1

Setting the Foundation: Studying Race and Structural Racism Responsibly

Welcoming participants to the first session of the workshop, discussant Trevon Logan (workshop planning committee member and Hazel C. Youngberg distinguished professor of economics at The Ohio State University) observed the growing interest among population health researchers in understanding the effects of structural racism on material conditions and outcomes at particular points in time and throughout the life course. He emphasized the value of social scientists learning from humanists and humanistic social scientists about race, racism, race-making, and structural racism as dynamic processes.

RACE, RACE-MAKING, AND THE USE OF RACE TO CONTROL POPULATIONS

Use of Narrative and Metaphor

Presenter Stephanie Li (Lynne Cooper Harvey distinguished professor of English at Washington University in St. Louis) championed the use of narrative and metaphor to better understand the complexities of race, which she described as “a social construction that determines what many perceive as essential aspects of identity” and that influences how individuals exist and relate. Reflecting further on the meaning of race, she noted that race has been categorized as an ideology, a structure, a history, a community, and an identity; however, class, gender, sexuality, and nationality “collude with race to make an indelible mark of social difference [that is] just out of reach.” She underscored that race remains the most important determinant of life outcomes, such as residence, salary, and life expectancy.

Given the difficulty of defining race, Li explored several insightful metaphors presented by literary scholars. For instance, she indicated that James Baldwin (1984) described Whiteness as a form of blindness to the violent history that “branded its inequalities” into the nation’s landscape. In Invisible Man, Ralph Ellison (1952) portrayed Black subjectivity as a “condition of invisibility.” Claudia Rankine (2015) compared Black life to a “condition of mourning” in a world in which a person can be killed for being Black. In other words, Li explained, Blackness “is inextricable from the imminent possibility of death . . . and is to live without shelter.” She also described the work of Edwidge Danticat (2016), who compared Black people in the United States to refugees, “as though [they] were members of a group in transit . . . who should either die or go somewhere else.” Li remarked that although these metaphors do not fully capture the complexities of race, they offer “modes of understanding” that reveal the injustices of the world, as all of these perspectives connect Black life to a lack of safety, as well as to a “systematic devaluation.” She emphasized that data on incarceration, the achievement gap, housing inequalities, and health disparities are readily available to support such narratives.

Asserting that racism infects all aspects of life, Li shared examples of the ways in which people communicate the danger and anxiety associated with race. For example, Ta-Nehisi Coates (2015) recounts being unable to enjoy an evening with a new friend because he was anticipating an attack that did not occur, in part because his “eyes were made in Baltimore . . . blindfolded by fear.” Li described Coates’s inability to be comfortable with a White stranger, even though he was never in danger, as an ingrained “anticipatory stress response,” having been taught that all White people pose a threat to Black people in a city that was segregated and violent, with

poor educational and job opportunities. Li underscored that the inability to trust a White person is the “cost of the vigilance required to exist” as a Black man in the United States; therefore, Coates’s narrative reveals the detrimental consequences of structural racism on basic human relationships. However, Coates’s writings on President Barack Obama describe a different experience with trust; the “kinds of traumas that marked African Americans of his generation . . . were mostly abstract for him” (Coates, 2017), owing to a different upbringing. Coates suggested that President Obama trusted White America, unlike many African Americans, who are “too crippled by [their] defenses” (Coates, 2017)—the same defenses that enable survival, Li observed.

In closing, Li explained that although racism begins in history and policy, it “resides in our bodies and our eyes.” She recalled Toni Morrison’s (1998) envisioning of race as a physical structure—a house—that defines the landscape and threatens to restrict movement. However, instead of escaping, Morrison aimed to transform this structure from “a windowless prison into which I was forced” to “an open house, grounded, yet generous in its supply of windows and doors” and further to “an out of doors safety where a ‘sleepless woman . . . could walk out the yard and on down the road. No lamp and no fear’” (Morrison, 1998, p. 4, 10). Li indicated that this freedom of movement symbolizes the liberation of race (versus its containment), and that Morrison challenged people to rebuild the structures of their lives, anchored in the strength of community instead of restricted by the absence of safety.

How Race Is Made and Remade Over Time

Serving as the session’s second presenter, Evelynn Hammonds (Barbara Gutmann Rosenkrantz professor of the history of science, professor of African and African American studies, and professor in the Department of Social and Behavioral Sciences at Harvard T.H. Chan School of Public Health, Harvard University) described her interest in historicized understandings of race and racism, as well as in the notion that race is continually being “made and remade” over time by custom, law, and scholarship.

Hammonds provided key definitions to introduce the concept of race and to frame the workshop’s discussions of structural racism (see Box 1-1).

Reflecting on the murder of Michael Brown in Ferguson, Missouri as a case study in racecraft, Hammonds highlighted the flaws in the explanation that Brown was shot because he was Black, which “veil[s] the work of multiple forms of racism that led a law enforcement official to shoot this young man to death. . . . Brown’s blackness did not pull the trigger. . . . Brown was not shot because he was black. He is black because he was shot” (see Benjamin, 2014). In essence, “race is the result of the power some people

have over others” (see Fields and Fields, 2012). Hammonds explained that the idea of inherent racial difference has continued to shape perceptions of race and articulations of these perceptions in contorted ways across centuries. Thus, understanding the historical nature of race and racism demands “map[ping] the relations of power, the patterns of contestation and struggle out of which such social constructions emerged” (see Holt, 2000). Hammonds reiterated that there is no single social construct of race.

Tracing historical discussions of race, Hammonds described W.E.B. Du Bois’s contribution of 60 data visualizations in 1900 to an exhibit in Paris that focused on the progress of African Americans since Emancipation. One of these infographics, “Valuation of Town and City Property Owned by Georgia Negroes,” plotted the value of property owned by African Americans from 1870 to 1900 within the context of political and socioeconomic events—the rise of the Ku Klux Klan in the 1870s; industrialism in the 1880s; and lynching, financial panic, and disenfranchisement in the 1890s. Property value increased until approximately 1900, at which point a decline began. This visualization prompts its viewers to consider the historical context in which African Americans acquired property:

Hammonds underscored that Du Bois collected data on a wide range of topics related to the actual lived experiences of African Americans in the context of a society that was structured by racial inequality and that reproduced this inequality over time. Du Bois addressed and raised problematic questions about race, even though society was not ready for this type of research at the time. Hammonds encouraged contemporary scholars to embrace Du Bois’s model of structural racism, as such social data illuminate the structural components of how race and racism are made and remade over time.

EMBRACING THE COMPLEXITY OF STRUCTURAL RACISM AND UNDERSTANDING THE INTERLOCKING FEATURES OF CULTURAL AND STRUCTURAL RACISM

The Racialized Social System

Presenter Eduardo Bonilla-Silva (James B. Duke distinguished professor of sociology at Duke University) explained that, while the use of terms such as systemic racism and structural racism has increased, much of society still fails to recognize how racism is systemic. To illustrate the systemic nature of racism, he presented a brief analysis of the police force, beginning with its history as an extension of slave patrols. Currently, he continued, as an “agency of racial and social control,” the police force chooses particular people to be officers and trains them in a racialized way, thus creating a “macho-military culture of ‘us’ versus ‘them’” and steering officers to use race-based policing. According to Menifield and colleagues (2019), this “explains the seeming contradiction of officers of color being as likely as their White counterparts to use lethal force against people of color.”

Bonilla-Silva indicated that this lack of understanding about the systemic nature of racism persists in part because society incorrectly conflates racism with prejudice. First, prejudice focuses on individuals’ psychology or attitudes, whereas structural racism is collective and societal, and extends beyond attitude to create an ideology. Second, the notion of prejudice is ahistorical (i.e., assumes the racism of today is not different from that of yesteryears), whereas structural racism has a historical beginning, retains

a set of practices that can change over time, and includes variations across locations. Third, prejudice is thought to revolve around overt actions, which neglects the common and covert ways that race matters. Lastly, prejudice focuses on the flawed morals of an individual, whereas structural racism has a material foundation (see Bonilla-Silva, 1997). Thus, he asserted that thinking about racism as a form of prejudice makes it impossible to understand the drivers of structural racism.

Although several alternative approaches to understanding structural racism exist, Bonilla-Silva emphasized the value of the “racialized social system” approach, which he developed in 1997. He explained that this theoretical framework is based on the following multidimensional premise: “the world-system was racialized in the 15th century, creating racialized social systems” in which “social, economic, political, and even psychological goods have been partially allocated by race.” Furthermore, race and racism are “social and political constructs that are mutually reinforced.” In other words, Bonilla-Silva continued, “race and racism coemerged and are codetermined.” Although races are constructs, they are “socially real,” he said, because belonging to the White race has positive consequences and belonging to a non-White race has negative consequences. As a result, races “develop different racial interests,” with the subordinate groups challenging their position in the system and the dominant groups defending the racial order. This creates an opportunity for racial contestation, which is “the struggle for position in the racial order, which transpires infrequently through concerted collective action . . . but often through individuals’ actions . . . or mostly through actors following the dominant racial script of a period” (see Bonilla-Silva, 1997). He stressed that any worthwhile structural theory of racism should recognize the collective practices and behaviors of members of a society; be both tied to history and cognizant of regional, local, and societal distinctions; be materialist; and consider individuals and their subjectivity, as well as how the racial structure is produced and reproduced.

Bonilla-Silva has further explored the complexity of structural racism and expanded the theory of the racialized social system in the years since his 1997 publication (see Bonilla-Silva, 2021). He explained that because “regular White folks” are fundamental to the maintenance of the racial order, understanding the “collective manufacture of Whiteness” is critical. He noted that the “White Habitus” molds individual Whites into Whiteness—the “hypersegregation” of White life (e.g., in residences, churches, friendships) reinforces Whiteness as a set of norms, culture, aesthetics, emotions, and cognitions (Bonilla-Silva, 2019, 2022). The Whiteness produced by this White Habitus, he continued, has become systemic. He underscored, however, that the “production of Whiteness and Blackness is always contingent”; in addition to being racialized, people are categorized by class,

gender, sexual orientation, political orientation, education, and levels of interaction, which affects their “racial sensibilities.” Therefore, Bonilla-Silva remarked that collective action through racial contestation is crucial to enable fundamental structural and cultural change.

Bonilla-Silva also described how understanding the systemic nature and historical context of racism is key to the methods and indicators society uses. For example, he discussed how an interpretation of residential segregation is based mostly on the index of residential dissimilarity¹ and the index of isolation. Using these indices to study racial segregation in neighborhoods, Vigdor and Glaeser (2012) revealed a decline in residential segregation since its peak in 1970. However, Bonilla-Silva emphasized that much progress remains to be made; although historical segregation patterns are beginning to change, they are changing because of gentrification, which only creates different realities within the same space. As an example, he portrayed Durham, North Carolina, as a city in which “whitopia” still dominates in places where Black and White people cohabitate spatially. In closing, Bonilla-Silva provided the following guidance to researchers conducting analyses of residential structural racism:

1. Because context and history matter, do not reify metrics;
2. Recognize that spaces and organizations are racialized;
3. Measure interracial contacts and their valence;
4. Examine (instead of assume) racial life in spaces such as neighborhoods (see Mayorga-Gallo, 2014); and
5. Consider power dynamics and the implications of arguments.

Defining and Measuring Cultural and Structural Racism

Serving as the final presenter of the session, Margaret Hicken (workshop planning committee member and research associate professor in the Institute for Social Research at the University of Michigan) explored how racism has been understood, measured, and modeled in population health and aging research, as well as how this aligns with conceptualizations of racism. She underscored that interdisciplinary scholarship—integrating research from the arts, humanities, and social sciences into population health research—creates stronger science, especially because public health and biomedical training are often misaligned with the reality of how structural racism shapes health and well-being. Reflecting on how researcher bias can influence this scholarship, she noted that racially diverse research teams are also highly beneficial. She emphasized, drawing from work by legal expert

___________________

¹ The index of residential dissimilarity measures the evenness of population distribution in a geographical area.

Charles Lawrence, that those with White privilege are often blind to the ways that structural racism affects population health, while scholars without White privilege have key insights about racialized processes. Without these diverse research collaborations, she cautioned, researchers cannot fully understand the processes they intend to measure—and science will not move forward.

Hicken expanded on the definitions of racism offered earlier in the workshop, first sharing philosopher Achille Mbembe’s (2003) interpretation that racism is “a technology aimed at permitting the exercise of biopower”—the tool that allows society to “regulate the distribution of death.” More specifically, Hicken described cultural racism as the socially accepted values, ideologies, and norms of a racialized society that are determined by the dominant power group. Cultural racism operates in the shared social subconscious and determines assumptions about who and what are important—that is, it shapes the answers to questions: Whose life counts? Who is fully American? Who deserves to live a long and healthy life? She explained that cultural racism also “acts as a distortion lens that renders racialized and racially hierarchical institutions neutral and rational.” Thus, Hicken continued, structural racism is the application of cultural racism; the social structure is composed of formal and informal interrelated institutions, and that when attempts are made to achieve equity in one institution, other institutions intervene to restore the White privilege set up by cultural racism. Furthermore, she indicated that society’s institutions “adapt to contemporaneous sociopolitical norms.” These institutional shifts will be replaced by others in a more civilized way of killing, according to Mbembe (2003), if the underlying cultural racism does not change. Hicken underscored that historical race-based policies continue to influence current policies because structural racism “includes the erasure of historical processes that could clarify the link between racialized groups and health.”

Turning to a discussion and evaluation of three specific strategies to measure and link cultural and structural racism, Hicken first depicted an approach that leverages individual-level reports and information—for example, reports of interpersonal prejudice or discrimination, anticipatory and perseverative thoughts and behaviors, beliefs about external regard for racialized groups, area-level composition of individual-level reports, skin tone, and documentation status. This type of information can be collected easily through interviews but is not meant to be a proxy for cultural or structural racism. She emphasized that these measures require a theory and a framework on how the constructs relate to cultural and structural racism within a particular context. Although each type of individual-level measure could be useful and could reveal possible connections between race and health, she continued, it is important to avoid mischaracterization of the measures, whose linkages to race and racism vary over time and place.

Hicken next described an approach that relies on multi-item indices, which are created by combining indicators in different ways to capture a construct. She asserted that an index significantly reduces the available information about constructs such as cultural and structural racism. For example, the critical information for a universal index of structural racism would be limited to a small number of formal institutions for which administrative data at a predetermined spatial level are available, by many unsubstantiated assumptions about the ways in which the components operate together over place and time, and with a narrow snapshot of contemporary life. These indices are difficult to interpret, she remarked, and thus create associations with health that are difficult to interpret. In some cases, the creation of indices may be useful in facilitating particular research questions, but she cautioned that indices are not indicators of structural racism.

Hicken commented that the field is moving toward more frequent use of an approach that captures specific features to better understand cultural and structural racism at local levels, both spatially and temporally—for example, via racial segregation (global and local residential, historical residential redlining, educational, and occupational); contemporary racial terror, surveillance, and control (police killings, mass incarceration, fines, and fees; child protective services; and vigilantism); and historical racial terror, surveillance, and control (mob violence, lynchings, enslavement, confederate monuments, Ku Klux Klan activity, and Jim Crow-era governance).

Hicken summarized that the key path forward for structural racism research centers on building interdisciplinary frameworks by integrating scholarship from the arts, humanities, social sciences, and population health; shifting away from atheoretical tests of racial group comparisons; and allowing for dynamic interactions among institutions. Further, underlying these frameworks and tests is the assumption that society is not moving toward equity. She championed the value of matching measurement and modeling to theory by reducing the use of universal, static, and temporally narrow indices; developing measures that reflect the spatially and temporally local nature of racism; embracing modeling approaches that allow for dynamic feedback loops and interactions among institutions over place and time; integrating historical information that could capture unmeasured or unmeasurable information about contemporary structures; and focusing on what information is actually captured by a measure, no matter what the measure may be labeled. She reiterated the value of creating diverse working groups and then challenged the current definition of academic success that privileges publishing alone or as lead author to achieve tenure and promotion, and the practice of waiting to discuss structural racism until tenured.

REFLECTIONS AND DISCUSSION

Continuing to serve as the session’s discussant, Logan expressed his support for Hicken’s assertion that scholarship from the humanities and the humanistic social sciences be integrated into population health and quantitative social science studies of structural racism—this interdisciplinary research could strengthen the theories that support the measures used as proxies for structural racism. He added that, because social science is ahistorical by nature, historians offer key insights into race as a dynamic process, as well as how people have understood race over time and place.

Logan offered further reflections on the first session of the workshop and on the topic of structural racism more broadly. He recalled sociologist Dorothy Roberts’s understanding of race as a political construct, which focuses on the division of resources among people through the political process, and noted that this definition captures another aspect of the complexity of structural racism because it reveals how race is operationalized at the individual and institutional levels, where power dynamics reign. Lastly, he noted that because individuals define themselves relative to other individuals, a relational process would help to better understand structural racism.

Opening the general discussion, Logan posed the following question: What makes the social construct of race real? Hammonds explained that because the United States was defined by the exclusion of various people from the body politic, “social” and “political” constructs have not been and cannot be separated. Furthermore, she continued, the biological component of race underlies both the social and political constructs of race. She referred to a letter that W.E.B. Du Bois received from a White physician in 1906 asking if “the Negro shed tears,” which illuminates the fact that meaning always has to be considered in context—there is no one way to define a social (or political) construct.

Logan wondered how to interrogate the data used to build measures that could capture structural racism. Hicken replied that the first step is to develop a research framework shaped by the arts, humanities, and social sciences. She noted that existing data are racialized, and researchers have access to a limited set of data—without a research framework, these data are difficult to interpret accurately. Next, she continued, researchers could use this framework to consider what the data have captured to better understand what is being measured and to create more appropriate research questions.

Frank Edwards (assistant professor of criminal justice at Rutgers University) described research from postcolonial theorists that reveals gaps in archives, which lead to gaps in the understanding of the forces that structure contemporary outcomes. Furthermore, he pointed out that data

collection is an “activity of power”; for instance, official statistics on police violence in the United States do not exist. In the context of these gaps in exposure to forms of structural racism, he asked how public health researchers could measure historical processes of structural racism more appropriately. Bonilla-Silva advocated for “undoing our silence” in the archives—for example on colonialism, genocide, and land expropriation in the United States. Jennifer Manly (workshop planning committee member and professor of neuropsychology at Columbia University) cautioned researchers against becoming too distracted by machine learning and artificial intelligence. More and better data are needed, she said, but when some data are missing and other data are imperfect, creative approaches help to better understand how structural racism operates.

Logan observed that measures of segregation are often focused on metropolitan areas, and he inquired about capturing geospatial aspects of systemic racism that might not align with existing theories. Bonilla-Silva suggested analyzing local racial formations to examine how the production of racial order varies depending on rural, urban, large, and small populations. He mentioned that understanding the “rules and regulations” that maintain racial order in a particular location are as important as having better metrics. In Latin America, for example, the historical racial order is stronger than in the United States, which explains in part why income and education data reveal that the gaps between White people and non-White people are slightly larger there than in the United States. He suggested developing new indices for these situations, as well as studying segregation in situ.

A participant posed a question about avoiding the use of measurements that reinforce White normativity and assimilation as solutions for anti-Black racism. Bonilla-Silva explained that the index that was useful 40–60 years ago is much less useful now in depicting the ways in which segregation matters. He suggested that instead of “hunting for racists,” researchers could examine the “depth of Whiteness”; such new metrics could produce a more robust measurement of structural racism. Hicken noted that consideration for spatial resolution is essential. Once the measure and spatial resolution are matched to the time and place where the research question is being asked, she said, segregation can be understood as a tool of structural racism—public and private entities systematically invest in some people while disinvesting in others. She encouraged researchers to evaluate how they are measuring segregation, whether this matches with the theory, and at what level segregation, which depends on place and time, is happening. Once the locations of investment and disinvestment have been identified, she continued, it becomes possible to target policy accordingly. Logan added that because the most popular measures of segregation in the social sciences are based on income inequality, researchers could consider what to

measure, in a geospatial sense, that would apply to an income distribution. He encouraged researchers to study segregation in rural areas, occupations, and schools moving forward.

Another participant asked whether intersectionality helps or hinders understanding of racism. Li responded that although intersectionality is critical to understanding the effects of racism, its presence also creates challenges when other forms of discrimination, such as sexism, classism, and ageism, cannot be isolated. Thus, intersectionality is valuable as an approach, she continued, but its impact is difficult to measure. Hammonds added that intersectionality is crucial, but different data (e.g., disaggregated data) would enhance understanding. For example, over the last 10 years, approximately 68 percent of women seeking abortions in Mississippi have been African American, but it is unclear why (Fadel, 2022). To understand this finding and ultimately to make related policy changes, the systemic forces affecting those women (e.g., lack of access to public transit, primary care, and health insurance) have to be understood first; if one wants to make possible safe and legal access to abortions, it is necessary to understand why the highest rates are happening for women of color. Logan pointed out that this example relates to the workshop’s earlier discussion about Michael Brown, and Hammonds reiterated that context is essential to reveal the structures that, for instance, prompt a law enforcement officer to see a Black face and think that the person can be killed with minimal interrogation about the crime. If race is relational, she added, then questions arise about what it means for White people to be White, as well as about the structures that produce Whiteness.

Li returned to the question that began this discussion—why and how these social constructs exist—and explained that they exist because of the narratives that underpin American identity and the founding of the nation, which relate to the concept of “White innocence.” She noted that all of the achievements in the United States are based on the foundation of freedom and the pursuit of happiness, which in reality are “built on mass exploitation and genocide and the plunder of bodies of color.” She emphasized the importance of challenging this notion of White innocence and changing the narrative that Blackness equates to violence, abjection, and disposability.