The Skeptics Society & Skeptic magazine

Sixth-century Byzantium was a city divided by race hatred so intense that people viciously attacked each other, not only in the streets but also in churches. The inscription on an ancient tablet conveys the raw animus that spawned from color differences: “Bind them! … Destroy them! … Kill them!” The historian Procopius, who witnessed this race antagonism firsthand, called it a “disease of the soul,” and marveled at its irrational intensity:

They fight against their opponents knowing not for what end they imperil themselves … So there grows up in them against their fellow men a hostility which has no cause, and at no time does it cease or disappear, for it gives place, neither to the ties of marriage nor of relationship nor of friendship.¹

This hostility sparked multiple violent clashes and riots, culminating in the Nika Riot of 532 CE, the biggest race riot of all time: 30,000 people perished, and the greatest city of antiquity was reduced to smoldering ruins.

But the Nika Riot wasn’t the sort of race riot you might imagine. The race in question was the chariot race. The color division wasn’t between black and white but between blue and green—the colors of the two main chariot-racing teams. The teams’ supporters, who were referred to as the Blue and Green “factions,” proudly wore their team colors, not just in the hippodrome but also around town. To help distinguish themselves, many Blues also sported distinctive mullet hairstyles, like those of 1970’s rock stars. Both Blues and Greens were fiercely loyal to their factions and their colors. The chariots and drivers were a secondary concern; the historian Pliny asserted that if the drivers were to swap colors in the middle of a race, the factions would immediately switch their allegiances accordingly.

The race faction rivalry had existed for a long time before the Nika Riot, yet Procopius writes that it had only become bitter and violent in “comparatively recent times.” So, what caused this trivial division over horse-racing teams to turn so deadly? In short, it was the Byzantine version of “identity politics.”

Modern sociological research helps explain the phenomenon. Decades of studies have demonstrated the dangerous power of the human tribal instinct. Surprisingly, it doesn’t require “primordial” ethnic or tribal distinctions to engage that impulse. Minor differences are often sufficient to elicit acute ingroup-outgroup discrimination. The psychologist Henri Tajfel demonstrated this in a landmark series of studies to determine how minor those differences can be. In each successive study, Tajfel divided test subjects into groups according to increasingly trivial criteria, such as whether they preferred Klee or Kandinsky paintings or underestimated or overestimated the number of dots on a page. The results were as intriguing as they were disturbing: even the most trivial groupings induced discrimination.^{2, 3}

However, the most significant and unexpected discovery was that simply telling subjects that they belonged to a group induced discrimination, even when the grouping was completely random. Upon learning they officially belonged to a group, the subjects reflexively adopted an us-versus-them, zero-sum game attitude toward members of other groups. Many other researchers have conducted related experiments with similar results: a government or an authority (like a researcher) designating group distinctions is, by itself, sufficient to spur contentious group rivalry. When group rewards are at stake, that rivalry is magnified and readily turns malign.

The extent to which authority-defined groups and competition for group benefits can foment nasty factionalism was demonstrated in the famous 1954 Robbers Cave experiment, in which researchers brought boys with identical socioeconomic and ethnic backgrounds to a summer camp, dividing them randomly into two official groups. They initially kept the two groups separate and encouraged them to bond through various group activities. The boys, who had not known each other before, developed strong group cohesion and a sense of shared identity. The researchers then pitted the groups against each other in contests for group rewards to see if inter-group hostility would arise. The group antagonism escalated far beyond their expectations. The two groups eventually burned each other’s flags and clothing, trashed each other’s cabins, and collected rocks to hurl at each other. Camp staff had to intervene repeatedly to break up brutal fights. The mounting hostility and risk of violence induced the researchers to abort that phase of the study.⁴ Other researchers have replicated this experiment: one follow-up study resulted in knife fights, and a researcher was so traumatized he had to be hospitalized for a week.^{5, 6}

How does this apply to the Blues and Greens? As in the Tajfel experiments, the Byzantine race factions had formed a group division based on a trivial distinction—the preference for a color and a horse racing team. However, for many years, the rivalry remained relatively benign. This was likely because the emperors had long played down the factional distinction and maintained a tradition of race neutrality: if they favored a faction, they avoided openly showing it. But that tradition ended a few years before the Nika Riot when emperors began openly supporting either one faction or the other. But more importantly, they extended their support outside the hippodrome with official policies that benefited members of their preferred faction. The emperors Marcian, Anastasius, and Justinian adopted official employment preferences, allocating positions to members of their favored faction and blocking the other faction from coveted jobs. To cast it in modern terms, they began a program of “race-based” affirmative action and identity politics.^{7, 8}

In nearly all the countries where affirmative action programs have been implemented, they have an invidious effect on the group that benefits, imbuing them with a sense of insecurity and defensiveness over the benefits they receive.

Official recognition of the group distinction enhanced the us-versus-them sense of difference between the factions, and the affirmative action scheme turned this sense of difference into bitter antagonism, which eventually exploded in violence. Procopius, our primary contemporary source, placed the blame for the mounting antagonism and the riots squarely on Justinian’s program of identity politics. It had not only promoted an us-versus-them mindset in the factions, it also incited vicious enmity between them, turning a trivial color preference and sporting rivalry into a deadly “race war.”

Considering how identity politics could elicit violence from randomly assembled groups like the Blues and Greens, it is easy to imagine how disastrous identity politics can be when applied to groups that already have some long-standing, historic sense of difference. Indeed, there have been numerous instances of this in history, most ending tragically. For example, Tutsis and Hutus enjoyed centuries of relatively peaceful coexistence in Rwanda up until Belgian colonialists arrived; when the Belgians issued identity cards distinguishing the two groups and instituted affirmative action, it ossified a formerly porous group distinction and infused it with bitter rivalry, preparing the path to genocide. Likewise, when Yugoslavia instituted its “nationality key” system, with educational and employment quotas for the country’s constituent ethnic groups, it hardened group distinctions, pitting the groups against each other and setting the stage for genocide in the Balkans. And, when the Sri Lankan government opted for identity politics and affirmative action, it spawned violent conflict and genocide that destroyed a once peaceful and prosperous country. This last example—Sri Lanka—is so illustrative of the dangers of identity politics that we’ll examine it in more detail.

Sri Lanka: How Identity Politics Destroyed Paradise

She is a fabulous isle just south of India’s teeming shore, land of paradise … with a proud and democratic people … Her flag is the flag of freedom, her citizens are dedicated to the preservation of that freedom … Her school system is as progressive as it is democratic. —1954 TWA TOURIST VIDEO

Sri Lanka is an island off India’s southeast coast blessed with copious amounts of arable land and natural resources. It has an ethnically diverse population, with the two main groups being Sinhalese (75 percent) and Tamils (15 percent). Before Sri Lanka’s independence in 1948, there was a long history of harmony between these groups. That history goes back at least to the fourteenth century when the Arab traveler Ibn Battuta observed how the different groups “show respect” for each other and “harbor no suspicions.” On the eve of Sri Lanka’s independence, a British governor lauded the “large measure of fellowship and understanding” that prevailed, and a British soldiers’ guide noted that “there are no historic antagonisms to overcome.” With quiescent communal relations, abundant natural resources, and one of the highest literacy rates in the developing world, newly independent Sri Lanka was poised to flourish and prosper. Nobody doubted it would outperform countries like South Korea and Singapore, with the British governor dubbing it “the best bet in Asia.”

It turned out to be a very poor bet. A few years after Sri Lanka’s independence, violent communal conflict erupted, culminating in a protracted civil war and genocide. By the time it ended, over a million people had been displaced or killed. Sri Lanka’s per capita GDP, which was on par with South Korea’s in 1960, was only one-tenth of it by 2009. As in sixth-century Byzantium, identity politics precipitated the calamity.

Turning a Disparity into a Disaster

At the end of British colonial rule in Sri Lanka, there was significant educational and income disparity between Sinhalese and Tamils. This arose by happenstance rather than because of discriminatory policy. The island’s north, where Tamils predominate, is arid and poor in resources. Because of this, the Tamils devoted their productive energy toward developing human capital, focusing on education and cultivating professional skills. This focus was abetted by American missionaries, who set up schools in the north, providing top-notch English-language education, particularly in math and the physical sciences. As a result, Tamils accounted for an outsized proportion of the better-educated people on the island, particularly in higher-paying fields like engineering and medicine.

Because of the Tamils’ superior education, the British colonial administration hired them disproportionately compared to the Sinhalese. In 1948, for example, Tamils accounted for 40 percent of the clerical workers employed by the colonial government, greatly outstripping their 15 percent share of the overall population. This unequal outcome had nothing to do with overt discrimination against the Sinhalese; it merely reflected the different levels and types of education achieved by the different ethnic groups.

When Sri Lanka gained independence, it passed a constitution that prohibited discrimination based on ethnicity. But a few years after that, an opportunist politician, S.W.R.D. Bandaranaike, figured he could advance his career by cynically appealing to identity politics, stoking Sinhalese envy over the Tamils’ over-representation in higher education and government. He launched a divisive campaign to eliminate the disparity, which spurred the majority Sinhalese to elect him. After his election in 1956, Bandaranaike passed a law that changed the official language from English to Sinhala and consigned students to separate Tamil and Sinhalese education “streams” rather than having them all learn English. As one Sinhalese journalist wrote, this divided Sri Lanka, depriving it of its “link language”:

That began a great divide that has widened over the years. Children now go to segregated schools or study in separate streams in the same school. They don’t get to know other people of their own age group unless they meet them outside.

Beyond eliminating Sri Lanka’s common “link language,” this law also functioned as a de facto affirmative action program for Sinhalese. Tamils, who spoke Tamil at home and received their higher education in English, could not gain Sinhala proficiency quickly enough to meet the government’s requirement. So, many of them lost their jobs to Sinhalese. For example, the percentage of Tamils employed in government administrative services dropped dramatically: from 30 percent in 1956 to five percent in 1970; the percentage in the armed forces dropped from 40 percent to one percent.

As has happened in many other countries, Sri Lanka’s identity politics went hand-in-hand with expanded government. Sinhalese politicians made it clear: government would be the tool to redress perceived ethnic disparities. It would allocate more jobs and resources, and that allocation would be based on ethnicity. As one historian writes: “a growing perception of the state as bestowing public goods selectively began to emerge, challenging previous views and breeding mistrust between ethnic communities.” Tamils responded to this by launching a non-violent resistance campaign. With ethnic dividing lines now clearly drawn, mobs of Sinhalese staged anti-Tamil counter-demonstrations and then riots in which hundreds—mostly Tamils—were killed. The us-versus-them mentality was setting in.

Bandaranaike was eventually assassinated by radicals within his own movement. But his widow, Sirimavo, who was subsequently elected prime minister, resolved to maintain his top priorities—expansive government and identity politics. She nationalized numerous industries and launched development projects that were directed by ethnic and political considerations rather than actual need. She also removed the constitutional ban on ethnic discrimination so that she could aggressively expand affirmative action. The existing policies had already cost so many Tamils their jobs that they were now under-represented in government. However, they remained over-represented in higher education, particularly in the sciences, a disparity that Sirimavo and her political allies resolved to eliminate. In a scheme that American universities like Harvard would later emulate, the Sri Lankan universities began to reject high-scoring Tamil applicants in favor of manifestly less-qualified Sinhalese with vastly lower test scores.

Just like Justinian’s “race” preferences, the Sri Lankan affirmative action program exacerbated us-versus-them attitudes, deepening the group divide and spurring enmity between groups. As one Sri Lankan observed:

Identity was never a question for thousands of years. But now, here, for some reason, it is different … Friends that I grew up with, [messed around] with, got drunk with, now see an essential difference between us just for the fact of their ethnic identity. And there are no obvious differences at all, no matter what they say. I point to pictures in the newspapers and ask them to tell me who is Sinhalese and who is Tamil, and they simply can’t tell the difference. This identity is a fiction, I tell you, but a deadly one.⁹

The lessons of the various affirmative action programs in Sri Lanka were clear to everyone: individuals’ access to education and government employment would be determined by ethnic group membership rather than individual merit, and political power would determine how much each group got. If you wanted your share, you needed to mobilize as a group and acquire and maintain political power at any cost. The divisive effects of these lessons would be catastrophic.

The realization that they would forever be at the mercy of an ethnic spoils system, along with the violent attacks perpetrated against them, induced the Tamils to form resistance organizations—most notably, the Liberation Tigers of Tamil Eelam (LTTE). The LTTE attacked both Sri Lankan government forces and individual Sinhalese, initiating a deadly spiral of attacks and reprisals by both sides committing the sort of atrocities that are tragically common in ethnic conflicts: burning people alive, torture, mass killings, and so on. Over the following decades, the conflict continued to fester, periodically escalating into outright civil war. Ultimately, over a million people would be killed or displaced.

The timeline of the Sri Lankan conflict establishes how communal violence originated from identity politics rather than the underlying income and occupational disparity between the groups. That disparity reached its apex at the beginning of the twentieth century. Yet, there was no communal violence at that point or during the next half-century. It was only after the introduction of affirmative action programs that ethnic violence erupted. The deadliest attacks on Tamils occurred an entire decade after those programs had enabled Sinhalese to surpass Tamils in both income and education. As Thomas Sowell observed: “It was not the disparities which led to intergroup violence but the politicizing of those disparities and the promotion of group identity politics.”¹⁰

Consequences of Identity Politics in Sri Lanka and Beyond

Sri Lanka’s experience highlights some underappreciated consequences of identity politics. Most notably, one would expect that affirmative action programs would have warmed the feelings of the Sinhalese toward the Tamils. After all, they were receiving preferences for jobs and education at the Tamils’ expense. Yet, precisely the opposite happened: as the affirmative action programs were implemented, Sinhalese animus toward the Tamils progressively worsened. This pattern has been repeated in nearly all the countries where affirmative action has been implemented: affirmative action programs have an invidious effect on the group that benefits, imbuing them with a sense of insecurity and defensiveness over the benefits they receive. That group tends to justify the indefinite continuation of these benefits by claiming that the other group continues to enjoy “privilege”—or by demonizing them and claiming that they are “systemically” advantaged. Thus, the beneficiaries of affirmative action are often the ones to initiate hostilities. In Rwanda, for example, it was Hutu affirmative action beneficiaries who perpetrated the violence, not Tutsis. The situation in Sri Lanka was analogous, with Sinhalese instigating all of the initial riots and pogroms against the Tamils.

One knock-on effect of identity politics in Sri Lanka was that it ultimately benefited some of the wealthiest and most privileged people in the country. The government enacted several affirmative action schemes, each increasingly contrived to benefit well-heeled Sinhalese. The last of these implemented a regional quota system that was devised so that aristocratic Sinhalese living in the Kandy region would compete for spots against poor, undereducated Tamil farm workers. As one Tamil who lost his spot in engineering wrote: “They effectively claimed that the son of a Sinhalese minister in an elite Colombo school was disadvantaged vis-à-vis a Tamil tea plucker’s son.” This follows the pattern of many other affirmative action programs around the world: the greatest beneficiaries are typically the most politically connected (and privileged) individuals within the group receiving affirmative action. They are often wealthier and more privileged than many of the individuals against whom affirmative action is directed. This has been well documented in India, which has extensive data on the subgroups that benefit from its affirmative action programs.

One unexpected consequence of identity politics in Sri Lanka was rampant corruption. When Sri Lanka became independent, its government was widely deemed one of the least corrupt in the developing world. However, as affirmative action programs were implemented and expanded, corruption increased in lockstep. The adoption of affirmative action set a paradigm that pervaded the government: whoever held power could steer government resources to whomever they deemed “underserved.” A baleful side effect of ethnicity-based distortion of government policy is that it undermines and erodes more general standards of government integrity and transparency, legitimating a paradigm of corruption: if it is acceptable to direct policy for the benefit of an ethnic group, is it not also acceptable to do so for the benefit of a clan or an individual? It is a small step to go from one to the other, a step that many Sri Lankan leaders and bureaucrats took. Today, Sri Lanka’s government, which once rivaled European governments in transparency, remains highly corrupt. This pattern has been repeated in other countries. For example, after the Federation of Malaysia expelled Singapore, it adopted an extensive affirmative action program, whereas Singapore prohibited ethnic preferences. Malaysia subsequently experienced proliferating corruption, whereas Singapore is one of the least corrupt countries in the world today.

Perhaps the most profound consequence of identity politics in Sri Lanka was that it ultimately made everybody in the country worse off. After World War II, per capita income in Sri Lanka and Singapore was nearly identical. But after it abandoned its shared “link language” and adopted ethnically divisive policies, Sri Lanka was plagued by violent conflict and economic underperformance; today, one Singaporean earns more than seven Sri Lankans put together. All the group preferences devised to elevate Sinhalese brought down everyone in the country—Tamil, Sinhalese, and all the other groups alike. Lee Kuan Yew, Singapore’s “founding father,” attributed that failure to Sri Lanka’s divisive policies, saying that if Singapore had implemented similar policies, “we would have perished politically and economically.” There are echoes of this in other countries that have implemented identity politics. When I visited Rwanda, I asked Rwandans of various backgrounds whether they thought distinguishing people by race or ethnicity ever helped anyone in their country. There was complete unanimity on this point: after they got over pondering why anyone would ask such a naïve question, they made it very clear that distinguishing people by group made everyone, whether Hutu or Tutsi, distinctly worse off. In the Balkans, I got similar answers from Bosnians, Croatians, Serbians, and Kosovars.

The Perilous Path of Identity Politics

Decades of sociological research and millennia of history have demonstrated that the tribal instinct is both powerful and hardwired into human behavior. As political scientist Harold Isaacs writes:

If anything emerges plainly from our long look at the nature and functioning of basic group identity, it is the fact that the we-they syndrome is built in. It does not merely distinguish, it divides … the normal responses run from … indifference to depreciation, to contempt, to victimization, and, not at all seldom, to slaughter.¹¹

The history of Byzantium and Sri Lanka demonstrates that this tribal instinct is extremely easy to provoke. All it takes is official recognition of group distinctions and some group preferences to balkanize people into bitterly antagonistic groups, and the consequences are potentially dire. Even if a society that is balkanized in this way avoids violent conflict, it is still likely to be plagued by all the concomitants of social fractionalization: higher corruption, lower social trust, and abysmal economic performance.

This article appeared in Skeptic magazine 29.2
Buy print edition
Buy digital edition
Subscribe to print edition
Subscribe to digital edition
Download our app

It is therefore troubling to see the U.S. government and institutions adopt Sri Lankan-style policies that emphasize group distinctions. Echoing Sri Lanka’s separate language “streams,” many American universities now have ethnically segregated orientation and graduation ceremonies. Some offer “theme houses”—dormitories segregated by ethnicity. An Illinois public high school offers separate mathematics classes for Black and Latino students. As the U.S. continues down the perilous path of identity politics, it is unlikely to devolve into another Bosnia or Sri Lanka overnight. But the example of Sri Lanka is a dire warning: a country that was once renowned for its communal harmony quickly descended into violence and economic failure—all because it sought to redress group disparities with identity politics.

Surveys and statistics are now flashing warning signs in the United States. A Gallup poll found that while 70 percent of Black Americans believed that race relations in the United States were either good or very good in 2001, only 33 percent did in 2021.¹² Other statistics have shown that hate crimes have been on the rise over that time.¹³ In the last year, we have also seen the spectacle of angry anti-Israel protesters hammering on the doors of a college hall, terrorizing the Jewish students locked inside, and a Stanford professor telling Jewish students to stand in the corner of a classroom. While identity politics have increasingly directed public policy and institutions, ethnic relations have deteriorated rapidly. This—and a lot of history—suggest it’s time for a different approach.