A Mathematical Explanation of Oppression
About a year ago, angry protestors filled numerous city streets across the United States holding signs that said, “Black Lives Matter”: a simple but somehow controversial statement. Ever since then, social media has been flooded with stories that continue to demonstrate that America being a society “beyond race” is nothing but a myth. Yet, some people are still not convinced that racial oppression is ingrained into American society.
When people think of complex social phenomena such as oppression, STEM rarely comes to mind as a field that could help explain these issues. However, mathematical modeling can help us objectively view past an inherently subjective society to identify and quantify controversial issues such as oppression. As Dr. Michael Olinick, Professor of Mathematics at Middlebury College explained in an interview, mathematics can be used as a tool for investigating patterns and deriving consequences from assumptions.
Dr. Olinick described mathematical modeling as a surprisingly straightforward process: First, take a problem in the real world that you hope to make sense of and write down your assumptions of the overall behavior of the issue. Then, utilize careful mathematical analysis to derive the consequences of such assumptions. Once the assumptions and consequences derived align with real-world observations, the model is complete. It’s similar to a chef creating a new recipe: they will experiment with different ingredients until they have a dish that is pleasing to their senses. However, the math can get more difficult to work with as assumptions get more sophisticated. Nonetheless, mathematical modeling can give us a different lens to look at situations where verbal analyses are challenging to interpret.
In 2013, a computer science professor at the University of St. Andrews named Ian Gent endorsed the mathematical model of another computer scientist named Karen Petrie. Her model, the Petrie Multiplier, shows that oppression doesn’t necessarily happen by malicious intent. Rather, it more so has to do with diversity. For example, in computer science, only 20% of professionals are women. If, in theory, men and women both made an equal amount of sexist remarks to each other as assumed in Petrie’s model, women would still receive more sexist remarks simply because there’s a greater proportion of men to produce these remarks.
While the Petrie model makes sense in a contained environment, it doesn’t account for the real-world disparity in workplace sexism experienced by men and women. In 2018, researchers from UC Berkeley published a research article in the Proceedings of the National Academy of Sciences, where they quantitatively demonstrated how stereotypes translate to unfair treatment. Similar to sexist remarks, stereotypes form by inferences about people’s social roles. In one instance, they applied their model to a U.S. study on how professors responded to mentorship requests. They found that the professors’ responses to the requests were predictable based on the race, ethnicity, and gender of the students. The major takeaway from the researchers’ modeling is that even from a quantitative standpoint, it can be seen that behavior influenced by stereotypes reinforces social inequities that occur based on race, ethnicity, and gender.
Research shows that structural biases toward specific groups, many of which are stemmed from stereotypes, can reinforce social inequities. In 2019, researchers from Columbia University, Princeton University, and Harvard University utilized agent-based simulation (a computational model that simulates interactions of entities to understand the behavior of a system) to demonstrate that “a key contributor to underrepresentation is the increased costs with being underrepresented in the first place.” The researchers’ simulated system illustrated that in male-dominated environments, such as computer science, women aren’t just more likely to receive sexist comments. They are also affected by sexist comments when objections to the comments don’t happen. This is because the comments naturally lead to the condition of the system worsening over time and perpetuate further imbalance of the system in question. Even when researchers implemented an equality policy to biased systems, the situation did not get better for the women because it didn’t directly address the issue at hand: the sexist men.
Since the models previously explained are highly generalized, they can also be applied to the relationship between Black and White people that was previously alluded to. The ratio of Black people to White people in the United States is approximately 1:6, which is even higher than the ratio of men to women in computer science. As a result, according to the Petrie Multiplier, Black people are 36 times more likely to receive racist remarks assuming that both groups are equally racist. If the agent-based model mentioned previously were applied to the system of Black and White people in the United States, it would most likely be found that without objections to the racist remarks, the system will only worsen over time, incurring a large cost to Black people and normalizing oppression. Therefore, the models described in this article could offer a possible explanation for why we observed the events of last summer. When the costs to a minority group become so high, they eventually erupt in frustration.
While this may appear to have dreary implications, the good thing is that people have the capacity to change over time. As Dr. Scott de Marchi, Professor of Political Science and Director of the Decision Science Program at Duke University said, “People are super clever… in ways that they don’t often get credit for.” Mathematical modeling can only paint a cartoon image of oppression based on current conditions. So, even if we don’t currently live in a “post-racial” society, it doesn’t mean that we don’t have the capacity to eliminate stereotypes, preferences, and prejudices from our society. When we work together to identify and reduce the factors that encourage the continuation of oppression, we are actively making an effort to reduce it. It’s all in the math.
