It was the summer of 1789, and a strange fever was sweeping across France. It wasn’t a plague, or at least not a plague of the body. To some, it a plague of the mind. Parisians had stormed the Bastille prison, bringing an end to the monarchy and unleashing the French Revolution, one of the most impactful events in history. Meanwhile, in the countryside, a wave of violent peasant uprisings also erupted.
For centuries, researchers have wondered why and how this happened. This series of rural uprisings, called the Great Fear, fully shattered the feudal system that dominated France for a millennium. Yet there’s no apparent reason why these peasant riots spread like they did. They seem, at the same time, coordinated and extremely chaotic.
In a new study, a team of researchers has looked at this pivotal moment in world history through an entirely new lens: epidemiology, the science of how diseases spread.
By treating the wave of riots like a viral outbreak, they used quantitative analysis to map its spread and uncover the logic hidden within the chaos. Their conclusion is almost as striking as the Great Fear itself: this was no mindless panic. It was a rational, politically-motivated insurrection that spread like an infection.
Revolution Fever
The French Revolution may seem like a singular event, but it erupted from a perfect storm of social, economic, and political crises that had been simmering for decades. French society was rigidly divided into three “estates,” with the clergy (First Estate) and nobility (Second Estate) enjoying immense privileges and paying almost no taxes. Meanwhile, the massive Third Estate, which included everyone from peasants to the educated middle class, shouldered the crushing tax burden.
This massive inequality was magnified by a severe financial crisis, as the monarchy’s lavish spending and costly wars, including its support for the American Revolution, pushed France to the brink of bankruptcy. At the same time, the powerful ideas of the Enlightenment — promoting liberty, equality, and the rights of citizens — had spread through France. Naturally, this led people to question the king’s absolute authority. The final straw was a series of poor harvests that led to widespread hunger and soaring bread prices, turning simmering resentment into a nationwide revolt against an oppressive system.
The epicentre of the revolution was in Paris. But rural unrest had also been brewing. In addition to the worsening grain shortage, there were rumors of an aristocratic “famine plot” meant to starve out the population. There was no large-scale plot, just bad management, but the result was all the same.
Harvests had already been poor since the massive 1783 Laki eruption in Iceland. Storms and floods also destroyed much of the harvest during the summer, prompting some historians to suggest the rural uprisings weren’t even directly connected to the central revolution.
To solve the centuries-old riddle, the researchers, led by author S. Zapperi, borrowed a classic tool used to track pandemics. They turned to an epidemiological model known as SIR — Susceptible, Infected, Recovered; they also added a Re-Infected. It’s a framework public health officials use to predict the spread of viruses like influenza or COVID-19. But Zapperi’s team repurposed it for a social contagion:
- A‘Susceptible’ town was one that hadn’t yet had a protest but was vulnerable to the “infection” of rebellion.
- An ‘Infected’ town was one where a riot was actively happening.
- A ‘Recovered’ town was a place where the protests had already occurred and ceased.
- And because some towns saw protests flare up more than once, the model included a ‘Re-Infected’ category.
The question was, what made a town “susceptible” and how did the “infection” of protest travel from one place to the next?
The Path of Protest
decomposition of the transmission network in which disconnected clusters are coloured differently (b). Image from the study.
As with any disease, the researchers were looking for patient zero. They painstakingly compiled data on riots from the Great Fear, mapping the precise location and timing of the uprisings. This gave them their patient zero — or rather, patients zero, the several locations where revolts first erupted. The map in their study shows these separate outbreaks, like distinct clusters of a virus, each with its own starting point and date. One wave began in Clisson on July 20th, another in Nogent-le-Bernard on the 21st, and yet another in Vesoul on the 22nd.
Next, they layered this protest data over maps of 18th-century France’s infrastructure—its road systems and, crucially, its postal networks. This allowed them to trace the physical pathways the rebellion could have taken. Finally, they gathered a trove of socio-economic data for every French town with over 2,000 people. They looked at everything: population size, literacy rates, average income, wheat prices, political participation, and local land laws.
A Rational Virus
By feeding all this information into their epidemiological model, they could rewind the clock and watch the Great Fear unfold. They could test different hypotheses by adjusting the variables. What if high wheat prices made a town more susceptible? The model would show how the protest might have spread. What if it was literacy? The model could simulate that, too. By comparing these simulations to the real historical record of what happened, they could determine, with stunning statistical power, which factors truly drove the spread of the revolution.
The results clearly showed that this was no random or irrational spread.
The team found that the towns most susceptible to the “infection” of protest were not the poorest, most isolated, or most ignorant places. Quite the opposite. The riots were far more likely to break out in towns with larger populations, higher average incomes, and greater literacy rates. The rebellion wasn’t fueled by the desperate and uniformed, but by people in communities that were more economically advanced, more connected, and better informed.
Furthermore, the idea that the protests were a panicked reaction to food shortages and rising bread costs (a long-held theory) also crumbled under the data. The model found no significant association between recent spikes in the price of wheat and a town’s likelihood of joining the revolt. The hunger for bread was not the main driver; it was a hunger for justice.
Justice. Vengeance. Fire and Blood
There was another striking connection that researchers found. The Great Fear was most virulent in regions where, if the legal documents proving a lord’s ownership of land were destroyed, that lord’s title was lost and the peasants could reclaim the land.
This suggests that in such places, attacking the lord’s castle and burning his papers was not an act of blind rage. It was a direct, strategic, and rational action with a clear potential benefit for the entire community. The peasants weren’t just lashing out. They were performing targeted legal and economic warfare. They knew which castles to attack and precisely what to burn.
By contrast, in areas where lords held land by absolute right or where peasants could only own land through explicit grants, the “infection” of protest was far less likely to take hold.
Analyzed through this lens, the Great Fear seems less like a fever and more like a conscious, fiery decision. The rural revolutionaries weren’t ignorant serfs; they were informed and cynical, with a thirst for vengeance. The system had wronged them and they wanted retribution.
Peasant revolts weren’t uncommon in France (or in much of Europe for that matter). There was much in common between the peasantry in the Great Fear of 1789 and previous rebellions. But the scale and virulence of the Great Fear was unprecedented. It was a stunning assault on a feudal system that had reached its limit. It undoubtedly had its irrational and feverish moments, but taken as a whole, it appears very calculated.
This is all the more striking because it feels very relevant today as well.
The study references similar epidemiological models being used to analyze modern protest movements, like the Arab Spring or the 2005 suburban riots in France. And surprisingly, the fundamental dynamics of how protest spreads from one location to another appear to have remained remarkably similar for centuries, even in our age of the internet and social media.
Then and now, protests spread not like a wildfire, but like a virus. Except this revolutionary ‘fever’ was not the sickness itself, but the calculated immune response of a society finally fighting the long-festering disease of injustice.
The study was published in Nature.
