The dying art of Madanpura kadhwa weaving, and why it matters

You can hear Madanpura before you see it. The mohalla sits behind the Ganges in the older quarter of Varanasi, a warren of lanes too narrow for a car, and from every second doorway comes the sound that has not changed in four hundred years — the soft, paced thud of the wooden beater, the rasp of the naksha being read, the dry click of a bobbin changing hands. There are three weaving mohallas of consequence in Banaras: Madanpura, Pilikothi, and Lallapura. Madanpura is where the kadhwa weavers still live. Kadhwa is the most labour-intensive brocade technique in the Banarasi tradition, and it is being lost faster than we know how to teach it. This is a reading about why, and about three men — Iqbal, Salim, and Habibullah — who are holding it together with their hands.

What kadhwa is, and why it cannot be hurried

Kadhwa, in the Banarasi vocabulary, is not the same as kadhua, the cutwork technique that is sometimes confused with it. True kadhwa weaving means that each motif on the saree is woven in place, individually, with its own discrete shuttle of zari or coloured silk. The weaver does not run a continuous extra weft across the ground of the fabric the way a cutwork weaver does. He brings the design thread into the cloth only where the motif is to appear, locks it, and ends it. Then he moves to the next motif and begins again. Watch a kadhwa weaver from above for an hour and you will see his hands enter and leave the warp perhaps forty times for a single buti. A saree with a dense kadhwa pallu and a kadhwa-bel border carries between forty and seventy thousand of these locked stitches. They are not embroidered onto the cloth afterwards. They are woven, in real time, into the bones of the fabric. A skilled kadhwa weaver produces between half an inch and two inches of finished cloth in a working day. A full saree takes between four and twelve months. There are no shortcuts. There has never been a shortcut. This is why kadhwa is dying.

Iqbal, who learned it from his father at seven

Mohammad Iqbal is fifty-eight. He learned to read a naksha from his father at the age of seven — not as a child in a chair, but as an apprentice standing at his father's elbow while a Tanchoi pallu was on the loom. He cannot remember a time when he was not a weaver. His own son, who is twenty-six, works for a logistics firm at the Varanasi cargo terminal and earns in a month what Iqbal earns in three. Iqbal does not blame him. He says, with the equanimity that only people who have spent their lives at a pit loom seem to possess, that the work is hard on the back, hard on the eyes, hard on the wage, and that he would not wish all three on his children. His daughter learned the loom briefly as a girl; she is now a tailor in Lucknow. When we ask Iqbal how many kadhwa weavers of his generation are still working in Madanpura, he thinks for a long time and says perhaps three hundred. When he was a young man, he says, the answer was four thousand.

What the weaver actually earns, in plain numbers

A handwoven kadhwa Banarasi that retails in a Mumbai boutique for three lakh rupees pays the weaver, on average, between forty and seventy thousand rupees for four to six months of work. This is a wage of perhaps four hundred rupees a day for a master craftsman, working in poor light, on a pit loom in a room with no fan, executing tens of thousands of locked stitches by hand. The remainder of the retail price goes to the master weaver who owns the loom (about twenty per cent), to the silk and zari supplier (about fifteen per cent), to the wholesaler in Banaras (about ten per cent), to logistics and finishing, and most of all to the retailer (often forty to fifty per cent). This is the arithmetic that explains why the children of Banarasi weavers do not become weavers. The wage cannot compete with a courier-firm salary or a call-centre headset. At Danyah we have rebuilt this arithmetic from the ground up — paying the weaver directly, by name, with a retainer and a per-yard fee on top of the design rate — and the difference per saree, paid into Iqbal's account, is roughly three times what the conventional supply chain offers. It is not yet enough. But it is the right direction.

Why the naksha drawer matters as much as the weaver

A kadhwa saree begins not at the loom but on a sheet of squared paper. The naksha is the design draft — a graph that maps every thread of weft and warp in the pattern, drawn by a specialist called a naksha-band. There are perhaps a dozen senior naksha-band masters left in Banaras. The oldest, Habibullah-saab, is seventy-three and works from a small room above a tea-shop in Pilikothi. His drafts are the ones our atelier returns to most often. He does not use a computer; he uses a pencil, a wooden ruler, and a memory that holds perhaps four hundred Mughal-era motifs in his head. When he goes, much of what he holds will go with him. We are recording his drafts now — photographing each one, naming it, archiving the meaning of each kalga and bel — but the gesture of his hand, the way he balances a jaal pattern on the page, that cannot be recorded. It can only be taught, and he has perhaps two students.

What the modern atelier can and cannot do

The narrative around vanishing crafts often ends with a luxury label arriving like a rescue. The truth is more complicated. A handwoven kadhwa saree is structurally expensive to produce, and no amount of marketing can change the labour mathematics of forty-seven thousand locked stitches by hand. What we can do, and what we are doing, is three things. We can pay the weaver directly and fairly, naming him on the certificate that travels with the saree. We can buy fewer pieces and price them honestly, so that one saree a month from Iqbal's loom pays him what fifteen pieces from a powerloom Banarasi would. And we can carry the weavers' faces into the rooms of the women who wear the sarees — through the passport that accompanies each piece, through the journal you are reading now, through the slow building of an audience that values four months on the loom over four hours at the printer. This is not philanthropy. It is the only commercial model that keeps the craft alive past the next generation, and we believe the math works.

Why this matters to the woman wearing the saree

It would be reasonable to ask why any of this should matter to a buyer who simply wants a beautiful saree for a wedding. The answer is that the difference between a powerloom Banarasi and a kadhwa Banarasi is not aesthetic alone — it is the difference between a textile that lasts twenty years and a textile that lasts a hundred. Mulberry silk woven on a pit loom, with hand-doubled warp and individually locked weft motifs, ages with grace because every thread has a little slack, every motif a little give. The cloth breathes with the wearer. Polyester does not. And the second answer, the one we cannot quite say out loud at the atelier but believe quietly, is that to wear a kadhwa saree is to wear an act of resistance against a kind of cultural amnesia. As long as the woman exists who wants this saree, the loom in Iqbal's house will turn. When she stops wanting it, it will stop.

Madanpura is not gone. It is, on a quiet morning, more alive than the streets around it. But it is held by perhaps three hundred hands, and those hands are ageing. If you have ever wondered whether your saree purchase has any consequence beyond your wardrobe, this is the answer. It does, for one weaver, by name, on one loom, in one mohalla, for as long as that loom turns.