Not long after his 19th birthday, David Vallat, a native Frenchman born to a secular family, converted to Islam. He was having an “existential crisis” and his new faith helped him curb his juvenile bêtises, or “bad behavior,” he told me. Few questioned his choice to convert either then or later when he joined the French army in 1992, to, as he saw it, protect the Bosnian Muslims in Yugoslavia. At the time, French troops were being sent to the Balkans as NATO peacekeepers, and Vallat jumped at the opportunity to join them. His motivation came from a promise he’d made to himself several years earlier after watching Nuit et Brouillard, a film about Auschwitz: He would not “stand idly by” in the face of another genocide. Yet the war was a shock: After escaping death twice in three days, Vallat considered returning home—“as a coward,” he said.

Then Vallat met Saudi Arabian and Qatari fighters on the frontline. They had a “momentum” he admired, a courage he desired. They taught him that if he sought to become a true Muslim, it didn’t matter if he lived or died in the war—Allah was waiting for him in paradise. He befriended several men who were members of the Armed Islamic Group (GIA), which sought to export the Algerian civil war to French soil. They told Vallat that if he wanted to be a true martyr, he needed to “train himself.” (Two of the men, Khaled Kelkal and Boualem Bensaïd, were later convicted of masterminding the 1995 Paris bombings.)

When Vallat returned to France in 1994, he began visiting a Salafi mosque his GIA friends had recommended. Salafism, an ultraconservative, fundamentalist strain of Islam that aspires to emulate the Prophet Mohammed and return to the religion’s supposed original ways, has been known to breed jihadists. At the mosque, he was told that modern-day Islam was a domesticated product of colonization, and that true Islam was that of combatants, of sacrifice, of blood. Anyone opposing the jihadists must be annihilated, he was told. He read the Koran and began learning Arabic.

Several months later, Vallat left France for an al-Qaeda base camp in Taliban-controlled territory in Khalden, Afghanistan, where he trained alongside Chinese, English, Yemeni, Malian, Turkish, Tunisian, Algerian and Filipino fighters in the “wild west,” as he called it: no state, no laws, just tanks, guns, and explosives. “There, they think only about death—how to kill and who to kill,” Vallat said.

After 10 months of training, the 22-year-old Vallat returned to France to stay with an Algerian friend he met at a Salafi cell in Chasse-sur-Rhône, a small town outside his hometown of Lyon. Working as a junior al-Qaeda operative, he prepared to return to Bosnia as an arms dealer and die for Allah “like the Americans in Normandy.” But on August 29th, 1995, just a month after a bomb exploded inside a Paris metro killing eight and injuring 100, French authorities raided the cell and arrested Vallat. They were so surprised to find a French native that his mother was asked to confirm that he was not an undocumented Algerian. He was sentenced to five years in prison for engaging in terror activities.

At first, Vallat was enraged. He considered himself a victim of colonial France’s anti-Islamic system. But, isolated from jihadi propaganda and his network of friends, he was forced to interrogate his own ideology, he said. He began to read two books a week and enrolled in university courses, studying French literature, history, geography, philosophy, and the Greek classics. Reading Machiavelli and Rousseau gave him a new political idea—what he later referred to as the “gift of humanism,” where the potential for human goodness is regarded as a more important force than anything divine.  “I started to understand that all humans … can make a choice to believe in God,” Vallat, now 45, told me when we met at a sun-soaked café in central Lyon. “We can decide what we want, and the majority of our choices can be made to benefit us here on earth. Really, it shocked me.”

By his fourth year in prison, he was “completely deprogrammed,” he said. He claimed he never returned to jihad, and there’s no evidence that he did. GIA radicals issued a fatwa against him in 1995 when he began to speak out against them; last year, the Islamic State issued a second one via Facebook.

Though Vallat comes from an older generation of jihad, he insisted that radicalization remains much the same today. And the problem is worsening: some 350 “Islamic terrorists” currently sit in French prisons; another 5,800 are under police surveillance, and an additional 17,000 have been classified as a potential threat. Manuel Valls, France’s former prime minister, has declared the country’s fight against radicalization the “biggest challenge of our generation.” France’s new president, Emmanuel Macron, reiterated Valls’s statement in a press conference last August, in which he said the fight against Islamic terrorism is France’s “top priority.” Since 2015, over 240 lives have been lost to terrorism, with the targets ranging from the offices of the satirical weekly Charlie Hebdo, to a kosher market, to a Jewish school outside of Toulouse, to the Bataclan concert hall, to the sidewalks of Nice on Bastille Day.

Following a year bookended by terror, Valls, looking like he hadn’t slept in months, addressed the French Senate in an emergency meeting in May of last year. “The radicalization of our youth is a deadly social model,” he said. “It’s the most serious challenge we’ve faced since World War II because it deeply damages the Republican pact,” a reference to France’s core principles of liberty, equality and fraternity. He proposed an €80-million, 80-point counterterrorism plan, to address France’s leading threat: homegrown radicalization, or the recruitment of young men and women on French soil into ISIS.

“The government realized they could no longer wait for radicals returning from Iraq or Syria,” a government employee for the Inter-ministerial Committee for the Prevention of Delinquency and Radicalization (CIPDR) told me when we spoke in her office in Paris. “We had to create something that would work with radicals already in France.”

The plan was to open an experimental “Center for Prevention, Integration, and Citizenship.” Radicalized men and women who’d been flagged by local prefectures for exhibiting withdrawn behavior were invited to voluntarily enter a program to “develop critical minds and appropriate citizenship and republican values,” according to its charter. If it went well, the government would open 12 more centers—one in each of France’s 13 districts.

***

Last September, nine residents showed up at a cream-colored 18th-century manor nested in rows of copper sunflower fields in Pontourny, a sleepy town in France’s Loire Valley. Aside from the video camera and a sign on the fence outside that read “No Filming,” there was little evidence that this was the site of an experimental government program. A team of 25 social workers, psychologists, special educators, and a Muslim chaplain, greeted them for what was expected to be a 10-month regimen. (The center was expecting three times as many people to show up.)

Residents, who were aged between 18 and 30 and came from all over France, received lessons in French history, philosophy, literature, media, and religion, all with the goal of teaching them to “muscle their intellectual immune systems,” as Gerald Bronner, a French sociologist who worked at the center in Pontourny, put it. They also participated in daily therapy, art, and music classes. Group conversations centered on democracy, religion and laïcité, the French concept dating back to 1905 that calls for the separation of religion from politics.

“You can’t tell someone, ‘What you think is bad, here’s good information,’” one social worker at the center told me in August. Instead, the center wanted to address what made the residents prone to their ideology in the first place. “We worked with each person on their history, job opportunities, home life, health programs, all to help them understand why they believe what they do and question whether it’s really the truth.”

The center did not intend to teach courses in religion, but a Muslim chaplain was brought in to meet with each resident individually. At first, no one would speak with him—they regarded him as unfaithful because he didn’t keep halal and worked with the French government, which they regarded as secular. But he decided to stay at the center, the social worker told me. The chaplain met with residents individually and in groups, and offered two workshops: the first was in Arabic so they could “better master the language of the sacred texts,” and the second a lesson on the history of Muslim civilization. “Secularism in France does not mean the rejection of religions,” the CIPDR clarified, “but, on the contrary, guarantees freedom of belief and worship of one’s choice.”

After a few months, the residents were eating non-halal food. Residents also received a rigorous training in French nationalism: They were asked to wear uniforms and sing La Marseillaise, France’s national anthem, each morning.

But deradicalization is a murky, unsettled science. A debate soon broke out among experts over how best to implement the program. Could radicalized youth be “cured” psychologically? Or was radicalization a structural problem, caused by inequality and segregation? What, for that matter, did it even mean to be radicalized?

When I posed these questions to the CIPDR, they admitted they were still working out the process. To be radicalized, they explained, was the “process by which an individual or a group adopts a violent form of action directly linked to an extremist ideology with a political, social or religious content that disputes the political, social or cultural order.” Part of the difficulty, though, is in creating a program that avoids falsely categorizing Muslims who are conservative but not radicalized. While French intelligence monitors mosques, neighborhoods, and online activities, often there’s no way to tell if someone has fully committed to jihad until it’s too late. And to Vallat, the problem extends in both directions. “Nobody knows they are radicalized,” he said. When I asked him his opinion of the government’s voluntary approach, he laughed. “What, I’m going to raise my hand and say, ‘Hi, I think I’m radicalized and I need a doctor?’”

The French deradicalization model was also unlike any other. Germany, Britain, and Belgium have developed programs that focus on further integrating radicals into their community. Saudi Arabia, on the other hand, focuses on finding jobs and wives for recruited jihadists. But in France, the idea was to take subjects away from their home environments.

To Vallat, the problem with a deradicalization program, especially one implemented en masse, is the ambiguity of terms. “Radicalization” is subjective; it’s not like being ill or suffering from addiction. The idea that someone can possess the “wrong” radical ideology presumes there’s some “right” corpus of values. The CIPDR claimed to be addressing this problem by using the term “disengagement” instead of deradicalization. “Deradicalization means that we are going to withdraw the beliefs of a spirit,” Bronner wrote in an email. “This is not really the objective of the center; everyone has the right to believe what he wants. Rather, we want to help these radicalized young people make a declaration of mental independence to better control certain processes of deceptive reasoning such as conspiracy theories.”

Bronner explained that he and his colleagues at CIPDR convened an independent team of psychologists, including the sociologist Farhad Khosrokhavar, to create a “psychometric test” to measure the extent to which the residents’ worldviews were changing. They were tested on their ability to identify conspiracy theories, and forced to examine how they may have fallen for a conspiracy. Many of their tools were modeled after the French government-designed program Epide, Bronner explained, which works with disenfranchised youth to better integrate them into French society. But the methods were controversial.

Esther Benbassa, a French senator of Val-de-Marne, told me the French program was a “total fiasco.” The problem, she explained, was not the government’s method but the model from the outset. “It’s a stupid idea to take young people from their homes. The problem is you need to re-socialize these people, not make them a bourgeois model.”

“Several errors were made,” Amelie Boukhobza, a clinical psychologist for Entr’Autres, an association that manages the state’s deradicalization cases, told me. “The issue of volunteering was very problematic.” But to Boukhobza, the “full-frontal” approach of “flag raising in the morning, courses in secularism, etc.,” was too aggressively nationalistic. “They’ve built a program in total opposition to the particular mental universe of the individuals. I don’t think it’s the right solution. Rather, they should propose not a counter-truth but something that can coexist.”

The issue of terminology, however, remains a problem. To push the term “disengagement” over “deradicalization” as the CIPDR does is still to assume that some ideas are safer than others. And to fully reject terms seems, in some sense, to admit the impossibility of deradicalization.

***

A few months after the program began, local protesters gathered outside of the château, carrying signs that read “Pontourny is in danger” and “Jihadist Danger.” They demanded the government shut down the center, fearing it would bring terrorism straight to their town. “Imagine having to tell local people that radicalized individuals are going to be living next to them and to tell them it ‘doesn’t matter,’” Jean-Luc Dupont, the mayor of Chinon, Pontourny’s next-door neighbor with a population of close to 8,500, said.

By February, all the residents had left the center, largely because of the protesters, the social worker explained. The media wrote the program off as “a total fiasco.” But the center, according to CIPDR, was not a total failure. “We were really starting to see an evolution in their thinking, but we didn’t have enough time,” the CIPDR said.

For now, the Pontourny Center is shut down indefinitely. The CIPDR said they were working on developing new methods and would release a plan next year. But as the country grapples with whether to institute a new program, radicalization remains something of an existential threat for France. A better understanding of the problem would require the French to look inward at an array of complex socioeconomic and racial issues—something Valls, at least, appeared unwilling to do when he announced to the press in November 2015 that he was tired of “social, sociological, and cultural excuses” to explain jihad. “To explain is to excuse,” he said.

Even a more sensitive approach, however, will have to address complicated questions about where to draw the line between radical and non-radical. Is someone who rejects principles of laïcité inherently radical, even if they aren’t violent? To Macron, France’s new president, the ambiguity appears not to matter. “There’s no place for naiveté, nor for fear of Islam which confuses Islamism and Islamic,” he said in a press conference at the end of August. Addressing the United Nations in New York last week, Macron emphasized his interest in maintaining France’s “exemplary relationship” with the United States to “fight terrorism in Africa and the Middle East.” How and when this will happen, however, remains to be decided. This week, the National Assembly will meet to decide on the French Senate's new law, “To Strengthen Internal Security and Fight Against Terrorism,” and in November, the French government plans to convene to discuss a new plan for a “disengagement” program. Still, no one seems able to answer the question of what deradicalization actually means.

Even Vallat, still a practicing Muslim with a wife and daughter, doesn’t have an answer. “There always remains something of the pathway created from radicalization,” he wrote me later over email. “For example, I never go into a protected place without immediately imagining how to take it by storm. When I see a group of soldiers or policemen on the street, I cannot help but think how I’d neutralize them. I know today I will never do it, but this regard (or “outlook”) persists.”


Research for this article was made possible with the support of a Heinrich Boell Foundation Transatlantic Media Fellowship.