There is a way of casting metal in which the tool is destroyed by the act of using it. You build the shape in wax, cover it in clay, heat it until the wax runs out through a vent, pour molten metal into the space the wax occupied, let it cool, and then break the clay to get the object out. Everything the mould knew about the shape has now been transferred to the metal, and the mould is in pieces on the floor.

This is lost-wax casting, and in India the tradition of it that runs across Chhattisgarh, Bengal, Odisha, Jharkhand and Telangana is called dhokra. The process is not exotic — it is one of the oldest metal techniques in the world, and it is still used industrially, for things like turbine blades. What is specific here is not the chemistry. It is what the technique does to the idea of a product line.

Building a surface out of wire

The Bastar version begins with a core: black soil mixed with paddy husk so that it does not crack as it dries, shaped roughly to the body of the figure and sun-dried. Then wax. Beeswax, softened and drawn out into a thread, is wound around the core coil upon coil, following the contours, building up limbs, bellies, trunks, the flare of a lamp. Small pieces of wax are pressed on for eyes and ornament. Nothing is carved and nothing is smoothed.

That is why the finished metal looks the way it does. The ridges are the record of the winding. Where the maker's hand paused, the coils bunch; where it moved fast, they stretch. A conventional foundry would grind those irregularities away. Here they are the ornament, and they are also the only signature the object carries — not a maker's mark, but the pressure of a particular hand on a particular afternoon.

The wax model is then invested: fine clay from termite mounds, then coarser clay, layer on layer, with a channel left for the metal and a vent for the wax. It is fired. The wax melts and leaves. Molten brass — usually scrap, bought from utensil shops and re-melted — is poured into the gap. The clay is broken off. What comes out is the only copy that will ever exist.

Five steps, one of which makes the sequence a dead end THE SEQUENCE a clay core, dried in the sun black soil and paddy husk; it fixes the volume, not the detail wax wound over it, coil on coil beeswax drawn into thread; every ridge here survives into the metal clay invested around the wax termite-mound clay, then coarser clay; vents left open fired: the wax leaves, the metal enters the mould is now the negative of a thing that no longer exists the shell is broken off and the sequence cannot be run again from here the same five steps in a factory wax pattern injection-moulded from a master, ceramic shell, fired, poured, shell knocked off identical sequence; the difference is that the pattern survives, so the next copy starts from step two Here, step one has to be done again every time — by hand, for each object.
The technique is not what separates this workshop from a foundry. What separates them is the fate of the mould, and therefore what "the same object" can mean.

The name travelled without the makers

In Bastar the families who cast this way are Ghadwa, a name traced to the liquefying and shaping of wax. But the word for the craft, dhokra, belongs to the Dhokra Damar — the itinerant casters of Bengal and Odisha, whose name the technique took as it spread along the routes they walked. So in Bastar the craft is called by the name of a community that is not the one practising it there, and the same is true in reverse across eastern India.

This is a small thing that turns out to be typical. The technique is older and wider than any of the names attached to it. What varies between regions is the metal available, the clay, the ritual use, and who the customers are — not the method. A figure cast near Kondagaon, a bell cast in Bankura and a lamp cast in Adilabad are the same process with different local constraints, and each of the three has its own geographical indication because the Indian system registers places, not techniques.

What a unique object costs

A craft whose every piece is unrepeatable is in an awkward position in a market built on repeatability. A buyer who wants five hundred identical elephants for a hotel lobby cannot be served honestly; the workshop can serve them with five hundred elephants that are all slightly different, and will be told the order is defective. A buyer who wants one good object is the natural customer, and there are fewer of them, and they are further away.

The cost structure follows from the same fact. Each piece carries its own core, its own wax, its own clay, its own firing, and its own hour of winding. There is no economy of scale available anywhere in the process — the five hundredth piece costs what the first one cost, in labour and in wax. What a workshop can scale is the number of hands, which is why the craft survives in families and clusters rather than in firms, and why the price of a piece bears little relation to its weight in metal.

Modern demand has pushed the object in a predictable direction. The figure that was made for a shrine, a household or a village ritual is now made as a decorative object, sized for a shelf and priced for a gift. Design institutions have worked with the cluster on pen stands, stationery and jewellery, which keeps the technique alive and changes the reason it is practised. The bell-metal lamp is still being cast. It is now more likely to be a paperweight.

The honest advantage

There is a version of this story in which the unrepeatability is a tragedy — the craft cannot compete with a moulding machine, so it declines. That is the story most craft writing tells, and it is only half true. The other half is that unrepeatability is the one thing a machine cannot counterfeit. A casting that carries the ridges of a hand, with the flash and the seam left on, cannot be reproduced by any process that requires a reusable tool. The object's weakness in a volume market is exactly its claim in a market for things that are not volume.

What follows is a marketing problem rather than a technical one: how to sell a single object to someone who is not standing in front of it. That is what the label, the cluster, the award and the export catalogue are for — not to protect a technique that is in no danger of being forgotten how to do, but to reach a buyer who can pay for the one thing the technique produces. The wax still goes in, the wax still comes out, and the mould still breaks. Everything after that is distribution.

Sources and checks

  1. Geographical Indications Registry, Chennai Bastar Dhokra — certificate dated 22 April 2008, applicant Chhattisgarh Handicraft Development Board
  2. Development Commissioner (Handicrafts) Craft cluster documentation for Kondagaon and Jagdalpur, Chhattisgarh
  3. Chhattisgarh Handicraft Development Board Bastar Dhokra — applicant records for the 2008 registration, and cluster artisan lists
  4. Bengal and Odisha craft surveys Dokra of Bankura and the Dhokra Damar casters — technique, materials, market
  5. National Museum, New Delhi Tribal and Adivasi metalwork from Bastar in the ethnographic collections
  6. Craft risk assessments Endangerment classifications for lost-wax casting clusters in central and eastern India

Last checked 1 November 2026 Yellowsea Huang Corrections