In 1932, Lebanon conducted a census. The results indicated that within the country’s famously expansive constellation of religious groupings, or confessionals, a slim majority of the population were Maronite Christians.

That was not a surprise. After World War I and the dissolution of the Ottoman Empire, France put Lebanon under its mandate. In 1926, Elias Peter Hoayek, the Maronite patriarch at the time, wrote a letter to the French minister of foreign affairs and defined the relationship: “The original idea that served as a basis for the establishment of the Lebanese state was to make it into a refuge for all the Christians of the Orient and an abode of undivided fidelity to France.”

In 1943, Lebanon declared independence from France and, in the interest of multi-confessional harmony, announced an unwritten power-sharing agreement called the National Pact. The major sects divvied up spheres of influence. The president would be a Maronite; the speaker of the Parliament a Shiite Muslim; and the prime minister a Sunni Muslim. Offices in the government, Parliament, and civil services were mandated at six to five in favor of the Maronites.

Nearly nine decades later, Lebanon has yet to conduct another census. There is no indication that it ever will. That multi-confessional system by which the country still operates—which defines not just parliamentary seats but also cabinet positions and employment throughout the public sector—is built atop murky population counts. Real, hard demographic data could disrupt the whole thing. There is a compelling bizarre argument to be had. Do you accept the lie, and relative stability, of an entrenched status quo? Or do you push toward truth, no matter the consequences?

As the United States lumbers forward to the 2020 Census, Lebanon’s story is a potent reminder of just how fraught, and strange, population politics can get. In America, the Census is a massive piece of administrative housekeeping; we fight over its details, but we take its existence for granted. In Lebanon, a census is so powerful that it is impossible to implement.

The 1932 census was not a paragon of technical data collection. Academics believe that under French supervision, the census may have purposefully overcounted Maronite-heavy emigrant populations in order to bolster Maronite numbers. If so, this was blatant colonial manipulation. But somehow, Lebanon, with its 18 recognized religious groups, was harmonious; “the Switzerland of the East” was a thing people actually said. The country seemed to be, in its very specific way, democratic.

Eventually, though, the structure cracked. In 1975, Lebanon’s brutal civil war, fought largely along ethnic lines, broke out. In 1989, the Ta’if Agreement ended the war and readjusted the balance between the Maronite and Muslim populations: Powers were removed from the Maronite president and given to the Sunni prime minister, and the divide in parliamentary representation was shifted to 50-50 Maronite-Muslim.

That shift in power came only after that bitter, long war. Another shift came after a political crisis in 2008, with the Doha Agreement, which granted more powers to the Shiite constituencies represented by Hezbollah. And still, the balance doesn’t represent reality.

Rania Maktabi is a professor at Østfold University College. In her 1999 paper, “The Lebanese Census of 1932 Revisited: Who Are the Lebanese?,” Maktabi wrote, “A Muslim majority on Lebanese territories did not primarily evolve over time. This majority has been manifest and explicit since the creation of modern Lebanon.… In other words, there was no demographic rationale for Christian political dominance.”

On the phone from Norway, Maktabi laughs when relating to me how often her 1999 paper is still cited. Then, well-practiced, she ticks off the concrete ways in which the census directly affects the country today. If you’re applying for Lebanese citizenship, you have to prove familial lineage representation in the census. When voting, you have to do so in the district in which your family was originally registered in the census. (For some people, that means traveling hours to parts of the country with which they no longer have any connection. Or it means not voting at all.)

Overall, Maktabi says, the census “is part and parcel of this elite-driven democracy—we have the 18 communal groups, and they have their elites which discuss [governance]. It has also created a manifest understanding that what has been should be, would always be, should always be!” The argument for the current system, even after the 15-year Lebanon Civil War, is still: stability. (Proponents of that line of thinking point, above all, to the fact that the current civil war in Syria has not triggered a similarly full-blown sectarian conflict in Lebanon.)

But, Maktabi argues, the system “creates stability in a negative way. Exclusionary stability. I think the Lebanese all understand we have to share power. But the state idea that Lebanon will always be Christian [dominated]—it does not complement the changing into a more democratic stability.”

Rabih Haber is the founder and CEO of the Beirut company Statistics Lebanon, and he says his company has counted every person in Lebanon. “We have a total census of the Lebanese citizens,” he boasts. “Every village. The composition of each and every village. Each and every building. We have all the details on everything.”

“Can I… see it?” I ask.

“I cannot share any data. No, no, no.” But, he says, “the data is not surprising for the average citizen.” Haber here implictly acknowledges the widespread understanding: The Muslim populations, both Sunni and Shiite, are undercounted. So, yes, “Christians still dominate,” he adds. Then, quickly and with no prodding, he engages me in the political aspect of the conversation: He defends the status quo. “It is an agreement, a compromise, in democracy,” he says. “Otherwise we will be in an unbalanced democracy.” If the multi-confessional system were shifted to represent the true numbers, Haber argues, “Muslims will maintain a majority, and Lebanon won’t be Lebanon anymore. Lebanon needs the acceptance of the other. [Lebanon] is a reconciliation between Christians and Muslims. If they oppose war, [the country] should admit Christians in authority.”

It’s an oversimplification, to be sure, but in a way, what Haber is saying is an echo of an argument being bandied about by nationalists everywhere from Israel to Hungary to the United States. If we can’t engineer the demographic makeup of our country, then it is not our country.

As Haber talks, I respond with a low simmer of ambivalent murmuring. Suddenly, sharply, he asks, “You agree with me?” I don’t get the sense that he’s asking me the question because he’s interested in what I’ll write in this piece. I get the sense he’s checking me. Like I’m a tourist in town and we’ve struck up a coffee-shop conversation, and now he’s trying to gauge where my personal allegiances really lie. I double down on the vague murmuring and politely, quickly, get off the phone.

In 2013, the Lebanese Association for Democratic Elections (LADE) built LebanonElectionData.org, a mapping project representing Lebanese demographics and based on voter lists (which are acquired, for a fee, from the Interior Ministry). The site breaks down voting, district by district, in a variety of ways, including by gender and religion. According to LADE spokesperson Haneen Shabshoul, local media were particularly interested in the project’s breakdown of voter over- and underrepresentation per district (meaning the percentage of voters in a given district compared to the percentage of Parliament seats allocated to that district). It was a direct gauge of the democratic nature of Lebanon’s elections. Why, in some districts of Lebanon, are votes worth more than in others?

The most recent elections came in 2018 and followed a series of legal changes purported to improve the democratic process. Shabshoul is dismissive of the changes and of the process in 2018 in general. “The elections were called representative,” she says, “but they were not representative at all. They were not democratic, by any means. It was disastrous.” LADE now continues to advocate for what it calls its “basket of reforms,” which includes greater demographic transparency.

Marc Farra, a programmer and open data advocate, helped build LebanonElectionData.org. Farra is proud of the project; he refers to it as an “effective census.” But, he adds, “I don’t feel that it matters that much, because these numbers exist in the public consciousness—numbers different than what the government is saying. But does it actually drive policy forward?”

“What about a proper, government census, then?” I ask.

“There is no faith in public institutions,” Farra says. “But maybe a real demographic census that would be, I don’t know, conducted by a third party trusted by the general population would allow policy to be driven forward away from sectarianism and towards secularism. My shtick is, I think there should be more open data in general. I think the government should be involved in responsibly publishing all the data that they collect. Why is the census hidden? The policy of obscurity, it’s just not giving activists enough tools to demand for real policy change.”

In 2016, The Economist acquired clear demographic data from Lebanon after voter lists were “posted on a website belonging to the Interior Ministry”—presumably accidentally, as they were then removed. As The Economist wrote, “The data show that Maronite Catholics, once the largest sect in Lebanon, now make up only 21% of voters. That crown has passed to the Shias, now 29% of those listed, followed closely by the Sunnis, who make up 28%.”

“Everybody knows that the Muslim population is greater,” Maktabi had told me. “It’s common sense.” But simultaneously, Maktabi says, Lebanese common sense dictates that “‘maybe we should not voice our arguments too [loudly]. Lebanon is fragile.’”

“Do you believe that?” I ask.

“Maybe some will say that I am too optimistic, too dreamy or something,” Maktabi answers, “but it’s overblown, the fear. I would say: The census will not make a revolution.”