Sikhara

Temples are classified from different points of view. According to the size, Alpa and according to the number of floors (or ekatala, dvitala, tritala etc; and according to their shape and design, chaturasra (square), (circular), dirga (rectangular), hasti-prshtha (apsidal), vrttayata (elliptical) shat-kona (hexagonal) and ashtasra (octagonal).

The old established nomenclature regarding the main styles of Indian Architecture has been subjected to a critical examination by F.H. Gravely and T.N. Rama-chandran in a Bulletin of the Madras Government Museum (M.S. Vol. Ill, I) and they challenged the use of the terms, Nagara, Vesara and Dravida for reasons stated below:—

According to the Manasara and the three main styles of temple architecture are defined as follows:—

1) Nagara is that in which the vimana was quadrangular throughout;

2) Vesara in which the vimana was crowned by a circular sikhara above the neck; and

3) Dravida in which the vimana was crowned by an octagonal or hexagonal sikhara above the neck; and the Manasara adds an apsidal form in the case of both Vesara and Dravida styles. They regarded that the Dravida style is applicable to monuments in the Dravida country; and they recognised that the earliest existing remains as Pallava.

They added that

“the term Vesara belonged to the Northern style hitherto unfortunately called Nagara by recent writers, in which the vimana was crowned by a circular amalaka”,

and that

“the term Nagara applied to the series of Chalukyan temples, which, owing to a strong superficial resemblance have been wrongly included by most recent authors among the Dravida”.

Any classification of style is intended to define and group the existing monuments into well recognised units based upon common characteristics. But the application of these terms to the Dravidian monuments of Mamallapuram has yielded the following results as found in the ‘Survey of Pallava Architecture in South India’ (Ancient India No. 14, pp. 129-30).

“The Draupati ratha illustrates the type of a single hut (Kutagara), square on plan, with four of the six angas of the vimana, the angas that are lacking being the prastara and griva below the sikhara. This represents the kuta type with a domical roof (sikhara) crowned by a single stupi or finial also square on plan (sama-chaturasra) the four sided sikhara, in conformity with the parts below, representing the Nagara order in its pure form. The bas-relief shrines in the Arjuna’s Penance (Ganga Avatarana) Sculpture (PI. XLVII A), and on the facade of the Ramanuja mandapa (PI. XLVII B) represent the same type. These are more complete alpa-prasadas in that they have the six angas including the prastara and griva; the latter being a clerestory raising the roof proper over the cellar; they are the same class of Sama Chaturasra Kutas with the Nagara sikhara representing the eka-tala (single storeyed) variety. The dvitala (two storied) variety of the Nagara order is represented by the Valaiyan Kuttai ratha, and the Northern Pidari ratha, square throughout with the difference that in the latter, the prastara of the second storey does not carry the hara of miniature shrines as in the storey below, while the former has the hara of salas and karnakutas over both talas.

The Southern Pidari ratha and the Arjuna ratha illustrate examples of Sama-chaturasra (square) vimanas of the dvitala variety with octagonal sikhara of the Dravida order. They constitute the misra (composite) variety of that order, four sided from base to prastara, including the talas with octagonal griva and sikhara. The Dharmaraja ratha (PI. XLIV A) illustrates a tri-tala (three storeyed) example with the same type of sikhara with all the three square talas intended to be functional. A pure variety of Dravida order, though hexagonal on plan from basement to finial is perhaps represented by the bas-relief at the front end of the Nakula-Sahadeva ratha (PI. XLVII D)”.

Enough has been quoted to show that these authors hold that even among the well-accepted Dravidian monuments, there were the Nagara and the Dravidian style in one and the same monument. Dravidian monuments have different kinds of sikharas—square, octagonal, round, wagon-roofed and all of them belong only to one order—the Dravidian.

Further, I think that Gravely himself has receded from his original rigid position in his next revised publication, an outline of Indian Temple Architecture, Bulletin of the Madras Government Museum (Vol. Ill Pt. 2). In the earlier Bulletin (Vol. Ill Pt. 1), it was stated in respect of item 3 of Plate I—the Mallikarjuna temple of the Mahakutesvara group (in the Northern style)—that it was “a Chalukyan Vesara temple, one of the Mahakutesvara group”; but Gravely reproduces as illustration the same monument on page 3 in the later Bulletin and styles it “Temple of the Northern form”. The curvilinear vimana and the Amalaka sikhara are certainly features of North Indian (North Hindu or Indo-Aryan) style. To call it Nagara as has been done by recent writers since the days of Fergusson need not be considered wrong. We are not now in a position to say what monuments the Manasara had in view when these stiff definitions were framed.

An interesting record from Holal is the label cut out on the capital of a finely carved pillar in the Amrite-svara temple. It is called, in the inscription in Kannada characters of about the 14th century, a Sukara pillar each costing 20 gadyanas (See also Ep. Carnatica, Vol. VIII part I—Ins. no. 275).

About the sculptor who made it, the record says that

“Bammoja, the pupil of Padoga of Sige, was a Visvakarman i.e. ‘the architect of the Gods’ in this Kali age, the master of the sixtyfour arts and sciences, the clever builder of the sixty-four varieties of mansions, and the architect who had the skill to raise the four types (styles) of buildings viz. Nagara, Kalinga, Dravida and Vesara”

(Madras Epigraphical Report, Southern Circle—“Nagara-Kalinga-Dravida-Vesara” chaturjati prasada vinilmita sutradhari”).

This local contemporary epigraphical testimony about the prevalence of these regional styles is of greater validity than academic, speculative definitions of silpasastra treatises. Moreover, the Kamikagama states that these styles are differentiated on a regional basis—that the Nagara or Indo-Aryan style was prevalent in the region from the Himalayas to the Vindhyas; the Vesara style was found in the basin of the Krishna; and the Dravidian style should refer to the region from the valley of the Krishna to Comorin. The monuments of the central region have been largely influenced by both the Northern and the Southern styles. Starting from the Kadamba style we pass through the Badami and Kalyani phases of the Chalukya and ultimately culminate in the Kakatiya and Hoysala styles of the 12th and 13th centuries. I agree more or less with a similar conclusion arrived at independently by the editors of “Karnataka Through the Ages” published by the Mysore Government, Bangalore (see the Chalukyan style).

None of these classifications agrees with the design of the existing monuments found today. There is great merit in the cautious approach of V.A. Smith who states:—

“The Dravidian or Southern style of Architecture is sharply distinguished from the Northern by the fact that its tower or spire is straight lined and pyramidal in form, divided into storeys by horizontal bands and surmounted by either a barrel roof or a dome derived directly from the ancient wooden architecture”

(p. 129—V.A. Smith—History of Fine Art in India and Ceylon).

It seems to me that no safe deduction about style could be drawn solely on the shape of the sikhara (see Notes on Sikhara).

The Chalukyan style

“The Chalukyan style may be equated with Vesara school of the Sanskrit texts. In the Vesara, there is a combination of the principles of South Indian (Dravida) and North Indian (Nagara) schools of architecture. As Mr. Cousens points out, the Chalukyan builders while retaining the storeyed arrangement of the Pallava tower (South India), reduced the height of each storey and covered it with such a profusion of ornamental detail that they eventually became greatly overcrowded. The result was that in later examples, these details are not apparent at first glance. At the same time, they borrowed ideas from the northern tower and so manoeuvred the central panel or niches on each storey as to form a more or less continuous vertical band. There is also a third element namely the apsidal character of the Buddhist chaitya as incorporated in the famous Durga temple at Aihole.”

(‘Karnataka Through the Ages’, p. 381, Mysore Government, Bangalore.)

It may be added that there are some centres like Badami, Aihole and Patta-dakkal where monuments of more than one style are found side by side.

Notes on Sikhara

Another writer discusses the Avani Kandarpa-Isvara-griham at Kilaiyur, and dates the Agastyesvaram to A.D. 875 and the Cholisvaram to A.D. 905 and he adds:

“The temple strongly partakes of Irukkuvel architectural elements and has been unduly (?) carried out under the immediate patronage of Irukkuvel chieftains and artisans. It is likely to belong to a date closer to Kodumbalur Muvar Koil which it greatly resembles and copies and thus indicates the extension of Irukkuvel architectural style into the mainland of Cola country. The fact that the sikhara is square unlike the normal later circular Cola sikhara is also a point in favour. The other temple here (Arunachalesvara or Colisvara) has a circular sikhara and is mentioned together with the first temple in the 22nd year of Aditya (c. 893 A.D.). However, the alignment of the parivara shrines does not cover this temple and only this temple has an Amman shrine facing south in its northern side. Further its periphery to the south is not well organised. The east niche shows a seated Subrahmanya stylistically different from that of Agastisvaram. All this shows the Agastisvaram was certainly built first and was followed by Colisvaram at a later stage of Aditya’s reign.”

(Paper submitted by K.V. Soundararajan to the Seminar on Indian Temple Architecture held at Varanasi—17th to 22nd November 1967—American Academy of Banares).

Kilaiyur together with Melappaluvur was the stronghold of the Paluvettarai-yars—not of Irukkuvels who flourished in the region round about Kodumbalur. This rules out the possibility of the Irukkuvels and their architectural influence in this region.

The separate Amman shrine was an addition in the later Chola period.

The three earliest inscriptions of Rajakesarivarman (alias Aditya I) clearly prove the existence of the twin shrines (iru-tali) in or before A.D. 884 (and 893). (See note and statements I and II below.)

The two devakoshta figures of Subrahmanyar, standing and seated, are of two varieties but not of different artistic style, they need not be uniform in all respects in order to assign them to the same period (see Four Chola Temples & Early Chola Art, Part I). How to reconcile his dating of Cholisvaram both to A.D. 905 and 893!

Temples in the Dravida desa have sikharas of different shapes—square, circular, octagonal and apsidal (wagon-roof). Avani Kandarpa Isvara-griham has two contemporary shrines side by side with square and circular sikharas. The Muvarkoyil had three shrines all of the same age and all only with square sikharas; but only two of the three exist now. The Kodumbalur temple is nearly a century later than the Avani Kandarpa Isvaram and there is no question of Kodumbalur influencing the architectural style at Kilaiyur.

Moreover the Jaina Panchakuta Basti at Kambadahalli in the Mysore State (E.C.A. I. Pis. 107-109) which may be said to belong to the Dravidian order has three shrines in a line built at the same time having three different sikharas—square, circular and octagonal.

The Dharmaraja Ratha at Mamallapuram and the temples at Tiruvallam and Konerirajapuram have octagonal sikharas. The temple of Sokkesvara at Kanchi, the temple at Viralur, the Arinjigai Isvaram at Melpadi and also the temple of Kadambavanesvara at Erumbur have circular sikharas.

So it seems unsound to classify monuments regarding their style on the basis solely of the shape of the sikhara.