Invisible Armor: How Violence Affects Self-Censoring Behavior in Local Office in the United States
Rebekah Herrick (Oklahoma State University), Sue Thomas (Pacific Institute for Research and Evaluation), Christine Bird (Oklahoma State University) and Sofia Collignon (Queen Mary University of London)
Like most current U.S. officeholders, mayors experience significant amounts of violence, both psychological and physical, and violent episodes are growing. From the most extreme examples, including the June 2025 murder of Minnesota state representative Melissa Hortman and her spouse to daily occurrences of psychological attacks such as harassment, many officeholders face some types of violence regularly. As well, conflict in local politics is on the rise and is especially intense when substantively engaging with racialized and gendered issues. Women, in particular, tend to experience more violence and more violence that is sexualized and identity-based than men in office.
Research across nations demonstrates that violence against officeholders causes psychological harm, decreases political ambition, alters politicians’ campaign styles, and diverts their attention from their work, and especially does so for women. These effects are likely to lead to avoiding actions and situations that may result in violence. This, also, is especially true for women officeholders
As a result, there are multiple warnings from political scientists, the media, advocacy organizations, and officeholders themselves about the effects of ongoing violence on the quality of democracy here and abroad. If experienced people choose to avoid routine aspects of their work or decline to run for or stay in office, the nature of our representation may deteriorate. Further, this trend has sparked concerns that violence will suppress the representation of women who are already underrepresented.
This paper addresses those concerns by examining the consequences of violence against officeholders. It focuses on mayors in the United States and uses an anonymous survey administered in 2024 to mayors in cities with populations of at least 10,000 residents. Mayors were questioned about one aspect of declining democracy - self-censoring behavior. That is, behavior that alters representation of constituents by limiting communication of views on policy and provision of access to constituents because of fear of violence. It includes, but is not limited to, publicly expressing policy or process views as well as frequencies of appearing in public and in the media, and declining to participate in social media interactions.
In our research, we focus on three types of violence:
Psychological violence which involves acts likely to harm the psychological health of individuals by inducing fear or harm to their sense of self-worth or well-being. This includes threats to cause harm.
Physical violence which involves activities that directly harm one’s physical well-being or property.
Identity-based violence- This category differs from the others because it is based on the content of the violence. Identity-based violence includes attacks rooted in gender and race, such as gendered criticism or racialized slurs. It also includes sexualized violence such as sexual comments and physical advances. It may be psychological or physical.
In brief, we find that violence is prevalent in the professional life of mayors. Further, a significant proportion of mayors engage in self-censorship – and that women mayors are more likely to self-censor than men. This is due to the fact that women experience more particular types of violence than men, not because all things equal, women are more likely to self-censor. That is, since levels of violence are not equal women engage in more self-censorship.
More details under the top-line findings include that fact that the most frequently reported category of violence, psychological violence, has the strongest relationship to self-censorship, Thus, our results emphasize that when violence is personalized —especially when it is related to mayors’ identity characteristics — the harm may be more consequential because it is demoralizing and dehumanizing.
Also interesting is which specific behaviors are avoided by mayors. For example, almost half of mayors avoid social media. A large minority avoid discussing certain issues in public and avoid disclosing their schedules.
Our survey results also provide some possible ways to address violence against officeholders. In it, we asked mayors what actions their municipalities took to respond to violence. The most common response concerned increased protection by law enforcement. Next, was security-related changes such as the use of metal detectors in council chambers, security camaras, bulletproof daises, and other infrastructure changes. Mayors also highlighted adoption of codes of conduct as well as setting and enforcing clear rules for public behavior in council chambers. Special training for officeholders about how best to handle the myriad types of violence was discussed. Several mayors noted that they were encouraged to work with the media to craft messages of public civility with encouragement of civil behavior. Producing guidelines, strategies, and resources for action that officeholders can take was mentioned by a few mayors. Finally, decisions to increase transparency and accountability were discussed as leading to officeholder safety. We encourage cities and other political institutions to consider these and other actions.
Ultimately, our study indicates that when mayors experience violence—even if that violence varies widely across context and in substance — there are strong patterns of mayors responding to that violence by disengaging from their position, their communities, and most troubling, the democratic process. The consequences of violent experiences may be less engaged in their role, their community, and the democratic process as a result. This relationship may be particularly strong in the context of local politics because to experience violence in one’s own community may have a larger effect on self-censoring behaviors than for officeholders in higher offices and broader constituencies.
Further, if the goal of violence against women in politics is to quiet women’s voices, it is having the intended effect. This pattern is especially troubling when persistent challenges recruiting women to run for office is considered.
Although it is unlikely that the levels of violence will be substantially lowered any time soon, there are actions that politicians and cities can take to reduce the detrimental effects of the violence on individual politicians as well as legislative bodies generally.
References
Collignon, Sofia, and Wolfgang Rüdig. "Increasing the cost of female representation? The gendered effects of harassment, abuse and intimidation towards Parliamentary candidates in the UK." Journal of Elections, Public Opinion and Parties 31, no. 4 (2021): 429-449.
Håkansson Sandra. “The Gendered Representational Costs of Violence against Politicians.” Perspectives on Politics. 2024;22(1):81-96. doi:10.1017/S1537592723001913
Herrick, Rebekah, and Sue Thomas. "Research note: Rise in violence against US mayors: 2017 to 2021." Social Science Quarterly 104, no. 2 (2023): 81-91.
Herrick, Rebekah, and Sue Thomas. "The Effects of Violence against US Officeholders." Political Science Quarterly 139, no. 3 (2024): 407-423.
Krook, Mona Lena. “Violence against Women in Politics.” (2020). Oxford University Press.
National Democratic Institute. Not the Cost: A Renewed Call to Action to End Violence Against Women in Politics. (2021) NTC 2021 ENGLISH FINAL.pdf. Accessed October 30, 2025.
Zetterberg, Pär, and Elin Bjarnegård. "Psychological Violence." Gender and Violence Against Political Actors (2023): 70-81.
Rebekah Herrick is a professor emeritus of political science at Oklahoma State University. She earned her PhD from the University of Nebraska in 1991. Her primary research interests concern issues of representation particularly as they relate to marginalized groups and legislatures. Her work has appeared in The Journal of Politics, Social Science Quarterly, Legislative Studies Quarterly, American Political Research, State Politics & Policy Quarterly, and Politics & Gender.
Sue Thomas is Senior Research Scientist at Pacific Institute for Research and Evaluation. Her work is at the intersection of social science, law, and health policy research. Additionally, Thomas has published several books, and many journal articles, book chapters, encyclopedia entries, and book reviews on women, politics, and policy and American government. In 2020, she received the Jewell Enduring Contribution Book Award from the APSA section on State Politics and Policy.
Christine Bird, PhD, JD is Assistant Professor of Political Science at Oklahoma State University. Her research specialty is how law interacts with public policy. She focuses specifically on agenda setting in political institutions and the role of elite interests legal-political processes. Her work has been published in peer-reviewed journals including Political Research Quarterly, Journal of Law and Courts, Interest Groups & Advocacy, American Politics Research and Polity.
Sofia Collignon of Queen Mary University, London, is an expert in the study of candidates, elections and parties and gendered violence against political elites. Her most recent research lays in the intersection of elite politics and public opinion. Her research uses mainly quantitative methods (surveys, panel data, survival and multilevel models) coupled together with interviews. She joined Queen Mary University, London in 2022.