uniquely American phenomenon. In Europe, employers usually called upon the police or other state agencies when labor conflicts were beyond their control.²⁶

Finally, the continuing persistence of the ancient macho honor code in the United States exemplifies the ethic of self-defense. The honor code's presumed historical route from the plantation South to the modern ghetto takes us along various high points of private violence. The latter part of that violent route is reflected in popular music, from early-twentieth-century blues to the hip-hop of the 1990s.²⁷ In order to better understand why black men adopted the honor code of their former masters, we have to take account of Frank Henderson Stewart's distinction between vertical and horizontal honor.²⁸ The first, the more stable of the two, is assigned to someone because this person is superior, usually in rank. In many types of hierarchical societies, only a few people are entitled to vertical honor, but in a slave society it extends in principle to even the lowest of nonslaves. By contrast, horizontal honor is assigned to someone by his peers, because this person is outstanding in some way or has special merits within his peer group. This is the type that is often fought over; moreover, it is associated with the ethic of self-defense. Slaves were denied both horizontal and vertical honor, and they were hardly able to defend themselves. At the same time, white Americans equated the right of self-protection with democracy. Hence, after emancipation, African Americans were attracted to the idea of striving for honor, because it signified that they were now independent citizens. They could obtain this cultural commodity in its horizontal form, within their peer group. This situation still prevails in today's inner-city neighborhoods, as analyzed by Elijah Anderson. The "code of the street," he observes, not only stresses the need to maintain respect and independence, but it is also associated with a relative lack of pacification: "the street code emerges where the influence of the police ends and personal responsibility for one's safety is felt to begin."²⁹ This is in line with Elias's theory, which claims that localized decreases in the public security that state organizations are able to offer may lead to the resurgence of an aggressive mentality. It is also in line with my slight modification of the theory: Strong countertendencies against the internal effectuation of a monopoly of force lead to a stagnation in the spread of more "civilized" standards of behavior in some areas of social life.

Thus, in the United States as a whole and throughout most of its history, the social pressures favoring a monopolization of force have been weak in comparison with those in European national societies. This tendency, which originated in the precocious emergence of democracy and continues into the present, may go a long way toward explaining America's high homicide rates. Even today, the states and the federal government allow more self-protection than is common in almost every European country. This is evidenced by the fact that U.S. politicians are unable—or unwilling, which largely amounts to the same thing—to ban guns.³⁰ It is an intriguing paradox that the country that boasted the most formidable military externally throughout the twentieth century was unable to do away with competing claims to authority internally. The United States need not be different from other modern countries forever. If institutional centralization would increase, this might, possibly with some delay, lead to a marginalization of the self-help ethos and an accompanying decline in homicide rates. It may even be that a measure of centralization, in the face of the terrorist threat, is already under way since 9/11.³¹ If so, the effects are as yet largely unknown.

As is true of any hypothesis, some questions remain. For example, if the precocious rise of democracy was of paramount importance, why should the strongest setbacks to a continuation of state formation processes have occurred in the elitist society of the nineteenth-century South? Perhaps slavery reinforced the ethic of self-help and the hostility to a central state. Today, obviously, the self-defense ethos is more dispersed throughout the country. There is no direct link between the prevalence of the ethic of self-help in a locality and its homicide rates. Instead of positing direct and simple causalities, my hypothesis refers to the dynamics of a social figuration. Specialists in various periods of U.S. history could provide more details and show which parts of the story are in need of revision. At the moment, I think the hypothesis is worth considering.

I first developed the argument presented here in lectures given in the spring of 2001 at Carnegie Mellon University, John Jay College, Cornell University, Syracuse University, and the University of California at Los Angeles. Subsequently, I presented a joint paper with Rick Rosenfeld at the 2002 meeting of the Social Science History Association in St. Louis. I am grateful to the audiences on all these occasions. Working with Rick Rosenfeld has stimulated my reflection about the issues involved, but the basic ideas of this contribution are my own. An earlier version was commented on by Stephen Mennell and Chris Quispel.

Pieter Spierenburg is affiliated with the history and criminology departments of Erasmus University Rotterdam, the Netherlands, and is a visiting professor (Spring 2006) at the University of California, Berkeley. He has published on executions, prisons, violence, and the culture of early modern Europe. His most recent book is Written in Blood: Fatal Attraction in Enlightenment Amsterdam (Ohio State University Press, 2004). He is currently writing a history of murder in Europe, from the Middle Ages until the present.

Notes

¹ Preliminary thoughts in Pieter Spierenburg, ed., Men and Violence: Gender, Honor, and Rituals in Modern Europe and America (Columbus, Ohio, 1998), 19–25. Eric H. Monkkonen, Murder in New York City (Berkeley, Calif., 2001), 157, seems to agree with this passage, but does not refer to it here.

² As the essay I react upon is not meant as a bibliographic exercise, my contribution should not be, either. I am citing only the literature relevant for my argument. For a more elaborate bibliography, see Pieter Spierenburg, "Violence and the Civilizing Process: Does It Work?" Crime, Histoire & Sociétés/Crime, History & Societies 5, no. 2 (2001): 87–105, and Manuel Eisner, "Long-Term Historical Trends in Violent Crime," Crime and Justice: A Review of Research 30 (2003): 83–142.

³ Alfred Blumstein and Joel Wallman, eds., The Crime Drop in America (Cambridge, 2000), 14–15.

⁴ J. Oskam et al., "Het gebruik van voorwerpen en wapens bij geweldletsels," Tijdschrift voor Criminologie 36, no. 2 (1994): 120–128, 120. The authors explain that the academic hospital offers the only emergency service for an area with approximately 560,000 inhabitants (during the entire period studied, presumably).

⁵ Peter Blastenbrei, Kriminalität in Rom, 1560–1585 (Tübingen, 1995), 71, 57–68. Rates calculated by me from his data. The period studied is 1560–1585.

⁶ Barbara H. Rosenwein, "Worrying about Emotions in History," AHR 107, no. 3 (June 2002): 821–845. It would require a separate essay to fully refute her arguments. Let me just mention two points here: (1) She uncritically lumps together, as sharing the same "narrative," studies making use of Elias and understanding him well, studies using Elias and badly understanding him, and studies not referring to Elias at all. (2) She claims that recent but apparently ahistorical research into emotions contradicts Elias's "model" of the emotions, but Elias has demonstrated precisely that human emotions cannot be studied adequately unless their historical development is taken into account.

⁷ Taiwan's homicide rate, for example, fluctuated between 1.17 and 2.38 during the years 1975–1998. See Pieter Spierenburg, "Violence and Culture: Bloodshed in Two or Three Worlds," Lecture at the Academia Sinica, Taiwan, April 13, 2005. Future scholars will probably take the European Union as a unit of comparison, but this would be anachronistic with reference to the pre-twenty-first-century period.

⁸ J[ohan] Goudsblom, "De paradox van de pacificatie," Amsterdams Sociologisch Tijdschrift 25, no. 3 (1998): 395–406. I am giving a very simplified summary here.

⁹ New England homicide rates in Randolph Roth, "Homicide in Early Modern England, 1549–1800: The Need for a Quantitative Synthesis," Crime, Histoire & Sociétés/Crime, History & Societies 5, no. 2 (2001): 33–67, 56. More data will be available in Roth's forthcoming book. On the difficulty of counting the killing of Indians and, in the West, Mexicans and Chinese, see Roger Lane, Murder in America: A History (Columbus, Ohio, 1997), 310. See, however, the quantitative analysis of race and homicide in Clare V. McKanna, Jr., Race and Homicide in 19th-Century California (Reno, Nev., 2002).

¹⁰ For the debate about violence and homicide rates in the West, see Clare V. McKanna, Jr., Homicide, Race and Justice in the American West, 1880–1920 (Tucson, Ariz., 1997); McKanna, Race and Homicide; David Peterson del Mar, Beaten Down: A History of Interpersonal Violence in the West (Seattle, Wash., 2002); Robert R. Dykstra, "Body Counts and Murder Rates: The Contested Statistics of Western Violence," Reviews in American History 31 (2003): 554–563.

¹¹ Bertram Wyatt-Brown, Southern Honor: Ethics and Behavior in the Old South (Oxford, 1982) and The Shaping of Southern Culture: Honor, Grace and War, 1760s–1890s (Chapel Hill, N.C., 2001); Edward L. Ayers, Vengeance and Justice: Crime and Punishment in the 19th-Century American South (New York, 1984); Kenneth S. Greenberg, Honor and Slavery (Princeton, N.J., 1996).

¹² All three quotations are from McKanna, Homicide, Race and Justice, 17 (my emphasis), 22, 76.

¹³ Norbert Elias, Über den Prozess der Zivilisation: Soziogenetische und psychogenetische Untersuchungen, 2nd ed., 2 vols. (Bern, 1969), vol. 2: Zweiter Teil: Zur Soziogenese des Staates.

¹⁴ Quoted in Michael A. Bellesiles, Arming America: The Origins of a National Gun Culture (New York, 2000), 215.

¹⁵ Private militias have existed in the United States since 1799. Bellesiles, Arming America, 224.

¹⁶ Only occasionally did state courts deal with African Americans. See Christopher Waldrep and Donald G. Nieman, eds., Local Matters: Race, Crime, and Justice in the Nineteenth-Century South (Athens, Ga., 2001).

¹⁷ Reproduced as an illustration to Rodriguez's contribution to Michael A. Bellesiles, ed., Lethal Imagination: Violence and Brutality in American History (New York, 1999), 138.

¹⁸ Sally E. Hadden, Slave Patrols: Law and Violence in Virginia and the Carolinas (Cambridge, Mass., 2001).

¹⁹ Ayers, Vengeance and Justice, 18. Compare Wyatt-Brown, Southern Honor, chap. 14. The jury system need not automatically lead to easy acquittals, but it tends to do so where violence is viewed with leniency, as in Ireland in the second half of the nineteenth century: "The jury system meant that crimes could only be punished severely if the community saw them as truly criminal." Carolyn Conley, Melancholy Accidents: The Meaning of Violence in Post-Famine Ireland (Lanham, Md., 1999), 40.

²⁰ Robert P. Ingalls, Urban Vigilantes in the New South: Tampa, 1882–1936 (Knoxville, Tenn., 1988), xvi.

²¹ William C. Culberson, Vigilantism: Political History of Private Power in America (New York, 1990), 3–4.

²² Ingalls, Urban Vigilantes; Lane, Murder in America, 69, 131–135.

²³ Michael J. Pfeifer, Rough Justice: Lynching and American Society, 1874–1947 (Urbana, Ill., 2004). He argues that the death penalty, technocratically administered, functioned as a substitute for the rough justice of lynching.

²⁴ Lane, Murder in America, 164–165.

²⁵ McKanna, Homicide, Race and Justice, 30–32.

²⁶ Employers sometimes relied on private agencies, such as the Worsted Committee in northern England, to detect workplace embezzlement. See Barry Godfrey, "Law, Factory Discipline and 'Theft': The Impact of the Factory on Workplace Appropriation in Mid to Late 19th-Century Yorkshire," British Journal of Criminology 39, no. 1 (1999): 56–71, and "Judicial Impartiality and the Use of Criminal Law against Labour: The Sentencing of Workplace Appropriators in Northern England, 1840–1880," Crime, Histoire & Sociétés/Crime, History & Societies 3, no. 2 (1999): 57–72.

²⁷ Chris Quispel, "The Rise of a Southern Culture of Honor and Violence and Its Persistence after the Great Migration," paper presented at the Fourth Seminar on Cultures of Violence, Rotterdam, April 26, 2005. See also Chris Quispel, Hardnekkig wantrouwen: De relatie tussen blank en zwart in de VS (Amsterdam, 2002), 62–63, 342–345.

²⁸ Frank Henderson Stewart, Honor (Chicago, 1994).

²⁹ Elijah Anderson, "The Code of the Streets," Atlantic Monthly, May 1994, 81–94 (quote), and Code of the Street: Decency, Violence and the Moral Life of the Inner City (New York, 1999). See also Loïc J. D. Wacquant, "Decivilisering en diabolisering: De transformatie van het Amerikaanse zwarte getto," Amsterdams Sociologisch Tijdschrift 24 (1997): 320–339.

³⁰ I agree with Monkkonen that guns by themselves are an insufficient explanation for homicide.

³¹ See Ayse Ceyhan, "Technologization of Security: Management of Fear and Uncertainty in the US and France," paper presented at the Fourth Seminar on Cultures of Violence, Rotterdam, April 26, 2005. Ceyhan emphasizes, however, that anti-terrorist measures mostly concern security technology, rather than institutional cooperation.

Content in the History Cooperative database is intended for personal, noncommercial use only. You may not reproduce, publish, distribute, transmit, participate in the transfer or sale of, modify, create derivative works from, display, or in any way exploit the History Cooperative database in whole or in part without the written permission of the copyright holder.