A version of this article was originally published in the Researcher's view segment on the University of Helsinki website under the title Understanding systemic discrimination patterns is the key to a more just world.
Patterns of discrimination, violence and inequality are some of the leading problems in today's world and major obstacles to achieving a more just world. Social customs have historically dictated that, for example, white males are entitled to hold property rights, while females and ethnic minorities have been given restricted rights. Women and ethnic minorities have themselves been classified as property in institutions of slavery and coverture. While in recent decades, legislation across the globe has sought to address these inequalities and human rights mechanisms have paid special attention to the rights of the most marginalized populations, domination of minorities by majorities has by no means ceased to exist.
In the modern world, women and gender-nonconforming individuals continue to face violence based on gender stereotypes, the pay gap between men and women has not been resolved, minorities face arbitrary police violence in the US and elsewhere, and ethnic minorities in practise lack state recognition for their cultures. These are just some examples of how discrimination and inequality continue to manifest themselves. What these manifestations have in common is how binary tags are applied. By binary tags, I refer to arbitrary markers, such as skin colour or gender, that people use to classify others into ingroup and outgroup members. While of course I do not suggest that aspects such as gender and ethnicity are arbitrary facts of individuals’ identity, they are arbitrarily used to divide people into categories of inferiority and superiority and to justify systems of hierarchy and domination.
Although these types of divisions based on binary tags have been condemned officially, they continue to shape social relations. As a researcher, I am interested in studying the impact of binary tags with scientific rigour and in finding a way to illustrate the mechanisms underlying systemic discrimination. This is important, because as societies, we must understand how systemic discrimination and violence operate in order to tackle them. This article seeks to provide a glimpse into my work using computer simulations which contribute to the debate on how systemic discrimination operates. Hopefully, too, it may be applied to the process of finding solutions on how we can create a more just world for all.
My research begins with the concept of social cooperation. Social cooperation, how to best achieve it and why it fails, are extremely important themes in social science research more broadly as well. Different models, such as the Checkboard model and the highly influential Prisoner’s Dilemma model have been commonly applied to study when and why cooperation fails. In my work however, I have been interested in another model called the Hawk-Dove game, which allows placing power at the centre of the analysis. I am particularly concerned with what happens to cooperation when an arbitrary binary tag is introduced.
To grasp the idea of the Hawk-Dove game, imagine two cars that are facing each other. Their aim is to intimidate the other off the road. When the flag is waved, the two cars drive towards each other. There are three possible outcomes: first, the winning car (the Hawk) stays on the road while the loser (the Dove) swerves off the road; second, both cars swerve off the road and there is a draw (Dove-Dove); or third, neither car swerves and there is a collision (Hawk-Hawk).
The Hawk-Dove game was central to my initial theoretical research on the impact of binary tags on homogenous populations. I presented this research in 2020 while working as a Berggruen Fellow at the Center for the Advanced Studies of the Behavioral Sciences at Stanford University. After returning to Finland, I developed a computer model to test my theoretical predictions.
The findings were insightful. The simulation (which you can see here), based on the Hawk Dove game, shows that when the population is homogeneous (meaning that members of the population are alike to each other), then decisions of whether to attack (be a Hawk) or to swerve (be a Dove) are random and strategic players make the choice on a case-by-case basis. However, when players are divided into two groups: red and blue, one group will always come to dominate over the other: attacking instead of swerving. Once this pattern becomes established, it is then impossible for one individual or a small group to change the status quo. These inevitable dynamics result in systemic discrimination and hierarchy within the game. You can read more here, and see the lecture here.
Initially, I relied completely on my computer model to develop these findings. However, scientific research depends on others being able to replicate the results. The novelty of the study attracted the attention of Dr Christopher J. Watts. He built his own computer simulation which enabled us to agree on the outcome of the research. Working together we showed conclusively that in populations with identical individuals, only differentiated by an arbitrary tag, majorities benefit from intense threats of violence. These could range from physical harm and incapacitation to property damage. Alternatively, minorities can benefit from lesser threats of harm. These could include verbal insults and hostile actions that are symbolic gestures. However, given the possibility of escalation of violence, and the fact that our models show that majorities gain the upper hand by this tactic, majorities will likely prevail over minorities.
The image shows that the majority tends to dominate when violence is intense, and minorities tend to dominate when violence is of low intensity. The pink cross in the figure shows that when the two groups are each 50% of the total population, and violence is mid-range, it’s a draw whether one group or the other dominates. Nonetheless, even in the case of two equal-sized groups, one group will always come to dominate over time from repeated interactions among members of the two groups.
Attribution: https://www.tandfonline.com/doi/full/10.1080/0022250X.2021.2012668
We found that in repeated encounters with threats of violence, a norm of discrimination always results. Members of one group dominate, and members of the other group are forced to submit. It should be mentioned that the binary tags in the simulations were wholly arbitrary. They did not mean that individuals were, for example, stronger, more aggressive, or better strategic actors. Even if we could associate binary tags with physical attributes, including size or capability, our models show that the tags labelling all individuals as a member of one group or another are sufficient to result in systemic discrimination. Discrimination is not intended, but results nonetheless. You can read the paper here. It is obvious that these patterns of discrimination do not merely exist in computer simulations. Rather, they are observable in real life and are historically familiar to us, across cultures with gendered and racialized customs.
This interactive simulation shows how in a population of actors with two groups and interactions including conflict, members of one group will always come to dominate over time.
Attribution: https://flowa.github.io/amadae-hawk-dove-binary-model/
While the findings only show that arbitrary binary labels are sufficient to lead to patterns of widespread domination and do not consider cultural and historical factors of discrimination, they are still highly important. We were able to build mathematical models showing how random tags build on systemic violence inevitably lead to discrimination. This finding provides insight into the extent to which the discrimination that many marginalised people continue to face today is reliant on the threat of violence and arbitrary binary tags. This research illustrates how identity politics that focus on removing these binary tags may help to challenge the ability of a majority to systematically dominate over a minority.