1 1Certain areas on earth are more sacred than others, some on account of their situation, and others because of their sparkling waters, and others because of the association or habitation of saintly People.
(Mahābhārata Anuśāsana Parva 108: 16–18)
Throughout the ages, many cultures have conceived of geographic space and expressed those conceptions in a variety of ways. One expression of these conceptions has been the establishment of sacred geographies. These sacred geographies were demarcated with the creation of temples and other architectural markers. This chapter provides an overview of the sacred sites in Gujarat spanning from circa third century BCE to the eighth century CE. It aims to bring forth an ever expanding religious landscape in Gujarat and diversity in the architectural forms. Certain physical locations were deemed more important and carried a sacred meaning for religions and communities, and structures mark these spaces as sacred. Temples serve as a bridge between the sacred and the profane and are centres for religious experiences as abodes of the divine. It is through art and architecture that the sacred is represented. The aim will be not to focus on the art historical perspective of sacred architecture of Gujarat, but rather, to trace the expansion of sacred geography, diversity, concentration of sites over time within different areas of Gujarat and linkages within sites.
The criteria for selection of this time frame as the first period is in the nature of the archaeological evidence. Though scanty, the evidence available indicates the beginnings of the demarcation of sacred space by communities in Gujarat through the construction of more or less permanent structures.
A combination of archaeological data and literary evidence indicates the existence of Hindu sacred sites at Dwarka, Somanatha and Padri. All the sites are located along the coast, within the Saurashtra peninsula of Gujarat. Dwarka and Somanatha are mentioned as sacred tirthas in the Mahabharata. Dwarka, famed as the city of Krishna, is mentioned as a sacred spot in the Mahabharata. ‘There is also Dvārāvatī which produces great merit and in which lives the slayer of Madhu’ (III, 88). One should go to Dwarka, as mentioned in the Mahabharata, with regulated diet and vows and one who bathes in the Pindaraka obtains the merit of giving away much gold and that Mahadeva is always present at this tirtha (III, 82). Excavations carried out at Dwarka revealed the existence of a temple (Temple I) dated to first century BCE to second century CE.1 The excavations were carried out in the forecourt of the Dwarkadheesh temple, and Temples II (third–seventh century CE) and Temple III (eighth–twelfth century CE)2 reveal the continued importance of this sacred site over the centuries and till date.
In the case of Somanatha, one has to rely solely on literary evidence as even though excavations reveal an early settlement at the site, there is no evidence for the early existence of a temple at the site. The story associated with Prabhasa (Somanatha) is that Soma married twenty-seven daughters of Daksa, and on account of his predilection for Rohini, the other wives complained to Daksa, who cursed Soma so that he was affected by phthisis. In order to get rid of the disease, Soma, on command of Daksa, proceeded to Sarasvati. He reached the greatest of tirthas, called Prabhasa, belonging to Sarasvati. After bathing here, he regained his effulgence, and hence, the place got its name Prabhasa (IX, 35).
In the Mahabharata, Prabhas Patan has been described as a sacred tirtha located on the coast of the sea (Vana Parva, Ch. 109). It mentions, ‘O Yudhistira, the brahmanas say that on the sea coast is Prabhasa – the tīrtha which is adored by the celestials’ (III, 880). Agni was always present at this tirtha, and those who bathe here with a subdued mind become pure and obtain the fruits of performing the Agnistoma sacrifice (III, 82). It is also mentioned as a tirtha, which is truly sacred, sin-destroying and a favourite place of Indra (III, 103).
A sacred site not mentioned in literary texts and containing evidence of an early shrine is Padri, located in Bhavnagar district of Gujarat, as discussed in the last chapter. Thus, while literary data bring forth the sanctity of Dwarka and Somanatha, archaeology proves the presence of sacred remains at Padri. From among the jars of the Harappan period found during excavations at the site, one of them has a painted figure with a horned head-dress, which has been identified as Pasupata.3 Dated to the early historical period, is a roughly oval structure with a number of post holes along its eastern periphery. The floor had two circular platforms in the northern and southern end with a diameter of 1.5 meters each. Two terracotta plaques of Lajjagauri were recovered from the adjoining area of this floor and it seems the structure had some religious significance.4 This structure has been identified as a temple dedicated to this goddess.5
All the three above-mentioned sites contain evidence of settlement from an early period dating to the Harappan and early Historical periods. Excavations at the site of the Dwarkadheesh temple revealed an occupational layer belonging to fourteenth–thirteenth centuries BCE with lustrous red ware and Harappan ware.6 The site seems to have been re-occupied in 900 BCE and continued up to 500 BCE.7 The third occupational layer corresponds to the second century BCE.8 In the period between 320 BCE and the first century CE, excavations brought to light red slipped ware, black ware, terracotta balls, and stoppers shell bangles, and iron. In Bet Dwarka was found an inscribed potsherd in Mauryan Brahmi, belonging to the same period,9 thus indicating trade contacts with northern India. Temple I was built some time later during this period.10 Similarly, excavations at the site of Prabhas Patan/Somanatha revealed occupational layers dating from an early period. Archaeological excavations at the site of Prabhas Patan reveal five cultural periods, starting from c. 2000–1800 BCE.11 At Somanathapattana, the beginnings of historical settlement date to the fourth century BCE, but religious structures such as temples emerge at a somewhat later period, in the fifth and sixth centuries CE.12
The location of Buddhist sacred sites in the region demonstrates a different trend. The few sites in this period are located more in the hilly tracts, with the concentration being in and around the modern-day city of Junagadh. The Deccan trap in the central part of Saurashtra is exposed extensively and most prominent among these is the Girnar hill, which is 1,117 metres high and is the highest hill in Saurashtra, followed by the Barda hill, which is 637 metres high. It is these two hills and their terrain that provide us with structural data indicating the presence of a Buddhist community in Gujarat.
Located within the present city limits of Junagadh, the Bawa Pyara caves are carved out of a single separate rock and contain viharas and a caityagrha dating to the Mauryan period.13
The caityagrha had four pillars supporting a flat roof, and the cave itself is 6 metres wide and 1.5 metres deep, and has a nearly semi-circular apse on the west. The caityagrha has a verandah in front of it, from which two cells are entered into, one on each side of the principal hall. There are six pillars in the verandah and each has simha brackets to the roof, the façade of which has very crude caitya window ornaments. At each end of the verandah are winged simha carved in low relief on the walls.14
The other Buddhist site with vihara and dating to the Mauryan period is located in the Barda hills at Ghumli.15 The site is located in the interior region of Saurashtra, between the Vartu and Bhadar rivers in the district of Jamnagar. The Barda hills are called Bahrot by the local Warli community, and interestingly, the caves in these hills were also home to the holy fire, the ‘Iranshah Fire’, of the Parsis for 12 years, which they carried with them from Sanjan, when the city was burnt by the Arabs, as has been recorded in the Kiss e Sanjan (authorship is recorded as 969 YZ or 1599 CE, and in its conclusion, the story is signed by a Parsi priest named Bahman Kaikobad or ‘Bahman Kaikobad Hamjiar Sanjana’).
The Buddhist caves are plain and devoid of much sculptural details and their simplicity suggests the main purpose of these was to accommodate the Buddhist monks during the retreat season. These caves lack evidence to suggest their utilisation as sites for ritual purposes or worship by the lay community on a large scale. The size of these caves suggests the presence of a small community of Buddhist monks. In the succeeding period under study, the proliferation of Buddhist structural remains in and around Junagadh speaks of a flourishing period of Buddhism.The monuments indicate an increase in the number of monks and lay followers as well.
In addition to the above-mentioned sites, the role of the royalty in the creation of sacred space in this period is seen in the presence of the Asokan rock edict.16 The site of the Edict is at a distance of 1.6 km from the main city of Junagadh and 3.8 km from the Gir forests. The rounded granite boulder with the edicts is situated at the entrance of the valley, about 0.8 km east of the town, near the edge of what was once the Sudarsana lake.17 The Edict, now situated in a building to ensure its protection, is located between the path connecting the Sudarsana lake and the holy Mount Girnar, which has been described as the hermitage of Rsi Dattatreya in the Brhat Samhita (XIV.11), on one side, and the city of Junagadh on the other. Those who entered the city via this path would have surely stopped to take notice of this boulder of stone, 3.6 metres high and 22.8 metres in circumference, inscribed with the edicts of Asoka.
The site probably would have marked the entry into the settlement sites, as is still the case. The selection of the site of Junagadh may have been due to varied causes.18 In addition to this, the site was connected to the coastal area, giving it access to trading activities, while the fact that Gir hills are the highest hills in Saurashtra and the city of Junagadh is nestled between the hills provide the site some protection. This is probably the reason why the dam at Sudarsana lake was constructed and later maintained by succeeding rulers. Added to this was the presence of raw materials for the fabrication of beads in Kathiawar in the form of agates, carnelian, onyx, and chalcedony besides chank shells.19
The second period under study witnesses not only an increase in the number of sacred sites, but a constantly expanding religious landscape, new forms of sacred sites and a concentration of Buddhist sites in Gujarat. Buddhist sacred space is not limited to the confines of rock-cut architecture, and there is evidence for free-standing architecture, mainly in terms of stupas.
An expanding religious landscape is seen in the existence of Hindu sacred remains at the sites in Karvan, Goraj and Nagara. The site of Karvan in the Vadodara district of Gujarat and situated at a distance of about 30 km from Vadodara is also known as Kayavarohan. The temple here is dedicated to Lakulisa, who is considered as an avatar of Shiva. Excavations at the site revealed a brick structure, the outer portion of which was plastered with lime. Also seen were a number of brick walls running north–south and east–west cutting at right angles. A brick wall around a Shiva linga was noticed at the site, and the size of the bricks was 40.64 cm × 24.3 cm × 8.8. cm and 40.64 cm × 25.4 cm × 7.62 cm, and according to Mehta, this fact clearly indicates that temples of brick existed in this part of the country from the Early Historic period II (0–fifth century CE).20 A number of Shiva lingas were also found at the site, which further attests to the sanctity of the site. The site is associated with Lakulisa, a Shaivite revivalist, reformist and preceptor of the Pasupata doctrine of Shaivism. The Avanti Khanda of the Skanda Purana mentions the installation of linga by Lakulisa and his four disciples at Kayavarohanesvara.21
Excavations at the site of Mahadevapura Goraj in Vaghodia taluka of Vadodara district brought to light remains of a temple. Vestiges of a fairly large brick size temple are deduced from the fact that the platform plinth measured 35 metres east–west and 24 metres north–south, and averages between 1.4 metres and 1.5 metres in height, and 29 courses were exposed in the plinth of the structure.22 A brick wall enclosing the temple on three sides was also excavated. The plinth was made of plain and moulded bricks, of which some were decorated. The pranala seen in the temple was oriented north–south and discharged into a brick built kunda having 36 brick courses. The temple’s construction has been dated to the Kshatrapa period (first–fourth century CE). It was dedicated to Shiva and was subsequently rebuilt as well.23
The site of Nagara is located at the Gulf of Khambat, and has antiquities which date the beginnings of the site to the Early Historic period II (0–fifth century CE).24 These include an uninscribed lead seal, copper plates, terracotta figure of a horse, mother goddess, chank bangles, and a beautiful marble image of a flying Gandharva. Also found at the site are images of Brahma, Parvati and Vishnu, and architectural pieces and pillars are strewn at various places such as Sati ni Deri, Moti Parsiwada, Laxman Hall compound near a well to the east of Dada Tatoo no Mohollo, Asapuri Mata temple, and Patwa Sheri,25 suggesting the earlier existence of some temple at the site.
While in the preceding period, there is an association of the Barda hills with Buddhism, in this period, there is scattered evidence pertaining to Hinduism. Loose sculptures of Shiva and Parvati, and a mandapa wall with three free-standing pillars and two pillars in the wall were noticed, and on the opposite side of the kunda, a matha was discovered at the site of Bhrgukunda, located close to Ghumli. Of the brick structures found at the kunda, one is a platform made of bricks measuring 43.8 cm × 27.7 cm × 8 cm, which is the common size of the Kshatrapa period.26 Hence, the site may have contained a temple in this period, and later came to have a matha as well. The site also has a temple complex to its credit, and continues to be an important religious site in the later period, as the hills contain the temple complex of Sonkamsari in the Maitraka period, and the site of Ghumli subsequently becomes the capital of the Saindhava dynasty, with the beautiful Navalakha temple to its credit.
An increase in the number of Buddhist sites is noticed in this period, and a concentration of sites is identifiable in and around Junagadh. Increasing complexity within Buddhism can be ascertained on the basis of the variety of Buddhist monumental remains. The repertoire of Buddhist remains consist now of caityas, viharas – both in rock, as well as bricks and structural stupas. In addition, at the site of Khambalida, the presence of carvings of Boddhisattvas are indicators of the existence of more complex forms of Buddhism. The majority of Buddhist monuments in Gujarat are viharas, as compared to caityagrhas or stupas. Rock-cut viharas are noticed at Bawa Pyara, Mai Gadechi, Talaja, Sana, Jhinjhurijhar, Khapara Kodia, Dhan, Uparkot, Khambalida, and Kadi Dungar among others; brick viharas are seen at Intwa, Devnimori and Vadnagar. Besides, caityagrhas are noticed at Bawa Pyara, Talaja, Sana, Khambalida and stupas are seen at Boria, Dhoraji (Rajkot)27 and Hadmatiya28 (Junagadh). At the stupa site of Patanvav, an inscription in the Brahmi script dating to the Kshatrapa period, which reads ‘Son of Dama …’, was also noticed.29 The site of Hadmatiya is located along the bank of river Sarasvati and the outer portion of the stupa, dated to the beginning of the Common era, was built of well-burnt bricks and the inner portion was filled with undressed stone.30
The increased importance of Junagadh in the Buddhist circuit can be attributed to the construction of the Boria stupa. The earliest stupa in Gujarat and datable to the second century, it contained some relics. In 1889, James Campbell discovered a brick mound known as Laha Medi in Boria valley, 10 km south east of Junagadh. It contained a stone coffer containing a stone pot in which were found a little copper pot, a silver box, and finally, a little gold box. In the gold box were found an aquamarine bead, a ruby, a sapphire, an emerald, two small pieces of wood, and a relic described as a sliver of bone, stone or clay. The presence of life-sized images in the mound surrounding the central coffer led Campbell to suggest that the relics had been housed in an older structure before being moved to the present location.31 This site may have been a part of the religious circuit encompassing the other two stupa sites of Sopara and Kolhapur.32 The find of a few sculptured marble slabs showed that some stone work existed probably as railing and crowning umbrella.33 The stupa had a brick core, which was about 9 metres in height, a central brick paved courtyard, a flanking verandah, and ranges of cells made of large bricks in mud. Other quadrangular brick structures were noticed – two on one flank and one on the other side of this monastery – and it was an impressive establishment.34 Also located at a distance of 3 km from Junagadh is the Intwa vihara, and this, along with the stupa at Boria is the first piece of architectural evidence pertaining to the use of bricks in the construction of Buddhist sacred sites.
At the site of Khapara Kodia, the chambers are cut in an east–west longitudinal ridge of trap rock, in which the eastern part is somewhat narrow or constricted and serves to emphasise the two basic wings of rock excavations on either sides.35 The two prominent component wings of the caves comprise the more or less compact oblong western wing, provided with a grid pattern of water tanks within and a roughly ‘L’-shaped wing, essentially fashioned to serve as habitational apartments.36 The site has large water tanks and a well-developed system of water storage and utilisation.
An expansion of Buddhist sacred geography of Junagadh is noticed in the increase of the number of sites: Khapara Kodia, Uparkot, Boria stupa, and the brick-built vihara at Intwa, demonstrating a concentration of Buddhist sites here. The concentration of Buddhist sites at the Girnar hill and its surrounding area seems to have been economically viable to be able to support such a large number of monks living in and around modern-day Junagadh. The establishment at Khapara Kodia was by no means small, and when taken together with the Bawa Pyara caves as well as the vihara at Intwa, it is clear that this area had a strong Buddhist presence, not seen elsewhere in Gujarat.
In addition to Junagadh, Buddhist monuments demarcated sacred space at other sites in Gujarat as well. The Khambalida caves, located 21 km from Gondal in Rajkot district, are the only caves belonging to this period that are carved on the exterior with life-size figures of Bodhisattavas, Avalokitesvara and Vajrapani on either side of the entrance to the caitya hall. They are depicted as standing below the Bodhi tree and surrounded by attendants37 and they are clear indicators of the presence of Mahayana Buddhists in the region. The caves at Khambalida, consisting of five groups dating to second–third century CE, contained besides viharas, a caitya hall with a stupa inside it. These caves are the only specimen of decorated caitya in Gujarat.
At a distance of a few kilometres from Dhank is the Jhinjuri Jhar ravine, which contains some caves. One of the caves has a verandah, which leads into a cell through a door and two narrower openings measuring 34 cm in width. In the right hand wall of this chamber is a door, which leads to a second room.38 A cave at Jhinjuri Jhar has two octagonal pillars with square bases and capitals. These pillars are connected to the pilasters by a low screen covered in the fashion of a Buddhist rail of large pattern.39
The site of Talaja has about 30 caves excavated at different levels out of a single hill and the site is located about 3.2 km from the confluence of river Shetrunji and the sea. It has 15–20 water tanks, and the largest of them, known as Ebhal Mandapa, measures 22.8 metres × 20.5 metres × 5.3 meteres and has four octagonal pillars in front, but none inside to support the roof. There is an absence of any partition wall that would divide the structure into an outer verandah and an inner wall. The façade of the cave has remnants of a primitive form of caitya window ornament.40
A large Buddhist monastic complex is seen carved out in the elevated plateau at the site of Sana. The site is located close to the sea with 62 caves on three spurs of an elevated plateau, which slopes down to the perennial stream of river Rupen on the west. The caves at different spurs are approached by rock-cut stairs, and of these, the northern arm is called Munda, which has 18 caves and 47 cells, the eastern arm is called Ravto with 15 caves and 30 cells, and the southern arm with 10 caves and 22 cells is known as Champagalo.41 This site contains three caitya halls as well, but the caves are devoid of any ornamentation.42 Here too, in the Champagalo hill, the largest vihara is known as Ebhal Mandapa, which originally had six pillars in front, and as in the case of Talaja, no pillars inside.43 A number of tanks were noticed near the viharas,44 indicating the site being settled by monks for a longer period and not on a temporary basis, as also the sheer number of caves suggest. Another site in Saurashtra is the cave site at Alech Patan, located in Jam Jodhpur taluka.45
In South Gujarat, rock-cut caves were noticed along Rampura bank of river Majoom in the Kapadvan taluka of Kheda district.46 In Nakhtrana taluka on the side of river Dharud, two caves were noticed, which have been architecturally dated to the third–fourth century CE.47 Cave number I is very simple, with a hall measuring 3.1 metres × 2.8 metres and its ceiling is domical. The second cave is the larger of the two, which is also simple, is east-facing and divided into two parts – namely, the main hall and the adjoining room.48
The other group of caves in South Gujarat is located at Kadia Dungar in Broach district, dating to 300 CE,49 and contains a total of seven caves.50 Kadia Dungar is a small elongated hillock between Jagadia and Netrang and the road between Ankleswar and Netrang is the only road linking it with other parts of Gujarat. An inscription found from cave number I informs us that these caves were carved out during the reign of the Kshatrapa ruler Viradamana.51 Of these, cave number II has a bigger hall, which is 9.4 metres long, and on the outer side of the cave, post holes for wooden railings are visible.52 These caves do not contain any caitya hall or images of Buddha and may have, thus, belonged to the Hinayana Buddhist monks.
Beginnings of brick-built Buddhist structures is noticed in this period. These are at the sites of Devnimori and Vadnagar, and the earliest structures at both the sites are the vihara. In the succeeding period, both sites develop into Buddhist complexes with stupas, monastery and votive stupas as well. In this period, the site of Devnimori had a large vihara to its credit, which was later enlarged twice and repaired as well.53 This site, along with the one at Vadnagar, provide us with the earliest specimens of Buddhist brick-built structures. Close to this vihara, the stupas, one main and four votive were constructed in the next period.
Recent excavation brought to light a monastery and other Buddhist remains at the site of Vadnagar, located in north-eastern part of Gujarat. ‘It appears there was a comprehensive Buddhist sanctuary from at least the third century BCE to the 11th century CE.’54 Among the antiquities found are a red sandstone broken head of Buddha of second century CE, a votive tablet with footprint impressions, six petalled flowers, and a crescent-shaped stone tray depicting the scene of the monkey offering honey to Buddha, sherds of black polished ware with inscription in Brahmi legend, dev-shri-ri-shi, shaka-sya and dha-ma, belonging to second–third centuries CE.55 A sandstone human head sculpted in Gandhara art style and terracotta human head pendants, which could have been used as cult objects, and the many fragments of incense burners56 are indicators of rituals associated with Buddhism at the site.
As for the structural remains, the stratigraphical position of the monastery suggests that it was built around second–third centuries CE.57 The monastery complex at the site was within the fortified area near the present Ghaskol Gate of Vadnagar. Small in size and square in plan, it measured 14.04 x 14.04 metres and near the north-west corner, there is an extended portion on the western side, consisting of two additional chambers. The plan of the monastery is a quadrangle, with an open square in the middle, surrounded by cells, nine in number originally. A small washing area at the south-west corner of the courtyard was found connected to a covered outlet drain, which passed through the west through the gap between the cells. According to Y.S. Rawat, the plan and construction of the cells around the courtyard followed a swastika-like pattern.58
Kantilal F. Sompura aptly describes the austere and simple Buddhist cave architecture encountered in Gujarat, ‘Nearly all the caves, with the exception of the Khambalida caitya cave, are devoid of decoration at the façade.’59 The sheer numbers of caves that have been found within the region of Gujarat are an indication of the presence of a large Buddhist community. In the second period under study, it is noticed that Buddhist sacred architecture came to include space for worship and rituals, as seen in the presence of the caitya halls and stupas.
The absence of images of Buddha at early Buddhist sites has been the centre of much scholarly debate. The inception of the debate on the origins of Buddhist images can be traced back to Alfred Foucher’s essay entitled ‘The Beginning of Buddhist Art’, published in Journal Asitique 1911, wherein the ‘aniconic theory’ was born. The absence of Buddha images in excavations at early Buddhist sites left archaeologists puzzled, and to find an answer to this, Foucher hypothesised both an interpretation of early Buddhist art in India to characterise their absence and a theory of the origin of the Buddha image.60 The depiction of trees, wheels and stupas in Buddhist works of art were interpreted as ‘aniconic’ substitutes for an anthropomorphic Buddha.61 Susan L Huntington, on the other contrary, states, ‘I think that Foucher and others who have followed him were wrong about what they believed they were seeing – or more correctly not seeing – in the arcaheological and art historical record.’62 Huntington disagrees with the aniconists, who contend that the non-figurative subjects in art are intended to substitute for a human figure.63 She argues that the trees, stupas, wheels, pillars, and other foci of narrative reliefs in the early artistic corpus are, in fact, icons. The art shows completely different types of subject matter and that these subjects do not require a Buddha figure nor even something that substitutes for him.64 The images depict sacred objects and places of Buddhism and the devotions accorded to them by lay practitioners and religious pilgrims.65 Trees, reliquary mounds, footprints, and other subjects were, then, not substitutes for a figurative image, but rather, objects of worship in their own right, and these reflect the core practices that lie at the heart of Buddhism.66 In Huntington’s perspective, the artistic rendering are representations of such monuments (stupa – which were created to house the bodily relic) and the relic veneration practices that occurred at them and not, as others have proposed – depictions of the death scene of Buddha. ‘The second type of relics, namely relic by contact, – anything that the Budha had ever touched or used … have been honoured and revered over the millennia by Buddhist devotees.’67 Such objects in art are, then, not viewed as substitutes for the Buddha images, but are the focus of veneration in their own right.68
For this period under study, there is limited evidence for the direct involvement and support of the royalty at any religious site. This is at the site of Intwa, which is located on a hill about 3 km to the north of Girnar. The vihara at Intwa is known as the Rudrasena vihara on the basis of baked clay sealing with a Brahmi legend recording that it belongs to the bhiksu Sangha of the Maharaja Rudrasena monastery. Excavations here brought to light a rectangular brick flooring measuring 19.7 metres × 57.1 metres.69 This can be considered a major development in Buddhist architecture in this region as, for the first time, an attempt is made to construct a vihara which is not carved out of natural rock.
The site near the Sudarsana lake with the rock edict continues to hold importance in this period as well. The rock was engraved during the reign of both the Kshatrapas, in the time of Rudradamana, as well as by the viceroy during the Gupta period. The inscription of the time of Rudradamana mentions that when, by the clouds pouring with rain, the earth had been converted as it were into one ocean, by the excessively swollen floods of the Suvarnasikata, Palasini and other streams of mount Urjayat, the dam breached. It measured 420 cubits long, just as many broad, and 75 cubits deep. All the water escaped. It also mentions that a dam was built under the rule of the Mauryas by the governor Tushaspha who was a yavana. When the dam broke again and the matter was brought forth to the Mahakshatrapa’s counsellors and executive officers, who, though fully endowed with the qualifications, were averse to a task as they found it futile on account of the enormous extent of the breach. The construction work of the dam was finally carried out by the minister Suvisakha, the son of Kulaipa, a Pahlava, who, for the benefit of the inhabitants of the towns and country, had been appointed by the king in this government to rule the whole of Anarta and Saurashtra.70
Presence of the boulder of stone with inscriptions pertaining to major ruling dynasties indicates the importance of Junagadh, Mount Girnar and its environs for almost a century. Evidence for the constant presence of the royalty and its involvement in the construction of sacred architecture are the construction of the vihara by Rudrasena and erection of a temple dedicated to Vishnu by the viceroy Chakrapalita.
Since the role of royalty is limited, we must then turn to archaeological data to determine the patrons of these sacred structures. Kumaran notes that the Kaolin ware used in Deccan in second century BCE to first century CE were reported from sites such as Rajarajeswara, Surpaneswara, Devnimori, and Dwarka – all invariably associated with religious sites.71 Finds of pottery belonging to various ceramic traditions help establish Gujarat’s trade contacts with various parts of the subcontinent. The glazed ware, with its concentration in the Saurashtra peninsula, shows contacts of this area with regions ruled by the Kushanas. On the contrary, the Vasai ware found in concentration in the Kachchh plains and Banaskantha district of Gujarat, both located along the border with Rajasthan, prove the interaction between these two regions.72 In addition to the above-mentioned ceramics, the find of the NBPW, knobbed ware from Hatab, Shamlaji and Devnimori, porcelain from Karvan, the red polished ware found along with Indo-Greek, and Kshatrapa coins and ceramics,73 suggest that local trade networks were established first, followed by overseas trade, which shows that Early Historic people of Gujarat had trade contacts not only with North, East and South India, but also with the Persian Gulf and Red Sea.74 Evidence shows that by the first century CE, Gujarat textiles were being traded in the Horn of Africa.75 There is material evidence from fifth century CE of patterned textiles of Indian origin, probably Gujarati in origin.76 The above evidence hints at the presence of a strong trading community who may have played a major role in the creation of sacred landscapes in Gujarat.
New elements that transform the landscape and imbue it with sanctity are memorial stones found in Kachchh and Saurashtra. Memorials are landscaped spaces created to memorialize individuals, places and events.77 The primary reason for sacredness was the act of memorial designation, of setting aside a space by a group of people. Sites were sacred because they would have meaning to the community or to particular individuals or groups.78 Ceremonies, pilgrimages and rituals acts of consecration create webs of meaning that continually redefine and intertwine the sacred and the profane.79 These are mostly dated to the Kshatrapa period in Gujarat and are spread over Kachchh, Rajkot and Jamnagar districts. In Kachchh, they are to be found at Dolatpur (236 CE and third–fourth centuries CE),80 at Vandh in Mandvi taluka (of the time of Rudrasimha)81 and at Andhau82 (Kshatrapa period). They are also noticed at Gunda83 and Mulvasar (of the time of Rudrasimha and Rudrasena), in Bhanwar taluka of Jamnagar district and at Gadhada (of the time of Rudrasena I), in Jasdantaluka84 of Rajkot district.
The five inscriptions found at Andhau were on a raised spot; the inscription in the vicinity of Mevasa in Rapar taluka of Kachchh was noticed on a small hillock, and another inscription was seen lying on a mound 0.8 km south west of Khavda.85 The inscription found at Andhau records the erection of a yasti in memory of A … ka, son of Satrumsaka, a sramanera (novice monk) by his son Dhanadeva, and the inscription belongs to the time of Rudradamana I.86 Another inscription from the same site, dating to 89 CE, records the erection of a yasti in the eleventh year in the reign of Samotik’s son.87 The stone inscription at Mulvasar, Okhamandal mentions the erection of a memorial stone. Similarly, the inscription from Vandh, in Prakrit language, records the erection of a memorial stone, as a funeral monument in memory of a woman who is described as the servant of the mother, as belonging to Atimutaka gotra and as an inhabitant of Kasadesa, by a person named Ajamitra.88 The stones found in Kachchh are in memory of individuals, but those found in Saurashtra record an event, mainly the digging of wells, as seen on the ones found at Gunda, Gadhada and Visavadar. The presence of these in Kachchh signifies the creation of a new type of sacred space, associated not with deities, but sites created to commemorate heroic deeds and sites to venerate the dead.
Of these, the Gunda inscription, dating to 181 CE, mentions digging of a well for the welfare of society by Senapati Bapaka’s son, Rudrabhuti Abhira.89 Similarly, the inscription at Gadhada, consisting of seven lines on the walls of a well, mentions the construction of a well.90
Thus, while some wells may have been for agricultural purposes, others may have been religious or social in nature.
These memorial stones are unique to Saurashtra and Kachchh, and are not found in any other region of Gujarat, either within this period under study or the next. Probably these were introduced in this region by the Kshatrapas. The limited number of these stones suggests that these were being erected probably only by those who were influential and not as much by the common folk. Memorial stones are also an indicator of worship of ancestors since these stones would be a constant reminder of an important individual of the clan or family in whose honour the stone was erected.
This period witnessed immense growth and development in sacred architecture and sculpture. Most of the monumental remains of this period are in the form of temples, and for Buddhism, the religious complexes from the sites of Devnimori and Vadnagar attest to the continued importance of this religion in Gujarat. There is a spurt in temple construction and sculptures and it can be said that the foundation of temple architecture was laid and artisans and craftsmen were becoming proficient and experimenting with varied styles of architecture.
While earlier sites of Karvan, Dwarka and Somanatha continue in this period too, there is an increase in the number of Hindu temples. Temple sites come up towards the interiors and coastal areas as well. At the site of Goraj, Vadodara district, a plinth of a brick-built structure was unearthed.91 The temple, even though constructed in the earlier period under study, was used and rebuilt in the Gupta and the Maitraka period. This is seen in the different stages of repair of the plinth and the find of a coin of Kumaragupta and Maitraka sculpture at the site.92 The site of Chachlana is located in the Kalyanpura taluka of Jamnagar district and it revealed remains of the Kshatrapa period. Found at the site were red polished ware, black on red ware and terracotta and stone sculptures. Found at the site is a sculpture of Kubera belonging to the fourth century CE, and this, along with the other architectural fragments from the site, suggests the remains of a temple dating to the fourth century CE.93 The site seems to have been an urban centre and probably an administrative centre and a strategic defence location since it is fortified.94
Some of the sites located in North Gujarat are Shamlaji, Than, Roda Polo, and Lakroda. The site of Shamlaji is located along the river Meshvo in the Sabarkantha district of Gujarat, and within the same district are located the sites of Lakroda. Goraj and Roda sites chosen as sacred sites are located in the hilly and forested areas, such as Shamlaji and Lakroda, unlike earlier sites that were only in close proximity to the coast.
At the site of Lakroda, the shrine and mandapika have been dated to the last quarter of the eighth century CE.95 At Roda, the temple complex contains seven temples,96 of which four are located within a stepwell at the site. This site was never mentioned either by James Tod, J. Burgess, Henri Cousens or Sankalia. The studies on Roda were initiated by P.A. Inamdar [1926], followed by U.P. Shah and M.A. Dhaky in the 1960s. The name ‘Roda’ literally means brickbats, and the site is located between the villages of Raisingpura and Khed Chandarani.97 Close to the well at one end are another three temples, with elaborate sculptural decoration while temple No. 7, a Shiva temple, is located near the curving of the river, has an austere doorway and is devoid of sculptural elaboration. This temple was originally connected to the river with stone steps, which are now ruined. Of the ones located next to the stepwell, two continue to survive, while the base is all that exists of the third temple. The sculptures collected from the site and preserved in the Baroda Museum and Picture Gallery point to an established religious site with temples dedicated to almost all in the Hindu pantheon.98 Temple-building activity at the site probably continued for at least two centuries as the sculptures in the museum and the thirteenth-century stepwell in a nearby village would suggest.99 Other sites in North Gujarat are those of Polo and Shamlaji, and the former comes to contain numerous Hindu as well as Jaina temples in the medieval period. Located deep within the Vijaynagar forest in Sabarkantha district, the city of Polo is believed to have been established in the tenth century by the Parihar kings of Idar, and was then conquered in the fifteenth century by the Rathod Rajputs of Marwar. The name is derived from ‘pol’, the Marwari word for ‘gate’, signifying its status as a gateway between Gujarat and Rajasthan. It was built between Kalaliyo in the east, the highest peak in the area, and Mamrehchi in the west, considered sacred by the local adivasis. At Polo, one notices only doorframes of an earlier period in the Shamlaji and Roda style,100 pointing to a temple at the site in the eighth century CE. Only the shrine walls survive and the blocks that are lying around point to the earlier existence of a mandapa as well. The temple seems to have been dedicated to Vishnu in one of his manifestations and the seat in the back wall of the shrine shows that it was not a Shaiva temple with a linga.101
Another temple site is located on the upper region of Saurashtra at the site of Than, where the temple dates to the first half of the eighth century CE.102 The shrine is located outside the courtyard to the north of the Surya shrine. Even though the Surya temple dates to a later period, an old doorframe fixed into the entrance porch, an eroded image of Revanta and a mutilated image of Surya are indicators of the presence of a shrine belonging to an earlier period.103
The vast majority of temples, labelled as ‘Maitraka Temples’ and dating to between the sixth century CE and the eighth century CE,104 emerge along the coastline of Saurashtra. These temples have been labelled after the then ruling dynasty of the Maitrakas, with their capital at Valabhi, but there is no evidence linking the temples to this royal dynasty. The temples running along the coast of Saurashtra are located between the sites of Dwarka and Somanatha, and the majority of the temples are seen between the Vartu and Bhadar rivers. Even though all the temple sites are located along the coastline, diversity in location is noticeable – some temples were erected at creeks (Miyani), yet others at hill tops (Ghumli) and yet others were directly on the sea coast (Khimesvara). There being no evidence to prove the involvement of royalty in the construction of these temples, one may attribute them to coastal communities involved in maritime activities, trade and fishing.
The earliest surviving stone temple is located at Gop. This temple is found in the interior and is located on a hilly outcrop. It seems to be looking down upon the surrounding area. The temple stands out in its surroundings and is clearly visible from a distance while on the road. This hilly outcrop is today surrounded by agricultural land, which is tilled by brahmanas. The temple site had to its advantage the availability of agricultural land as well as mineral resources. This could probably explain the choice of this site for the construction of a temple.
The temple complex of Boricha is also located away from the coast and is built around what is now called a Yajnakunda, where all temples contain Shiva lingas. They seem to be architecturally Dravidian in style, and of these, only one contains a path for circumambulation. Scattered around the temple area are images of Nandi and also seen is an image of Ganesha. It is no longer under worship and is nestled in a village whose people practice agriculture. Found close to the temple is a well – but a natural one without any steps.
Khimesvara is a temple complex located about 122 metres away from the sea. Earlier, it was probably closer to the sea, but is more distant now because of the silting over the centuries. The site seems to have been religiously and architecturally an important site since it is a temple complex and contains within itself varied architectural styles. Interestingly, it is the only site in Saurashtra where an example of the Valabhicchandaja105 style is to be found, with the main temple being dedicated to Shiva.
Another site that shows continuity of worship is Miyani. This site is located on a creek and temples are to be found on either side. The creek was used as a sheltering harbour for local small craft, particularly fishing vessels. Besides temples belonging to the period under study, also seen are two temples exquisitely done in the Chalukyan style. The older temples are to be found on the slope treaded to reach the top of this hill and are located on either side of the path. The choice of this site was probably because of the view one could get from the top. From one side, the sea can be easily seen up to its horizon, and from the other side, the settlement area is visible. The site must have been important because of its location at the creek, as well as the view it could afford of the surrounding area. This is the only site where six temples, mostly in groups of two, covering different periods, are found. The temples stand testimony to the continued importance of this site over a long period of time.
The site of Adodar is located 3–4 km inland from the shore and seems to have been continuously worshipped. This site too is a temple complex and is currently mainly a Shaivite shrine, even though it was dedicated to Surya earlier.
A temple site that continues to be dedicated to Surya is Pata. Found here is also a newly built shrine, dedicated to Navadurgas. Even though the small shrine is new, the images of the goddesses and that of Ganesha seem to be old. They are very plain and simple and do not have any finesse. The images of the goddesses are an interesting mix, with some being represented in the motherly aspect and others in the warrior goddess aspect.
Beyond the site of Somanatha are another three temple sites – namely Kadvar, Sutrapada and Pasnavada. The temple at Kadvar is of Vaishnavite affiliation where currently an original image of Varaha is under worship, and those at Sutrapada and Pasnavada are dedicated to Surya.
All of them are located at a short distance from the coast, and the temples at Sutrapada, Kadvar and Pasnavada are surrounded by agricultural land and located in the interior of these villages.
The location of the majority of temple sites along the coast prompts one to look for associations of these with trading networks and coastal communities of Gujarat. Gujarat had many ancient small settlements connected by caravan routes. Amreli, Devnimori, Dwarka, Girnar, Karvan, Somanatha, Pindara, Vadodara, Vala, and Vasai were all flourishing as small industrial and administrative townships.106 The importance of the trading community in the construction of sacred architecture and their maintenance is seen in the inscriptions of the time of Toramana and the Mandasor inscription. The copper plates of the time of Toramana record donations of commodities by local and long-distance merchants to a Vaishnava temple of the deity Jayasvamin.107 Ranabir Chakravarti notes that donations to this temple by a mercantile organisation of local and long-distance merchants who came from Kannauj, Ujjayini and Mathura indicate that the district headquarters of Vadrapali ‘can be viewed as a nodal point in the overland supra local trade network’.108 The Mandasor inscription mentions the construction of a temple by a guild of silk weavers who had migrated there from the province of Lata in Gujarat.109
While for Hinduism, we have architectural remains, the database for Buddhism consists mainly of inscriptions, sculptures and archaeological remains. Buddhist sites of Devnimori and Vadnagar develop into huge complexes with viharas and stupas and votive stupas as well. The presence of monastic complexes revealed through archaeology at the sites of Devnimori and Vadnagar, and with the aid of inscriptional data at the site of Valabhi, stand testimony to the flourishing condition of Buddhism in Gujarat. At the time of the visit of Xuanzang, a number of Buddhist schools were flourishing in Gujarat. S. Beal’s Si-yu-ki notes that there were 10 sangharmas with 300 monks in Bharukaccha,110 10 sangharmas with 100 monks at Kachha,111 100 monasteries with 6,000 monks at Valabhi,112 and in Saurashtra, 500 sangharmas with 3,000 monks.113 Image worship was also common among the Sammitiyas of Valabhi.114 This complexity mentioned by Xuanzang is well corroborated by the presence of Buddhist sites spread across Gujarat.
At Vadnagar, the preceding period has Buddhist structural remains in the form of monasteries, but belonging to this period are additional two small votive stupas. The stupas, one square and the other circular in plan, were uncovered near the north-eastern part of the monastery, and have been dated to the fourth–fifth centuries CE.115 The elevation of stupa square in the plan shows that it was erected in a terraced fashion, with square platforms in diminishing dimensions raised over one another with the topmost surmounted by a semi-circular solid dome (anda)., the circular base of which measures 2 metres in diameter. The second stupa, circular in plan, measures 2.24 metres in diameter and only the lower three courses are extant.116
Located at about 20 km from the site of Vadnagar are the Taranga hills, which have Buddhist caves ascribable to this period. The earliest archaeological remains from the site were reported in 1938 and these were the image of Tara and a few Buddhist images, including Avalokitesavara Padmapani found in two shrines, locally known as Dharanmata and Taranmata. In addition, a small panel depicting four Dhyani Buddhas was reported from Jogida hill inside a natural rock shelter.117 These hills have ruins of cave dwellings, two Buddhist shrines and the remains of an ancient dam. There is a natural spring at the right bank of a seasonal stream, which is the only perennial source of water available at present118 and the Taranmata and Dharanmata shrines are located on a high terrace just above this bank. One of the constructions on the right side of the rivulet seems to be an altered base of a Buddhist stupa as two large-sized terracotta images of Buddha have been recovered from its top surface. Based on stylistic similarities with images from Devnimori, these have been dated to the last quarter of the fourth century CE.119 There is a path that connects the Jogida hill site with Dhyani Buddhas and the shrine of Taranmata.120 Evidence of the continuity of Buddhism well into the eighth–ninth centuries CE is from the site of Taranmata, where the marble image of Tara has been dated to eighth–ninth centuries CE.121 From the twefth century CE onwards, the site becomes an important centre of Jaina pilgrimage122and has a Jaina temple complex to its credit, which is located on the opposite bank of the river with the site of Taranmata temple. But the legend of the place continues in the name itself, and the Kumarapala Pratibodha of Somprabhacharya, composed in V.S. 1241, records that King Veni Vatsaraja, a Buddhist and contemporary of Jaina monk Arya Khaputacharya, built a temple to the goddess Tara in the city which came to be known as Tarapur after her and Tarapur is identified with Taranga.123
Excavations at the site of Devnimori brought to light a mahastupa, four votive stupas, a vihara, and what may have been an apsidal temple. A number of images of Buddha in dhyanamudra and other decorative architectural pieces were also recovered during the excavations.124 This site is discussed in detail in the chapter dealing with parallel existence with the Hindu site of Shamlaji.
In the region of Kachchh, the Buddhist caves found at the site of Siyot have been dated to between the eighth and eleventh centuries CE. Most of the caves were simple and small, and the main cave has a door and faces north. One of the chambers, chamber C, has a pradaksinapatha in the rear end, and in chamber A, the entrance door facing west has a niche. Excavations at the site brought to light more than a thousand fragments of clay sealings. Of these, 50 were intact, and on one fragment, the upper remaining portion contains the figure of a spire of a temple,125 further indicating the probable existence of a Buddhist temple as well.
In the coastal east and southern area, after Kadia Dungar in the preceding period, the other Buddhist site is at Kampiya village in the Navasari taluka of South Gujarat. Found at the site was some earthen work and a small votive clay object, and this, along with inscriptional data, points to the existence of a Buddhist community at the site. The copper plate grant of Rashtrakuta Dantivarmana, dating to Saka 789, records the donation of the village of Chokkhakuti to the vihara at Kampilya,126 and mention of this vihara is found in the inscription of King Dhruva, dating to Saka 806.127 Though the site has not been excavated, inscriptional data and remains at the site suggest, ‘the occupation of the site of the monastery’.128
In the earlier two periods, a gradual concentration of Buddhist sites was noticed around Mount Ginar and Junagadh. For the third period under study, the copper plate grants of the Maitraka dynasty, totaling to 98, dating from 499 CE to 740 CE, confirm the presence of a Buddhist community at the Maitraka capital of Valabhi. A study of these reveals that a major concentration of Buddhist viharas existed at Valabhi and its surrounding areas. Inscriptional data reveal the existence of 14 such viharas, all except two were under a viharamandalas. The main viharamandalas were the Dudda viharamandala and the Yaksasura viharamandalas. Under the Dudda viharamandala were the Dudda vihara, viharakuti of Buddhadasa, Yakkamali vihara, Gohaka vihara, Skandabhata vihara, Vimalagupta vihara, and the acharya Sthiramati vihara. The Yaksasura viharamandalas contained the Yaksasura vihara, Purnabhatta vihara and Ajita vihara. In addition to the viharas in the above two viharamandalas, the other viharas in Valabhi were the Bappadiya vihara and the Abhyantarika vihara.
Taking this piece of evidence into account, along with that of the earlier period, it becomes apparent that a concentration of Buddhist sites in both periods is found at the provincial capital (Junagadh), and then, at the ruling capital of the Maitraka dynasty (Valabhi). While the data seem to suggest that these were the creation of the royalty, a careful analysis reveals that the rulers donated to already established viharas. Besides the Dudda vihara, no other vihara was constructed by the ruling dynasty. The common factors at the site of Junagadh and Valabhi are the rich agricultural land in their surroundings, and the fact that they were well-established towns where trading activities took place. Virji mentions that agriculture was the chief occupation, and besides agriculturists, in the country, also a rich class of traders and industrialists existed.129 Dandin’s Dasakumaracarita mentions, ‘there is a city named Valabhi in Saurashtra. In it there is an owner of ships (nāvikapati) named Grhagupta, who can vie with Kubera in riches … A merchant prince named Balabhadra from Madhumati (Mahuva) comes to Valabhi’.130 Both the sites had major rivers flowing nearby, and while Bhadar is still the major river in Junagadh district, the river Ghelo is important in Bhavnagar district. The Buddhist monasteries then seem to have been concentrated at economically viable sites where both trade and agriculture were carried out. The two sites demonstrate a trend wherein these sites were important even prior to them being established as ruling capitals. At Junagadh, the first evidence of prosperity are the Saurashtra janapada coins, which are then followed by the earliest caves and the rock edicts of Asoka. In the case of Valabhi, donation of land to already established monasteries hints at the site being selected by the Buddhists prior to its becoming the ruling capital of the Maitraka dynasty.
Direct reference to the erection of a temple by the royalty in the region, and probably the only one so far, is the inscription pertaining to the Gupta period, inscribed on the same boulder of rock that contains inscriptions of the Mauryan and Kshatrapa period. The inscription pertaining to the Gupta period, besides mentioning the repair of the dam, clearly points to the construction of a temple by the royalty in the vicinity of the dam. This is the only other piece of evidence of the direct involvement of the royalty in the erection of a temple in the region. It mentions that in due course of time, there came the season of clouds, bursting asunder, when much water rained down unceasingly for a long time, by reason of which the lake Sudarsana suddenly burst – making the calculation in the reckoning of the Guptas, in a century of years, increased by 30 and also six more, at night, on the sixth day of (the month) Prausthapada. And these other rivers which take their source from the mountain Raivataka, and also, this Palasini, beautiful with its sandy stretches, all of them the mistresses of the ocean, having dwelt so long in captivity, went again, in due accordance with the scriptures, to their lord the sea. And having noticed the great bewilderment, caused by the excess of rain, Urjayat, desirous of appropriating the wives of the mighty ocean, stretched forth as it were a hand.
Then, Cakrapalita, the governor, having sacrificed to the gods with oblations of clarified butter and with obeisance, and having gratified the twice-born with presents of riches and having paid respect to the citizens with such honours as they deserved, and to such of (his) servants as were worthy of notice, and to (his) friends with presents, in the first fortnight of the month … belonging to the hot season, on the first day, he, having practiced all the above respectful observances for two months, made an immeasurable expenditure of wealth, and, [built an embankment] a hundred cubits in all in length, and sixty and eight in breadth, and seven (?) men’s height in elevation, … of two hundred cubits.
The second part of the inscription mentions that the viceroy also built a temple. It states that Chakrapalita, who is endowed with his own good qualities, (and) whose life is devoted to (the worship of) the feet of (the god) Govinda, … by him, with a great expenditure of wealth and time [there was built a temple] of that famous (god Vishnu) who carries the discus, … enemies, (and) who became (incarnate and) human by the exercise of his own free will. (Thus) by Chakrapalita, who is of a straightforward mind, there has been caused to be built a temple of (the god) Chakrabhrit, in a century of years, together with the thirty-eighth (year) … the time of the Guptas.131
While a number of temples are erected in different parts of Gujarat in this third period under study, none contain any evidence of the involvement of the royalty in their construction. This is the only inscriptional evidence in Gujarat to prove the role of the royalty in the erection of a temple.
The other inscription with evidence for the role of royalty in the creation of sacred space are the copper plates of the time of Toramana from Sanjeli, Gujarat, dating to the end of the fifth–beginning of the sixth century CE. The three inscriptions record grants to a temple built by the queen mother Viradhikya.132 These are the only pieces of direct evidence for the involvement of royalty in the creation of sacred space. But in the larger region of Gujarat, this role can be said to have been limited in nature as no other sacred site, besides the above-mentioned ones, contain any form of evidence of the role played by the royalty.
The chapter demonstrates a constantly expanding sacred landscape in Gujarat, with Buddhist and Hindu remains. While Jainism does also exist in the region, it has been excluded since the data pertaining to Jainism are mainly sculptural, and hence, it is discussed in the succeeding chapters. A combination of data available in terms of architectural remains, inscriptional data and literary tradition displays the diversity and complexity of religious sites in the period under study. Besides an ever-expanding religious landscape, clear concentration of sacred architecture is identifiable as well.
In the case of Buddhism, a clear concentration of sites could be identified in and around the site of Junagadh in the second period, and in the third period under study, the focus shifts to Valabhi as inscriptional evidence informs us of the existence of numerous monasteries at Valabhi and in its vicinity. Junagadh and its environs with Buddhist structural remains seem to have formed a religious circuit and it is probably the pilgrims visiting the sites that were the focus of address in the Asokan edicts engraved on the Girnar rock.
A gradual expansion of Hindu sacred space is noticed as well. A proliferation of temples occupying sites of coastal Saurashtra seem to have formed a religious circuit, with Dwarka and Somanatha at either ends. The temples also bring forth the diversity in locations as well as deities worshipped – Shiva, Surya and Vishnu.
There has been a general association of the construction of temples in the early medieval period with the ruling dynasty and Bhakti. In the opinion of B.D. Chattopadhyaya, regional states emerged from the seventh to tenth centuries, marked by complex changes in religious dimensions of society and it is believed that Bhakti and the worship through Bhakti of God as lord located in a temple was the key ideological strand of the period.133 D. Desai concludes that at numerous feudal centres, temples were constructed in permanent material such as stone for the first time in fifth century CE, inspired by the growing importance of Bhakti and the newly established Smarta Puranic religion, which was associated with the new social set-up.134
A similar viewpoint is expressed by H. Kulke, who also lays stress on the political importance of temples, which were a potential media used by the political elite. ‘It is most likely that through the construction of these temples rājās tried to create a new and centralized ritual structure, focused on the new state temple and its royal cult.’135 A study of the temples located at various sites in Gujarat reiterates the inference drawn by Ray that ‘The assumption by historians that kings established temples and donated to brahmanas to seek legitimization of their rule and that these religious shrines were agents of acculturation is not substantiated by available data.’136 Even though the numerous temples along the coastline of Saurashtra, which have been labelled as Maitraka Temples do not provide any evidence of the involvement or contribution of the ruling Maitraka dynasty in the spate of temple construction activities, neither is there any evidence to substantiate the existence of any ‘imperial temple’ that can be ascribed to this dynasty. The inscriptional data further verify the view to reassess temple construction in light of Bhakti, feudalism and the ruling elite. Of the 98 copper plate grants ascribed to the Maitraka rulers spanning from 499 CE to 740 CE, there is clear absence of the mention of any of the temple sites or even the construction of any temple by the rulers. The notion of the presence of ‘cultic centre’ or the role of temples in ‘cultic’ centres is negated by the archaeological, literary or inscriptional evidence, as none of these even remotely indicate the existence of a temple at the capital of the Maitraka dynasty located at Valabhi. Even in the case of the succeeding ruling dynasty, namely, the Saindhavas, it is noticed that they chose their capital to be located at an already established sacred site. Their capital of Ghumli was located at the foothills of the earlier temple site of Sonkamsari. In the regin of aother major ruling dynasty of the medieval period, the Chalukyas/Solankis, a similar trend in noticed. The pattern of Chalukya patronage also suggests that the connection between deity and dynasty was not clear-cut. Chalukya political consolidation did not rely exclusively on the patronage of Shaiva temples and complexes, but encompassed Jaina establishments as well.137
Thus, without the role of the royalty, the existence and continuity of sacred sites need to be studied and analysed in terms of the varied communities of the region. These sites are created, maintained and patronised by communities and since there is no direct evidence for the involvement of the royalty, it becomes imperative to locate the sacred sites within the context of settlements and identify the resource base that enabled the construction and maintenance of these sacred sites. The settlement pattern, with the aid of archaeological and inscriptional data, shall be analysed, to determine whether trade, agriculture or craft and manufacturing activities were being carried out at a site, or whether a combination of two or more existed at a site. Also, in looking for the support base of the religious sites, it must be taken into consideration that Gujarat was economically viable and the support came, in all probability, from various social groups. The issues of patronage and continuity shall be addressed in the following chapter.
The Buddhist architectural remains demonstrate ever evolving forms of architecture to accommodate the ritual needs. While the earliest structures are simple monastic caves, over time, one notices the presence of caityagrhas, stupas and votive stupas. From small monasteries, the religious remains expand to full-fledged religious complexes replete with varied architectural features, as noticed at Devnimori and Vadnagar. The sheer number and type of sites demonstrate a diversity in the forms of Buddhism prevalent as well as the large following and support it gathered in Gujarat. Sites being located mostly on trade routes, Devnimori, Vadnagar and Siyot connect these Buddhist sites with other parts of the subcontinent through communities that visited and supported these.
This chapter took into consideration only a part of the database comprising architectural remains. The region also has to its credit a vast number of sculptures scattered at different sites. While temple sites were the domain of coastal Saurashtra, sites with sculptural remains were found to be located mostly in northern Gujarat. An analysis of these sculptural remains shall be undertaken in Chapter 3 to further demonstrate diversity in types of sacred sites, existence of various sacred sites not available in the archaeological records and diversity in deities worshipped within different areas of Gujarat. Jainism shall also be discussed in the chapter pertaining to sculptural remains since, as yet, no structure has been clearly identified as Jaina in any of the three periods under study, even though the sculptural remains point to a strong Jaina presence in the region.
1Debala Mitra, ed, Indian Archaeology – A Review 1979–80, New Delhi: Archaeological Survey of India, 1983, p. 24.
2Indian Archaeology – A Review, 1979–80, p. 24.
3Indian Archaeology – A Review, 1990–91, p. 9.
4Indian Archaeology – A Review, 1991–92, pp. 21–2.
5Vasant Shinde, The Earliest Temple of Lajjagauri? The Recent Evidence from Padri in Gujarat, East and West, 44 (2–4), 1994, pp. 481–5.
6Indian Archaeology – A Review, 1979–80, p. 29.
7Indian Archaeology – A Review, 1979–80, p. 29.
8Indian Archaeology – A Review, 1979–80, p. 29.
9Kuldeep K. Bhan, Recent Explorations in Jamnagar District of Saurashtra, Man and Environment, 10, 1986, p. 10.
10Indian Archaeology – A Review, 1979–80, p. 28.
11Indian Archaeology – A Review, 1971–72, pp. 12–13.
12Himanshu Prabha Ray, Sailing to India: Diverse Narratives of Travel in the Western Indian Ocean, Athens Dialogues, 3.15, http://athensdialogues.chs.harvard.edu/cgi-bin/WebObjects/athensdialogues.woa/wa/dist?dis=35 (accessed on 29 August 2015).
13Kantilal F. Sompura, Buddhist Monuments and Sculptures in Gujarat – A Historical Survey, Hoshiarpur: Vishveshvaranand Institute, 1965, p. 32.
14Sompura, Buddhist Monuments and Sculptures in Gujarat, p. 7.
15Annual Report of the Department of Archaeology, Rajkot: Government of Gujarat, 1972–73, p. 4.
16E. Hultzsch edited, Inscriptions of Asoka, Corpus Inscriptionum Indicarum, Vol I. New Delhi: Archaeological Survey of India, 1991 (reprint).
17K.V. SoundaraRajan, Junagadh, Delhi: ASI, 1985, p. 12.
18It would also seem as though the area was agriculturally important, as according to Romila Thapar, one of the major agricultural regions in the Mauryan period was Saurashtra in the west. Romila Thapar, The Mauryas Revisited, Calcutta: K.P. Bagchi & Comp., 1987, p. 13.
19U.P. Shahi, Urbanisation in Gujarat: A Geographical Analysis, Gorakhpur: Institute for Rural Economic Development, 1989, p. 23.
20R.N. Mehta, Archaeology of the Baroda, Broach and Surat Districts up to 1300 AD, Unpublished PhD Thesis, MS University, Vadodara, 1957, p. 204.
21N.P. Joshi, Regional Trends in Some of the Mediaeval Brahmanical Sculptures of Malwa, in M.D. Khare (ed.), Malwa Through the Ages, Bhopal: Directorate of Archaeology & Museums, Government of Madhya Pradesh, 1981, p. 112.
22B.M. Pande and Narayan Vyas, An Early Temple in Gujarat – Excavations at Goraj, Puratattva, 20, 1989–90, p. 108.
23Pande and Narayan Vyas, An Early Temple in Gujarat, p. 108.
24Mehta, Archaeology of the Baroda, p. 204.
25Mehta, Archaeology of the Baroda, pp. 152, 204.
26Kuldeep K. Bhan, Archaeology of Jamnagar District up to 1300 AD, Unpublished PhD Thesis, MS University, Vadodara, 1983, p. 111.
27Annual Report – Gujarat Department of Archaeology, 1975–76, p. 3.
28Annual Report – Gujarat Department of Archaeology, 1971–72, p. 11.
29Annual Report – Gujarat Department of Archaeology, 1975–76, p. 3.
30Annual Report – Gujarat Department of Archaeology, 1971–72, p. 11.
31Himanshu Prabha Ray, The Return of the Buddha – Ancient Symbols for a New Nation, Delhi: Routledge, 2014, pp. 104–5.
32Henry Cousens, Architectural Antiquities of Western India, London: The India Society, 1926, p. 8.
33Cousens, Architectural Antiquities of Western India, p. 7.
34Debala Mitra, Buddhist Monuments, Calcutta: Sahitya Samsad, 1971, p. 141.
35K.V. SoundaraRajan, Junagadh, New Delhi: ASI, 1985, p. 32.
36SoundaraRajan, Junagadh, p. 33.
37Sompura, Buddhist Monuments and Sculptures in Gujarat, p. 7.
38Sompura, Buddhist Monuments and Sculptures in Gujarat, p. 12.
39Sompura, Buddhist Monuments and Sculptures in Gujarat, p. 19.
40James Burgess, The Cave Temples of India, London: H. Allen and Co., 1880, pp. 201–2.
41Le Thi Lien, Buddhist Monuments and Antiquities of Gujarat, Unpublished MA Dissertation, MS University, Vadodara, 1992, p. 53.
42Mitra, Buddhist Monuments, p. 146.
43J. Burgess, Reports of theAntiquities of Kathiawad and Kacchch, Varanasi: Indological Book House, 1971, p. 149.
44Lien, Buddhist Monuments and Antiquities of Gujarat, p. 56.
45Annual Report – Gujarat Department of Archaeology, 1972–73, p. 3.
46Annual Report – Gujarat Department of Archaeology, 1973–74, p. 4.
47Annual Report – Gujarat Department of Archaeology, 1969–70, pp. 3–4.
48Annual Report – Gujarat Department of Archaeology, 1969–70, pp. 3–4.
49Yasmin Sindhi, KadiaDungar – An Archaeological Site, Unpublished MA Dissertation, MS University, Vadodara, 1999, p. 14.
50Sindhi, KadiaDungar, p. 3.
51Sindhi, KadiaDungar, p. 25.
52Sindhi, KadiaDungar, p. 17.
53R.N. Mehta and S.N. Chowdhary, Excavations at Devnimori (A Report of the Excavations Conducted from 1960–63), Vadodara: MS University, 1966, p. 10.
54Y.S., Rawat, Recently Found Ancient Monastery and Other Buddhist Remains at Vadnagar and Taranga in North Gujarat, India, in Barbara Andaya (ed.), Bujang Valley and Early Civilizations in Southeast Asia, Malaysia: Department of National Heritage, Ministry of Information, Communication and Culture, 2011, p. 240.
55Rawat, Recently Found Ancient Monastery, p. 228.
56Rawat, Recently Found Ancient Monastery, p. 230.
57Rawat, Recently Found Ancient Monastery, p. 226.
58Rawat, Recently Found Ancient Monastery, p. 226.
59Sompura, Buddhist Monuments and Sculptures in Gujarat, p. 24.
60Susan L. Huntington, Buddhist Art Through Modern Lens – A Case of Mistaken Scholarly Trajectory, in Julia A.B. Hegewwald (ed.), In the Shadow of the Golden Age: Art and Identity in Asia from Gandhara to Modern Age, Berlin: E.B. Verlag, 2014, p. 80.
61Huntington, Buddhist Art Through Modern Lens, p. 84.
62Huntington, Buddhist Art Through Modern Lens, p. 80.
63Huntington, Buddhist Art Through Modern Lens, p. 86.
64Huntington, Buddhist Art Through Modern Lens, p. 88.
65Huntington, Buddhist Art Through Modern Lens, p. 90.
66Huntington, Buddhist Art Through Modern Lens, pp. 79–114, 93.
67Huntington, Buddhist Art Through Modern Lensp, p. 97.
68Huntington, Buddhist Art Through Modern Lens, p. 98.
69B.Ch Chabbra, Intwa Clay Sealing, Epigraphia Indica, 28, 1949–50, pp. 174–5.
70F. Kielhorn, Junagadh Rock Inscription of Rudradaman: The Year 72, Epigraphia Indica, 8, pp. 36–49.
71R.N. Kumaran, Second Urbanization in Gujarat, Current Science, 107 (4), 2014, p. 584.
72Kumaran, Second Urbanization in Gujarat, p. 583.
73Kumaran, Second Urbanization in Gujarat, p. 584.
74Kumaran, Second Urbanization in Gujarat, p. 584.
75Giorgio Riello and Prasannan Parthasarthi (eds), The Spinning World: A Global History of Cotton Textile, Oxford: Oxford University Press, Pasold Research Fund, 2009, p. 163.
76Rielloand Parthasarthi (eds), The Spinning World, p. 163.
77Erika S. Svendsen and Lindsay K. Campbell, Living Memorials: Understanding the Social Meaning of Community Based Memorials, September 11, 2001, Environment and Behaviour, 42 (3), 2010, p. 319.
78Svendsen and Campbell, Living Memorials, p. 326.
79Svendsen and Campbell, Living Memorials, p. 319.
80V.V. Mirashi, Daulatpur Inscription of the Reign of Chashtana, Journal of Oriental Institute, 28, 1978–79, pp. 31–7.
81P. R. Srinivasan, Three Western Kshatrapa Inscriptions, Epigraphia Indica, 37, 1967–69, p. 142.
82Srinivasan, Three Western Kshatrapa Inscriptions, p. 141.
83Prof. G. Bühler, A New Kshatrapa Inscription, Indian Antiquary, 10, 1881, p. 157.
84R.D. Banerji and V.S. Sukthankar, Three Kshatrapa Inscriptions, Epigraphia Indica, 16, 1921–22, p. 238.
85J.M. Nanavati and H.G. Shastri, An Unpublished Kshatrapa Inscription from Cutch, Journal of Oriental Institute, 11 (3), 1962, p. 237.
86Srinivasan, Three Western Kshatrapa Inscriptions, p. 141.
87Bühler, A New Kshatrapa Inscription, p. 157.
88Srinivasan, Three Western Kshatrapa Inscriptions, p. 142.
89Bühler, A New Kshatrapa Inscription, p. 157.
90Banerji and Sukthankar, Three Kshatrapa Inscriptions, p. 238.
91Indian Archaeology – A Review, 1982–83, p. 31.
92Pande and Vyas, An Early Temple in Gujarat, p. 108.
93Bhan, Archaeology of Jamnagar District up to 1300 AD, p. 386.
94Atusha M. Bharucha, The Archaeology of the Settlements of the Kshatrapa Period, Unpublished PhD Thesis, Deccan College, Pune, 1997, p. 124.
95Michael W. Meister and M.A. Dhaky (eds), Encyclopedia of Indian Temple Architecture, Delhi: American Institute of Indian Studies, 1991, p. 357.
96Kantilal F. Sompura, Structural Temples of Gujarat (up to 1600 AD), Ahmedabad: Gujarat University, 1968, p. 86.
97Jayaram Poduval, The Roda Group of Temples – The Pioneering Monuments of the Architecture Heritage of Gujarat, http://historyandarts.blogspot.in/2007/01/roda-group-of-temples.html (accessed on 28 July 2015).
98Poduval, The Roda Group of Temples.
99Jayaram, The Roda Group of Temples.
100Michael W. Meister (ed.), Encyclopedia of Indian Temple Architecture: North India: Foundations of North Indian Stylec.c.250 BC to AD 1100, Delhi: Oxford University Press, 1988, p. 357.
101Henry Cousens, Somanatha and Other Medieval Temples in Kathiawad, Varanasi, Delhi: Indological Book House, 1986 (reprint), p. 48.
102Sompura, Structural Temples of Gujarat, p. 86.
103Meister (ed.), Encyclopedia of Indian Temple Architecture, Vol. II, Part 1, p. 384.
104Meister (ed.), Encyclopedia of Indian Temple Architecture, pp. 197–206.
105Meister and Dhaky (eds), Encyclopedia of Indian Temple Architecture, p. 194.
106Kumaran, Second Urbanization in Gujarat, p. 582.
107R.N. Mehta and A.M. Thakkar, Copper Plates of the Time of Toramana, Vadodara: MS University, 1978.
108Ranabir Chakravarti, Three Copper Plates of the Time of Toramana: Glimpses of Socio Economic and Cultural Life in Western India, in Ellen Raven (ed.), South Asian Archaeology 1999: Proceedings of the Fifteenth International Conference of the European Association of South Asian Archaeologists, held at the Universiteit Leiden, 5–9 July 1999. Groningen: Egbert Forsten (Gonda Indological Studies 15), p. 397.
109J. F. Fleet (edited) Corpus Inscriptionum Indicarum – Inscriptions of the Early Gupta Kings and their Successors, Vol. 3, Calcutta: Superintendent of Government Printing, 1888, p. 149.
110Samuel Beal (translated), Si-yu-ki Chinese Accounts of India II, Calcutta, 1958, p. 260.
111Beal (translated), Si-yu-ki Chinese Accounts of India II, p. 266.
112Beal (translated), Si-yu-ki Chinese Accounts of India II, pp. 266–7.
113Beal (translated), Si-yu-ki Chinese Accounts of India II, p. 269.
114Sukumar Dutt, Buddhist Monks and Monasteries of India, London: George Allen & Unwin, 1962, pp. 228–9.
115Rawat, Recently Found Ancient Monastery, p. 229.
116Rawat, Recently Found Ancient Monastery, p. 229.
117Rawat, Recently Found Ancient Monastery, p. 231.
118Rawat, Recently Found Ancient Monastery, p. 234.
119Rawat, Recently Found Ancient Monastery, p. 235.
120Rawat, Recently Found Ancient Monastery, p. 237.
121U.P. Shah, GujaramaBaudh Dharma, Svadhaya, Book. I. No. iii. Vadodara: Svadhyaya, 1964, pp. 317–21.
122Rawat, Recently Found Ancient Monastery, p. 231.
123Sompura, Buddhist Monuments and Sculptures in Gujarat, pp. 29–30.
124Mehta and Chowdhary, Excavations at Devnimori.
125Lien, Buddhist Monuments and Antiquities of Gujarat, p. 66.
126D.R. Bhandarkar, Plates of Dantivarman of Gujarat, Samva 789, Epigraphia Indica, 6, 1900–1901, pp. 285–94.
127A.S. Altekar, A New Copper Plate of Dhruva II, Epigraphia Indica, 22, 1933–34, pp. 66–7.
128M.G. Dikshit, History of Buddhism in Gujarat, Journal of Gujarat Research Society, 8 (2 and 3), 1946, p. 111.
129K. Virji, Ancient History of Saurashtra, Bombay: Konkan Institute of Arts and Sciences, 1952, p. 219.
130M.R. Kale, Daśakumāracarita of Dandin, Delhi: Motilal Banarsidass, 1966, p. 332.
131J.F. Fleet (ed.), Corpus Inscriptionum Indicarum – Inscriptions of the Early Gupta Kings and their Successors, Vol. 3, Calcutta: Superintendent of Government Printing, 1888, pp. 61–5.
132Mehta and Thakkar, Copper Plates of the Time of Toramana.
133B.D. Chattopadhyaya, The Making of Early Medieval India, New Delhi: Oxford University Press, 1994, p. 29.
134D. Desai, Social Dimensions of Art in Early India in Proceedings of Indian History Congress, Gorakhpur: Indian History Congress, 1989–90, pp. 21–56.
135H. Kulke, State in India, 1000–1700, Delhi: Oxford University Press, 1995, p. 14.
136H.P. Ray, The Arcaheology of Sacred Space – Introduction, in H.P. Ray and Carla Sinopoli (eds), Archaeology as History in Early South Asia, New Delhi: ICHR & Aryan Books International, 2004, p. 355.
137Alka Patel, Architectural Histories Entwined – The Rudra Mahalaya/Congregational Mosque of Siddhpur in Gujarat, Journal of the Society of Architectural Historians, 63 (2), 2004, pp. 144–63, 154.
