Why Vigilance Committees?

Although there were many examples of private justice during the colonial and revolutionary period, the private enforcement most associated with “classic” organized vigilantism in the U.S. emerged in the 1820s and 1830s in frontier communities like Vicksburg, Mississippi.^{Footnote 13} In these areas, groups formed with the express intent of protecting civic values in a context where crime and vice had, according to members, undermined local social order. Though participants claimed that this preservation often required violent coercion, especially public execution or lynching, according to accounts promulgated by the committees, the groups were also supposed to help keep a watchful eye over local community interests and enforce law in lieu of the state.^{Footnote 14}

Vigilance committees were often quasi-formal associations, typically composed of forty to fifty individuals, and were usually organized at the county or township level, at times with alliances to other groups in nearby areas.^{Footnote 15} They frequently possessed a constitution or founding charter. For example, the Northern Indiana Regulators, a “mutual protection” association organized in Noble County in 1852 to combat counterfeiting by local “blacklegs,” had explicit rules for membership, were organized into formal “squads,” and outlined mechanisms for pursuing and prosecuting criminals.^{Footnote 16} In the case of the most famous (and largest) vigilance committee in American history—that in San Francisco in 1855–1856—the committee performed a number of municipal functions in addition to serving as private law enforcers.^{Footnote 17}

Vigilance committees relate to the political order in two ways. First, they attempt to provide redress for social wrongs in circumvention of, but not opposition to, state authority.^{Footnote 18} Vigilantes frequently do not see themselves as opponents of the state per se, but rather as defenders of a popular and legitimate form of political authority that, for whatever reason, existing institutions are not seen as providing.^{Footnote 19} This characteristic helps us differentiate vigilance committees from other forms of organized private coercion, like those opposed to the state (such as rebel militias or organized criminal protection rackets), those founded on the basis of economic rather than political authority (such as private security firms), and those with official relationships with law enforcement (such as volunteer police organizations like the Neighborhood Watch).^{Footnote 20}

Second, vigilance committees also shape the political beliefs and expectations of their members. Much of the recent literature on vigilantism in Latin America and Africa explores the effects of democratization and social transition on community membership, showing how critical moments in the expansion of rights or contraction of economic security lead to vigilante activity through threats to extant norms or values.^{Footnote 21} Brown, in turn, has argued that the ideological basis of American vigilantism is found in traditional norms of self-help, revolutionary participation, and popular sovereignty.^{Footnote 22} Just as important as the ideological roots of vigilante behavior, however, are the effects—organized committees also helped propagate an engagement toward state authority that view direct political action as the sine qua non of proper citizenship. Although they may not be the only route to this kind of vigilant citizenship, in communities undergoing moments of social ambiguity and institutional instability such committees were used to establish civic identity. For example, they were at the heart of racial social control in the Jim Crow South,^{Footnote 23} enforcement of vice laws during the Progressive Era,^{Footnote 24} anti-radical suppression during World War I and the inter-war period,^{Footnote 25} and political mobilization during the Civil Rights era.^{Footnote 26}

Current Accounts of Vigilance Committees

There are several existing accounts of vigilantism. Each of these highlights different and distinct conditions that give rise to instances of private coercion, though, as discuss later in this section, they all share similar assumptions. This section presents the most prominent of these existing explanations, while the next section presents our own approach, based on what we call social frontier theory.

One approach stresses the role of economic conflict.^{Footnote 27} According to this perspective, vigilantism is essentially an attempt by a dominant class, usually considered “respectable” and “decent,” to manage an influx of poorer residents who might threaten property values or resources in areas where the law was weak or ill-enforced. A variant of this argument has been offered recently by Phillips, in an analysis of the Autodefensa movement in contemporary Mexico. It argues that inequality creates a demand for security on the part of poorer residents: if their demand is unmet by the state, they turn to private vigilante organizations instead.^{Footnote 28}

In addition, others have also investigated the relationship between private violence in economic conflict in their analysis of claims clubs, private ordering institutions formed by land claimants in areas where private property rights were insecure to manage disputes and protect against encroaching squatters.^{Footnote 29} These perspectives contend that such clubs, which were widespread in the Midwest of the nineteenth century, were private methods of ensuring such property rights were enforced within the community.

For these economic theories, in other words, vigilance committees are institutions that selectively offer enforcement to members and in which individuals participate to supply a remedy for the lack of credible commitments produced by the perceived ineffectiveness of the law. In addition to generating local enforcement, therefore, pooling coercion in a club also indirectly allowed them to maintain economic hierarchy and fend off challengers to the status quo from newcomers. This approach thus leads to the following hypothesis:

Another explanation, derived from law and economics, argues that such institutions would emerge as the consequence of local or spontaneous ordering. In this approach, such institutions either emerge when communities share strong norms of adjudication or when they are capable of producing enforcement based on well-known reputations and repeated interactions.^{Footnote 30} Private forms of protection can operate as a kind of “club good,” in which members of exclusive organizations are able to secure collective action because they are able to provide particular benefits to those who join. Stringham, for instance, points out ways in which residents in San Francisco established private policing efforts in the nineteenth and twentieth century on the basis of subscription patrols.^{Footnote 31}

In these cases, private enforcement institutions (which may or may not coexist with state governance) are usually the result of local scale, common knowledge of rules and rule systems, and shared incentives for the creation of institutions to mitigate uncertainty. The local settlement of property rights, in particular, is a domain in which these conditions are likely to prevail, but any context in which individuals can converge around a set of rules which clearly mitigate conflict will witness such local ordering institutions. Empirically, local orders are most likely to emerge in conditions of social and economic homogeneity rather than difference — since the kinds of bargains that produce spontaneous order depend on close, tight-knit communities of relative equals, we should see them emerging in spaces characterized by shared social culture and minimal economic stratification. Thus:

A third account is functionalism. Since vigilantism is marked by a non-state attempt to create state-like law, vigilance committees compensate for a state that is absent or where existing institutions prohibit the pursuit of certain kinds of criminals. Recently, for example, some observers theorize prison social order as emerging in response to inadequate governance.^{Footnote 32} According to this theory, made popular in earlier accounts, citizens join vigilance committees either because there are no local courts or legal procedures available to residents, or because those that do exist are weak and new.^{Footnote 33}

The functionalist understanding of vigilantism is shared by many public choice scholars, who view private coercive institutions like vigilance committees as providing governance roles in substitution for the state. One of the most intriguing versions of this thesis is found in the recent work of Hadfield and Weingast, who demonstrate how decentralized forms of law (a close analogue to our concept of vigilantism) can emerge in stateless areas simply through the presence of a publicly known and flexible system of normative classification, which a shared authority can use to adjudicate disputes and provide social order.^{Footnote 34} A key assumption for these approaches is that where state authority does exist, such mechanisms are unnecessary and unworkable, since private forms of normative classification would compete with those provided by centralized state law. Thus:

A final approach focuses on social deviance to explain where and when vigilante committees are most likely to form. In this view, vigilante membership is associated with conditions where racial or communal norms are violated or endangered.^{Footnote 35} For example, vigilante committees might arise in communities when white residents are outnumbered by African Americans, and turn to private enforcement to try to gain control over a large and potentially antagonistic population.^{Footnote 36} Similarly, cultural codes of honor or shame may drive the demand to punish those who offend the status quo—in these communities, the law may not provide direct recompense for the abridgment of social norms and vigilante committees might be formed to patrol deviance informally.^{Footnote 37} For example, the threat of miscegenation was frequently an excuse for lynching, particularly when connected with rape.^{Footnote 38} Although they emphasize different mechanisms and causal processes, scholars working in this tradition focus on the ways in which cultural norms of appropriate behavior or markers of status come in conflict with existing legal process, with the result that “violence, when exercised by a ‘collective,’ became an expression of communal values.”^{Footnote 39} Historically speaking, many of the scholars that focus on episodes of vigilantism have identified these norms with the desire by whites to maintain racial social dominance,^{Footnote 40} although others have made similar arguments about ethnicity more broadly.^{Footnote 41} Thus:

These theoretical approaches, while deploying distinct causal mechanisms, all ultimately rely on some similar assumptions. First, they all argue to some extent that vigilance committees are either in competition with or substitutions for centralized legal institutions. In the functionalist approach, committees provide law when the government is not doing so effectively, while in the economic conflict model state law is present but is unable to resolve the conflicts plaguing communities. In social deviance models, vigilance communities do what the law cannot—police individual morality—while in local ordering, informal adjudication is often seen as a more efficient alternative to state arbitration. Vigilance committees in each theoretical approach are more likely to emerge in conditions of state weakness (though not necessarily absence): either the state government is too remote or inefficient to serve as the means of providing order or it is an incapable or suboptimal source of law.

Second, with some important exceptions,^{Footnote 42} these theories also agree that the identities and interests of participants in vigilance committees are relatively static and well defined. In economic conflict approaches, for example, the groups or classes that form institutions to protect their resources and authority are composed of individuals with already clear collective interests. In social deviance and functionalist approaches, in turn, groups of people with already well-bounded identities (e.g., whites, property-owning classes, etc.) attempt to correct the misbehavior of others or establish law where there is none. In turn, for local ordering models the assumption that individuals share pre-existing interests or norms is key to explaining the emergence of adjudication institutions.^{Footnote 43}

The Social Frontier Theory of Vigilance Committees

We theorize a different relationship between both formal state legal institutions and vigilantism as well as vigilance committees and identities/interests. Rather than emphasizing the ways that such committees reflect stable, pre-existing conflicts under conditions of state weakness like the theories just mentioned, we instead argue that vigilante committees are most likely to emerge where civic membership is ambiguous or undergoing transformation and where political institutions are unstable and changing—what we call social frontiers. Social frontiers encourage the formation of vigilance committees by creating uncertainty over who belongs and who gets to participate. Vigilance committees offer a way to address this uncertainty by making new civic identities. We explain our claim as follows.

Research Design

Our empirical strategy is twofold. First, we use both primary and secondary sources to delve into one specific case of vigilance committee, the large Attakapas movement in Southwestern Louisiana in 1859. This case analysis allows us to illustrate in detail how the mechanisms of ambiguity over group membership and institutional rules worked in practice and how vigilance committees used coercive organization to confer civic identity. We selected this case not only because it was one of the largest and best known examples of antebellum vigilantism, but also because it poses a fairly hard test for our theory. This region of Louisiana was quite stratified and racially diverse, putting the onus on us to identify the problem of unstable institutions and social ambiguity as opposed other potential causes for vigilance committee formation like racial or class conflict as key in the mobilization of the committee.

Second, to explore whether our claim holds when controlling for alternative explanations, we turn to a large-N analysis of the antebellum period. We use multiple sources to create an original historical dataset of vigilance committees across the United States: the U.S. census, Integrated Public Use Microdata Series (IPUMS), National Historic GIS (NHGIS), and Brown’s Reference Brown1975 data.^{Footnote 66} We also use several imputation techniques to see if our results are robust (refer to the online appendix).

Case Study: Vigilantism in Antebellum Louisiana

There were well over 150 vigilance committees active in at least 131 counties in antebellum America, and one of the best documented is the large Attakapas movement in Southwestern Louisiana in 1859. This region of Louisiana was in the midst of widespread social/structural transformation in the decades prior to the Civil War. What had been a highly rural world of Acadian small-landholding farmers and herders in the first decades of the century had been transformed into a center for large sugar plantations, some of which were owned by wealthy English-speaking Americans in the 1830s and 1840s. In addition to the relatively novel presence of slaves in the area, towns in Lafayette, St. Martin, and Vermillion had also become the destination for small (but increasing) numbers of European immigrants, further complexifying the social landscape.^{Footnote 67} Moreover, the area that abutted the Sabine River and Texas, called the “Neutral Ground,” was the site of continual conflict over state and territorial jurisdiction and a favorite destination for fugitive slaves.^{Footnote 68} Southwestern Louisiana was, in other words, an archetypical social frontier.^{Footnote 69}

In January of 1859 twelve owners of small plantations near Côte Gelée met to form a committee to combat a rash of thefts supposedly plaguing the area. The committee was primarily concerned by the perceived corruption in the court system, which was failing to convict individuals who “were perfectly known to society” as being responsible for the crimes.^{Footnote 70} Organizing themselves under a local notable, the group produced a constitution in which they committed themselves to using coercion to banish offenders from the region. They chose banishment as a tactic for two reasons. First, because “it was of an almost primitive simplicity; It was short but it was clear.”^{Footnote 71} Banishment allowed the committees to use a publicly visible sanction and to monitor whether or not the punished individual was meeting their demands. Second, exiling residents reflected a “common sense” form of punishment that was “closed to the gummy eloquence of attorneys.”^{Footnote 72} It allowed committee members to preempt challenges to their control over local civic life by simply removing threats directly.

Their first target was the son of Alsatian immigrants involved in the cattle trade; the group accused him of theft, whipped him, and forced him to sign a statement agreeing to leave the state. The group began targeting other supposed thieves through night riding and raiding.^{Footnote 73} At least 37 individuals were punished over the next nine months by the vigilance movement, which had taken on a highly organized, quasi-military cast.^{Footnote 74}

Why did the vigilance committees emerge in Southwestern Louisiana in 1859? The reasons given by the chronicler of the movement are suggestive. He contends that the initial committees were concerned not with establishing law in the absence of state institutions. Rather, the movement was responding to the “absurdities of our criminal legislation” to prosecute effectively well-known criminals. Specifically, the state was not considered absent. The procedural requirements of due process were found to be cumbersome and even limiting, allowing offenders to take advantage of “the almost unlimited right of recusation and the power of witnesses” to avoid guilty verdicts.^{Footnote 75} The vigilantes also claimed to take into consideration various institutional solutions: “the Committee of Vigilance,” claimed a manifesto published in the Avoyelles Pelican, “will put into the hands of the law all persons guilty of crimes that can be reached by law; but in cases where there is not sufficient proof to insure their condemnation before the Court, the committee will judge themselves.”^{Footnote 76} By strategically short-circuiting those offenses existing institutions couldn’t cover with summary forms of judgment and punishment the committees were not becoming substitutes for the state, but rather were correcting its failures to protect property from theft.

The protection of property was not the only goal of the groups. Equally important was the desire to “purge . . . society of the foul elements which it contains, and... to mark with the stigma of the whip or to punish by exile the dishonorable men whose presence is an insult to our morality and a danger for us and our families.”^{Footnote 77} Indeed, the problem of who, exactly, belonged in the local community was very much a matter of contention for residents of the Attakapas during the period. Not only did the influx of Anglo-Americans into what had previously been primarily French speaking areas create confusion in cultural and economic life (particularly during an era of rising American nativism), but existing fraternal and religious organizations were marked by factionalism and meager attendance in the 1840s and 1850s.^{Footnote 78} The question of disorder was intimately tied to matters of background, cultural identity, and a general debate over whether or not Acadians and Creoles could become “decent” and “law-abiding” members of American society.^{Footnote 79}

In creating the vigilance committees, the “honorable” citizens of the Attakapas region wanted to remove the miscreants from their midst; they were responding to an environment in which “no one knew if the next passer-by on the road would be a scoundrel or an honest man” and where “all the props of society were lacking.”^{Footnote 80} Driving out bandits would both deter other criminals as well as draw a firm boundary between who was a reliable citizen and who was not. The vigilance committee was seen as the only way to secure these aims.^{Footnote 81}

Who counted as honorable, then? Honorable citizenship seems to have been drawn in part in ethnic and nationalistic terms—of the 60 vigilantes for whom we have demographic information, 51 (85%) were either native Acadian or Creole in ancestry (most of the rest were Anglo-American transplants). Because they were mostly well-off members of the white elite, the vigilantes claimed to represent the best “citizens” of the parishes at large; “they are honest men, whose characters are above suspicion in the community,” wrote the Opelousas Courier; this implied that “when an issue is raised between honest men and acknowledged rogues and villains, we hope the citizens . . . will not long debate as to which party they belong!”^{Footnote 82}

In contrast, many of the targets of the committees, possessed what the vigilantes viewed as a “deviant” or hard-to-discern social identity. The vigilance committees particularly singled out slaves, “mulattoes,” Germans, Spanish, Free African Americans, and French nationals; almost 57% of the 37 known vigilante victims were immigrants of various backgrounds or were viewed as “colored,” while only 13 (or 35%) were Acadian or Creole. In the Attakapas region, high levels of physical mobility and complex ethnic identities created particular difficulties for those hoping to delineate community membership and for those trying to preserve individual reputations as law-abiding citizens. For instance, it was often difficult to tell who were the rightful owners of cattle, which was constantly at move over the sparsely settled prairies. As a proclamation issued by one committee in March 1859 warned, “race, color, black and white, all blended and grouped in this army of pillage, making it very difficult to discover the true perpetrators of criminal acts.”^{Footnote 83} A major task of the committees was thus to identify specific individuals as threats, and thereby draw a boundary around the community of honorable citizens in the social frontier of Southwestern Louisiana.

Were the vigilance committees in the Attakapas regions simply manifestations of deeper, long-standing ethnic or racial struggles between the Acadians and Creoles and everyone else? There is very little evidence to suggest that this was the case. On the one hand, despite continuing problems, by the late 1850s many Anglo-Americans had largely managed to integrate themselves into Acadian/Creole life, serving in political and economically powerful positions throughout the Attakapas and intermarrying with local families.^{Footnote 84} As a result, they shared with the Acadians and Creoles concerns over local civic membership; fully 11% of the vigilance committee members were native Anglo-Americans. On the other hand, many of the victims were themselves Anglo-Americans, Creoles, or Acadians. Indeed, some of the victims may have actually been related to prominent vigilantes.^{Footnote 85} Though vigilantes were primarily French-speaking Louisianans, the committees actually embodied a broader political identity, one made up of self-proclaimed “honorable” native citizens rather than simply “native” versus “foreigner.”

The choice of some of the targets that we cite here might suggest that class conflict played an important role. Brasseaux suggests that most targets of the committees were poorer farmers and that the committee was supported by larger planters.^{Footnote 86} Yet several pieces of evidence point away from this direction. First, several of the key figures attacked by the committees were themselves wealthy property owners. Victims Olivier Guidry and the Herpin brothers, for instance, were members of large and relatively prosperous planting families in the region. Second, vigilantes themselves made clear that the movement had little to do with class conflict; a report from a local paper sympathetic to the committees noted that “the impression that none but poor men have been driven from the country . . . is very erroneous: of those who were first driven away by the regular action of the committees, the majority were men of means . . . some of whom we know to possess considerable property.”^{Footnote 87} Moreover, some of the targets of the committees had allies among Louisiana’s elite. Robert Wyckliffe, the state’s governor, for instance, made several attempts to disband the vigilance committees in response to reported abuses of respectable local citizens,^{Footnote 88} while two of the strongest opponents of the committees, Amadee Martel and Adolphe Olivier, were influential local officials.^{Footnote 89}

The desire on the part of native Acadians, Creoles, and Anglo-Americans to use extralegal violence—and thereby help shore up the political power of committee members through realigning honorable citizenship—was also driven by unstable political institutions in the region. Much of this instability was a function of long-standing complexities of Louisiana’s legal regime; H.L. Griffin notes that “many of the settlers [in the Attakapas] were, naturally, strangers to American customs and laws, and hence could not understand them; many others were criminally inclined and not only did not want to understand them, but actively violated them.”^{Footnote 90} Southwestern Louisiana’s court system was still in the process of transitioning from the French code to American common law, and the clash of legal cultures made it difficult for locals to agree on a single, unifying set of rules to address conflicts. For years following admission to the United States, for instance, Louisiana courts had to rely on Spanish and even French precedent to help adjudicate property disputes and civil matters, frequently creating confusion over which rule system would apply.^{Footnote 91} The vigilance committee was a technique of forging a politically powerful class that crosscut ethno-linguistic barriers while also articulating a new vision of civic membership based on co-participating in violence.

In this effort, the committees seem largely to have been successful, at least in the long run. Members who first gained prominence in the vigilance committee’s activities went on to become civic leaders during and after the Civil War and gained control over much of the political life of the Attakapas parishes. For instance, Alcibiades DeBlanc, a successful attorney from St. Mary Parish and activist known as the “King of the Cajuns,” served as a Supreme Court Justice in Louisiana and was considered one of the most important political figures in Louisiana during much of the early 1870s.^{Footnote 92} Moreover, the committees seemed to pave the way toward consolidating a specifically Southern national and, ultimately, racial identity. DeBlanc was also the founder of Louisiana’s earliest violent racist vigilante group in 1868, the Knights of the White Camelia—which included a number of other former Attakapas vigilantes—and became a close ally of Anglo-American conservatives in New Orleans and elsewhere in the struggle against Reconstruction.^{Footnote 93}

In short, the vigilance committees in Attakapas were not mechanisms for adjudicating disputes among well-defined groups and individuals in a world where state capacity was weak. Instead, they were attempts to confer local civic identity on participants and block access to the rights of citizenship to socially ambiguous populations in conditions of social ambiguity and institutional instability. Indeed, in the runup to the Civil War, local vigilance committees became a staple of mobilizing local populations and monitoring treason and sedition throughout western and southern Louisiana.^{Footnote 94} In other words, what had begun as an attempt to punish local offenders fortified a conceptual schema based on vigilant citizenship that may have aided mobilization against the Union and helped build a broader, Southern identity.

The case of the Attakapas vigilance committee is not exceptional, however. Indeed, we would contend that the relationship between ambiguous social identity and institutional instability was also critical, for example, in San Francisco’s^{Footnote 95} and Iowa’s^{Footnote 96} vigilance committees of the 1850s, as well as many others in the Midwest and middle border states.^{Footnote 97}

Large-N Analysis

We test our hypotheses on the period 1850–1860, the years before the Civil War, because it presents both a compelling and difficult historical case for explaining the emergence of vigilance committees. First, the antebellum United States was the seedbed for the most important vigilance committees in American history, such as the committees in San Francisco and the Attakapas in the 1850s, by far the largest of the nineteenth century.^{Footnote 98} Focusing on that historic period allows us to model this phenomenon at its very roots.

In addition, the Civil war profoundly changed the American political landscape. With few exceptions,^{Footnote 99} most scholars associate American vigilantism with the post-Civil War struggle of whites in the South to reassert control over African Americans using extra-legal means. While our theory would expect that social frontiers such as those found in the Reconstruction and post-Reconstruction South would indeed bring about the formation of vigilance committees, by limiting our analysis to the uncharted waters of the antebellum period we are forced to explore whether the same conditions would apply prior to the reconfiguration of race relations during the Civil War.

Moreover, unlike theories based solely on the post-Civil War data, our theory is general and parsimonious, applicable to both the antebellum era (when the main ambiguity over civic membership had to do with American national identity) as well as the more complex case of the Reconstruction period (when this ambiguity shifted toward race).

We estimate a Cox proportional hazard model with a time-varying design,^{Footnote 100} with only our variables of interest (Model 1) and with all the relevant variables to test for alternative hypotheses (Model 2, refer to table 1).^{Footnote 101}

Table 1

Cox proportional hazard models' estimates of vigilance committee formation, 1850-1860. Standard errors in parentheses.

Note: *p<0.1; **p<0.05; ***p<0.01

Measurements

Our unit of analysis is the county-year. Counties enter the dataset when they are incorporated or in 1850, whichever comes first. Our dependent variable is vigilance committee formation. We focus on the county-year unit of analysis because doing so allows us to capture the administrative unit at which many vigilance committees were organized. That is, with the exception of larger cities and towns, law enforcement in this period was primarily managed by county sheriffs and militias,^{Footnote 102} and many vigilance committees also adopted the county as their organizational frame of reference. Moreover, while their jurisdictional tasks varied somewhat, focusing exclusively on counties (as opposed to also including other administrative units such as cities) allows us to compare the trajectories of units at a similar administrative level more accurately, on average, than mixing counties and cities in the sample.

To identify the vigilance committees, we rely on the list of American Vigilantes Movements compiled by Brown. The list uses the following criteria to identify committees: the groups “took the law into their own hands,” were “regular and illegal,” and existed for some period of time.^{Footnote 103} These groups are purely non-governmental—they do not include, for example, sheriff’s posses—nor were they technically legal or authorized by state law, thus excluding most private prosecution and anti-horse thief associations. The list we use triangulates across multiple sources, including counties’ histories, memoirs, and newspapers of the period.

To capture institutional instability, we use Borders Revision, a dummy derived from NHGIS data. For each year, the variable acquires the value of 1 if the county has undergone a revision of its borders in the past two years, and 0 otherwise. When the borders of counties changed, it almost always indicated that either new institutions were being constructed (e.g., new counties were formed out of unincorporated land) or that existing institutions were in flux (e.g., older counties were undergoing some jurisdictional revision, often at the behest of state representatives). Given the centrality of county sheriffs for law enforcement, border revisions are a simple but effective way of systematically capturing instances when legal institutions in a given county were present but unstable.

To measure the presence of mixed and ambiguous populations, we use the variable Ethno-National Fractionalization. The variable is computed as the classic heterogeneity index: for a classification i where p _i is a proportion of the overall population, ethno-linguistic fractionalization is $1 - \mathop \sum \nolimits _{i = 1}^np_i^2$.^{Footnote 104} Taken from U.S. Census reports, ethno-nationalist fractionalization indicates the rough probability that any two individuals drawn from the population are also from different ethno-nationalist backgrounds, coded as the number of native-born American residents and the number of foreign nationals residing in a given county.

This measure effectively captures mixed and ambiguous populations for several reasons. First, it captures the key axis along which civic membership was designated: vigilant citizenship. The more ambiguous the ethno-nationalistic character of the local population—the more frequently strangers and foreigners interacted in a given community—the more difficult it was to determine this belonging and the less likely any given specific divide (say, between Irish and Anglo-Americans) was to serve as a master cleavage. Second, the index avoids the problem of “groupism”—the predisposition to treat collectivities as entities rather than the result of social practices.^{Footnote 105} Since the index is built on very broad census categories constructed out of birthplace rather than the more fine-grained but essentialized and unstable classifications of national membership found in the census records, it tracks a salient and general demographic divide without treating those more precise classifications in particular localities as equivalent over time and space.

We expect that the probability of committee formation is higher when social frontier conditions are present—that is, when there is institutional instability and highly mixed and ambiguous populations. Thus, we expect the coefficient for Ethno-National Fractionalization X Borders Revision to be positive and significant.

To assess the economic conflict approach, following H1, we use a measure of economic inequality: the Gini coefficient. Bound between 0 and 1, the index takes an occupational ranking from the IPUMS Census Population Schedule sample and measures how unequal the distribution of well-paying jobs and low-paying jobs in a given county.

We also include a Manufacturing index, which captures the value of capital invested in manufacturing firms across a county. Higher values indicate a more industrialized and status-diverse locale. This measure controls for the possibility that more economically complex areas might have systematically different patterns of establishing social order, as predicted in H2.

To test functionalism, H3, we measure the degree to which citizens have access to state institutions. The variable Urbanization captures the proportion of the county’s population that live in cities of 2,500 inhabitants or more. We use this variable to capture the extent to which professional, bureaucratic law enforcement resources (and other public services) were readily available to residents because, for the most part, only cities had full-time police forces in the antebellum United States.^{Footnote 106} While rural counties did provide law enforcement in a variety of ways, even those with relatively effective legal institutions still lacked a cadre of state officials to whom they could turn to administer the law.^{Footnote 107}

To control for the role of social deviance described in H4, we turn to several measures. First, the variable Slavery assumes values between 0 and 100, reflecting the proportion of slaves to total inhabitants. This captures the propensity that honor codes and racial hierarchy drove vigilance committee formation, as predicted by social deviance theories. Second, Ethno-National Fractionalization can be used to test for nativism variants of H4. Counties with a both clear divide between “native” and “foreigner” (e.g., local conditions in which a distinct minority or majority of the local population wishes to police the deviance of outsiders) will have relatively high levels of homogeneity. If this variant of H4 is correct, high levels of ethno-national fractionalization should be negatively associated with vigilance committee formation.

This analysis seeks to capture and compare the theorized structural conditions that bring about the formation of committees, rather than precipitant events such as episodes of criminal activity. Historical data on levels of crime and disorder, particularly for remote areas in the United States in the nineteenth century, are notoriously unreliable and unsystematic, so adding them to the model would introduce a noisy independent variable. Moreover, because crime statistics are usually collected by state bureaucracies or law enforcement agencies, higher reported rates often merely track higher levels of state capacity rather than objectively increased rates of disorder. Finally, vigilance committees tend to respond to threats that they interpret as being important, so using (noisy) crime statistics to predict committee formation might be particularly misleading. For example, counterfeiting and horse thievery were common throughout Illinois, Ohio, Iowa, and Missouri in the 1830s and 1840s and involved relatively large-scale organized crime networks.^{Footnote 108} The mere presence of this criminal activity in a given town was not enough to activate vigilantism however; instead, vigilantism clustered in borderland areas where thieves were likely to seek refuge rather than in areas where the actual thefts took place. Crime incidence statistics do not reliably correlate to how a threat is interpreted in such conditions.^{Footnote 109}

Results suggest that the presence of a social frontier is correlated with an increase in the risk of committee formation: the coefficient for Ethno-National Fractionalization × Border Revisions is positive and statistically significant across model specifications. Figure 1 parses out the effects of the presence of social frontiers, by plotting two survival curves: one with conditions of social frontier (black line, with Ethno-National Fractionalization at the seventy-fifth percentile and Border Revisions equal to 1) and one illustrating the lack of conditions of social frontier (gray line, with Ethno-National Fractionalization at the twenty-fifth percentile and Border Revisions equal to 0). The black line is steeper than the gray line, showing that in the presence of social frontier conditions the risk of committee formation increases more rapidly than it does in the absence of those conditions.

Figure 1

Estimated probability of the years a county will go before witnessing the formation of a vigilance committee (survival curve), comparing counties with no social frontier (gray line at the top) and counties with social frontier (black line at the bottom). Continuous lines represent 95% confidence intervals.

The coefficient for Border Revisions captures the effect of institutional instability when there is no social ambiguity (thus, when Ethno-Fractionalization is equal to 0). The coefficient for Border Revisions is not significant in either model. Our result suggests that institutional instability in the absence of ethnic fractionalization is not conducive to the formation of vigilance committees.

The coefficient for Ethno-National Fractionalization captures the effect of social ambiguity when there is no institutional instability (thus when Border Revisions is 0). When there is no institutional instability, the presence of social ambiguity increases the risk of committee formation as the coefficient for Ethno-National Fractionalization is positive and significant. However, the effect of the presence of frontiers is stronger than the effect of the presence of ethnic fractionalization without institutional instability. Specifically, each percentage increase in ethnic fractionalization increases the yearly hazard of committee formation by 3.01% in the absence of institutional instability and by 6.7% in the presence of institutional instability. These effects suggest that the probability of committee formation is 55 percentage points higher in the presence of social frontier conditions than it is in the presence of social ambiguity alone. The coefficient for Ethno-National Fractionalization does not reach statistical significance in Model 2.

Each percentage increase in economic inequality reduces the risk of committee formation by 2.48%, contrary to H1. Results fail to support the other alternative explanations as we operationalized them, as the coefficients for the measures we are employing fail to reach statistical significance.

Conclusions and Implications

We have argued that the emergence of vigilance committees depends on the presence of social frontiers. We find that vigilance committees are not the product of conflict between well-defined groups, classes, or factions, but rather the joint product of socially defined ethno-nationalist heterogeneity and politically defined institutional instability. In the United States, committees could confer a new civic identity based on norms of vigilant citizenship.

Several implications follow from our analysis. First, our findings reinforce the emerging scholarly consensus that, because vigilance committees were not simply substitutes for an absent state, as a functionalist approach would suggest, but instead were correctives for institutional instability and social ambiguity, simply expanding state capacity is not always effective at countering private enforcement.^{Footnote 110} This runs directly contrary not only to the conventional view of vigilantism as a substitute for a weak state, but also to general political economic models of state formation that pose clear tradeoffs between state monopoly control over enforcement and private alternatives.^{Footnote 111} It also expands on important recent work explaining contemporary vigilantism as a response to threats to local social order from post-war reconstruction,^{Footnote 112} neoliberal politics,^{Footnote 113} or the dislocations of globalization^{Footnote 114} by showing how the making of such order also necessarily entails practices of defining and conferring civic identity.

Moreover, although the ideological content of that membership reflected a version of “vigilance” that was distinctly American, the spread of these types of organizations across many democratizing nations in the past thirty years suggests that the social frontier hypothesis might apply elsewhere. For instance, one fruitful setting to explore committee formation is Mexico, where at least 70 municipalities have experienced vigilantism since 2013.^{Footnote 115} Here the drug economy has generated a great deal of uncertainty over the application of the law (rather than an absence of law, per se) as well as questions over who is and is not counted as a “criminal.” Social frontiers might also help explain contemporary cases in Nigeria,^{Footnote 116} Guatemala,^{Footnote 117} and India,^{Footnote 118} where moments of institutional transformation and instability have combined with ambiguity over community membership to generate vigilante justice.

Supplementary Materials

The below supplementary materials are available at https://doi.org/10.1017/S153759271800107X