The Deccan in the fourteenth century witnessed a pattern similar to that of Bengal: the political expansion of the Delhi sultanate, the diffusion of Persianate culture, and rebellion followed by the establishment of not one but two independent regional sultanates.
In 1342, as his caravan lumbered southwards from Delhi to the Tughluqs’ co-capital of Daulatabad, Ibn Battuta described the road he was using as ‘bordered with willow trees and others in such a manner that a man going along it imagines he is walking through a garden; and at every mile there are three postal stations … At every station [dawa] is to be found all that a traveller needs.’24 Such a description indicates Muhammad bin Tughluq’s earnest attempts to link north India politically with the Deccan. But the Moroccan also recorded that those roads were infested with bandits, armed tribal groups and other miscreants, suggesting the sultan’s failure to build the sort of transport and communication infrastructure that is essential for any large state. The ever-resilient world traveller did manage to wriggle out of dangerous encounters with bandits on the road to Daulatabad, as he did many other times in his global travels. But there is no doubting the fragility of the lifeline that linked the Deccan to Delhi, suggesting that the sultanate’s attempt to annex the plateau was doomed from the start.
The Tughluq rulers inherited from their Khalaji predecessors two distinct ways of governing newly conquered regions: direct rule and indirect rule – a strategy not unlike nineteenth-century India, where British imperialists divided conquered territory into directly administered provinces and indirectly ruled ‘princely states’. In the annexed regions of the northern Deccan, stretching from the Vindhya Mountains down to the Krishna River, Muhammad bin Tughluq imposed a colonial idea – that is, a tightly controlled direct rule, symbolized by the name-change of the Deccan’s provincial capital from Devagiri to Daulatabad, meaning ‘city of government’. The idea was to plant colonies of northern immigrants, establish mints and strike coins on the same standard as those of Delhi, and partition the land into revenue parcels assigned to trusted iqta‘dars. Whereas many of these men were colonists from the north, administrators co-opted local society wherever possible by redefining as iqta‘dars those native chieftains who already occupied the land.
To the south of the Krishna, by contrast, the Tughluqs established an imperial idea, a system of indirect rule. Here they planted no colonists, appointed no governors, established no mints and made no effort to reach the base of the agrarian society, remaining content to identify powerful figures at the top of local political hierarchies and redefine them as tribute-paying chiefs, designated by the Perso-Arabic term amir, ‘commander’. Annual tribute, not regular taxes, was what they expected from territories south of the Krishna. By the end of Muhammad bin Tughluq’s reign, however, some tributary states were teetering on the verge of collapse. This was true of the Hoysala rajas, who had been conquered by Malik Kafur in 1311. But their capital, Dwarasamudra, lay some 650 kilometres south of Daulatabad, putting them practically beyond the Tughluqs’ reach. What is more, Malik Kafur’s looting of the Hoysala capital had so crippled the dynasty’s credibility across the south-western Deccan that its client-chieftains were soon commanding their own armies and carving out autonomous regions of effective control.25
Among the strongmen who emerged in this politically volatile environment were the five sons of Sangama, an obscure chieftain who seems to have been in Hoysala service in south-eastern Karnataka in the early fourteenth century. Already in 1313 one of his sons had emerged in the records as a politically active chieftain. By 1327, the same year the Tughluqs established direct administration in the northern Deccan, another of his sons, Muddamma, asserted his authority in the present Mysore District. In the next decade, both the Sangama brothers and the Tughluqs were picking up the pieces of the disintegrating Hoysala state, with Muhammad bin Tughluq co-opting chieftains formerly subordinate to the Hoysalas and installing them over their territories as Tughluq amirs. Ibn Battuta met three sons of one of those chieftains who had died resisting Delhi’s authority. The sultan had made all of them imperial amirs, the Moroccan reported, ‘in consideration of their good descent and [the] noble conduct of their father’.26 Sometime in the 1320s another of Sangama’s sons, Harihara, was enlisted in Tughluq imperial service as an amir, at least nominally. Although a contemporary observer called him a ‘renegade’ (murtadd), referring to his subsequent renunciation of former association with Tughluq authority,27 his loyal service in the Delhi sultanate persisted in folk memory. A Sanskrit chronicle composed around 1580 states that Muhammad bin Tughluq had ‘bestowed’ the entire Karnataka country (i.e. the south-western Deccan) on Harihara and his brother Bukka because the sultan, in his wisdom, had recognized the two men as eminently trustworthy and deserving of imperial patronage.28
However, the entire edifice of Tughluq authority south of the Vindhyas, both direct and indirect, began to crumble in 1336 when a Telugu-speaking chieftain of obscure origins led a successful uprising in Sultanpur, the Kakatiyas’ former capital of Warangal, which Prince Ulugh Khan had seized only thirteen years earlier. The imperial governor of the city was forced to flee, which permanently ended the sultanate’s control over the eastern Deccan, while other rebellions arose across the rest of the plateau. In that same year, Bukka asserted his autonomous control over parts of modern Kolar District. By 1339 Harihara had begun styling himself ‘Lord of the Oceans of East and West’, referring to his control over widely dispersed regions of the southern plateau between the Coromandel and Malabar coasts.29 In 1342 Ibn Battuta identified him as the overlord of the port of Honavar in northern Malabar. Significantly, he made no reference to Harihara’s former loyalty to the Tughluqs.30
By early 1344 most of Karnataka had accepted the Sangamas’ rule amidst the continued crumbling of both Hoysala and Tughluq authority. In February 1346 all Sangama’s sons gathered at the important Śaiva centre of Śringeri (in modern Chikmagalur District), where they celebrated Harihara’s coast-to-coast conquests. This meeting anticipated the final collapse of Tughluq authority on the plateau, the disappearance of the Hoysala house altogether – their last inscriptions were issued just two months later – and the formation of a new state, Vijayanagara, which would have an illustrious place in Indian history. Finally, in February 1347 one of the brothers, Marappa, declared the Sangama family’s principal deity to be Virupaksha, a form of Śiva. He also publicly presented himself as, among other titles, ‘sultan’, or in its Kannada formulation, ‘Sultan among Indian kings’ (hindu-raya-suratalah).31 In 1352, his brother Bukka adopted the same title, as did Harihara two years later, in an apparent arrangement of joint rule. In 1354 Bukka reaffirmed Virupaksha as the Sangama family deity, implicitly making that god the fledgling kingdom’s state deity. In 1355 he styled himself simply as ‘sultan’, instead of the earlier ‘Sultan among Indian kings’.32 When Harihara died in 1357, Bukka became the sole ruler of the new kingdom, now identified as Vijayanagara, ‘City of Victory’. The next year he added grandiose Sanskrit imperial titles to his name; in 1374 he even sent an ambassador to Ming China.33 Such was the meteoric rise of Vijayanagara, a hybridized sultanate-like state that would soon sprawl over nearly the entire southern half of the peninsula, and whose four dynasties, the Sangamas being the first, would survive down to the mid seventeenth century.
Meanwhile, a similar drama was playing out to the north of the Krishna. Although the Tughluq revolution of 1320 had overthrown the Khalaji rulers in Delhi, many former Khalajis remained entrenched as local administrators in Tughluq service. This was especially true in the Deccan, where such officers viewed the new dynastic house as usurpers. Among these was one Zafar Khan and his three brothers, who were nephews of a former high official in the deposed Khalaji court. In 1339, just when Harihara and his brothers emerged as dominant players amidst the chaos of the crumbling Hoysala dynasty further south, Zafar Khan and his brothers joined an anti-Tughluq uprising in which the forts of Gulbarga, Bidar and Sagar were all briefly seized. In addition to former Khalaji officers, Daulatabad’s settler-colonists who had migrated from north India since 1327 – many of them forced to move against their will – felt increasingly alienated by Muhammad bin Tughluq’s high-handed governance of his Deccan province. In 1344 the sultan dismissed Daulatabad’s popular governor, to whom Deccanis had looked to defend them against the sultan’s increasingly erratic and arbitrary actions. Things came to a head the next year, when 1,500 cavalrymen and a large number of commanders were ordered to be transferred from Daulatabad to Gujarat. Aware that the transfer amounted to a death penalty, given what the sultan had done under similar circumstances in the neighbouring Tughluq province of Malwa, these cavalrymen and officers openly rebelled, seizing and executing those government officers who had been sent to investigate them. The year 1345 was also the last that imperial coins were minted anywhere in the Deccan, indicating the demise of Tughluq rule across the entire plateau. Two years later Zafar Khan triumphantly strode into Daulatabad’s great mosque, where he was crowned Sultan ‘Ala al-Din Hasan Bahman Shah.
In short, in 1347 two independent kingdoms had emerged on either side of the Krishna, each founded by a family of obscure origins. The realm of ‘Ala al-Din Bahman, known eponymously as the Bahmani kingdom, covered the northern Deccan, where the Tughluqs had exercised a direct colonial rule. To the south of the river, the Sangama brothers based themselves in the new capital of Vijayanagara and ruled over territory vacated first by the Hoysalas and then by the Tughluqs. In the brief span of just eight years, between 1339 and 1347, the founders of these two dynasties had radically redrawn the plateau’s political map. Upon expelling Tughluq power from the Deccan, the Sangamas and the Bahmanis both established new, transregional states that sprawled across the plateau’s three vernacular zones: the Marathi, the Telugu and the Kannada. The two ruling families also styled themselves ‘sultans’. By using this title, which denoted supreme power in Persianate political discourse, the dynasties expressed not only their claims to political independence and supremacy but also their aim to participate in the wider Persianate world.
If these two states shared similar political origins, having arisen from simultaneous rebellions against their Tughluq overlords in Delhi, they differed vastly in how they legitimized their claims to rule. For the Bahmanis, as with other former provinces of the Delhi sultanate, the blessings of a Sufi shaikh were considered essential for ensuring a new state’s prosperity and legitimacy. According to Deccani sources, Shaikh Nizam al-Din Auliya himself had implicitly lent his blessings to the Bahmani enterprise, even though the state would not emerge until several decades after his death. In an anecdote current in the Deccan in the late 1500s, the great Chishti shaikh of Delhi had just finished a meeting with Sultan Muhammad bin Tughluq at his khanaqah in Delhi when he found Zafar Khan, the future founder of the Deccan’s Bahmani sultanate, waiting outside. ‘One sultan has left my door,’ the shaikh is reported to have remarked, ‘another is waiting there.’34 The anecdote illustrates the theme of an eminent shaikh predicting future kingship for some man, with that ‘prediction’ serving as a veiled form of appointment to royal power. For in the discourse of Sufism, sovereign authority was in fact held by these spiritually powerful shaikhs. In the words of ‘Isami, a contemporary chronicler who witnessed the launching of the Bahmani state, ‘Although there might be a monarch in every country, yet it is actually under the protection of a fakir [i.e. a Sufi].’35 By this logic, such shaikhs effectively leased political sovereignty to kings and charged them with the messy business of worldly governance while they themselves withdrew to a life of austerity, spiritual discipline and teaching.
However, a shaikh’s prediction of kingship was no guarantee of a sultanate’s well-being. As in Bengal, independent rulers in the Deccan needed the continuing legitimacy underwritten by a resident Chishti shaikh spiritually descended from Shaikh Nizam al-Din Auliya. ‘Isami recorded that with the latter’s death in 1325, the city and empire of Delhi had sunk to desolation, tyranny and turmoil. But the Deccan, he maintained, suffered no such fate. Because one of Nizam al-Din’s leading disciples, Burhan al-Din Gharib, had migrated down to Daulatabad in 1329, the city prospered thanks to the shaikh’s radiant, beneficent presence.36 When Burhan al-Din died in 1337, that protective presence passed to his leading disciple, Shaikh Zain al-Din Shirazi (d. 1369), by whose actions the newly launched Bahmani state was transformed from a rebel movement into a legitimate Indo-Islamic kingdom. The very robe worn by the Prophet Muhammad on the night he ascended into Paradise – a robe subsequently passed on through twenty-three generations of holy men until finally received by Zain al-Din – was allegedly bestowed upon Zafar Khan when he was crowned sultan in 1347.37
Sultan ‘Ala al-Din Bahman Shah’s earliest successors to the throne sought the support of other Sufi shaikhs, one of whom, Shaikh Siraj al-Din Junaidi (d. 1379/80), shifted his residence from Daulatabad to Gulbarga when the Bahmani capital moved there. The shaikh presented a robe and turban to each of the first three Bahmani sultans during their respective coronations. Sufi shaikhs thus played roles fundamental to the diffusion of sultanate systems in the Deccan, as elsewhere in India. Whereas Khalaji and early Tughluq invasions beyond the Vindhya Mountains had lacked any moral basis, being undertaken simply for booty or tribute, the extension of the Sufis’ notion of spiritual sovereignty lent moral legitimacy to the planting of a sultanate’s political authority. No longer available for plunder with impunity, such land – its people, its produce and its fixed assets (including Hindu temples) – now merited state protection. In classical Islamic discourse, the presence and blessings of great Sufi shaikhs could transform yesterday’s Abode of War (Dar al-Harb) into today’s Abode of Peace (Dar al-Islam), thereby legitimizing the transplantation of Indo-Muslim rulership from one region to another within South Asia. As vessels into which divine favour was believed to have been poured, great shaikhs thus exercised a quasi-political dominion over the lands in which they resided. If Shaikh Burhan al-Din Gharib’s arrival in colonial Daulatabad had inaugurated Delhi’s legitimate rule of the Deccan, legitimate independent rule there began when Shaikh Zain al-Din Shirazi bestowed the mantle of the Prophet on ‘Ala al-Din Hasan Bahman Shah in 1347.
To the south of the Krishna, meanwhile, the new state of Vijayanagara asserted very different claims to legitimate rule. Its authority was based on a goddess cult that had emerged as early as the seventh century on the southern banks of the Tungabhadra River, a major tributary of the Krishna. At that time the site was known simply as Pampa’s tirtha – or the ‘crossing’ of the river goddess Pampa – where passing chieftains would halt and make votive offerings during military campaigns. By the ninth century the first stone temple had appeared at the site, dedicated evidently to this goddess. By the early eleventh century, donations were being made to the male deity Mahakala Deva, the violent aspect of Śiva. By the twelfth century, a temple complex dedicated to Virupaksha, who represented Śiva’s more universal and benign aspect, had emerged. Unlike the earlier phase, when she was merely protected by Mahakala Deva, to whom she was in no way subordinate, the river goddess Pampa was now reduced to a subordinate status as Virupaksha’s consort. Moreover, south Indian texts had begun describing Pampa’s marriage to Virupaksha in terms paralleling the all-Indian story of Śiva’s marriage to the goddess Parvati.38
In this way, over the course of 500 years a regional shrine had gradually become Sanskritized, as a local river goddess was pulled up into, and transformed by, the big world of pan-Indian Śiva-worship. At the same time, the site grew ever more important as a pilgrimage centre. From the thirteenth century on, politically ambitious or already dominant rulers in the area had begun cultivating closer ties with the shrine and its deities, whereas in earlier days only passing chieftains had done so. In the early fourteenth century, the short-lived kingdom of Kampili tried to reap the ritual and political benefits of an expanding goddess-cum-Śiva shrine by building a state near the site. But in 1327 Sultan Muhammad bin Tughluq crushed that fledgling kingdom. Soon thereafter, though, the Sangama brothers proclaimed themselves to be the site’s ‘protectors’. And when those chieftains established their capital at the site, by then called Vijayanagara, their family deity Virupaksha was elevated to the status of state deity. Important state documents now bore the ‘signature’ of this deity, while the Sangama kings heaped lavish architectural patronage on his temple complex by the banks of the Tungabhadra [see Fig. 4]. Hundreds of copper-plate inscriptions recording land grants state that the donation was carried out ‘in the presence of Virupaksha on the banks of the Tungabhadra River’ and conclude with a large ‘signature’ of the god ‘Śri Virupaksha’, written in Kannada script. As owner of all state land within his kingdom, the deity in this way participated in formal transactions and attested to their validity.
From the moment of its launching, then, Vijayanagara was a culturally hybridized kingdom. Its religious origins and legal foundations were embedded in a state cult focused on a local form of the great god Śiva, just as the Kakatiya kingdom had been grounded on one devoted to another form of that deity. On the other hand, the kingdom’s ruling institutions, architecture and its wider, transregional posture suggest a desire to assimilate the ideals and idioms of the Persianate world. Vijayanagara’s system of land revenue assignments, called nayamkara, required high-ranking officials, or nayakas, to combine military and tax-collecting duties. Having no known precedent in earlier kingdoms of the Deccan, the nayamkara institution was apparently modelled on that of the iqta‘ used by the Tughluqs and their political descendants in the Deccan, the Bahmani sultans. Vijayanagara also incorporated and lavishly displayed Persianate architectural motifs – arches, domes, vaulted arcades, squinches, and so on – together with the image of the sultan as a supreme king of kings.39 Paralleling Turko-Mongol political traditions, claims to sovereignty at Vijayanagara were extended to all male members of the ruling family, meaning that royal successions were marked by tensions and sometimes violent conflict.40