By late February, the lower slopes of the Matamuhuri river basin in Alikadam take on a particular signature. Thin grey columns of smoke rise from the conical mud-plastered kilns that stand at the edge of almost every homestead in Morong Para, College Para, and Palong Para. Inside each kiln, bamboo racks hold tier upon tier of harvested tobacco leaves. Outside, stacks of split timber wait to feed seventy-two-hour fires. Before dawn, men and women move between kilns and sorting yards. After dark, children tend the heat.
This is not subsistence farming. It is an industrial commodity system assembled inside the Chittagong Hill Tracts, and the smoke that hangs over Alikadam during the curing weeks is its most visible signature.
Alikadam upazila in Bandarban district sits at the southern end of the CHT, where the Matamuhuri river cuts through the lower hills before turning towards the sea. It is home to Mro indigenous communities whose agricultural life has been organized, for generations, around jhum rotational shifting cultivation that moves across the hillslopes in concert with forest cycles and soil recovery.
Since the mid-1980s, something else has moved in alongside it: Commercial tobacco, contracted and procured by British American Tobacco Bangladesh and affiliated companies, and now covering an estimated 85-90% of cultivable land at peak season.
I spent the late months of 2025 doing ethnographic fieldwork across three villages in Alikadam. What I found was not a story of farmers freely choosing a profitable crop. It was a story of a frontier being assembled through contracts, through debt, through the invisible labour of women and children, and through forests burned, one kilogram of cured leaf at a time.
The contract that cannot be left
In College Para, only two of 62 households hold direct company registration cards. The remaining 60 participate in the tobacco economy as wage labourers, informal subcontractors, or dadon borrowers, advance-credit takers whose seasonal income arrives pre-mortgaged to the company-linked trader who extended it.
Mujibur Rahman, a Bengali settler farmer in College Para, has been growing tobacco for six years. He described the economics with a kind of exhausted clarity. The advance he receives in October, such as seeds, fertilizers, pesticides, sometimes cash, is logged against his account before a single leaf has been harvested.
By the time the curing season ends in June and he appears at the company collection point, a substantial portion of his gross payment has already been notionally spent. What remains, after the mahajan intermediary takes his margin and the debt is partially settled, is rarely enough to cover the next season without another advance.
"Every year we say this is the last tobacco season. But then we have the debt, and there is nothing else that pays. So we begin again."
This is not a personal failure of financial planning. It is the design of the system.
Advance payments in the contract farming architecture function not as credit in any neutral sense but as a binding mechanism; they anchor the household to the next season before the current one has concluded.
The grading system performs the same function at the other end: Price differentials between leaf grades are wide, the criteria are opaque, and the determination happens at the company collection point, under the company's gaze, without independent appeal.
What the contract formally offers is a guaranteed buyer, a predictable income stream, and extension support. What it does not advertise is the architecture of asymmetric risk that underpins it. The company bears none of the harvest risk, none of the input cost risk, and none of the health risk. All of these are transferred downward, onto the household, through the contract itself.
Women sort every leaf. No one records their name
Tobacco companies in Alikadam register contracts exclusively in the names of male household heads. Company representatives interact almost entirely with male farmers. Income from leaf sales is controlled by male household heads. This is the formal structure of the tobacco economy.
The actual structure is different. Ching Mro, a thirty-eight-year-old Mro woman from Morong Para, has participated in tobacco cultivation for six consecutive seasons. Her name appears on no contract and no payment record.
During the curing season, she and her fourteen-year-old daughter alternate night shifts at the kiln, monitoring heat levels and rotating the bamboo leaf racks. She estimated working between 14 and 18 hours a day during peak periods, with persistent headaches and dizziness she attributed to smoke inhalation.
"We work from before dawn until after midnight during curing season. My husband signs the paper with the company, but it is my hands that sort every leaf."
Her experience is not exceptional. Across all three villages, women and adolescent girls maintained curing vigils through the night, performed transplanting alongside men in the early season, and absorbed the leaf-sorting work as male household heads managed company interactions. This labour is essential to the production of commercially viable tobacco. It is entirely absent from the corporate ledger.
Women hired as daily labourers during peak harvest earn between Tk400-500 per day, consistently lower than the Tk700-800 received by men performing equivalent physical tasks. They are disproportionately assigned to tasks involving direct contact with wet tobacco leaves and agrochemicals without gloves, without masks, without any occupational health provision.
The forest pays for the curing
Each taandur kiln requires a sustained fire for up to 72 hours to cure a batch of tobacco leaves. With thousands of kilns operating across Alikadam, Lama, and Naikhyongchhari at peak season, the fuelwood demand is enormous. There is no commercial supply chain for this fuel. It comes from the surrounding hills.
Field observations in all three villages confirmed that fuelwood collection for tobacco curing has become a primary driver of forest extraction in the upazila. Seasonal labourers carry wood on bamboo poles and motorized three-wheelers along tracks descending from the forest reserve. The forest that has historically provided Mro communities with food, medicine, fibre, and ecological regulation is being converted, load by load, into curing fuel, and the cost of this conversion does not appear anywhere in the company's annual sustainability report.
The Matamuhuri river receives it differently. Chemical fertilizers and pesticides applied intensively across tobacco fields are entering the river system through surface runoff and irrigation drainage.
Farmers in all three villages reported declining catches from the river over the past decade. The aquatic ecology that has sustained Mro food systems is under chemical pressure that accumulates, season by season, below the threshold of any single measurable event.
Inclusion as dispossession
For Mro households in Morong Para and Palong Para, the most consequential aspect of tobacco expansion is not the crop itself but what it has replaced. Jhum cultivation, rotational shifting agriculture across hillslopes, was not merely a farming technique. It was a food system, a land governance system, and a repository of ecological knowledge accumulated over generations.
Knowledge of forest cycles, soil conditions, seed varieties, the relationships between different crops and the hillslope ecology. As tobacco has come to occupy most of the available land and most of the available labour, this knowledge is eroding with a generation that no longer practises what it describes.
Ba Mro, a sixty-three-year-old elder from Palong Para, spoke at length about the seeds that are no longer grown because the land is under tobacco, the forest patches cleared for kiln fuel, and the young men and women who leave the village after the season because tobacco income is insufficient to cover what they owe. His conclusion carried the weight of a long observation:
"Our children do not know the old ways of the field. They know tobacco. And tobacco knows only the company."
The Mro experience of tobacco expansion is structurally distinct from that of Bengali settler farmers, many of whom arrived in Alikadam as part of state-sponsored settlement programmes following the 1970s and who arrived with more extensive credit networks and, in some cases, greater proximity to company registration systems.
This is not a simple story of settlers against indigenous communities. It is a story of class formation cutting through ethnicity, with corporate tobacco as the structuring force producing a small stratum of workable participants, a larger stratum in chronic debt, and a further stratum displaced into wage labour on land they previously farmed.
What the smoke obscures
The official representation of tobacco in the CHT is a story of rural economic opportunity: Cash advances, contract security, and lump-sum payments at season's end that alternative crops have not, it is argued, reliably delivered.
The farmers of Alikadam are not unaware of this argument. Several repeated a version of it themselves, often immediately before or immediately after, describing the debt cycle, the health symptoms they have normalized, and the forests they have watched thin.
“What else is there?” is not consent. It is the sound of a structural choice, a landscape in which alternatives have been systematically rendered less viable, in which the only guaranteed buyer is the tobacco company, in which the advance that arrives in October has already decided the October after that.
The regulatory apparatus that might constrain this system exists on paper. Environmental law covers fuelwood extraction. Pesticide regulations exist. Customary land rights for indigenous communities are nominally protected under the 1997 CHT Peace Accord.
What exists on paper and what operates in the hills of Alikadam are different things. The state is selectively absent in roads and security, absent in the forest, absent at the grading board, absent when a Mro family's informal land arrangement with a registered Bengali farmer converts into something permanent.
In Alikadam, the smoke that rises from the taandur kilns each February is not just burning tobacco. It is forests converted to fuel, unregistered labour converted to leaf grade, and a community's agricultural future converted into another season of advance credit. Calling this development requires a very selective definition of the word.
Nur Nishat Anjum conducted ethnographic fieldwork in Alikadam Upazila, Bandarban, in 2025 as part of an MSS thesis at the Department of Anthropology, University of Dhaka. Names of research participants have been changed.