Ramesvara Group of Temples at Boudh, Part 3


BY: SUN STAFF - 12.10 2020

Matengeswar Temple, Boudh

The last in a study of Orissan temple architecture, by Ramesh Meher, Birbhum, West Bengal.

Significance of the Stellate Plan

Frequently questions are raised about the explanation of the stellate plan and triangular placing of these three [Ramesvara] temples. Not only the garbhagriha, but also the arghapatta of the Siva lingas within have this stellate design.

K.C. Panigrahi further emphasizes that the star like shape of the shrines and the saktis of the Siva lingas enshrined in them indicate that both the temples and the deities were made in the form of Mandalas or mystic figures with the help of which the Tantrikas wanted to attain their Siddhis. [ ] K.C. Panigrahi has said to associate the Boudh temple with the Tantrikas is not believable. The Tantrikas do not make a Mandala or a yantra on the principle of intersecting squares.

At Budhi Komna, a small village in the Kalahandi district of Orissa, there is a brick temple locally known as the Patalesvara, which the department of Orissa State Archaeology has renovated by way of restoration, having stellate ground plan, like the temple at Boudh. A temple in the Bolangir district also has the same ground plan like at Boudh and Budhi Komna. According to the view of K.C. Panigrahi and R.D. Banerjee, the stellate plan of the shrines and the saktis of the lingas and the triangular placing of the temples are due to the Tantric significance, but then why there is only one stellate temple each at Budhi Komna and Kansil in the place of three identical temples, out to be placed in a triangular formation? So it may be believed that the arrangement of the Boudh temples in a triangular formation may have any implication other than Tantric.

Further the Tantric nature of the shrines might have been indicated by the execution of erotic figures of bigger dimension and prominence, but why are there such small erotic figures at Boudh, as observed by Donaldson? On the other hand, we cannot imagine the erotic figures as indicatives of the Tantric significance of the temples. There are so many temples at Bhubaneswar, Puri and other parts of Orissa, full of bigger erotic figures, but certainly devoid of any Tantric significance.

On the other hand, it may be of significance that all the known examples of temples built on a plan of two intersecting squares have been discovered in Western Orissa. Formally this part of Orissa was included within Kosala, which also comprised Durg, Raipur, Bilaspur and Raigarh districts of M.P. A number of stellate temples have been noticed in that region of M.P, with the possible exception of one. The plans of these temples do not reproduce exactly what has been found in Orissa. In M.P., the central element on each face of the temple is rectangular instead of being triangular. These differences apart, the principles guiding the layout of the temples of Orissa and M.P. are fundamentally similar, as such they seem to constitute a group by themselves and represent a Kosalan version of the Rekha temple sytle. S.K. Saraswati refers to a list of ten kinds of circular temples enumerated by the samaranjgana satradhara.

The plans of a few of the temple types are said to have been reached by rotating the square ksetra all around or in different direction. S.K. Saraswati finds in this description an obvious allusion to the stellate plan of the temples at Boudh and other places. Krishna Deva, who also considers the samaranjgana satradhara description of some temples of the Circular class as referring to the stellate layout of the plan, thinks that the way in which the text has treated and extolled in these temples makes it clear that the circular temples formed the Metropolitan Malava type.

He further believes that the original nucleus of the Amaresvara temple at Onkar Mandhata in the East Nimar District (M.P.) being assignable to the second half of the 10th Century, is the earliest of such temples. In this temple he recognizes the beginning of the bhumija mode of the Nagara temple style. It is evident that both S.K. Saraswati and Krishna Deva were unaware of the existence of a plan characterized by oblique projection between the rahas in a group of Kosalan temples, of a date earlier than the stellate temples at Boudh or Malwa. Therefore it may be understood that this plan, which became a dominant feature of the bhumija class of temples originated in Kosala, and not in Malwa.

From the above discussion it may be seen that the star shape of these temples at Boudh is not to meet the ritualistic need of Tantrikas, but to confirm to a standardized plan evolved in Kosala. It may also be due to a desire to introduce a novelty that the arghyas inside these temples were made star like.


Comparative Study and Date of these Temples

All monuments of temple architecture in Orissa and Central India represent a regional demonstration of the Nagara temple style and have certain common features, being derived from the same model. All temples of Orissa have curvilinear spires and square plans with projected angles of sikhara type or of rekha order, and it ultimately became the dominant form of the temple architecture in Orissa. Now, the earliest temple represents the nature products of that type. But the Orissan temple architecture, by reason of its own distinct individualities and long history of evolution, soon came to acquire for itself a separate nomenclature, i.e. the Kalinga style.

Prof. R.D. Banerjee has drawn our attention to an inscriptions of the pre-Muslim period in the temple of Amriteswara at Holal in the Ballary district, in which mention has been completed of four classes of temple: Nagara, Kalinga, Dravida and Vesara. Prof. Banerjee's observation has further been supported by another scholar, Mr.D.P. Ghosh, who has exposed certain well-marked peculiarities distinguishing the Orissan group of temples from the sikhara temples of North India, Central Provinces, Rajputna, and Gujrat.

Regarding the Ramesvara group of temples at Boudh, about which we have discussed their architectural features, decorative motifs as well as the iconography, certainly played a significant role in the long evolution of Kalinga temples. Each of these temples have the common features of indigenous substyles of temple architecture of Central India and Orissa, and pave the way for matured Kalinga style, which is marked perfectly in the Lingaraj temple of Bhubaneswar.

There are no paleography or epigraphic sources available for the determination of the dates of these temples of Boudh. However, these temples can be co-related on the basis of their architectural features, their decorative motifs, sculptures and the iconography of their images to one or other of the monuments of which the chronology is known, as analytical study of the dated and datable temples and cumulative results. When applied to study these undated temples as cognates of one or other of the date and datable temples, we may not be able to find out the exact date of their construction, but we can place them to particular period as cognates of the particular temple of which the date is known. Such a chronology, though approximate, is borne out by the logic of the evolutionary process experienced by the architectural movement through different period of Orissan history.

The stellate plans of the three Siva temples of Boudh represent a pure Kosalan version of the rekha temple style. Among such stellate temple of South Kosala, it has been observed that temple at Kharad and Pallari are the earliest one. Buddhi Komna, Kansil and Dhobini temples are a step further to Kharad and Pallari, whereas Boudh has the most modified, developed and latest temples among them. It is more fully adapted to the Orissan architectural tradition and exhibits a further elaboration of the decorative programme evolved at Gandharadi. On the other hand, a comparative style of its features with the dated temples of Eastern Orissa can clearly point out the time of its construction. While the temples of Boudh, which have a stellate plan having seven Konarkas on the body, can not be easily compared with the Kalinga temples of Bhubaneswar in architectural elevation, they possess a super affinity towards the decorative motifs, sculptural programme and iconographic peculiarities of other Orissan temples.

The number of pabhaga mouldings used to change in long evolution of Orissan temples. Boudh is a step forward to Gandharadi, having five horizontal mouldings at pabhaga. A new moulding is added here in between patta and vasanta, i.e. kani. As a result of which there is now khura, kumbha, patta, kani and vasanta or pabhaga at Boudh. But kani has not yet got its point shape, rather the edge is flat.

A champaka leaf from the Kumbha links up with a chaitya on the muhanti of its khura as at Gandharadi. The top three moulding are linked together by vertical bars decorated with diminutive standing figures separated by this band of scroll work. This arrangement of five mouldings became prevalent at Bhubaneswar from the temple of 10th Century A.D.

Among those temples most popular are Gauri and Mukteswar of 10th century A.D., Brahmeswar and Rajarani of 11th Century A.D., Lingaraj and other temples of 12th and 13th Century A.D. Boudh provides a close similarity though not absolute with the pabhaga mouldings of Mukteswar. On the Gouri temple, the Kanika has four mouldings, but the anartha has five mouldings where a talagarbhika is inserted, unlike Boudh. There are no vertical bars or champaka leaf and it is less ornamentally decorated than Boudh.

While this arrangement of four mouldings on the kanika and five on the twin piers of the anartha is repeated on the Muktesvara, like Boudh there is no talagarbhika inserted beneath the anartha niche, merely a shallow indentation. The moulding on the anartha is more ornamentally decorated than on Gauri, with the champaka leaf added to the kumbha as is possessed by Boudh.

Though the temples of 11th and 12th Century possess the five mouldings of pabhaga uniformly on kanika and anartha, tala garbhika are eliminated from the anartha, but by this time kanika has already achieved its perfect pointed shape. There is no vertical bar linking these upper three mouldings as at Boudh but there is a very small figure motif, generally erotic carved at the base of the coming pidha in the centre of patta and vasanta. The kirita design on the khura is now transformed into a small vajramundi with the niche, housing various figures, motifs and the crowning vajramastaka linked to the champaka leaf above.

Therefore the decorative plan for the pabhaga achieves its nature form on the temples dated from 11th Century A.D. So, considering pabhaga is the most valuable source to date an undated temple, Boudh should be placed in 10th Century as a cognate member of Mukteswar and its group. By reason of the unique stellate plan of Boudh, it cannot be compared with other Orissan temple in respect of paga division on the bada. However, the vertical alignment of the bada can be discussed with three pagas on each side of the points in the star shaped plan. We may discuss the other decorative and sculptural motifs as possessed by jangha at Boudh and can compare them with other dated Orissan temples. [ ]

The kanika pilaster of the Panchayatana temple at Ganeswar (dedicated to Visnu) is decorated in the exact manner as at Boudh, but the standing female figure in the shaft is encased within a niche. These attaint dwarfs are also soon on the upper raha niche at Mukteswara. Above all the decoration of the bada at Boudh combined the characters that of Mukteswar, Gauri and Ganeswarpur temples, dated to 10th Century A.D. Vidya Dehjia rightly placed them in one group belonging to the culmination phase of Orissan temples. The notable feature of these temple is that with the transformation of the paga from a vajramundii to khakhara mundi is thus complete, now the jagratas with kalasa in between crowns the khakhara mundi in the place of a vajramastaka.

The bada at Boudh was full of images of Gods and Goddesses. Unfortunately some of the projecting blocks with the figures have been knocked of from the pilaster and carried away. There are several detached images most important to other eastern Orissan temples, such as Nataraja and Ganesa. The images of Nataraja of Siva as the Lord of dancers is a frequently occurring motif in the temple of Bhubaneswar. Most important among them are Vaital, Sisiresvar, Muktesvara, Rajarani and Papanasini. Numerous images of Ganesa are also seen in Parasurameswar, Vaital, Sisireswar, Mohini and Bharatesvar, etc., belonging to different periods of history of Orissan temples.

The gandi at Boudh is effectively demarcated from the bada by the baranda and the silhouette, though containing only five bhumis bends in sharply near the top rather than curving gradually as an early temple. The decoration continues the vertical alignment of the bada with three pagas on each side of the points in the star shaped plan. The kanika or tip of each point has four barandis in each the five bhumis. The barandis are decorated primarily with chaitya or floral motifs, as in earlier temples. The Kosalesvara temple at Baidyanath, Ganeswarpur and Mukteswara have the similar pattern of kanika division at the gandi.

With the coming of the 11th Century the kanika paga became divided into five bhumis having five bhumibarandikas in the place of four, as in earlier temples. The middle paga consists of superimposed moulding continuing up the height of the gandi decorated with this scroll motif, exquisitely carved so as to produce a lace like incrustation, which accentuates the vertical thrust of the paga by obliterating the horizontal division of the barandis. The anartha paga of Ganeswarpur and Mukteswar is decorated in a similar fashion as on the above side middle paga at Boudh.

So far as the inside paga at Boudh is concerned, which is divided into uneven barandis by a pheni with padma pista in the place of ribbed amla, it is more influenced by Chhatisgarh as at Khanod and Pallani than by the Orissan tradition. On the front side, just above the porch is a sukanasa, which supports a beautifully executed large bho. The construction of the sukanasa is dictated here by the necessity of the plan, where the gap between the gandi of the shrine and the porch is to be filled. However, as in Boudh, Mukteswar also has the bho motifs, whereas there is single bho motif resting in the front of the shine at Boudh. There are four bho motifs on each of the four sides of the gandi at Mukteswar. Due to the angler projection at the middle of other three sides at Boudh, it is also impossible to have bho motifs on every side as at Mukteswar.

The doorframe inside the porch has three hands of scrollwork, which continues across the lintel above. Donaldson observed that the number of jambs on either side of the door is generally standardized at three initial experimental phase, and from the 10th Century the scroll motifs are likewise standardized. Thus the door frame at Boudh is based purely on the tradition of Orissan temples of 10th Century A.D.

Gajalaxmi rests on the dvaralalata bimba panel and rectangular navagraha slab above it, a group as at Kotitirthesvara temples which belongs to the Muktesvara group and at Mukteswara itself. Laxmi is being bathed by two elephants, seated in lalitasana. This is typical in post-9th Century temples at Bhubaneswar, whereas Laxmi is found seated in rigid padmasana in earlier temples. On the earliest temples there are only eight grahas represented, Ketu being absent, and it is not until the 10th Century that the number is increased to nine, possibly due to the popularity of the Astottari system of astrological calculation in the early period, in which Ketu is not included.

It is quite likely that the Vimsottari system of calculation prescribed by Varahamihira was introduced into Orissa by the Somavamsis. This system incorporates Ketu among the Grahas. The earliest Astagraha slab was attached to the Southern raha niche of the Laksmanesvara temple, now housed in Orissa State Museum, whereas the earliest navagraha slab appeared on the temples of Budhikomna and then at Boudh.

At Budhikomna, except Rahu, all eight-seated graha are depicted ardhaparyanka, whereas they are in bhumisparasa mudra at Boudh, as at Muktesvara and Kotitirthesvar of Bhubaneswar belonging to 10th Century A.D. Rahu is reduced to a head only as at Muktesvara. Ketu is not represented as a serpentine from the waist down, but appears in bhumi sparsamudra. This shows that the temples of Boudh were a later construction than Budhikomna and belong to the age of Kottitirthesvara and Muktesvara. Muktesvara is an exact duplicate of Boudh so far as the trisakha door jam, Gajalaxmi and Navagraha panel are concerned.

In general, the door frame at Boudh is relatively flat, the jambs being nearly flush with one another rather than progressively stepped, as in the architectural traditions, and there are no large figures flanking the door on either side, except two Saivite dvarapalas housed in arched niches at the base of the jambs. Originally in earlier temples these dvarapala used to be housed in small niches, but later on they became arched and eventually transformed first into vajramundi, and then into a Pidhamundi in the 11th and 12th centuries.

Above all, Boudh and Muktesvara possess in common the same subdivisions at the base, such as khura, kumbha, patta, kani and vasanta, with a leaf design in the kumbha linked up with a chaitya arch above, the same form of rounded corners and the same kind of sunken panel marking the transition between the bada and sikhara. We also find in them the same form of the nine grahas, not eight as in earlier temples, in which Rahu has been represented by a head, but not by a half-bust figure, the same types of doorkeepers, alasakanya, kirtimukha, chaitya archs, the dwarf with uplifted hands as if supporting the sikhara, and the same form of pouncing lion-riders.

The decorations consist of a peculiar type of interlacing scrolls at the sikhara. The pitchers carved in alto relief in its vimana, a large chaitya arch flanked by the two Saivite image holding a chain with a belt at the end dropped into a lotus medallion is to be found in those two places. The miniature images of lakulisa in various mudras too occur at Boudh and Muktesvara.

Thus in the overall architectural and decorative programmes of the shrine, porch and door frame, Boudh, though sharing various motifs with Kosalan temple style, is uniquely Orissan and has a fair similarity with Muktesvar, Panchayatana, Vishnu temple of Gonesvarapura, Gauri and Kottitirthesvara temples of the 10th Century. A. D. So, it deserves to be regarded as a cognate member of the Muktesvar group and belongs to this epoch also.


It is known from the above discussion that the architectural features of the temple are important, like other notable temples of Orissa. The triple Siva temples of Boudh possess striking similarities, particularly with Muktesvar and other temples of Bhubaneswar belonging to the 10th Century A.D., which were constructed under the Somavamsis' patronage.