From Kingdoms to Transregional States: Exploring the Dynamics of State Formation in Pre-modern Odisha

Introduction

This article seeks to investigate the history of the emergence of early kingdoms in the middle of the first millennium ce, followed by the more complex state systems, which were spatially spread over more than a subregion/region (maṇḍala) from around the middle of the eighth century onwards (see Figure 1). ² By the early tenth century, larger political enterprises or transregional states spanning substantial parts of present-day Chhattisgarh, Andhra Pradesh and Odisha, or several subregions/regions, were made possible. The trend continued into the middle of the second millennium under the Later Eastern Ganga and Gajapati dynasties. The historical processes leading to these developments and the new requirements that they entailed—such as an elaboration of the structure of administration and legitimation—and an understanding of their underlying social foundations will constitute the subject matter of this endeavour.

OPEN IN VIEWER

Figure 1.

Source: The author.

States under the Bhaumakaras (736–930 ce) and Somavamsis (930–1112 ce), and more particularly the Later Eastern Gangas and Gajapatis, indicate a certain correspondence with imperial formations insofar as they were transregional states spread over territories through conquest, which accommodated a measure of built-in unevenness in terms of developed, developing and underdeveloped spatial segments. The simultaneous emergence of a region-specific caste system, regional language and script, with an imprint of neighbouring regions, their use for official communication and a regional style of temple architecture, imbibing elements from the different subregions, combined to give the emerging transregional state the necessary coherence and stability. Largely owing to the emergence of a regional language and script and their convergence with the territorial spread of these states, they have been also characterized as ‘vernacular polities’. ³ This article hopes to contribute towards a comparative perspective of the state across regions in pre-modern India.

As in the other regions of South Asia, state formation in Odisha at different junctures was constituted by the complex interplay of multiple historical forces. In terms of the trajectory and chronology of the forging of kingdoms, or later transregional states, it may demonstrate variance compared to the neighbouring regions, forcing us to discern its specificities. The gradual coming together of localities (janapadas) and subregions (maṇḍalas) at different stages of state formation inevitably imposed interactions, overlaps and intersections, and a wider network of cultural transactions with other regional societies in the subcontinent. ⁴ During the early historical period, the evolution of state–society was spatially segmented across localities and largely centred on the coast. ⁵ In the course of monarchical state formation and the transition to complex state–society from around the middle of the first millennium, chiefs desirous of becoming classical Indian kings and brahmanas who collaborated with them in the said project had to adapt to and integrate numerous autochthonous groups and local deities through a long-drawn-out process of acculturation. The region witnessed the influx of a large number of brāhmaṇas from different parts of North India. ⁶ Royal patronage within the emerging ruling lineages extended to numerous deities and their believers, influenced as much by exigencies of political requirements as by convergences and cultural plurality. ⁷ Brahmanization of the indigenous population and the tribalization of Brāhmaṇic culture moved in tandem all through, leading to the shaping of shared cultural spaces. ⁸

Towards the Forging of Transregional States

Odisha is a partly mountainous region, being part of the Eastern Ghats and an extension of the Chota Nagpur Plateau and the Central Indian complex. It is broadly characterized by thick forests and river valleys that open into the coastal plains. The mountainous nature of large parts of the region influenced the proliferation of local and subregional polities in the early medieval period. Disparate localities and subregions experienced internal transformation from around the middle of the first millennium ce. During the fourth–seventh centuries, there was a horizontal spread of state–societies—the rise of the Matharas, Vasisthas and Pitrbhaktas, Early Eastern Gangas, Nalas and Vigrahas in what is today’s coastal northern Andhra Pradesh and adjoining parts of Odisha, for instance. These early medieval polities evolved through change coming from within as well as cultural transactions with other regions. The varied subregions were known as Kaliṅga, Kōngoḍa maṇḍala, Dakṣiṇa and Uttara Tosali, Khiñjali maṇḍala, Kodalaka maṇḍala and Khijjiṅga maṇḍala, among others. The term maṇḍala seems to represent the idea of a subregion and is encountered in epigraphic sources not only in Odisha but all over the east coast during this period. Continued manifestation of the process across spatial segments and also away from the littoral was registered between the seventh and tenth centuries. Most of the ruling families, such as, the Eastern Gangas, Sailodbhavas, Bhaumakaras and the Bhanjas of Khiñjali and Khijjiṅgakoṭā, had humble local origins. Their origin myths mostly point in that direction and unambiguously suggest that they were increasingly conforming to a pan-Indian tradition of explaining away their origins by invoking mythical pasts. ⁹ Expressions such as astadasatavika-rājya, Vindhyāṭavinātha and Sakala Goṇḍramādhinātha used by different rulers and lineages unmistakably suggest the local presence of autochthons across spaces. Whereas the term aṭavi stood for the tribes, Gondrama is said to have been derived from the Gond tribe. References to the Sabaras and Pulindas in the inscriptions of the region further affirm it. ¹⁰ Interestingly, the same period also witnessed the formation of a large political enterprise under the Bhaumakaras of central coastal Odisha, who, after combining Kōngoḍa and both the Tosalis, the lower valleys of the Rushikulya and Mahanadi, moved into the hinterlands. ¹¹

The rise of monarchical states and socio-economic transformation through the subregions appear to have been simultaneous processes. Agrarian and social changes were essential for the extension and consolidation of political authority. The migration of brāhmaṇas through land grants of various types—individual, collective and scattered—especially in the tribal belts, or the maxim of bringing virgin land under cultivation (bhūmichhidranyāya) opened these areas to peasantization and concomitant changes, including the emergence of occupational groups/castes, hāṭṭas (exchange centres) and towns. It quickened the social and economic integration of wider areas across localities and subregions, and gently helped in the creation of a regional agrarian base by the turn of the millennium. The rise of janapadas (agrarian localities) was an ongoing process. The coming together of a number of janapadas would have facilitated the rise of maṇḍalas. The Matharas, Vigrahas and Manas during the fourth–seventh centuries were able to sustain smaller kingdoms with fewer kinds of taxes, fewer officials and shorter longevities, ¹² whereas the agrarian spread and growth by the ninth and tenth centuries made it feasible to have territorially expansive states, with greater longevity and evolved political structures. The Bhaumakaras and Somavamsis provide good examples of this in Tosali (central coastal Odisha) and Dakshina Kośala (parts of modern Chhattisgarh and western Odisha), respectively. The Bhaumakaras (eighth–ninth centuries) made the earliest effort at territorial expansion and integration of larger areas beyond their homeland (janakabhū). They united the two Tosalis and Kōngoḍa maṇḍala and extended their sway over some of the other maṇḍala polities.

It may be pertinent to mention that the intermediate stage between the early kingdoms and the Bhaumakaras is represented by the Sailodbhavas in the littoral and the Sulkis of Kodalaka maṇḍala in the interior of Odisha, who also happened to be one of the sāmanta intermediaries of the Bhaumakaras. While the Sailodbhava records, given the historical background of the region, yield a fuller and more coherent picture of developments, the latter owing to its peripheral spatial location and late transition to the historical phase in the subregion, largely owing to the presence of the autochthons, provides a varying pattern. Their records initially mention only the bhogika (an intermediary) and purogan (inhabitants of a settlement) and gradually move on to include the viṣayapati, pustakapālas and karaṇa, together with the janapadan (inhabitants of the locality or common peasants), and register the growing complexity of society. ¹³ The Sailodbhava inscriptions, representing a relatively settled area in contrast, refer to sāmanta, mahāsāmanta, mahārāja, rājanaka, rājaputra, antaraṅga, daṇḍanāyaka, uparika, viṣayapati, tadāyuktaka, vyavahārin, karaṇa and cāṭa-bhaṭa. ¹⁴ Their land grants, besides referring to graded officials, rural settlements and a land measure (timpira), also invoke comparisons of royalty with epic–puranic heroes and bear allusions to their performance of Vedic sacrifices. ¹⁵ These details in the inscriptions gradually expanded with the coming of the Bhaumakaras and their successors.

The transformation of the tribal chiefs to institutionalized rājas was synchronous with the gradual universalization of the local deities. Kulke posits that the problem was largely addressed through the royal patronage of autochthonous deities as family or tutelary deities of the emerging local or subregional states. ¹⁶ In the process, the rājas tried to appropriate the territoriality of the deities and the loyalty of the subjects. With the passage of time, the deities were raised to a higher level of ritual elaboration along with the political consolidation of the rajas. However, the visual aspects of the deity, their tribal priests and the basic hymns (mantras) and rituals were retained, while introducing and superimposing brāhmaṇas and Brāhmaṇic rituals in his/her worship, facilitating the integration of the deity with the pan-Indian pantheon. Consequently, the coexistence of the popular and Brāhmaṇic forms helped in shaping both the worlds. Cultural communication impacted both ends of the spectrum. It helped to tie the king’s legitimacy both downwards and sidewise. ¹⁷

The Bhaumakaras who began their rule around ad 736 ruled from Jajpur (Guhadevapataka) in central coastal Odisha and gradually expanded their influence. Several maṇḍalas in the river valleys and hilly hinterland of coastal Odisha became a part of their circle of sāmantas. These lineages included the Sulkis of Kodalaka maṇḍala in the Brahmani valley (Talcher–Dhenkanal area), Bhanjas of Khiñjali maṇḍala in the Boud valley (Boud–Khondmals), Tungas of Yamagarta maṇḍala and the Nandodbhavas of Airavata maṇḍala, both of whom succeeded the Sulkis. The Mayuras of Bonai maṇḍala (parts of Sundergarh), with varāha titles, and the Adi Bhanjas of Khijjiṅga maṇḍala (parts of Mayurbhanja and Keonjhar districts) emerged on the political scene almost contemporaneously and acknowledged the suzerainty of the Bhaumakaras. Most of these dynasties used the Bhauma Era (which is generally recognized to have begun in 736 ce) in their land grant charters, adorned themselves with epithets such as ranaka and mahāsāmantādhipati (the great intermediary chief), and proclaimed having obtained the privilege of pañchamahāśabda (five different sounds). There are a few Bhaumakara grants that were made in Kodalaka and Kongoda maṇḍalas, which suggest the exercise of their authority in these territories. The combined testimony of such evidence suggests their sāmanta or intermediary status in the face of the rise of a powerful overlord. Interestingly, these states, while acknowledging the lord’s pre-eminence, exercised independent power in their domains largely owing to their local origins. They built independent legitimation structures as can be seen in the case of the Sulkis, issued their donative charters without reference to the overlord and at times, with the use of superior titles such as mahārajādhiraja and paramabhaṭṭāraka, they even provoked the dominant power to take punitive measures. ¹⁸ The later titles were privileges of superior overlords, and their simultaneous use with lesser titles unmistakably suggests the vacillating relationship between the kings of maṇḍala states and their overlord. However, this arrangement seems to have been mutually beneficial. The acceptance of the Bhaumakara lordship allowed the maṇḍala states to ward off the danger of political extinction for some time, while it whetted the Bhaumakaras’ political aspirations to the chakravartin status.

The Somavamsis, successors of the Bhaumakaras, came to power in Dakshina Kosala following the Panduvamsis. In the early stages of their history, they remained confined to that region, but by the middle of the tenth century (around 930 ce), they had conquered Khinjali maṇḍala, which was a sort of buffer between Dakshina Kosala and Tosali/Utkala, and extended their control over the coastal territories of the Bhaumakaras. Consequently, they forged a regional polity for the first time, bringing together several subregions (Kongoda, Tosalis/Utkala, Kodalaka, Khinjali, among others) including parts of western Odisha and Chhattisgarh. It may be of interest that Mahasivagupta Yayati II, in his Jatesinga Dungri copper plate charter, describes himself as ‘the self-chosen husband of Kalinga, Kongoda, Utkala and Kosala and as the wind in carrying away the cloths of Gauda and Radha…’ ¹⁹ The last two territories represent parts of modern Bengal. The fact that they emerged in interior Dakshina Kosala and moved on to forge a supra-regional state after the conquest of the coastal plains complicates the usual notion of core and periphery. Nevertheless, they were not alone in charting such a successful venture. The contemporary histories of the Hoysalas of Karnataka and Kakatiyas of Telangana also provide good instances of it. During their rule, deśa, maṇḍala, khaṇḍa, bhukti and viṣaya constituted the administrative divisions and there was an expansion in the list of state functionaries. During the same period, we do not come across the maṇḍala kingdoms, indicating their integration at least for some time within the regional polity. The success of the political experiment of the Somavamsis was premised on continuous agrarian expansion across the maṇḍalas, with all that it entailed, including the emergence of an agrarian region. These tendencies were accentuated under the Later Eastern Gangas who came from the south, that is, northern coastal Andhra Pradesh, at the beginning of the twelfth century and continued into the middle of the fifteenth century as the next dominant supra-regional power.

The Later Eastern Gangas of Kaliṅga rose to the status of the ‘Imperial Gangas’ of Odisha during the reign of Anantavarman Chodaganga (1078–152 ce). His kingdom extended from the river Bhagirathi (Hooghly) in the north to the Godavari (Gautami Ganga) in the south. However, their upward trajectory can be traced from the times of Vajrahastadeva III, who ascended to power in Kalinga in 1038 ce. He was the first ruler in the family to assume imperial titles such as mahārajādhirāja (the great king of kings), paramamaheśvara and paramabhaṭṭāraka. He seems to have declared his independence from the Cholas and the Somavamsis, entered into diplomatic and matrimonial ties with the Cholas and the kingdom of Dahala, and went on to create a new power structure based on the Nāyaka system. It is on this foundation that his successors moved on to consolidate the state. Chodaganga after the conquest of the Mahanadi delta shifted his capital from Kalinganagara (Mukhalingam) to Kataka (modern Cuttack), which was more central to the newly acquired territories. Administratively, the Nayaka system was strengthened in their homeland. The Nāyakas were crucial to his expansionist designs and their numbers expanded conspicuously during his reign. ²⁰ It needs mention that the institution was socially inclusive and cut across varṇa lines. At the centre, the Pañca-Pradhana (five ministers) including the Mantrin, Purohita, Senāpati, Yuvarāja and Dauvārika were instituted, which was twice extended to sixteen and twenty under Bhanudeva I and Bhanudeva III, respectively. Ministers were known as Patras and even as Mahāpātra or Brihatmahāpātra. Subsequently, there was territorial reorganization with the introduction of the dandapatas (provinces) by Bhanudeva I in the Ganga territories. This signalled the movement towards centralization against the semi-autonomous maṇḍala system. Parikṣās or governors were appointed from the centre and were transferable and accountable to it. ²¹ These initiatives remind one of the administrative reorganization and vertical spread of the state under the Cholas in the south during the eleventh century. ²² Interestingly, Kaliṅga remained a separate province and its governors were known as Kaliṅga Parikṣā. They controlled the powerful Nāyakas and linked them with the central government. The kings used significant epithets such as Vijigishu and claimed to be performers of hiraṇya-garbha and tulā-puruṣa mahādānas in their donative copper plate charters, providing us glimpses of their self-image. ²³

From the basic dependence on the Kalinga/Telugu warriors organized by the Nāyakas, with the passage of time coastal Odisha too contributed to the formation of their military might. Besides the elephants, the cavalry too was lavished with state patronage, and they are liberally incorporated in the sculptural art at Konarka. The forts at Jajpur, Chatia, Choudwar, Cuttack and Sarangagarh (Baranga) were built to protect the Mahanadi delta, the fertile core of the Odishan possessions. In the south under the Gangas, Godavari became the boundary marker, while later under the Suryavamsi Gajapatis, the Odishan southern frontier moved to the river Krishna and beyond, albeit temporarily, with the Godavari–Krishna delta being the bone of contention between the Odishan rulers and Vijayanagara. The transregional state of Odisha did, in fact, encompass territories across regions in Chhattisgarh, Bengal and Andhra Pradesh at different points in time between the coming of the Somavamsis and the Gajapatis (tenth–fifteenth centuries), leading to the making of a territorially expansive state. Conventionally, expressions such as the first regional state under the Somavamsis, the Ganga Empire and the Imperial Gangas have been continuously used, whereas more recent usage would suggest the use of concepts like supra-regional or transregional state to be more appropriate. ²⁴ Transregional states may be situated between usual kingdoms and the celebrated states as empires.

It may be more appropriate and useful at this juncture to look into the ideological and social foundations of these emerging polities and their anatomies, rather than getting distracted by issues surrounding the typologies of state.

Brāhmaṇas and Brāhmaṇic settlements, Vedic–sastric–epic–purāṇic ideas, vihāras, maṭhas and temples provided the necessary ideological coherence to the evolving sociopolitical order through time. It must be mentioned that Buddhism flourished in the Sirpur area of Chhattisgarh and coastal Odisha at least up to the turn of the millennium. Sculptural narratives on the early temples such as at the Parasurameswara group of temples at Bhubaneswar and Sirpur, representing episodes from the epics and purāṇas, and the early inscriptions provide ample evidence of the spread and internalization of sastric-epic-purāṇic ideas and imageries. ²⁵ The conception of royalty, their self-images and their association with and desire to be identified with deities and temples so as to touch a chord with believers is also epigraphically attested to. Epithets such as Kośalādhipati or Kośalendra, while implying identification of the rulers of Dakshina Kosala with their territory, also sought to identity them with the subregional deity in the temple at Baidyanath known as Kośaleswara. Comparable structures of legitimation can be seen across the various subregions. ²⁶ These developments finally culminated in the cult of Jagannatha, with its vertical and horizontal linkages, under the Later Eastern Gangas and Gajapatis. ²⁷

The cult of Jagannatha appears to have been consciously supported to appropriate the deity’s territoriality in the newly conquered region, as well as to provide an alternative to Liṅgarāja who was patronized by the Somavamsis earlier. This comes through unmistakably because Codaganga, despite remaining a devotee of Śiva, undertook the construction of the Puri Jagannatha temple. This marked a shift in their structure of legitimation by changing the focus from their earlier association with Śiva-Gokarṇeśvara on Mahendra mountain. Hermann Kulke has done path-breaking work on the cult of Jagannatha and its changing forms through time. ²⁸ The cult synthesized various elements which brought together deities with a strong territoriality, that is, Virajā-Durgā of Jajpur, Śiva-Liṅgarāja of Bhubaneswar and Purushoṭṭama-Jagannatha of Puri. Anangabhima III presented himself as the son of Purushoṭṭama, Rudra and Durgā in an inscription dated 1216. ²⁹ During the first half of the thirteenth century, Jagannatha proceeded to emerge as the supra-regional deity (rāṣṭra devatā) and Anangabhimadeva III dedicated the state to Lord Jagannatha and consequently declared himself as his deputy (rāuta). The next stage unfolded with the coming of the Gajapatis, who seized power from the Gangas in 1434. The founder of the lineage pronounced himself as the first servitor (ādiśevaka) of the Lord. These shifts, we are informed, did not signal the unmaking of the Ganga–Gajapati rulers, but symbolized strategic manoeuvring on the part of the rulers to ensure the continued support and loyalty of their intermediaries and subordinates. Kings were now no more expressing their will, but essentially functioning as representatives of the regional deity/cult and carrying out his wishes, non-compliance with which was tantamount to sacrilege with its attendant consequences. Kings and their activities were now placed beyond the range of criticism, and they were invested with additional authority.

Social and Economic Foundations

The Anatomy of Kingdoms and Transregional States

At this juncture, it may be useful to investigate the structure of the earlier kingdoms and later-day transregional states in order to have a clear picture of the elements of continuity and change through time. In the case of Odisha, kingdoms usually emerged as maṇḍala states, with distinct geographical boundaries and identifiable sociopolitical structures. ⁴⁶ Kulke’s works have demonstrated the transition from chieftainship to ‘Hindu Kingship’ with simultaneous changes in the structure of legitimation across several of these subregional states in the early medieval times. ⁴⁷ One may add that these spaces gently evolved as historical entities/maṇḍalas with their own social compositions and art forms and other concomitant cultural traits, as seen above. The political and cultural boundaries mostly converged. In that sense they had an inner coherence and shared history, leading to a sense of affiliation, howsoever incipient. The ruler and the ruled shared certain commonalities, and the king partly derived his political validation from the people owing to this shared inheritance. ⁴⁸

Empires are usually conquest states, emerging from the foundations of a core territory with earlier experience of statehood. They are characterized by developed, developing and underdeveloped spatial entities because, in the process of acquiring desired resource areas to augment its economic basis, the victor also inherits intermediate territories, not necessarily at the same level of historical growth or social formations. Flowing from these developments, empires generally resort to creating pretensions of uniformity, including an ideology, to bind its subjects in a shared experience to ensure the continuance of the political enterprise. ⁴⁹ However, such states also exclude and hierarchize in the process of including peoples and cultures. Empire states are based on and perpetuate differences among peoples and cultures. ⁵⁰ Deriving from these characteristics of empires, we notice that the transregional states which emerged in Odisha in the early medieval period and after, in fact, seem to resemble them on many counts. The temples of Virajā, Lingarāja and Jagannātha successively represented the ideological dimension of the state under the Bhaumakaras, Somavamsis and Later Eastern Gangas, respectively. ⁵¹ The strategies of domination involved the accommodation of differences among people, as can be seen in the incorporation of Vishnu and his symbols in the Śaiva temple at Liṅgarāja and its replication on a wider scale at Puri Jagannātha. ⁵² The same practice can also be interpreted as the sustenance and manipulation of differences among the indigenous people to win over their affiliation.

The transregional states relied on diversity in their domains to shape a new order. Admittedly, under the Bhaumakaras and Somavamsis, spaces in Angul and Dhenkanal districts and in and around Bolangir and Sonepur districts, respectively, seem to have evidently opened up to the spread of the plough and of rural settlements along with other symmetrical developments, ⁵³ but they were not at the same level of growth as the territories on the littoral. Further, subregions such as Khinjali and Khijjiṅgakoṭā too registered visible growth during the eighth–tenth centuries, pointing to the uneven and gradual surfacing of varied spaces into historical limelight. It makes the more general point that the region presented a picture of a patchwork quilt rather than a seamless carpet. ⁵⁴ It is in these developments that tensions between centralizing and decentralizing tendencies or differences within the transregional states can be identified. Subregions came together in the making of a regional or transregional state, but it was not the same as having lost their identities once and for all. Their continued persistence can be seen in the domain of culture ranging from art, dance and music to culinary habits and choice of marriage circuits. And therein lays the roots of the history of subregional contestations. The expansion of administrative units, taxes and officials or the collection of taxes and mapping of the new territories by the bureaucracy through various stages did not, and could not, undo these inherent contradictions. Consequently, the western hinterland broke away both under the Somavamsis and the Later Gangas, despite efforts at incorporation. The Later Eastern Gangas’ integration of western Odisha in the thirteenth century was not an unqualified success, as it ceded after a brief interlude and the conquest of Khinjali maṇḍala too was temporary. Similarly, the nature of their control over Dandabhukti (Midnapore) is not very clear. ⁵⁵

The original homeland appears to have had an attraction for the Somavamsis and the Later Eastern Gangas insofar as the Bolangir–Sonepur region reports more than two dozen of the Somavamsi land grant charters (out of a total of around forty), many of which were issued even after their movement into coastal Odisha. The temples in Kaliṅga continued to be patronized even after the shift of the capital to the Cuttack–Puri region by the Gangas. Land grants continued to be made in Kaliṅga subregion and the Nāyaka system too found its best manifestation there. The Kaliṅga Parikṣās were frequently transferred to retain full royal control over the strategic division and not allow them to strike roots. To elaborate, over a period of 170 years, there were thirty-nine Parikṣās in Kaliṅga, which amply drives home the point. Kings, queens and important officials regularly visited Mukhalingam, the earlier capital in Srikakulam district in Andhra Pradesh, in order to retain their linkages and control over the region. Temples at Mukhalingam and Simhachalam, near Visakhapatnam, were continuously patronized. All this would indicate the privileged position of the homeland in the history of the transregional states. In the case of Simhachalam, it may also have been inspired by the desire to make royal authority more tangible in the frontier area. The continuance of official designations in the administration and the inherited cultural imprint in the newfound political centre unmistakably emblematize the continued preference for the place of origin. Their presence did not diminish in the usual course of day-to-day interactions with the conquered peoples and territories. State patronage of Śaivism under the Somavamsis had a Dakshina Kosala antecedent, and the influence of temple architecture from that subregion during the same period has already been discussed. On the provenance of their records, it appears that the activities of the Gangas were largely confined to the coastal plains. Their land grants and temple inscriptions are reported mainly from the undivided districts of Ganjam, Puri and Cuttack in Odisha, besides the districts of Srikakulam and Visakhapatnam in Andhra Pradesh, as the state extended on the coast up to the Godavari and Rajahmundry. ⁵⁶ The states under consideration were large political units, born out of territorial aggrandizement, in which incorporation of peoples and cultures and privileging of spaces and associated cultural attributes moved hand in hand. ⁵⁷

The transregional states under the Somavamsis and Gangas endured for about 200 and 400 years, respectively, thereby demonstrating considerable durability. The continuous focus on central coastal Odisha or the Puri–Cuttack region allowed it to increasingly emerge as the cultural centre. Not only was the cult of Jagannātha with its vertical and horizontal linkages located there, it also witnessed the spread of Odiā language and script. A few inscriptions from the time of the Later Eastern Gangas show the sprinkling of proto-Odiā characters, which finally by the mid-fifteenth century found its fullest manifestation in Sarala’s Odiā Mahābhārata and the Gajapati inscriptions. Interestingly, Sarala’s work has the Cuttack–Puri region as its spatial focus, though it draws its inspiration from the Sanskrit epic Mahābhārata, and in that sense it regionally appropriates a pan-Indian tradition and represents its vernacularization, both linguistically and culturally. ⁵⁸ It is the convergence of such developments with the gentle unfolding of a regional caste system and supra-regional polities which has made the Ganga–Gajapati times an important reference point even in modern Odisha. The cult of Jagannatha accommodated Śiva, Vishṇu and Śaktī, drawing on the inheritances of varied peoples and cultures. It seems to have provided a moral foundation to the power of the Gangas and Gajapatis and an ideological coherence to the state. States such as the Ganga transregional state did not exist in isolation; they interacted, competed and emulated contemporary comparable political systems. In this case, Vijayanagara with its temples, including Virupaksha, the tiered Mahānavami platform where Goddess Durgā was worshipped along with royal ceremonies, and the cavalry being central to the war machinery, provides a good instance. The Gangas and later the Gajapatis fought many battles with them for the control of the Godavari–Krishna delta. For the Gangas, the Mahanadi delta in central coastal Odisha and Kalinga (with the Rushikulya and Vamsadhara valleys) were important resource bases, both as their homeland and the centre of their territories in the littoral extending farther south. The contemporary allusion to the sultans of Delhi and rulers of Vijayanagara and Odisha as aśvapatis, narapatis and gajapatis, respectively, was also a recognition and reflection of their comparative resources and strengths.

There was an underlying vibrant economy, based on both agrarian and non-agrarian production, markets, merchants and towns. From the middle of the first millennium ce, it gradually opened up and expanded through the subregions and subsequently provided for an integrated regional economy with networks beyond its borders. The peasants paid various taxes but were not structurally dependent by any consideration, nor were there any signs of a subsistence economy. Politically, it is again a story of integration through stages and not fragmentation from the top. ⁵⁹ There are instances of the expansionist state located in central coastal Odisha intervening in the maṇḍala states, as in the case of the Bhaumakaras and even later. Nevertheless, the maṇḍala states having emerged from below or subregional societies cannot be seen as being analogous to vassalage; at most, they were semi-autonomous states. There is no visible manifestation of the feudal ladder either in terms of a well-graded hierarchy, military service or transfer of resources. ⁶⁰ The term sāmanta, at least in the present case, increasingly appears to mean an intermediary and not a feudatory. The maṇḍalas were historical regions which evolved through the simultaneous interplay of several historical processes during the early medieval centuries and beyond. From the time of the Bhaumakaras onwards, they began to be increasingly integrated into larger political enterprises, and in the process, these emerging cultural entities came to be accommodated within the expanding frontiers of aggrandizing states, which cut across maṇḍalas or historical regions. ⁶¹ Deriving from it, the states from the time of the Bhaumakaras to the Later Gangas and Gajapatis, which appear to be situated between usual kingdoms and the classical empires in terms of their spatial reach and political structure, can possibly be best understood as transregional states. ⁶²

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.