Abstract
Objectives: Drawing on the “legacy effect” framework and the intergroup contact hypothesis, this study examines whether 1) slavery is associated with anti-Black hate crimes, 2) racial segregation is associated with anti-Black hate crimes, and 3) racial segregation moderates the relationship between slavery and anti-Black hate crimes. Methods: Using several data sources, including the 1860 and 2000 U.S. Census and the 2000-2016 Uniform Crime Report, negative binomial regressions are employed to explore these research questions. Results: In support of the “legacy effect” and intergroup contact hypothesis, the findings from the analyses reveal that both slavery and racial segregation are significantly associated with increases in anti-Black hate crimes. Moreover, the relationship between slavery and anti-Black hate crimes is amplified by racial segregation. Conclusions: The results from this research suggest that historical practices and institutions, such as slavery, can influence the contemporary maltreatment of Black citizens. Specifically, the racism and discrimination inherent in slavery has manifested in the form of modern-day anti-Black hate crimes. Additionally, the relationship between slavery and anti-Black hate crimes is amplified when racial segregation prevents racial and ethnic groups from building meaningful relationships that could demystify the culture of Black inferiority created by the institution of slavery.
Introduction
Existing scholarship has consistently demonstrated that Black and Brown citizens disproportionately experience criminal justice outcomes, including victimization, crime, and punishment. To gain a better understanding about these disparities, scholars have explored and found that a host of individual- and aggregate-level factors are associated with the differential experiences of Black and Brown citizens, including race, gender, and socioeconomic status. Yet, despite this line of research, questions remain about the root causes of the racial disparities that characterize justice outcomes in America. For example, in the wake of the 2020 murders of George Floyd, Breonna Taylor, Ahmaud Arbery, and others, there has been public outcry to address the continued brutalization of Black bodies. I argue that in order to gain a better understanding of the racialization of criminal justice outcomes, it is necessary to acknowledge the failure of the criminological discipline to adequately account for the historical context within which these disparities may emerge. Therefore, to fully center race within the criminal justice system and account for the disparities that people of color continue to experience, scholarship must consider the historical institutions that have given rise to the racial structure that exists today.
To understand these and similar issues fully, it is necessary to consider the institution that remains the most striking example of the disparate treatment of Black people in United States history. Slavery—an institution in which Whites decimated Black life and culture while building generational wealth for them and their descendants—set a precedent for how Black people would be treated over time. During slavery, Black people were not only forced to work on cotton, tobacco, sugar, and rice plantations under harsh conditions (Ruef and Fletcher 2003) but were also subjected to starvation, whippings, and lynchings. Slavery, then, provided both economic resources to the owners of enslaved people and created a violent culture rooted in the perceived inferiority of Black people when compared to White people (Kolchin 2003; Bruce 1979).
At the conclusion of the Civil War, which marked the end of slavery, over 4 million Black people had been enslaved (see Schwartz 1970). While the organizational structure created under the institution of slavery had been formally abolished, the social, political, and economic hierarchies created under the institution of slavery persisted. In addition to providing cheap labor to White Americans, Black Americans continued to experience social, political, and economic subjugation in America for more than 100 years after the abolition of slavery. Specifically, Black Codes, Jim Crow Laws, legal and de facto segregation, and the rise of the Ku Klux Klan facilitated the perpetuation of violence and discrimination targeting Black people. More recently, the era of mass incarceration, or the New Jim Crow as termed by Michelle Alexander (2012), has further amplified racial disparities through the disproportionate incarceration of Black Americans when compared to White Americans. Ruef and Fletcher (2003) refer to this process as the institutional legacy of slavery, suggesting that there is a “reproduction of material resource and cultural conditions from a social institution despite the fact that the institution has been formally dismantled” (p. 447). Related research has been largely supportive of the assertion that the “legacy of slavery” can be maintained through the manifestation of subsequent systems of inequality. Specifically, research has found that the legacy of slavery continues to be consequential, particularly for Black Americans, across several outcomes, including occupational stratification (Ruef and Fletcher 2003; Sacerdote 2005), poverty inequality (Schiller 2008; O’Connell 2012), education (Bertocchi and Dimico 2012), and legal executions (Vandiver et al. 2006).
While this line of research has furthered our understanding of how historical institutions can give rise to the unequal treatment of Black Americans, voids remain. Expressly, scholarship has been limited in exploring the relationship between the legacy of slavery and contemporary acts of victimization towards Black people. Given the violence used against Black people throughout slavery to maintain dominance and instill fear, this relationship warrants further attention. Therefore, the current study assesses whether (1) the legacy of slavery is consequential for contemporary acts of victimization towards Black people, and (2) the conditions under which this relationship may manifest. Building on theory and prior research, the current study explores three specific research questions. First, is slavery associated with anti-Black hate crimes? Second, is racial segregation associated with anti-Black hate crimes? Third, does racial segregation moderate the relationship between slavery and anti-Black hate crimes?
Investigating these research questions contributes to prior research on the legacy of slavery, racial segregation, and hate crimes in several notable ways. First, it historically centers how we think about the impact of race in America, particularly within the criminal justice system. We cannot fully understand and therefore address issues related to the disparate treatment of Black and Brown people without considering factors that may be associated with current crime and justice trends. Historical events, such as slavery, should be incorporated in these discussions because they have the potential to provide a more complete framework for how race matters contemporarily. Second, this research responds to calls to disaggregate hate crimes by bias type (Powers and Socia 2019; Gladfelter et al. 2017). If hate crimes resulted from the same stimulus, there should be similar rates of hate crimes across bias. However, in 2016, the FBI reported that 50.2 percent of hate crimes were motivated by anti-Black bias, constituting the largest category of racially motivated hate crimes. By specifically exploring whether slavery is associated with anti-Black hate crimes, this research may illuminate what factors are consequential for the most prevalent form of hate crimes in America. Finally, by investigating racial segregation as a potential moderator, this research has the potential to provide a better understanding of the conditions under which the potential relationship between slavery and anti-Black hate crimes is amplified. Taken together, these contributions can further elucidate not only how historical institutions, such as slavery, can influence anti-Black hate crimes but also how the racial composition of an area matters in the explication of this relationship.
Background and Theoretical Foundation
Historical Continuity and the Legacy of Slavery
Prior research has consistently demonstrated that cultural traditions can become ingrained in the fabric of a place by replacing outdated institutions with new systems of inequality (O’Connell 2012; Ruef and Fletcher 2003; Tilly 1998). Using a school of thought termed “institutional legacy of slavery”, scholars have begun exploring the relationship between slavery and contemporary outcomes (Ruef and Fletcher 2003). In this line of work, the conceptualization of slavery has taken on two forms—individual-based or place-based perspectives. The individual-based perspective emphasizes the development of enslaved people and suggests that the institution of slavery may have a direct impact on formerly enslaved people in contemporary times. For example, Ruef and Fletcher (2003) used interviews conducted with formerly enslaved and free Black people during the late 1920s and 1930s to examine whether slavery continued to influence the social status of southern Black people after emancipation. They found that Black people who were free in the antebellum South had significantly higher status attainment after slavery than enslaved people, suggesting that the legacy of slavery persisted decades after slavery's abolition. Specifically, enslaved people continued to be overrepresented in unskilled occupations and semiskilled agriculture after emancipation, while Black people who were free during slavery were overrepresented in all high-ranking occupations. Enslaved artisans and craft workers, for example, achieved higher occupation attainment after emancipation than enslaved people who worked in the field (Ruef and Fletcher 2003). This work emphasizes that the dismantling of slavery did not necessarily free enslaved people from subjugation. Instead, the previous oppression experienced by formerly enslaved people morphed into newer institutions meant to maintain a social hierarchy where White Americans could secure better economic placement than Black Americans. This perspective, however, largely overlooks how slavery may have shaped the character of a place.
The current study draws on the place-based perspective, which suggests that the legacy of slavery not only influences future outcomes of enslaved people but also permeates places by creating societal norms and/or practices that allow the mistreatment of Black people who are far removed from the institution of slavery. Tilly (1998) provides the mechanism for which cultural transmission might occur, arguing that places reproduce inequality across generations by mapping social categories onto hierarchal power relations. The resulting hierarchy is maintained because dominant groups develop a monopoly on valuable resources. To this end, several studies have employed a place-based perspective for understanding the contemporary effects of slavery. To illustrate the point, Reece and O’Connell (2016) used spatial data analysis techniques to explore the relationship between historical concentrations of enslaved people and the Black-White ratio of public school attendance. They found that Black students were more likely to attend public schools in areas that were characterized by higher historical concentrations of enslaved people. These findings demonstrate that White children enjoyed access to a wider array of school opportunities when compared to Black children. Such disparity parallels the inequities that characterized the institution of slavery, which prohibited Black children from learning to read or write.
The legacy of slavery is not limited to contemporary effects on inequality. According to Ruef and Fletcher (2003), the legacy of slavery has also manifested in “the enforcement of Black codes by police and the harassment of Blacks by White patrol and vigilante groups convened to regulate African American behavior” (p. 448). One notable example of such harassment is the “lynching era” which spanned from 1880 to 1930. American lynchings were public gatherings characterized by torture, mutilation, burning, and/or hanging until the death of primarily Black victims (Levin 2002). These “powerful tool[s] of intimidation” (Brundage 1997:2) served as overt symbols of the overarching oppression and terror experienced by Black folks. Between 1882 and 1968, Black lynchings accounted for 72.7% of all lynchings (Tolnay and Beck 1995). Racialized violence, such as lynchings, is a residual effect of slavery and represents the perpetuation of Black people's perceived subordinate status in society. In other words, “this pattern contributes to the equilibrium in a system of stratification, despite superficial transformation” (Ruef and Fletcher 2003:448). Here, I contend that the explicit racism inherent in the ownership and lynching of Black people has evolved, albeit superficially, into something less explicit. That is, the legacy of slavery highlights the contemporary institutions and practices that arise from reproducing the social, economic, and political context surrounding slavery. Under new names, these new practices and institutions are created to replicate similar patterns of stratification, terror, and violence that were fundamental to slavery.
A small body of research has examined slavery's role in predicting contemporary acts of violence. One line of inquiry has suggested that slavery is consequential for state-sanctioned violence. Lofquist (2002), for example, found that membership in the confederacy was the strongest predictor of modern death penalty intensity. Similarly, Vandiver et al. (2006) used post-Gregg 1 execution and death sentence data to explore the relationship between slavery and executions. Their analyses revealed a positive relationship between slavery and modern executions, finding that 90.6% of post–Gregg executions occurred in states that supported the institution of slavery, irrespective of the state's prior membership in the confederacy. Other scholars have explored the relationship between slavery and violent crime. Hackney (1969) argued that violence had been highly associated with Southerners. Moreover, Gouda and Rigterink (2017) found that the proportion of enslaved people in the 1860 population was significantly associated with higher violent crime rates in all Census years from 1970–2000. Collectively, this work suggests that there is a relationship between slavery and violence, indicating that violence continues to be ingrained in places that traditionally had higher concentrations of enslaved populations. What remains unclear is whether this relationship extends to one of the most prominent contemporary examples of victimization of Black Americans—hate crimes.
Hate Crimes
The victimization of people based on their membership in or identification with a particular group is termed a hate crime. To fully understand the prevalence of these events, Congress passed the Hate Crime Statistics Act (HCSA, 1990). HCSA required the federal government to collect data on the nature and prevalence of hate crimes or crimes motivated by race/ethnicity, religion, or sexual orientation, including crimes of murder, non-negligent manslaughter, forcible rape, aggravated assault, simple assault, intimidation, arson, and destruction, damage, or vandalism of property. Since its enactment, the HCSA has been updated to include acts against disability and actual or perceived gender identity. Consistently, official statistics have found that hate crimes are more likely to result in physical attacks against the victim and are less likely to result in property damage, such as vandalism and arson (see Lantz and Kim 2019 and U.S. Department of Justice 2007). To this end, scholars have used official statistics in their exploration of the determinants of hate crimes. Scholars have suggested that hate crimes, as suggested by the name, are motivated by hate, prejudice, and/or bias. Disha et al. (2011) explored hate crimes against Arabs and Muslims across U.S. counties in the months before and after September 11, 2001 (9/11). Their analyses revealed that hate crimes targeting those groups saw a sharp increase in the months following 9/11, a period when anti-Muslim and anti-Middle Eastern sentiment had intensified. Similarly, King and Sutton (2013) found that high-profile trial verdicts and lethal domestic terrorist attacks are associated with increases in hate crimes motivated by religion and race.
Other lines of work have explored contextual factors in the explication of hate crimes. Levin and McDevitt (2002), for example, suggested that hate crimes are higher in communities with job scarcity. Gladfelter et al. (2017) used Pennsylvania Human Relations Commission (PHRC) data to explore the relationship between social disorganization and racially motivated hate crimes. Their analyses revealed that anti-Black hate crimes were higher in advantaged and predominantly White locales, unlike anti-White hate crimes, which were higher in disadvantaged and residentially unstable communities. Several studies have found associations between the size of the White population and hate crimes committed against Black people, suggesting that White people use hate crimes as a mechanism to protect their spaces (Jenness and Grattet 1996; Lyons 2008; Sampson and Raudenbush 2004). To this end, Lyons (2008) found that anti-Black hate crimes were most prevalent in largely White areas characterized by strong communal ties and recent Black in-migration.
Finally, a small body of literature has explored the effects of historical factors on contemporary instances of hate crimes and compliance with hate crime mandates. Notably, King et al. (2009) explored whether the contemporary enforcement of hate crime laws and compliance with hate crime mandates are inversely related with prior levels of lynching. Their analyses revealed that the size of the Black population moderates the relationship between past lynchings and (a) police's compliance with federal hate crime law, (b) police reports of anti-Black hate crimes, and (c) the likelihood of prosecuting a hate crime case. Despite the breadth of literature exploring the determinants of hate crimes, research has often excluded other historical considerations that may further explicate hate crimes, particularly those targeting marginalized groups. While lynchings provide a foundation for understanding how historical factors may influence the continued mistreatment of Black people, research has largely overlooked the institution that gave rise to America's racial structure—slavery.
Slavery and Anti-Black Hate Crimes
Given this void in the literature, the current study explores the relationship between slavery and anti-Black hate crimes. There are several reasons why this relationship might exist. First, violence was a prominent component of slavery and hate crimes (Tolnay and Beck 1995; Rice 1975; Mullin 1992; Fogel and Engerman 1995; Campbell 1991). Second, in addition to the violent nature of slavery and hate crimes, hate crimes also illustrate a level of bias, stereotyping, and discrimination against Black people that is similar to that of slavery (Messner et al. 2004; Perry 2001; Sandholtz et al. 2013; Strom 2001). Consistent with this idea, Levin (2002) suggests “slavery is important to a contemporary analysis of hate crimes because the hateful precepts that justified it still resonate in the stereotypes that contribute to acts of discriminatory violence” (p. 228). I assert that, in many ways, slavery framed how we view race in America. During slavery, skin color conveyed social standing in America, where all Black people, regardless of status, were assumed to be enslaved until proven otherwise. Equating skin color with social, political, and economic status did not end with the Union's victory in the Civil War. In contemporary times, skin color is often associated with one's wealth, status, and criminality. Despite these points, explorations of how race matters for criminological outcomes have not adequately considered the institution that shaped the ways we view and treat Black people. Slavery is essential to a more comprehensive understanding of the nexus between race and the criminal justice system because it provides a historical point of reference from which contemporary disparities have emanated.
Third, both slavery and hate crimes exhibit similar patterns of social control. Historically, slavery was an institution rooted in the control and subjugation of Black people with the expressed purpose of maintaining the social, political, and economic hierarchy of White supremacy. Even though slavery was abolished in 1865, many of the characteristics of slavery persisted in the places where slavery was most prevalent. I argue that hate crimes create and cultivate a comparable stratification in contemporary times by instilling fear and sending the message that Black people are not welcome in certain places. Finally, contemporary acts of bias and discrimination may manifest in places that have a historical legacy of bias and discrimination. This idea is supported by Acharya and colleagues (2016), who argue that ideology about race and social standing can become ingrained through socialization. In other words, aspects of history—particularly slavery—may be woven into the culture and continue to have what is referred to as a “legacy effect”. Thus, hate crimes may be thought of as the evolution of overt displays of racial animus (King et al. 2009).
Despite these parallels, prior literature has not adequately addressed how the legacy of slavery may extend to the explication of hate crimes against Black people. To my knowledge, one scholar has tested the relationship between the legacy of slavery and anti-Black hate crimes. Gunadi (2019) found that a one-percentage-point increase in the 1860 enslaved population is associated with 0.018 more anti-Black hate crime incidents per 100,000 population. While this work provides a foundation for the current study, it does not account for several relevant control variables. Building on prior research and the legacy of slavery assertion, the current study explores whether the cultural norms and racial attitudes inherent in the institution of slavery are associated with the contemporary victimization of Black people through anti-Black hate crimes.
Slavery, Anti-Black Hate Crimes, and Intergroup Contact Hypothesis
In addition to the limited empirical attention exploring the relationship between slavery and anti-Black hate crimes, a paucity of research has examined the association between slavery, racial segregation, and anti-Black hate crimes. Therefore, the current study draws on the intergroup contact hypothesis to develop two additional hypotheses about the salience of context in explaining the prevalence of anti-Black hate crimes. In his work, The Nature of Prejudice, Allport (1954) proposes intergroup contact theory, which explains how contact between groups is consequential for perceptual outcomes across racial groups. According to the intergroup contact hypothesis, “prejudice, unless deeply rooted in the character structure of the individual, may be reduced by equal status contact between majority and minority individuals.” (Allport 1954:281). Allport (1954) asserts that greater contact among groups of different backgrounds, particularly racial backgrounds, could effectively reduce prejudices and stereotypes between those groups. Empirically, scholars have found that intergroup contact can alleviate prejudicial beliefs while also increasing social and intellectual development, openness to diversity, commitment to racial understanding, and cultural awareness (Stotzer and Hossellman 2012; Chang 1999, 2000; Duster 1995; Pascarella et al. 1996). Stotzer and Hossellman (2012), for example, suggested that greater integration provides opportunities for individuals to meet and interact with people from diverse backgrounds, thereby fostering more tolerance and acceptance.
An alternative way of conceptualizing this relationship is to consider how limited contact (or the lack of integration) is likely to breed intolerance and hate. In other words, Allport's (1954) intergroup contact theory suggests that racial distrust, fear, exclusion, discrimination, and violence are likely to arise in places where there is more segregation and less contact between groups. Brameld (1946) reinforces this perspective by suggesting that when groups “are isolated from one another, prejudice and conflict grow like a disease” (p. 245). Similarly, Sigelman and Welch (1993) posit that racial segregation [can be expected to be] a source of ignorance and ignorance [will] breed derogatory stereotypes and racial hostility” (p.781). Consistent with this idea, scholars have found that limited contact amongst racial groups creates social distance, which often fosters intolerance, discrimination, and prejudice (Allport 1954; Iceland and Wilkes 2006). Specifically, segregation appears to be influential in shaping the treatment and feelings towards members of the out-groups. Rothwell (2009), for example, found that segregation increased the likelihood of stereotypical views of Black people's intelligence levels. Notably, residential segregation has also been linked with violence, especially in urban areas. Krivo et al. (2009) found that increases in segregation are associated with intraracial homicides in White and non-White neighborhoods, with violence being more pronounced in Black and Hispanic neighborhoods. As it pertains to the current study, racial segregation has been extended to explain the occurrence of hate crimes and has been largely supportive. Lyons (2008) revealed that anti-Black hate crimes are most prevalent in White homogenous communities characterized by the in-migration of Black people. Moreover, Gladfelter et al. (2017) found that anti-Black hate crimes are more likely to occur in White homogenous and residentially unstable areas. Building on prior research and intergroup contact hypothesis, the current study explores whether the separation of racial groups fosters an environment that increases the likelihood of Black victimization through anti-Black hate crimes.
Additionally, prior research has not examined how slavery may interact with racial segregation to influence the treatment of Black people in certain places. I argue that the legacy of slavery created the foundation of the perceived inferiority of Black people. This inferiority is likely heightened in segregated places where Black people and White people are less likely to form meaningful relationships that could lessen the stereotypes of Black criminality, fear, and inferiority created by slavery. It follows logically that anti-Black hate crimes could be the outward display of the convergence of slavery and racial segregation in specific locales. Therefore, the current discussion proposes that segregation has been damaging to Black people in ways similar to that of slavery; and as a result, the interaction between slavery and segregation may amplify the maltreatment of Black people in specific locales.
Data and Methods
Data
This study incorporates data from several sources. The 1860 U.S. Census provides the most up-to-date data for the percentage of enslaved people at the time of emancipation (Acharya et al. 2016); the 2000 U.S. Census provides data on population demographics for U.S. counties; 2 the 2000 Massachusetts Institute of Technology Election Data and Science Lab provides data on county-level election returns for the 2000 U.S. Presidential Election; the 2000 Uniform Crime Report (UCR) provides data on the index crime rate; and the 2000–2016 Uniform Crime Report Hate Crime Statistics Data provides information on hate crime incidents. 3 The final analysis is limited to 1,302 counties from states within the U.S. Census-defined South: Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, South Carolina, Tennessee, Texas, Virginia, and West Virginia (see Reece and O’Connell 2016). Oklahoma was excluded from the analysis because it was not a state in 1860 and lacks the geographical boundaries necessary to connect with contemporary county boundaries (O’Connell 2012; Reece and O’Connell 2016).
Measures
Dependent Variable
Anti-Black hate crimes is the dependent variable. Consistent with prior research, it is measured as the overall count of criminal acts motivated by bias or hate towards Black people in a county from 2000–2016 (see Disha et al. 2011; Gladfelter et al. 2017; Lyons 2007, 2008; King et al. 2009). Table 1 demonstrates that the average number of anti-Black hate crimes across the 1,302 counties included in the study is 6.12.
Descriptive Statistics for Dependent and Explanatory Variables.
Key Predictor Variables
Slavery is measured as the percentage of enslaved people in 1860. Scholars have commonly used this measure as a proxy for slavery's prevalence (see Reece and O’Connell 2016 and Acharya et al. 2016) and is preferred over the alternative operationalization that uses the contemporary population of Black people for two reasons. First, the use of the contemporary Black population overlooks the sizable amount of Black migration both in and out of the South since 1860; failing to account for this change can compromise the validity of the measure (see O’Connell 2012). Second, contemporary Black population sizes are used in the racial threat, defended neighborhoods, and intergroup contact literatures (O’Connell 2012). In this case, operationalizing the legacy of slavery as the contemporary Black population size while also incorporating the contemporary Black population size in the measure of racial segregation presents issues with multicollinearity, therefore, limiting the ability to draw meaningful conclusions from the results (O’Connell 2012). To distinguish the effects of the legacy of slavery from contemporary racial segregation, the current study operationalizes the legacy of slavery as the percentage of enslaved people in 1860.
The current study uses a new set of enslaved proportion estimates for contemporary units to account for county boundary changes over time. Instead of aggregating spatial units, this method uses a redistribution process described by O’Connell (2012), which creates more nuanced estimates of the distribution of historical enslaved populations. Specifically, “information from neighboring counties was used as an indication of the distribution of slaves within a historical county that subsequently split into counties” (Reece and O’Connell 2016: p. 49). O’Connell (2012) asserts that “this approach ignores population concentration shifts across counties within a cluster, yet is preferable to assuming slaves were evenly distributed across space” (p. 725; see also Downey 2006). Assuming an even distribution is problematic because it negates that slavery was predominantly agricultural and, therefore, centrally located in places with fertile land. According to Table 1, the average 1860 enslaved population across the 1,302 counties in the study is 28 percent.
Racial segregation is measured using the Black-White isolation index, which describes “the extent to which minority members are exposed only to one another” (Massey and Denton 1988, p. 288). This index measures the likelihood that the average Black person in a county will come into contact with other Black people. As suggested by Allport's (1954) intergroup contact hypothesis, the incorporation of racial segregation as the isolation index is used because it can capture the extent to which the separation of racial groups contributes to the prevalence of hate crimes in a county. Using the 2000 U.S. Census, the isolation index is computed as the weighted average of the Black population in each county. The isolation index is a continuous variable, ranging from 0 to 1, where 0 represents no isolation and 1 represents complete isolation. The calculation for the isolation measure is as follows:
Control Variables
To account for factors that may confound the potential effects of the key independent variables, the current study includes several county-level control variables in the regression analyses. Because the dependent variable is the pooled anti-Black hate crimes in a county from 2000–2016, the control variables are measured in the year 2000, which helps establish causal order (see also King et al. 2009). County-level control variables include: percent age 15–24, percent high school degree, percent rural, percent democrat, percent unemployed, population density per 10,000, index crime rate, and legislative mandate. 4 Percent age 15–24 measures the percentage of the population that is 15–24 years of age is included as a control variable because prior research has demonstrated that younger people are disproportionately involved in serious crime (Disha et al. 2011; Moffitt 2007; Rosenfeld 2002). Table 1 demonstrates that across the 1,302 counties, approximately 13 percent of people are between the ages of 15 and 24. Additionally, in line with McVeigh and colleagues' (2003) argument that individuals in the United States with more education are less likely to express racial prejudice, the percentage of the population age 25 and older with a high school degree is included as a control variable in the analyses (see also Bobo and Kluegel 1993). Because predominately rural counties are likely to have fewer incidents and fewer resources available to direct at hate crime enforcement, percent rural is included as a control variable. Table 1 demonstrates that, on average, 64 percent of counties included in the study are rural.
Democratic strongholds are more concerned with hate crime reporting because of the party's affiliation with certain civil rights issues (Disha et al. 2011); as such, percent Democrat represents the percentage of individuals who voted for the Democratic candidate—Al Gore—in the 2000 U.S. Presidential Election. According to Table 1, on average, 41 percent of counties within the study voted for the 2000 Democratic presidential candidate. Percent unemployed represents the percentage of individuals 18 years or older that are unemployed in a county (see Lyons 2007; Stacey et al. 2011). This is included as a control variable because places with higher levels of unemployment may not have the available resources to influence the enforcement of hate crime laws (see also McVeigh et al. 2003 and Disha et al. 2011). Population density per 10,000 represents the number of people per square mile per 10,000 people within each county. Existing research has shown that hate crimes are more prevalent in places where people are more densely crowded together.
Consistent with prior literature that suggests that hate crimes are a function of general crime patterns (Disha et al. 2011), index crime rate represents the total number of index crimes—murder, forcible rape, robbery, aggravated assault, burglary, larceny, and motor vehicle theft—per 100,000 people in each county in the year 2000. Finally, prior research has demonstrated that some agencies recognize the need for hate crime policy but cite administrative reasons for the lack of policy creation; other agencies deny the presence of hate crimes, and thus there may be a need for policy mandates (see Jenness and Grattet 1996 and Gladfelter et al. 2017). Because official statistics regarding hate crimes can vary from agency to agency, using data obtained from the U.S. Department of Justice (2020), legislative mandate measures whether a county is located within a state with statutory mandates for law enforcement agencies to report hate crime incidents to the attorney general. As demonstrated in Table 1, 60 percent of the counties in the study are required to report hate crimes. To account for multicollinearity among the predictor variables, several diagnostics were performed. Appendix 1 includes a bivariate correlation matrix. In line with prior research that suggests that racial segregation is lower in places where slavery was more pronounced, the bivariate correlation suggests that slavery and racial segregation are significantly and negatively correlated (see Liberson 1963; Masey and Denton 1993; Taeuber and Taeuber 1965). Further examinations of multicollinearity were conducted using Variance Inflation Factor (VIF). All of the coefficients from the assessment were below 4, suggesting that multicollinearity is not an issue for the regression analyses.
Analytic Strategy
The dependent variable, anti-Black hate crimes, is measured as a non-negative count of the number of anti-Black hate crimes in a county from 2000–2016. The distribution of count data typically has large counts toward zero and a positively skewed tail, suggesting evidence of overdispersion (see Lyons 2008). Overdispersion was investigated by assessing whether the mean dispersion parameter is equal to zero (H0: K = 0) against the alternative hypothesis that the mean dispersion parameter is greater than zero (Ha: K > 0). These findings demonstrate that the null hypothesis can be rejected, which assumes that the variance and the mean are equal (K = 0). Therefore, negative binomial regression is the most appropriate approach because it accounts for overdispersion.
Moreover, prior research has demonstrated that spatial dependency can be characteristic of hate crime data for several reasons. First, hate crimes can be felt across community lines because of their retaliatory nature (Craig 1999; Lyons 2008). Second, given the difficulty of defining units spatially, an error may be introduced when defining communities at the aggregate level (Morenoff et al. 2001). To avoid biasing the key independent variables and account for spatial autocorrelation, a spatial lag is included in the models. The lag represents the average anti-Black hate crime in contiguous counties and assesses the independent effect of spatial proximity on county levels of anti-Black crimes (Lyons 2008). Finally, because the units of analysis (counties) are clustered within states, the independence assumption is violated (see King 2007). Thus, to account for the potential of correlated errors and inflated values, the analyses include state dummy variables.
Results
Direct Effects of Slavery and Racial Segregation on Anti-Black Hate Crimes
Model 1 in Table 2 presents the negative binomial regression effect of slavery on the number of anti-Black hate crimes, while controlling for theoretically relevant factors. In support of Hypothesis 1, the effect of slavery (b = 1.951) is significantly associated with increases in the likelihood of anti-Black hate crimes. Anti-Black hate crimes are higher in counties where the size of the enslaved population was larger in 1860. Despite its abolition over 150 years ago, the “legacy effect” of slavery extends to the present-day victimization of Black people in the form of hate crimes. In other words, the legacy of slavery created the framework and structure within which Black people were to be viewed and treated; the findings from this study suggest that this framework and structure continue to be salient even after slavery's abolition. Instead of the explicit ownership of Black people by White people, anti-Black hate crimes are now used to express bias, hate, and discrimination toward Black people.
Negative Binomial Regression of Direct and Interaction Effects of Slavery and Racial Segregation on Anti-Black Hate Crimes.
Note: Dummy variables for states are included in the analyses but are not shown in the tables. Alabama, Delaware, Florida, Georgia, Louisiana, Maryland, Mississippi, Tennessee, and Texas are statistically significant, indicating that the state effects account for some variation in anti-Black hate crimes. *p
In addition to the direct effect of slavery on anti-Black hate crimes, Model 1 in Table 2 provides support for the contact hypothesis, which suggests that limited contact between racial and ethnic groups will foster animus between groups and that animus will present itself in several deleterious ways (Allport 1954). In support of Hypothesis 2, the analysis reveals a positive and significant relationship between racial segregation and anti-Black hate crimes (b = 1.743). Specifically, anti-Black hate crimes are higher in counties characterized by increased racial segregation. This is in contrast with work that suggests that victimization might be lower in racially segregated places because of a lack of suitable targets (see Bhavnani 2014). Instead, the findings from this analysis suggest that despite the availability of targets, anti-Black hate crimes are one of the ways in which prejudice and discrimination may manifest when racial and ethnic groups have limited contact with one another.
With respect to the control variables, two measures demonstrate a significant and negative relationship with anti-Black hate crimes: percent rural (b = − 3.122) and percent unemployed (b = -.129). More specifically, the results reveal that anti-Black hate crimes are significantly lower in rural counties. This is consistent with prior research that has found that rural locales are less likely to have crimes of all types (Disha et al. 2011). Moreover, in line with the argument that places with higher unemployed populations might not have the resources to devote to hate crime enforcement, the number of anti-Black hate crimes significantly decreases as the unemployed population increases. Several control variables exhibit a significant and positive relationship with anti-Black hate crimes: percent age 15–24 (b = .058), population density per 10,000 (b = .153), and index crime rate 5 (b = .003). Consistent with prior research demonstrating that younger people disproportionately commit serious crime (see Disha et al. 2011; Moffitt 2007), the analysis revealed that anti-Black hate crimes are higher in counties with larger percentages of the population aged 15–24. The analysis also reveals that anti-Black hate crimes are higher in counties with greater population density, suggesting that there are more available targets who are vulnerable to becoming victims of hate crimes in places that are densely populated. Also, in line with prior research, the analysis reveals that anti-Black are higher in counties that have a higher index crime rate (see McVeigh et al. 2003 and Disha et al. 2011).
Moderating Effect of Racial Segregation on the Slavery and Anti-Black Hate Crime Relationship
Model 2 in Table 2 provides the analysis for the interaction effect of slavery and racial segregation on anti-Black hate crimes, while controlling for theoretically relevant factors. Model 2 in Table 2 shows support for Hypothesis 3, indicating that the relationship between slavery and anti-Black hate crimes is amplified in racially segregated counties (b = 4.112). 6 Thus, anti-Black hate crimes are more likely to occur in places characterized by both a higher concentration of enslaved populations in 1860 and greater racial segregation. More specifically, slavery appears to be particularly consequential for the emergence of anti-Black hate crimes in counties where intolerance, discrimination, and hate toward Black people are allowed to fester because of limited interactions between racial groups. This suggests that both historical and contemporary contextual processes work jointly to influence contemporary acts of violence towards Black people.
The findings demonstrated in Model 2 are further illustrated in Figure 1, which visually represents 1) how the likelihood of anti-Black hate crimes changes depending on the level of slavery in a county and 2) how the likelihood of anti-Black hate crimes changes depending on the combined level of slavery and level of racial segregation in a county. 7 , 8 Overall, Figure 1 shows that regardless of whether a county is characterized by low or high racial segregation, the likelihood of anti-Black hate crimes increases as a county's enslaved population increases from low to average to high. In other words, the likelihood of anti-Black hate crimes is lowest in counties characterized by lower levels of slavery and highest in counties characterized by higher levels of slavery. Furthermore, within counties with low, average, and high in levels of slavery, the likelihood of anti-Black hate crimes is lower in counties with low racial segregation and higher in counties characterized by high racial segregation. Taken together, Figure 1 demonstrates that anti-Black hate crimes are most prevalent in counties with the greatest legacy of slavery and the highest level of racial segregation.

The likelihood of anti-black hate crimes at low, average, and high levels of slavery and racial segregation.
Discussion and Conclusions
The current study explores how America's “original sin"—slavery—continues to define the contemporary experiences of Black Americans. The results from this study highlight three key findings. First, slavery is significantly associated with increases in anti-Black hate crimes, suggesting that anti-Black hate crimes result from a legacy of the racism, prejudice, and discrimination created through the institution of slavery. Second, racial segregation is significantly associated with increases in anti-Black hate crimes, suggesting that prejudice and discrimination are fostered in counties where racial groups cannot create meaningful relationships. Finally, the relationship between slavery and anti-Black hate crimes is amplified in counties that are racially segregated, suggesting that the prevalence of anti-Black hate crimes is associated with both the prejudice, racism, and discrimination inherent in slavery and the fear, mistrust, and hostility that arises from racial groups being segregated. That is, the inability to build meaningful relationships with people from different racial backgrounds because of racial segregation amplifies the positive relationship between the legacy of slavery and anti-Black hate crimes.
The results from this study underscore the importance of exploring the contemporary effects of historical institutions and highlight the need to consider how historical practices are particularly relevant for understanding how race in historical contexts translates into the contemporary treatment of Black people. Although this study demonstrates how slavery set the foundation for Black people's continued victimization, more work is needed in this area to fully situate and understand race within the U.S. criminal justice system. We must continue to draw parallels between criminal justice practices and historical events, such as chain gangs, slave patrols, and lynchings. Without considering these historical practices, it will be challenging to gain a comprehensive understanding of the disparities faced by African Americans.
The work explored here provides a foundation for future research and policy recommendations that address race-related issues. First, curricula within the academic setting should not only address the existence of historical institutions but should also extend beyond the historical context of certain time periods and delve deeper into the lasting impact of institutional policies and practices, such as slavery. Restructuring how American history has been traditionally taught can challenge knowledge systems that are not wholly accurate. This has the potential to lay the foundations of social justice, blunting the blows of racism, prejudice, and discrimination built in an education system that has not always painted an accurate picture of the past. To address issues related to disparities within the criminal justice system, this knowledge should also be a critical part of trainings for criminal justice practitioners. Moreover, policymakers should also work to outlaw practices that have traditionally led to segregated cities and neighborhoods across America. For example, issues arising from practices such as redlining and gentrification should be addressed. The negative effects of gentrification include residential and commercial displacement, loss of affordable housing, and decreases in diversity; these effects are especially pronounced for low-income residents and racial and ethnic minorities (Atkinson 2000; Lees et al. 2007; Nyden et al. 2006). Efforts to revitalize these areas should not include displacing racial and ethnic minorities. Instead, there should be efforts to increase Black citizens' ability to rent and own land and housing in places that they have traditionally been denied access. The resulting increase in contact among racial groups can foster positive relationships among the groups and possibly decrease hate crimes.
Finally, a major criticism of hate crime research is whether official hate crime statistics reflect the prevalence of hate crimes or whether they represent law enforcement's willingness to enforce these types of behavior. Arguably, hate crime statistics represent both (King et al. 2009). Future research should explore the prevalence of hate crimes using different measurements of hate crimes, including the NCVS and NIBRS. In order to have more comprehensive hate crime reporting, state legislatures should require all jurisdictions to collect and report hate crimes to the federal government to be included in the UCR. Law enforcement personnel should also be uniformly trained to identify hate crimes as such. According to the Department of Justice, 17 states and Puerto Rico do not require hate crime data collection. As a result, the gathering of hate crime data has been highly inconsistent in certain places. Requiring all jurisdictions to collect hate crime data will provide a more comprehensive understanding of hate crimes. In turn, researchers can provide more accurate hate crime trends and strengthen analyses using hate crime data.
In conclusion, the findings from the current study highlight how far the tentacles of an unjust, inhumane institution such as slavery can stretch and evolve. Moreover, the protracted influence of slavery is not limited to acts of racial hate and prejudice. Although signs are no longer posted over restaurants and water fountains warning Black people that they do not belong, systems of redlining and gentrification disguised as revitalization efforts also communicate who is welcomed where. This revitalization has come at the cost of excluding Black and Brown populations from living in certain areas. Although America no longer finds it acceptable for law enforcement to turn hoses on Black citizens who peacefully protest, present-day Black Lives Matter protests are often met with excessive police force. In contrast, White citizens who riot and storm the United States Capitol are treated much more leniently. We no longer violently rip Black people from their families to till the mill; however, Black families are increasingly torn apart by the disproportionate stopping, arresting, and incarcerating of Black people. Though it is no longer socially acceptable to lynch Black people in front of angry mobs of White people, Black people still face grossly unfair treatment at the hands of law enforcement. Whether it is the killing of George Floyd or Breonna Taylor, discriminatory and racially biased treatment remains a constant fear of minority citizens. Although we do not actively cut off the fingers of Black people attempting to read and write, Black children are often relegated to underfunded and under-resourced schools, which only exacerbate existing disparities.
Reece and O’Connell (2016) posit that “history is involved in place development.” I extend this assertion by arguing that history is central to race relation development. This study suggests that slavery in conjunction with racial segregation help explain the occurrence of hate crimes targeting Black people, demonstrating that historical practices can have a demonstrable effect on the current lived experiences of Black populations. It is necessary to continue to assess critically how we have institutionalized historical practices. It is equally important to consider the impact that these policies and practices will have on future generations. Although slavery as an institution is no longer a part of American life, the racism and violence inherent in it have fundamentally shaped this nation. As a result, Black people continue to feel its impact 160 years after abolition. This work has endeavored to connect historical events with contemporary outcomes, and future scholarship should build on the findings presented here and further explore how past racial antagonism manifests itself in contemporary times. Such an approach can better situate how historical practices are associated with present-day issues of inequality in crime and justice.
Footnotes
Acknowledgments
I would like to thank Dr. Brian J. Stults and Dr. Heather A. O'Connell for their helpful contributions in advancing this project.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article
Notes
Author Biography
Appendix 1. Correlation Matrix
| Slavery | Racial Segregation | Percent Age 15-24 | Percent HS Degree | Percent Rural | Percent Democrat | Percent Unemployed | Population Density per 10,000 | Index Crime Rate | Legislative mandate | |
|---|---|---|---|---|---|---|---|---|---|---|
| Slavery | 1.0 | |||||||||
| Racial Segregation | −.77*** | 1.0 | ||||||||
| Percent Age 15-24 | .18*** | −.25*** | 1.0 | |||||||
| Percent HS Degree | −.05* | .11*** | −.28*** | 1.0 | ||||||
| Percent Rural | −.02 | .05 | .−24*** | .47*** | 1.0 | |||||
| Percent Democrat | .34*** | .52*** | .15*** | .02 | −.04 | 1.0 | ||||
| Percent Unemployed | .20*** | −.39*** | .49*** | −.10*** | −.09*** | .42*** | 1.0 | |||
| Population Density (per 100,000) | −.03 | .05 | .03 | −.37*** | −.47*** | .12*** | −.05 | 1.0 | ||
| Index Crime Rate | −.02 | −.08** | .06** | −.33*** | −.44*** | .10*** | .03 | .59*** | 1.0 | |
| Legislative Mandate | −.08** | .16*** | −.01 | −.10*** | −.10*** | −.27*** | −.18*** | .07** | .04 | 1.0 |
Note: *p
