War Escalation Is a Persistent Risk, New Study Finds

Researchers led by Aaron Clauset, a computer-science professor at the University of Colorado Boulder, examined yearly battle deaths in civil and interstate wars mainly between 1946 and 2008. The statistical model shows that wars between nations do not follow a steady trajectory; each additional year of fighting carries roughly a 10 percent chance the conflict’s severity will double and about a 1 percent chance it will increase tenfold. The study concludes that escalation dynamics—feedback loops among battlefield decisions, alliances, and political commitments—can drive a limited clash into a far larger war.

Research finds some wars face a persistent risk of escalation War is hell. CU research finds some conflicts have a persistent risk of escalating

Historical examples illustrate the pattern. General Douglas MacArthur expected U.S. forces to return from the Korean War by Christmas, yet Chinese intervention that winter prolonged the conflict. More recently, President Donald Trump described the 2025-2026 Iran confrontation as a “short-term excursion,” but fighting has persisted for over five months with periodic spikes in violence. These cases show how leaders often underestimate the potential for wars to expand.

Research finds some wars face a persistent risk of escalation

• The analysis tracked annual combat deaths rather than total war tolls, allowing observation of escalation and de-escalation as dynamic processes.

Research finds some wars face a persistent risk of escalation

• Across the sample, no systematic trend toward growth or decline in violence was detected.

Research finds some wars face a persistent risk of escalation

• The probability estimates (?10 % chance of doubling, ?1 % chance of a tenfold increase per year) describe a recurring risk, not a deterministic rule.

Research finds some wars face a persistent risk of escalation

• Civil wars tended to see a decline in violence over time, whereas interstate wars displayed a persistent risk of continual escalation.

Research finds some wars face a persistent risk of escalation

Clauset emphasized that the findings do not provide a forecasting tool capable of predicting exactly when a war will surge or how many casualties will ultimately occur. He suggests that leaders treat every armed conflict as carrying the possibility of substantial escalation, regardless of its initial intensity, and consider the “ratcheting effect” where rising casualties make compromise harder and retaliation more likely.

Research finds some wars face a persistent risk of escalation

• “War is inherently very unpredictable once it gets started,” — Aaron Clauset.

War is hell. CU research finds some conflicts have a persistent risk of escalating

• “Each additional year of fighting carries roughly a 10% chance of doubling in severity and about a 1% chance of increasing by 10 times.” — Aaron Clauset.

War is hell. CU research finds some conflicts have a persistent risk of escalating

The study’s probabilistic view of escalation offers a cautionary framework for decision-makers who might label a conflict as “limited,” “controlled,” or “short-term.”

Research finds some wars face a persistent risk of escalation

Clauset notes that richer, more granular data—from individual battles, military orders, and political negotiations—could eventually refine the understanding of escalation mechanisms. Until such data are available, the historical record provides a stark warning: the ultimate size of a war may be shaped less by initial intentions than by the unpredictable chain of reactions that follows its onset.

Research finds some wars face a persistent risk of escalation