May 31, 2021 1:24 pm
The massacre against the black community of Tulsa, Oklahoma, lasted less than twenty-four hours, but has been causing damage for decades. A century later, the extent of the economic losses suffered by African Americans is still not clear, but important progress has been made recently, also thanks to new digital tools that allow scholars to extrapolate data on the effects of the massacre from the census registers. .
On May 31, 1921, a crowd of more than a thousand white men showed up in front of the prison where a black boy suspected of molesting a white woman was being held. More than fifty African American men arrived to help the police defend the building. They were sent away and retreated to Greenwood, the predominantly African American area. Towards evening the white crowd stormed the neighborhood, burning down houses, shops, churches, a school and a hospital. Despite efforts by local authorities to cover up the violence – at least 300 African Americans are estimated to have died – journalists, historians and residents of Tulsa have managed to document what happened to the city’s wealthiest blacks and their properties. Entrepreneurs John and Loula Williams lost the Dreamland Theater. The Gurley Hotel, owned by Ottowa and Emma Gurley, was razed to the ground.
Yet so far it has been difficult to calculate the long-term economic effects of the massacre. How can the consequences of racist violence on culture and institutions be measured years after the violence? How did the massacre change the social and economic trajectory of Tulsa’s blacks?
We found some answers by analyzing the censuses made before and after the massacre. From the archives we have extrapolated individual stories that reveal the economic hardships of people who remained in Tulsa after the events of 1921, such as those of Gibson Van Dyke and Roy Drew. In 1920 Van Dyke, a 36-year-old black man, ran a guesthouse in Greenwood. He and his wife Bertha were among the few blacks to stay in the city. Ten years later Gibson was no longer in his business, and was working as an unskilled bricklayer on an extremely low salary. The Van Dyke trajectory contrasts with that of Roy Drew, a black man who worked in a general store in 1920. In 1930, Drew appears in the Kansas City records as a pharmacy owner.
The archive data also tells a broader story. The 1920 census provides a comprehensive picture of Tulsa’s reality one year before the massacre. At the time the city was an agricultural hub and enjoyed large profits from oil, so much so that it was renamed “Magic City”. Tulsa was a peculiar city because the wealth was shared by whites and blacks.
Greenwood earned the nickname “Black Wall Street”, not for financial institutions but for its wealth. In 1920 it was rare to come across a distinct community of wealthy blacks. Black children in Tulsa went to school more than those in other cities in the region. Before what historians consider to be the worst outbreak of racist violence in US history took place, a robust middle class was born in Greenwood.
Subsequent census data show a stark contrast to the prosperity of the decade. The income of black families progressively decreased from 1920 to 1930 and then up to 1940. How can we establish that the contraction was due to the massacre and not to the change in local or national conditions? History is not a laboratory to carry out controlled experiments, but comparing the data of Tulsa with those of some cities similar to Tulsa – such as Oklahoma City, Wichita and Little Rock – it emerges that in these three cities the incomes of blacks in the same period remained constant, despite the impact of the great depression. In 1920, the earnings of Tulsa’s African American households were 9 percent higher than the 14 cities surveyed in our analysis. In 1940 they were 7 percent lower. In the same period, the evolution of the incomes of whites in Tulsa followed the same evolution as the other cities: it remained constant from 1920 to 1930, before recording a small decline from 1930 to 1940. Overall over the two decades, the incomes of Tulsa whites were States taller compared to the cities considered.
The Tulsa that was rebuilt after the massacre was very different from that of 1920
The ability to process large amounts of data is revolutionizing historical knowledge, because it allows us to answer crucial questions about communities and societies in ways unthinkable with previously used techniques, which were based only on individual histories. The United States Census, in particular, is a valuable record for reconstructing past economic trends, especially those relating to the lives of ordinary people. Furthermore, the digitization of the archives allows us to investigate specific contexts by analyzing the life of each individual and collecting complete information on age, ethnicity, family ties, housing and employment. Although the census has always been a decisive source for understanding the conditions of specific places on an individual level, today machine learning-based techniques allow researchers to follow large numbers of people over the years, discovering who has remained in the same place. locations and those who have moved in search of a better life.
According to archival documents, the Greenwood neighborhood was rebuilt a few years after the massacre. Many buildings were rebuilt in the same position and style as the original ones, giving the impression of a return to normalcy. A local historian wrote that “everything was more proposed than before”. But that was not the case: the rebuilt Tulsa of 1930 was very different from that of a decade earlier.
The damage is evident not only by analyzing incomes but also by looking at the indicators of social well-being. By 1920, blacks from Tulsa were more likely to own a home, be married, and have a job than blacks from other similar cities. In 1930 and 1940, however, they were lagging behind by all indicators. Our model indicates that in 1940 the probability of a black man from Tulsa being married was 6 percent lower, and the chance of having a job was 9 percent lower. Furthermore, the average income was 12 percent lower than in a hypothetical context in which the massacre had never occurred.
Violence has changed the social structure of the city. Some wealthy blacks left Tulsa. Ottowa and Emma Gurley, for example, moved to Los Angeles, where they opened a new hotel. But how prevalent was the decision to leave?
A comparison of who stayed and who left reveals that Tulsa black men with high salaries were more likely to leave the community after 1920 than Tulsa whites and blacks from similar cities. This migration brought about a substantial change in Tulsa, a city that before 1920 attracted many blacks from other parts of the country.
The techniques of analysis also allow us to observe the effects of the slaughter over the generations. Our research linked the children who lived with their parents in 1920 to their condition in 1940. The massacre and its aftermath had profoundly different effects on black children whose families remained in Tulsa than those who left. In 1940, those who had chosen to leave after the massacre 88 percent more than those who remained. Conversely, the effects of migration for the whites of Tulsa were negligible and in any case negative.
Our data clearly shows the opposite fortunes between those who stayed and those who left. Take the example of Clarence Knight, who in 1920 was a tailor in Greenwood and whose wife, Daizy, worked as a maid. The Knight family remained in Tulsa after the massacre. Their son, Ralph, did not finish high school and in 1940, at the age of 28, he was working as a bellboy in a hotel. Blacks who left Tulsa found better opportunities to thrive. In 1920 Sylvester Holman, a resident of Tulsa with his wife Annie and children, worked as a porter in a haberdashery. Twenty years later one of his sons, Clarence, was registered in the Nashville, Tennessee census: he had graduated from high school and embarked on a career as an insurance agent.
The scars of 1921 are still very deep in the city of Tulsa and in the people who experienced the massacre firsthand. The data captures an important aspect related to the long-term economic consequences of racist violence in the United States. Within just 24 hours, the promising trajectory of an entire community was radically altered, causing effects that lasted for generations.
(Translation by Andrea Sparacino)
International has a weekly newsletter that chronicles what’s going on in the United States. You sign up who.