6 6The formulation of the role of the religious shrine in the ancient period, as discussed in this study, is different from that which is generally accepted by historians of ancient India. In current historical writing, the emphasis has been on the spread of Sanskritic culture around the fourth–fifth centuries CE through movements of brahmanas at the behest of the emerging political elite and the consolidation of a feudal order. The temple is often associated with the newly emerging states in what is termed the early medieval period. In lieu of generous donations, the brahmanas and temples are said to have provided legitimisation to the rulers.1 There is unanimity also in the association with migrations of brahmanas and their role as priests in consolidating the new cults, frequently described as of Puranic affiliation. How valid are these claims? These formulations have been countered elsewhere and the arguments will not be repeated here.2
The data presented in Chapters 2 to 6 have convincingly demonstrated that temples and religious edifices not only represent the prevalent religious beliefs and practices, but also, bear testimony to the social and economic conditions of which they were a crucial component/element. The construction of religious structures implied the involvement of people who were attached to these and performed their delegated tasks. Resources were then required not only for rituals associated with deities, but also, for the maintenance of those individuals whose livelihood depended on the religious structures such as temples.
Another contribution of this study has been the use of inscriptions as source material for the study of religion. Most of the epigraphs had earlier been utilised mainly to reconstruct the genealogy of the various ruling dynasties, or in the case of early medieval India, to prove the emergence of feudalism as also regional polities. No doubt inscriptions provide valuable information about the reign of the Mauryas, Kshatrapas and the Gupta dynasties in Gujarat as elsewhere. Inscriptions at the site of Girnar prove that the Maurya, Kshatrapa and the Gupta dynasties considered the site to be significant for recording their exploits. On the great rock at Girnar, in addition to the Asokan edicts, there is another inscription of Rudradaman that records that a dam was built near the rock edict during the time of Chandragupta Maurya, and in the time of Asoka, the Sudarsana lake was adorned with conduits. The lake was repaired again under the rule of Rudradaman, and once again, during the rule of Skandagupta of the Gupta dynasty. This inscription has been utilised by historians to reconstruct the rule of these major dynasties over Gujarat. Other inscriptions that are used for similar purposes are the Mulavasar inscription and the Jasdan inscription pertaining to the rule of the Kshatrapas. An interesting point in the Girnar inscription is the mention of the construction of a Vishnu temple in the vicinity of the dam. According to Campbell, the only trace left of the original temple is a pilaster built into the wall to the right as one enters the modern Damodar temple, which was constructed in the fifteenth century.3
Gavin Flood rightly states, ‘temples were also centres of learning, popular devotion and pilgrimage. We need to understand not only the architecture and formation of temples, but also their social and religious context. We need to understand how the temples were built, what went on inside them and what they represent’.4 The archaeological data more importantly include an analysis of the location of religious architecture within the social domain and cultural landscape. Religious architecture is an indicator of interaction with diverse interest groups, such as worshippers, ritual specialists, patrons, artisans, and so on.5 Religious shrines transformed the social geography of the region and building new temples stimulated economic growth.6
Religious structures of Gujarat thus need to be viewed and studied from a different perspective, as shown in this book. It is not merely the temple as a structure that is important, but rather, locating it in broader terms and placing it within the social domain and wider networks. Temples and images carry narratives of their own, a story of continuities, of reassertions and of new identities. They symbolise the fact that at a particular point in time, the religious needs of the people underwent change, and the structures are witness and symbols of those changes and continuities. In all the studies carried out for Gujarat, the issue of coexistence and sharing of ideas between religions in terms of art architecture and ritual, as well as sacred space has not been an issue of concern. Sites in Gujarat demonstrate the parallel existence of Hinduism, Buddhism and Jainism at sites of Shamlaji/Devnimori, Junagadh and Vadodara/Akota, to mention a few.
An in-depth study of temples, religious imagery and traditional lore and literature demonstrates the role of the community in the reinvention and re-creation of religious identities at certain sites, which helped these continue well into the twenty-first century. For instance, at Shamlaji in North Gujarat, a post-Gupta image of Vishnu was found by a boy belonging to the Bhil community, and this image was enshrined in a temple constructed by a merchant. While some sites have a long thread of continuity (Dwarka, Somanatha and Miyani), some were reclaimed (Shamlaji) and others were abandoned (Ghumli and Boricha). Of the ones that survived, one again notices involvement of varied communities, such as that of the Jadeja clan at Miyani and Bilesvara, local village community at Pata, and Bhils and tribals at Shamlaji, which in turn, defines their networks within and outside the region as well. These are some of the perspectives adopted in the book while analysing religious sites in Gujarat.
Sacred sites need to be located in the context of settlement sites in order to enable identification of the support base of religious sites. Even though archaeological reports on excavations and explorations have been reported over time, a comprehensive study of these in context of religious sites has not been carried out. The settlement sites in this study have not been classified as fortified/non-fortified, or urban–rural dichotomy, but rather, in terms of economic activities such as craft production, salt manufacturing and iron extraction. The bases for site selection demonstrate variation and could be either political or due to economic activities, such as agriculture, craft, trade, or a combination of these. A study of the environment and topographic conditions of the region aids in a comprehensive understanding of the factors that played an important role of the communities that were essential for the continued sanctity of a sacred spot or religious structure. On the basis of architectural and inscriptional data, the involvement of communities in the continued sanctity of sites has been discussed, which vary from village community, to merchants, traders, monks nuns, and the ruling elite as well. For example, while sites along the coastline of Saurashtra with temples, such as at Kadvar, Mangrol and Porbandar served as intermediate ports and also subsisted on fishing, at Dwarka and Valabhi, agriculture and bead manufacturing formed the backbone of economic activities, and at the sites of Roda, Akota and Broach, agricultural activities were of prime importance.
Yet another crucial, but less analysed source is the numerous sculptural finds found scattered and reported from various sites in Gujarat. Religious imagery greatly aids in the reconstruction of sacred landscapes, religious beliefs, practices, and developments where architectural and monumental remains are lacking and nonexistent. They bring forth the diversity in religious sites of the region. The first and foremost source of information here are the punch-marked coins that contain depictions and symbols of not only deities, but interestingly, also layout plans of viharas and stupas, which are otherwise not archaeologically traceable for the period these coins belong to. With the help of tables and maps, the iconographical developments, heterogeneous in nature, have been traced in the period under study as each sub-region within Gujarat demonstrates varied preferences for iconographical images. For instance, mātṛkā images were more popular in North Gujarat, as seen at Shamlaji, Roda, Palej, and Ambaji, while Bhairava and Lakulisa images are found in the southern area of Gujarat at the sites of Avakhal, Vadaval, Karvan, and Vadodara. Images of interest found in the region are those of Lajjagauri at Valabhi, Pavi Jetpur, Dhank, etc. and Panchagnitapas Parvati at Karvan and Vagpur, which are small in size, but clear indicators of diversity in deities worshipped. Another image is of Visvarupa Vishnu found only at the site of Shamlaji and Devnimori in North Gujarat. Buddhist images demonstrate growing complexities within the religion, for example, at Dhank, Amreli, Valabhi, Taranga, and Gogha and the Jaina images from Akota, Vaviya, Mahudi, and Ahmedabad are the only evidence for the strong presence of the community.
The role of religious texts such as the Mahabharata and the Skanda Purana in establishing the continued sanctity of certain sites also receives attention. These texts highlight multiple religious processes in Gujarat, which involved a process of interaction and dialogue between local beliefs and practices, on the one hand, and the Brahmanical religion, on the other. Some of the themes covered in this study include continuity of Vedic rituals, a further elaboration of rituals in the Skanda Purana as compared to the Mahabharata, festivals involving pilgrims on a large scale, water as sacred sites in Skanda Purana, incorporation of local beliefs and practices, Somanatha and Dwarka in literature and archaeology, and religious developments in Gujarat in a wider pan-Indian context on the basis of study of Visvarupa Vishnu and Panchagnitapas Parvati images. The narratives weave sites within Gujarat with each other, as well as place them in larger religious circuit through myths, stories and legends.
An important aspect of the early temples of Gujarat is their location near the seacoast. These coastal temples were dedicated to a variety of deities, ranging from the non-Sanskritic fertility goddess Lajjagauri, whose shrine dated to first century BCE was excavated from the site of Padri in the Talaja tahsil of Bhavnagar district of Gujarat hardly 2 km from the Gulf of Khambat to temples of Surya or Sun and other gods along the Saurashtra coast from the sixth century onwards. The shrine at Padri also needs to be placed within the larger context of increased hierarchy of settlement in the area at this time. Padri was known for extraction of salt, while Hatab was the largest site, located close to the sea with an area of over 40 hectares. Archaeological exploration in the region has provided evidence for 22 early historic sites located in a linear pattern along the Shetrunji river and a multi-tier settlement hierarchy.7 In the fifth century CE, the lower Shetrunji valley in Bhavnagar district emerged as the core region of the early Maitraka rulers (475–767 CE), as evident from inscriptions, with the political centre at Valabhi at the head of the Gulf of Khambat.
From the 10th to 13th century, the primary route was along the coast from Dwarka on the Gulf of Kachchh to Somnath on the Saurashtra coast and Bhavnagar at the head of the Gulf of Khambat. The coastal centres of Somnath and Dwarka were well-known for their magnificent temples, though the beginnings of these sites date to the early centuries of the Common era. We start with the nature of the early temple in western India and changes in its interactive circuits across the ocean over time.
This section addresses the theme of cultural integration through the institution of the religious shrine or Hindu temple, especially with reference to those located on or near the west coast of India. Traditionally, the Indian coasts have been portrayed as inhospitable regions in historical writing, lacking natural harbours and afflicted with a shallow continental shelf and turbulent swells of waves. Did this deter travel across the ocean and attempts to control the waters? Gujarat has a coastline of 1,600 km, most of which lies in Saurashtra. Saurashtra is bounded by the Gujarat plains in the east and north east, by the Gulf of Kachchh and Little Rann on the north, and on the south east by the Gulf of Khambat. The Arabian Sea borders the entire southern seaboard. As recently as 1970, sailing vessels carried 30 per cent of the total trade. It had one large international port at Kandla in Kachchh and 46 smaller ports, which handled 40 per cent of the total traffic. Thus, sailing vessels were an integral part of the maritime cultural landscape of Gujarat. In the 1950s, the Swedish maritime ethnologist Olof Hasslof introduced the term sjobruk or maritime cultural landscape, signifying demarcation and utilisation of maritime space by communities for settlement, fishing, shipping, pilotage, and so on.8 A significant feature of this maritime landscape was the religious shrine and it is the role of the shrine that forms the focus of this section.
A discussion on the coastal shrine is critical for this book in order to dispel the myth that due to restrictions stipulated in the Law Books or the Dharmasastras on maritime travel, the Hindu population turned to ‘agrarian pursuits and production, away from trade and maritime transport’.9 Temples are best described as ritual instruments and their function is ‘to web individuals and communities into a complicated and inconsistent social fabric through time’.10
It is the strands of this social fabric that need to be understood and appreciated. It is significant that while the origin myths of most of these temples associate their founding with a royal patron, there is little historical evidence for this during most of their existence. Another association is genealogical and the fact that a particular community has a special relationship with the deity enshrined in the temple. Monastic and temple-centred religious institutions formed an important intermediate group between the state and the family. Thus, the temples and monasteries were not merely centres of devotion and worship, but were also principal institutions in the period from the ninth to thirteenth century for establishing laws and enforcing them on their members.
In addition to their role as adjudicators in society, religious shrines were also consumers of a variety of commodities used in ritual, as well as important locales for trading activity, as indicated by shops and markets within or in the vicinity of temple premises. The building of new temples stimulated economic growth, thereby transforming both the geographic and social landscapes of the region.11 At the same time, there are several instances of a differential tax on commodities required for religious purposes.
Of interest to this discussion is the copper-plate charter of Vishnusena in Sanskrit, issued from Lohata in the Kathiawar region. D.C. Sircar, the editor of the inscription has identified Lohata with the town of Rohar on the Gulf of Kachchh. The find-spot of the copper plates is unknown, but on palaeographic grounds, it is dated to sixth–seventh centuries CE. The inscription states that the king Vishnusena was approached by the community of merchants from Lohata to endorse customary laws prevalent in the community and which had been continuing for several generations. The king assures protection to the community of merchants established in the region and endorses their continued functioning. The inscription then provides a detailed list of 72 trade regulations or customary laws to be followed by the merchant community.12
Some of the regulations are of great interest to this discussion. For example, it is specified that merchants staying away for a year were not required to pay an entrance fee on their return (rule 52). Other clauses specify duties that were to be paid. A boat full of containers (bhanda-bhrta-vahitrasya) was charged 12 silver coins, but if the containers were for religious purposes, they were charged only one and a quarter silver coin (rule 53). In the case of a boat carrying paddy, it was half this amount. The exception to this was a boat carrying buffaloes and camels, where no reduction is allowed (rule 54). Other items, which were frequently transported by boat included dried ginger sticks, bamboo, wine, leather, and bulls. The variety of taxable objects mentioned in the inscription is an indication of the diverse nature of trade in the region. These included oil mills, sugarcane fields, wine, cumin seed, black mustard, and coriander. The inscription also refers to a tax on dyers of cloth, weavers, shoemakers, and retailers hawking goods on foot. Others such as blacksmiths, carpenters, barbers, potters, etc. could be recruited for forced labour under the supervision of officers. Thus, the record makes a distinction between commodities meant for religious purposes and the temple, as opposed to those to be sold in the market and underscores a differential in taxation.
Another early centre of religious architecture on the west coast is the site of Cotta Chandor in Chandor district in south Goa on the banks of the river Paroda leading to the sea, which was subjected to excavation for two field seasons in 2002–04.13 The complete plan of a brick temple complex, datable from third to eleventh century CE, was unearthed and five phases of structural activity were identified. Though three phases of construction were identified, these were marked by continuity of religious beliefs, and in the last phase, the sculpture of Nandi was added to the temple complex. Politically, the Kanara coast was controlled by the Kadambas from 350 to 550 CE and several families are known who ruled from centres further inland, such as at Banavasi and Halsi. The Silaharas followed the Kadambas in Goa from 750 to 1020 CE, but the Kadambas remerged in the tenth century. The Panjim plates refer to king Guhalla Deva of the Goa Kadambas (750–1020 CE) undertaking a pilgrimage to the temple of Somnath on the Saurashtra coast, but hardly had he reached halfway, when the mast of his ship broke and he was forced to take shelter with a ruler friendly to him. This was the port of Goa where a rich Muslim merchant by the name of Madumod of Taji origin and the wealthiest of all the seafaring traders, came to the help of the king. In return, the king gave him much wealth. This record tells us, for the first time, of Arab traders settled on the Goa coast in the eleventh century CE.14 These references are important as indicators not only for the presence of religious shrines in coastal areas of western India and donations of land to the temples, but more importantly, of travel and pilgrimage by the coastal route that provided interconnectedness to the shrines.
It is evident from the discussion above that the crucial element in the Asian coastal landscape was the religious shrine. It is important to locate this shrine in context, both physical and social in order to unravel the multiple levels at which sacred sites interacted with a diverse range of communities and negotiated between these. Another aspect of the shrine is its horizontal expansion and additions made to it over time to house a variety of functions of interest to this paper, such as ghatikasthana or centre of learning, which came to be incorporated in temples, especially in peninsula India from the eighth–ninth centuries onwards.
Eleventh-century inscriptions from the temple in Thirumukkudal, on the banks of the Palar river near Kanchipuram in Tamil Nadu, indicate the existence of a Vedic pathshala attached to the temple, as also a medical centre termed athura saalai and arrangements for distribution of medicinal herbs. Inscriptions from the Buddhist site of Kanheri near Mumbai and a temple in Gujarat would suggest that this practice may have earlier beginnings. Sixth-century inscription from Kanheri refers to a donation by vaidya or physician. Three copper plates from central Gujarat dated in the reign of the Huna Toramana (fifth–sixth centuries CE) record gifts made to the temple of Jayaswami or Narayana belonging to the queen mother by the trading community of Vadrapalli. Vadrapalli was probably located 8 km to the west of Sanjeli and signatories to the donation included traders from Ujjain, Kannauj, Mathura, and perhaps, Mandasor in central India. A goldsmith constructed a lake near the temple. The copper plate states that itinerant mendicants visiting the temple, as also devotees, should be provided with medicines (bhaisaja).15 These are aspects which need further research and analysis. In this concluding chapter, we would like to keep the focus on the Gujarat coast and its place in the maritime world of the first millennium CE.
In this section, it is suggested that the fishing and sailing communities formed a distinct group and were the crucial component of sea travel in the western Indian Ocean, extending from the west coast of India to the Red Sea and the African coast. Fishing was the traditional occupation of coastal groups in several pockets of the Indian Ocean and this is an adaptation that dates to at least the fifth millennium BCE in several areas.16 These communities adopted numerous occupations associated with the sea: fishing and harvesting other marine resources, salt-making, sailing, trade, shipbuilding, and even piracy. These maritime communities are to be distinguished from merchants and traders involved in oceanic trade. Merchants and traders in some cases certainly owned ships and watercraft, but they neither manned nor sailed these. More often, however, goods and cargoes were entrusted to the skipper of the vessel, who was then responsible for their sale and profit. It is important to stress the participation of these different communities and diverse ethnic partners so as to highlight the varied nature of seafaring activity in the western Indian Ocean.
The Periplus Maris Erythraei perhaps provides the first detailed description of local boats in the Indian Ocean in the early centuries CE and several types extending from the East African coast to the west coast of India are referred to.17 The Periplus informs us that: ‘Two runs beyond this island [Menuthias = Zanzibar?] comes the very last port of trade on the coast of Azania, called Rhapta [sewn], a name derived from the aforementioned sewn boats, where there are great quantities of ivory and tortoise shell’ (section 16). There is a strong tradition of archaeological research in this part of East Africa, though objectives and strategies have changed over the last six decades. Current research indicates that prior to 2300 BP, the entire littoral was occupied by hunter-gatherer-fishing communities using microlithic stone tool technology and utilising marine resources. The period from 2300 to 1500 BP witnessed the gradual emergence of the first farming and iron-using groups.18 In recent years, Felix Chami has found archaeological evidence for artefacts associated with early trade on Mafia Island, and, not far away, on the mainland, near the mouth of the Rufiji River, which he dated to the first few centuries CE.19 Interestingly, the Periplus informs us that Rhapta was under the firm control of a governor appointed by the Arabian king of the Yemeni centre of Muza, where taxes were collected, and it was serviced by ‘merchant craft that they staff mostly with Arab skippers and agents who, through continual intercourse and intermarriage, are familiar with the area and its language’.20
A range of communities participated in maritime activity at this time, such as the Nabataeans, Sabaeans, Homerites, and Arabs, in addition to Indians. Trade in cloth, wood and agricultural products sustained the Indian Ocean network, as is indicated by the presence of guilds of weavers, potters, oil millers, and so on, in the list of donors mentioned in the inscriptions from the Buddhist monuments of peninsular India.21 This is further substantiated by the botanical evidence from the archaeological excavations at Berenike on the Red Sea coast, which included imports from South Asia, such as pepper, coconut, Job’s tear, and possibly, rice.22
Berenike was a multicultural site and the inhabitants came from throughout the ancient world, including Egypt, the Mediterranean, Axum, sub-Saharan Africa, south Arabia, Nabataea, Palmyra, and perhaps, India. Greek was the lingua franca in early Roman times and most texts are in that language, though ostraka found in the city indicate a substantial Roman military presence. The Nikanor Archive belonging to a family of camel owners involved in the transport of goods between the Nile valley and the Red Sea coast indicates the participation of elite families resident in Egypt in trading activity.23
Data from shipwrecks indicates that pepper had been imported into the Mediterranean at least from the second millennium BCE. It was a critical ingredient in medicines, had culinary applications and was also used in funerary and religious rituals. Pepper has been found at several sites on the Red Sea coast, but perhaps was the most noteworthy import to Berenike and included black pepper, white pepper or black pepper that had ripened and long pepper. Black pepper was a product of South India, while long pepper was cultivated in North India, and in all probability, shipped from the west coast centre of Bharuch or Barygaza in the present state of Gujarat.
In the context of Gujarat, the Periplus refers to Syrastrene or Saurashtra; Eirinon or the Rann of Kachchh beyond which lies Barake or the Gulf of Kachchh (section 40). This is a dangerous gulf to navigate for ‘not only are the waves there very big and oppressive, but the sea is choppy and turbid, with eddies and violent whirlpools’ (section 40). After this, the author mentions the Gulf of Barygaza identified with the Gulf of Khambat (section 41). The region is described as ‘very fertile’ and ‘in the area there are still preserved to this very day signs of Alexander’s expedition, ancient shrines and the foundations of encampments and huge wells’ (section 41). Somewhat later, the author refers to ‘old drachmas engraved with the inscriptions, in Greek letters, of Apollodotus and Menander, rulers who came after Alexander’ being found in the market of Barygaza (section 47).24
Does the Periplus indicate the ethnic identity of traders who traversed the sea lanes of the western Indian Ocean? Muza at the mouth of the Red Sea is described in the Periplus Maris Erythraei as a port of trade without a harbour (section 24), but with a good road-stead for mooring, and teeming with Arabs – shipowners or charterers and sailors (section 21). Leuko Kome on the Red Sea coast was the harbour of the Nabataeans where craft, ‘none large’, loaded with freight from Arabia (section 19). While discussing Muza or Mocha on the Yemen coast of the Red Sea, the text refers to Arabs – shipowners or charterers and sailors – who ‘trade across the water and with Barygaza, using their own outfits’ (section 21). Muza is termed ‘port of trade’ even though it lacks a harbour. Another important harbour and storehouse at the entrance to the Red Sea was the island of Socotra, which was settled by Arabs and Indians and even some Greeks, who sail out of there to trade (section 30). Merchants from Barygaza or Bharuch customarily traded with Oman and centres in the Persian Gulf and brought in supplies of copper and logs of teak and ebony. The Malabar Coast owed its prosperity to Greek shipping, and also, to traffic from Ariake or Gujarat (section 54). On the east coast, the Periplus refers to Poduke identified with Arikamedu near Pondicherry where local boats mix with those that sail down from the region of the Ganga and from across the Bay of Bengal.
Thus, it is evident that diverse local communities in Gujarat, as elsewhere, participated in seafaring activity, as reflected in references to local watercraft in the Periplus Maris Erythraei. A reading of the Periplus also establishes the absence of state control over maritime trade, either in the region of the Red Sea or further afield, though it is true that the local polities attempted to extract revenues from the sale of trade commodities at market centres.
The ethnographic data indicate that the coastal settlements participated in trade with other centres further south and also across the Ocean, though each port specialised in a particular route. Mandavi, Porbandar and Veraval trade with east Africa, Porbandar and Veraval with south Arabia, Southeast Asia and the Persian Gulf. Thus, route specialisation is an important aspect of seafaring activity that has far-reaching implications for a study of the past as well.25 Another critical factor is the sailing season, which in Gujarat comes to end in June, after which the monsoon winds become too dangerous until end-July.
One aspect that is missing from the Periplus is any reference to the belief systems of the local communities. Here again, ethnographic data show participation of the sailing communities in a variety of rituals and beliefs. There are rituals performed at each stage of construction of the watercraft and are also compiled in a book titled Nauka Navghatan Pujanam or worship for new ship construction. The communities are linked to coastal temples dedicated to a range of deities and often donate regularly for maintenance as also performance of rituals. A third feature in the religious landscape is the memorial stone or paliyar set up in memory of those who lost their life at sea. It is also significant that temple structures often double as markers to identify the coasts and are often associated with local legends of saviours at sea. Thus, religious architecture fulfils both religious and navigational purposes and this would have been important in an earlier period also,26 though the primary resource for an understanding of the weltanschauung of the shipper and the sailor remains oral traditions.
Rather than posing a dichotomy between oral and literate traditions, it is the interface between the two that is significant. It is suggested here that in the historical period, one significant use of writing was for trading activity. The shared culture that extended across not only South Asia, but also the Indian Ocean was part of a literate tradition, which was by no means controlled by the ruler or the brahmana, but included Buddhist and Jaina monks, navigators and trading and crafts groups. Writing facilitated storing of information, cumulative knowledge promoted new genre of cultural and artistic expression and aided ordering of information under numeric and alphabetic heads and the use of maps.27 Of interest to this book is the use of writing as a marker of identity and valuable insight into this aspect of seafaring activity is provided by recent data from the island of Socotra, as discussed in the next section.
Socotra is a small archipelago of four islands about 250 km east of the Horn of Africa, off the coasts of Yemen and Somalia in the western Indian Ocean. The largest island in the group is also called Socotra. The 132-km-long island is often described as the most isolated place on Earth, though this account does agree with its central location in the maritime networks of the early centuries of the Common era, as evident from archaeological excavations conducted over the years. The excavations at Socotra need to be viewed in the context of the marine resources of the island. One of the earliest descriptions of the island is to be found in the Periplus Maris Erythraei (section 30), which refers to it as Dioscurides. Several varieties of tortoise shell found on the island are referred to, and are said to be in demand by the shippers from Arabia and the west coast of India, who exchanged big cargoes of tortoise shell for rice, grain, cotton cloth, and female slaves. The other resource referred to is cinnabar, which is collected as an exudation from the Dragon trees.
One of the oldest known archaeological sites on the island of Socotra has been located in the vicinity of the modern village of Rakuf in the eastern part of the island. Remains of a workshop used for the manufacture of flint tools were found. Close to the site was a small cemetery, comprising of ten dolmen-shaped grave structures comprising burial chambers of stone cists unique on the island. These burials had scant grave furniture and as a result the excavators found it difficult to provide a precise date. The graves were very different from those found in other parts of the island, but structurally similar ones have been found in south Arabia and in the Persian Gulf and Oman.28 On the basis of analogies with these remains, the burials on Socotra have been dated from the second half of the first millennium BCE – a date which corresponds well with the evidence from Greek sources regarding beginnings of trade contacts with the island. It also indicates colonisation of the island by local communities with close links to the south Arabian coast.
Contemporary to these burials is the graffiti found on limestone outcrops on the eastern coast of the island at Eriosh. The graffiti includes drawings of outlines of feet, purely geometric shapes and animal and human forms. There are attempts at imitating a script, but no identifiable South Arabian inscription could be found. Another type of stone structure found extensively on the island of Socotra is what has been described as ‘boundary walls’, though the function of these is not quite clear. It would seem that they demarcated plots of land growing incense trees such as aloe and cinnabar. In many cases, in the vicinity of these walls were found stone structures and cemeteries.
One of the most interesting discoveries was that of a settlement along the northern littoral of the island 2 km south of the modern village of Suq. An analysis of the archaeological data from the site indicates two phases of settlement: an earlier phase dated to the early centuries of the Common era; and a second phase from around the tenth to thirteenth century CE. The archaeological data also suggest similarities with the Arabian Peninsula and Persian Gulf.
In the first century CE, Hoq cave on the northern face of the island of Socotra was known in a wide region by merchants travelling between Africa, India, the Middle East, and Far East. Explorations in the 2000-metre-deep cave in 2001 led to the discovery of 200 graffiti, drawings and small offerings dated to the late second to fourth centuries CE. Of these, 192 are in the Brahmi script, 1 in Kharosthi, 1 in Bactrian, 3 in Greek, 1 in Palmyrene Aramaic, and 20 odd in Axumite or ancient Ethiopian. The graffiti and inscriptions are not randomly engraved in the cave; instead, they occur in clusters at specific sites, most likely representing deliberate choice of location. In the graffiti written in the Brahmi script at site 14, for example, individuals adopt several terms to identify themselves, such as sea-captains; yavana, a term that is frequently also used in the epigraphs of the western Deccan caves or by their place of residence such as Barygaza or Bharuch on the west coast of India.29
These overwhelming cases of self-representation of individuals/groups from the Indian subcontinent may be compared with contemporary epigraphs on the west coast of India,30 both with reference to content and locations and add a hitherto under-researched perspective to maritime activity in the western Indian Ocean, which has largely centred on discussions of pottery and identifications of coastal centres mentioned in early Greek writing.31
Cosmas Indicopleustes (‘India-voyager’), a Greek sailor from Alexandria who travelled to Ethiopia, India and Sri Lanka in the early sixth century, provides a good example of interconnectedness of the western Indian Ocean. In later life, he became a monk, probably of Nestorian tendencies. Cosmas tells us that he was a native of Egypt, probably of Alexandria, never received a complete education (II, 1). He was a merchant (II, 54 and 56) in early life, perhaps importing spices and made many voyages. He knew Palestine and the area around Mt. Sinai (V, 8, 14, 51–52), had been to Socotra (III, 65), and had navigated in the Mediterranean, Red Sea and Persian Gulf (II, 29). He had rounded Cape Gardafui and sailed off Somalia (II, 30).32 Cosmas even mentions in book 2 another merchant, Menas, a friend of his, who also became a monk. His book, the Christian Topography of Cosmas Indicopleustes is dated to 550 CE. The surviving manuscripts of the text include an uncial manuscript of the ninth century, written in Constantinople (Rome: Vaticanus Graecus 699 (V), and No. 1186 of the Greek Mss. of the Monastery of St. Catherine at Sinai (S), dated to the eleventh century.33
The fourth century saw the rise of both the Christian holy man and Christianity as a civic institution in Egypt. Archaeological work at Abu Sha’ar, located 20 km north of Hurghada on the Red Sea coast, located a fort built in the late third–early fourth centuries CE as a result of the reorganisation of the eastern defences of the Roman Empire. After the fort’s abandonment in the early fifth century, it was occupied by Christian communities and converted into a church and the north gate became the principal entrance. The presence of graffiti, Christian crosses and two major ecclesiastical inscriptions in Greek attest to the importance of Abu Sha’ar as a pilgrimage centre in Upper Egypt. In addition, it was ideally located to facilitate travel to Sinai for St. Catherine’s Monastery or to Aila (Aqaba at the northern end of the Gulf of Aqaba) and onwards to Jerusalem.34
It would seem that from the fourth to seventh centuries, the focus of maritime activity shifted further south along the Red Sea coast to Adulis. Adulis, with the two harbours of Diodorus island and Orienê, is located at a spot where the Red Sea winds are less ferocious. At present, the site is located 7 km from the coast – though there is evidence that it was connected with the sea by a silted channel of the River Haddas. The sixth-century work of Cosmas Indicopleustes refers to the town in the Aksumite period and contains a sketch map showing Adulis a little way from the coast, clearly connected with Aksum. There are also indications for the presence of a church at the site.
The ceramic evidence, especially the presence of ribbed body sherds, largely from amphorae and costrels with some coarse wares from the kilns in Aqaba indicates continuing contact with the Jordanian Byzantine/Umayyad tradition with a probable seventh-century date. Aqaba amphorae have a wide distribution in the Red Sea area, having been found in fourth-century contexts at Berenike, while at Abu Sha’ar, the type occurs from at least the fifth century, and at Qana, they appear in the Upper Period dating to the sixth–seventh centuries CE. The easternmost distribution is at Kamrej on the west coast of India. Some of the amphorae bear Christian monograms, suggesting that the contents may sometimes have had a role in church liturgy.35
Amphorae jars were used for the long-distance transport of foodstuffs in antiquity, and when found in archaeological excavations, provide indicators of trade and maritime contacts. The torpedo jars, dated from third to eighth century CE, are of Mesopotamian origin and have a body resembling a torpedo. They look like amphorae without handles and probably had similar uses. Torpedo jars are common and ubiquitous between the fourth and the tenth centuries CE, having a very wide distribution in the western Indian Ocean, stretching from Iraq, along the Arabian and Iranian coasts, down the western coast of India as far as Sri Lanka, and along the east coast of Africa.36
In the Indian subcontinent, they are clustered around the north-western coast, particularly Gujarat and Maharashtra. Especially significant is their find at a Christian monastery site on the island of Sir Bani Yas, 170 km south west of Abu Dhabi in late sixth/early eighth century CE context.37
Thus, it is evident that trade networks across the western Indian Ocean indicate continuity, though the affiliation of shrines and sites often change over time.
In the final analysis, rather than ruptures and discontinuities, occasionally revived by external stimulus, a complex trading network involving a variety of groups including artisans, craftspeople and transporters marks the early history of Gujarat. The history of these communities is evident from the donations that they made to religious establishments. Thus, trade and trading activity cannot be studied in isolation from the diverse religious landscape that developed in the region. The attempt in this book has been to highlight the location and archaeology of religious structures, thereby placing the temple in its wider social base. It also needs to be appreciated that the temple was by no means unique; instead, it was part of a diverse sacred geography. For a comprehensive appraisal of crafts and communities in western India, the intertwined strands of religious architecture, economic activity and political intervention need to be examined and understood. An overview of this activity as presented in this book amply demonstrates the participation of communities of western India in a variety of trading networks, local, regional and transoceanic in the first millennium of the Common era.
1‘The rapid growth in the number and networks of temple centers, whose origins certainly date to pre-Gupta times, become understandable when we begin to appreciate how closely they were linked, as were gifts and land grants to Brahmins (brahmadeyas and agraharas) with the formation of subregional and regional kingdoms and their legitimation, consolidation of their resource bases, and the forging of linkages for social integration across communities’ (B.D. Chattopadhyaya, Historiography, History and Religious Centre – Early Medieval North India c. 700–1200 AD, in Visakha N. Desai and D. Mason (eds), Gods, Guardians and Lovers – Temple Structures from North India, Ahmedabad: Mapin Publishing, 1993, p. 42).
2Himanshu Prabha Ray, The Shrine in Early Hinduism: The Changing Sacred Landscape, The Journal of Hindu Studies, 2, 2009, pp. 76–96; H.P. Ray (ed.), Archaeology and Text: The Temple in South Asia, New Delhi: Oxford University Press, 2010.
3James M. Campbell, History of Gujarat, Gurgaon: Vintage Books, 1989, p. 70.
4Gavin Flood in H.P. Ray, Archaeology and Text – The Temple in South Asia, Delhi: Oxford University Press, 2010, p. xiii.
5H.P. Ray, Archaeology and Text – The Temple in South Asia, Delhi: Oxford University Press, 2010, p. 4.
6Cynthia Talbot, Pre Colonial India in Practice: Society, Region and Identity in Medieval Andhra, Delhi: Oxford University Press, 2001, p. 87.
7Ashit Boran Paul, The Early Historic Settlement and Subsistence Pattern in the Shetrunji River Basin, Bhavnagar District, Gujarat, Puratattva, 30, 1999–2000, pp. 99–105.
8C. Westerdahl, The Maritime Cultural Landscape, International Journal of Nautical Archaeology, 21 (1), 1992, p. 5.
9A. Wink, Al-Hind: The Making of an Indo-Islamic World, New Delhi: Oxford University Press, 1990, p. 72.
10Michael Meister, Ethnography and Personhood: Notes from the Field, New Delhi, Jaipur: Rawat Publications, 2000, p. 24.
11Cynthia Talbot, Precolonial India in Practice: Society, Region and Identity in Medieval Andhra, New Delhi: Oxford University Press, 2001, p. 87.
12D. Sircar, Charter of Visnusena samvat 649, Epigraphia Indica, 30, 1953–54, pp. 163–81.
13Derek Kennet and J.V.P. Rao, The Early Historic Brick Temples at Chandor, South Asian Studies, 17, 2001, pp. 97–107.
14M.M. George, The Kadamba Kula, B.X. Furtado, Bombay, New Delhi: Asian Educational Services (reprinted), 1931/1995, p. 171.
15K.V. Ramesh, Three Early Charters from Sanjeli in Gujarat, Epigraphia Indica, 40, 1973–75, pp. 175–86.
16Himanshu Prabha Ray, The Archaeology of Seafaring in Ancient South Asia, Cambridge World Archaeology Series, Cambridge: Cambridge University Press, 2003, Chapter II.
17L. Casson, The Periplus Maris Erythraei: Text with Introduction, Translation, and Commentary, Princeton: Princeton University Press, 1989.
18Colin Breen and Paul J. Lane, Archaeological Approaches to East Africa’s Changing Seascapes, World Archaeology, 35 (3), 2003, pp. 469–89.
19Felix A. Chami, The Graeco-Romans and Paanchea/Azania: Sailing in the Erythraean Sea, in Paul Lunde and Alexandra Porter (eds), Society for Arabian Studies Monographs 2 Trade and Travel in the Red Sea Region. Proceedings of Red Sea Project I held in the British Museum, October 2002, pp. 93–104.
20Casson, The Periplus Maris Erythraei, p. 61.
21Himanshu Prabha Ray, Monastery and Guild: Commerce Under the Satavahanas, New Delhi: Oxford University Press, 1986, p. 112.
22R. Cappers, Archaeobotanical Evidence for Roman Trade with India, in Himanshu Prabha Ray (ed.), Archaeology of Seafaring, New Delhi: Pragati Publications, 1999, pp. 51–69.
23Steven E. Sidebotham, Berenike and the Ancient Maritime Spice Route, Berkeley, CA, Los Angeles and London: University of California Press, 2011, p. 70.
24Casson, The Periplus Maris Erythraei, pp. 75–81.
25Nancy Pinto Orton, Sea-Going Trade in Early Historic Gujarat (ca. 100 BC – AD 500), Doctoral Thesis, University of Pennsylvania, 2001, pp. 100–1. Nancy Pinto Orton, Red Polished Ware in Gujarat: Surface Collections from Inland Sites, in S.A. Abraham, P. Gullapalli, T.P. Raczek and U.Z. Rizvi (eds), Connections and Complexity: New Approaches to the Archaeology of South Asia, Walnut Creek, CA: Left Coast Press, 2013, pp. 195–222.
26Orton, Sea-Going Trade in Early Historic Gujarat, pp. 78–85.
27Jack Goody, The Power of the Written Tradition, Washington, London: Smithsonian Institution Press, 2001, p. 144.
28V.V. Naumkin and A.V. Sedov, Socotra, Topoi, 3 (2), 1993, pp. 581–2.
29Ingo Strauch, Foreign Sailors on Socotra: The Inscriptions and Drawings from the Cave Hoq, Bremen: Hempen Verlag, 2012.
30Ray, Monastery and Guild, 1986.
31Eivind Heldaas Seland, Archaeology of Trade in the Western Indian Ocean, 300 BC–AD 700, Journal of Archaeological Research, published online 2nd April 2014.
32The region which produces frankincense is situated at the projecting parts of Ethiopia, and lies inland, but is washed by the ocean on the other side. Hence the [139] inhabitants of Barbaria, being near at hand, go up into the interior and, engaging in traffic with the natives, bring back from them many kinds of spices, frankincense, cassia, calamus, and many other articles of merchandise, which they afterwards send by sea to Adulê, to the country of the Homeritcs, to Further India, and to Persia.
33Cosmas Indicopleustes, Christian Topography, J.W. McCrindle (translated), London: Hakluyt Society, 1897.
34S.E. Sidebotham, Archaeological Work in the Eastern Desert and Along the Red Sea Coast, in Marie-Françoise Boussac and Jean-François Salles (eds), A Gateway from the Eastern Mediterranean to India, New Delhi: Manohar, 2005, pp. 109–10.
35David Peacock and Lucy Blue (eds), The Ancient Red Sea Port of Adulis, Eritrea, Oxford: Oxbow Books, 2007, p. 95.
36Roberta S. Tomber, Rome and Mesopotamia – Importers to India in the First Millennium AD, Antiquity, 81, 2007, pp. 972–88.
37R.A. Carter, Christianity in the Gulf During the First Centuries of Islam, Arabian Archaeology & Epigraphy, 19, 2008, pp. 71–108.