Galteshwar: The Octagonal Temple on the Mahisagar

The octagonal mandapa is the first thing anyone notices, and it is genuinely unusual in this region.

What the norm is:

• North and western Indian temples of this period almost always use a square plan for the sanctum and for the hall before it
• The square is the standard module of the vastu tradition, laid out on a grid, and the whole system of proportions runs from it
• Halls may be open or closed, pillared or screened, but the underlying plan is square

What Galteshwar does:

• The mandapa is laid out on eight sides
• The pillars are arranged accordingly, and the ceiling is worked to close an octagonal space rather than a square one
• The sanctum retains the conventional plan, so the departure is in the hall

Why this is a hard thing to do. Roofing a square is straightforward, and Indian builders had a well developed vocabulary for it, corbelling inward course by course to close a ceiling.

An octagon changes the geometry of every course. It requires the mason to solve a different problem at each level, and it means the standard templates do not apply.

Nobody builds an octagonal hall by accident, or because it is easier. Somebody at Sarnal wanted it, and had craftsmen capable of delivering it.

Octagonal and stellate plans do appear elsewhere in India, particularly in the Deccan, where Chalukya and later Hoysala builders worked extensively with polygonal and star shaped plans. That is the direction the comparisons point, and it is a long way from Kheda district.

The plan is not the only thing that is out of place. The carving raises the same question.

What has been observed:

• The treatment of the pillars and of the ceiling work has been compared by scholars to the Hoysala idiom of the southern Deccan
• The density and depth of the ornament, and the way surfaces are cut, differ from the Gujarat norm of the same period
• Some elements of the mouldings and figure work read as belonging to a southern vocabulary

How this could have happened. There is no inscription at the site settling it, so what follows is inference rather than record.

• Craftsmen moved. Master masons and their teams travelled long distances for major commissions, and a patron who could pay could bring in a workshop from elsewhere
• Patrons travelled too, and a ruler or merchant who had seen buildings in the Deccan could ask for something like them
• Trade routes connected Gujarat to the Deccan and the coast, and Gujarat's ports made it one of the best connected regions in India
• The idea that regional styles were sealed off from each other is a modern textbook convenience, not a description of how building actually worked

What the temple demonstrates. This is the useful point, and it applies well beyond Sarnal.

Indian architecture is taught through clean regional categories, Nagara in the north, Dravida in the south, with sub styles neatly assigned to dynasties. Those categories are real and useful.

But buildings like Galteshwar, and like Teli ka Mandir at Gwalior with its southern roof, are the evidence that the people doing the work moved between the categories freely. The boundaries are ours. The masons had feet.

Galteshwar belongs to one of the great periods of Indian building, and knowing what else was being built at the time puts it in context.

The Solanki or Chaulukya period in Gujarat, roughly the tenth to thirteenth centuries, produced:

• The Sun Temple at Modhera, with its stepped tank and assembly hall
• Rani ki Vav at Patan, the stepwell of seven levels, now a World Heritage Site
• The Rudra Mahalaya at Siddhpur, once enormous, now fragmentary
• The Dilwara temples at Mount Abu, Jain temples whose marble carving is among the finest anywhere
• A very large number of smaller shrines, tanks and stepwells across Gujarat and Rajasthan

What the period is known for:

• Extraordinary surface carving, worked to a fineness that in the marble examples approaches lacework
• Water architecture of a sophistication unmatched elsewhere in India, in a region where water management was the central practical problem
• Torana arches, ornamental gateways standing free before temples
• A confident and consistent regional idiom, usually called Maru Gurjara

Why so much of it is damaged. Gujarat's temples took heavy losses in the medieval period, from raids and from the campaigns of successive sultanates, and a great deal of what survives is fragmentary. Somnath was destroyed and rebuilt repeatedly. Rudra Mahalaya stands in pieces.

Galteshwar's missing shikhara puts it in that group, though the causes at any specific site are often not recorded.

What survives is enough to establish that Gujarat in these centuries was building at the very highest level available anywhere in the subcontinent, and that a small temple on the Mahisagar was part of that culture rather than an outlier in a provincial backwater.

The siting of the temple follows a rule that operates across India, and it is worth stating because it explains a great many temple locations.

The rule. A sangam, the meeting of two rivers, is a preferred site for a shrine. Galteshwar stands where the Galati joins the Mahisagar.

Why confluences:

• They are treated as points of heightened power, where two distinct things become one
• They are natural sites for bathing, which is the basic act of pilgrimage
• Rites for ancestors are performed at confluences across India
• A confluence is a fixed and unmistakable landmark, which matters for a tradition that maps its geography by places rather than addresses

The great examples run from Prayagraj, where the Ganga and Yamuna meet, through Devprayag and the other prayags of Uttarakhand, to Triveni Sangam at Somnath and countless local confluences with a shrine on the bank.

What the Mahisagar is. The Mahi is one of the peninsular rivers that flow west, like the Narmada and the Tapti, rising in Madhya Pradesh and reaching the Gulf of Khambhat. It has considerable religious standing in Gujarat, and Mahisagar is both a name for the river and the name given to a district.

The practical effect at Galteshwar. The temple sits above the river with ghats below, and the setting is a large part of why it is worth visiting. It is a quiet stretch of water in agricultural country, and on an ordinary weekday you are likely to have both the temple and the bank almost to yourself.

Getting there

• Near Sarnal, Thasra taluka, Kheda district, Gujarat, on the bank of the Mahisagar
• Roughly 85 km from Vadodara and about 65 km from Nadiad, by road
• Anand and Nadiad are the nearest railway stations of size, and Vadodara and Ahmedabad are the airports
• A car is the practical choice. The last stretch is a rural road

When to go

• October to March for comfortable weather
• Mahashivratri, when the temple sees its largest local gathering
• Shravan Mondays
• Morning, for light on the carved interior, which is dim without it
• After the monsoon the river is full and the setting is at its best

What to look at

• The octagonal mandapa, from the centre, looking up at how the ceiling closes
• The pillars and their carving, which is the part that raises the southern question
• The plinth mouldings on the exterior
• The river frontage and the ghats below
• Take your time. This is a small building and the reward is in detail

Practical notes

• The temple is both a protected monument and a working shrine. Remove footwear where indicated and follow the sanctum rules
• Do not touch or lean on the carving, and do not climb the plinth for photographs
• Carry water. Facilities at the site are minimal
• There is no substantial food nearby. Eat at Nadiad, Anand or on the highway

Combining the visit

• Dakor, one of Gujarat's major Krishna shrines, is a short drive away and is the usual pairing
• Vadodara for the Laxmi Vilas Palace and the museum
• Champaner and Pavagadh, a World Heritage Site, are within reach of Vadodara
• Anand and the dairy country of Kheda

A closing thought. Gujarat's famous Solanki monuments are famous for good reason and are properly visited. Galteshwar is the one where you can stand alone under an eight sided ceiling on a river bank, look at carving that came from somewhere it should not have, and think about how much moved around this country long before anyone wrote it down.