STUDY REVEALS QUEENS MUCH MORE LIKELY TO WAGE WAR 

A widespread cultural assumption portrays men as naturally aggressive and women as inherently peaceful, compassionate and reluctant to use violence.

History, however, is rarely that simple.

A major peer-reviewed study of European monarchies found that states governed by queens participated in interstate wars more frequently than states governed by kings. The result challenges the comforting but inaccurate idea that political violence is exclusively—or even inevitably—male.

The study does not establish that women are naturally more violent than men. Nor does it prove that modern countries led by women will be more likely to go to war. What it does demonstrate is that leadership, aggression and the exercise of power cannot be understood through simplistic gender stereotypes.

What did the researchers study?

The research was conducted by Oeindrila Dube and S. P. Harish and published in the Journal of Political Economy in 2020 under the title Queens.¹

The researchers constructed a historical dataset covering European monarchies between 1480 and 1913. Their primary sample included:

• 193 reigns;
• 192 distinct monarchs;
• 18 European political territories;
• 34 queen-led reigns involving 29 distinct queens; and
• 3,586 yearly observations.

Queens governed during approximately 18 per cent of the reigns examined.¹

The period ended in 1913 because monarchs generally exercised much greater personal authority over decisions about war before the First World War than most constitutional monarchs do today.

The central finding

After accounting for differences between territories and attempting to address the circumstances under which a queen was able to inherit the throne, the researchers estimated that states ruled by queens were 38.8 percentage points more likely to participate in a war in a given year than states ruled by kings.

This distinction is important. The finding was not merely that queens were "38.8 per cent more likely" to fight. The study reported a 38.8-percentage-point difference against an average annual war-participation rate of approximately 29.6 per cent during the period studied.¹

The researchers also produced simpler estimates showing an 11-to-13-percentage-point difference. Their larger estimate came from an instrumental-variable analysis designed to reduce the possibility that queens were more likely to inherit power during unusually peaceful or unstable periods.²

How did the researchers address selection bias?

Comparing kings and queens directly presents a difficult problem.

Under the succession rules used by many European monarchies, women did not have the same opportunity as men to inherit power. A woman might become queen only after male heirs had died, when there was no surviving son, or under unusual political circumstances.

Those circumstances could themselves influence the likelihood of war.

To address this problem, the researchers examined factors including:

• whether the previous monarch had a first-born son; and
• whether the previous monarch had a sister who could form part of the succession line.

These variables helped the researchers estimate the effect of queenly rule separately from some of the political circumstances surrounding succession.

They also performed sensitivity tests, removed individual queens from the analysis one at a time, considered co-rulers separately and tested whether succession disputes were driving the findings. The central relationship between queenly rule and war remained.¹

This does not make the study beyond criticism, but it makes it substantially stronger than a simple comparison of famous kings and queens.

Were queens starting wars or being attacked?

One possible explanation was that neighbouring rulers perceived queens as politically or militarily weak and were therefore more willing to attack them.

The results offered some support for this theory—but only in particular circumstances.

Unmarried queens were attacked more frequently

Among unmarried monarchs, queens were more likely than kings to participate in wars in which their territory had been attacked.

The researchers suggested that unmarried queens may have been regarded by neighbouring powers as vulnerable or easier to defeat.

This is important because some of the additional conflict experienced during queenly reigns may have resulted from the actions of hostile foreign rulers rather than the queen's own aggression.

Married queens were more likely to be aggressors

The pattern changed among married monarchs.

Married queens were more likely than married kings to participate in wars in which their own state was the attacking party.

The researchers found that the overall queen effect was driven disproportionately by participation on the attacking side, rather than simply by queens being invaded by others.¹

Their estimates therefore did not support the argument that the entire result could be explained by queens being innocent targets of male aggression.

Why might marriage have increased a queen's capacity for war?

The authors proposed an institutional explanation based on the division of responsibilities within royal marriages.

Queens frequently placed their husbands in positions involving:

• military command;
• taxation and finance;
• diplomatic relationships;
• administration; and
• the formation of foreign alliances.

Kings were generally less likely to give equivalent political or military authority to their wives.

A married queen could therefore retain supreme political authority while delegating military, financial or administrative duties to a husband or male consort. That arrangement may have increased the state's ability to raise revenue, form alliances and conduct military campaigns.

The researchers described this as an asymmetry in the division of labour. Marriage may have provided queens with additional governing capacity in a way that marriage did not provide to kings.¹

This explanation concerns the structure of royal government. It is not an argument that husbands secretly caused every war attributed to a queen. The authors tested whether male ministers or advisers appeared to be the real decision-makers and concluded that their evidence was more consistent with the decisions reflecting the queens' own policies.

Queens were also more likely to gain or preserve territory

Greater war participation did not necessarily mean worse political outcomes.

The study found that queen-led states were less likely to lose territory than king-led states. They were more likely either to gain territory or preserve the territory they already controlled.

The researchers also found no evidence that queens experienced greater internal instability during their reigns.

In other words, the findings do not portray queens as irrational rulers creating chaos for its own sake. In the competitive environment of early modern Europe, war was often used to expand territory, defend dynastic interests, obtain resources and increase state power.

Queen-led governments appear to have been both more willing and, in some respects, more capable of pursuing those objectives.¹

It does not prove that queens caused every war

The outcome measured was whether a state participated in war during a particular year. Some wars had already begun, involved coalitions or arose from long-running territorial and dynastic disputes.

"Participated in more wars" is therefore more accurate than saying queens personally caused or declared every conflict.

It cannot automatically be applied to modern democracies

The study examined hereditary European monarchies between 1480 and 1913.

Modern leaders operate within cabinets, parliaments, constitutions, defence alliances, international law and electoral systems. The study's authors expressly warned against directly applying their historical estimate to modern political leaders.¹ Other studies of women in national security leadership have also produced mixed findings, underscoring that modern political institutions differ substantially from hereditary monarchy.³

The number of queens was relatively small

Although the dataset covered more than four centuries, it contained only 29 distinct queens. The researchers used specialised statistical methods and extensive sensitivity testing to address this limitation, but it remains an important qualification.

The most accurate conclusion is not that women are more violent. It is that the claim that women in power are inherently peaceful is not supported by this historical evidence.

The "women are wonderful" stereotype

Psychological research has found that people often evaluate women more positively than men when judging women and men as broad social categories.

Alice Eagly, Antonio Mladinic and Stacey Otto found that participants generally associated women with more favourable traits and evaluations.⁴ This tendency later became widely known as the "women are wonderful" effect.⁵

Related research on "benevolent sexism" has shown that apparently positive beliefs about women—such as assuming women are naturally pure, gentle, morally superior or in need of protection—can still produce distorted and restrictive attitudes.⁶

The assumption that women are naturally peaceful may sound complimentary. However, it can prevent society from recognising women's full moral agency, including their capacity to exercise power, make aggressive decisions or perpetrate violence.

At the same time, the corresponding stereotype that men are naturally dangerous can cause men to be treated as presumptive offenders, even when they are victims.

Why this matters to domestic-violence advocacy

Interstate warfare and domestic violence are very different subjects. This historical study should not be presented as direct evidence about violence within families or intimate relationships.

Its broader lesson is nevertheless relevant.

Gender stereotypes can interfere with the accurate recognition of victims and perpetrators. In Australia, national data on family, domestic and sexual violence show that both women and men experience serious harm, though the patterns and prevalence differ.³ When society assumes that men are naturally violent and women are naturally harmless:

• male victims may not be believed;
• female aggression may be minimised;
• self-defence by men may be interpreted as offending;
• authorities may begin with gendered assumptions instead of evidence; and
• support services may be designed around an incomplete picture of domestic and family violence.

Rejecting the myth that all women are peaceful does not require portraying women as uniquely violent. It requires applying the same standards of responsibility, evidence and compassion to everyone.

Violence has no gender

The study of European queens does not reverse the stereotype by proving that women are worse than men.

It does something more valuable: it shows why the stereotype itself is unreliable.

Women can be peaceful or aggressive. Men can be peaceful or aggressive. Both can exercise power responsibly, abuse power, protect others or perpetrate violence.

Domestic-violence policy should therefore be based on conduct and evidence—not assumptions about which gender is capable of harm and which gender is entitled to be believed.

Violence has no gender. Neither should recognition, accountability or support.

References

1. Dube, O., & Harish, S. P. (2020). Queens. Journal of Political Economy, 128(7), 2579–2652. DOI: 10.1086/707011. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

2. Blandhol, C., Bonney, J., Mogstad, M., & Torgovitsky, A. (2022; revised 2025). When Is TSLS Actually LATE?. National Bureau of Economic Research Working Paper No. 29709. https://www.nber.org/papers/w29709. ↩

3. Australian Institute of Health and Welfare. Family, domestic and sexual violence summary. Australian Government. Accessed 16 July 2026. https://www.aihw.gov.au/family-domestic-and-sexual-violence. ↩ ↩²

4. Eagly, A. H., Mladinic, A., & Otto, S. (1991). Are women evaluated more favorably than men? An analysis of attitudes, beliefs, and emotions. Psychology of Women Quarterly, 15(2), 203–216. DOI: 10.1111/j.1471-6402.1991.tb00792.x. ↩

5. Eagly, A. H., & Mladinic, A. (1994). Are people prejudiced against women? Some answers from research on attitudes, gender stereotypes, and judgments of competence. European Review of Social Psychology, 5(1), 1–35. DOI: 10.1080/14792779543000002. ↩

6. Glick, P., & Fiske, S. T. (2001). An ambivalent alliance: Hostile and benevolent sexism as complementary justifications for gender inequality. American Psychologist, 56(2), 109–118. DOI: 10.1037/0003-066X.56.2.109. ↩