In Eurasia and North Africa, Islamic armies burst out of tribal Arabia and conquered preexisting, often ancient, imperial structures built on extracting agricultural surpluses from fertile lands. In Sub-Saharan Africa, however, the methods and ideology of Islamic governance could not overcome Sub-Saharan Africa’s obstacles to constructing enduring territorial states. Islam could not make African soils richer nor its terrain easier to control.

But Islam was nevertheless a force for state-building. The distinctively African-Islamic solution was to build states by selling salt, gold and slaves to the Middle East in exchange for warhorses (and other goods). Since these states were largely organised around man-hunting rather than steady control of territory, they did little to build law and order, public goods, or national cohesion.

The consequence was that people came to identify as Muslim while still maintaining older loyalties of lineage, ethnicity and region. Islam was added to Africa’s heterogeneity. Islam would build states, but brittle ones.

All this generated long-run disadvantages that continue to haunt post-colonial states. Inheriting this fragility, governments struggle to police vast semi-arid frontiers or suppress peripheral rebellions.

As Islamic preachers marshalled modern communications to excite religious fervor, older conflicts have been recast as righteous jihad. The wider wars of the Muslim world then added further fuel: scattering weapons and fighters, denouncing rulers as unIslamic. Excluded young men are recruited to wage violence in the name of righteous jihad.

The region now accounts for over 50% of global deaths from terrorism. At the time of writing, Mali’s capital is under attack by jihadis.

Islamic Trade Networks

The Sahelian-Sudanic belt had a distinctive ecological package that aided early state formation.

This zone largely escaped the tsetse fly, which transmits trypanosomiasis and causes sleeping sickness in cattle and horses. As Marcella Alsan shows, tsetse-suitable areas were less likely to sustain large domesticated animals, intensive cultivation, higher population density, or political centralisation.

The Sahelian-Sudanic belt also lay within Africa’s cereal zone. Millet and sorghum could be grown, hoarded and taxed - supporting political extraction. Yet cereal yields remained relatively low, so states leant more heavily on trans-Saharan and Red Sea commerce.

Rulers and war-bands increasingly exchanged human captives for Arab and Barbary warhorses (which survived longer without the TseTse fly). These imported mounts provided a major military advantage, enabling riders to encircle a village, set it ablaze, chase down runaways, tie captives to their horses, and carry off children crammed into small sacks. Beyond slave-raiding, horses enabled leaders to build up control across a far larger expanse, subjugating peripheral communities who were captured, assimilated or escaped to rugged terrain.

As rulers procured yet more warhorses and profited from Islamic trade networks across the Sahara and Red Sea, they built up larger empires. In 1324, the Malian king Mansa Musa led a gold-laden slave caravan to Mecca to perform the hajj. This exchange in gold, slaves, and warhorses also underpinned the 15th-century Songhay Empire. Early states thus drew strength from ecology and trade networks built on trading gold, salt, and also slaves.

Military Might without Public Goods

Slave-raiding enabled militarily superior groups to seize captives and scale up, but these were not cohesive, developmental states. Raiders plundered peripheral communities and gained military strength, while the continual scramble for slaves generated destructive competition, as states and militias warred over people, generating endemic insecurity.

Eurasian rulers likewise built war machines and glorified militarism, but their ultimate goal was to acquire more territory. Governments sought to amp up agricultural production, extract revenue and prevent internal rebellion. Ultimately, this generated greater bureaucratic state capacity.

Sub-Saharan African wars were different - seizing people. And as Dincecco, Fenske and Onorato demonstrate, historical warfare in Sub-Saharan Africa is associated with both higher fiscal capacity and higher civil conflict. Wars helped rulers extract revenue, but did not build broader peace.

Powerful states existed, but excelled in raiding, rather than building law and order, public goods, or national cohesion. Without a strong sense of the national community, people’s strongest loyalties remained local.

Islamisation and Fractured Ethnicities

Trading with Arabs across the Sahara, Sahelian-Sudanic rulers usually converted to Islam. Clerics offered literacy, spiritual potency, and a unifying identity of religious brotherhood.

But the region’s encounter with Islam was distinctive. In Eurasia, Muslim conquerors took over older agrarian states, with established institutions of administration and taxation. In the Sahelian-Sudanic semi-arid belt, by contrast, many communities were highly mobile (like the Tuareg traders and Fulani pastoralists), settlements were sparse, and rulers struggled to project power at scale. Thus even as people converted to Islam, many continued to see themselves as different ethnicities.

In the late 18th century, Fulani scholar ‘Uthmān dan Fodio preached reform, denouncing Hausa rulers as unjust and insufficiently Islamic. His movement not only attracted religious students but also ex-pastoralist freelance fighters, eager for loot and female captives. Victorious jihadis then established emirates and the Sokoto Caliphate became the largest 19th-century state in Sub-Saharan Africa.

Sokoto consolidated territory, built walled towns, established slave agricultural plantations, textile manufacturing, as well as Sharia courts. Learned scholars debated who precisely could be enslaved. But administration remained highly personalised. There was no standing army, no police force, no local coinage, no imperial buildings, and little durable public infrastructure. The state was not an impersonal bureaucracy, but rooted in local administrators and lineages.

Nor did Sokoto ever resolve the problem of cohesion. When dan Fodio died in 1817, allegiances fractured so badly that a second jihad had to be fought to re-establish the right to rule.Beyond the reach of cavalry, large zones of woodland and uplands sheltered dissidents, runaways, and raiders.

Older distinctions remained so salient that early Sokoto rulers used them to demarcate Islamic legal categories of enslavability. Commander of the Faithful Muhammad Bello explicitly identified peoples and regions such as Gwari, Bauchi, Borghu, and Nupe, along with Maguzawa communities, as legitimate targets for enslavement.

Subsequent colonialism, post-colonial independence and international intervention in the wider region have reinforced earlier militarism, but struggled to police vast territories and build up impartial public administration.

Poverty, Low Schooling & High Fertility

A growing quantitative literature suggests that African regions historically under Islamic rule now tend to have lower human capital, higher fertility, and weaker economic development.

Areas once governed by Islamic states now experience higher infant mortality, fewer years of education, and lower density of nightlights. Bauer, Platas, and Weinstein find this result, even after controlling for geography. They argue that regions under Islamic rule were treated differently under European colonialism, receiving less missionary education and educational investment as well as more indirect rule.

I would add that like many other world regions, Muslim African societies often resisted rival religions, seeing these as threatening their own legitimacy and prestige.

At independence, Muslim Africans were concentrated in poor, rural interior regions. They also tended to be less educated. That initial disadvantage has persisted. Even after controlling for parental education and location, Muslim Africans also have markedly lower educational mobility. Alesina and colleagues show that of Nigerian parents with no schooling, 80% of Christian children but less than 50% of Muslim children will complete primary school. Muslim Africans are also less likely to move to more prosperous places.

This gap in educational mobility is much larger in regions where Muslims are a majority, surrounded by other Muslims. So this is not just a story of a poor start, something is preventing uplift.

Muslim Africans also tend to have more children, even than Christians of the same ethnicity, find Phoebe Ishak and Mark Gradstein. These differences are strongest Muslims are a majority.

Muslim African fertility is higher because such women are less educated, younger at age of marriage, younger at age of first birth, less likely to work, and less able to achieve their preferences. Muslim husbands tend to want more children, and impose their will. In Chad, women under 39 typically have 2 years of schooling, and only 5% of married women use modern contraceptives.

Patriarchal beliefs are not only widespread, but routinely legitimised by Islamic preachers. When Mali’s government amended the law such that wives were no longer obliged to obey their husbands, Islamic organisations mobilised tens of thousands of people to march in opposition, forcing the president to backtrack. 60% of Malian respondents now say girls and women should not decide for themselves whether and when to marry.

Muslim African polities were thus initially disadvantaged by lower schooling and under-development, and have struggled to catch up. As preachers repeatedly reinforce men’s righteous dominance, while condemning contraception, Muslim African women bear more children than they actually want. This results in an expanding youth bulge, but without corresponding gains in job creation, productivity or state capacity. More young men struggle to secure status.

Muslim & Christian Nigerians have different moral priorities

Islamic preachers may be reinforcing development disadvantage by championing early marriage and large families, while reiterating that our primary objective is to reach paradise.

In World Values Survey data, Muslim Nigerians put greater weight on obeying religious norms (rather than doing good to others), reaching paradise, and are usually more intolerant.

Muslim Nigerians also express far more patriarchal beliefs, consistently valuing female housewives, boys’ education, and men’s right to jobs.

Polygyny

Polygyny carries prestige across Muslim West Africa, and is strongly supported. It also appears to have major developmental outcomes, with women in polygynous countries marrying 5.1 years earlier and having 2.2 more children.

Polygynous unions can also be fraught with rivalry, conflict and competition. Analysing Africa’s Demographic Health and Surveys, Opoku Ahinkorah shows that polygyny is a major predictor of intimate partner violence.

Polygyny also steepens marriage inequality, tilting the marriage market towards older men with livestock, land and status. As richer, established men procure additional wives, bride prices rise. Using georeferenced data on more than 800 African ethnic homelands, Koos and Neupert-Wentz find that ethnic groups sharing more border with polygynous neighbours suffer substantially more intergroup conflict. Young childless men from polygynous groups are also more likely to feel unfairly treated and more likely to view violence as justified.

In Northern Nigeria, families may prefer to match with richer, often already-married older men, squeezing out young men. Higher marriage inequality then predicts more Boko Haram violence, and more abductions.

Commanding Right & Forbidding Wrong

Preachers have leveraged radio, cassettes, and now smartphones to stir religious fervour, urging believers to command right and forbid wrong. As Muslim publics grew more orthodox and pressed for sharia, politicians cloaked themselves in Islam. In northern Nigeria, 12 states have expanded sharia, promising Allah’s blessings, while hisba agencies patrolled markets and punished deviance.

Islam can be intensely consoling. In my interviews, Muslim men describe collective prayer as profoundly peaceful and affirming, taking heart that our sufferings are merely temporary, and that emulating the Prophet brings righteous purpose, as well as eternal paradise. Moral policing can also be thrilling - denouncing others as lax, presenting oneself as pious. All this can be tremendously reassuring.

With this Islamic revival, old grievances are recast as righteous jihad. In the 1960s and ‘70s, many struggles across the region were expressed as ethnic, regional or separatist. But especially after 9/11 and the Iraq war, frustrated men have used the language of holy war, denouncing leaders as corrupt and unIslamic, while justifying violence with calls of ‘Allahu Akbar’.

Rural peripheries have proved easy targets. In borderlands, where the state is weak, ethnicities press grievances, desertification creates conflicts over scarce pasture, terrorists find safe passage for smuggling, training and illicit economies.

Interviewed combatants from across the Sahelian-Sudanic belt have shared that they joined because they were economically frustrated and were persuaded by Islamic militias promising material gains, community belonging and rewards in the afterlife.

“I joined because I felt frustrated with the conditions of everyday life, the life that I lead. I was so poor and vulnerable, with a family to take care of. I thought that in the group I would have a better status as a scholar, I would be better off and in the worst case scenario I would die as a martyr.” Moustapha, 39 years old, Niger

“We were told we will get money and wives when we join the group. We were told we will go to heaven when we die in battle.” Goroma, 23 years old, Nigeria.

All this created a tinder box - ignited by the contingent, proximate trigger of the Libyan civil war. After Libya collapsed Tuareg fighters who had previously been employed by Muammar al-Gadaffi then returned to Mali with their weapons, seeking autonomy in the northern region and joining militant jihadis to launch an insurgency. Conflict further inflamed ethnic divisions and triggered a military coup. Mali’s military junta then pursued support from Russia. Terror attacks continued to escalate, and in April 2026 Islamists stormed the capital.

Why Did Sub-Saharan Africa’s Earliest States Became So Fragile?

The Sahelian-Sudanic belt is the site of Sub-Saharan Africa’s earliest political centralisation, now marred by low human capital and jihadist insurgency.

My suggestion is not that Islam made states weak. Rather, Islam entered a region where population sparsity, mobility, and low agricultural surpluses made territorial administration difficult. Given these constraints, Sahelian-Sudanic rulers built power joining a wider commercial network, trading slaves, salt, gold and warhorses across the Sahara. And although people converted to Islam, this did not erase older loyalties of lineage, ethnicity, and region. They remained deeply divided.

In the 20th and 21st centuries, governments have struggled to raise human capital, build state legitimacy, overcome ethnic polarisation, control vast semi-arid frontiers, or suppress peripheral rebellions.

Islamic revival may be further entrenching this disadvantage by encouraging believers to prioritise paradise, early marriage, and large families. With the transnational Islamic revival, jihadists enlist new recruits with offers of quick wins and eternal rewards. Excluded young men wage violence in the name of righteous jihad.

Related Essays

• How does Religion shape Education & Fertility in Africa?

• The Prestige-Piety Feedback Loop

• The Paradise Premium: How the Afterlife affects Economics

• Demographic Overstretch

• What’s the Most Under-Rated Engine of Gender Equality?

• Male Violence and the Myth of Female Freedom

• Why was Ancient Nubia less controlling than Ancient Egypt?

• The Origins of States

Excellent New Books on African History!

I greatly enjoyed Sharman’s (2026) Ties That Bound: Slavery and Power in Africa and Reid’s (2025) The African Revolution: A History of the Long Nineteenth Century. Reid superbly focuses on Africans’ agency, their prerogatives and ambitions. Sharman prevents a very compelling account of why endemic slave-raiding proved so devastating and destructive.

Postscript: what about Ethiopia?

Ethiopia also saw early state-development, as its highlands supported denser agrarian settlement, grain-cultivation, and territorial statehood. However, the state was still highly fragmented: soldiers were not paid but survived through plunder, so undertook opportunistic raids. Neighbouring polities paid tribute in ivory and slaves.

Ethiopia now has far higher population density and crop yields than Sahelian states, while it has reduced fertility much faster. Yet income per head remains low. So Ethopia does not refute my argument about the specific trajectory of the agro-pastoral Sahelian-Sudanic Muslim belt.

A tremendous new paper by Xiaoxia Huang, a PhD candidate at Syracuse University, shows this operating within the Chinese civil service.

Xiaoxia Huang uses administrative data from the Chinese National Civil Service Examination, covering 238,086 job ads, application records for 100,931 postings, and 742,563 shortlisted candidates.

From 2005 to 2015, the share of job adverts explicitly preferring male candidates grew from 0.05% to 15%, with virtually no postings preferring women. From 2016 to 2024, adverts increasingly preferred male candidates, reaching a high of 30%. There were also some adverts for women, but these were usually paired “split-gender” postings, for otherwise identical jobs.

What led to this shift?

In 2009, women outpaced men in college degrees, and often applied to the public sector. But rather than welcome new talent, many hiring units appear to have reaffirmed male bias.

Huang finds that as women’s share of shortlisted candidates rose, hiring units subsequently issued more adverts for men. Importantly, this pattern does not appear in the pooled cross-sectional comparison. It only emerges once Huang compares each hiring unit to itself over time, using hiring-unit and year fixed effects.

This effect was especially pronounced in lower-level bureaucracies, where about a quarter of county and city level adverts expressly preferred men.

A rise in male-preferred job adverts is associated with 40-45% smaller candidate pools, and with 4-5% lower exam cut-offs.

Xiaoxia Huang thus offers striking evidence of how the Chinese civil service has consolidated male advantage by blocking female competition.

Building Networks in China - Baijiu, Karaoke, and Sexy Ladies

Another important mechanism, explored in my earlier essay, is that Chinese companies explicitly seek men for outward-facing sales jobs, where they are expected to feast with clients so as to forge camaraderie and reciprocity (‘guanxi’).

Private and public sector fraternities are then mutually reinforcing. By drinking baijiu, banqueting and singing karaoke, men affirm networks of loyalty and reciprocity.

Chinese Social Media Feminism

Can women fight back? In my own interviews, Chinese women have shared how users increasingly decry sexism and validate righteous resistance on Xiaohongshu (“Little Red Book”). In the post below, one poster says “A woman + 985 = male”.

985 means a prestigious university.

So an exceptionally clever and qualified female is equal to a mediocre male.

Another user invites women to share their experiences of employment discrimination. One replies that she went to a recruitment fair where they openly told women, “don’t bother applying” (i.e. we want men).

Meritocracy requires Countervailing Power & Persuasion

While social media enables women to highlight unfairness and legitimise feminist consciousness, these conversations tend to remain trapped in their own echo chambers, while men vent on Baidu Tieba. Resistance is further constrained by authoritarianism, and the CCP’s tendency to preserve social harmony by repressing dissent. Seldom seeing successful resistance, many women remain despondent. Absent independent organisations, women have little option but to navigate around bigotry.

Western feminists operated in far more permissive environments. The 1960s-70s counter-cultural movement and civil rights, condemned structural discrimination and condemned bias as categorically unfair. Harnessing this moral crusade, feminists persuaded their peers to regard sexism as similarly unjust. Liberal democracy proved equally crucial: turning to investigative newspapers and independent courts, women secured legal victories, won public support and created powerful deterrents. Because mass media had relatively high barriers to entry, these ideas became part of the wider cultural tide. By persuading the state to uphold merit and punish discrimination, female education thus became politically consequential.

But herein lies the Great Gender Divergence… In much of the world education does not break patriarchal gatekeeping, because men either discourage women from leaving the house or by closing ranks at the top.

(Xiaoxia Huang presented her paper at The Empirical Gender Network Conference, organised by Sole Prillaman, hosted at Stanford’s King Center on Global Development).

Related Essays

• Building Networks in China - Baijiu, Karaoke, and Sexy Ladies

• The Emergence of Chinese Feminist Consciousness

• Structural Transformation Enables Women to Rewrite the Script!

• China’s Economic Slowdown & Social Media

• The Emergence of Chinese Feminist Consciousness

• Economic Development & Cultural Change

• “Violent Intimacy” by Tiantian Zhen

• India’s Untapped Talent

But here lies an under-rated asymmetry. While an Uzbek factory can quickly install foreign machinery, it’s far harder to assemble rules-based governance. The UK’s Metropolitan Police only became impartial after decades of parliamentary wrangling, press scrutiny and civic activism, which slowly reshaped expectations. This was an uphill slog, lambasting corruption, forcing officials to treat everyone alike.

That asymmetry helps explain why cities can outrun the state, creating large ‘shadow economies’. This can include criminality. If formal job creation is weak and policing is ineffective, expanding cities can swell with people stuck in saturated informal economies, crowded in poorly serviced favelas, routinely exposed to violent predation. Criminal gangs profit from growth - running illicit markets, fighting for valuable turf, bribing politicians to look away. Under these conditions, the road to prosperity can widen the scope for violent predation.

Samuel Huntington made a related point in Political Order in Changing Societies. He was writing about political contestation rather than ordinary criminal violence, but emphasised the same problem of sequencing. When rapid economic and demographic change outrun governing institutions, this drives disorder. Building on his insight, I suggest that in weak states, rapid urbanisation often culminates in criminality.

Village Surveillance

“What is a village but a sink of localism, a den of ignorance, narrow mindedness, caste and communalism”, asked Dr. Ambedkar.He rightly emphasised that villages can be oppressive, but this intense surveillance also tends to enforce order..

In the village, everyone knows your name, your business, and your mother. Whatever happens, others will soon discover. A young woman from Haryana shared that within three minutes of leaving her house, her parents would receive this news by observant neighbours. Impoverished villages also tend to be highly dependent on each other for support and loans in a crisis. This mutual dependence may be reinforced by collective rituals, like weekly worship and celebratory festivals, which all reinforce community cohesion. Combined, this mutual dependence, social policing and fears of ostracism can encourage pro-social behaviour.

East Asia Escaped the Trap

Early 20th century China was quite lawless. In 1923, bandits hijacked a train in Shandong, abducting over 300 passengers. East Asia escaped this trap by combining job-creating economic growth and repressive, muscular states. Cheap educated workers found employment in factories and offices, while states iteratively enforced the rules. A woman from Shanghai shared with me that her grandparents often worried about kidnappers, but this is no longer the case thanks to strong security.

Migration in Weak States

When states are weak and wage jobs are scarce, urbanisation can magnify disorder.

Rio de Janeiro’s slum-dwelling population rose from 7% in 1950 to 22% in 2010. In 2000, 26% of Rio de Janeiro’s residents were migrants. They usually built self-help housing in unserviced favelas, where land tenure was insecure, neighbours were strangers, and everything was unpredictable. Only 7% of Brazilians say most people can be trusted.

During my stay in São Paulo, my hosts warned me not to take my phone as we walked two blocks to the supermarket. Thieves cruise on motorbikes, snatching phones in the hope of accessing online banking. Just last year, a guy was shot on that same street - for his phone. As I realised while being punched in the face: if other people’s welfare is devalued, their bodies are just obstacles, inconveniently attached to something of value.

Latin American cities struggle with high crime. Soares and Naritomi argue that this reflects especially high inequality, combined with fewer police, fewer judges and modest incarceration. When a large bulk of the population struggles to find formal employment, while others have substantial wealth, this obviously creates incentives for criminality. Meanwhile, weak and corrupt state institutions fail to provide effective deterrence.

Ed Glaeser adds another layer with his idea of “poor urbanisation”. Developing countries have urbanised at lower levels of wealth, depriving municipalities of tax revenue to provide basic services and public order. This is surely correct. But low income is only part of the problem. Rising incomes do not automatically diminish corruption, bribery or gang warfare. In weak states, growth can coexist with criminality. Elites can enrich their cronies, intimidate opponents, expand illicit markets, and buy impunity. The deeper problem is not just premature urbanisation, but insufficient job-creation in the formal economy and weak states.

Oil booms in Brazil actually amplified violence, amid weak policing. During the 2000s, GDP growth was nearly 100% higher in southeastern Brazil’s oil-producing municipalities. These regions also saw major increases in inequality and violence.

The struggles of informality, the lure of illicit gains, and the cloak of impunity, can all encourage young men to take dangerous risks.

The Machismo Trap

In areas where states are weak, toughness becomes defensively necessary, habitual and even glorified. Young men may project dominance to gain respect, deter attacks and avoid humiliation. Brazilian funk and Mexican reggaeton sometimes glorify retaliation, dominance and revenge. DJ Helinho’s “Kit Salve Da Quebrada” threatens to break arms, bind bodies, and shave heads. In communities where bravado is respected, studious boys may be mocked and maligned. Entire communities may come to prioritise brawn over books.

Biology comes into play. Boys’ and young men’s bodies surge with testosterone, amping up energy, aggression and competitive zeal. If channelled into sport and purposeful work, testosterone can unlock competitive advantage, but without such guardrails, young men may be more enticed by illicit gains. I call this the “Machismo Trap”.

Venancio (now 60) comes from Ceará (northeast Brazil) - where state presence is minimal and Catholic saints’ festivals incorporate indigenous rain rituals. Boys may be taught fearlessness by mounting an ox, while men typically carry knives. Unrestrained, Venancio’s father used to beat the entire family - one time whipping the cattle ropes against his son’s back.

A gay black man from a favela in Sao Paulo narrated his experience of violence - from his father’s beatings to school fights.

“I had be strong to show I was stronger. I wasn’t violent, but I had to be violent. If you do something that affects someone’s masculinity, they’ll start a fight and that will become a murder. Everything goes around masculinity: the man needs to show their strength, their power, they’re reckless, otherwise they’ll treat him as weaker. That’s how things works, you need to learn how to use violence… Everything is about honour”.

A young man from Rio’s City of God [favela] shared that he was previously part of a pick-pocketing child-gang. At age 8, a drug trafficker stubbed a cigarette out on his face. The following year, his neighbour was publicly tortured. Since gang leaders ruled the roost, some youths belted out rap music celebrating dominance. Chatting over lunch, another favela-dweller told me that girls often sought thuggish partners for protecters. Many of his childhood friends are now dead. When I confessed that I had never seen or heard of violence in my home town, they were shocked by this cultural difference. “Not even school fights?”.

Where states fail to protect, citizens turn to extralegal violence, like militias and vigilante lynching.

This can breed an escalating cycle - not just between criminal gangs but also with the police. Latin American security forces tend to respond aggressively. In Rio, new recruits are purposefully inducted with brutal training to “toughen them up”, “like pit bulls, crazy to bite people”. Meanwhile, weak oversight sustains impunity. Likewise in South Africa, the police force still use suffocation.

When States Make Violence Risky

When states make punishment more certain and public order more credible, violence often falls. Look at Sweden. After a surge in gang shootings, voters shifted right, the government increased spending on policing, lowered the age of criminal responsibility, and gave police wider search powers and expanded surveillance. As the risk of punishment rose, deadly gun violence declined.

San Francisco has seen a similar shift. As voters backed Proposition E, police gained broader powers over vehicle pursuits and surveillance technology (e.g. drones), with fewer reporting requirements. Crime fell.

The reverse is also true. In Britain, austerity weakened both prevention and deterrence. Between 2010-2019, budget cuts led to widespread closures of both youth centers and police stations, creating two distinct natural experiments. Youth centre closures triggered a surge in teenage crime, while shutting London’s police stations led to spikes in violent crime. Together, these studies nicely demonstrate how youth clubs help channel young energy productively, while effective policing provides a powerful deterrent. When the state retreats, the most disadvantaged neighbourhoods suffer.

When Cities Outrun the State

Economic growth can raise incomes, but only a capable state can enforce urban security. When cities outrun formal job creation and rules-based governance, they get nasty.

Migrants head to cities in search of opportunity, but if they struggle to find waged work and punishment is uncertain, they may be tempted by illicit gains. Criminal gangs prosper from illicit markets, gun down opponents, extract protection rents, and bribe politicians to look away.

Under these conditions, powerful syndicates permeate state institutions - bribing judges, politicians, and law enforcement officials, who then selectively target certain criminal groups while protecting others. In Latin America, Feldmann and Luna suggest that criminal networks enter society’s ‘bloodstream’.

The challenge, especially in societies undergoing rapid demographic change, is to build responsive policing and credible deterrence before criminal governance takes hold and young men come to revere aggression.

Otherwise, the gains from growth are lost to extortion, gunfire and intimidation. Young men are killed in gang wars, girls are assaulted, and entire communities come to prize toughness over trust and cooperation.

Related Essays

• Why are So Many Latin American Women Beaten & Murdered?

• Why are Homicides so High in Latin America?

• Fight Club, USA

• Did Austerity Scar Young Men?

• Lead Poisoning must be Tackled, Worldwide

• Towards a Scientific Study of Sexual Violence

Female labour force participation provides a reliable proxy for whether the society permits women to move in the public sphere, but this is not sufficient for gender equality. In low-income Africa, women may move independently, but her life is still marked by relentlessly drudgery and insecurity. Too often, poor women wake before dawn and labour long after dusk, only to be stuck in the same place.

Domestic Drudgery

Across many Sub-Saharan African countries, access to safe water and electricity have radically improved. Ghana has been a clear leader, alongside wider economic growth.

But absolute levels remain low, especially in rural hinterlands where households live far from pipes, grids, and tarmac roads. Only 22% of people in the DRC have access to electricity.

While Sub-Saharan African women work at high rates, many are also spending long stretches fetching water and firewood, and such labour is often shared with children. In Rwanda, 35% of surveyed households report that collecting water is a 30 minute round trip. And that’s just one journey. Imagine several buckets, every day, before cooking, washing, or bathing can even begin.

In Niger, Rwanda, Tanzania and Ethiopia, less than 10% of households have fridges. This is partly a function of household poverty but also state capacity, and a reliable electricity grid.

Dinkelman and Ngai have a nice paper, tracking African women’s time use. They find that South African and Ghanaian housewives often spend as much time as American housewives did in the 1920s, but under much harsher conditions, with fewer labour-saving appliances, and much more time spent on cooking, cleaning and laundry. In Uganda, Tanzania and Ghana, women spend 32-48 hours a week in home production, then only 25 hours a week in marketised work.

In richer economies, many activities once performed within the household have long since been marketised. At the supermarket, we buy food that has already been grown, milled, processed, packaged, and transported by specialists (whether that’s breakfast cereal, bagels or sausages). Shopping online, we can treat ourselves to new clothes, then wash and dry with machines. And if those (divine) appliances break, we hire someone to fix them. As housework shrinks, mothers and fathers can spend much more playing and caring for their kids, enjoying shared leisure.

But when I lived in rural Africa, it was radically different. In Zambia, I can honestly say that I rarely saw parents ‘playing’ with their children. There was simply too much work - whether that’s hauling buckets back from the pump, scrubbing clothes, making fire, getting vegetables from the market (since we didn’t have a fridge).

What if you wanted to make dinner? Raw cassava can be poisonous because it contains cyanogenic glycosides that release cyanide. In rural Luapula, it was often dried in the open, pounded into flour, then cooked over fire as a thick porridge (nshima). In Northern Province, we would head to the bush, forage for a very specific tree with edible roots (ubusala), then boil it for three hours, yet the nutritional value was meagre. High inputs, low returns. In The Gambia, we would sit with a large calabash (dried gourd) and sort through the rice for grit.

Poor Sanitation

72% of households in Niger report that they have no toilet or latrine. This means they do not even have an open pit or bucket. Instead, they are going to the bush.

Open defecation is not only risky in terms of safety, but also contaminates the local area, sometimes resulting in outbreaks of cholera.

When kids get sick, mothers usually step in to provide much-needed care.

Half of Sub-Saharan Africans work in farming

In African countries where more women work, a large share of that work is still farming. Yet African agricultural productivity is extraordinarily low. In Zambia, the agricultural value added per worker is just $244, compared to the US’s $85,000.

But without any money, how can you leave an abusive relationship? You are trapped.

Working in agriculture does not necessarily improve women’s status

In Luapula, men are widely perceived and revered as providers. As Sarah (30-year-olds, rural homemaker and mother of four) explained:

It’s men who provide for us, who buy us everything. Husbands help parents [financially] but we women don’t do it much. We don’t really concentrate on school. … We get pregnant. You’re finished now, it’s over. … A boy will make her pregnant and he will return to school; whereas the girls’ parents will refuse to educate her again. They’ll say “You will get pregnant again,” thinking it’s a waste of money.

Her mother, Mary (55), likewise maintained:

Rural girls - paying for their education is a waste of money. They’re just interested in men.

In villages, where most women bear many children and work on farms, and virtually no one has social media or televisions showing alternatives, then marriage and motherhood are often taken for granted as aspirational goals. Whereas rural girls, without social media or exposure to alternatives, didn’t necessarily see the value in education or envy such advances. As Nsenga remarked,

“In the village they just give birth and it’s all over. It’s because of early marriage. There’s nothing else they see and aspire to.

And while global elites may host glitzy conferences about ‘valuing women’s work’, I will report that this hasn’t punctuated rural Zambia. Cooking, cleaning, sweeping, fetching water and firewood, as well as weeding and planting (for subsistence agriculture) are commonly referred to as ukwikala (just sitting).

Wage labour with a contract is extremely rare in Sub-Saharan Africa

Most African women who do escape agriculture ultimately become self-employed, e.g. selling donuts or hawking dried fish in the market. Over 80% of SSA women workers are in what the International Labour Organization calls ‘vulnerable employment’ - either working on their own account or contributing to a family enterprise.

African cities have boomed in population size, but manufacturing jobs remain scarce. Instead, ever more young people crowd into saturated, low-productivity informal services - buying and reselling goods from Tanzania or South Africa - where profits continue to dwindle. All these challenges are compounded by demographic growth.

On meagre earnings, women are often extremely reluctant to leave abusive relationships. I saw this starkly in Kitwe, where I lived in a shanty compound with the elected local councillor. One day, a woman came to our house with a terribly swollen eye, pleading for help. USAID had funded a one-stop shelter for survivors, but she had no wage, no savings, and children to feed. Desperate, she returned home.

Another time, driving down a dirt track beside the fields, we passed a sixteen-year-old girl who was covered in bruises, gashes, and hay. She’d been badly beaten by her ill-tempered husband, but she was illiterate; where could she go?

When I visited a school, a 17 year old girl approached me saying that she was being raped by her uncle (also her foster-parent). Again: where could she go?

Scarcity and Harassment

Where waged work is scarce, employers gain the upper hand. Workers may be reluctant to leave or even voice grievances.

When Precious Zandonda surveyed 160 female staff at Barclays Bank and the Ministry of Community Development and Social Services in Ndola, most reported that their male colleagues made discriminatory remarks, slighted and ignored them, or habitually shade ‘suggestive stories or offensive sexual remarks, either publicly or privately’. 9% reported men ‘implied faster promotions, reward or special treatment” if they were sexually cooperative. One shared,

“I work in an office surrounded by male colleagues. Every other day I receive comments about my breasts, my thighs when am wearing a skirt and general talk about bedroom issues”.

For me, the most striking part about this dissertation is that the workplaces are named and published. Perhaps no one anticipated accountability.

Poverty is a major barrier to gender equality

In high-income countries, employment can be a source of great fulfilment, friendships and independence. Seated comfortably at her desk, surrounded by photos of her loved ones, she typically engages in cognitively engaging work, allowing for creativity, skills and pride. Colleagues and bosses generally play by institutional rules, knowing she has outside options, and that if things turn ugly she could complain to HR, consult a union, or pursue litigation. Returning home, yes, there are chores, but much of the pain has been marketised: step into the shower, fetch supper out of the fridge, chuck dishes in the machine. Easy mode.

In low-income countries, it’s the opposite. Productivity is extremely low, so earnings are threadbare. Much of the population is stuck on family farms, or scratches out tiny returns through petty trade, seldom sufficient to accumulate savings. She may also bear heavy (and physically painful) domestic burdens because the state is incapable of providing infrastructure and her family is too poor to afford market substitutes. Poor women thus wake before dawn and labour long after dusk, only to be stuck in the same place.

Only a privileged few have secured formal jobs, and may endure humiliation and harassment rather than risk dismissal. On the Zambian Copperbelt, you’ll hear a common refrain, especially among women: “ukushipikisha”. It means to endure.

Related Essays

• Does it Really Matter if Female Labor Force Participation is Miscounted?

• What indicators help track gender relations?

• Demographic Overstretch

• Sub-Saharan Africa’s Economic Stagnation

Note: all figures use the latest available Demographic and Health Surveys, most are from 2015 or later. A few countries have older data like Chad (2014) or Namibia (2013). All scatters exclude countries which only have older data.

In many SSA countries, “improved water” is not necessarily piped water, it can be boreholes and protected wells. People still have to walk and collect. Some jumps in this measure can reflect DHS methodology changes in water classification.

That, I think, helps explain a central challenge in much of Sub-Saharan Africa. Before European imperialism, much of the region was politically fragmented, comprising numerous competing polities, with weak authority over their hinterlands. Europeans then redrew borders and scaled up political units very rapidly, bringing together ethnically diverse populations that had little prior experience of shared rule or common institutions. Nairobi, for example, was only founded in 1899 as a railway depot, yet within decades it became the administrative centre of a vast colonial territory. Post-independence leaders thus inherited diverse populations without the slow historical accumulation of national institutions to govern at scale.

From 1960 to 2024, Sub-Saharan Africa’s population rose from roughly 230 million to about 1.3 billion. Western technologies and externally-financed health interventions saved millions of lives. This includes vaccination campaigns, malaria control, insecticide-treated bed-nets, indoor residual spraying, antiretrovirals and humanitarian relief. Without these interventions, especially against HIV, malaria, measles, tetanus and diarrhoea, mortality would have remained far higher.

But there was a catch. Mortality fell faster than incomes, taxes, schools and courts scaled up. Municipalities across Sub-Saharan Africa often lacked the resources to police expanding populations, process court cases, or prevent environmental hazards - like lead pollution. Schools are built, yet teachers and health care workers are often absent. This creates severe learning crises: 94% of Sub-Saharan African youth are not reaching basic skills in maths and science, while even among those enrolled in secondary school, 89% still fall below that threshold. With poor services and checkered accountability, 53% of Nigerians do not trust the police. One in five Africans remain under-nourished. What should have been a triumph of human survival has become a story of institutional overstretch.

Ed Glaeser makes a related argument about ‘poor urbanisation’: cities in low- and middle-income countries are expanding at relatively low levels of wealth, leaving municipalities unable to finance core services or uphold public order. He suggests that cheap food imports can enable urban growth without productivity improvements in domestic agriculture. Let me suggest a complementary mechanism: mortality fell rapidly through imported health technologies and donor-backed public-health interventions, while fertility remained high. Populations surged, while already weak states were increasingly overstretched.

Demographic Overstretch in Zambia

Back when I lived in rural Luapula (Zambia), people would say ‘balitulekelesha’ - they have abandoned us. The village has grown, but it’s still two hours walk from the nearest health clinic, two hours drive from police posts. So who should teenage girls turn to when their school teacher keeps groping them?

A year ago, I received a message on Instagram about a Zambian friend with 5 children. Abandoned by her husband, she had become destitute, living in a shack. She couldn’t simply start trading as she lacked the tiny capital to buy dried fish for re-sale. Obviously, I wanted to send money, but even that was difficult since she didn’t have her own phone and there was no local infrastructure for digital payments. Instead, another relative had to travel six hours by taxi then coach to Lusaka to collect the cash in person.

I saw the same pattern when living in a shanty compound in Kitwe, Zambia’s second largest-city. Walking home in the evenings, it was pitch black. No street lights. Weak environmental regulation of nearby copper mines also created risks of pollution. To wash, I pinned a sack of maize across a bush then threw a bucket over my head. One time my eyes stung from a toxic leak.

Schools revealed a similar pattern. For three months, I sat at the back of local classrooms. Teacher absenteeism was rife. We were meant to have six classes a day, teachers typically came to two. If they went to do other business, they faced zero penalties. One time the neighbouring class grew so noisy that our teacher walked over and beat them all with a hose pipe. Ironically, she was writing on the board about ‘human rights’. Curious, I asked one student what she thought of the subject? “Ukulemba fye”, she replied - meaning it’s just something to learn for the exam, not actually enforced.

These are among the more benign consequences of demographic overstretch. When I was travelling with small-scale traders across the Democratic Republic of Congo, the coach ahead toppled and we had to dig it out of the mud, but that was surely the country’s least egregious form of state weakness.

African civil servants and parliamentarians recognise these challenges. When I shared an office at Zambia’s Ministry of Health, the most common refrain at donor-funded workshops was “The problem is implementation”. But if the state does not deliver public goods, then legitimacy is severely strained.

Overstretched states struggle with nation-building

France, Thailand, the US, Russia and China have all used schools, story-telling, symbolism and public goods to cultivate national pride, teaching children to see themselves as part of a wider national community. That project has been far harder in African polities where political centralisation came very late, linguistic diversity is very high, and ethnic particularism remains pervasive. It is even more of a challenge if teachers do not even show up.

The state then struggles to foster a sense of unity, trust, and confidence in state impartiality. 40% of Batswana said they do not trust people from other ethnic groups. 30% of surveyed Basotho and Mauritanians said they would dislike neighbours from a different ethnicity. Low trust creates frictions for markets, encourages ethnic favouritism, sharpens grievances, and can fuel unrest.

Climate stress is intensifying agro-pastoral conflicts in northern Nigeria, but these tensions are magnified by demographic pressure, as more people try to make a living from semi-arid and fragile terrain, pushing up food prices. Yet despite rising dissatisfaction, Northern Nigeria’s security forces fail to contain jihadi insurgencies.

The consequences of such failure can be deadly - Sudan’s civil war has displaced 11 million people, eastern Congo faces rebel violence and foreign-backed militia, while the Sahel is becoming the epicentre of global terrorism.

Demographic overstretch slows progress towards gender equality

While Sub-Saharan African women often move freely in the public sphere and achieve high rates of labour force participation, many remain extremely vulnerable to male predation. Analysing Demographic and Health Surveys of the 20 most populous African countries, David Evans and colleagues find that “one in four adolescent girls report having experienced violence, and one in seven report having experienced sexual violence in the previous year”. In Montserrado (Liberia), 35% of women reported early sexual engagement [statutory rape], 40% of male assailants worked at schools.

Just this week, Médecins Sans Frontières published a report on sexual violence in Darfur. One 40 year old woman is quoted as saying,

“Every day, when people go to the market, there are four or five cases of rape. When we go to the farm, this happens. Men, they will cover their heads, and they will rape women. If there is more than one woman, they can try to escape. When a woman is alone, it is difficult for her to run away and escape. A few days ago, a woman tried to defend herself against rape, nearby, and she lost her tooth… There is no way to stop the rapes. The only way is to try to stay home, and to not go out as much”.

Ed Glaeser is right that poor urbanisation creates severe governance challenges. But I would go one step further: where population booms collide with low incomes and weak state capacity, the result is demographic overstretch.

In 15th Century Paris, Christine de Pizan crafted beautiful manuscripts, championing female heroes, lampooning misogyny. England’s Queen Elizabeth I kept Christine’s books in her royal library and hung City of Ladies tapestries on her palace walls. European aristocracy noblewomen likewise passed her works among themselves like precious heirlooms. But below the upper bourgeoisie, her brilliant works remained unknown, largely forgotten. Without printing, de Pizan could not bypass the state and church. Her rebellion failed.

The Printing Press famously enabled large-scale ideological persuasion - both by dissidents like Martin Luther and also establishments like the Catholic Church. Writing in pamphlets and periodicals, Europe’s Enlightenment thinkers later lobbied for liberty and secularism. Governments also got in on the act: mandating mass education and standardising curricula. Circulating libraries brought novels to the middle-classes, enabling female authors to celebrate loving companionship. In this new era of nation-building, villages were no longer isolated but swept into wider intellectual tides.

No longer isolated in remote peripheries, Muslims in Southeast Asia and Sub-Saharan Africa bought radio and cassette sermons to learn from clerics trained at prestigious centres of Islamic scholarship. Sufism and traditional spirituality, like offering rice porridge to ancestral spirits, were increasingly condemned as bid’ah (deviant innovation). Soviet satellites, meanwhile, followed Moscow and Westerners imitated Hollywood.

All these technologies required relatively high capital investments in production and circulation, and were therefore usually controlled by organisations - national governments, book publishers, radio producers, cassette makers. Whether communist, nationalist or Islamist, gatekeepers in patriarchal countries were invariably men. They tended to amplify ideologies that reaffirmed their righteous dominance.

Smart phones smash the barriers to both entry and dissemination. No longer do women need men’s permission or approval before seeking out, sharing and promoting ideas. With a device and headphones of her own, she can bypass gatekeepers, privately pursuing whatever she finds appealing or affirming, then share it with friends, encouraging wider emulation. Fathers, clerics, and broadcasters may still scold her behaviour, but they can no longer so easily police what she sees, admires, or reposts to her peers.

Smart phones enable radical decentralisation: anyone can become a content creator. But they do not entail a particular ideological direction. They simply erode older monopolies over what people can see, admire and imitate. In Malaysia, TikTok carries both BlackPink and petitions to censor subversive movies.

So while some currents tug towards piety and control, others glamourise female freedom. Algorithms may then compound those initial differences, sorting users into polarised worlds. Young men and women from Hohhot, São Paulo, Mexico City, Puebla, San José, Istanbul and Seoul have described swiping into worlds that are more glamorous, entertaining and amusing than their immediate vicinity - whether that’s irreverent comedy skits, influencers’ glamorous travels, or K-dramas in which doe-eyed guys show tender devotion. These are all alluring adverts for women’s freedoms and status. In Istanbul, a Kurdish woman invited me to her favourite church. “Do you have Christian friends?” I asked. “Of course!,” she grinned, fully immersed, “I watch Desperate Housewives.”

But can feminist revolutions be won with YouTube shorts?

Vitriolic messages may ferment righteous anger, but passive scrolling does not furnish young women with the charisma, competence or clientelist networks to move upwards in business or politics.

In private conversations, South Korean women repeatedly described male bosses promoting junior men, then bonding over drinks in bars staffed by sexy hostesses. But they nonetheless stay silent, for fear of ostracism. Reposting content to like-minded friends doesn’t necessarily persuade wider society that men’s dominance and impunity for abuse are unjust. The individual costs of dissent thus remain extremely high, and fears of such backlash inhibit women workers from speaking up. As Kim sighed, “I must do 24/7 nunchi” (i.e. be polite to maintain harmony).

Many young men still remain in their own self-adulating message boards - where entitlements, grievances, and sexist humiliation garner constant applause. So while some women may seek more gender equal partnerships, they often face supply-side constraints. Recognising that mismatch, many young women in big cities are increasingly choosing to remain single, unwilling to cook and clean without love: “Don’t settle for less”.

My reading of the available evidence is that smart phones have not ushered in a feminist revolution. They have, however, made ideological persuasion far more decentralised, with rival visions battling for power.

Related Essays

• Cultural Leapfrogging: Swiping Past Tradition

• Cultural Leapfrogging - Turkish Style

• The Great Splintering: How Digital Tech Shattered Cultural Control

• Did the Communications Revolution amplify Feminism?

• The Global Rise of Singles

• Why are Marriage & Fertility Plummeting in MENA?

• Ghosting the Patriarchy: Female Empowerment and the Crisis of Masculinity

• What Prevents & What Drives Gendered Ideological Polarisation?“Flowers of Fire” by Hawon Jung (review)

In Baroness Casey’s National Audit on Group-based Child Sexual Exploitation and Abuse (CSE) (2025), she documents chronic failures of policing, social services, local councils, and political leadership. Police and local governments repeatedly denied abuse; dismissed victims; suppressed whistleblowers; failed to record basic offender characteristics; cited fear of community tensions and accusations of racism as reasons for inaction; and resisted external scrutiny by obstructing public access to data.

Absent a full inquiry, the true scale of abuse remains unknown. Where suspect ethnicity has been recorded, such as in Greater Manchester, South Asians are over-represented relative to their share in the local population. But for the vast majority of cases, the police do not report ethnicity. The lack of systematic data means that political discourse becomes marred by anecdotes, speculation, and aggressive vitriol.

Below, I present a pathway forward, showing how standard methods can answer the most important questions. Any government committed to gender equality and women’s safety will doubtless provide full support.

Estimating ethnicity when records are incomplete

The lack of data is not insurmountable. Economists studying immigration in the United States routinely infer ethnic origins from names linked to large population datasets. Ran Abramitzky and Leah Boustan use historical census data and surnames to study how Jewish and other European immigrants assimilated over the twentieth century. Because names carry persistent signals of cultural origin, researchers can identify group-level patterns even when ethnicity itself was never directly recorded.

Muhammad Usama Polani (my collaborator) and I suggest applying a similar approach to British data on group-based child sexual exploitation.

The 2011 Census for England and Wales contains around 56 million records linking names with self-reported ethnicity. Kandt and Longley (2018) developed an ‘Ethnicity Estimator’ - the probability that particular names belong to different ethnic groups. Such models cannot determine an individual’s ethnicity, but if given 1000 names, they could say what percentage are probably Hispanic.

This approach has already been used to fill gaps for English hospital admissions. Kandt and colleagues successfully used name-based predictions, validating against records where ethnicity was recorded.

Geography further improves accuracy, as shown by Elliott and colleagues (2009). Name-based probabilities are more informative when combined with data on where someone lives and the ethnic composition of that neighbourhood. Since Bradford is home to many British Pakistanis while Tower Hamlets has more British Bangladeshis, an “Ahmed” from Bradford is statistically more likely to be Pakistani than an “Ahmed” from Tower Hamlets. Subsequent work has provided improvements.

Suffixes - like ‘-ewski’ and ‘oğlu’ - also provide clues about cultural origins. Sood and Laohaprapanon (2018) trained such a model on millions of names, achieving around 85% accuracy in predicting ethnicity.

Analysing UK police data

These methods could be applied to the UK’s national police datasets on child sexual exploitation - including the earlier CPAI database and the more recent COCAD system. By linking suspect names to large population datasets, we can estimate the ethnic distribution of suspects, e.g. 25% Hispanic.

Some police forces do record ethnicity. These cases are tremendously valuable because they allow the probabilistic estimates to be validated and refined.

Historical reconstruction

This approach could also be extended historically. Although COCAD was created in 2023, it draws on police data that goes much further back. If the earlier CPAI records also contain suspect names, the same ethnicity classifiers could be used to those records.

Unfortunately, earlier cases were not classified consistently. But analysts have already read and labelled COCAD cases, probably adding explanatory notes. AI could learn from these examples and classify older CPAI case narratives in a consistent style. Human analysts could then validate the model’s accuracy before applying it across the full dataset.

Institutional failures and attrition

The 12-week COCAD snapshots suggest substantial attrition. Within this timeframe, only around 5% of suspects are charged or summoned, while roughly 30% of cases are closed because victims do not support prosecution and 23% because of insufficient evidence. Casey’s National Audit documents similar institutional failures.

All this can be studied systematically: which cases moved forward, and which stalled. Did prosecution rates vary across regions or ethnic groups?

If the government were to provide access to older CPAI records that could be reclassified and linked to outcomes, researchers could examine whether patterns changed over time, such as in response to greater scrutiny.

Court records & victims’ testimonies

Court records provide another rich source of evidence. Roughly 400 cases of group-based child sexual exploitation have gone to court in Britain. Their transcripts contain detailed evidence on offenders’ coordination and institutional responses.

Large language models could help extract structured information from these transcripts, identifying recurring patterns. Since the number of cases is manageable, the resulting dataset can be cross-checked by humans.

Survivors’ testimonies also remain paramount, alongside nationally representative surveys measuring attitudes towards gender norms, victim-blaming and whistleblowing.

Evidence and Accountability

Building a transparent national dataset is an essential step - towards evidence-based discussions, greater institutional accountability, and ultimately stronger protection for vulnerable children.

I write this in support of all victims, everywhere.

Related Essays

• What do Epstein’s Emails, the Catholic Church & Grooming Gangs have in Common?

• Why Does Horrific Abuse Persist in the Most Gender Equal Countries?

• Can Women Walk Home Safely?

• Breaking Barriers, Building Nations: The Power of Neighbourhood Diversity!!

Discussions on boys falling behind in school usually focus on delayed entry, more male teachers, more technical education, and more support for atypical male jobs like health and education. Richard Reeves’s best-selling book Of Boys and Men has been catalytic, but says very little about cultures of violence.

Yet teachers repeatedly insist that the biggest problem is violence. 56% of surveyed US teachers said they experienced some form of physical violence from students during the past year, while 80% report verbal or threatening aggression. 60% said student behaviour as worsened.

In Germany, 47% of teachers report psychological or physical violence among students at their school, rising to 69% in socially disadvantaged schools. The French Education Ministry records increased assaults on teachers - last month, a 14 year old repeatedly stabbed a 60 year-old art teacher in front of 20 other students. During a bag search last summer, another teacher was stabbed to death.

Over in Britain, 24% of surveyed teachers say they’ve been pushed or shoved in the past year, while 62% report recent verbal abuse. 1 in 7 British secondary school teachers say they are attacked by pupils at least once a month. In 2023/24, there were a million suspensions at state schools. Teachers at two primary schools in Greater Manchester recently went on strike against ‘almost daily’ pupil attacks: biting, kicking, ripping skin off hands, throwing chairs through windows, and bringing knives. Allegedly, one was “locked in the room by students so I’ve not been able to get out”. Scotland’s largest teaching union reports that a quarter of its members in Aberdeen were physically assaulted last year, while a third see violence every single day.

Strikes and surveys are not necessarily representative. Distressed teachers may be more likely to express their grievances and mobilise for industrial action. But all this does raise alarm bells, which societies might take seriously if they value human capital. Classroom havoc is obviously disruptive, undermining everyone’s learning, also causing bullying, trauma and physical harm.

“We still have teachers experiencing on a daily basis, objects being thrown at them and being pushed, punched, slapped, kicked and bitten” - Scottish trade unionist.

“Increased anxiety, more pupils becoming more introverted, more pupils ‘playing along’ so as not to be targeted themselves.” - British teacher.

The Machismo Trap

Yet it can be difficult for adolescents to escape cultures of violence. Naturally, we all seek status and social approval, but in communities where this is achieved through brutal aggression and physical dominance, boys may project toughness in order to gain respect, deter attacks and avoid humiliation. Bravado may even be glorified - like Russian men uploading videos of their fights and YouTubers streaming house invasions for clout. Academic boys may then be mocked, maligned, belittled and bullied. Entire communities may come to prioritise brawn over books. I call this the Machismo Trap.

How often do men get into fights?

Over the past three decades, surveyors have dramatically improved our understanding of violence against women. Yet male victimisation remains strangely neglected, even though men suffer the vast majority of assaults and homicides worldwide. Official statistics are also patchy. In countries with weak state capacity, both lethal and non-lethal violence are routinely under-reported. When I was attacked and left covered in blood in Mexico, the police showed little interest.

So, how can we systematically study men’s violence? Surveys are useful, but must be triangulated with wider data. A thug who smacks randoms in the street seems unlikely to diligently complete our questions. If the survey is in person, what administrator is brave enough to enter areas under criminal governance, and request a full compendium of brutality? Even if the survey is representative, will men truthfully report their own violence or victimisation? Given social desirability bias, questions must be designed carefully.

Fight Club, USA?

This week, I collaborated with YouGov America to design several questions. With the caveat that violent people are unlikely to answer polls, the survey usefully depicts significant variation.

27% of young men say they got into fights at least a few times a year while growing up. 11% say they fought once or twice a week.

Older women usually say they never got into fights, but this gender gap is smaller among younger generations.

Imagine a young man was publicly disrespected but walked away without fighting. In your local area, would he be viewed positively or negatively?

With this question, we wanted to gauge whether retaliation would be respected. Personally, I would have preferred the wording to be that he ‘hit back’, so as to signal that such behaviour could be acceptable, and thereby reduce social desirability bias.

In response, most Americans say this would be ‘positive’. There is some variation, but overwhelmingly a lot of support for ‘turning the other cheek’.

Putting this all together, we see “The Machismo Trap”. Americans whose friends get into fights are more likely to say that walking away would be viewed negatively. This fuels a vicious cycle: if boys grow up believing that respect comes from intimidation, then classrooms become battlegrounds.

If every classroom of 20 pupils includes 2 with a proclivity for fighting, this may be highly disruptive. Discussions on boys falling behind in education might take this Trap more seriously, and engage with the rich empirical literature on how to promote emotional regulation, reduce the social returns to retaliation, and ensure that policing creates effective deterrents. Such moves would benefit everyone.

How does male violence vary worldwide?

Going forward, I am keen to study male violence across all regions, triangulating responses from surveys with official crime statistics.

So, what do you think of the following questions:

1. How often do you get into physical fights?

2. How often do people you know get into physical fights?

3. How often did you get into physical fights while growing up?

4. Imagine a young man who is insulted in public and fights back. In your community, how would others see him?

5. Do men in your community get more respect if they win fights?

6. How often do you watch fighting matches, eg boxing/ wrestling?

7. Men gain protection from being part of a group known for toughness. Dis/agree.

8. Can the police be trusted to punish criminals?

9. Would you report theft to the police?

Related Essays

• Beyond Brawn: Reimagining Masculinity

• Lead Poisoning must be Tackled, Worldwide

• Did Austerity Scar Young Men?

• Why Do Some States Protect Women While Others Enable Abuse?

• Why are So Many Latin American Women Beaten & Murdered?

• Can Women Walk Home Safely?

• Male Violence and the Myth of Female Freedom

Warlords organised coalitions of armed men to seize new territories, capture labour, extract tribute, and return with concubines. The route to power and prestige thus lay in commanding militias who unleashed bloodshed in the anticipation of honour and booty. Masculinity thus became synonymous with dominance, and women’s security depended on the strength of their male protectors.

The earliest surviving writings typically commemorate conquest. Egyptian hieroglyphs celebrate pharaohs’ victories, Sumerian cuneiform glorifies warrior-king Gilgamesh; the Book of Moses narrates the Hebrew conquest of Canaan. Chinese bronze inscriptions commemorate campaigns against rival states, Homeric epics enshrine martial glory as male virtue, Assyrian annals list their captives, Roman triumphs revel in the humiliation of enemy defeat, while Mayan murals record subjugation. Across civilisations, elites inscribed victory as proof of divine favour and masculine glory.

If cities expanded but struggled to monopolise the use of force, then criminal gangs tried to gain the upper hand - using blackmail, intimidation and protection rackets to extract advantage. Men rallied into coalitions, which projected collective toughness, asserted dominance, defended shared honour, or else were humiliated as vulnerable losers.

Several new studies track violence through the ages - using skulls, bones and court records. Eisner and Peng (2026) show that cranial trauma was pervasive in prehistory. Baten, Benati and Sołtysiak (2022) find that everyday interpersonal violence in the Middle East rose during early urban competition but declined once these polities developed more effective internal governance. In the Andes, Arkush (2022) similarly documents waves of escalation followed by pacification under Inca imperial consolidation. In medieval England, Eisner and colleagues (2025) use coroners’ inquests to compare homicides between cities and show that killings were highest in Oxford, where young unattached male students enjoyed clerical privilege and weak enforcement.

Tying the knot, these datasets suggest that violence flourished when organised armed groups competed for resources without restraint - neither constrained by overarching empire or credible rule of law. Brutality only declined when credibly constrained by effective institutions.

Cracked Skulls from Prehistory

Ancient burials often included skeletons with evidence of trauma - either unhealed lethal blows or healed damage, indicating recurrent warfare. Until recently, studies typically focused on individual sites, making it difficult to discern broader trends, but Eisner and Peng’s fantastic new paper aggregates the evidence across all time periods and regions. (If you’re in the Bay Area and heard a scream of delight, I can only apologise).

Across Europe, the Andes, the Levant, North China and pre-contact Mesoamerica, roughly 10–25% of adults show evidence of cranial trauma. In Neolithic Europe, 11% of skeletons bear violent head injuries, in the Southern Levant it’s over 25%. In pre-contact South America, rates cluster around 12–15%.

Caveat: all archaeological evidence must be viewed cautiously, since we only have access to that which survived. Any sample is of course a tiny fragment and partial.

Male Coalitions, Female Risk

Men were typically two to four times more likely to exhibit cranial injuries, suggesting they were the primary actors in organised combat.

Crucially, rates of female trauma rise and fall alongside male trauma. In societies with high overall violence, women also show elevated rates of injury. While this study doesn’t tell us whether women were being murdered in war or by their husbands, it’s certainly clear that violent ecologies have large spillovers, with more victimisation of women. This aligns with my earlier analysis of pre-industrial small-scale societies and contemporary patterns in Latin America.

State Formation and Competition in the Middle East

Baten, Benati and Sołtysiak analyse 3,539 individuals from across Turkey, Mesopotamia, the Levant and Iran, from 12,000 to 400 BCE, measuring interpersonal violence using cranial trauma and weapons-related injuries.

Trauma was relatively low during the Neolithic, when population density was sparse and groups could avoid each other. Violence peaked during the Chalcolithic period (c 4,500–3,300 BC), when early urbanisation and pro-state formation intensified resource competition.

Curiously, trauma seems to have declined during the Early and especially the Middle Bronze Age (c 3,300–1,500 BC). This was precisely when Mesopotamian polities developed consolidated territorial control, codified law, bureaucratic administration and expanding trade networks. The decline persists even after controlling for battle sites, military innovations and climate shocks. Baten and colleagues suggest that greater state capacity may have reduced everyday interpersonal violence.

However, violence rose again between the Late Bronze and Iron Ages (c 1,500–400 BC) - amid droughts, political fragmentation, mass migration, large-scale warfare, and imperial expansion.

Baten and colleagues suggest that early state formation coincided with greater violence, but once states achieved credible internal enforcement, then interpersonal violence declined within their territories. The implication is that stronger states can reduce violence, within their own walls. Though this is fragile.

When did Violence peak in the Andes?

Elizabeth Arkush synthesises Andean datasets, tracking cranial and facial trauma over time, distinguishing between healed injuries (repeated fights) and perimortem, unhealed wounds (likely lethal blows).

After 400 BC, trauma rose alongside the proliferation of fortified hilltop settlements. Small-scale societies fled to the hills and erected protections, suggesting constant fear of attack. Sites such as Chankillo, with massive stone walls, parapets and controlled access points, reflect escalating inter-group conflict. In some highland regions, between 1000-1450 CE, over 30% of skeletons show evidence of trauma.

From the mid-15th century, violence appears to have declined. The consolidation of Inca imperialism dampened local cycles of inter-community warfare.

When Men Fight, Women Suffer

28% of all men show head or face trauma, compared with only 19% of women. Men were also more likely to accrue multiple wounds, suggesting repeated attacks. Women’s cranial trauma was often at the back or side of the head - presumably as they were fleeing or crouched in fear.

Across large samples, there is a strong positive correlation between male and female trauma. This holds across time and between different societies.

Spectacles

Andean leaders also created spectacles to affirm their political dominance. As Arkush details, they publicly paraded prisoners of war, hoisting up trophy-heads, carving victories into temple walls, and lauding this as part of spiritual regeneration.

From the first millennium BC, monumental centres across the northern and central Andes were saturated with violent imagery. At Cerro Sechín, granite panels depict rows of warriors and their dismembered bodies: severed heads, detached limbs, exposed innards. Temples were designed to evoke predation, with visuals of fanged mouths, feline deities, and divine consumption.

The Nasca (of Peru’s south coast) carefully prepared trophy heads by removing brains, drilling holes through fontal bones, applying paint, then carrying them in public processions. Trophy heads sometimes sprouted plants, to convey the link between decapitation and regeneration.

Even when attacks lessened during the first centuries of the common era, the Nasca continued to elevate trophy heads as symbols of warriors, political authority and magical fertility. Dominance and brutality gave men prestige.

Medieval Murders, Oxford’s Impunity

Comparing cities across Medieval England, Eisner and colleagues reveal the importance of enforcement.

First up, they use coroners inquests to reconstruct recorded homicides in 14th century London, York and Oxford. Now, you’re probably imagining charming towns with participatory assemblies, guilds, courts, churches, and civic administration.

Um.

London and York’s estimated rates of homicide were 20-30 per 100,000 inhabitants - on par with contemporary Mexico. By contrast, Oxford’s was 90-100 per 100,000 inhabitants. Though for context, it was only a small town of 6,000 souls, so that worked out as 5 or 6 killings a year.

Over 90% of victims were male. Most killings were executed by groups. As was common across pre-modern Europe, attacks sometimes involves disfiguring someone’s face, so as to destroy their honour.

Why was Oxford so violent?

University towns were often dangerous. Large numbers of young males came alone and unaccompanied, then organised into student associations, divided by factional loyalties - ‘Northerners’ versus ‘Southerners’. 75% of all recorded victims and perpetrators were described as clericus/cleric - usually referring to individuals affiliated with the university, mostly students.

Legally classified as clerics, they were immune from prosecution under common law. Church courts were rather lenient, generally favouring penance, fines, or spiritual sanctions. In sum, a diverse motley of young men, high on testosterone, and getting off scott free.

Institutions really mattered (so my pal Acemoglu should be happy). Arrest rates were highest in London, lower in York, and lowest in Oxford. Credible governance and deterrence evidently proved pacifying, and that’s equally true today.

From Mobs to Magistrates

Smashed skulls remind us that organised violence was pervasive for most of human history. In such ecologies, men gained power by assembling mobs and militias, forging solidarity and glorifying domination. Bigger armies and sharper military innovations enabled ever greater extraction - of land, labour and tribute. Commanding force got you status.

When state enforcement was weak, private groups wrestled for advantage, battering rivals with stone axes, obsidian-edged macuahuitls, talwars and kris daggers.

As rulers monopolised force and courts gained authority, the returns to aggression began to fall. Only under these conditions could men (and much later women) pursue status in other domains - administration, commerce, innovation, and literature. (Or at least until AGI turfed us all out).

Anyone who tells you that there was some primordial feminist utopia where we all lived in harmony may be better suited to writing fiction.

Postscript

The Bayeux Tapestry depicts a pivotal moment in British history when it was conquered by Norman warlords. It includes 202 horses and 55 dogs; how many women?

To give another example of martial glory, let me share that my ex-army father decorated our home with spears, swords and paintings of battles. He then taught us how to fire guns, crossbows and slingshots. Though the only form of violence I pursued is savage trolling ;-)

When the UN compares countries’ progress towards gender equality, it usually focuses on women’s attainment in the public sphere: schooling, jobs and parliamentary seats. The World Values Survey similarly measures support for women’s paid work and leadership. This data is extremely useful, but omits a fundamental driver of patriarchy: violence.

For most of human history, power has been won by militarised coalitions who conquered new territories, raided for slaves, patronised patriarchal religions, while expanding their populations by glorifying maternity. Over the past century, progress has been real, but regionally uneven. In the most patriarchal societies, male status is safeguarded through female seclusion - confining them to the home, preventing any possibility of resistance. In another set, women move freely in markets and public spaces, but men exercise impunity for harassment, assault, trafficking, and gang violence. She is either secluded or vulnerable. Only in a third set have women escaped family control and also walk down the streets feeling safe.

In 2024, 98% of Singaporean women reported feeling safe walking alone at night. In South Africa it is just 24%. That gap captures something fundamental about women’s lives, and surely merits greater attention?

If the state fails to constrain criminal networks or predatory officials, then violent male coalitions can accumulate power. Opportunists gain advantage through brute force, sustaining a premium for male aggression. Yes, most victims of homicides are usually male, but that does not signify female empowerment. This loss of life reflects male competition within violent hierarchies. Criminal gangs run rampant, exploiting others for forced work and sex, publicising brutal punishments. Everyone lives in fear, most especially the poorest. Recognising danger, people may ally with known thugs or stay silent for fear of reprisals. Trapped by violence, they struggle to build institutions that enforce ‘the better angels of our nature’.

So while it’s important to track women’s progress in employment and politics, such metrics can be misleading. They omit pervasive harassment, homicides, conflicts and coercion. And here we see a marked class divide, with the poorest girls being most vulnerable to male aggression.

Let me illustrate several different forms of impunity: Guatemala’s religious conservatism, Haiti’s criminal gangs, Nigeria’s ritual intimidation, as contrast with Europe’s recent divergence. These examples are clearly much more severe than Letterman’s hair sucking, but they capture the same underlying principle: male impunity.

Impunity in Guatemala

Over a third of Guatemalan women experience sexual violence in their lifetime, yet the courts remain reluctant to impose punishment. From January 2018 to October 2023, prosecutors and judges dismissed 6,697 allegations of sexual violence against girls under 14. In that same period, there were just 102 rape convictions involving pregnant girls under 14.

Maria, age 12, was repeatedly abused by her grandfather and became pregnant. But instead of being supported by her family, they actually turned against her. “The whole family is against Maria because, as a good Christian, she should forgive,” shared a lawyer.

If you read Human Rights Watch’s report, you may feel puzzled. It rightly highlights prosecutory failures, fragmented data systems, lack of staffing and resources. But its recommendations are overwhelmingly technocratic - focusing on better laws, larger budgets and improved protocols. This omits religion, culture, and state weakness.

Guatemala is extremely socially conservative. Teenage fertility is high (68 per 1,000), while female labour force participation (40%) is well below the regional average. Only 74% of women make decisions concerning their own health care, while 80% of parliamentary seats are held by men. 81% of Guatemalans say religion is extremely important, 49% say religion is an important quality for children, while 60% emphasise children’s obedience.

As one female Guatemalan judge remarked,

“[In those regions] men still have the ingrained idea that they are in charge, not women.”

In societies that idealise obedience, hierarchy and the broader collective, then victims of assault are usually urged to reconcile, forgive and remain silent. Impunity isn’t just a function of administrative backlog, it reflects a social order that abets male abusers.

Criminal gangs in Haiti

After Haiti’s President was assassinated in 2021, armed groups have struggled for control, causing pervasive violence, intimidation and fear. The UN estimates that 90% of Port-au-Prince is now controlled by criminal groups, who use rape as a ‘weapon of terror’ to ‘punish’ girls and women from territories controlled by their rivals. A Haitian security expert explained, “Leaders, mid-level members, and rank-and-file members all rape girls and women just because they can, and nobody stops them. At MSF’s clinic, 58% of survivors experienced a group assault, with an average of three perpetrators.

“I was raped by four men while walking down a street” said a 25-year-old woman who had been looking for water for her children. “They were Gabriel’s men [from the G-Pèp criminal group]. They didn’t used to do this, but now they do whatever they want to all of us.”’

“It was near 10 a.m. More than 10 bandits arrived... Two grabbed me by my arms and dragged me to another house... [There], they took me to a room where there were six [other] girls. They tied me to a chair ... five men raped me that day. They hit me in the head with their fists several times... I spent five days in that house and every day I was raped by different men... The other girls were also raped and beaten.. I still have many nightmares about what those men did to me. I wake up in the middle of the night screaming in fear and sweating” (Bridget, 14).

Sorcery in Tracking Networks

Over the past 20 years, girls and women from Edo State, Nigeria, have increasingly been trafficked to Italy for sex work. To prevent their escape and instill compliance, criminal gangs routinely use ritual sorcery. Adeyinka and colleagues detail that girls may be taken to a juju doctor to swear a sacred oath, pledging never to reveal their trafficker’s identity, and repay the cost of their trip, or else the spirits will punish their families and subject them to madness. During this ritual, the juju doctor may take the women’s knickers, pubic hair or nails - as tools by which to control their bodies.

“I was scared and shaking… The man then told us not to cry, that anyone who cries in that place will die within 24 hours. She [the trafficker] then brought out the paper with my pubic hair and with that of the other girls, she had now written our names on each one. She gave it to the man [juju ritualist] and told him: “these are the girls,” and she wants him to make sure that we pay” (minor, quoted by Adeyinka et al).

As with all religions worldwide, fears of divine punishments can be used to enforce submission. Taken to Agadez in Niger and then Tripoli, women report being detained in warehouses, beaten, raped, sold between captors, and thereafter trapped in criminal networks in Italy. Victims have been reluctant to testify for fear of juju punishments.

Transnational trafficking, gang rule and religious hierarchy are clearly very different, but there is a common pattern: where the state does not constrain male violence, women are vulnerable to coercion.

Europe’s Divergence

Getting to Denmark isn’t just about state policing, but also public attitudes. Back in the 1970s, British police might turn a blind eye to men’s impropriety, dismissing it as ‘just a bit of fun’. Institutional disregard then enabled repeated attacks. An intrepid BBC reporter decided to turn the tables: pinching men’s bums, showing that this was actually weird and unwanted.

Europe has since made ginormous strides towards gender equality:

• 82% say it is unacceptable for men to ogle, catcall or whistle at women;

• 78% find it unacceptable for men to make suggestive comments about female colleagues appearances at work;

• 73% disagree that women often exaggerate claims of rape.

• 63% totally reject the proposition that if a woman suffers sexual violence while under the influence of alcohol or drugs then she is at least partly responsible.

• 79% say it’s unacceptable for a man to control his partner’s activities (e.g. mobile phone use or social media).

• 92% say a man occasionally slapping his wife or girlfriend is unacceptable.

• 72% totally disagree that domestic violence is a private matter.

Analysing both World Values Survey and Eurobarometer, I find that countries scoring highly on one dimension of sexism score highly on another. Compared to Western counterparts, Eastern Europeans tend to be more patriarchal. They are more likely to blame victims, dismiss the notion that ‘no means no’, and say that university is more important for boys.

In many former Soviet countries, feminism was suppressed and remains heavily stigmatised. Liberalism in Western Europe and North America was vitally important, enabling the spirit of ‘68 to spread like wild fire across college campuses, igniting the counter-cultural movement and feminist revolution.

Women’s movements can be catalytic - overthrowing taboos, galvanising public pressure, and demanding government reforms. Indeed, they are a major predictor of state action against gender-based violence. Whereas in Russia, where feminism is heavily stigmatised, the Dumas decriminalised wife battery.

But while Brits and Germans strongly condemn harassment and assault, they are now witnessing rising rapes, low convictions and institutional failures. Women’s safety in the public sphere is perennially vulnerable and requires constant activism.

Do Women Feel Safe Alone at Night?

For the past 300,000 years of human history, male coalitions have exercised dominance and often enjoyed impunity. When #MeToo went viral in 2017 it was nothing short of revolutionary: joining forces, women publicly supported each other, condemned abusers, and collectively signalled that victims would no longer be dismissed, disregarded, shamed or stigmatised.

Yet social scientists have been slow to catch on. Our analyses of progress towards gender equality still focus on economic activity and political representation, omitting a far more basic question: can women walk down the street alone at night and be at ease? Or, for yours truly, can she run into the woods and just enjoy nature?

Related Essays

• “Violent Intimacy” by Tiantian Zhen

• “Flowers of Fire” by Hawon Jung (review)

• Did Communism Smash the Patriarchy?

• “Putin vs. The People” by Greene and Robertson (review)

• Stopping Sexual Harassment

• Why Do Some States Protect Women While Others Enable Abuse?

• Male Violence and the Myth of Female Freedom

Last month I hiked through the Swiss mountains, paraglided into Davos, snuck through security, then gave a talk on the macroeconomic, social and political implications of falling fertility.

Feedback is always welcome. Summary below 👇

My Essays on Fertility

• How to boost birth rates

• Why is fertility collapsing, globally?

• How does Low Fertility Affect Economic Growth, Worldwide? Jesús Fernández-Villaverde

• Why don’t men want so many kids?

Relationship frictions

• The Global Rise of Singles

• Why are Marriage and Fertility Both Falling?

• Why are Childless Singles on the Rise in East Asia?

• What Prevents & What Drives Gendered Ideological Polarisation?

• Ghosting the Patriarchy: Female Empowerment and the Crisis of Masculinity

• Why are Marriage & Fertility Plummeting in MENA?

• Cultural Leapfrogging - Turkish style

Intensive parenting and the educational arms race

• What caused East Asia’s rapid fall in fertility? (over the 1960s and 70s)

• The Danger of Single Country Case Studies

• Why has fertility plummeted across East Asia?

• Fertility is a collective action problem. Can it be tackled with taxation?

• Why has Fertility Plummeted? Insights from Hong Kong

Films and prestige

• Lights, Camera, Ovulation!

Religion

• How does Religion shape Education & Fertility in Africa?

• Why is African Teen Fertility So High?

Rejecting assumptions of universal male dominance, academics began asking when, where and why women became oppressed. Explanations ranged from plough-cultivation and patrilineal clans to private property, state-development and colonialism. Sub-Saharan Africa and Southeast Asia’s small-scale matrilineal and bilateral societies were often presented as counter-examples - where women inherited land, lived near kin, ran markets, and exercised spiritual authority. ‘Warrior women’ of the Eurasian steppe are likewise esteemed for their military prowess.

But have these celebrations of female freedoms omitted a major constraint: extreme violence? By reading a wealth of historical ethnographies and analysing the Standard Cross-Cultural Sample (a comparative dataset of pre-industrial societies), I have discovered that many cultures renowned for female prominence also experienced pervasive homicide, assault and warfare. Organised male coalitions amassed power through slave-raiding and labour coercion, sustaining ecologies of fear.

Inter-disciplinary Consensus

Across economics, anthropology and evolutionary theory, I notice a common assumption: where women inherit land, choose partners and move freely, the society is seen as more gender equal.

In A Natural History of Patriarchy, David-Barrett suggests that our ancestors once lived ‘gender harmony’, and draws on contemporary hunter-gatherers as exemplars of equality. Female dependence on male providers, he speculates, only arose with the transition to agriculture, which raised fertility while exacerbating food insecurity.

Graeber and Wengrow in The Dawn of Everything similarly emphasise autonomy in pre-state societies. Drawing on Jesuit accounts of Wendat and Montagnais–Naskapi, they describe women as having “full control over their own bodies, and that therefore unmarried women had sexual liberty and married women could divorce at will”. Nuer women in the Horn of Africa are likewise portrayed as exercising freedom over their sexuality and reproduction.

Economists Sarah Carmichael and Auke Rijpma construct a composite index of “Family Constraints on Women’s Agency”, drawing on ethnographies recorded between 1860 and 1950 from nearly two hundred societies. They combine patriliny, patrilocality, polygamy, cousin marriage, marriage payments, inheritance, extended households and clan structures. Their framework implies that when women are free from patrilineal control, they exercise greater autonomy.

These accounts all assume that where women inherit property, choose partners and move freely, the society was relatively gender equal. Perhaps the past was not so bad?

Violence

Even if a woman is not subordinated to lineage elders, she may still be vulnerable to violent predation. Sexual permissiveness does not mean genuine choice if she can be abducted, enslaved and raped. When men can get what they want via coercion, they need not charm women with displays of devotion or companionship. Brute force wins.

Even if women can choose their husbands, in dangerous places they may strategically partner with aggressive men who can provide protection. Conflict societies also tend to glorify male aggression, raising boys to be tough, show readiness to retaliate, and intimidate potential foes. Male coalitions gain power through organised violence, then consolidate dominance through spectacular punishment.

Once we recognise how violence shapes culture, we may wish to revisit rosy narratives. Take pre-colonial Southeast Asia. “Men of prowess” gained power by militarism and labour coercion, while men could sell wives and children into debt-bondage. “Virtually every large trader was, simultaneously, a slave-raider. Every military campaign, every punitive expedition was a campaign for captives who could be bought, sold, or held”, writes James Scott.

In peripheral small-scale societies of Southeast Asia, Barbara Andaya details how young men went on head-hunting expeditions to mark their passage to adulthood, while young women celebrated men’s violence. In 1679, Governor Padtbrugge observed that men’s tattoos tracked the number of men they had killed, which their wife copied in pride. An 18th century Malay manuscript records young Abung women rejoicing at raiders’ return.

Northern Thai villagers recall that girls could be captured as “play wives” by passing militia, and folklore tells of village parents hiding daughters by smearing their faces with fish paste. In Palembang and Jambi, east Sumatra, rulers raided local stores of pepper and carried off their “prettiest daughters”.

Reflecting on this evidence, I began to suspect that cross-cultural research has radically understated historical patriarchy by ignoring pervasive violence.

Does ‘Female Agency’ track Safety?

Curious, I turned to the Standard Cross-Cultural Sample (SCCS) - a widely used anthropological dataset of 186 small-scale societies. Each society is coded on variables such as descent, marriage, gender norms, political organisation, homicide and warfare. This sample was designed to minimise regional clustering and shared historical ancestry, thereby strengthening cross-cultural comparisons.

Like all such datasets, this coding relies on observations recorded at particular time periods and reduces social complexity to discrete categories. It cannot capture internal variation (for example, elites versus marginalised sub-groups) nor long-term change. Nevertheless, it provides the most systematically comparative evidence currently available. Separate research teams coded different measures of violence, thereby reducing the risk that any single interpretive bias might drive the results.

Below, I explore whether conventional indicators of female freedom correlate with violence.

1) Violence was pervasive

In 80% of coded societies, male toughness was explicitly glorified.

In 80% of coded societies, wife beating was reported as common.

Institutional rape was present in 53% of the societies coded on that variable. Rape, or gang rape, was a socially legitimate tool of punishment, discipline, or warfare.

The Montagnais–Naskapi - caribou-hunting people of northeastern Canada described by Graeber and Wengrow as respecting female liberty - are coded as practising institutional rape.

2) Warfare is correlated with interpersonal violence

Not all pre-industrial societies were violent or warlike. But societies that frequently waged war also tended to exhibit more interpersonal violence - more assaults, coercion and homicides.

Ember and Ember (1994) suggest this pattern could be partly driven by warring societies socialising boys for aggression. Similarly, I find that societies with high homicides and assault tended to glorify toughness.

3) Over half of pre-industrial societies score highly on male aggression

Male aggression is coded on a cumulative Guttman scale, with the highest score assigned to societies that institutionalised rape and raided for wives. Over half of pre-industrial societies in the SCCS reach this maximum level.

The Yanomami, Chuckchi, Kazakh and Somali all score high on male aggression, warfare, interpersonal violence, glorified toughness and institutional rape. In the absence of a centralised monopoly of force or reliable guarantee of security, Somali patrilineal clans defended pastoral herds and lineage honour through threats of retaliation and feud.

Only 19% of coded societies did not glorify toughness. These were typically more peaceful, eeking out livelihoods in peripheral forests and rugged uplands - like India’s Garo and Adivasis.

4) Conventional indicators of female agency are statistically unrelated to violence

Standard measures of ‘female agency’ (including patrilineal descent, patrilocal residence, restrictive pre-marital sex norms, and arranged marriage) are statistically unrelated to measures of violence.

The graph below plots the Carmichael–Rijpma ‘Family Constraints on Women’s Agency’ index against violence. Two components account for most of the variation: patrilineal descent and patrilocal residence.

I also plot female property ownership against violence. As you see, there is no relationship. Even if women lived near their natal kin, owned property, or were free from arranged marriage, they might still be vulnerable to male predation.

The Bemba in Southern Africa were shifting cultivators, valuing not property but ‘wealth in people’. They practised matrilineal descent and matrilocality. A groom typically performed several years of labour for his in-laws to demonstrate readiness for marriage and then lived nearby. Women could divorce and return to their natal kin, often remarrying.

In the 19th century, this co-existed with warfare: Bemba chiefs expanded their territory by raiding neighbouring groups and taking captives. In 1832, Gamitto reported that ‘the Bemba.. are today powerful and feared by all for their fierceness’. In 1895, [Chief] ‘Ponde’s men captured Chitimbwa’s village, killed the chief and carried away rich spoils of cloth, salt and women’.

In the Standard Cross-Cultural Sample, wife-beating is coded as common. Still today, in Luapula (the Bemba-heartland, where I used to live), over half of women say that wife beating can be justified.

The Igbo of southeastern Nigeria are often presented as evidence of a more egalitarian precolonial order. The language is largely gender neutral, political authority was decentralised and women organised powerful assemblies. In village-wide gatherings (mikiri), women shared grievances, regulated markets and coordinated sanctions. If a man abused his wife or violated market rules, women might gather at his compound, bang pestles against his walls, shame him publicly, and create a din until he submitted. This was called ‘sitting on a man’. During the Women’s War of 1929, the same networks mobilised against Warrant Chiefs (colonial appointees).

British colonial officers and missionaries can be rightly faulted for neglecting these significant institutions. Yet later correctives seem to have ignored the wider context. In the SCCS, the Igbo rank among the highest for homicides, with frequent inter-village warfare and retaliatory violence.

The Fon of Dahomey are coded with the maximum score for female power. By the 1840s, they had 6,000 female professional soldiers - recently popularised in the Hollywood film, “Woman King”. Women ran markets and the Queen Mother (Kpojito) was politically influential.

Yet the Dahomey also had interpersonal violence, institutional rape and wife-beating. This was a slave-raiding state, which gained strength from organised warfare. Elite female prominence co-existed with slave-raiding, coercion and enslavement.

Vast populations lived under insecurity and retaliatory violence. It follows that regimes which successfully monopolised force and pacified society, thereby reducing the returns to male aggression, created a necessary (though not sufficient) condition for greater gender equality

Violence Entrenches Patriarchy

Male violence isn’t just downplayed in the historical record, it’s often omitted from discussions about contemporary societies. For example, the World Economic Forum’s Gender Gap Report tracks health, education, employment and political representation. All four are important, but omit a hugely important obstacle emphasised in so many of my interviews across the world: male aggression, coercion and brutality.

Economists may suggest that property ownership raises women’s bargaining power, but this surely breaks down if one can be captured and enslaved. Likewise, sexual permissiveness does not mean female agency when women may be abducted by male predators.

“In the 1990s, Russia was like a Wild West, many of my friends’ dads were killed by the mafia” - shared Julia from Vladivostok. In the shadow of organised crime, working class men learnt to project brutal dominance, uploading videos of their fights, impressing peers with physical aggression. Desperate for security, Russians favoured tough male leaders. Worldwide, militarism and geopolitical threats consistently boost demand for macho leaders.

Today, in Russia-occupied areas of Ukraine, there are reports of killings, abduction, torture, electric shocks, and gang rape. Victims of sexual abuse range in age from four to 82. This reflects a recurring pattern of sexual violence as a weapon of war.

Over in Latin America, 52-58% of people say they live in the presence of criminal groups, while 12–16% report some form of criminal governance. Across the continent, high homicides strongly predict high femicides. Across Mexico, Costa Rica and Brazil, women often described older men as being intensely ‘machista’ - controlling and quick to anger. Some interviewees attributed their self-doubt and nervousness to early experiences of fear and intimidation.

Amid climate stress and state fragility, the African Sahel has seen a surge in Islamist militancy and armed conflict. The region now accounts for 51% of global terrorism-related deaths.

Protection from Male Predators

In their eagerness to challenge myths of universal male dominance, scholars scoured historical records for evidence of women’s prominence. But across 186 pre-industrial societies, I find that standard indicators of female agency (like matriliny, sexual permissiveness and female political participation) can co-exist alongside homicides, warfare, and institutional rape. Although scholars were right to highlight women’s political participation, this obscures a much harsher reality.

Violence was endemic across many non-state societies. Celebratory narratives of ‘kick-ass’ women often obscure head-hunting, slave-raiding and intimidation.

Going forwards, I suggest that measures of gender equality should incorporate indicators of generalised violence, as a major threat and brutalising force.

‘Getting to Denmark’ doesn’t simply mean gender quotas for political elites, but building states that enforce rule of law and protect vulnerable girls from predators. I would love to see more research on how this can be achieved, worldwide.

Related Essays

• An Intellectual History of the Patriarchy

• “Gendered Species: A Natural History of Patriarchy” by Tamas David-Barrett (Review)

• Rose-Tinted Patriarchy

• Against Rose-Tinted Patriarchy

• Why are So Many Latin American Women Beaten & Murdered?

• Why Do Some States Protect Women While Others Enable Abuse?

• Wartime Patriarchs

Notes

1. This is a nice example of a useful null finding. These are under-reported in academia, as shown in a new paper by Ryan Briggs.

2. Male toughness and institutional rape are coded on a cumulative Guttman scale. On this metric, you can have societies glorifying toughness without institutionalising rape, but societies institutionalising rape will glorify toughness. Since this creates a mechanical correlation between these two variables, I also incorporate other, independently coded indicators of violence, such as homicide and assault.

3. My violence index is a min–max normalized row mean of three standard indicators: homicide, assault, and interpersonal violence. All substantive conclusions are identical when I expand the index to include wife beating and institutional rape, and when I restrict the three-item index to societies with at least two coded components. I also replicate the null results using homicide and assault separately, confirming that the results are not driven by index composition.

4. I do find a correlation between restrictive premarital sex norms and homicide, but this relationship is driven by 21 societies that are both extremely restrictive and extremely violent. If these cases are removed, the correlation disappears. Looking more closely, this cluster of societies tend to be pastoral economies with organised violence, waging wars for resources, and socialising boys to be aggressive. Otherwise, there is no relationship between pre-marital sex norms and violence.

5. In “Better Angels of Our Nature”, Stephen Pinker suggests that small-scale societies had extremely high rates of war deaths and homicides.

But this pattern is far from universal. Why has cultural leapfrogging been so rapid in Turkey, why do young men often remain more conservative, and what does this mean for coupling?

Turkey’s Cultural Fist-Fights

For over two centuries, Westernising reformers have clashed with social conservatives over authority, women’s visibility, and autonomy. Still today, male honour remains tethered to female seclusion. Even online, Turkish men have very few female friends. Ideals of gender segregation help explain why - despite economic growth, soaring education and falling fertility - only 36% of women participate in the labour market. Men continue to dominate state power, comprising 80% of parliamentarians.

After the 2001 economic crisis discredited secular elites, Erdoğan’s Justice and Development Party (AKP) rose to power - blending market reform with Sunni revivalism. Yet Turkey continues to struggle with weak business dynamism and high inflation. The AKP maintains its grip partly by favouring cronies and punishing dissent: jailing opponents, curtailing competition, weakening judicial independence, and replacing university rectors with loyalists.

But while the AKP maintains support among older voters, younger cohorts appear to be experimenting with alternatives.

4 Drivers of Cultural Liberalisation

Turkish liberals benefit from four forces:

First, when a governing regime fails to deliver, its ideology often takes a battering. Acemoglu and colleagues find that unsuccessful democracies lose support, and I suggest this holds for governing ideologies more broadly. Citizens don’t just lose faith in the ruling party, they also lose faith in its underlying mantra - whether that’s communism, progressivism etc. Since Turkey’s AKP has been in government since 2002, and is widely blamed for economic mismanagement, Sunni orthodoxy is also losing favour.

Second, compared to neighbouring Muslim countries, Turkey always had a stronger support base for secularism. Today, only 41% of Turkish parents cite ‘religious faith’ as an important quality for their children. Whereas in Egypt, 82% underscore this priority and are much more insistent on ‘obedience’. The Egyptian school system also provides many more hours of religious education.

Third, Turkey has seen skyrocketing university enrolment - from 1.7 million students in 2008 to over 4.1 million by 2020. Away from home, young people are socialising more freely, carving their own spaces. Female graduates are also more likely to be employed.

Fourth, female graduates in big cities are increasingly engaging with progressive Western values. Smart phones and cheap data mean they can circumvent parental surveillance and national broadcasters, then gravitate to channels that validate their aspirations. On Instagram, TikTok and Spotify, liberal women can speak and be heard. I call this ‘culturally leapfrogging’.

Just as artists once flocked to Paris - the epicentre of prestige in the 19th century - and repackaged prestigious styles for domestic audiences, Turkey’s feminist entrepreneurs are re-imagining female status, happiness and autonomy. They are asserting their own prerogatives, reassessing intimacy and prioritising their own happiness - to large online audiences. For the first time in Turkish history, liberal women can really engage in large-scale ideological persuasion. Their weapons of choice? Comedy, music, and glamour - fun, aspirational, online social-bonding.

Digital Footprints

Google Trends data helps us triangulate surveys, which are often irregular, marred by non-responses and social desirability bias. For example, some may report daily prayer, but nonetheless skip a few.

Searches for ‘prayer times’ suggest the extent to which Muslims incorporate religiosity into their daily life, by dutifully checking the appropriate times for Fajr (dawn), Dhuhr (midday), Asr (afternoon), Maghrib (sunset), and Isha (night).

Analysing Google Trends, I can visualise Turkey’s regional heterogeneity. In Istanbul, locals are 2.4 times more likely to search for ‘Tinder’ relative to ‘prayer times’ (namaz vakitleri).

Below, each province’s bar shows the ratio of its Google Trends search index for Tinder relative to its index for Prayer Times (’Namaz Vakitleri’), with both terms independently scored by Google based on search concentration rather than raw volume. For more details see footnotes.

Reflecting growing embrace of US norms, searches for ‘ADHD’ and ‘lip filler’ have spiked - in both Istanbul and Konya. As we spend more time online, our feeds and algorithms, rather than neighbourhoods, play a growing role in shaping cultural expectations. And this is entirely consistent with my qualitative research…

Konya

In Konya, I was introduced to a young hijabi, yet within five seconds of meeting, I instantly knew she was plugged into Western social media. Her gestures gave the game away - fingers clicking like the supremely confident Nicki Minaj. Sassy, astute and assertive, Ece became a great friend, generously helped me interview a range of people across Konya.

Ece’s friends - fluent in English - have formed a bookclub, stacking their shelves with Sally Rooney, Dolly Alderton, Virginia Woolf, Rebecca Kuang, Gabrielle Zevin, and Kazuo Ishiguro. These novels portray lives that are so radically different from that of their mothers and grandmothers. Instead of enduring abuse, hierarchy and domesticity, they are continually choosing protagonists who contest claims to power, while re-evaluating intimacy. Love, friendship, and even family ties are constantly open to scrutiny: “Is this actually good for me?”. Women’s desires for recognition and pleasure are cast as entirely legitimate. That may sound mundane to Western readers, but in central Anatolia it is absolutely radical.

With three universities, young people come from all over Turkey, mixing without parental supervision. Cafes - once male preserves - now host young women in crop tops. As Ece explained,

“If you go to cafes, everyone is young, no one will care. My friend [from Istanbul] was really shocked about the situation, like “omg it’s crazy”…

“Today, one of my friends sent me a photo, it’s ‘celebrating winter’ [She laughed at this coded description of Western Christmas decorations].

Male Comics

When I study world history, I realise that cultural liberalisation has usually been spearheaded by men - contesting strictures, legitimising dissent. This isn’t to diminish women’s contributions: it merely reflects the reality of patriarchy: men exercise greater authority and act as gatekeepers.

Comedy is a powerful solvent. Laughter triggers endogenous opioid release, forging social bonds. We relax, connect and feel that we’re all in on the joke. In that moment, it becomes easier for relatable and charismatic figures to broach taboos and persuade their peers.

When orthodoxy is mocked, it loses gravitas. Sacred tenets are pulled off their pedestals and deviance becomes more acceptable.

On Netflix, Cem Yılmaz and Hasan Can Kaya mock the absurdities of tradition - family authority, sexual hypocrisy, and macho posturing. While they never target Islam itself, satire is nevertheless a powerful weapon of subversion.

Mocking the Patriarchy

Marshalling this same weapon - irreverent comedy - women have been sharing their own ideas. Yasemin Sakallıoğlu, 37, with over seven million Instagram followers, pokes fun at women’s never-ending drudgery, cooking and cleaning for religious festivals. As a middle-aged mum, she’s extremely relatable - never aggressive or polemical. Much like U.S. women’s magazines in the 1970s, she brings feminist ideas to a mass audience through irreverent, self-effacing humour.

While some economists maintain that women’s low employment in the Middle East, North Africa, Anatolia and South Asia is due to care work, Yasemin Sakallıoğlu is rocking our priors. Patriarchal societies that idealise female domesticity are highly effective in ensuring women perform more housework.

Over on Instagram, femininity is being redefined, with glamorous Turkish influencers chronicling their exotic adventures. Spotify’s top streaming artists include Melike Şahin and Gülşen, who champion women’s independent desire, autonomy and pursuit of happiness. Popular anthems are widely remixed on TikTok and Instagram, signalling wider acceptance.

Beyond beats, banter and lip filler, many Turkish women also look up to articulate, knowledgeable journalists, like Nevşin Mengü. As Ece remarks from Konya, this connectivity shows a whole new world of possibilities:

“Before, you only knew your family, or your sister’s husband. Now, you can see EVERYTHING!”.

In a country where a third of women have experienced domestic violence, viral terms like ‘gaslighting’ name harms and help legitimise exit.

Bodies and Backlash

In societies where women’s bodies are a dangerous source of ‘fitna’ (moral corruption), they are encouraged to hide their bodies and cover up:

“I was constantly ashamed of my body. I felt like my body was not my own. When I was going through puberty, I was constantly told to cover up, to wear bigger clothes, to shut my legs. My father made me feel so disgusting. And this affected my sexual relationships, because I felt like my body was not my own” - Yildiz, Istanbul.

Staying in Istanbul, I joined a local gym, yet rarely saw any women. Asking more broadly, many expressed extreme discomfort. Here, the Turkish Women’s Volleyball team has been catalytic! In 2023, they won the Women’s European Volleyball Championship, becoming a major source of national pride (and attention). Sportswomen demonstrate their strength and agility, casually wearing shorts. But this has stirred backlash. After photographs circulated of Ebrar Karakurt embracing another woman, online bullies came out in droves, with one Islamist newspaper calling her “a national shame.”

In 2021, when the team was competing in the Summer Olympics in Tokyo, a prominent preacher sharply criticized the team for not adhering to his conception of how a Muslim woman should behave. Preacher Ihsan Senocak, with 1 million followers on Instagram, blasted condemnation:

“Girl of Islam! You are not the sultan of the courts; you are the sultan of faith, virtue, chastity and decency”.

Cultural fist-fights continue. A spokesperson for Turkey’s volleyball federation spoke out in her defence, praising her “spirit of a fighter to represent her country”, adding “Everyone’s private life concerns them only. All the rest is hokum.”

Ghosting the Patriarchy

Social media is a double-edged sword. On the one hand, it has enabled Turkey’s female graduates to leapfrog towards egalitarianism. But persuading their male peers remains hard, because many young men appear to be gravitating towards icons that affirm their own status. In 2022, searches for Andrew Tate surged in both Istanbul and Konya. Interest in Menzil - a conservative Naqshbandi Sufi community - is often even higher.

Smartphones may deepen separation - especially in societies where men and women already keep their distance. As attention shifts to hyper-engaging apps, face-to-face connectivity may decline. Over time, this makes it harder to build new friendships, networks and personal charm. Yet if society liberalises such that young people can choose their own partners, selecting for companionship, then weak interpersonal skills create a real barrier to coupling.

Since the early 2000s, Turkey’s marriage rates have fallen, divorce has more than doubled, and fertility has dropped to 1.5. Last week, Ece went on a first date but was distinctly unimpressed:

“He was very kind, but not funny. I wasn’t mesmerised. And he said, ‘you’re in business, “that’s so great for a woman”’. I thought “What the hell?”.

While he was trying to be supportive, Ece found his approach antiquated and patronising. “Thank u, next”, I joked in reply (rightly anticipating one of her favourite songs). Thus even if Turkey’s young female graduates struggle to shift electoral outcomes, they are increasingly saying ‘no’ in their personal lives.

Will Cultural Leapfrogging Go Global?

So will TikTok catalyse worldwide secularisation? Not necessarily. Cultural prestige often tracks perceived political competence. When religious regimes falter, the pews tend to empty - as in Spain after Franco or Chile amid Catholic sex scandals. But where secular elites are blamed for corruption and economic stagnation, religious movements can harness grievance and present themselves as purer, more just alternatives - as in Egypt in the 1970s and 2010s, or Bangladesh’s Jamaat today.

Social media does not determine the direction of cultural change; it amplifies a raft of anti-establishment voices, who then battle to present a more alluring alternative.

Related Essays

• Culture isn’t a Fossil, it’s a Fist-Fight

• Can Men and Women Be Friends? Evidence from 1.3 Trillion Facebook Connections

• Why are Marriage & Fertility Plummeting in MENA?

• Cultural Leapfrogging: Swiping Past Tradition

Footnotes

• To explain the Methodology, Usama Polani used Google Trends to compare the prominence of “Tinder” and “Namaz Vakitleri” in each Turkish province’s search activity.

• When you query the Google trends API in R, it does not give you raw search counts, it gives you a 0-100 index where 100 is assigned to the province where that term is most prominent relative to all searches, and other provinces are scaled proportionally.

• Usama fetched each term separately across 81 Turkish provinces for 2023 and then computed the ratio of the indexes. A ratio above 1 means Tinder is more prominent in that province’s search activity; below 1 means that prayer times dominate. Istanbul’s ratio of 2.4 means that Istanbul scores 2.4 times higher on the Tinder index than on the prayer-times index, so it ranks much closer to the top on Tinder prominence than it does on prayer-times prominence.

• Clearly, there are many other differences between provinces. For example, larger cities might have more prayer apps, Tinder swiping or audible azan calls to prayer. Therefore Usama ran 6 additional tests to gauge whether this within-Turkey difference in search habits was picking up an underlying pattern. The same gradient in relative search intensity Usama found using the Tinder/prayer times index ratio across provinces also occurs even as you change the religious terms, so modest dress (tesettür), umrah pilgrimage, Diyanet or Quran. There is a consistent pattern. The correlation in province rankings for the ratio across terms was often above 0.8.

• Usama then ran tests to identify whether a Ramadan religiosity effect existed and varied across provinces. He confirmed that during Ramadan the spread of Tinder/prayer times ratios across provinces compresses but the provincial ranking persists.

• Analysis over time also confirms the findings. Looking at search activity from 2016-2023, Konya sits relatively flat on the ratio while Istanbul and Izmir show significant increases.

• When looking directly at cities above 1 million, İzmir (4.5m population) has a Tinder/prayer ratio of 3.41 while Konya (2.3m population) has a ratio of 0.46. So even with both being major cities, Tinder is over 7 times more prominent relative to prayer times in İzmir’s searching behaviour than that of Konya.

• Usama also regressed each province’s Tinder/prayer times ratio on log(population). Without implying causality, this accounts for 24% of the variation.

• When a direct religiosity measure is added, each province’s Google trends index for tesettür, the explained variance becomes 42% with both population and the term for modest dress independently being significant.

• There are multicolinearity issues in this approach: the index for the religious term is correlated with the prayer times index, both are capturing religiosity. So the regression is partly picking up the relationship between the regression and the denominator the ratio. This is both mechanical and also the main thrust of the analysis: religiosity terms covary strongly across provinces. Even removing these mechanical correlation issues, the results hold.

• Usama also investigated how AKP voting share may be related to the Tinder/prayer ratio if the former also proxies in part for religiosity. In the full sample there is no clear role for AKP voting share to explain variation in the ratio but that may be because Kurdish provinces are deeply religious and vote for HDP/DEM, not AKP. If Usama excludes those provinces then AKP vote share becomes one of the strongest predictors of the variation in Tinder/prayer times ratio although of course many other factors are influencing both political choices and attitudes. Usama also ran cross-country analysis and finds similar results: Egypt is much more uniformly religious on my Google trend metrics than Turkey which is consistent with other surveys on religious attitudes.

As Epstein’s emails hit global headlines, many were bewildered by his plethora of allies. So many powerful men - titans of finance, tech, politics, and academia - cultivated his friendship and sought invitations to “wildest nights”. Email chatter is, of course, perfectly legal. The deeper problem is that powerful men socialised with a known pedophile, while escaping any kind of judicial inquiry or reputational damage. Protecting their own, governments seldom pressed too closely. As Julie Brown documents in “Perversion of Justice”:

“In liberal circles, political analysts who studied the case often focused on Epstein’s connection to Donald Trump, while conservatives tried to link the scandal to Bill Clinton”.

Studying sexual abuse worldwide, I see a recurring pattern: where powerful fraternities leverage their institutional power to block accountability, violent predators walk free. Seldom seeing accountability, victims are often reluctant to invest years in stressful, adversarial court trials - only to be dismissed. Instead, they remain silent. I call this a “Despondency Trap”.

To illustrate these dynamics, consider two prominent cases of institutional cover-ups.

The Catholic Church

“It is all slander”, declared Pope Francis, dismissing accusations that Chilean bishops had covered up pedophilia. Senior Cardinals likewise dismissed such “gossip”. For centuries, the Church entrenched its moral supremacy by protecting bishops, forgiving their sins, and quietly moving them between parishes.

In 1970s and 1980s Latin America, the Catholic Church maintained popular prestige by supporting human rights campaigns against the brutal military dictatorships. This legitimacy was preserved and perpetuated by elites who were often educated at Catholic universities and monopolised the means of ideological persuasion.

But this hegemony was subsequently challenged. By 2018, 85% of Chileans were connected to the internet with a smart phone, allowing evidence of abuse to go viral. Priests came to appear hypocritical, as they had presented themselves as righteous while privately abusing minors. Trust in the Church rapidly plummeted.

With mounting public pressure, Chile’s National Public Prosecutor opened nationwide criminal investigations, raided dioceses, and subpoenaed cardinals - bringing 229 investigations to trial. For the first time, Chilean law enforcement treated the Church as subject to ordinary criminal investigation. Just like struggles for accountability in Ireland, the Church only lost its chokehold and became accountable when it was knocked off its righteous pedestal.

Globally, the Vatican changed tack on sexual abuse. Rather than suppress dissent, it began to ally with victims. In 2018, the Vatican hosted a conference in which survivors shared how they had been shamed and condemned by Church leaders:

“The first thing they did was to treat me as a liar, turn their backs and tell me that I, and others, were enemies of the church” (Chile).

The pattern is clear. Whether it is loyalty to the Church or US political parties, institutions dig in their heels and defend their own. This exact same process of institutional defensiveness currently looms over quests for justice in my homeland.

Britain’s Grooming Gangs

In Britain, thousands of girls have been sexually abused by organised groups of men. Across towns and cities of England, vulnerable girls were showered with gifts and affection, isolated from their friends and family, then plied with drink and drugs, bullied, raped, and trafficked as part of large criminal networks. Yet when victims, journalists and researchers sought help or pressed charges, they were dismissed and derided.

In Rotherham, journalist Andrew Norfolk chronicled widespread child sexual exploitation. In response the local government sought an injunction, asking the police to launch a criminal inquiry into the leak! Despite 256 separate allegations by 50 complainants, no officer lost their job. Thanks to the absence of accountability, the violence persisted – within 15 years, an estimated 1,400 children were sexually exploited.

Abuse persists. In one northern city, a British Pakistani politician told me how she had uncovered evidence of senior men in her community grooming a young boy, she reported it to fellow representatives, and instead of investigating abuse, she lost her appointment, received violent threats, and her house was burgled - with thieves only taking her computer.

In Baroness Casey’s National Audit on Group-based Child Sexual Exploitation and Abuse (CSE) (2025), she reveals repeated failures of policing, social services, local councils, and political leadership. Statutory agencies chronically neglected credible evidence of organised child sexual exploitation. Police and local governments:

• Minimised or denied the scale of abuse;

• Dismissed victims, treating them as wilfully complicit;

• Suppressed internal reports and whistleblowers;

• Failed to record basic offender characteristics, including ethnicity;

• Cited fear of community tensions and accusations of racism as reasons for inaction; while

• Resisting external scrutiny by obstructing public access to data.

Absent a full inquiry, the true scale of abuse remains unknown. Where suspect ethnicity has been recorded, such as in Greater Manchester, South Asians are over-represented relative to their share in the local population. But for the vast majority of cases, the police simply do not report ethnicity.

Progressives thus face a difficult bind. On the one hand, they share a commitment to women and girls’ safety. On the other, there are genuine fears that scrutiny could inflame community tensions or bolster support for the far right.

Just last year, Prime Minister Keir Starmer dismissed calls for a grooming gang inquiry as jumping on a “far-right bandwagon”. Victims’ quest for justice were illegitimate because of the identity of those accused. London mayor Sadiq Khan likewise rejected requests for extra funding as “playing party politics”. When Labour Minister Mike Tapp called for an inquiry into the link between ethnicity, religion and rape, leftists condemned it as racist.

Fearing such accusations and ostracism, many have self-censored. Casey’s Audit emphasises,

“We found many examples of organisations avoiding the topic altogether for fear of appearing racist”.

Grooming gang survivors have often expressed deep disappointment with government inaction. Fiona Goddard resigned from the victim liaison panel, saying “the secretive conduct and conditions imposed on survivors has led to a toxic, fearful environment, and there is a high risk of people feeling silenced all over again”. Ellie-Ann Reynolds also resigned in dismay that the inquiry’s broader remit would “downplay the racial and religious motivations behind our abuse.”

How Can We Prevent Abuse in the 21st Century?

When governing regimes feel threatened by accusations of abuse, they tend to crack down on dissent to maintain legitimacy. Victims are pressured into silence - purportedly for the sake of the greater good. Over and over, they prioritise institutional self-presentation. Just as Chilean Catholics and US political parties protected their own, Western progressives have been reluctant to discuss, let alone investigate, crimes that implicate ethnic minorities. Even after two decades of grooming gangs, British sociologists, criminologists and mainstream media remain weirdly silent.

This strategy is short-sighted. Institutions that are supposed to serve the general public forfeit legitimacy if they show favouritism, while others feel unheard. As Acemoglu and colleagues demonstrate, “successful democracies breed their own support”. Governing regimes only sustain support when they deliver growth, services and safety for the broad majority. Extending this logic, political parties may expect electoral defeat if they fail to uphold accountability and public safety.

The real threat is not scrutiny, but impunity.

While romantic love is experienced worldwide, there is enormous variation in the extent to which it is celebrated or suppressed. In regions where marriages are arranged by kin or coerced through brutal violence, her wants and welfare count for little. If divorce is stigmatised, she cannot credibly threaten exit and must then endure any abuse.

But where men must compete for a woman’s hand, they must demonstrate devotion and emotional companionship. A man who cherishes his wife tries to make her happy, supporting her ambitions and sharing the washing up.

What follows is a speech I performed at my German friends’ wedding: two thousand years of history through the lens of marriage, starting with Ancient Rome to the Reformation, Wars of Religion and subsequent Romanticism, all the way to 1970s counter-cultural liberalisation.

To keep it vaguely romantic, I have omitted some parts of German history.

Classical Inheritance: Marital Monogamy

Male citizens of Athens and Rome were only permitted one legal wife. Preventing polygamous hoarding gave every free man the opportunity to get a wife, sire children and pass on property. Egalitarianism among male citizens may have aided fraternal camaraderie and state legitimacy.

Importantly, however, normative monogamy is not sufficient for female choice nor male competition. Daughters’ marriages were still arranged by families, while elites could procure infinite concubines - via slavery. Only with the demise of labour coercion did Europe achieve de facto monogamy.

In the early Middle Ages, the Catholic Church began enforcing a dramatic transformation of marital norms. They banned cousin marriage, which eroded clan-based kinship in Western Europe; and they required consent from both the bride and groom, which undercut arranged marriage outside the aristocracy. The wedding ring was placed on the fourth finger - a direct line to the heart, representing their unity. The Church also transformed marriage into a sacrament - a profound union, considered equal to baptism or communion.

Economic shifts encouraged a shift towards male competition and companionate marriage. By the 15th century, serfdom had disappeared in north-western Europe, replaced by wage-labour markets. Young men and women often left home to work on farms, workshops, and manor houses. In Württemberg, most women in their late teens and early twenties did not spend their days supervised by kin - they were workers. Mixing, mingling, flirting and listening to gossip, single women could assess potential matches. Just like any functioning market, men had to compete. Oafs, thugs, paupers and bores might well remain single.

Couples then established separate households, away from extended kin. This privacy and mutual dependence permitted greater emotional intimacy.

Yet Catholic theologians’ highest ideal was actually celibacy. Marriage was not framed as a site of mutual fulfilment, but rather a concession to human weakness, a remedy for sin. Sex, even within marriage, was marred with suspicion, tarred with lust and loss of self-control after the Fall. Meanwhile, misogynist proverbs ran rampant:

“If you find things going too well, take a wife”.

The Reformation: Marriage as Companionship

Protestant Reformers got lucky: their theological movement coincided with printing technology, university networks, political fragmentation. Luther’s pamphlets sold like wildfire, racing through personal networks, while universities became hotbeds of radicalisation, turning students into fervent reformers!

Protestants praised marriage based mutual affection and companionship. This catalysed an explosion of marriage sermons, household manuals, and advice literature - urging husbands to rule wisely, love their wives, govern gently and cultivate mutual affection. English Puritan Robert Cleaver (d.1613) wrote,

“The husband is not to command his wife in manner, as the Master his servant, but as the soul doth the body, as being conjoined in like affection and good will; just as the soul in governing the body tendeth to the benefit and commodity of the same, so ought the dominion and commandment of the husband over his wife . . . tend to rejoice and content her”

A wife should be emotionally mature and financially able to support a household - at least over the age of 20. And as for sex, Protestants condemned prostitution and raided brothels, but regarded sex within marriage as entirely legit! Together, these forces - monogamy, consent, ideals of mutual respect, and wage work among youth - incrementally pushed marriage toward female choice, male competition, and emotional companionship. Neil Cummins’s research on English wills reveals this shift – from the 17th century, men increasingly referred to their wives with affection.

Small print: Germany was by no means a feminist utopia. Guilds were the ultimate frat club, which locked women out. Every institution, every mode of power and persuasion was dominated by men. In religiously-contested territories, rivals struggled to demonstrate their superior power to catch and snare Satan’s agents. Areas with intense wars of religion saw more witchcraft trials: vanquishing the devil.

Towards Liberty!

From the sixteenth to the seventeenth century, Europe was torn apart by religious wars. Convinced by their claims to absolute truth, salvation and authority, Protestants and Catholics sought to entrench their dominance and persecute heretics. Europe’s zealots were on the rampage - razing towns, torturing suspects. God’s war proved immensely destructive.

Exhausted by generations of bloodshed and disillusioned by imposed orthodoxy, European philosophers began to question whether beliefs could, or should be coerced. Protestantism had already encouraged ‘sola scriptura’ - knowing God by reading the Bible. Now, this principle was applied more broadly, emphasising individual conscience.

In England, John Locke (1689) pushed for religious tolerance, arguing that since people’s personal beliefs cannot be changed by force, and spiritual beliefs are private, then coercion only breeds hypocrisy and violence. English people increasingly accepted that people could differ - in beliefs, ideas and expression - and this was fine.

Simultaneously in German lands, theologians like Philipp Spener were advancing a similar transformation. Pietists used believers to cultivate inner faith, rather than outward conformity. True Christianity was not about dogma, but heartfelt sincerity and inward conviction.

Meanwhile, the print industry was booming! Pamphlets, newspapers, and cheap prints circulated across Europe, while coffee houses and debaucherous inns became lively hubs of discussion, stocked with opinionated newspapers and biting satire that savaged the airs and graces of the nobility. Elites were kicked off their pedestals and disagreement became part of ordinary life.

‘Free speech’ was gaining traction, though more as an aspiration than reality - championed by underdogs, but suffocated by those in power. The French and American Revolutions were part of this wider shift, reflecting men’s growing demands for political, religious, and sexual freedoms.

Romanticism, the Self and Subjectivity

If private sincerity mattered more than external conformity, shouldn’t inner feelings have moral weight? Building on this culture of tolerance, private conscience and debate, European philosophers, poets and artists embraced Romanticism: moral value was no longer defined by the church, crown or custom, but rather the Self!

Romanticism elevated subjectivity: a man’s emotions, imagination and longing. Turner’s storms, Goya’s nightmares, and Friedrich’s Lone Wanderer invited viewers to engage emotionally - as if feelings have some moral validity. Value lay not in sacred law, but one’s mind.

This radical culture of the “I” flourished in the small university town of Jena (Germany), where writers and thinkers bucked all kinds of convention - in marriage and philosophy - to pursue emotional authenticity, following their own desires. Freedom was revered as a personal imperative. The sovereign male self - expressive, inward, and restless - became the new template of cultural prestige.

Thanks to expanding print networks, rising literacy, circulating libraries, letters, and salons, this ideal spread with speed and gusto! Young educated men in Leiden, Copenhagen, and Geneva came to see themselves as leading protagonists.

Rewriting the Script!

One of Europe’s most popular novelists of the early 19th century - Johann Wolfgang von Goethe - celebrated not just inner feelings, but love itself. In “The Sorrows of Young Werther”, Werther’s intense longing for Lotte is morally legitimate. The tragedy is not his love, but societal constraints.

As economic growth expanded the middle class, women writers embarked on a hitherto unrecognised revolution. For the first time in human history, they could reach the literate masses and engage in large-scale ideological persuasion.

Sophie von La Roche (raised as a Pietist) wrote a best-selling novel in 1771, “The History of Lady Sophie Sternheim”. When our heroine enters bourgeois society, she is surrounded by male suitors, each competing with wealth, flattery and power. She herself evaluates men’s sincerity, respect, and devotion. A powerful noble lavishes her with attention, but does not respect her refusal. “No” does not compute. Ultimately, our protagonist judges male behaviour, evaluates suitors, and accepts the man who respects her judgement.

From the 18th century, European bourgeois ‘civility’ was largely defined in terms of capacity to make engaging conversation, including with women. Female choice raised the returns of becoming polite, witty, agreeable, and sociable!

At the same time, marriage was subject to institutional reform. Under Frederick the Great, clerical authority over marriage weakened as the state expanded control over divorce, inheritance, and schooling.

1970s Counter-Cultural Liberalisation!

In May 1968, the streets of Paris became a battleground for liberalism, as students, workers, and residents stacked breeze blocks and overturned cars around the Sorbonne, defying police tear gas to demand educational reform, free speech, and sexual liberties.

Dissent spread across university networks - in West Berlin, Munich and Frankfurt. New counter-publics analysed power and oppression, forming magazines and newspapers like Liberation (1973) in France, Lotta Continua (1969) in Italy, and tageszeitung (1978) in West Germany.

Since Europeans lacked a culture of female seclusion, university students mixed freely. But this also made women vulnerable to violent opportunists. While young men wanted to indulge in sexual freedoms, women sought rights and protections. Activists demanded “Mein Bauch gehört mir” (my belly belongs to me). They demanded respect for their choices and protection from abuse. Popular magazines and sitcoms helped these feminist ideas go mainstream - with trusted characters pursuing careers.

Love, Power & Exit

Europeans came to cherish emotional companionship, but only through two millennia of contestation, in which power was leveraged for wider persuasion. Greco-Romans insisted on marital monogamy, the Church’s demand for consent, Luther’s elevation of partnership, the subsequent turn towards inner conscience, as well as Romanticism’s celebration of emotional authenticity.

All these cultural shifts were pioneered by men, yet inadvertently strengthened women’s status. Once men had legitimised greater freedoms for themselves, women grabbed the pen and pushed for wider application. After Pietism and Romantics extolled the Self, women novelists celebrated their own prerogatives. Two centuries later, as young men lambasted traditional authorities, women expanded those ideals - denouncing shame, demanding autonomy, while simultaneously gaining the economic means to exit. And thus Europeans learnt to love, as equals.

Congratulations to the bride & groom! I am also available for bar mitzvahs, nikahs, and funerals.

Postscript: Modern Implications?

(Omitted from the wedding speech, since it’s a little dark)

European courtship historically rewarded persuasion and companionship. This sharply diverges from societies where female choice is trumped by violent coercion or clan consolidation - like Boko Haram’s abduction of school girls, Kazakhstan’s trend of bride-kidnapping, and South Asia’s suppression of divorce. If men can secure sex and social approval through coercion, skills like pleasing women become superfluous.

Now, if you accept that a culture of courtship disciplined male behaviour, feminist readers will surely wonder how to keep those incentives intact? Here, let me raise two risks.

First, technology may rewire rewards. Hyper-realistic, personalised online entertainment, chatbots and pornography offer low-cost gratification without reciprocity. For men and women who already struggle in competitive mating markets, these substitutes dampen the rewards to developing social skills. Already, we are seeing a rise in solitude and singles.

Even with AI substitutes, most people still seek physical intimacy. Whether that is pursued via persuasion or coercion partly depends on expected punishment. If the state turns a blind eye - or worse, covers up rape - then opportunists can persist with impunity. Last year, 2% of adults in England and Wales said they had experienced unwanted sexual touching or exposure. Simultaneously, there was a 7% increase in rape. In 2024, Germany saw a 9% increase in rape and sexual assault. When men get away with abuse, they have less incentive to play by the rules.

When the government retreats, parents can impose their own prerogatives. 60% of British Pakistani and Bangladeshi marriages are cross-country - usually arranged by kin.

If AI-enabled gratification expands while institutional protection weakens, boys and men may believe there are diminishing returns to being nice.

From Aceh’s hisba (religious police) enforcing weekly prayers and clearing beaches, to Salafis engaging in da’wa missionary activities, to 20% of Indonesian schoolchildren enrolled in madrasas, to TikTok megastars advocating liberal tolerance, to young men and women sharing favorite books late into the night at trendy cafés - Indonesia abounds with cultural clashes. Eager to understand these struggles for state power, prestige, and persuasion, I interviewed over a hundred fishermen, traders, entrepreneurs, university students, police, Salafis, artists, feminist activists, clerics and kyai (religious teachers).

Next up? I’ve just begun a British Academy Fellowship exploring the “Islamic World’s Great Gender Divergence” - why some Muslim societies are more gender equal than others. Simultaneously, I’m embarking on a great expedition to my favourite place: California!

Stanford University has invited me back as a Visiting Associate Professor in Economics! I’m so excited to learn from brilliant minds - like Alessandra Voena, Neale Mahoney, Lukas Althoff, Marcel Fafchamps, Walter Scheidel, Lisa Blaydes, Michele Gelfand, Hesham Sallam, Muriel Niederle, Anna Grzymała-Busse, Hakeem Jefferson, Soledad Artiz Prillaman, Saad Lakhani, and Suhani Jalota. Perhaps you recall our previous podcasts? Rest assured, I’ll be asking for more! I’m also trying to catch-up with Bay Area tech pros - who can advise me in using computer vision to track global cultural variation through a million paintings at the MET.

To comply with my J-1 visa, I must pause this Substack side hustle. So I’ve given paid subscribers a year’s complimentary membership - entirely gratis. Posting may be infrequent, as I really need to finish writing The Great Gender Divergence.

When I started this book project, I was quite frankly terrified. No single person could ever understand, let alone articulate, the causes and consequences of global cultural variation. While Large Language Models excel at summarizing public data, they still struggle with comparative rigor. And if this is the age of Artificial Super Intelligence, I can hardly expect to out-compete!

Fears loom large, and punishment fires not from the barrels of enemy guns, but the world’s multi-platform armies of critics. Trepidation about bullies and attacks might help explain why women have tended to specialise in narrow topics - leaving ‘Big Books’ to (more confident) men.

My secret weapon? You - my wonderfully generous readers - have been phenomenally helpful in sharing papers, facilitating introductions, and always rocking my priors. Wherever I venture in this world, you’ve been in my pocket - spurring me onwards, encouraging this quest. It is most certainly a team effort, racking up enormous debts of gratitude.

Thank you all so much,

Alice

P.s. If you’re near Stanford, do say hello.

My brain - in all honesty - is much the same. Despite my best efforts to study global cultural evolution, there are muddy patches.

But perhaps you can help me tackle these gaps? Here, I’m especially eager to trace people’s digital footprints - the little clues they leave while living their lives online. How can brilliant computer scientists leverage massive advances in machine learning to study cultural change? Allow me to present some ideas:

• How does the Caste System change upon Migration?

• Did Saudi Arabia’s export of Wahhabism change the Islamic world?

• Can We Track the Great Gender Divergence on TV?

• How Might we Measure Cultural Flux in Autocracies?

• What’s the Impact of Religious Schooling?

How does the Caste System change upon Migration?

My hypothesis is that India’s ultra-low female employment and ultra-low divorces have emerged from the caste hierarchy, jati policing and Islamic legacy. Indians tend to secure trusted networks of cooperation by raising their children to marry within jati. This motivates arranged marriages and strict policing. Since jatis gain status collectively, by emulating upper caste ideals of purdah (“Sanskritisation”), many uphold strict policing and suppress female mobility. This effect is strongest in North-Central India, where Muslim rulers raised the prestige of female seclusion.

Yet the Indian diaspora - from the UK to Malaysia to Mauritania - is totally different - showing high rates of female employment, less religosity and declining significance of caste. 32% of Indian Americans said that “they do not personally identify with a caste”. The share is much higher among the second generation. Moreover, Indian Americans tend to celebrate a range of festivals, including US Independence, Christmas and Valentine’s.

What explains these differences?

Drawing on my globally comparative research, I suggest that as Indians move to different economies, they can do business with a broader set of partners, and are no longer beholden to a singular jati. Indian Americans usually say their friends are diverse, rarely from their own caste. In Southall (UK), for example, one Sikh trader shared that he sources from at least 30 different traders, of diverse ethnicities. With so many prosperous connections and exit options, jatis may lose their power to police. If families are no longer fixated on endogamous marriages, young women may be permitted greater freedoms. Simultaneously, families can secure status by investing in education and prioritising upwards mobility.

Why do only 38% of US Indian Hindus describe themselves as “very religious”, compared to 71% of US Indian Muslims? I suggest that Hindus never developed religious education for the masses, so seldom carried this institution overseas. Nehru - the Secularist - explicitly to prohibit divisions and imposed strict proscriptions, rejecting religion in government schools. Religion is instead expressed through devotional chants at temples, H-pop songs, big festivities (like Holi), family rituals (like Raksha Bandhan), and online gurus. Hindu faiths haven’t really developed educational institutions. But there is one exception, which proves the rule: Bali. Here in Indonesia, the Hindu minority seeks to preserve their faith by inculcating it in schools - stoking the fire.

On the other hand, survey responses could be misleading. Eager to appear more progressive, upper castes elites may be overstating network diversity. Further, if Hindus continue to internalise ideas of purity and pollution, or if upper caste groups predominate, they may still discriminate. Soumitra Shukla studies US and European multinationals hiring in India, and finds significant caste bias. It would be wonderful to see replications within the West.

Research ideas:

1. Leverage big data from Meta to map connections across the Indian diaspora. Using surnames to identify caste, estimate intra-caste connections and marriages.

2. Why do British Hindu and Sikh women work at high rates? Does this also hold within the US? Speculatively, I hypothesise that jati becomes less significant overseas, thus families permit greater female freedoms, including love marriages and higher female employment? One might explore these correlations through large-scale surveys - within the US, UK and Malaysia.

3. Why are Indian Hindus and Muslims similarly religious, but US Hindus are much more secular? One could test my hypothesis by comparing children’s after-school enrolment in religious education both within India and the diaspora.

4. In Britain, Hindu and Sikh women are just as likely to work as work as white British women. Families prioritise not Sanskritisation through purdah, but upward mobility through education and wealth. By contrast, British Muslim families typically rely on a single male earner. How do different cultural values affect upward mobility and neighbourhood prosperity?

5. Compare migrants from the Middle East, North Africa and South Asia - countries which share similarly high rates of religosity, alongside low female employment. What percentage enrol their children in religious education (either full-time or as an after-school club). How does this vary by religion?

How did Saudi Arabia’s export of Wahhabism change the Islamic world?

Wahhabism is a distinct Islamic sect, urging strict adherence to the Quran and Hadith, purging all other innovations (bid’a) and polytheism (shirk). Under the principle of “al-wala’ wal-bara”, you should show loyalty (wala’) to true believers while actively disavowing (bara’) and distancing oneself from disbelievers/ polytheists/ innovators. This includes Sufism, worship of saints and shrines, as well as birthdays and Valentines. Importantly, Wahhabism isn’t just a set of theological beliefs, but a mandate to ostracise other Muslims for deviations.

Ever since the 1960s, Saudi Arabia has leveraged its oil wealth to propagate Wahhabism worldwide. By funding mosques, madrasas and religious scholarship, KSA globally denigrated Sufism, while ostracising non-Wahhabi Muslims, disavowing corrupting influences and innovations.

Empirically, we have a wealth of qualitative and historical studies, but these are rather piecemeal, like discrete research on an elephant’s tusks, tail and trunk. We really need to understand the interconnections of financial and cultural flows.

Based on my globally comparative research, I anticipate huge effects. Imagine there are two equilibria:

• In “A”, Wahhabis are extremely niche and generally ignored. Since they’re the minority, they can’t threaten others with ostracism. No one cares.

• “B” is Wahhabi-majority. They organise politically and have the resources to create club goods. Wahhabi bullying works, since everyone seeks status and social approval.

Saudi funding for Wahhabism likely shifted the equilibria from “A” to “B”, which then allowed endogenous processes of “al-wala’ wal-bara”, loyalty and disavowal. This could help explain the post-1970s Global Islamic Revival.

Causality may be a little murkier, however. Some aspects of Wahhabism have been adopted by other sects. Just like warfare, there’s inter-group competition: clerics from each sect extol religious education, telling parents that they will be Judged for any such failure.

Research ideas:

• How much money has Saudi Arabia spent on proselytising? This includes flows to mosques, madrasas, as well as funding scholarships at the Islamic University of Medina (IUM).

• Can you digitise historic Wahhabi syllabi at the Islamic University of Medina via OCR, then use track diffusion in madrasa curricula or local fatwas?

• Track Islamic discourses over the past 50 years using Natural Language Processing on digitised fatwas, sermons, and books (e.g., from Al-Azhar vs. Medina libraries). Measure shifts in keywords like ‘bid’a’ or anti-Sufi rhetoric.

• Explore the demise of Sufism via shrine destructions, surveys on adherence, or conflicts. Proxy with Google Images/satellite data, detecting pixel shifts in destruction, correlating with financial flows or return IUM scholars.

Can We Track the Great Gender Divergence on TV?

Someone clever could explore whether cross-national differences in television content contributes to divergent rates of female employment. This would likely require:

• Building a comprehensive corpus of popular television scripts,

• Undertaking a content analysis with natural language processing,

• Constructing a ‘Progressive Index’, comparing across cultures, and then

• Tracking the lagged rate of female employment.

Drawing on my globally comparative research, I have sketched likely charts. Correctives welcome!

Can we Track Cultural Flux in Autocracies?

Kevin Hart, Pete Davidson and Zarna Garg are all performing next month at Riyadh’s comedy festival. MBS is engineering top-down cultural change by imprisoning dissidents and making secularisation appear more permissible and prestigious. But are Saudis really laughing? Or is he stretching the boundaries of permissibility?

One could use surveys, but I worry about non-responses and social desirability bias. Eagerly, I went to the Saudi Embassy, requesting permission to undertake qualitative research. Alas, they never replied. Not easily defeated, I must change tack… Can we leverage people’s digital footprints to track their revealed preferences?

Research Ideas:

• Examine social media metadata. Harnessing GPS locations, hashtags, computer vision and sentiment analysis, you could track the vibes. On Snapchat, you can search for posts by location.

• Use streaming data (from Netflix and Spotify) and book sales to track what stories and ideas are gaining popularity. Apply Natural Language Processing (NLP) to book reviews and comments.

• Track consumption data - e.g. rising demand for women’s workwear and athleisure.

What’s the impact of religious schooling?

The 1949 Constitution of West Germany made religious education compulsory, exposing state high-school graduates to a 1,000 hours of religious education (quadruple the time they spent learning physics). Between 1972 and 2004, this obligation was progressively weakened, with states permitting students to study non-denominational ethics. What happened next? Aold, Woessman and Zierow find students then became less religious, less likely to pray, or attend religious services.

Indonesia mirrors this same trend, as the growth of religious schooling spawned demand for religious politicians, campaigning to uphold the faith. In Sub-Saharan Africa, European-run missions appear to have increased educational gender equality and female household autonomy. In Kerala, competition between religious and secular groups may have amplified female enrolment.

Belief in Heaven, Karma or nirvana may motivate young people to forgo economic returns in exchange for salvation. I call this “The Paradise Premium”. In my research in Leicester (a city in the UK), every Muslim I interviewed enrolled their children in after-school madrasas. Does this have any impact on female employment, upward mobility, or the broader macro-economy?

How might we test this empirically? This is tricky - disentangling cause and effect. Religious parents are doubtless more inclined to enrol their children in faith schools - so as to preserve their own values over the generations. How do we tackle endogeneity?

Research ideas:

• Peer match students using observable characteristics, then follow long-run outcomes. One might compare Gujarati Muslim and Hindus in Leicester, noting their religious education, time spent on home-work, expressed priorities, and long-run economic outcomes.

• How do after-school religious schools impact the macro-economy?

• Suppose the government relaxes restrictions on faith schools/ clubs, permitting more establishments. How does this affect long-run economic activity, especially of women?

• Caution! When comparing faith and secular schools, recognise that there can be ‘de facto’ faith schools. Suppose an ecological disaster forces a large Buddhist population to relocate as refugees. In host communities, some schools may become predominantly Buddhist, despite being designated as ‘secular’.

Puzzles!

Madura is a glorious island with abundant delights, but poor infrastructure limits people’s capacity to harness these resources. Just as great roads would aid connectivity, I’m hoping that LLMs can help us connect the dots and trace global cultural evolution at scale.

Thank you for being so patient while I’ve been away doing qualitative research across five regions of Indonesia. These big trips are hugely important, as they enable me to understand cultural change on the ground. Truthfully, it was tremendous and there is much to come…

Related Essays

• Beyond Caste: How Migration Enables Love & Economic Prosperity!

• What Caused the Global Islamic Revival? - new paper!!

• Netflix and Change: Tracking Cultural Flux in Autocracies

• Can We Track the Great Gender Divergence on TV?

• The Paradise Premium: How the Afterlife affects Economics

• What do Parents Want?

• Are men and women becoming more politically polarised?

• Is there a rise in hostile sexism?

• Does this hold worldwide?

• Which groups are deviating from the median - singles/ poorest/ ethnic variation?

• Is this due to economic frustrations, online persuasion, or less socialising?

• What are the possible solutions?

You can watch on YouTube, listen on Spotify/ iTunes, or read the transcript below.

Read more

Years later, at a women’s rights conference in Sweden, Azadeh bounded up to me, introducing herself as an avid reader of my research. Instantly, we became pals, as Azadeh shared stories from western Iran, explaining why patriarchy persists her homeland and what enables the diaspora to achieve liberal freedoms. Again and again, we see the cardinal forces driving cultural evolution: power, punishment, and prestige.

Read more

1. Once women are in control of their fertility, they only procreate if they expect rewards - whether that is personal fulfillment or social approval. Thus any analysis must consider what gives people pleasure and status, as well as their outside options.

2. People are more likely to have kids if it’s FUN - relative to alternatives. As entertainment-technology improves, it generates a tantalising menu of ways to spend our lives. Babies must then tussle - in this dynamic trade-off.

3. Economic and housing incentives from governments could encourage more births, but these must outcompete alternatives, which are ever more alluring (especially for women!). Government would need to offer more (and more) to compensate for parents’ massive, irreversible investments, where are the returns are uncertain, yet the adjustment costs are high!

4. Community comes into play. If everyone else has kids, this shifts expectations and creates club goods. Families organise barbecues and trips to the seaside - sharing the care, while revelling in amusements. As a new dad remarked to me,

   “I took Jack to a local church that has a soft play, bumped into a friend and his baby, met an engineer mum and fellow cycling enthusiast, and suddenly we’re all just mates! Everyone had a great time!

   But the odds of that being a reliable occurrence, and there being enough fellow parents who you vibe with, get worse when fewer people have kids. Negative feedback loop!”

5. Films could make parenting seem more desirable - a pathway to love, fulfilment, status and social approval. Visualise this:

   

   That said.. cultural engineering is extremely difficult in the 21st century: in this age of infinite personalisation, people can swipe past anything they dislike. Preaching is strictly limited to the choir.

6. The rise of singles and solitude interact with all the above, as it’s much harder to raise a child alone and doing so might hurt one’s dating prospects. Even if late 30s eventually find love, the reproductive window is much smaller. Cash incentives can’t trump biology.

While there’s no silver bullet, anything that addresses 1-6 would probably help. And although my theory is global, it interacts with local factors. Interventions will be most effective if they target country-specific binding constraints. Italy’s young adults are daunted by precarity, Koreans are highly polarised, while Egyptians keep their gender-segregated distance.

My qualitative research has taught me one thing: culture is malleable! So surely this issue is fixable! Looking forward to learning from brilliant minds innovating in this space. Questions and critique are always welcome!

Curious to hear more? I discussed these conundrums on the BBC’s Today Programme, 54 minutes in.

My Essays on Fertility

• Why is fertility collapsing, globally?

• How does Low Fertility Affect Economic Growth, Worldwide? Jesús Fernández-Villaverde

• Why don’t men want so many kids?

Relationship frictions

• Why are Marriage and Fertility Both Falling?

• Why are Childless Singles on the Rise in East Asia?

• What Prevents & What Drives Gendered Ideological Polarisation?

• Ghosting the Patriarchy: Female Empowerment and the Crisis of Masculinity

• Why are Marriage & Fertility Plummeting in MENA?

Intensive parenting and the educational arms race

• What caused East Asia's rapid fall in fertility? (over the 1960s and 70s)

• The Danger of Single Country Case Studies

• Why has fertility plummeted across East Asia?

• Fertility is a collective action problem. Can it be tackled with taxation?

• Why has Fertility Plummeted? Insights from Hong Kong

Films and prestige

• Lights, Camera, Ovulation!

Religion

• How does Religion shape Education & Fertility in Africa?

• Why is African Teen Fertility So High?