How they built it in granite.

Brihadeeswara is, by mass, almost entirely granite. The plinth, the walls, the vimana, the floor of the courtyard, the corner-shrines, the Nandi's mandapa — all granite, all in courses dressed to within a few millimetres of true, all laid without mortar. The approximate total mass of dressed stone in the temple is 130,000 tonnes, of which the vimana alone accounts for 43,000.

None of this stone is native to Thanjavur. The lower Kaveri plain is alluvium for tens of metres before the bedrock; the granite was brought in. Where it came from, how it was moved, and how the building was assembled remains one of the more interesting engineering questions in pre-modern South Asian construction.

Why granite.

The Cholas inherited a Pallava and Pandyan temple tradition that worked predominantly in granite, with subordinate carving in softer stones (sandstone, limestone) where these were available. Granite is structurally exceptional: high compressive strength, low porosity, excellent weather resistance, and — for the carver — a uniform crystalline grain that allows fine sculptural work without internal cleavage planes.

The Tamil country has substantial granite reserves in the rocky country north and west of the Kaveri delta — the hill country of Sittannavasal, Pudukkottai and the Salem plateau. These quarries were active in the Pallava and Chola periods and remain in use today.

The quarry, sourced.

The principal granite quarry for Brihadeeswara is now generally identified as Sittannavasal, in the rocky country approximately 80 km north-west of Thanjavur. The identification is based on petrographic analysis of the temple stone, which matches the fine-grained grey biotite granite of the Sittannavasal beds, and on the presence of large abandoned quarry faces near the village that show characteristic eleventh-century cutting marks.

Secondary sources of stone — particularly the darker, finer-grained black granite used for the Nandi and for some of the principal sanctum sculptures — are likely Salem- country quarries somewhat further north. The total quantity moved over seven years of construction is in the region of 130,000 tonnes, or roughly 50 tonnes per day if sustained six days a week.

The transport problem.

Moving 50 tonnes of granite a day from a quarry 80 km away, in 1003 CE, is a non-trivial logistical operation. The largest individual blocks in the temple are over five metres long and weigh upwards of forty tonnes. These cannot be carried by oxen, elephants or any cart of the period. The only feasible long-distance transport for blocks of this size is by water.

The most plausible route involves floating barges down the Kollidam (a distributary of the Kaveri) from a quarry-side loading dock, across the delta network to Thanjavur, and through a dedicated canal — possibly the Vadavaru — into a holding basin near the temple. The smaller blocks would have been moved overland on greased timber rollers, drawn by elephants or by human teams.

Building the raft.

With no bedrock to dig to, the Chola engineers built upward from a raft. A broad, shallow assembly of interlocking granite slabs was laid directly on compacted river sand across the entire footprint of the temple. The raft is approximately 0.6 m thick; the slabs are dressed to fit tightly along their edges, with no mortar. The whole assembly distributes the load of the structure above it like an inverted ship's hull.

This is, structurally, an extraordinarily forgiving choice. Sand resists vertical load well but allows the assembly to settle uniformly under monsoon flooding and seismic activity. Where a more rigid masonry foundation would have cracked, the Brihadeeswara raft has simply sunk — by an estimated 60 cm over a thousand years, with differential settlement of less than 4 cm across the entire footprint.

Dressing and dry-jointing.

The granite courses of the temple are dressed to a finish accurate to within about 5 mm. Each block is cut to fit precisely with its neighbours, often with interlocking shoulders cut into the stone — a course of granite locks into the course below it via keyed projections that prevent lateral slip. No mortar is used anywhere in the principal structure. The weight of the upper courses, distributed through the interlocking joints, holds the assembly together.

Dry-jointed granite is extraordinarily stable under compression but brittle under tension. The Brihadeeswara structural strategy is to keep the entire vimana under compressive load: there are no cantilevered elements, no horizontal spans of consequence, and the hollow interior of the upper tiers is filled with rubble that adds dead weight rather than introducing tensile stress.

“The Brihadeeswara structural system represents the maximum scale that can be achieved with dry-jointed granite construction. After it, larger Hindu temples were built only by introducing iron reinforcement or by reducing the central tower in scale and elaborating the surrounding precinct.”— Pierre Pichard, IFP/EFEO, 1995

Lifting the courses.

For the lower courses, lifting was done by simple inclined ramps and timber sledges, with elephant or human haulage. For the higher tiers, where the working height became excessive, an external earthwork ramp was almost certainly used — wrapping around the building as it rose, with the ramp dismantled and rebuilt at each successive stage. The ramp would have been the largest single piece of temporary infrastructure on the project, larger by volume than the temple itself.

The final great lift, of the 80-tonne kalasam at the top of the vimana, has its own documentation: the Sarapallam ramp tradition, the six-kilometre approach, the clay-bound fill from the borrow pit. This last lift is discussed in detail elsewhere on this site.

Why it has lasted.

The combination of granite, dry-jointing and the sand-raft foundation has proven remarkably durable. Pichard's 1995 structural assessment, repeated by the ASI's 2010 photogrammetric survey, finds the building in fundamentally sound condition. The main long-term degradation processes are biological (lichen, micro-organism colonisation of the stone surface), atmospheric (acid-rain etching of the carved surfaces), and thermal (cyclical expansion and contraction in the surface skin of the stone).

None of these have produced structural problems. The dry-jointed granite courses are still locked together by their own weight; the foundation raft has settled uniformly; the kalasam is still in place. The building, for engineering purposes, is essentially unchanged since 1010.

Further reading

• Pierre Pichard, Tanjavur Brhadisvara: An Architectural Study, IFP/EFEO, 1995 — the structural reference.
• R. Champakalakshmi, Trade, Ideology and Urbanization: South India 300 BC – 1300 AD, Oxford, 1996 — for the logistics of stone supply.
• ASI, Brihadeeswara Temple — Conservation Programme 2007 – 12, official report.
• K. R. Srinivasan, Temples of South India, National Book Trust, 1972 — for comparative granite construction across the Dravidian tradition.