A Hoard of Copper Plates: Patronage and the Early Valkhā State

Historians of the state, society and economy in India from 600 to 1300 CE have extensively drawn upon and investigated copper-plate charters for their study. In other words, copper-plate charters recording the king’s grant of villages or land to brāhmaṇas or temples are one of the basic historical sources of early medieval India. With these charters numerous revenue-free settlements (agrahāra, brahmadeya, devadāna, etc.) were created in almost every part of India. These charters are markers of significant shifts in polity, economy and society. In the pre-600 CE period too, land grants inscribed on copper plates form the bulk of evidence for the study of polity, economy and society. The horizontal spread of the state society accompanying the institution of land grant frequently resulted in the formation of a monarchical state polity from c. 300 to 600 CE. State formation in regions without prior experience of the state was a continuing process. ² We have profusion of donative records, available from disparate regions of India during five centuries from c. 200 BCE to 300 CE. ³ The donors were, however, mainly persons from occupational and professional groups, mostly merchants and women who donated largely to Buddhist and Jaina establishments. From around c. 300 CE, the nature of the donor as well as the pattern of patronage underwent changes. As mentioned earlier there was a spurt in the number of monarchical powers not only in the Ganga valley or in the deltas of the Godavari and the Krishna, but also in remote areas or fringe zones. Land grants to brāhmaṇas and the genesis and growth of temple-building activities, the two perceived markers of patronage since 300 CE, can be appreciated in the context of the need to legitimize the emerging political structure and society. ⁴ Rulers derived their legitimation through patronage to brāhmaṇas, local deities and religious centres among others.

The period from c. 300 to 600 CE saw the beginning of royal elites who, in order to supplement their political control, engaged in construction of genealogies and giving out their resources to brāhmaṇas, temples and monasteries. Among the many new and small kingdoms which surfaced during the time of the Guptas was the kingdom of Valkhā in Central India, located on the banks of the Narmada. During this period and after, a horizontal spread of state society has been suggested implying transformation of pre-state polities into state polities. ⁵ In the case of the Valkhā kingdom, it appears that in the process of transition from pre-state to state, it can be placed in a category where with the formation of the kingdom in around the mid-fourth century CE, Valkhā has just transcended the pre-state stage and could be placed in the genre of an early state. Claessen and Skalnik divided the early state into three types, the inchoate, the typical and the transitional, each having certain dominant features. ⁶ In the following pages an attempt would be made to ascertain the category where one can situate the Valkhā state.

We seek to understand the early character of the Valkhā state through the lens of twenty-seven copper plates found together in a hoard and five others published in a scattered manner. Like most other hoards, the story of their discovery falls in the category of a chance find. While cultivating a field in a settlement known as Risavala, near present Bagh (Madhya Pradesh), in 1982, the hoard was found with the only exception being that in lieu of coins which are mainly discovered in hoards, here we had a huge copper container, covered with a copper lid. On uncovering the cover, a hoard of twenty-eight (one was broken) neatly arranged copper sheets, which were fully covered with patina as well as soil, saw the light of day. It goes to the credit of K.V. Ramesh and S.P. Tewari who edited the plates in 1990 and revealed the names of the rulers of Valkhā. ⁷ Before the discovery of these twenty-seven copper plates, five other copper plates of rulers of this region were discovered and published but scholars could not locate the kingdom of Valkhā. ⁸ These plates belonged to Bhuluṇḍa, Svāmīdāsa and Rudradāsa. In the hoard the other rulers are Bhaṭṭāraka and Nāgabhaṭa. It is clear that the first ruler Bhuluṇḍa had a non-Sanskritic name (use of hard sounds in the name is instructive) while the others used Sanskritized names such as Svāmīdāsa, Rudradāsa and so on. The plates do not give us any clue to genetically relate these rulers and this belies any prospect of formation of any dynasty. The only common factor relates to the name of the kingdom. Thus, any sort of conjecture is possible. On the basis of the similarity in the recording of these charters we may imagine a loose connection. Their inscriptions are in all probability dated during the Gupta reckoning and they acknowledge the superiority of the paramabhaṭṭāraka, most probably the Gupta emperor, and use a less grandiose title mahārājā. All the charters begin with paramabhaṭṭāraka – pad = ānuddhyāto. This expression also figures in the inscriptions of many of the feudatories of the period. ⁹ In a study of courtly culture in early medieval India, Daud Ali remarks ‘the focus on feet in courtly circles developed into a rich language of power’. He has shown that this terminology spread to kingdoms throughout the subcontinent in a huge number of inscriptions and became a common political idiom. ¹⁰ Pādānudhyāta is commonly translated as ‘meditating at the feet of’ but recently Ferrier and Torzsok ¹¹ have offered a different explanation for the use of this term. According to them, the correct understanding appears to be ‘favoured by the feet of’ where feet is an honorific, the whole expression meaning ‘favoured by the respected’. Thus, the expression seems to imply that the subordinate ruler derives his authority and power from his overlord, in this case, perhaps the Gupta rulers. With the penetration of Gupta power in the region, emergence of a state structure following the pattern of the Gupta administrative system is noticed. Along with the Guptas, the Vākāṭaka influence could also be discerned in the administrative terminologies of the Valkhā kingdom. As these rulers were of local origin, it is possible to show how the Gupta and Vākāṭaka incursions gave rise to a transformation of the indigenous elites of the region. Since the inscriptions are dated we can locate their contemporary Gupta and Vākāṭaka rulers.

The kingdom, as it appears from its land grants, ¹² was situated on both sides of the Narmada river, at the southern periphery of the important Gupta strongholds in central India (Airikiṇa, Eran). Eastern Malwa around Eran came under the Gupta rule during the reign of Samudragupta ¹³ (c. 335–375 CE) and continued to be associated with the Gupta realm till the reign of Budhagupta (c. 475–496/500 CE). ¹⁴ In fact his Eran Stone Pillar inscription of the year 165 or 483/4 CE speaks of the rule of his governor Suraśmichandra between the Kālinidī (Yamuna) and Narmada (Kālinidī Narmmadayor = mmadhyaṁ) embracing eastern Malwa. From an inscription inside a cave at Udaygiri in Vidiśā, ¹⁵ we learn that Chandragupta II (c. 375–414 CE) was present in eastern Malwa with an ardent desire to conquer the world (kṛitsna-pṛithvī jayārtthena). The Tumain (ancient Tumbavana) inscription of Kumaragupta I, year 116, ¹⁶ is indicative of Gupta occupation also in the west of Eran, in the Sagar district of Madhya Pradesh. This entire region came under the occupation of the Guptas perhaps during the reign of Samudragupta and was consolidated by his successors Chandragupta II and Kumaragupta I till Budhagupta as none of the Gupta rulers are known to have made any conquest in this region. Tumbavana being located on a very important trade route was a coveted region. In fact, the inscription talks of a merchant (sādhu) family. Thus the north of the Valkhā kingdom saw the continuous presence of the Guptas till Budhagupta’s reign around 496 CE. ¹⁷ However, the last known Valkhā king Nāgabhaṭa was a contemporary of Skandagupta (c. 454–480 CE). To the south of the Valkhā realm lay the northern frontier of the Vākāṭaka kingdom (c. 300 CE–500 CE) embracing the modern regions of central India and the northern Deccan. The Vākāṭakas first established themselves in the Vindhyan region, which included a large part of the Bundelkhand and Bagelkhand tracts. Then they extended southwards towards Deccan. There were parallel collateral branches, one founded by Sarvasena, at Vatsagulma or modern Wasim in Akola district, Maharashtra, the other led by Rudrasena I had its seat of power at Nandivardhana (Nagardhan/Nandardhan near Ramtek or Nandapur in Nagpur district of Maharashtra). Some of the major rivers delineating the landscape of the Vākāṭaka territory include Wardha, Narmada, Tapti, Godavari and their tributaries such as Purna, Dudna and the Wainganga. We have place names in epigraphs, such as, Bennātaṭa and Bennākaṭa, indicating their association with the river Wainganga (Bennā). ¹⁸ This perhaps influenced the Valkhā realm and thus in their inscriptions we have such expressions as Narmmadā-pāra-kule, Narmmadā-dakshiṇa-taṭe, Narmmadā-Uttara-taṭe ¹⁹ in the context of recording the location of the donated land properties. The Narmada valley is not one continuous plain but is broken up into parts, separated by hills. North of the Narmada is the Dhar upland, a hilly and forested area. Bagh is located near Dhar. The donated villages perhaps also fall in the region of the Nimar plain that lies on both sides of the Narmada and Nimar upland, which is entirely south of the Narmada. ²⁰ The area is rich in mineral deposits. Thus it actually was a buffer zone between the two important monarchical polities of the fourth and fifth centuries CE India with a strong forest and mineral resource base. Following B. Subbarao ²¹ we can place the Valkhā kingdom in an area of relative isolation which was not environmentally unfavourable and so the area displayed a pattern of survival and progress. Though not located on the main routes of communication, the region could be linked with arterial routes through narrow passageways. Some of the areas of donated villages, both on the north and south of the Narmada had concentration of Chalcolithic sites such as Mahesvar and Navdatoli ²² which suggest a strong antecedent of village farming culture. The identifiable donated villages from the charters of the Valkhā kings can be located in the Narmada valley falling within 72°–80° longitude and 20°–24° latitude. ²³ Apart from Valkhā being the name of the kingdom, we have such terms as Valkhādhishṭhāna ²⁴ in the grant of Bhuluṇḍa, year 50 (369/370 CE), and Valkhā Vāstavya ²⁵ in the grant of Svāmīdāsa, year 63 (382/383 CE). The use of the suffix adhishṭhāna implies an administrative centre at Valkhā which therefore may have been the capital of the kingdom of the same name. Valkhādhisṭhāna has been identified with present day Balkhar in Dhar district of Madhya Pradesh by the editors of the charters. We have the famous archaeological site of Mahesvar ²⁶ (west Nimar) near Balkhar. Linkages between the Mahesvar area on the Narmada with regions on the Jamuna river was possible traversing the plateaus and then connecting with river Chambal. The area of the Narmada where Mahesvar lay on the north and Navdatoli on the south was one of the fording points. The emergence of the Valkhā kingdom around Mahesvar on the Narmada may underline the traditional significance of this zone as a major crossing point of the Narmada. ²⁷ Being contiguous with the Gupta and the Vākāṭaka kingdoms, it is natural that the local chiefs were influenced by the monarchical political structure. Thus, these new rulers emerged in the history of what may have been a peripheral area of those times. They represented the appearance of local monarchies whose background is uncertain.

Romila Thapar had suggested that in the threshold period things are announced and they take a definite shape in later times. ²⁸ This is applicable to the Valkhā state which saw the beginning of land donations in the form of copper-plate charters in western central India which took a definite structure in the later period. Land donations are linked with agrarian expansion. This in turn is crucial to the emergence of state society. New areas were brought under the fold of cultivation and this changed the political economy of the region. There are many instances in Indian history where settlements were created after reclamation of forest tracts or marshy lands. It is evident that in case of Valkhā state formation was possible because it had attained the technological level of settled agriculture. But the process needs to be studied. Surajit Sinha ²⁹ has shown that in a much later period internal development processes in the tribal belt took the form of aspirations to meet the Rajput model with corresponding adaptations. Whether the emergence of the Valkhā kingdom was also a part of the internal development of a pre-state polity with aspirations for monarchical state structure as seen in the neighbouring regions is a question to ponder upon. Thus through a reading of these charters we seek to understand the emergence and growth of the Valkhā state. With donations the donee assumes a significant position and so the nature of patronage of the Valkhā rulers would also be central to our study.

It is evident that this was the initial phase of the land grant charters in India. Therefore, it is natural that the structure of later land grants is not followed here. We know that rather than the early Gupta rulers, it was their contemporary Vākāṭakas and their subordinates who issued land grants. It is important to note that the earliest copper-plate charters written in Prākrit belonged to the Pallava kings of Kāñchī and assignable to the middle of the fourth century CE. ³⁰ The first charter of Bhuluṇḍa was issued in the year 47 of the Gupta era which means c. 367 CE. ³¹ Thus this was perhaps the earliest Sanskrit copper-plate charter issued by a local ruler. However, we do not have a long genealogy of the king, benedictory and imprecatory verses per se, and a record of the king’s prowess and conquest. What we do have is the name of the Brāhmaṇa donee specifying his lineage (gotra) or in some cases the donation is made towards a religious establishment, locational information of the village or land to be granted, prohibition of civil and military officials from entering the grant, and an affirmation that the grant was made as long as the moon, sun and the stars lasted (śāśvatam-āchandr-ārkka-tārakā-kālīnena), that is, the grant is made in perpetuity. It should be noted that an almost similar expression śāśvata-āchandr-ārkka-tārakā-bhojye is found in the D̄amodarpur copper plate inscription of Kumāragupta I, year 124, that is, 444 CE. ³² Incidentally Damodarpur is a village in west Dinajpur district in West Bengal and thus the record belongs to northern Bengal, far removed from the Valkhā area. However, the earliest example of the idea of perpetuity of a donation could be seen in the Nagarjuni Hill cave inscriptions of Daśaratha (c. 220 BCE) where the expression ā-chandram-shīliyam (ā-chandramāḥ-sauryam) ³³ is used. Thus, the notion of permanence was always associated with donation whatever might have been the mode of expression.

It is significant to note that in these copper plates we have the term agrāhāra (agra + āhāra), implying that the specific land was donated for the sustenance of the brāhmaṇas, in lieu of the much familiar agrahāra used in the later records. Consequently, the term agrahāra evolved from agrāhāra which is unique to these copper plates only. Along with agrahāra another term used is brahmadeya. It will be worthwhile to remember that the janapadaniveśa section of the Arthaśāstra refers to the fact that the brāhmaṇas were granted brahmadeya lands to be enjoyed hereditarily and these were exempted from all kinds of daṇḍa and kara (brahmadeyānyadaṇḍakarānyābhirīpadāyakāniprayachchhet). ³⁴ Therefore, the expression brahmadeya can be traced back to Arthaśāstra. Many of the charters mention concurrently Brahmadeya and Agrahāra, for example, brahmadeyāgrahāra ³⁵ in a grant of Rudradāsa, year 69, the two terms being clearly differentiated in the later records. This again reinforces the idea that due to the early nature of the state-specific and concrete connotation of each term was not crystallized during the reign of the Valkhā rulers.

It is difficult to agree with the editors about the total absence of boundary markers as the key element suggesting the initial stage of land grants. ³⁶ Though we do not have any clear-cut demarcations, as evident in later grants, yet the perception of a fluid line in the form of either the river Narmada or any village or simply referring to the forest region or a forest deity (Vanavāsinī) as a boundary, is found in the charters. This suggests a concept of alienating the granted land from the other through a tangible marker and here the mention in one epigraph of the term simā along with a marshy or wet land ³⁷ (for example, in the grant of Bhuluṇḍa, year 57, the expression Ulladana sīmāyaṁ…. saha maṇḍala kachchhena brahmadeya) should also be taken note of. Gradually, in the later records of Rudradāsa in the year 67, we find that some concrete idea of boundary markers were evolving and thus it is stated that a field entrusted with the potter Āryyadāsa is situated in the south-western boundary of Dāsilakapallī. This in reality is a pointer to the embryonic character of the state as later examples of land grants with definite boundary markers may be seen as an attempt by the state at gradually organising the donated lands in such a way that there remained no chance for encroachment of any other plot by the donee. As a result, ownership or enjoyment of the donated land became more precise and secure and this was ensured by the state. To cite two examples: in about a little more than a 100 years from the grant of Svāmīdāsa, the Gunaighar (in Comilla area of south eastern Bangladesh) grant of Vainyagupta (c. 507 CE) exhibits a serious attempt on the part of the ruler to precisely mention the boundaries of the donated lands. ³⁸ More stringent boundaries could be perceived in the records of the Pāla rulers of Kāmarīpa (c. end of tenth or early eleventh century CE to twelfth century CE). ³⁹ The markers help to somewhat reconstruct the landscape of the rural settlements described in the charters and show how the boundary markers of the area can indicate a wide social cross section of people inhabiting a particular space. ⁴⁰ The total absence of land measurements, of course, points to the initial stage of the system of land grants in the region.

Moreover, unlike the format of the standard charters, no imprecatory verse is present but at the same time the rulers fear some kind of hindrance to the donee and so we have specific injunctions such as na vyahantavyam (grant of Bhuluṇḍa, year 59, l. 9) meaning the grant is not to be violated or transgressed, ⁴¹ na vyāsedhya (grant of Bhuluṇḍa, year 55, l. 9) implying no hindrance or prohibition to be caused to the donee ⁴² and pratishedhashcha na kāryah ⁴³ (grant of Bhuluṇḍa, year 50, l. 7) indicating that nothing should prevent the execution of the grant. ⁴⁴ Thus, these three terms used in three different charters of Bhuluṇḍa, are employed in the sense of a caution to any encroachers of the grants. We know that revenue rights were an important right bestowed on the Brāhmaṇa recipients of royal land grants. None of the charters refer to any revenue term except one. The charter of Bhaṭṭāraka records the grant of a plot of land along with paddy fields in the village of Suhasahanānaka in Aśvasśati pathaka to Brāhmaṇa Revati Śarmman of Vatsya gotra along with Udraṅga (s-odraṅga), implying that the latter was a source of income. Incidentally this was the first use of this privilege in the land grants of India. There is much debate among scholars regarding the actual meaning of the term. The meaning ranged from the share of produce collected usually for the king, additional tax, tax on permanent tenants, to ground rent and so on. ⁴⁵ In a recent study on Udraṅga ⁴⁶ it has been observed that grant of land together with Udraṅga was mostly confined to western India and Deccan. Reference to Udraṅga is not known in the inscriptions of Orissa and Assam. We have the mention of Udraṅga in a few grants from Bengal where it was used in the sense of a tax and a type of land. ⁴⁷ Normally in the later grants Udraṅga is always associated with another revenue term Uparikara which has been explained as additional taxes, minor taxes or tax paid by the temporary tenants. ⁴⁸ In a grant of Subandhu ⁴⁹ (c. 487 CE) from Bagh, Udraṅga is mentioned along with Uparikara. The absence of any mention of common tax names could be understood from the point of view that the rulers did not feel the necessity of separately mentioning the normal taxes. Taxation had not developed into a well-defined system with a complex apparatus to ensure its regular flow. It was only in the case of introduction of an unknown tax that the necessity to record the name was felt.

We know that the royal order regarding the grant of a village or plot/plots of land was not always issued by the king himself, but was often conveyed through an intermediary often called dutaka (literally messenger, but here executor) who was generally a high officer, sometimes even a prince. ⁵⁰ In the case of these charters the presence of a dutaka is not very regular. Moreover, what is noteworthy is that in the early charters, the dutakas, wherever they were present, were officially either an ārakshika (a police officer or the chief of the king’s bodyguards) or a pratihāra (officer in charge of the defence of the palace or the city gates). Only in one grant of Nāgabhaṭa of the year 134 do we find a bhāṇḍāgārika (officer in charge of the treasury or royal store house) as an executor (dutaka). What is significant is that the dutakas of the early copper plate charters were all security personnel and thereby the king’s trustworthies. Moreover their names are indicative of their autochthonous origin. So we have persons with names such as Gomika, Haṭaka, Āddyakarṇṇa, Varāha, etc., as dutakas. From the time of Mahārāja Rudradāsa ⁵¹ (year 68, that is, 388 CE) there were other persons who acted as dutakas but they did not have any precise official designation. They had the honorific Bhaṭṭi added to their name, which became Sanskritized. Consequently, we now have as dutakas Bhaṭṭi Isvaradatta, Bhaṭṭi Rudrada˜sa or simply Jayanatha. The addition of the title of respect Bhaṭṭi may suggest that brāhmaṇas were given the post of dutaka. The irregular mention of dutaka suggests a kind of fluid administrative setup where the presence of an executor with respect to grant of lands/villages was not an essential requisite. Interestingly, the first plate of Bhuluṇḍa records the execution of the deed in the presence of five artisans (pañcha kārukam cha samukham). ⁵² Pañcha kārukam could mean five groups of artisans, namely, goldsmiths, blacksmiths, brassmiths, carpenters and stone masons. ⁵³ It is said that the artisans were the descendants of the five sons of Viśvakarmā. The question that naturally comes to our mind is why was it important to mention the artisans as witnesses? Were the artisans an important social group whose witness merited record in an official document? Was it something like pañchakula? ⁵⁴ That artisans were important in this region is evident. We have an entire village in the name of the lohakāras, that is, iron mongers and the lohakārapalli is referred to as a purva-bhujyamānaka, that is, the land was formerly under the enjoyment of Lohakarapallika in a grant of Svāmīdāsa, year 65 (384/85 CE). ⁵⁵ But after this instance, none of the records mention groups of artisans. Did the pañcha kāru as an important socioeconomic group lose its relevance so as to merit complete silence in the later records? We have no answer.

The use of the expression bhujyamānaka (being enjoyed) is significant. In this grant of Mahārāja Svāmīdāsa one plot of land which was under the enjoyment of Lohakārapallikā was transferred to an individual brāhmaṇa called Matujja. We have such earlier cases also where Mahārāja Bhuluṇḍa, (year 54, c. 373/374 CE) granted villages which had been earlier brought under the enjoyment (purva-bhujyamānaka) of a donee, in this case god Bappa Piśāchadeva and then re-granted to the same deity with an exalted status of devāgrāhāra. The term bhujyamānaka is not ubiquitous in the charters of early India and perhaps the earliest reference to this term has been found in the copper plates issued by Mahārāja Bhuluṇḍa. However, it needs to be mentioned here that the Yavatmal plates of the Vākāṭaka ruler Pravarasena II (year 26) renews an earlier grant of land which was already being enjoyed (pīrvva-bhujyamānikā bhīmiḥ). ⁵⁶ The use of the term bhujyamānaka could also be seen much later in the second half of the seventh century CE in the Ashrafpur copper plate grants of Devakhadga belonging to the Khadga dynasty of south-eastern Bengal. ⁵⁷ Thus, it appears that ownership of the land was retained by the king and the right of enjoyment was transferred from one donee to the other. ⁵⁸ This also signifies different kinds of land relations.

Though in a nascent stage, the mention of various territorial divisions in the charters is quite remarkable. Consequently, we have rāshṭra, bhukti, pathaka, āvāsa. Again for villages we have such name endings as padraka, garttā, ānaka, vāṭaka, pallikā and so on. ⁵⁹ Importantly, except for garttā all the name endings for villages are common to the Vākāṭaka records suggesting influence from the Vākāṭaka territory. Garttā (lowland), however, is representative of local flavour. The suffixes denoting rural settlements are more or less synonymous suggesting small village settlements. Though Palli is ordinarily taken to be a ‘tribal’ ⁶⁰ village, it was not invariably such a settlement, as B.D. Chattopadhyaya argues. ⁶¹ It could also mean a hamlet in the context of a peasant village. In the case of these charters, two references to Pallikā and one to Palli have been found but it is not possible to identify them either as a peasant village or a settlement of tribal people only on the basis of their names such as Palāśapalli, Lohakārapallikā and Dagdhapallikā. Lohakārapallikā was definitely a settlement of ironsmiths and as mentioned earlier, they were also in possession of land. With the penetration of the agrahāra system of land donation resulting in settled agricultural practices, palli/pallikās would eventually be transformed into peasant villages or administrative units. In one case Dasilākapallī is mentioned as a rāshṭraka (Bhuluṇḍa, year 47) in one inscription and pathaka (Rudradāsa, year 67) in another, both being territorial units without any properly identified hierarchy. The fluidity of usage of administrative or land vocabulary from the Guptas and Vākāṭakas comes through when we come across such divisions which contain both the village name ending and the administrative name such as Udumbaragarttāpathaka or two divisions together such as Navarāshṭrakapathaka. So, it was possible to use any of the terms for denoting a territorial unit. This is a definite marker of early state where administrative divisions are not yet well-defined. Though other divisions are known, the editors feel that āvāsa/vāsa occur for the first time as a territorial name suffix in our charters. ⁶² The village names are quite interesting with pronounced references to flora, bird names and therefore are instructive of the environment of the granted land bearing a strong indigenous element. As examples we have Śalmalīpadraka, Palāśapallī, Kharjurika, Nimbapadraka, Pippalojjharā, Arjunapanktika, etc., related to flora and Kokilavasaka and Kukkutananka based on bird names. As a conjectural piece of evidence it may be suggested that there was some physical basis to these names and the landscape of these villages was dominated by the flora that their names indicate. Thus, Śalmalīpadraka at one point of time must have been strewn with silk cotton trees which were also economically viable. The landscape of the kingdom as understood from a study of these charters could have boasted of woods.

A look at the nature of donation by these rulers indicates that during the reign of Bhuluṇḍa, the first ruler, most of the grants known as devāgrahāra were donated for the services of the gods and goddesses, that is, temple rituals. ⁶³ That the temples were already active is clear from the fact that the charters speak of persons related to temple as deva-karmāntika (a temple officer), deva-parichāraka (commonly servant of the temple, perhaps personnel who look after the temple), deva-prasādaka (perhaps meaning one who sustains himself through the favour of the god, in other words an official in the temple), deva-karmina (an officer in charge of the affairs of the temple or a temple priest) and devakīya-karshaka (cultivators associated with temple lands). These are of special significance in the context of the role of religious establishments in undertaking agricultural production. ⁶⁴ The mention of these officials/personnel related to temples with varying designations perhaps indicates the beginning of a temple organization in the region. The purpose of these grants was always the performance of Bali, Charu and Satra rites and providing perfume, incense and garlands (gandha-dhīpa-mālyopayogādishu) to these temples. A reading of the inscriptions would show that during the reign of Mahārāja Bhuluṇḍa (c. 367CE–c. 379 CE) three temples refer to Nārāyaṇa, four to the autochthonous deity Bappa Piśāchadeva, two temples to mother goddesses (Bhagavatinām) also called mahāmātriṇāṁ in the same plate (Great mothers) (Bhuluṇḍa, year 50) and one to God Mahāsenadeva. Regarding these gods and goddesses a few points could be noticed. Mahārāja Bhuluṇḍa refers to himself as the devotee of Nārāyaṇa and he also installed the temple of Nārāyaṇadeva at Valkhā adhishthāna, that is, the capital city. Consequently, his personal association with Nārāyaṇa is established. It is noteworthy that in the first inscription of Bhuluṇḍa dated Gupta year 47 (CE 366–67) and even in the others a long account of Vishṇu and his incarnations are given. ⁶⁵ According to Michael Willis this explains the relationship of subordinate kings to the Vaishṇava cult of the Gupta court. ⁶⁶ With respect to the mother goddesses we find that in the first instance he established ‘Great Mothers’ at a place called Navataṭāka. According to B.D. Chattopadhyaya the prefix Maha’ indirectly refers to the exalted status of the goddesses which justifies their receiving royal patronage. ⁶⁷ Perhaps Chattopadhyaya is justified as another grant given to the devakula ⁶⁸ (temple) of mother goddesses, established by Pāśupatāchāryya Bhagavat Lokodadhi and not the Mahārāja himself, so the prefix ‘Mahā’ is missing. In the case of Lord Mahāsenadeva (Svāmī) also, the term used is asmābhiḥ pratishṭhāpitaka, that is, in this case also the temple is established by the Mahārāja. That the ‘new rājā’ was not associated with the indigenous deity Bappa Piśāchadeva is evident from the fact that three of the grants (years 51, 54 and 57) to this deity were given by Bhuluṇḍa, only at the request of Bhojikā Bhaṭṭa Bandhulā who established the temple of Piśāchadeva at the capital city of Valkhā. At the request of Bhojikā Bhaṭṭa Bandhulā, they were given the status of devāgrahāra, ⁶⁹ which implied many facilities, and the elevation of a local deity to the status of other deities of the brahmanical pantheon with the usual rites and rituals. This may seem to be what Chattopadhyaya calls the ‘brahmanical mode of appropriation’. ⁷⁰ Moreover, the establishment of a temple by a lady is also rare. Piśācha relates to an evil spirit. Perhaps by worshipping the lord of Piśāchas, here Bappa perhaps meant father or lord, that is, Valkhā was being protected from the malevolent eye. Patronage legitimised power and helped the emergence of a stratum of ruling elites, more importantly in this case a woman. Another grant of Mahārāja Bhuluṇḍa, year 59, donated at the request of one Innapāda also talks about the installation of Bappa Piśāchadeva not in Valkhā but in a vishaya called Narmmadāparapāra (Narmmadāparapāra-vishaye iha-aiva pratishthapitaka-svami-Bappa Piśāchadeva). We have no idea where this vishaya could be located but the interesting fact is that the vishaya is named after the river Narmada. What is worth noting is that an erstwhile brahmadeya-kshetra of the village Garjanānaka, was granted to the god.

It is significant to note that the choice of the other deities such as Mātrikā and Kārtikeya was perhaps not random but thoughtfully selected. A later grant of the Chālukyas, the Amudalapadu plates of Vikramaditya I, dated around 660 CE shows that these gods had some specific roles within the royal cult, variously augmenting the favour, prosperity and power of the royal family. The record begins with an account of how the founders of the Chālukya dynasty increased in favour through the grace of the Sapta Mātrikā (seven mothers), obtained prosperity through the protection of Kārttikeya, and overpowered enemies by the very sight of the Varāha emblem that the Chālukyas acquired by the grace of Nārāyaṇa. ⁷¹

(saptamātribhirbhivarddhitānāṁ kārttikeyaparirakshaṇaprāptakalyānaparaṁparāṇaṁ bhagavatnārāyaṇaprasādasamāsādita varāhalānchhanekshaṇkshaṇavasikritaseshamahibhritāṁ).

Holding centre stage was Nārāyaṇa, the god by whose grace the kings enjoyed power. For example Mahārāja Bhuluṇḍa, too, as his first record reads, was the dāsa of Nārāyaṇa, his Svāmī. Another significant aspect of these grants to temples demands our attention. In most of the cases the Paśupatas are said to be attached to these temples. We know that the Paśupatas were a Śaivite sect. But here they are present in the context of the temple of Nārāyana, as well as Bappa Piśāchadeva. Paśupata was one of the earliest Śaiva sects prevalent in India and its popularity in western India is well known. ⁷² Their presence here implies that sectarian differences did not surface in these areas during the fourth century CE and hence they are present in a Vaishṇava context. The centrality of Lord Nārāyaṇa in the kingdom of Mahārāja Bhuluṇḍa, is thus beyond question. Though it was just the beginning of cult formation in the region yet it seems that Nārāyaṇa surpassed the local god Bappa Piśāchadeva because of the ruler’s association with him. The king as the patron was instrumental in giving centrality to the cult of Nārāyaṇa and not otherwise. Unlike in Orissa the dominant autochthonous deity was not the tutelary deity of the Valkhā kingdom. ⁷³

The installation of these images/temples required regular services, which in turn demanded resources which came from grants of villages and lands. In all the cases we have a combination of bali, charu and satra, which was a part of the pañchamahāyajña, and offerings of sandal paste, incense, flowers and garlands which formed a part of the ritual of worship. It is difficult to understand its implication in a temple setting. The pujā and the yajña were entwined with the services required for deities in the form of temple servants (devakarmins) and members of other sects. It has recently been shown by Willis that when these three are mentioned in a pujā context, they are not simply parts of the pañchamahāyajña. According to him bali, charu and satra ‘were a discrete and well ordered triad, quite separate from other types of offerings. Priests have carefully and deliberately moved sacrifices from the domestic environment to the temple and attracted funding to support these rites in a new location.’ ⁷⁴ Satra activities took place at the temple, they were understood to be part of the pujā of the deity and that the charitable provision involved clothing as well as food. It is true that epigraphic evidence shows that in the beginning this concept was associated with Nārāyaṇa. Thus, from c. 367 CE to c. 379 CE temples in the Valkhā kingdom received considerable patronage. ⁷⁵ One must remember that this was the beginning of temple-building activity in north India.

Now let us turn our attention to copper-plate charters of other rulers. What is surprising is that from about the time of the next ruler Svāmidāsa whose earliest grant dates to c. 383 CE till the time of Nāgabhaṭa (c. 454 CE) none of the charters donate land to any temple. Did the temples lose relevance to the succeeding rulers? Perhaps not, as the beneficiaries of the grants are all brāhmaṇas and among the brāhmaṇas a large chunk were probably priests. They were surely offering pujā to the deities. But here the kings chose the brāhmaṇas as they could realize that giving grants to a temple meant the prosperity of the institution of temple whereas giving grants to individual or groups of brāhmaṇas meant elevation of an individual or a group from a simple purohita to a land holder in some cases to a landed group. Consequently, there is a prominent visibility of the brāhmaṇas in the form of land grants. Brāhmaṇa donees were introduced to the village population through these grants. This was done as the ‘new political elites’ needed legitimation and this benefited the brāhmaṇas. Kulke has shown that regional kingdoms systematically settled Brāhmaṇas near the political centres of the kingdom towards the end of the first millennium CE and these brāhmaṇas acted as ideological and administrative specialists. ⁷⁶ This process began in the Valkhā kingdom in the middle of the first millennium CE when brāhmaṇas were entrusted with the task of a dutaka. The rural landscape underwent changes. Brāhmaṇas emerged as the dominant caste group in the brahmadeya villages. Plots of lands which were originally entrusted to non-brāhmaṇas, such as, potters, merchants or belonging to a village of the iron mongers (lohakārapallī), were given to the brāhmaṇas. Thus, we have evidence of individual ownership of land by occupational groups other than the brāhmaṇas. From the charters it is not possible to discern whether these people were in possession of land before the rise of the Valkhā state but it could be categorically said that in about fifth or early sixth centuries CE, increasing evidence of individual ownership from other areas could be inferred from the copper plates in the form of boundaries of a given piece of land. A classic case could be the Gunaighar (near Comilla in Bangladesh) copper plates of Vainyagupta issued in the Gupta year 188 (507 CE) where we have reference to a large number of privately owned plots. ⁷⁷ However, gifts of villages to the brāhmaṇas led to a significant proliferation of brahmanical settlements. That new settlements were coming up is evident from the names of places as navataṭāka and navarāshṭraka. Moreover there was a migration of brāhmaṇas too. Though in these charters we have no reference to large-scale migration from other places but we have numerous references to a group called Āryya Chaturvaidyapādas, that is, proficient in four Vedas, who resided at Valkhā and were given villages on both sides of the Narmada. This group of brāhmaṇas became wealthy and with the possession of so many villages turned into landed elite. To make use of their newly found wealth, they must have moved to new locations. In another instance it is said that a group of brāhmaṇas which was already a part of another agrahāra was now given the habitation site of a village as per brahmadeya norms. So they moved from their original habitation to this one. As the group of brāhmaṇas was large, the establishment of the habitation site implied the termination of cultivation from large tracts of land. The expression used is krishṭāvasannakagrāma-dhāna and this is the meaning given to the term by the editors. ⁷⁸ If it was indeed so, then it is rather surprising that cultivable lands were converted to sites for habitation at a landscape that did not boast of a fertile topography. Consequently, here is a case where even at the cost of losing agrarian resources the rulers felt the desperate need of establishing a brahmanical settlement. K.M. Shrimali, ⁷⁹ however, has given an alternative reading to the term. According to him ‘it would perhaps be more appropriate to read the concerned term as āvāsana rather than avasannaka, which would mean a dwelling or living on the boundaries of a village’. The problem however is that a close look at the inscription itself would show that the term is avasannaka and not āvāsana. ⁸⁰

We might divide the plates into two phases; a single king Bhuluṇḍa dominates the first phase where the religious context of the charters becomes important. We notice a symbiotic relation between the different sects of Puranic brahmanism. Thus, the Pāśupatas, Pāśupatācharyas, Bhagavachchhishṭas, Mantragaṇāchāryas figure together in the charters. ⁸¹ Perhaps the new political elite were trying to grapple with a contemporary situation where transformation from Vedic brahmanism to the sectarian pluralism of the period was in the making. We have seen that recording donations on a copper plate was at an experimental stage. In the second phase we have a set of four rulers who were aware of the virtues of Bhumi dāna, especially to brāhmaṇas. There was perhaps a change in the belief system, where temple donation was relegated to the background and donation to brāhmaṇas became primarily important. The term belief is not taken here in the sense of religion or faith but it means the belief in the method of procuring political supremacy. In the case of Bhuluṇḍa, temple donation meant a place in the brahmanical monarchical system following his overlord the paramabhaṭṭāraka, identified with the Gupta monarch where similarity in religious ideology with the lord would give him a firm footing in his newly acquired territory. On the other hand, for Bhuluṇḍa successors, the Valkhā kingdom was already created and the need of the hour was to make the brāhmaṇas, the highest social group as well as increase the resource base through agrarian expansion.

Though nearby, Bagh as a Buddhist site looms large from about the middle of the fifth century CE; however, in the inscriptions, we have no indication of any kind of presence or patronage to any Buddhist establishment. Along with the Brahmanical gods and goddesses, worship of dominant autochthonous deity in the form of Bappa Piśāchadeva could be seen. The sectarian elements were not predominant and there was no incompatibility between local cult and brahmanical sacred centres. The shift in patronage from temples to brāhmaṇas exhibits early forms of political–social transformation. The actual administrative structures inevitably contained the carry-overs of the Gupta and Vākāṭaka administrative systems. An attempt to introduce terminologies related to administration of land used by the two dynasties could be seen. In fact several administrative terms have a distinct connection with the terminologies found in the Vākāṭaka records, this again speaks of patronage to brāhmaṇas. In the choice of language for the grants they followed the Gupta tradition, thereby indicating a north Indian orientation. The use of the Gupta era speaks of the possible penetration of the Gupta power and cultural matrix in this area, without however leading to a direct incorporation of this area in the Gupta realm.

It appears from a reading of all the charters that this newly emerged local polity did not seek validation through linkage with a respectable ancestry though it was in a state society stage. This is exceptional as various studies on political processes have shown that the creation of genealogies or establishing linkages with mythical rulers was the common trend of new political powers. ⁸² The Valkhā state was evolving and we have gradual visibility of the brahmanical norms. The evolving character of the state is also evident from the gradually discernible presence of state officials in the charters. Strengthening central power and increasing the number of appointed functionaries, which means a gradual process of bureaucratization, is characteristic of all early states. Agrahāra as an institution became an agency of the ruler which could be used for legitimacy. The study of the charters suggest that it is not possible to posit the Valkhā state in any one of the much accepted three types of the early state as suggested by Claessen and Skalnik. The Valkhā state was truly a combination of a typical early state and an early state judged to be transitional. While we have little mention of trade and markets, state ownership of lands was gaining ground. Therefore, we have the enjoyer of lands who could be easily displaced from his possession with his right of enjoyment transferred. Though specific punishment is not mentioned yet caution is pronounced regarding any kind of hindrance to the donee or in the execution of the grant. The most ubiquitous feature of the early state is the emergence of a two-class system of the ruler and the ruled. Here, through the prism of copper-plate charters, we perceive a three-tier system, where the king is the apex political authority, then we have a set of people who are known as bhujyamanaka (enjoyer of the land), usually brāhmaṇas and then the peasants and other social groups. A new set of temple officials were created whose designations are instructive of two functions—service and favour. Thus, we have persons working in and for the temple who are known as deva karmin, deva parichāraka, deva prasādaka and so on. The Valkhā state was in the process of articulation of the modes and norms for granting lands. So, the grants are stated to have been made as per the krama (custom, rule sanctioned by tradition) or nyāya (a general or universal model) evolved for the purpose and the stipulations are of an appropriate nature (uchitayā). This demonstrates that there were customs and usages which derived their validity as much from the recognition accorded by the existing political authority as from the force of practice coming from the past. The early state of Valkhā grew out of its own internal dynamics and aspirations for monarchical political structure acted as an added incentive for the rise of the early state of Valkhā.