Check out this new article by Malicki Zorom and colleagues. They build a mathematical model that explains how terrorism spreads and persists in the Central Sahel, and test which counter-terrorism strategies actually work. The focus is on three groups in the population: people vulnerable to radicalisation, active terrorists, and internally displaced persons. Their goal is to work out the conditions under which terrorism dies out and when it becomes a long-term feature of the system.
To do this, they build a compartmental model similar to those used in epidemiology. The population is divided into susceptible individuals who could be radicalised, active terrorists, and people displaced by conflict. Individuals move between these groups according to specific rates. For example, susceptible people become terrorists at a rate determined by contact with existing terrorists, and some terrorists can be deradicalised and go back into the general population. Violence also forces people out of their homes and into displacement. The model is expressed through a system of nonlinear differential equations. The authors estimate parameters from data sources including ACLED, the Global Terrorism Index, and UNICEF. They examine stability, bifurcations, and sensitivity through techniques such as Sobol variance decomposition. They also run 20-year simulations to test how the system behaves under different policy scenarios.
The results are really interesting. A key finding is the importance of the basic reproduction number, R0, which measures how many new terrorists one terrorist produces in a fully susceptible population. If R0 is below one, terrorism fades out. If it is above one, terrorism becomes endemic. The model shows that in the Sahel, R0 is well above this threshold. They also find that military interventions alone reduce terrorism by less than 20 percent. Strategies focused only on prevention work better over time but are still slow. The most effective approach is a combined strategy of prevention, deradicalisation, and targeted military pressure, which the model shows can reduce terrorism by more than 80 percent. Sensitivity analysis confirms that no single intervention can shift the system. What matters is the interaction between security measures, social development, and rehabilitation.
For policy, the message is clear. Military action cannot shift terrorism dynamics on its own. Long-term stability depends on reducing the social and economic conditions that drive susceptibility and on building sustained deradicalisation pathways. The model gives a clear quantitative target. Bringing R0 below one should be the central focus of policy, and this can only be achieved with integrated, multi-sector strategies. For research, the paper opens the door to richer models that account for socioeconomic differences, ideological variation, and the financial networks supporting extremist groups.