The queue begins before dawn. A mother in a faded sari clutches her five-year-old's hand too tightly. The child, who should be dreaming, is instead practicing letter recognition -- coaching centre certificates rolled under his arm like credentials for a job he's too young to understand.
Outside the school gate, hundreds wait. The air smells of sweat and anxiety. This isn't an exam hall. It's a selection camp for first-graders.
This scene could be from Satyajit Ray's Pather Panchali -- not the part where Apu and Durga chase the train through mustard fields in pure wonder, but the other parts.
The parts where hunger is a constant companion, where dreams are luxury goods, where a mother's face carries the weight of a world that offers no mercy.
Ray's genius was showing us poverty without ever reducing his characters to statistics. He gave us the poetry of ordinary struggle. But in today's Bangladesh, the struggle has lost its poetry. It has become a laboratory experiment, and our children are the rats.
The ghosts of policy past
On March 15, 2026, a lawmaker asked in Parliament: "Will we keep suppressing merit all the time?"
By March 17, the education minister announced the lottery system -- introduced in 2019 after years of activism -- would be scrapped.
Lightning speed for a government decision. Warp speed for a decision affecting millions of children.
Let us remember why the lottery existed. In 2018, then-Education Minister Nurul Islam Nahid revealed what everyone knew but no one said aloud: It cost Tk10 lakh to buy a first-grader a seat in a reputed school. Teachers ran coaching businesses. Parents falsified birth certificates to buy their children another year before entering the admission wars. Children as young as four were pushed into exam prep.
The lottery was never perfect. As educationist Prof Manzoor Ahmed noted, it was simply "better among the bad" -- a humane pause in an inhumane race. It silenced the coaching advertisements. It gave a poor child from Kamrangirchar the same statistical chance as a rich child from Gulshan. It was not justice, but it was a ceasefire.
Now the ceasefire is broken. And the question no one in power has answered: What research, what evidence, what stakeholder consultation informed this reversal?
The minister mentioned "extensive discussions" and a "month-long review." But with whom? Which parents? Which teachers? Which child psychologists who understand that a five-year-old's brain cannot be reduced to an admission score?
The world has already solved this
Here is the tragedy: We are conducting experiments that the world has already completed. The results are in. The solutions exist. We simply refuse to read the literature.
- Czechia: The geography of dignity
Travel to any small town in Czechia. A child turns six. Between January 15 and February 15, their legal representative registers them for school. That's it. No coaching centres. No exam pressure. No Tk10 lakh donations.
Czech law is simple: A municipality must provide a school for every child residing in its territory. Catchment areas are clearly defined. If the distance from home exceeds four kilometres, the state provides transport.
There are over 4,200 basic schools serving one million pupils -- an average of 233 children per school, with 27.7% of schools having fewer than 50 students, ensuring even the smallest village has access.
The enrollment process includes a teacher's interview lasting maximum 20 minutes -- a gentle conversation to understand the child's readiness, not a test to reject them. If the school conducts play-based activities, it extends to 60 minutes. The goal is inclusion, not exclusion.
And if parents want to choose a school outside their catchment area? They can. But the local school must accept them. No child is left without a seat.
The result: 90% of Czech children complete basic education in their local schools. The country doesn't need lotteries or entrance tests because it has something better -- a guarantee.
- Los Angeles: Competition that elevates
Now cross the Atlantic to Los Angeles. In 2010, facing declining enrollment and charter school competition, the Los Angeles Unified School District did something radical. Instead of defending traditional catchment boundaries, they redesigned them.
They created "zones of choice" (ZOC) -- clusters of 5-10 high schools within geographic areas. Families rank their preferences. A centralized algorithm assigns places. Oversubscribed schools use lotteries.
The results, tracked by the National Bureau of Economic Research, are remarkable.
By the sixth year, students in ZOC neighbourhoods improved their English language arts scores by 0.16 standard deviations compared to peers outside the zones.
Four-year college enrollment jumped by 25% -- almost entirely driven by students enrolling in California State University campuses. Between-neighbourhood inequality in educational outcomes collapsed.
Why did it work? Because schools now had to compete -- not on their ability to select privileged students, but on their ability to improve all students.
Families ranked schools based on academic quality, giving schools the right competitive incentives. The lowest-performing schools improved the most. Choice, properly designed, didn't cream-skim the advantaged. It lifted everyone.
- Brighton: The fairness formula
Across the Atlantic, the English coastal city of Brighton offers another lesson.
Their hybrid model gives priority to children within catchment areas but uses lotteries for oversubscribed popular schools.
Crucially, they give admissions priority to children eligible for free school meals -- actively using policy to reverse inequality rather than reinforce it.
The common thread across Czechia, Los Angeles, and Brighton is this: None of them pretend that a five-year-old's "merit" can be measured by a test.
They understand that early childhood is a period of uneven development, that IQ is not fixed at age five, that a child's performance reflects parental resources more than innate potential.
So they build systems that guarantee access, manage choice transparently, and invest in school quality rather than student selection.
The blueprint for Bangladesh
We are not Czechia. Our density is higher, our resources thinner, our challenges more complex. But their principles are universal. Here is what an evidence-based path forward could look like:
- Immediate (2026): Suspend the decision to reintroduce tests. Form a national task force not of bureaucrats alone, but of teachers, parents, education economists, and child psychologists.
Their mandate: Study the global models. Publish the findings. Invite public comment. Treat education policy as a matter of national consultation, not ministerial whim.
- Pilot phase (2027-2028): Do not flip a switch on the entire country. Launch pilot projects in 2-3 dense Dhaka neighbourhoods. Map all schools within defined zones.
Allow parents to rank preferences. Use transparent, algorithm-managed lotteries for oversubscribed schools. Mandate minimum quality standards for all schools in the zone, creating healthy competition to improve.
- Long-term vision (2029+): Simultaneously invest in a Czechia-inspired vision. Create a master plan to ensure every neighbourhood has a "good enough" school. This means tackling teacher training, infrastructure, curricula, and -- crucially -- the corruption and inefficiency that have long plagued the sector.
The goal is to eventually make catchment areas work: So no parent feels they must fight for a seat outside their community because the local school is genuinely excellent.
The choice before us
Seventy years ago, Satyajit Ray showed the world that Indian cinema could tell stories of ordinary lives with dignity. Pather Panchali was born of struggle -- shot intermittently over three years, funded by pawned jewelry, created by a crew averaging age 21.
Out of limitation came a work of startling originality that continues to question us: "What stories are we telling today, and to whom? Are we still willing to look at ordinary lives with dignity, or are we content with noise and distraction?"
The mother waiting in that pre-dawn queue deserves dignity. Her five-year-old, clutching coaching certificates he cannot fully read, deserves dignity. The child in Kamrangirchar with the same potential as the child in Gulshan deserves dignity.
Our policymakers face a choice. They can continue treating education as a laboratory -- flipping systems on a whim, responding to a single parliamentary question with a nationwide reversal, making children the subjects of endless experiments. Or they can learn from the world, adopt proven solutions, and build a system worthy of our children's potential.
The train will always pass, the children will always run, and the mothers will always watch in silence. The question is whether we build a world where the train stops for every child -- not just those who can afford the ticket.
Zakir Kibria is a Bangladeshi writer, policy analyst and entrepreneur based in Kathmandu, Nepal. His email address is zk@krishikaaj.com.