Lords of the Deccan

Southern India from the Chalukyas

to the Cholas

Anirudh Kanisetti
 Praise for Lords of the Deccan
‘Rarely has the history of peninsular India, of this period, been told as anything more than a dry and
ceaseless monotony of battles between obscure and unimaginable rulers. Anirudh Kanisetti’s Lords of
the Deccan has lifted the history of south India out of the dusty archives of Indian archaeology and
epigraphy. Meticulously researched and narrated with a style that is at once lively and
judicious, Lords of the Deccan synthesizes a wide array of innovations in recent scholarship with the
older tradition of political history. Kanisetti harnesses his impressive skills as a storyteller to breathe
new life into his subject, deftly interweaving the careers of individual kings, the structures and
networks of noble families, and the great transformations in religious, cultural and literary life into a
single coherent and riveting account of south India in this crucial period, which saw the region enter
historical centre-stage and take on many contours still palpable today.’
– Daud Ali
‘Lords of the Deccan is the completely thrilling and game-changing debut
of a major new talent. Anirudh Kanisetti is a superb writer and a talented storyteller as well as an
impressively judicious and subtle historian. He breathes life into the rajas, scholars and soldiers of
two nearly forgotten medieval dynasties and resurrects for us a whole extraordinary world with flair,
nuance, clarity and sophistication.’
– William Dalrymple
‘Lords of the Deccan is an assured and supremely entertaining account of a clamorous, tumultuous
and little-known period of Indian history – the early medieval Deccan period. Navigating these
unchartered waters with considerable confidence and panache, Anirudh Kanisetti brings to vivid light
both the savagery and the song of these riotous five hundred years. He does so, moreover, by
dismantling the opaque language of historiography for the
lay-reader and giving them the tools to truly understand the dialect of power – the temple building,
the ode-writing, the painting and even the very manipulation of language and religion. With Lords of
the Deccan, Kanisetti has claimed for himself a place of pride in the cartography of Indian history
writing.’
– Ira Mukhoty
‘Ambitious in its scope and rich in depth and detail, Lords of the Deccan is an outstanding debut.
With his evocative retelling, Anirudh Kanisetti restores medieval south India to the prominence and
centrality it deserves in general imagination – marrying old learning to new perspectives on
everything, from kingship to the evolution of religions.’
– Manu S. Pillai
‘In shining a light on the Deccan in the second half of the first millennium, Anirudh Kanisetti has
given us a marvellous book. It brings to life the Chalukyas, the Pallavas, the Cholas, the Rashtrakutas
and others, and in the process gives us a peninsular perspective of Indian history. This is also a story
of imperial achievements in which language, literature, faith, sculpture and artistic achievement weld
seamlessly with political and military stratagems. A wonderful read.’
– T.C.A. Raghavan
‘Anirudh Kanisetti takes us on a fascinating journey into the past to revive five
hundred years of glorious rule of early medieval Deccan rulers and brings the
Chalukyas, Rashtrakutas and Cholas back into popular historical consciousness.
Kanisetti’s engaging narrative style and lively account brings to life the triumphs, defeats and
rivalries of those lords and ladies of the Deccan who have left behind a legacy of magnificent
architecture and sculpture; sumptuous textiles and jewellery; evocative poetry and literature.’
– Rana Safvi
And to the generations who have shaped the world before us.
 Introduction

Every monsoon, rain clouds bathe the cool, dark surfaces of an ancient
temple in Ellora, Maharashtra. Peals of thunder echo in its cavernous halls,
like the bells that once greeted throngs of devotees.
There’s something dazzlingly different about this gigantic temple. You
see, it isn’t a building of the kind you and I might be used to. It wasn’t
assembled bottom up from the ground, brick by brick, stone by stone.
It was excavated.
It is called the Kailashanatha, the Lord of Kailasha, because generations
of awestruck visitors have seen it as a manifestation of the mountain upon
which the god Shiva lives. To fashion it, thousands of sculptors carved up
an enormous basalt cliff face, removing two million cubic feet of rock
(enough to fill two Olympic-sized swimming pools). They did so in barely
twenty years in the ninth century CE, with a plan breathtaking in its scale
and attention to detail, leaving behind a monolith the size of a football field
and about half the height of the Leaning Tower of Pisa. A monolith in the
shape of a spectacular south Indian temple, with the weight of its
superstructure cascading down in wider and wider tiers, decked with
sculptures of frolicking deities. The Kailashanatha is a single sculpture so
large that it approaches the size of modern buildings. As a monolithic
structure, it is unlikely to be matched in size and beauty for the rest of
human history.
This extraordinary edifice was made by people who thought themselves
every bit as modern as you or I. They were a vibrant, warlike, sophisticated
people. They were ruled by men who claimed the majestic title of Sri-
Prithivi-Vallabha, the Beloved of Sri (the goddess of fortune) and Prithivi
(the goddess of Earth). Their empire dominated the ancient Deccan plateau
at the heart of India, especially the states of Maharashtra and Karnataka,
today an area almost as large as Germany and many times more populous.
At their peak, these Vallabha emperors received the prostrations of hosts of
vassal kings from Madhya Pradesh, Chhattisgarh, Odisha, Andhra,
Telangana and Tamil Nadu, and dominated most of India south of the
Narmada river. One Arab merchant, visiting this medieval superpower in
the ninth–tenth centuries, mentioned the lord of the Deccan in the same
breath as the Abbasid caliph, the emperor of China, and the Byzantine
emperor.1 Another visited the glittering capital of the Deccan, Manyakheta,
and left us with an account of its wonders:
… in that city there are for the ordinary people one million elephants which carry the
merchandise … In this temple there are about twenty thousand idols made of a variety of
precious metals, and carved stones mounted with artistically-worked precious jewels … [There]
is an idol whose height is twelve cubits and is placed on a throne of gold in the centre of a
golden cupola, the whole of which is set with jewels like white pearl, ruby, sapphire, blue and
emerald stone.2
Both these accounts – even allowing for some exaggeration – leave no
doubt that in the eyes of the medieval world the Deccan was the wealthiest
and most powerful of all the kingdoms of the Indian subcontinent.
This book is a story about this time when the Deccan ruled India: an epic
journey through five hundred years of a history that has long been
forgotten.

Our tale begins in the sixth century CE, a few decades after the collapse of
the Western Roman Empire in Europe, only a few years after the
disintegration of the Gupta empire of northern India. In the dry and arid
heartland of the Deccan, cattle raids, banditry and abduction were
ubiquitous. Here, an obscure clan of chalke (crowbar)-wielding
agriculturists or pastoralists3 learned the difficult lessons of war and
diplomacy, and began to battle their way up the shifting hierarchies of
India’s kingdoms. Within the space of three generations, they declared
themselves a new imperial dynasty – the Chalukyas – established a mighty
citadel in the sandstone cliffs of Vatapi in northern Karnataka, and exploded
on to the historical stage by defeating the dominant ruler of north India.
That collision, which occurred in 618 CE on the shores of the Narmada river,
is where this book begins. It will set the stage for a half-millennium of
Deccan dominance.
In the first part of this book we will watch these Chalukyas, masters of
medieval Indian geopolitics, at work. They understood very well that there
was little wealth to be scratched out of the arid lands of the Deccan. Soon
after their emergence on the medieval Indian stage, they went to war north,
south, east and west, ruthlessly raiding their wealthier neighbours and
breaking into the networks of the Indian Ocean trade. We will watch how an
empire was made, attempting to peek, through the dust of centuries, into the
minds and hearts of the men and women at its centre. We will accompany
them in these wars, gaining a singular look into the machinations of
medieval Indian power, and the glories and tragedies associated with it.
We’ll see how this power shaped and was shaped by the turbulent religious
and social tides of medieval India, observing these upstart Chalukyas –
constantly looking for new propaganda to rally their unruly vassal chiefs
and subjects – ally with the rising tides of bhakti devotion to Shiva the
Destroyer; patronize the use of Sanskrit literary texts in south India; and
embark on a wave of monumental building projects, establishing some of
the oldest surviving temples in the subcontinent. By the mid-eighth century,
this project of dynastic aggrandizement had elevated the Chalukyas to the
heart of a sprawling network of vassal kings, governors, trading ports and
pilgrimage sites that dominated the Deccan plateau and much of India’s
western coast. A cadet Chalukya line ruled a kingdom of their own in
Andhra, on India’s east coast. This vast agglomeration of people could
mobilize resources of such a scale that a Chalukya vassal, acting on his own
initiative, was able to smash the Umayyad Caliphate’s attempt to conquer
Gujarat in 737 CE, defeating a seemingly invincible army that had seized
Sind and parts of Gujarat, and even reached the outskirts of Ujjain in
modern-day Madhya Pradesh.
In the second part of this book, as the world changes, we will watch the
Vallabha emperors of the Deccan – a title now held by a clan called the
Rashtrakutas – lead the plateau to a splendid apogee. During their time, the
Abbasid Caliphate in the west and the Tang dynasty of China to the east
oversaw an age of flourishing trade with the Indian subcontinent. The cities
of the arid plateau and the ports of western India, under Rashtrakuta
control, began to trade in everything from indigo and perfumes to exotic
poisons, fruits, animals and spices. The Rashtrakutas took advantage,
inviting Arab merchants to serve as the governors of harbours humming
with activity, ordering fine Sanskrit verses to be composed in their honour,
and importing the finest horses in the world to serve in their armies. We will
follow them as they project this power into the rest of the subcontinent.
They will lead armies of marauding south Indians into the Gangetic plains,
nearly a millennium before the rise of the Marathas. They will manipulate
the politics of Madhya Pradesh, Andhra Pradesh and Tamil Nadu from their
seats of power in Maharashtra and Karnataka, moving pawns around a vast
geopolitical chessboard. We will watch them compile great grammatical
treatises in Kannada, and hear how their poets mounted the first serious
challenge to literary Sanskrit, once India’s dominant language of power and
prestige. We will watch them commission the dazzling Kailashanatha
temple and the splendid imperial city of Manyakheta, projects that cost tens
if not hundreds of millions of dollars in that day’s currency. And we will see
how, like their predecessors, they used religion as royal propaganda,
patronizing a unique form of Jainism that once ruled south India as the
equal of Hinduism.
However, the rest of India did not sit idly by as the Deccan threw its
weight around the subcontinent. In the third and final part of this book, we
will see how the rise of new challengers to its dominance left the medieval
Deccan world in burning ruins. We will watch the military disasters that led
to the collapse of the Rashtrakutas – easily some of the most calamitous
upsets in Indian history – and observe how, in the bloody anarchy that
followed, the Chalukyas returned to once again restore order to the Deccan.
Their return came not a minute too soon. In the deep south, a new power
rose to challenge them: the imperial Cholas, perhaps the most famous of all
south Indian dynasties today. Over the course of twenty years, we will
accompany the terrified courts of the Deccan as they watched the Cholas
burn and conquer kingdoms through India’s east coast and attack Indonesia,
an unheard-of feat for any medieval Indian polity. And we will see how
those bloodthirsty conquerors finally met their match in the gritty,
determined Chalukya Vallabhas of the Deccan. We will witness the clash of
these two south Indian superpowers, the culmination of centuries of social
and political evolution in the Deccan and the Tamil country. Both
commanded armies numbering in the tens of thousands, both ruled
enormous, sophisticated courts in bejewelled temple-studded capitals. Both
hated each other with a burning passion that razed cities to the ground, tore
families apart, killed kings and left kingdoms in ruins.
In the midst of this chaos, a new generation of Chalukyas and Cholas
tried, at last, to put an end to the violence. Betraying his father and brother,
a new Vallabha, wiser and warier than his predecessors, backstabbed his
way to power with the help of a Chola ally. But just when it seemed that –
at long last – peace would return to the land, one of the most shocking
political upsets of south Indian history unfolded in the Tamil country,
ending any hope of peace and reconciliation between the two great
geopolitical regions. As vassals of the Chalukyas rose to challenge their
dominance across the Deccan, we will leave the plateau on the brink of a
century of renewed war, that would only end with Delhi’s invasions of the
south and the rise of Vijayanagara and the Deccan Sultanates.

Most Indians today are unaware that the Deccan has such a dramatic and
world-changing past. Our understanding of the history of this vast, diverse
subcontinent is based on an obsession with ‘imperial moments’ – often
fleeting moments in history when north India is able to impose its
dominance over other regions. In our school textbooks – which retain a
disproportionate influence in shaping our identities and sense of the past –
we leap five hundred years from the Mauryas of the second century BCE to
the Guptas of the third century CE. We then jump six hundred years from the
end of the Gupta empire directly to the arrival of the Turkic sultans in north
India in the twelfth century, and thence move neatly to the Mughals, the
British and then Independence. Somehow, in this subcontinent that is as
large as, more populous than and exponentially more diverse than western
Europe, we are used to ignoring the histories of entire peoples, eras and
regions when thinking about how India became India.
This is a ludicrous way to contemplate the subcontinent’s history.
Ignoring the history of the Deccan in recounting the history of India is like
ignoring the history of France or Germany in telling the history of Europe.
This book aims to do something about that. It is the story of India between
two north Indian ‘imperial moments’, the half millennium or so after the
end of the Gupta empire and before the establishment of the Delhi
Sultanate. In order to do so, it roots itself unabashedly in the Deccan. Yet it
does not seek to replace a north Indian ‘imperial moment’ with a south
Indian one, but instead seeks to develop a more complicated and
interconnected narrative of the history of this enormous and diverse land
between the Himalayas and the Indian Ocean.
Our modern obsession with ‘imperial moments’ makes it difficult to
appreciate the scale and uniqueness of the subcontinent’s history. By now,
every region of India has its legends of great and glorious monarchs who
‘made contributions’ to a medieval or ancient culture implicitly connected
to a contemporary political identity. This view reduces our past to a stale
series of moralistic stories and figures who serve black-and-white
conceptions of linguistic, regional or religious glory. In this view, Indian
kings were not living, breathing human beings like you or I, but flawless
paragons, images based on little more than tiresome sermonizing as to what
constitutes ‘greatness’.
On the other hand, a new trend in popular history attempts to paint them
as sexy influencers similar to what one would find in HBO’s Game of
Thrones; another makes them out to be enlightened crusaders for human
rights a thousand years before the concept existed. These caricatures are
boring substitutes for the vibrant and diverse lives that our ancestors
actually lived.
This book takes a somewhat different approach to thinking about and
writing about the past.
 Imagine that somehow, in the year 3020, all that remains of India from
2020 are ads issued by the Union government in newspapers in Uttar
Pradesh; Instagram posts from posh art galleries in New Delhi; and
recordings of grand galas attended by the who’s who of society, industry
and the art world in Mumbai. Future historians decide to engage with this
evidence from their past in three ways. One group diligently collects the
government’s ads into a neat chronology, declaring it the most effective
government that ever existed, the most flawless and intellectual political
leadership ever, an exemplar in pandemic management, economic recovery
and social harmony for all who came after. Another pores over the
Instagram posts and waxes eloquent about the amazing art that filled the
museums and dreams, starry-eyed, of the generous patrons who must have
showered money upon the talented artists who created them. Another
fashions gorgeously produced sensory experiences that hardly anyone can
afford to purchase, filled with tear-inducing nostalgia about the lavish
events and luxurious clothing of this long-forgotten golden age. Crowds of
people shake their heads sadly and yearn to go back to those days.
Under such circumstances, it would be easy to ignore the humdrum
struggles of the millions of other people who lived in India in the 2020s and
didn’t get to leave behind the fragments of evidence that our imaginary
thirty-first century historians are so enamoured with. It would be easy to
forget about Kolkata, Chennai, Bengaluru, Hyderabad.
This – fragmentary evidence fit into feel-good nationalist sermons or feel-
cool romantic narratives – is the state of popular writing about medieval
India today. A thousand years on, we have forgotten how to imagine a past
India as searing and real as the India we inhabit today.
So how will this book be any different? We cannot magically conjure up
more evidence than actually exists from more than a thousand years ago.
Not a lot survives from the time this book explores, partially due to the lack
of systematic archaeological study. As far as actual evidence goes, we’re
mostly stuck with royal land grants full of kingly boasts of generosity,
religiosity, and administrative, sexual, artistic and military prowess. We
have imposing temples covered with sculptures. We have literary and
sculptural portraits of glittering court life. Like our imagined 2020–3020
scenario, a thousand years on, all the evidence we have from this distant
time was shaped by a tiny and supremely well-off social elite and their self-
promotion. If we are to have a realistic understanding of our history, we
need to interpret this keeping in mind that these are mere fragments of a
vast and complicated world similar to ours, inhabited by individuals who
shared the same fundamental human impulses – including the urge to
pretend they were less imperfect than they actually were.
And so the kings and queens you will meet in this book are neither
flawless paragons nor sexy influencers. Instead, they are much like people
you might see around you today. This book will help you understand their
activities – war, politics, intrigue, patronage – as they were intended and as
they were perceived at the time, from battlefield savagery to temple
building to literature and sculpture. The book has no heroes or villains, no
‘superior’ or ‘inferior’ regions or cultures or morals. Instead, it is a tale of
the grand forces of nature and randomness, and the tiny humans who dare
to make history out of it all. It will explore the complexity of power and
people in medieval India, so similar to our India. We will discern a close
alliance of religion and politics; ruthless violence against dissenters and
rivals; relentless narcissism and ambition; stark inequality; monumental
architecture; seductive glamour; and ravishing, unparalleled, immortal art.
It is not a dull, comforting history, but a vivid, fascinating past far closer to
the reality we inhabit.

The five hundred years of history through which we’ll journey are among
the most misunderstood in our modern understanding of the past. Of late, it
has become fashionable to accept a colonial-era tripartite division of India’s
history: a ‘Hindu’ period, a golden age, called ‘ancient’; a ‘Muslim’ period,
a dark age, called ‘medieval’; a ‘British’ period, enlightened, modern. The
archaeological and academic consensus does not support this simplistic
division. Since the early twentieth century at least, generations of scholars
have excoriated it as a deliberate fabrication intended to portray the British
as ‘rescuing’ a Hindu India from ‘Mahomedan’ tyranny. This image comes
across very clearly in works such as Robert Sewell’s A Forgotten Empire:
Vijayanagara – A Contribution to the History of India, which remain
popular today for lack of accessible modern writing about the period.
Just like the stereotypes of the rich and powerful which we saw above,
this tripartite division obscures a past that is far more complicated than we
might think. Historical India is a unique and fascinating world, with
features far more profound than the religion a bunch of royals happened to
follow. Drawing on more systematic and objective appraisals of the
evidence, the scholarly consensus now recognizes an ‘ancient’ period
stretching from the third century BCE to the fifth–sixth centuries CE,
involving a deep connection to Central Asia coupled with religious and
political efflorescence. This era gradually transitioned into an ‘early
medieval’ period during the seventh to twelfth centuries, associated with
radically new ways of organizing polities and societies in the subcontinent,
as international trade grew and religions became vastly more complex and
politically involved. This was followed by a ‘late medieval’ period from the
thirteenth to the sixteenth centuries involving deeper cultural, political and
religious engagement with the Persianate world. The epoch transitioned into
one of increasingly powerful and globalized states in the ‘early modern’
period from the sixteenth to the nineteenth centuries, finally bringing us to
the ‘modern’ period of the subcontinent’s history with the onset of
colonialism.
Over the time this book covers – the early medieval period, roughly the
seventh–twelfth centuries CE – the courts and battlefields of India witnessed
events contemporaneous with the most dramatic changes in global history:
the birth of Islam in Arabia, the Abbasid ‘Golden Age’ in Iraq, the
formation of the Holy Roman Empire in Germany, the Great Schism
between Catholic and Orthodox Christianity, the Song dynasty’s economic
revolution in China, the rise of neo-Confucianism, high chivalric courtly
culture, and the Norman conquest of England. Through this time, India saw
supremely important and influential developments in religion, art,
architecture, literature and political economy. The major geopolitical
regions of this subcontinent – the mountains of the Himalayas, the Gangetic
plains, the coasts of Odisha and the Tamil country, the arid pastures of
Rajasthan, the Malwa plateau, and of course the Deccan – saw new forms
of political, social and religious organization, and developed increasingly
complex interrelationships within themselves and with each other. These
new ways of doing things would shape the subcontinent into a form we can
recognize today.
Hundreds of new cities and towns grew during the medieval period, and
rose to prominence. Many of them still survive in some form today: the
cities of Dhara, Kalyana, Vatapi, Thanjavur, Kanchi, Old Goa, Banavasi,
Mamallapuram, Khajuraho, Warangal, Halebidu and Kannauj. Salons and
courts reverberated with the recitation of marvellous literature. Artisans
made sumptuous textiles, paintings and jewellery. Their products adorned
the bodies of cultured aristocrats and talented dancers participating in rich,
diverse and sophisticated material cultures. Thousands of elaborate
sculptures and temples were assembled in increasingly complex and awe-
inspiring forms. They were paid for by the wealth that perfumed lords and
ladies wrung from a growing population of emaciated agriculturists.
All this was nourished by trade and cultural exchange with the rest of the
world: and the drama, depth and spectacle of medieval India is easily on par
with the global events mentioned above. India had its own William the
Conqueror equivalents who invaded ancient island kingdoms and made
them their own; its own religious movements whose waves of devotion
transformed the lives of millions; its own charismatic, ruthless emperors
who remade entire countries in their images; its own powerful,
uncompromising religious sects drawing on centuries-old philosophical
traditions, whose struggles for power and influence led to the making and
breaking of kingdoms. Yet their relationship with the rest of the world – or
even their relationship with the rest of Indian history – is poorly understood,
at least in the domain of popular history.
This book seeks to remedy this gap in our conception of our past, because
no story of these transformations that shook India and the world can be told
without giving the medieval Deccan the centrality it deserves.
The erasure of the Deccan from our historical consciousness is one of the
strangest reversals of fortunes in Indian history, especially given what its
contemporaries thought of it.
Think about the last major Indian empire before this period, that of the
Guptas. The Guptas are increasingly present in popular ‘historical’ fiction
in India, always a good metric of the recognizability of a historical dynasty
to a general audience (not coincidentally, Maratha, Mughal, Rajput and
Chola fiction are also high up on the bestseller list). The Guptas are often
thought of as the Indian Golden Age, the pinnacle of India’s religious
history before the coming of the Turks, as well as the apogee of Indian
architecture and martial power and art and poetry. Yet on every one of these
counts they pale in comparison with their successors in medieval India in
general, and the medieval Deccan in particular.
At their peak, almost the entirety of India south of the Narmada would
pay the Vallabhas tribute and acknowledge their overlordship – a record
matched by no Deccan or south Indian polity before or since. In
comparison, Gupta influence was predominantly felt in Uttar Pradesh and
Bihar, though they exercised some control in Madhya Pradesh and Gujarat
and may have received tribute from as far away as regions in modern-day
Pakistan in the northwest of the subcontinent. These north Indian kings only
ever managed to attack southern India once, under the reign of the ruthless
Samudragupta. Compare him to the five Deccan Vallabhas who attacked
north India, one (Indra III) reaching as far as Kannauj. One (Vijayaditya I)
may have made it as far as the Ganga before being captured, another
(Dhruva I) smashed two of north India’s most powerful armies near the
confluence of the Ganga and Yamuna. At least two (Krishna III and
Someshvara I) sacked, burned and subjugated Madhya Pradesh before
turning their attentions elsewhere. To the medieval Deccan mind, the
Gangetic plains were little more than a stage for the display of their
intimidating military might before the shocked eyes of the subcontinent’s
other kings.
And who were these other kings? Barely a century after the collapse of
the Guptas, their political legacy in north India almost completely vanished,
to be replaced by other dynasties – the Maukharis, the Pushyabhutis, the
Palas, the Pratiharas and, later, the Chandellas, the Paramaras, the
Gahadavalas, the Chahamanas and many others – some of whom we will
meet through the course of this book. Many of these medieval dynasties
arguably left a far deeper imprint on India’s literary and aesthetic culture
than the Guptas ever did.
Yet all of them pale before the extraordinary power and influence of the
emperors of the Deccan, who shaped the fate of many modern-day Indian
states, some of which exceed the size and population of European countries.
The Kakatiyas of Telangana and the Hoysalas of Karnataka, both of whom
still occupy a hallowed place in regional memory, were vassals of the
Deccan Vallabhas; they broke free in the chaos that engulfed the Deccan in
the thirteenth century, before the invasions of the Delhi Sultanate. The
Kadambas and the Shilaharas, their vassals in Goa, founded the great port
of Gopakapattinam, which would eventually draw the avarice of the
Portuguese and form the nucleus of what is now Old Goa. The bloodthirsty
campaigns of the Vallabhas even, inadvertently, led to the emergence of the
famous imperial Cholas of the eleventh century. How ironic that when
south India is remembered at all in popular Indian history it is only the
Cholas who feature in it, totally overshadowing the Chalukyas and the
Rashtrakutas, their accidental preceptors and deadliest rivals.
All this is just the tip of the iceberg, a simplistic comparison of the
military and political aspects of power that loom so tall in our modern
historical consciousness. You’ll see here, as we explore the other aspects of
medieval Indian power, that the lords of the Deccan really made India.
Hinduism as we know it in southern India might not have existed if not
for them. Think of temple visits, for example: when the Deccan began its
long trek to imperial power, the idea of enshrining Hindu gods in temples
was still new, and primarily a north Indian one. As part of their propaganda
projects, Chalukya kings lavished patronage on this new ‘Puranic’ form of
Hinduism, creating a religious practice focused on pilgrimages and on ritual
worship at temples built by kings. This decision was hugely significant in
the fractious religious landscape of medieval south India, where religious
sects professing many different rituals and routes to salvation competed for
influence, land, patrons and devotees. The patronage of the Chalukyas, and
kings like them, would swing the balance of power decidedly in favour of
the many religious practices that we now call Hindu. But in the Deccan, this
‘Hinduism’ coexisted with an innovative form of Jainism that drew on a
similar set of practices: organized monasteries, temples, public rituals,
pilgrimage. South Indian Jainism, patronized by the Chalukyas’ successors
to Deccan overlordship, the Rashtrakutas, was a unique, warlike form of
this ostensibly peaceful religion, and was very popular with the Deccan’s
glamorous military aristocracy. The staggering plurality and contestation
that marks medieval Deccan religion is an important counterpoint to our
popular notion of India as an eternally or unchangingly Hindu country.
As with religion, so with many other aspects of medieval India. The
Deccan, as the subcontinent’s dominant power, straddled axes of exchange
stretching from Kashmir to Kanyakumari, from Sind to Bengal. Ideas and
migrants poured in, drawn by the Vallabhas’ wealth, and they mingled with
the peoples of the heart of the subcontinent, arriving at uniquely Deccan
forms that would in turn spread out to influence the rest of India.
For example, the earliest Deccan temples – some of which are the oldest
still-standing temples in the subcontinent today – freely took elements from
both north and south India. But Deccanis soon began to innovate entirely
new plans of their own. As the Deccan grew ever more socially and
politically complex and more influential, we begin to see Deccani temple
design elements as far away as Madhya Pradesh, and vice versa. As part of
a broader medieval Indian trend towards more elaborate temples based on
iterations of simple patterns, the Deccan began to build spectacular star-
shaped shrines that have never been equalled since.
The Chalukya emperors also adapted and reworked north Indian Sanskrit
political propaganda to suit the Deccan, sparking off generations of
imitations and a feverish era of Sanskrit literary production across southern
India. Over the centuries, as south India’s cities and courts grew ever larger
and wealthier, its aesthetics and its grand religious and poetic ideas,
enriched the vibrant intellectual culture of Sanskrit, adding to the corpus of
thousands of texts on everything from grammar to architecture to political
theory. The successors of the Chalukyas, the Rashtrakuta emperors, turned
this dynamic on its head: they took a south Indian Sanskrit grammatical
treatise and used it to create a grammar for Kannada, the language of the
people of Karnataka, the heart of their empire. In doing so, they set off an
explosion of regional literature that lasted off and on for a thousand years –
the first time any vernacular language successfully challenged Sanskrit’s
dominance of court culture. If not for their activities, it is doubtful that
many south Indian languages (with the exception of Tamil) would have the
hallowed literary tradition they now do.
And so on, and so on. Through this book, we will delve into the lives of
the men who ruled the Deccan and the complex ways in which this region
changed over half a millennium. We will explore how art, literature,
religion and architecture were influenced by these stories, holding these
fragments of history up as a mirror to a dazzlingly complicated past. And
we’ll maybe begin to fill in that all-important gap in our concept of what
Indianness should mean to us and to the rest of the world.
Though modern Indians have forgotten the Deccan, early modern Indians
– in particular, the people of the Deccan Sultanates and Vijayanagara –
looked to its ancient emperors for inspiration. The awe-inspiring temples
they built across the land impressed many an Adil Shahi sultan, who built
palaces in the Chalukya capital of Kalyana and commissioned texts drawing
on the culture of their courts, attempting to connect themselves to the
Chalukyas’ prestige. The rival kings of Vijayanagara used the Chalukya
dynastic crest, the boar, as their imperial standard; a Vijayanagari dynasty
explicitly claimed the title of ‘Chalukya Chakravarti’ (Chalukya emperor)
for some of its members, and attempted to directly or indirectly control
Kalyana; and the empire even reassembled a full-fledged Chalukya temple
tank in the royal centre of the city as a sign of its great antecedents.
All this goes to show that the lords of the Deccan were supremely
influential people in Indian history. They are worth understanding on their
own terms, as the proud rulers of a proud empire untouched by twentieth-
century ideas of how a single ‘Indian’ history should look. But I should
clarify again that this book is not a panegyric to them, and that medieval
India was a dark, violent and unequal place. The medieval lords of the
Deccan were not saintly devotees or noble conquerors or brilliant
masterminds. They were human beings doing the best they could,
responding to personal disasters, trying to get rich and be happy and find
companionship and see beauty in a difficult world. Changing the course of
human history was just an unplanned side effect of all that.

A few brief words on the research that went into this book: It began through
a quest to understand the history of Andhra and Karnataka, the states where
I was born and live in respectively. I was driven to write it by the strange
lack of any accessible modern writing on this period.
I am a researcher, writer and digital public humanities scholar, and still a
historian-in-training. I would not have been able to write this book without
the efforts of generations of academics who have unravelled various aspects
of Deccan history. I’ve read them as comprehensively and critically as I can
in order to assemble their insights here in an attempt to bring the medieval
Deccan to life in a way that resonates with modern readers. The works of
Daud Ali, Durga Prasad Dikshit, A.S. Altekar, G. Yazdani, Adam Hardy,
Sheldon Pollock, Shonaleeka Kaul and Whitney Cox were integral to this
project. I used studies of inscriptional evidence and medieval texts by
Cynthia Talbot, Ronald Inden, Richard H. Davis, K.A. Nilakanta Sastri,
Noboru Karashima and Y. Subbarayulu; studies of the interactions between
India and the eastern Indian Ocean by Tansen Sen, Hermann Kulke and
Kenneth R. Hall; translations of medieval Kannada texts by R.V.S.
Sundaram; works on the political economy of medieval south India by
Jayashri Mishra, R. Champakalakshmi, R.N. Nandi, Aruna Pariti and Meera
Abraham; and my own critical study of inscriptions from the period, as
collected in the Epigraphia Indica, Epigraphia Carnatica and Indian
Antiquary.
A limitation of my research has been my lack of knowledge of Sanskrit
and Old Kannada, which has forced me to rely on translations. Scholars
have also tended to focus on textual evidence to understand this period,
partially due to the lack of systematic archaeological excavations – an issue
highlighted by Jason Hawkes and Derek Kennet. As a partial remedy, I
have attempted to use materials from art history to trace out social, political
and economic trends. I offer challenging reinterpretations of existing
evidence and complicate the romantic portrayals and misconceptions we
have of medieval India, and I explain the reasoning and the evidence for
doing so in detail in my notes.
Some of what I have chosen to depict may be difficult to establish
definitively, given the highly fragmentary state of the evidence. In
particular, it is extremely difficult to reconstruct the lived experiences of
people from the period, which has forced me to indulge in reasonable
speculation. In general, I try to phrase these speculations as tentatively as
possible, usually in the form of questions. Explanations and evidence
supporting these speculations can be found in the notes. In general, they are
based on comparative study from other parts of the medieval world as well
as inferences from Sanskrit literature and inscriptions from across southern
India. I do not claim that my narrative is definitive or the last word on the
subject: this is, first and foremost, a work of popular history.
But I have kept you here for too long. Can you hear a great drum beating,
the sounds of a forest in panic as animals flee from the approach of
thousands of people? The Vallabha calls for our attention. A terrible battle
is about to unfold, and we must witness it. In our ears are the faint echoes of
trumpets, horns and the roar of armies. Our time here in the twenty-first
century is drawing to a close. Let us rush to medieval India, a world that is
in some ways forgotten, but in many more ways a world that is all around
us.
Abu Zayd al-Sirafi, Two Arabic Travel Books: Accounts of India and China, ed. and trans. Tim
Mackintosh-Smith (New York University Press, 2014), 39.
Jayashri Mishra, Social and Economic Conditions Under the Imperial Rashtrakutas (Commonwealth
Publishers, 1992), 190–91. The original text has been edited to improve flow.
The origins of the name ‘Chalukya’ are hotly debated; this is one version presented by Professor
Durga Prasad Dikshit. See Durga Prasad Dikshit, Political History of the Chalukyas of Badami
(Abhinav Publications, 1980), 23–24.
 Part I
Dawn: The Rise of the Chalukyas
 1
Harsha’s Laughter
Winter, 618 CE1

There must have been thousands – perhaps tens of thousands – of men in

the forests of the Vindhya hills near the southern bank of the Narmada river.
Mingled with their shrieks and shouts, the throbbing of their leather drums,
and the blaring of horns and trumpets were the rumbling and trumpeting of
hundreds of war elephants, the neighing of horses, the bellowing of bulls.
The men had come here to kill and be killed.
On the northern bank of the Narmada was another enormous gathering of
men. There must have been hints of strange perfumes and unfamiliar
flowers in the air, the sounds of instruments from a distant land.
Two great armies were on the verge of irrevocably transforming the world
that millions of South Asians inhabited. Two brash young emperors had
come here, to the foothills of the Vindhya mountains dividing the southern
peninsula of the Indian subcontinent. They were about to set in motion a
chain of events leading to one of the most epic stories the subcontinent
would ever see – and later forget.
The older2 of the two emperors, twenty-seven-year-old Harsha (Joy) was
the overlord of the thousands of settlements and hundreds of petty kings of
the lush plains of north India. He was there, on the banks of the Narmada
river, to tear through the Vindhyas and punish3 the other, Pulakeshin II
(Great Lion4) of the Chalukya dynasty. Perhaps he thought he would swat
away this challenger once and for all, as he had brushed away so many
others through a lifetime of cunning diplomacy and brutal wars.
Meanwhile, Pulakeshin was the unlikely overlord of an unlikely power –
the dry, rocky plateau of the Deccan in south-central India, sparsely
populated, torn by war and raids. His reasons for gathering his allies and
vassals and coming here were more complicated.
The stakes were high. Harsha was the most powerful warlord north India
had seen in centuries, the King of Kings of one of the world’s most
expansive, economically productive and culturally sophisticated regions.
Harsha’s ‘camp’ was really a moving miniature city complete with stables,
armouries, stores, workers, attendants and soldiers.5 At its centre, in a field
demarcated with tents and fencing, surrounded by troops and attendants,6
Harsha appeared before his overawed vassals in a silk dhoti, a diaphanous
upper garment embroidered with gold stars loosely tied at his waist, his
chest and head piled with pearls and rubies and flowers.7 Before embarking
on this campaign, this potentate of fabulous wealth had probably spent a
small fortune in a public sacrifice, giving away gold and cows to delighted
priests who promised him all sorts of success in return.8
Pulakeshin, on the other hand, had less of an aura of fame and glory. His
life up to this point had been one of Hamlet-esque drama as he struggled to
keep his father’s throne from rivals within his family and without. Being
from the Deccan, he was almost certainly duskier and shorter than his rival.
The chiefs of this arid land, Pulakeshin included, could hardly match the
spectacular finery in which the flashy lords of the north paraded around in
618 CE. Indeed, when he had first seized his crown barely ten years earlier,9
Pulakeshin’s existence would barely have been acknowledged by Harsha.
But that was no longer a problem. This Deccani upstart had challenged
the older north Indian for control over a wealthy and profitable territory on
the subcontinent’s west coast: Lata, modern-day southern Gujarat. Whoever
controlled it was guaranteed access to the thriving ports of the Indian
Ocean. With this move, far more than merely being acknowledged, he had
made Harsha and his army come to him.

About a quarter-century earlier, where sandy rivers snake through the basalt
rock and sandstone of the Deccan, was born the wailing child who would
one day challenge north India’s most powerful warlord.
As the newborn screamed, the queen’s most senior attendants, members of
prominent aristocratic families,10 rushed to bring the good news to the
eagerly waiting father. On the way, they may have stopped briefly to thank
the nude lotus-headed goddess Lajja Gauri, patroness of mothers and
childbirth.11 The thrilled king ordered the baby’s horoscope cast by the
expert astrologers of his court.12 Drums and trumpets were sounded,13 and
messengers and heralds sent out to proclaim the good news in the
marketplaces of small towns and under sacred trees in tiny hamlets. While
the humble inhabitants of these settlements might not have had much to say
about the matter, the wealthy and who’s who of the royal family’s fortress
town, Vatapi (wind-swelled) – modern-day Badami in Karnataka – sent rare
and expensive gifts for the child, to ingratiate themselves with the king and
queen. The kingdom that Vatapi controlled over was a network of chiefs
and petty rulers, traders and townsmen stretching for a few hundred
kilometres around this town in the Malaprabha river valley, in modern-day
northern Karnataka.14
Even as priests chanted sacred verses in celebratory sacrifices, spooning
ghee into flames, much singing and dancing must have broken out within
the palace’s maze of courtyards and rooms.15 The little prince received the
name Ereya, and probably participated in a series of loud rituals, appearing
before crowds on the auspicious occasions of his first haircut, first birthday,
and so on. Such public rituals were a key source of political power in sixth-
century India: they conferred the prestige and legitimacy crucial to
establishing an individual’s position in a socio-political pecking order. So
these early rituals of Ereya’s life – with their chanting priests and offerings
of clarified butter – established his high status, and declared to the world
that he would head his family’s political network in the future.
The other luminaries of that network – the men and women of the palace,
court officials and the families of assorted Deccani vassal chiefs – also
attended these ceremonies; their participation and interactions served to
renew and clarify their position within the pecking order to the court and
public.16 With their families and attendants carrying gifts of perfumes and
semi-precious stones17 in baskets, their modest processions of bullock carts
and palanquin bearers wound their way through the bustling streets of
Vatapi. Meanwhile, the town’s denizens decorated their homes with leaves,
flowers and fresh thatch, even as dancers, acrobats and storytellers poured
through the streets.
We can only imagine what these aristocrats thought as they watched the
royal couple and their new child. They knew that if the status quo were to
remain, their descendants would inherit their political position unchanged.
Their sons would be Ereya’s vassals, just as they were his father’s.18 Their
daughters would be his wives and wives’ attendants.19 Their families would
pay tribute and appear at the baby’s ever-expanding repertoire of public
rituals as he grew into a king. These vassals were all men (and very
occasionally women) with their own mini-courts, their own mini-armies
and their own mini-administrations, with an array of rights and
obligations.20 They worked within a tangled web of allegiances to advance
their own interests. The most powerful of the lot, such as those close to the
royal clan, must have been pleased enough with this state of affairs, which
generally allowed them lives of leisure.21 There were, however, any number
of opportunities for change, for those who dared seize them. After all, the
royal family itself had set an example mere decades ago.

In the early sixth century, when Ereya’s family were little more than a group
of agriculturists or pastoralists, the Deccan was ripe for military adventures
and conquest. There were several petty chiefdoms and only one or two
powerful kingdoms around. The Kadambas of Banavasi, nestled away
among the imposing hills of the Western Ghats, were one of the latter
category; they were generally deferred to as the overlords of parts of
southwest Karnataka. But their authority outside of their territorial cores
was weak and disputed by dozens of squabbling leaders. In the Deccan,
warfare was a fact of life. Raiding and pillaging were fairly common.
The existence of so many petty chiefdoms, none able to secure a decisive
edge over their rivals, had up to this point prevented the emergence of long-
lasting Deccan superpowers.22 Local groups and leaders were generally left
to their own devices provided they merely submitted some sort of tribute to
other, more powerful chiefs, who in turn paid tribute to a distant dynasty,
such as the Kadambas. Often even that was not forthcoming or demanded:
the vast and unruly lands of the plateau hardly merited the expense it
required to attack and keep control over them.
But that was all about to change.
 We do not know exactly how Ereya’s grandfather’s early career panned
out, merely that he was not born into the splendid circumstances his
grandson was. We don’t even know the origins of his family name: there
seems to have been some connection to the crowbar, chalke, pointing to
very humble origins,23 likely as a leader of an agricultural group in the
valley of the Malaprabha river, in the northern part of the modern state of
Karnataka. But as the mighty Gupta empire of northern India began to fade
away, to be replaced by smaller and more dynamic regional kingdoms in the
Punjab, Malwa and Bengal–Bihar, a new kingdom was soon to rise in the
Deccan.
Crowbars were heated and hammered into spears, as the cultivators and
pastoralists who lived around the valley of the Malaprabha gradually began
to heed the orders of this ambitious chief of the Chalke clan. In the early
years of this man’s career, the agriculturists who followed him were
probably clad in little more than a loincloth, armed with just a spear, bow or
a crudely shaped sickle as they attacked villages first, then towns, then
more powerful warlords, seeking to conquer fertile land and seize loot. With
each successful raid they grew progressively wealthier and more scarred,
decorated with the weapons and jewellery of fallen foes. They traded the
red dirt of the farm for the scented body paints of the wealthy, hired
attendants to oil them, bathe them, cook for them. Perhaps they married
cultured ladies from the families of small-town aristocrats, perhaps they
kidnapped women from the poorest of villages and enslaved them.
A new ‘aristocracy’ must have slowly formed around the rising Chalke
chief, slowly growing in size and confidence, negotiating marriage ties and
political alliances as a new kingdom emerged in this land, protected from
rivals by the imposing sandstone cliffs of the Malaprabha valley. Ereya’s
grandfather seems to have gone about it all with a combination of ability,
ruthlessness and good luck. Agricultural produce and looted treasures soon
swelled the Chalkes’ coffers and those of their officials and commanders.
They were now wealthy members of the landed, leisured elite. Experts in
the performance of rituals, dealers in luxury goods, priests and poets and
musicians and dancers and artists rushed to the Chalke court from the older
kingdoms of south India and the northern Deccan. These foreign specialists
brought with them an aura of legend and ancient tradition, and were a great
way to impress Ereya’s grandfather’s new subjects.
As his horizons began to expand, the Chalke chief decided to make his
intentions and his power clear to other, distant rivals. The Malaprabha river
valley was no longer enough to contain his ambitions. He wished to declare
himself a truly sovereign monarch, equal in rank to well-established and
more prestigious kings, such as the aforementioned Kadambas of Banavasi,
the dominant rulers of southwest Karnataka. His sovereignty over the
swathes of territory he controlled in northern Karnataka needed to be
sanctioned by religion to be recognized by his subjects, vassals and rival
kings. It was time for the chief to become something more – for that humble
name to transform into one that would bedazzle the subcontinent for the
next five hundred years and beyond. The humble Chalke clan would be
transformed into the Chalukya dynasty.
And so, a spectacular sacrifice was organized to elevate him to royal
sovereignty, to higher caste and class status, a ritual badge that would
catapult the Chalke chief over those petty rulers and chiefs surrounding
him. It was time for him and his descendants to rise to the institution of
Indian kingship. He chose to declare this message by performing the
ashvamedha, the horse sacrifice, an undertaking of such complexity and
expense that it was reserved only for the most powerful of rulers.24 In doing
so, the chief offers us a glimpse into how medieval Indian lords used
religion and religious rituals to make potent political claims.
The Ashvamedha, its origins long forgotten,25 had been handed down
orally through hundreds of generations. Bits and pieces were constantly
being added to and removed from it based on the sacrificers’ interest,
knowledge and inclination.26 It had never before been performed in this part
of the Deccan, and to undertake it, the Chalke chief must have secured the
services of foreign priests at great expense.
It consisted of two parts. The first, quite dramatically depicted in epics
such as the Ramayana and Mahabharata, involved setting loose a white
stallion with black spots27 to wander at will for a year, followed by an army.
During this time the Chalke chief and his court would gather at his town of
Aryapura (modern-day Aihole), not far from Vatapi, which the chief and his
sons had been decorating with palaces and shrines funded by the loot from
their military campaigns. Here, the court was supposed to pay for four ritual
offerings daily28 as well as recitations of legend and poetry29 at the site
demarcated to sacrifice the horse later.
The horse, meanwhile, roamed the land, trailed by its accompanying
army, eating and sleeping and galloping as it saw fit (with the sole
restriction of not being allowed to mate, thus preserving its generative
powers for the sacrifice).30 In theory, if it entered the territory of any other
independent ruler, he had the opportunity to capture the stallion – gaining a
great deal of prestige – but only if he also defeated the army following it.
However, if the other ruler didn’t stop the horse from wandering around in
his territory, he accepted its owner’s suzerainty over him. More realistically,
the sacrificial horse was probably deliberately guided to territories that its
owner had already subjugated, or wished to conquer.
The horse’s travels paralleled a king’s enjoyment of his kingdom’s
expanse. The land over which it moved symbolized the territory that the
king controlled, and that of other lesser kings who had ritually agreed that
their right to rule their land stemmed from his. These subordinate rulers
thus served to establish the polity’s pecking order and the king’s pre-
eminent position within it as the king of other kings – raja-dhi-raja. But in
the second part of the ritual, when the animal returned to the owner’s court
– in this case the Chalke court at Aryapura – at the end of the year, events
took a more macabre turn, at least as the ritual books describe it.31
A large grid was demarcated with paint, powders and rope, and
consecrated through the lighting of a fire and the pouring of water.32 Within
it, primary and secondary fire altars were set up according to exact
specifications as to their shape and position – and even the number, size and
shape of the baked clay bricks used to make them.33 Clay pots shaped like
animal and human heads34 were prepared to contain the many ingredients
required for the sacrifice. These would be fed to the flames with pincers,
ladles and spoons shaped like nostrils and collarbones and arms,
metaphorically representing the body of the sacrificer,35 the Chalke chief.
Most importantly, dozens if not hundreds of priests with specific roles – the
chanters, the repeaters, the chief officiants, their assistants – were
assembled for this great and difficult undertaking. Fortunes must have been
spent to bring foreign specialists from far and wide to conduct the sacrifice
and impress the novelty and grandiosity of the ritual upon the chief’s
subjects.
When the stallion returned, the sacrifice began. The royal creature was
rubbed with ghee by the king’s wives, who would also braid its mane and
drape its body with 101 golden beads.36 It was then strangled and its corpse
covered with a sheet. Other animals – 609 according to one source37 – were
also sacrificed along with it, everything from buffaloes to dogs. The
screams of animals were drowned out by the great drone of the chanting of
the Vedas.
After ‘mourning’ the dead horse and fanning it with the yak-tail fan
reserved for kings,38 the chief queen would lie under the sheet39 while the
rest of the king’s women joked obscenely (but ritually) with the officiating
priests and hundreds of female attendants circled them, ‘their hair unbound,
singing, dancing and slapping their thighs’,40 all acts of public taboo
breaking that would never have been allowed otherwise. Under the
supervision of priests, all of these ‘transgressive’ acts were symbolic of the
harnessed powers of fertility, rebirth and renewal.41
After this wild night, in the cold light of the next day, the horse’s corpse
was marked with lines of golden needles42 along which the priests sliced it
into pieces and cooked the best cuts. Its ribs were separated,43 its flesh
roasted and its blood apparently poured into the sacrificial fires from a
dismembered hoof.44 After it was eaten by the attendees of the sacrifice, the
power of the stallion was believed to have entered the royal bloodline.
The performance of such a ritual was loaded with other meanings as well.
At a metaphysical level, the grid upon which the sacrifice was performed,
the vessels used, the ritual slaughter and the verses chanted were believed to
have connected and transformed both the world of mortals and the world of
the gods.45 The world was thus reordered; sovereignty was established,
success and victory were ensured.
This is what some ritual manuals, which were over a thousand years old
even by the sixth century CE, had to say about the Ashvamedha. We have no
idea whether the Chalke chief did all that the manuals demanded; all we
know for sure is that he performed some version of this sacrifice, and made
sure that everyone knew it.46 And going by the inscriptions he issued, the
Chalke chief intended his horse sacrifice to be a declaration of
independence and imperial ambitions.47 There can be no doubt it
accomplished that successfully: his birth name was completely forgotten,
and all the sources we have call him by the new, royal title he adopted.
Henceforth he would be accompanied everywhere by bearers of yak-tail
fans, shaded from the harsh Deccan sun by the bejewelled parasol reserved
for sovereign monarchs: independent kings who bowed to none but the
gods. From now to his death, Ereya’s grandfather would be Maharaja ‘Great
Lion’,48 Pulakeshin I Chalukya.
But though Pulakeshin had now attained the rank of an independent
monarch, he seems to have thought that the term ‘maharaja’ was a little too
generic, a little too common. He thus chose a new title that set him apart
from all his contemporaries, and suited his improbable rise to power. He
declared himself ‘Sri-Prithivi-Vallabha’,49 Fortunate Lord of the Earth, or
Fortune’s Favourite, Earth’s Beloved – the consort of both Sri, the goddess
of fortune, and Prithivi, goddess of Earth, thus the equal of the god Vishnu
himself. It was a title that would become so closely tied to the Deccan that
for the next five hundred years its emperors would be referred to as Sri-
Vallabha, Vallabha-Raja or just Vallabha, Fortune’s Favourite, Beloved
King or Beloved One, in popular parlance as well as their own
propaganda.50
 Chalukya propaganda thus portrayed Pulakeshin I and his heirs as
practically Vishnu on Earth – a claim first made by the Gupta emperors in
north India about two hundred years earlier.51 To call themselves Sri-
Prithivi-Vallabha, Fortune’s Favourite, Earth’s Beloved, was a means of
identifying themselves with Vishnu, who in his mythology ‘embodied the
virtues of kingship’ by ‘preserving the Earth and the prosperity of the
righteous’.52 In these legends, Vishnu earns the right to marry both
Sri/Lakshmi, the goddess of fortune, and Prithivi, the goddess of Earth. Just
as Vishnu was their husband and lord, with the right to command them as
he saw fit, so, too, was the Chalukya king.
Through the rest of his reign, this new Chalukya king, Pulakeshin I,
would continue to bolster the clan’s standing through the performance of a
series of splendid ancient royal sacrifices53 – the Hayamedha, Agnistoma,
Agnicayana, Vajapeya, Bahusuvarna, Paundarika and Hiranyagarbha54 – all
complex and expensive undertakings in their own right, but clearly worth it
for the prestige and legitimacy they conferred. He also ordered the erection
of temples, establishing his privileged relationship with their divine
inhabitants.55 Though these activities were centuries old in other parts of the
subcontinent, and were even known in parts of the Deccan that had been
home to kingdoms, they were unknown in the Malaprabha river valley. By
performing all of these, Pulakeshin I sought to distance and elevate himself
above his subjects, claiming his authority over them came from the gods
themselves, a relationship established and strengthened through these rituals
and titles.
Through all this, a profound transformation in his societal and ritual status
unfurled, giving him access to new sources of political power. As a newly
minted high-caste lord, whose farmer and pastoralist ancestors would never
have had the status or wealth for any projects of this sort, Pulakeshin I now
made sure his contemporaries saw him as nothing less than the ideal Indian
king of theory and legend, building a sophisticated propaganda system all
his successors would follow.
Pulakeshin I, we are told, ‘delighted in dharma’,56 a ‘very specific
ordering of people, places and things’57 that was integral to the Puranic
Hinduism he was seeking to embrace, a socio-political and religious
institution codified in Sanskrit texts, centred around temples and dependent
on an organized system of social and economic relations. Many aspects of
this – not just the aforementioned religious practices – were relatively new
to the denizens of the Malaprabha river valley. Their religious life up to this
point still revolved around megaliths and folk deities of the sacred springs
they depended upon for agriculture and lush forage for their animals. The
concepts of regular tax, feudal lordship, a highly stratified society, caste and
Sanskrit education were relatively new to the vast majority of its
inhabitants. Their lord Pulakeshin I, however, claimed to have studied
Manu, the most prominent of Brahmin dharma thinkers, assorted Puranas
narrating the legends of the gods, and the entirety of the Mahabharata and
Ramayana epics.58 How a petty agricultural chief-turned-king managed to
do all this is unclear; these claims were probably meant to establish him as
cultured enough to belong to the aristocratic classes and castes that had
once looked down upon him, from whose ranks he now looked down upon
his subjects.
As a natural part of his ascension, the Chalukya king also claimed the title
‘Brahmanya’, Friend to Brahmins.59 Brahmins were one of the few groups
with access to elite forms of education in the medieval Indian world,
including the cosmopolitan Sanskrit language, ritual knowledge, and the
business of running a state.60 For nearly a millennium they had dominated
India’s oral and written knowledge systems. From around the first century
CE, they had increasingly diversified from their usual religious roles into
other occupations useful to fledgling states like that of the Chalukyas. One
land grant from around this time describes them as skilled in everything
from fortress construction to battle tactics to administration and performing
sacrifices.61 And so Pulakeshin invited them to settle in Chalukya territory
en masse, granting groups of families land in new Brahmin settlements
(agraharas) or endowing them with the revenue of smaller villages in
return for working in his towns and religious establishments. This was an
important addition to the relatively simple socio-political arrangements that
had prevailed in the Malaprabha river valley before the rise of the
Chalukyas.
Henceforth, Pulakeshin’s children and grandchildren would be educated
by and surrounded by families who claimed or had inherited a similarly
high-caste identity. This new-found status was also useful in other ways.
Chalukya clan members were now eligible for marriage with the oldest and
most prestigious families from within their domains or beyond; they could
be assured of potent ritual services and excellent education; and they could
feel part of a prestigious and well-connected group of people who sat atop
the hierarchy of hundreds of peoples, including tribes, sub-castes, not-castes
and not-yet-castes who toiled over the land.62 Medieval Indian kings were
neither the hapless dupes of Brahmins nor the protectors of powerless pious
folk they are often made out to be in popular culture. First and foremost,
they were for themselves and their own.
Pulakeshin I, his sons, vassals and allies watered battlefields with blood,
constructed new irrigation systems, and fed their loot and agrarian produce
into sacrificial flames and new temples. Religion acted as an engine of
sorts, converting material power into prestige that justified their radical
reorganization of land and people, thus laying the foundations for even
more material power. By around
543 CE,63 the clan had set up their primary capital at the great fortified
citadel of Vatapi (modern-day Badami), commanding the entrance to the
Malaprabha valley, the core of their influence and power. Situated on top of
breathtaking cliffs that could only be accessed through difficult and well-
guarded paths, this was a site of considerable strategic and religious
significance. Not only was it an impenetrable base of operations, but the
area had also been a site of active religious practice for thousands of years.
To this day it is dotted by massive sacred stones, remnants of an Iron Age
culture of ancestor veneration long since forgotten or incorporated into
Hinduism and other Indian religions.
Over the course of a single lifetime, the Chalukya dynasty thus introduced
a series of radical political, social and religious changes that had hitherto
not been seen in the Deccan. The foundations were laid for the emergence
of a new superpower.

Pulakeshin I’s grandson, the prince Ereya, must have grown up hearing
tales of his lineage with pride and wonder. He and other young aristocrats –
some siblings, some cousins, some the children of senior court officials –
wandered between thorny trees and huge boulders of the Malaprabha valley.
Courtiers and attendants were on call for whatever the young prince might
want to know. Perhaps they told him of the ancient megalithic shrines of the
valley, left by those long gone, carved out of massive sheets of rock.64 Once
venerated by long-forgotton ancestors, they were still held sacred by the
Chalukyas’ subjects, who worshipped them alongside the newly introduced
gods in the family’s shrines: Shiva, Vishnu, the Seven Mothers. Perhaps he
was also taught about the Kadambas, who used to control a mighty
kingdom to the south, not far from the western coast,65 who were now on
the verge of total defeat at his father’s hands.
His mother’s brothers, the Sendraka chiefs, may have visited; they were
former vassals of the Kadambas, who had abandoned their overlords due to
the military successes of the Chalukyas. They had married their sister to
Ereya’s father, Kirti-Varman I, the son and successor of Pulakeshin I, as a
symbol of their allegiance. Perhaps Ereya also saw embassies from
Chalukya allies such as the Gangas, the lords of Kolar near modern-day
Bengaluru, whose lands bordered the Tamil-speaking kingdoms of the
south-eastern coast.66 Maybe there were even representations from the
distant lords of the northern Deccan, the Kalachuris. One of their princesses
was Ereya’s grandmother by marriage, a later wife of his grandfather
Pulakeshin I and the mother of his uncle, the dashing general Mangalesha.
It was a fairly charmed existence, no doubt, especially compared to the
hand-to-mouth existence of the cultivators and pastoralists who were the
subjects of his father. Ereya would never have to worry about access to
food, shelter or sex. Sitting in his family’s fort on top of Vatapi’s
spectacular sandstone cliffs, the boy could easily have convinced himself
that the world as far as he could see was his. He might, at best, have to
share it with his brothers. One of them had been born with a crooked back.
The other was still just a baby.
Had the young prince’s life continued this way, he would gradually have
been entrusted to more senior teachers, studying war, politics, art, languages
(especially Sanskrit, that ultimate symbol of royal sophistication) and
religion alongside a cadre of other princes and boys of similarly elite
backgrounds.67 Sanskrit and formal religious education – indicative of the
status they had earned through their conquests and performance of
prestigious sacrifices – were still a novelty for his family.
Ereya would have learned to command and fight, got married, had some
children, worked as a governor or general, and eventually been consecrated
king. His father was still very much in his prime, having developed
Chalukya influence to the point where it was a major regional power. Their
core territory stretched over a vast radius (for the time) of about 240
kilometres,68 and from there the Vallabha led armies to defeat and subjugate
the lords of the southern Deccan (modern-day Karnataka) – smashing the
Kadambas and, as mentioned, marrying a princess of the Sendrakas, former
vassals of the Kadambas. This was a major upset by the standards of the
day: the Kadambas were famed patrons of architecture and learning, and
had been connected to pan-Indian developments since well before the
Chalukyas even existed. They had been the first southern Indian kings to
use courtly Sanskrit and systematically perform public rituals in the early
fifth century CE.
The Vallabha also sent raids west, across the hilly rainforests of the
Western Ghats, as far as the Konkan coast69 (in modern-day Maharashtra).
Some of his defeated rivals were reinstated as his vassals, others lost their
territories to Chalukya clan members or new lineages of vassal kings related
to the Vallabha through marriage. Under Ereya’s father’s rule, Vatapi grew
and bloomed, and new cave temples were excavated into the sandstone
cliffs – an innovation likely adopted from the eastern or north-western
Deccan. Religious and artistic patronage thrived alongside vicious politics
as families competed for the royal largesse.
But then, the king died suddenly in 597 CE, when Ereya was barely an
adolescent.70
This was a critical moment in the story of the Chalukyas. These former
peasants had risen to a position where they were the envy of the Deccan’s
powerful chiefs. At a moment when these terrified lords had been falling
over themselves to submit to, ally with or marry into the Chalukya clan, this
new-found dominance was suddenly thrown into question. The Kadambas,
arguably the oldest and most powerful kingdom in the southern Deccan, had
only just been defeated in war and forced to submit. The Chalukyas seemed
to have regarded them with a mix of envy and hatred, and spent much blood
and money attempting to grind down their power – even, at one point after
defeating them, ordering buildings raised over the foundations of destroyed
Kadamba edifices.71 Less of a challenge were the Mauryas of the Konkan
and other chiefs such as the Nalas, also subdued in war.72 Other alliances
(especially those sealed with marriages to the dead king, such as the
Sendrakas73) were now in question, as were other diplomatic treaties and
political arrangements the Chalukyas could no longer credibly uphold. A
child king was not going to be able to demand the submission of grizzled
old warlords with swords older than he was.

Under these circumstances, the dead king’s younger half-brother, a man of

considerable dash and verve, now seized the throne as Maharaja
Mangalesha. Ereya, his siblings and their mother(s) were incorporated into
Mangalesha’s new court – demoted and moved to less impressive quarters
in the palace, but otherwise unharmed. With the prospect of a child ruler
gone, and with a capable and energetic commander in place, the Chalukyas
retained their geopolitical dominance over their neighbours. Following in
the footsteps of his father and elder brother, Mangalesha developed the
Chalukya armies into one of the Deccan’s most capable and feared fighting
machines. Incidentally, the Chalukyas had an odd way of keeping their
armies in shape: defeated generals were forced to wear women’s clothing74
and mocked in open court, which apparently ensured they would either win
battles in future or kill themselves to avoid public humiliation. Chalukya
vassals and defeated rivals – Kadambas included – very wisely decided not
to challenge the new king, whom they had probably already met in battle.
Mangalesha, however, could not afford to rest. Though his accession had
stabilized the fledgling polity, the expected ‘duty’ (dharma) of an Indian
king – according to the Sanskritic court culture his family was promoting –
went much further than providing stability and security to his subjects. In
theory, a truly successful king was so good at establishing and nurturing the
dharma that ‘he and his domain overflowed, so to speak, into the adjacent
kingdoms’.75 In practice, powerful Indian polities sought advantage over
their neighbours by ‘overflowing’ their armies through their borders. This
confluence of ideology, war and the personal avarice of kings is aptly
summarized in a quote from the Manusmriti, used by the scholar Richard
Davis in his study of violent looting in medieval India: ‘Chariots, horses,
parasols, money, grain, cattle, women, all kinds of goods, and base metals
all belong to the one who wins them.’76 Similar declarations from other
texts were probably familiar to aristocrats like Mangalesha, should they
require religious sanction for self-aggrandizement through war.
The dynasties of the northern Deccan, the seat of the plateau’s oldest
kingdoms, are unlikely to have expected a serious attack from the
Malaprabha valley, which had been little more than a backwater until
recently. The most powerful dynasty there, the Kalachuris, had married one
of their daughters to the first Chalukya king, Pulakeshin I, perhaps
assuming this great honour would be enough to keep these upstarts quiet.
On the contrary, this lady – Mangalesha’s mother – seems to have actually
encouraged her son77 to attack her brother, the Kalachuri king.
The new Vallabha now led a marauding raid into his uncle’s territory. The
Kalachuri army of cavalry, infantry and elephantry was routed and driven
off the battlefield by Mangalesha’s attack.78 The Chalukyas then sacked the
Kalachuri camp, capturing much treasure. The defeat was wholly
unexpected: the women of the Kalachuri king’s household had accompanied
the army to the battlefield to witness their anticipated victory. Mangalesha
captured the women, his panegyrists declaring that ‘Mangalesha enjoyed
the Kalachuri ladies along with their prosperity’.79 These women, likely
from noble families with their own rank within the Kalachuri political
network, were forcibly incorporated into Mangalesha’s court and suffered
(at the very least) depression and homesickness,80 and probably humiliation,
emotional and physical abuse, or sexual assault at the hands of his courtiers
and attendants. The Kalachuri king himself retreated, living to fight another
day. This raid had been a humiliation, but it had hardly shaken the
Kalachuris’ grip over the northern Deccan.
Mangalesha returned to Vatapi in triumph, lavishing his supporters with
loot. Even gods that were seen as friendly enough to him got a cut.81 Now,
lured by the potential of all this treasure, an array of service providers
began to migrate to the Malaprabha river valley.82 Sculptors came to work
for the Chalukyas, some of them from Kadamba and even Kalachuri
territory, bringing their distinct styles with them. On the Vallabha’s order,
splendid new cave temples were now cut into the soft sandstone cliffs of
Vatapi, dedicated to the god Vishnu the Preserver, the archetypal deity of
kings.
The inauguration of Mangalesha’s new projects, coming soon after
Pulakeshin I’s series of prestigious and painstaking royal sacrifices, is a
clear sign that sixth-century Deccanis were attuned to the subcontinent’s
latest innovations and trends (such as the use of Brahmin settlements and
religious propaganda to support royal power) and could harness them to suit
their own context. Sacrifices and religion and temples hardly seem
‘innovative’ to us today, but as far as the Chalukyas were concerned, they
were introducing the most modern of foreign ideas to their homeland to
tighten their grip on wealth and power.

What young Ereya thought of all this is unknown. The Chalukya kingdom
was doing well, its leader was clearly able and well liked, it was
safeguarded from its enemies, its vassals did not challenge it. Growing up
in the town of Vatapi as the king’s nephew must have been a pleasant
experience, and he could have relied on urbane company and an excellent
education. Had he not pressed his claims to the throne, Ereya may have
been guaranteed a comfortable career, perhaps as a general or governor, and
could eventually have retired into peaceful obscurity as the leader of his
own minor branch of the Chalukya family. But kingship was irresistible to
any ambitious and capable prince.83
Trouble began to brew as he came of age. Ereya was certainly well
connected on his mother’s side – she was a Sendraka, a family who were
once vassals of the Kadambas of Banavasi – and had powerful friends in his
own right. It is possible that he was betrothed or married to a princess of the
Gangas of southern Karnataka, and he probably knew many senior
Chalukya officers, generals, vassal lords and allied chiefs personally.
Perhaps the danger of his position began to dawn on him as Mangalesha’s
sons, his cousins,
grew up and began to challenge his seniority among that generation of
Chalukya princes. Perhaps the king saw a popular nephew who could be set
up as a power centre opposed to him and his sons.84 We will never know
what bitterness and acrimony erupted in the court.
What we do know is that Ereya was forced to flee Vatapi,85 perhaps with
little more than a few companions. He now desperately began to rally
support for a coup86 against his uncle before Mangalesha’s loyalists could
seize and defeat his ragtag group of companions. As always, the lords of the
Deccan gleefully pounced on this opportunity for personal advancement,87
including the Bana family (based near modern-day Kolar88) and the Alupa
family (based near modern-day Shivamogga89). A coalition of these minor
chiefs soon began to gather around the prince. The prospect of installing a
young, pliable ruler who was beholden to them for his throne must have
been tantalizing, despite the risk of retaliation from the reigning Vallabha,
Mangalesha.
Meanwhile, others attempted to independently break away from Chalukya
authority, including – dangerously – both the Kadambas and the Gangas.90
Even closer to home, a vassal ruler in the Konkan, who had been appointed
by the Chalukyas after some preliminary campaigns in the region, began to
put on the airs of an independent sovereign. Mangalesha had to move
immediately to nip this idea in the bud. The rebel had fortified some islands
off the coast, apparently confident that they were impregnable. The
Vallabha, like some sort of Indian Caesar,91 built a bridge of boats, crossed
the tossing seas, and slaughtered the inhabitants.92
Ereya, meanwhile, moved from chiefdom to chiefdom. His entourage
trekked through the Deccan, through vast fields burning under the
unrelenting sun, through forests shrill with the call of crickets, winding over
hills and between cliffs, feeling the blistering summer heat and the rains of
the monsoon on their skin as they watched the sun rise and set on
unfamiliar vistas. The rebel prince needed to travel everywhere he could
think of in order to find potential supporters: visiting settlements of mud
and thatch and manors of wood, negotiating future rewards in return for
immediate military or political support. Everything from loot to territory to
offices, titles and marriage alliances were likely on offer. It must have been
a difficult and stressful endeavour, and the Chalukya prince probably spent
more than a few nights sleeping under the starry skies surrounded by a
small company of guards, hearing the distant roars of tigers in the dark. But
slowly Ereya’s camp grew. He learned to fight and command, to flatter and
persuade, to threaten and crush, to loot and reward as he grew into royalty
the hard way. Soon, Ereya was in open rebellion, warring with his uncle and
cousins, seeing on battlefield after battlefield the bloody, broken corpses of
soldiers he may have recognized from Vatapi, handing out jewellery and
weapons to his new vassals, as men flocked to his banner. A final
confrontation with Mangalesha inched ever closer.
In 609 CE, Pulakeshin I’s son and grandson clashed near a village called
Nadanuru. A few confused hours of dust and turmoil later, the victor
summoned Nadanuru’s elders to attend on him. Under a makeshift roof of
wood and cloth, he offered them an auspicious gift of land and the revenue
of all taxes collected in the vicinity: perhaps the first grant he had ever had
the authority to make.93 The sacrifice of a horse’s life had once made
Ereya’s grandfather Pulakeshin I in the eyes of the world. The sacrifice of
his uncle’s life had made the young prince Pulakeshin II in the eyes of the
world.

If Pulakeshin’s neighbours and new vassals had thought the young king
would be less assertive or militarily capable than Mangalesha, they were
soon proven spectacularly wrong. Pulakeshin II would respond to every
challenge with deviousness, resolve and a firm grasp of both politics and
geopolitics.
Almost immediately after his accession, with his authority over the
Chalukya core territories – let alone its extended network of current and
former vassals – still in question, two adventurers from the northern Deccan
attempted to dethrone him.94 Pulakeshin’s response (according to an
inscription by his court poet, found today on a temple atop Meguti hill in
Aryapura/Aihole, near Badami) was straight from a political manual,
involving a mixture of conciliation, gifts, dissension and force – the four
strategies Indian princes were taught to apply to potential problems. The
young king bribed one of the attackers to join his side, with his ‘host of
elephants’,95 rewarding him with high honours and titles. Reinforced,
Pulakeshin then pounced on the other and routed him.
Of course, the entire inscription may have been composed to portray
Pulakeshin II as conforming to theoretical notions of a king in Sanskrit
literature, just as his grandfather Pulakeshin I claimed to have studied the
Vedas. But this careful curation of his royal image, coupled with bold and
decisive action to protect his power, was a sign of things to come.
Pulakeshin II now turned his attention to the huge challenge of restoring his
clan’s shattered political network. A message needed to be sent and his new
followers rewarded with spoils. And so his eyes alighted on his family’s
ancient foe, the Kadambas. After two generations of war, the realization
seemed to have dawned that the Deccan was not big enough for both of
them.
Denizens of the fortress of Banavasi, the Kadamba capital, were generally
accustomed to watching white geese swim in the Varada, a tributary of the
Tungabhadra river which served as the town’s moat.96 But all those
beautiful birds were put to flight by the roar of the besieging Chalukya
army, which surrounded Banavasi’s earthen ramparts, cut off food supplies,
and possibly even blockaded the river with boats. Flocks of serene white
geese were replaced by a turbulent ocean of tents, banners, a host of
loincloth-clad infantry, with their gleaming spears and painted shields, and
the looming silhouettes of war elephants. The invading army was so large,
the Chalukyas tell us, that Banavasi, a fortress on land, looked like it was a
fortress on the sea.97 Engulfed by the Pulakeshin tsunami, the main
Kadamba line abruptly (and likely bloodily) vanishes from the historical
record. Their lands were seized and handed to Pulakeshin’s mother’s family
– the Sendrakas – as well as other clans, as a reward for their support in his
war for the throne and to ensure their future loyalty.98 Clearly, those who
sided with the Chalukyas would be rewarded splendidly sooner or later.
Those who did not would be crushed.
The Gangas were among the first to recognize this fact. Pulakeshin II of
the Chalukyas of Vatapi was evidently a son-in-law any king could be
proud of. Possibly through a renewal of an engagement brokered during his
father’s reign, a Ganga princess was married off to him.99 To her family, the
marriage was an investment in a potentially rosy collective future. After all,
any children she might have could be expected to help the Gangas should
they ever come to the throne, or act as a powerful pro-Ganga lobby at court
if they did not.
A whole host of other dynasties followed in the Gangas’ footsteps, tying
their fortunes to the re-emerging Chalukyas through the exchange of
daughters and promises of fealty and tribute. Pulakeshin would also have
taken care to cultivate or replace his uncle’s senior commanders and
officers, granting them titles and wealth to ensure they stayed loyal. The
court was probably reorganized so that the new Vallabha’s partisans
dominated, and the bases of Chalukya power were once again secured. A
true pan-Deccan superpower was emerging. Pulakeshin II was no longer
content to call himself Maharaja: he and his successors would be
Maharajadhirajas, Great Kings of Kings, emperors.100
Pulakeshin II now set his eyes on the grand prize that was the northern
Deccan – that heartland of empires past and present, watered by the great
river Godavari and its tributaries, its shoreline connected to the rich trading
ports of the north and west. This was not a new target for Chalukya
attention: both his father and uncle had raided and conquered bits of it,
especially along the Konkan coast, but their authority was fading
(Pulakeshin’s own rebellion against Mangalesha had encouraged a
Chalukya vassal in the region to make a bid for independence, as we saw
earlier). Pulakeshin II would now go about this project much more seriously
and systematically, setting himself on a path that would, one day, lead to his
meeting Harsha on the banks of the Narmada river.
India’s west coast had been a major trading region of the southern
peninsula for centuries. But it had waned somewhat in importance since the
heyday of the Roman Empire in the first and second
centuries CE, when fleets of hundreds of ships would arrive every year from
the Roman province of Egypt to buy Indian luxuries. Now, less than a
hundred years after the collapse of the Western Roman Empire, the Eastern
Roman or Byzantine Empire was beginning to try to restore trading links to
India and China, and the Sassanian dynasty of Persia began to expand its
own presence in the Indian Ocean.101 The Chalukyas seemed to have
realized that the coasts were becoming increasingly wealthy and could not
be left in hands that were not absolutely loyal to them.102 Control over
lucrative trading ports was a grand prize to hand out to loyalists, a
guaranteed source of trade income, and most importantly, a source of plush
luxury goods that were crucial for maintaining a court of sufficient
splendour for a potentate of Pulakeshin’s rank.
The great island city of Puri, ‘the Lakshmi of the Western Ocean’,103 once
stood near the Elephanta island off modern-day Mumbai. It was a serious
threat to Chalukya authority in the Konkan, and it would need to be taught
to heed the might of the lord of the Deccan. Pulakeshin found a way to
attack the fortified island, perhaps hiring merchant ships to transport his
infantry from the mainland to the island, or conscripting fisherfolk to build
galleys104 for his attack. Soon after, his ships’ hulls sliced through the
choppy waters of the Arabian Sea in squadrons, ‘like arrays of rutting
elephants’,105 perhaps making an amphibious landing close to the city walls,
or breaking into the harbour at an unexpected time, surprising its sleeping
guards. It was undoubtedly a bloody and difficult fight – but, like the fall of
Banavasi, the dramatic collapse of this seemingly impregnable citadel sent a
message. The Konkan would henceforth be ruled by Pulakeshin’s maternal
uncle, a Sendraka prince,106 a family that had time and again proven its
usefulness to this nephew. As a Sendraka from faraway southern Karnataka,
this uncle had no power base of his own in coastal Maharashtra and thus
had all the more reason to stay loyal to Pulakeshin if he wanted to survive.
The coast secured, the young emperor now turned his attentions to the
interior of the northern Deccan, dotted by ancient Buddhist monasteries
(such as Karle, near modern-day Lonavala), dating back to the origins of
urban settlements and flourishing trade in the region nearly half a
millennium ago.107 Through this land snaked the ancient highway
connecting the markets of northern and southern India, called the Dakshina-
patha – the Southern Path. (The modern term ‘Deccan’ comes from
dakshina/dakkhina, south.) Nashik, the seat of the Deccan’s earliest
empire,108 the Satavahanas, was developed into a de facto second Chalukya
capital.109
However, in order to truly rule the northern Deccan, Pulakeshin could not
be satisfied with merely defeating the powerful Kalachuris and sacking
their treasures, as Mangalesha had done. He systematically integrated this
foreign land into the Chalukya imperial network, using the tools of politics
and economics, far more long-lasting than ephemeral and costly military
activities. Pulakeshin handed land to his collaborators there, the terms of the
grant (a matter of public record) incorporating local religious formulae.110
This suggests that he took care to respect local traditions and cults,
participating in the region’s seasonal festivals while also performing the
annual royal rituals of the Sanskritic world, increasingly becoming
recognizable across the subcontinent due to the activities of kings such as
Pulakeshin. Chalukya prestige and patronage continued to bring ever more
artisans and religious teachers from the northern Deccan to the Malaprabha
valley,111 binding the two socio-political networks closer together.
This policy of integration did not take long to bear fruit. Pulakeshin II’s
influence soon spread far and wide, and players far away from the nucleus
of his power were bending over backwards to please him. For instance, as
far as 320 kilometres further inland from Nashik, the centre of Chalukya
power in the northern Deccan, were the volcanic basalt cliffs of Ellora, a
flourishing pilgrimage centre. Here, ‘some individual or group with
importance’ took it upon themselves to commission a cave temple
celebrating Pulakeshin and the Chalukyas,112 its sculptures unmistakably
similar to those at Vatapi. For an actor so distant from the Malaprabha
valley to invest the time and money needed to execute such a work is a sign
of how profitable joining this new polity could be, suggesting an
unprecedented northward expansion of the Chalukya clan’s influence.
These dramatic developments were watched with great alarm in western
India, particularly in the region around the Gulf of Khambat (earlier
Cambay), the southern part of modern-day Gujarat, the kingdom of Lata.
The Latas would certainly have heard about Pulakeshin’s imperial
ambitions through groups of itinerant traders and workers, if not directly
through diplomacy or spying. Their territory was dotted with excellent port
facilities, and was a major entrepot for maritime traffic travelling via the
Persian Gulf. It was an even more lucrative prize than the Konkan.
Pulakeshin, a man who ‘treated his neighbours with contempt’,113 seems to
have seen the area as within his rightful geopolitical sphere of influence,
and its rulers sent him tribute, perhaps hoping it could buy them peace with
the powerful lord of the Deccan. This would be the spark that lit the powder
keg, for Pulakeshin was not the only Indian emperor with an interest in
controlling Gujarat.

Harsha, Great King of Kings of north India, was an unconventional sort of

emperor. His family, like the Chalukyas, came from a humble background,
though he was a descendant of merchant townsmen114 rather than cultivators
or pastoralists. Unlike the Chalukyas, this clan’s military capabilities had
been forged in the heat of north Indian wars, which meant they had not only
fought Alchon Huns from Central Asia but had also been entangled in the
dramatic political developments of the previous century attending the final
collapse of the Gupta empire – a saga of espionage, conspiracy, betrayal and
war.115 Suffice it to say here that by 606 CE, Harsha, supposedly only fifteen
years old, ended up on the throne of Kannauj, north India’s most wealthy
and prestigious city at the time,116 after the sudden death of both his elder
brother and his brother-in-law.117
By 618, at which time Pulakeshin II had subjugated most of the Deccan
and was throwing his weight around Gujarat, Harsha had successfully
forced most of north India to submit to his imperial formation, spreading his
influence across its thousands of thriving settlements ruled by hundreds of
vassal dynasties. Over this time, his army is supposed to have grown from
5000 to 60,000 war elephants, from 2000 to 100,000 cavalry,118 and
incorporated innumerable masses of infantry. Even if this is an
exaggeration, Harsha was clearly seen by his overawed contemporaries as
one of the subcontinent’s dominant rulers, if not the dominant ruler. In
comparison, Pulakeshin’s territories had far fewer large towns and
productive agricultural regions; few chiefdoms and kingdoms could afford
to build massive irrigation works in the Deccan yet. The arid plateau also
had many more pastoralists and herders than the lush plains of the north,
and was in general more sparsely populated. We do not know how much
military might Pulakeshin could summon, but it is doubtful that he could
have matched Harsha’s.
 Having expanded his influence as far as Bengal, commanding the ports of
India’s east coast, Harsha now wanted to control its west coast as well,
potentially linking his territories to flourishing coastal trade routes in both
directions. This threat may have been the trigger for the Latas to send
tribute to Pulakeshin in the first place. If so, it was a dangerous gambit: as
one scholar puts it, ‘the sovereign of the Deccan must have considered to be
his natural birthright … unlimited access to the ocean ports of the Gulf of
Cambay [Khambat]’.119 Apparently deciding that tribute was not enough,
the Chalukya emperor now attacked and conquered a part of Lata (southern
Gujarat) and installed a relative of his as ruler.120
Up to this point, Harsha must have watched with growing surprise and
interest as his audacious rival survived every challenge thrown at him, but
the invasion of Lata must have been the last straw. It was the winter of 618
CE,121 little more than eight years since Pulakeshin II had come to the throne.
There would be no better chance for Harsha, who had been emperor for
twelve years, to put him in his place. After consulting astrologers and
calculating the outcomes of the campaign, planning out the route to be
followed, rallying his armies and vassals, and arranging for supplies along
the way, the order was given.
‘At the close of the third watch, when all creatures slept,’ Harsha’s court
poet tells us, ‘the marching-drum was beaten with a boom deep as the
gaping roar of the sky-elephants. Then, after a moment’s pause, eight sharp
strokes were given upon the drum, making up the number of leagues in the
day’s march.’122
In the dead of night, Harsha’s camp – like those of many Indian
monarchs, a capital city constantly on the move through his vast empire –
roared into activity, trumpets and horns blaring, officers ordering their
troops about, war elephants brought out of their stables, hosts of cavalry
horses shaking their manes as they were roused from slumber. Tents were
uprooted and packed, their pegs packed into bags, fastened along with water
bags, awnings and screens on the backs of transport elephants and
camels.123 Foragers scurried around gathering heaps of fodder for the army’s
animals, while the poorer families of the area hurried to steal the grain left
behind, bemoaning their tiny grass huts uprooted by the enormous
temporary city.124 Through the night, an enormous din of sound and clouds
of dust were raised, and by the next morning, the mass of thousands of
people and animals was ready to move. At its forefront were Harsha’s
vassals, seated on beautiful female elephants, ‘riders holding up bows
striped with gold leaf, swords grasped by servants, chowries [yak-tail fans]
waved by betel-bearers, sheafs of javelins in cases under the charge of those
who sat in the back, and saddles curving with scimitars and bristling with
golden arrows’.125
Before them, on the tallest and most beautiful elephant at hand, shaded
from the red sun of the dawn by the enormous white parasol reserved only
for emperors, appeared Harsha-Vardhana Pushyabhuti. After first
acknowledging the bowing ‘company of feudatory kings’ according to their
rank with gestures and compliments126 as per Indian courtly protocol,127 he
would have carried out a review of the army. And then these hundreds of
thousands of feet and hooves began their march south.
In the Deccan, Pulakeshin was doing the same, his hereditary guard
drilling with sharp spears every morning as troops from his vassals and
allies, as well as tribes of pastoralists and forest dwellers, began to trickle
in. These tribal warriors were generally outside the system of king and
vassal lord but could be enticed to join military campaigns. Far from being
‘primitive’ in any way, Adivasi tribes were keen observers of cultural and
political trends, and often played pivotal roles in the affairs of sedentary
states. They saw the settled peoples of the subcontinent as a useful source
of prestige, tribute and luxury goods,128 and sometimes worked for them as
mercenary troops and scouts. Harsha was already one of the subcontinent’s
most famous monarchs; to face him in battle was a great opportunity for
loot and glory, and Pulakeshin would need the skills of these fearsome
warriors to even the odds.
Though little is known about precisely how ancient and medieval Indians
fought battles, contemporary manuals describe huge, heavy formations and
counter-formations (vyuhas) organized according to complex rules.129 North
India, with access to vast amounts of infantry, elephantry and cavalry, was
especially suited to this sort of fighting. The Deccan could not muster or
feed the same numbers of infantry, nor did it have access to the overland
routes of the horse trade, emerging as they did from Central Asia. If
Pulakeshin had fought Harsha in north India, his army would easily have
been surrounded and crushed. But in 618, to punish Pulakeshin for his
audacious move on southern Gujarat, Harsha had to cross the Narmada
river into the Deccan – which tilted the odds in Pulakeshin’s favour. With
his control over the northern Deccan solidified by his policy of economic
and political integration, Pulakeshin could now easily scout out Harsha’s
route of attack and contest the northern emperor’s attempts to cross the
great river. Even if Harsha’s probably larger army were to cross the
Narmada, this unwieldy force of infantry, cavalry and elephants had to enter
the thickly forested Vindhya foothills in order to capture or defeat
Pulakeshin – potentially negating their advantage in numbers.
 In addition to local intelligence and favourable terrain, Pulakeshin had
access to the tanks of the ancient world.130 elephants. These deadly creatures
possessed the mass and sheer shock value needed to break almost any
immobile enemy formation sent against them. Indian kings could deploy
elephants in a variety of ways depending on the kind of logistical
capabilities and strategic requirements they were working with. The
cheapest, most plentiful and most poorly trained beasts were rarely allowed
into the front lines. Instead, they carried teams of common soliders ‘armed
with knives, daggers, pots of oil, stones and other weapons and missiles’131
and acted as mobile archery platforms. Aristocrats, who could afford larger
elephants and trained staff, carried gilded bows with assistants to hand them
extra missiles, as well as an expert mahout to guide the animal and goad it
into attacking. Their better-equipped elephant corps, wearing howdahs, tusk
spikes and warpaint, could be used both at range and in close-quarter duels,
when the animals’ riders would swap their bows for javelins and long
elephant lances.132 Should a war elephant be unleashed on an enemy
formation, it would be deployed with foot soldiers to guard its legs and
protect its rear whenever it took a breath from the slaughter.133 In a pinch,
these immense creatures – specially captured, trained and fed to fight –
could even be armoured, gathered into squadrons and used as heavy
cavalry, as well as to screen formations of infantry and horsemen.
In the Vindhya foothills,134 though, none of those elaborate manoeuvres
were possible. But the elephants still played a significant role.
 In the lush green foliage of a Deccani winter, dozens of Chalukya
elephants must have cleared a path through the moist undergrowth, soon
churned into mud by miles of marching soldiers. Thousands of bare feet
squelched in the mud and trampled down the grass and insects. As the sun
bathed the Vindhyas, Chalukya scouts – perhaps allied forest tribes – may
have fanned through the seemingly endless forests, using sounds and drums
to signal the approach of the enemy. Harsha’s army was probably carefully
tracked and avoided until the very last moment. Ambushes and sneak
attacks may have been planned to stretch their supply lines and worsen
morale. Thickets might conceal ambushers with bows and spears, while
hills and ravines might hide elephants to trample incautious groups of
infantry. When Harsha’s army, like the Mughals centuries later, was
disoriented and exhausted in the unfamiliar Deccan, Chalukya forces may
have aimed to lure them into a brutal, decisive confrontation.
Pulakeshin’s troops were ideally adapted to close-quarter combat. His
elephants, in the hundreds, were fed huge quantities of alcohol before
battle, spikes fitted to their tusks, great bells hung around their necks,
howdahs tied to their backs. The animals were then bunched together into
close formations for massed charges, brave mahouts desperately trying to
guide them with the fearsome ankusha, a sharp elephant goad dug into their
eyes and temporal glands. To the hypnotic beating of battle drums, the
elephants were followed by bands of elite hereditary warriors wearing
loincloths and minimal armour, also drunk on alcohol.135
Harsha’s court poet describes his infantry as wearing topknots and spotted
red coats, ears adorned with ivory rings.136 The north Indian emperor
commanded them from the back of his elephant Darpasata, a massive
animal whose head was adorned with a ‘crested crown of gold’.137 But
beyond this, there is little we know for certain of the confrontation between
the two young emperors. Pulakeshin’s inscriptions, and those of other
medieval royals, paint pictures of horrifying battlefield violence. They
describe elephants colliding, tusks gleaming with blood. The hulking beasts
would attempt to force each other to topple, their senses dulled by alcohol,
as their riders stabbed each other with lances. The screams of men trampled
underfoot and gored by tusks, the squeals of fallen elephants whose bellies
were pierced by the cruel spears of the Chalukya infantry, must have filled
the forest.
But eventually – perhaps after months or weeks, or perhaps after a few
disastrous ambushes and confrontations – Harsha seems to have realized
that he would have to cut his losses. Forcing the Deccan to accept his
suzerainty was not worth sacrificing an entire army. Perhaps he intended to
renew the conflict another time, but that time would never come, as
rebellions and easier pickings called his attention east and kept him there
till the end of his reign.
Pulakeshin II, unlikely lord of the Deccan, had defeated the subcontinent’s
superpower.

And so, as Harsha ordered his retreat, as the Vindhyas reverberated with
the sound of retreating drums and the piercing blast of victory trumpets,
Pulakeshin was left to giddily proclaim his astonishing victory.138 As a
Chalukya court poet put it, Emperor Harsha, whose name meant Joy, had
lost his laughter in the Deccan. All of a sudden, it was clear not only to
Pulakeshin’s vassals, not only to his family, but to the entire subcontinent,
that the Deccan had arrived. As kings and emperors reeled from the news,
Pulakeshin claimed the splendid title of Parameshvara,139 Paramount Lord.
Lata and the northern Deccan were his. He now turned his attention to the
rest of India south of the Narmada.
The date of the battle between the Chalukya ruler Pulakeshin II and his rival Harsha has relatively
recently been conclusively narrowed down to this window. See S.S. Bahulkar and Shilpa Sumant,
‘The Bijapur-Mumbai Copperplate Grant of Cāl.ukya Ruler Pulakeśin II, Dated April 04, 619 ce,
Mentioning His Triumph over Emperor Harṣavardhana,’ Annals of the Bhandarkar Oriental
Research Institute XCIII (2012): 205–09.
This is assuming that Professor Durga Prasad Dikshit is correct in inferring Pulakeshin II was five
years old in 597, which would have made him about twenty-five in 618. See Durga Prasad Dikshit,
Political History of the Chalukyas of Badami (Abhinav Publications, 1980), 54. However, this is
based mostly on his assumption of an ‘appropriate’ age for Pulakeshin to overthrow his uncle
Mangalesha. Meanwhile, Harsha was (according to his court poet Banabhatta) just fifteen when he
became the king of Kannauj in 606, which would have made him twenty-seven in 618. This is,
again, doubtful, because it would have made Harsha’s sister, the widow of the king of Kannauj and
one of the main drivers behind his accession to that throne in the Harshacharita, just twelve years
old at the time. We cannot be absolutely certain of their ages.
This book follows Bakker’s depiction of this conflict as arising from the two clashing over the
control of Gujarat. See Hans Bakker, The World of the Skandapurān.a: Northern India in the Sixth
and Seventh Centuries, Groningen Oriental Studies, Supplement, Vol. 4 (Brill, 2014).
Alternative translations include ‘Lion Maned’, this is the translation used by Professor Dikshit. See
Dikshit 1980, 33.
Prithwis Chandra Chakravarti, The Art of War in Ancient India (University of Dacca, 1941), 99–102.
Daud Ali, Courtly Culture and Political Life in Early Medieval India (Cambridge University Press,
2006), 38–43.
Vasudeva S. Agrawala, The Deeds of Harsha: Being a Cultural Study of Bana’s Harshacharita
(Prithivi Prakashan, 1969), 62–63.
Chakravarti 1941, 97.
Ali 2006, 54. Citing Epigraphia Indica VI, 6. Hereafter EI.
Agrawala 1969, 88. To what extent can the Harshacharita be used as a source for the life that
Pulakeshin II may have lived? It is probably a useful source, with some caveats. Harsha and
Pulakeshin were almost exact contemporaries, and at the time of their birth neither seemed
destined for imperial sovereignty. However, both were part of families that were ‘royal’ and thus
likely steeped in the same cultural milieu (Ali 2006) of the Sanskrit cosmopolis. See Sheldon
Pollock, The Language of the Gods in the World of Men: Sanskrit, Culture and Power in
Premodern India (University of California Press, 2006). For Pulakeshin’s family this culture would
have been only recently adopted.
References to horoscopes being cast, announcements being made, recitals of song and dance, vassals
visiting and so on are present in the Harshacharita. These are also tropes associated with a royal
birth in most Sanskrit courtly literature of the period – which means that the royals of the Deccan
were likely well aware of the ‘proper’ rituals attending the birth of a crown prince. This
assumption is further bolstered by the Chalukyas’ performance of many major Vedic and Puranic
royal sacrifices – aligning themselves with a prestigious foreign culture – for at least two
generations by the time of Pulakeshin II and Harsha. As such, allowing for Banabhatta’s
somewhat hyperbolic writing style and regional differences in practice, the events presented here
may serve as a broad-strokes reconstruction of Pulakeshin’s early life. Of course, there must have
been considerable variation in the details and order of elements between what actually happened
and what is presented here, but these speculations are useful and instructive in suggesting the
possible outlines of the events of the time.
For a detailed study, see Carol Radcliffe Bolon, Forms of the Goddess Lajja Gauri in Indian Art
(Pennsylvania State University Press, 1992).
Agrawala 1969, 88.
Ibid., 99.
Ronald B. Inden, Imagining India (Indiana University Press, 2000), 240.
Ibid.
The depiction of medieval Indian states as ‘imperial formations’ which used rituals as a socio-
political ‘world-ordering mechanism’ within more amorphous polities is based on Ronald Inden’s
work. See Inden 2000, Chapter 6.
Agrawala 1969, 88–89.
Ali 2006, 36.
Ibid., 51–52.
Ibid.
Ibid., 51.
Though the Satavahanas of the first century ce were a powerful polity, they do not seem to have
retained a predominant position in South Asian geopolitics for more than a few decades during the
reign of Gautamiputra Satakarni, probably c. 60–85 ce.
Dikshit 1980, 23–24. The clan will hereafter be referred to as the Chalkes, but it is worth noting that
there may be many other possible origins for their future dynastic title.
Narendra Nath Bhattacharyya, Ancient Indian Rituals and Their Social Contents (Manohar, 1975), 1.
Ibid., 1–2.
Ibid., 2.
Ariel Glucklich, The Strides of Vishnu: Hindu Culture in Historical Perspective (Oxford University
Press, 2008), 111–12.
Ibid., 4.
Ibid., 3.
Ibid., 112.
Bhattacharyya 1975, 5–6.
Kapila Vatsyayan, The Square and the Circle of the Indian Arts, Second Edition (Abhinav, 1997),
24–27.
Ibid.
Ibid., 27.
Ibid., 27–28. They also represented the primal Cosmic Man, thus establishing a correspondence
between the microcosm and macrocosm in order to ritually manipulate both.
Bhattacharyya 1975, 4.
Glucklich 2008, 112–13. Modern Hindu organizations such as the Arya Samaj argue that no animal
was actually sacrificed and that the entire ritual is entirely symbolic – but archaeological
excavations at ancient sites such as Kaushambi have discovered hundreds of bones of various
sacrificed animals. Even if they were not sacrificed for the Ashvamedha itself, the idea that ancient
rituals did not involve bloodshed is clearly not borne out by the evidence. As for whether historical
performances of the Ashvamedha were indeed performed exactly as described, the evidence does
not permit us to make a definitive statement. It should also be noted that other Indian philosophical
traditions, such as the empiricist Carvakas, roundly criticized the Ashvamedha and described its
creators as ‘buffoons’. See Bhattacharyya 1975, 2, footnote 9.
Glucklich 2008, 112–13.
Ibid., 113.
Stephanie W. Jamison, ‘Roles for Women in Vedic Srauta Ritual’, in Goddesses and Women in the
Indic Religious Tradition, ed. Arvind Sharma, Brill’s Indological Library, Vol. 24 (Brill, 2005), 1–
17, 9.
Bhattacharyya 1975, 18–22.
Glucklich 2008, 113.
Bhattacharyya 1975, 5.
Glucklich 2008, 113.
Vatsyayan 1997, 30.
Glucklich 2008, 114.
Bhattacharyya 1975, 1.
Dikshit 1980, 33.
Ibid., 34.
Ibid.
See John Michael McKnight, ‘The Gupta Temple Movement: A Study of the Political Aspects of the
Early Hindu Temple’, PhD dissertation (McMaster University, 1973).
Susan Locher Buchanan, ‘Cāl.ukya Temples: History and Iconography’, PhD dissertation (Ohio State
University, 1985).
Ibid., 37.
Buchanan 1985.
Ibid., 22.
Ibid., 13.
Ronald Inden, ‘Hierarchies of Kings in Early Medieval India’, Contributions to Indian Sociology 15,
nos. 1–2 (1981): 99–125, 102, https://doi.org/10.1177/006996678101500108.
Buchanan 1985, 13.
Ibid.
Ali 2006, 49.
Malini Adiga, The Making of Southern Karnataka: Society, Polity and Culture in Early Medieval
Period, AD 400–1030 (Orient BlackSwan, 2006), 107.
To Pulakeshin as to many later Indians such as the Rajputs and Marathas, ritually attaining high-caste
status and militarily establishing their sovereignty went hand in hand, and to conflate the two made
great sense in their political circumstances.
Adiga 2006, 36.
It’s almost certain that growing up in the area, Pulakeshin II would have seen dolmens; they were
everywhere, and in some cases still modified and used well into the eighth century ce. See
Srikumar M. Menon, ‘The Curious Case of the Galaganatha Dolmen: Possible Links between
Megalithic Monuments and Early Temples at Aihole’, Heritage: Journal of Multidisciplinary
Studies in Archaeology 2 (2014): 54–73.
Dilip K. Chakrabarti, The Geopolitical Orbits of Ancient India: The Geographical Frames of the
Ancient Indian Dynasties (Oxford University Press, 2011), Kindle Edition, Location 1188.
Chakrabarti 2011, Location 1633.
Ali 2006, 53.
Gary Michael Tartakov, ‘The Beginning of Dravidian Temple Architecture in Stone’, Artibus Asiae
42, no. 1 (1980): 39–99.
Dikshit 1980, 42.
Ibid., 54. It is doubtful that Ereya could have been that young. Assuming his father was alive when
the Chalukya capital was moved to Vatapi, he must have been at least fifty-six years old in 597
(especially given that Dikshit depicts him as one of Pulakeshin I’s commanders), which seems a
rather elderly age to have an oldest child aged five. This is especially so given how prolific and
frequent Indian royal marriages were. Professor Dikshit is probably off by a decade or so.
Cathleen Ann Cummings, ‘A Study of the Iconographic Program of the Lokesvara (Virupaksa)
Temple, Pattadakal’, PhD dissertation (Ohio State University, 2006), 203.
Dikshit 1980, 40.
Ibid., 41.
Hsüan-tsang, Buddhist Records of the Western World, Vol. II, trans. Samuel Beal (London, 1884),
256.
Inden 1981, 103.
Manusmriti 7.96, quoted and translated in Richard H. Davis, ‘Indian Art Objects as Loot’, The
 Journal of Asian Studies 52, no. 1 (1993): 22–48, 27.
This is implied by her triumphant declaration in 601 ce that she had spent the entire fortune of her
birth family, the Kalachuris, in a victory procession to honour her marital family’s gods including
Shiva Makuteshvara. See R.N. Nandi, Religious Institutions and Cults in the Deccan (c. A.D. 600–
A.D. 1100) (Motilal Banarsidas, 1973), 12.
Dikshit 1980, 56.
Ibid., 57.
For a similar example, see David Smith, ‘One Man and Many Women: Some Notes on the Harem in
Mainly Ancient and Medieval India from Sundry Perspectives’, Cracow Indological Studies 14
(2012):
1–16, 8.
Dikshit 1980, 60.
Cummings 2006, 203.
This trope is ever-present in Indian courtly literature and even in political manuals. See Ali 2006, 53–
54.
This, at least, is the version that Ereya presents. See EI VI, 9.
Ibid.
EI VI, 9. The terms used in the inscription are apparently taken from the Arthashastra or a similar
manual. ‘Power of good counsel’ (mantrashakti) refers to ministers and councillors, and ‘power of
energy’ (utsahashakti) to a rallied population and army.
K.V. Ramesh, ‘Pulakesin II – His Career and Personality’, in The Chalukyas of Badami: Seminar
Papers, ed. Nagaraja Rao M.S. (The Mythic Society, 1978), 53–65, 56.
Chakrabarti 2011, Location 1734.
Ibid., Location 1633.
Dikshit 1980, 76.
The Roman emperors Trajan and Hadrian both used bridges made of boats, as depicted in their
victory columns in Rome.
Dikshit 1980, 58.
Ramesh 1978, 55–56.
Dikshit 1980. 73. Professor Dikshit’s identification of the two as Rashtrakutas is quite tenuous.
EI VI, 9.
EI VI, 9. Though hamsa is translated as ‘swan’, this follows Vogel’s argument that the term actually
refers to the bar-headed goose, as swans are not native to the subcontinent. See Jean Phillippe
Vogel, The Goose in Indian Literature and Art (E.J. Brill, 1962).
Ibid., 10.
Dikshit 1980, 76.
Ghulam Yazdani, The Early History of the Deccan, Vol. I (Oxford University Press, 1960), 213.
Hereafter EHD I.
Ramesh 1978, 56.
Pius Malekandathil, Maritime India (Primus Books, 2010), 4.
Ranabir Chakravarti, ‘Merchants, Merchandise and Merchantmen in the Western Seaboard of India:
A Maritime Profile (500 bce–1500 ce)’, in The Trading World of the Indian Ocean, 1500–1800, ed.
Om Prakash (Pearson, 2012), 53–116.
EI VI, 10.
While Professor Dikshit thinks this was accomplished by the ‘Chalukyan navy’ (Dikshit 1980, 77),
 there is little evidence that any imperial formation based in the interior of the Deccan
systematically invested in a state navy until the early modern period.
Ibid.
Dikshit 1980, 96.
For a detailed study, see Kathleen D. Morrison, ‘Trade, Urbanism, and Agricultural Expansion:
Buddhist Monastic Institutions and the State in the Early Historic Western Deccan’, World
Archaeology 27, no. 2 (1995): 203–21.
Carla M. Sinopoli, ‘On the Edge of Empire: Form and Substance in the Satavahana Dynasty’, in
Empires: Perspectives from Archaeology and History, eds. Susan E. Alcock et al. (Cambridge
University Press, 2001), 155–78.
Mirashi 1955, lx.
Buchanan 1985, 122.
Ibid., 442. Cited in Cummings 2006, 211.
Buchanan 1985, 123.
Hsüan-tsang 1884 (Vol. II), 256.
Ronald M. Davidson, Indian Esoteric Buddhism: A Social History of the Tantric Movement
(Columbia University Press, 2002), 39.
For a reconstruction of the events leading up to Harsha’s accession, see Hans Bakker, The World of
the Skandapurān.a (Brill, 2014).
Bakker 2014, 70–71.
See Shankar Goyal, ‘Acquisition of the Maukhari Empire by Harsha’, Proceedings of the Indian
History Congress 62 (2001): 1090–91, http://www.jstor.org/stable/44155859. Shankar argues,
accurately, that Harsha had no actual claim to the throne of Kannauj either through his sister or in
his own right, and that even if he managed to win the sympathies of the city’s elite – perhaps
through his Buddhist connections – he may have had to conquer the rest of the Maukhari kingdom
by force.
Hsüan-tsang 1884, 213.
Bakker 2014, 108.
Dikshit 1980, 78.
Bahulkar and Sumant 2017.
Banabhatta quoted in Chakravarti 1941, 99. Of course, this was a description of a generic day’s
march for Harsha’s moving court, not a military campaign, so the number of leagues and the
composition of the force tasked with punishing Pulakeshin may have been different from what is
presented here.
Ibid.
Ibid., 100
Ibid.
Ibid., 101.
Daud Ali, ‘Aristocratic Body Techniques in Early Medieval India’, in Rethinking a Millennium:
Perspectives on Indian History from the Eighth to the Eighteenth Century: Essays for Harbans
Mukhia, ed. Rajat Datta (Aakar, 2008), 69–93.
A. Aruna, State Formation in the Eastern Deccan, 7th Century A.D.–13th Century A.D. (Bharatiya
Kala Prakashan, 2000), 101.
Pradeep P. Barua, The State at War in South Asia (University of Nebraska Press, 2005), 7.
Ibid., 12.
Chakravarti 1941, 52.
Konstantin Nossov and Peter Dennis, War Elephants (Osprey Publishing, 2008), 18.
Ibid., 19.
Prashant Srivastava, ‘The Harsha-Pulakesin II War’, in Select Battles in Indian History, Vol. I, eds.
K.K. Thapyal and S.N. Mishra (Agama Kala Prakashan, 2002), 126–30.
Hsüan-tsang 1884 (Vol. II), 257.
Agrawala 1969, 24.
Ibid., 50.
Pulakeshin himself was unable to do anything further with this victory. Perhaps he realized he would
be unable to confront Harsha in a pitched battle on north Indian terrain. In any case, his future
campaigns were in the south. So the battle was apparently not as decisive as it is often thought to
be, but the outcome was probably very unexpected.
Bahulkar and Samant 2017.
 2
Pulakeshin’s Blood
Spring, 619 CE

Fresh from the victory over Harsha, Pulakeshin II and the Chalukyas set out
to dominate as much of southern India as they could. This wasn’t an easy
task: rivals and traitors and nemeses were everywhere. And far away in the
deepest south, a storm of songs and devotion was being unleashed. The
name of an enigmatic, erotic, violent god was on everyone’s lips: Shiva, the
Auspicious One, the Destroyer.

In the early centuries CE, the coastal belt of Andhra, to the east of the
Chalukya dominions of the seventh century, had been host to hundreds of
Buddhist monasteries, stupas and pilgrimage sites. Dozens of now-forgotten
Buddhist schools – Aparamahavinaseliya, Bahusrutiya, Ahirasamghika,
Mahisasaka and more – had vied for the patronage of local communities
and wealthy merchants. Buddhist monasteries ran massive religious and
commercial establishments, and were perhaps at least as influential in the
daily lives of the Andhra people as the then-fledgling apparatus of
monarchical rulership. From Andhra, Buddhist artisans, pilgrims, monks
and merchants had exchanged ideas with the Tamil country (also known as
Tamilakam), the northern Deccan, the Gangetic plains, and even Gandhara
and thence with Central Asia and China. Buddhism’s position as the most
prestigious and powerful of South Asian religions had then seemed
unassailable.
But by the second half of the first millennium CE, as the Chalukyas were
rising in the Deccan, the influence of this ancient religion was waning.
Buddhism was becoming increasingly dependent on patronage from
military aristocrats to survive. But few kings or queens of Andhra’s ruling
dynasties now declared a Buddhist affiliation, preferring newly prominent
gods and goddesses, such as the Saptamatrika or Seven Mothers, the avatars
of Vishnu, and especially the family of Shiva. This new preference, as
we’ve glimpsed through the activities of the Chalukyas in the Malaprabha
river valley, was based on careful political calculation and deliberate
institution building as much as on personal devotion. We shall return to this
theme soon. For now, it’s sufficient to say that by the early 600s CE, the
petty dynasties of Andhra were constantly at war with each other and with
the Tamils to their south.
Meanwhile, on the shores of the Narmada, Pulakeshin II was realizing
that his victory over Harsha was not as decisive as he had expected. Harsha,
whatever his deficiencies as a military strategist, was by far the shrewder
and more experienced diplomat. To Pulakeshin’s northeast, Harsha allied
with their common neighbours1 to stymie any expansion in that direction
and retain them as buffer states between the two emperors’ competing
spheres of influence. Meanwhile, to Pulakeshin’s northwest, the rulers of
northern Gujarat began to lean towards Harsha under a diplomatic barrage.
The Chalukya was far too cunning to try to subjugate them outright and
thus be drawn into a conflict on a terrain of Harsha’s choosing. Nor was he
foolhardy enough to repeat Harsha’s strategy and attempt to cross the
Narmada.
And so Pulakeshin II’s eyes alighted on the east: the coastal plains of
Andhra. Its ports were connected to the trade circuits of the eastern Indian
Ocean, guaranteeing luxury goods and tariffs. Its lands were fertile, good
prizes for his loyal followers, and its rulers were weakened by wars with
their southern neighbours, the Tamil dynasty known as the Pallavas.2 When
the lords of Vengi – the fertile heartland of Andhra, that bounteous delta
between the rivers Krishna and Godavari – received word of the Chalukya
emperor’s march downriver from the Deccan plateau to the Andhra plains,
they must have summoned their vassals and ordered the mustering of
troops.
 By winter 618–spring 619 CE3 (mere months after his victory against
Harsha) Pulakeshin’s elephants were hammering at the gates of the fortified
city of Pithapuram, capital of one of Vengi’s major dynasties. Soon after, he
faced another local king on the shores of the Kolleru lake, famous today for
its waterfowl. A brutal slaughter was inflicted on this army, supposedly
leaving the shores of the lake littered with corpses and turning its waters red
for days.4 An uneasy peace was settled, with the Andhra kings accepting
Chalukya overlordship. However, many other challengers – including the
tribes of the Eastern Ghats – still threatened their authority there.
Nevertheless, the Chalukyas now had a foothold on both shores of the
peninsula, east and west. The Konkan had been secured by the conquest of
southern Gujarat and the appointment of viceroys directly related to
Pulakeshin in those regions. A similar strategy would need to be followed
in order to secure the conquest of Vengi. During his return to Vatapi,
Pulakeshin set this programme in motion. At Alampur, the sacred
conjunction of the Krishna and Tungabhadra rivers about 200 km from
modern-day Hyderabad,5 he paused to divide up parts of conquered Vengi
among his vassals.6 They were granted the right to rule over the land and
make war against its natives to expand their holdings. He then continued
onwards to his clifftop fort to celebrate with his sons and queens. Showered
with flowers and amidst the roaring acclaim of his subjects, Pulakeshin
visited the cave temples of Vatapi to thank his clan’s gods.
Chalukya domains were now more expansive than they had ever been:
most of the Deccan, from the upper reaches of the Kaveri river to the
Narmada, owed them direct or indirect allegiance. Southern Gujarat and the
Konkan coast were controlled by Chalukya viceroys, and Chalukya-
affiliated warlords ruled over swathes of Andhra on the east coast. Despite
Pulakeshin II’s by-now legendary military reputation, the ruling of such a
wide and diverse domain required bold ways of stating, justifying and
enforcing political power.
One way he did this was by embarking on the construction of temples. In
doing so, Pulakeshin II demonstrates to us why elites across the Indian
subcontinent were abandoning Buddhism and moving towards Puranic
Hinduism (associated with encyclopaedic texts called Puranas, quite
distinct from Vedic Hinduism).7 After all, there are few more effective ways
to cloak avarice and political ambitions than by using the garb of devotion.

Today, it is difficult to imagine Hinduism without temples, or even to

imagine temples as being more than places for worship. That was not the
case during the time of the early Chalukyas. Then, the idea of building
permanent shrines to ‘Hindu’ gods was somewhat of an innovation.
Buddhists, on the other hand, were used to worshipping idols and
congregating around permanent religious institutions such as monasteries or
stupas, and folk religion in the subcontinent always had a strong idol-based
aspect to its worship of yakshas and other fertility deities. But orthodox
Hinduism, even up to the early centuries CE, had still relied primarily on
Vedic rituals – expensive and difficult undertakings for kings, and rather
ephemeral in how they displayed royal power and political messages. These
rituals were performed in temporary altars, built specifically to perform a
sacrifice and destroyed after.
But by the fourth century, aided by the patronage of the Gupta emperors
of northern India, Hindu cults adopted permanent iconic representations of
gods and introduced radical theological innovations in a bid to increase
popularity and capture patronage. In the Vedic worldview, the gods, who
always remained invisible presences through the rituals, needed periodic
sacrifices from mortals to help them defeat anti-gods and bring rains. In the
new, temple-based, Puranic Hinduism, gods such as Shiva and Vishnu were
embodied as idols and were declared Supreme. They no longer needed
sacrifices to help them order the cosmos, but went about their business
according to a plan that mortals could not hope to understand. Mortals
could also bring them to Earth as permanent residents in temples, not just as
temporary visitors to sacrifices. There, they could propitiate these mighty
deities with new rituals that were presented as reworked versions of Vedic
sacrifices: periodic, daily offering of animals, food, flowers and so on. This
innovation proved popular with rich and poor alike. Royals across northern
India soon started to build shrines to Hindu gods, while continuing their
patronage of existing Buddhist and Jain institutions.
However, temples did not spread only for devotional reasons. Major
temples and the deities they enshrined were associated with legends
contained in the Puranas, and were believed to have potent boons to offer
devotees and donors. They attracted pilgrims and, more importantly,
commercial activity, creating hubs where India’s flourishing religious cults
could preach, squabble, sing, celebrate and make money. The popularity of
temple-based Hinduism also led to the gradual creation of new religious
communities, more amenable to supporting the activities of warlike
aristocrats who claimed a special relationship with the gods8 – all of which
were in stark contrast to what Buddhists in south India were willing or able
to offer, as will be seen later. Building a temple could also help royals curry
favour with existing groups of worshippers, monastic communities,
pilgrims and so on. The crowds of devotees that visited temples would
frequently hear of the land grants and military exploits of kings, and
associate them with the inscrutable world-ordering activities of deities, in
contrast to the moralistic tales of Buddhas and Bodhisattvas that they might
hear in a stupa. As one scholar puts it, the rise of ever more elaborate
temples and the growth of royal power inevitably went hand in hand.9
This was why it was so important for medieval kings to build temples,
though we may often misinterpret these activities as stemming purely from
devotion. Early Chalukya temples, in particular, show us how medieval
kings used these buildings as a crucial aspect of their power. Chalukya
temples are replete with subtle and not-so-subtle political messages.
Pulakeshin I’s earliest attested activities were Vedic sacrifices, meant to
establish his status amongst other elites conversant with their meaning –
kings, Brahmins, generals, poets, astrologers – rather than the average
Deccani. By the end of his reign, he had begun to seek more permanent
means to express his power to the mass of his subjects. His imperial title,
Sri-Prithivi-Vallabha, had already established his relationship to Vishnu.
But he also used the soft sandstone of the cliffs of Vatapi to state his
personal relationship to the other great god of temple-based Hinduism –
Shiva – whose cults were already popular in the northern Deccan.
Pulakeshin I’s sculptors were ordered to make sculptures totally unique to
the Chalukyas, expressing royal support for these new religious practices.
One such image, a spectacular eighteen-armed image of a dancing Shiva10
still welcomes visitors to the cliffs of Vatapi (modern-day Badami). The
god’s arms swirl around him like a blooming lotus, his left foot poised an
instant away from striking the ground.
Pulakeshin I’s two warlike sons (Mangalesha and Pulakeshin II’s father)
also engaged in a spurt of cave temple building in Vatapi and other
Chalukya towns in the Malaprabha valley, such as Aryapura (modern-day
Aihole). Ancient sacred springs and wells and dolmen burial sites, where
the indigenous peoples of the land had congregated for centuries to
celebrate nature’s rhythms, were now incorporated into the sites of flashy
new temples.11 Of course, the Chalukyas by no means confined their
patronage to Hindu gods; the religious composition of early medieval India
was very much in flux, and they built Buddhist and Jain temples to appeal
to those audiences as well.12
Today, the constructions of the early Chalukya kings are among the oldest
known temples in southern India, and they tell us a fair bit about the
political clashes and competitive religious environment of the sixth century.
For example, the Kadambas of Banavasi, their deadly rivals, associated
themselves with the Saptamatrikas, goddesses of victory and liberation with
roots in ancient cults. Pulakeshin I’s sons went out of their way to use
images of these Seven Mothers in their cave temples,13 competing with the
Kadambas to claim the favour of these popular deities.
Mangalesha commissioned an image of the original Sri-Prithivi-Vallabha,
Varaha, the man–boar form of Vishnu, who had earned the love of the
goddess Earth by rescuing her from a demon at the time of the Great
Deluge. The sculpture, evidently meant to establish a visual parallel with
the king, can be seen in Cave 3 in modern-day Badami. To this day it
retains traces of its original lustrous blue paint. Mangalesha’s campaigns
against the Kadambas and Kalachuris, it seems to say, are analogous to
Vishnu’s rescue of Earth from darkness. This Chalukya king’s association
with Varaha would go on to become one of the subcontinent’s most iconic
visual motifs: his nephew Pulakeshin II would use the boar as his battle
standard, and it would continue to be used intermittently for nearly one
thousand years after, even during the time of the Deccan empires of Bijapur
and Vijayanagara.14

Pulakeshin II inherited some of his uncle Mangalesha’s skill at visual

propaganda, introducing innovations of his own to Chalukya political
messaging. He, however, wanted to make a new sort of temple – not the
dark, concealed cave temples that the Deccan was familiar with, but elegant
new structures that embraced open air, space, and light. His sculptors and
sthapatis (‘establishers’, or master architects) oversaw the removal of great
blocks of sandstone from the cliffs of Vatapi. These were then assembled
into some of the oldest surviving free-standing temples in southern India.
Their clean lines were adorned with subtle sculptural motifs – artistic
representations of gods, myths and miniature shrines. Built to support their
own weight through careful balancing and positioning of joints, Pulakeshin
II’s temples have endured nearly 1400 years of erosion.
 Their design borrows from both north and south to make something new,
something uniquely Deccan. The shape of Pulakeshin II’s temples, using
the distinctive south Indian tiered superstructure ascending in ever-smaller
layers from a wide base, seems to reflect a south-centric worldview. And
yet, there are many northern influences, which would only have been
included by sculptor guilds at Pulakeshin II’s express wish, in order to
impress audiences in Vatapi. For example, in sculptural panels on the walls,
Shiva can be seen standing straight and calm in the samabhanga posture
with a snake in his right hand and a trident in his left, all of which are
‘common North Indian attributes’.15 These influences came by way of the
northern Deccan, a region that was now firmly in Chalukya hands, though it
had been a foreign country just a generation ago. Religions and goods from
there were now becoming more popular in the Chalukya home territories, as
these once-poor lands grew into prosperous towns and were integrated into
the subcontinent’s webs of trade and religious exchange. Pulakeshin’s
splendid new buildings must have been a great hit with the increasingly
religious and cosmopolitan people of the Malaprabha valley, a shrewd
investment of the loot he had gained from his military campaigns.
Pulakeshin II’s temples thus tell us a great deal about the complex ways in
which medieval Indian elites went about the business of solidifying and
perpetuating their power. Given what temple-based Hinduism could offer to
the wealthiest and most powerful people, the decline of monastic Buddhism
at this time makes much more sense.
Pulakeshin II also brought innovations to other aspects of Chalukya
power. Temples helped structure social activity in urban centres, with their
consistent daily and seasonal rhythms where the people of the Malaprabha
river valley could congregate. But their audience was limited to a radius of
a few dozen miles, at most. For faraway subjects and vassals, more concise
religio-political messages were needed.
Pulakeshin II now proved himself the equal of his grandfather in
propaganda, inventing a legendary backstory for the Chalukyas, erasing
their humble chalke-wielding past. This was interwoven with land grant
formulae once used by the Kadambas, ensuring that the aura of glory that
surrounded that old dynasty would now accrue to his clan.
According to Pulakeshin, the Chalukya were a clan ‘nourished by the
breasts of the Seven Mothers … who have acquired an uninterrupted
continuity of prosperity through the protection of Karttikeya [the war god],
who have had all kings made subject to them at the sight of the boar-crest
which they have acquired through the favour of the divine Narayana
[Vishnu]’.16 He declared that they were a race of heroes sprung from a pot, a
chuluka, filled with water from the Ganga by an ancestor who defended the
gods from demons.17 In another version, the Chalukya progenitor was
Brahma the Creator himself.18 In another, there is no pot, but there was a
hero whose name is Chalukya.19 Finally, Pulakeshin also adopted for
himself a bevy of titles designed to inspire awe among his vassals and
rivals, including Satyashraya, Refuge of Truth – this would be the title by
which he was remembered by his successors for centuries after.
The point of all this intense activity in architecture, iconography and
political propaganda was manifold. It established that the Chalukyas had
rescued the Deccan from all the darkness and anarchy that came before, just
as Varaha had rescued the Earth. It established that they were no ordinary
mortals, but the favourites of the most popular gods – even the gods of their
erstwhile rivals. It established that they ruled over north as well as south,
that their power extended into areas that had never bowed to the might of
the Deccan before. And, most importantly, it laid the popular and
institutional foundations for many more generations of Chalukyas to build
temples, make land grants and reorder the Deccan as their ancestors had.

As Pulakeshin went about the business of creating an empire, world-

changing events were afoot elsewhere in the 620s. A devastating war had
broken out between two of the world’s superpowers: the Eastern Roman
Empire and the Sassanian Empire of Persia. The Sassanian Empire was left
a shadow of its former self, tottering on the brink of anarchy. Large swathes
of the Levant and modern-day Turkey and Iraq were in ruins, disrupting
trade to the little town of Mecca in the Arabian Peninsula. In 622, a middle-
aged former merchant, Muhammad, fled from there to Yathrib (modern-day
Medina) with his followers. He brought with him a new religion: Islam,
‘Submission to God’.
This, of course, was of no importance to Pulakeshin II, who still had
challenges to overcome in Vengi, while the ambitious Tamil kings were
cooking up trouble to the south. Dealing with the first task was the
responsibility of Pulakeshin’s most trusted lieutenant, his younger brother
Vishnuvardhana. This Chalukya was a far wittier and more cynical
individual than Pulakeshin. He invariably called himself Kubja-
Vishnuvardhana in his own inscriptions, the word kubja meaning
‘hunchback’ – this in an era when court residents delighted in boasting
about their physical features and attractiveness.20 Indeed, he sometimes
called himself Vishamasiddhi, Overcomer of Obstacles,21 a reference both
to his military abilities and physical handicap.
The brothers were extremely close. Pulakeshin had, after all, defeated
their uncle Mangalesha and fought for their right to primacy in the Deccan
almost single-handedly. Vishnuvardhana, in return, worked alongside him
as a governor and general at the head of the Chalukyas of Vatapi. He very
likely earlier participated in the brutal sacking of Banavasi, the Kadamba
capital, and Puri, the great port on the Konkan coast. He is known to have
served as Yuvaraja, heir apparent and designated regent, on multiple
occasions, and may have held that title when Pulakeshin was away
campaigning against Harsha and Vengi in the years 618–19. Given that his
elder brother had already spent years developing and consolidating his own
personal power, Vishnuvardhana is unlikely to have seen himself as a
candidate for the throne of Vatapi. Instead, he seems to have calculated, like
the other chiefs of the Deccan, that his talents would be more profitably
employed in attempting to seize glory under Pulakeshin’s shadow. The
easiest way to do that, of course, was through conquest.
 By the mid-620s, when Pulakeshin was busy with his construction
projects and strengthening his grip over the Deccan at large,
Vishnuvardhana was in the process of completely subduing Vengi,22 which
was proving remarkably recalcitrant to Chalukya authority. The dynasty
Pulakeshin had so bloodily defeated at the Kolleru lake was still around;
they were nominally Chalukya vassals, but caused no end of trouble to the
brothers as the pair tried to enforce their authority on the other fiercely
independent powers of Andhra. An even more serious challenge came from
the Adivasi forest peoples of the country.
Vishnuvardhana’s inscriptions refer to them as the Durjayas (Difficult to
Conquer),23 occupying a land called ‘Giripaschimasima’24 – ‘the place west
of the hills’ – though we have no idea what the people of this powerful
tribal coalition called themselves. It would seem that the Chalukya armies
could conquer cities and agrarian land easily enough, but were far less
effective in hilly and forested terrain, especially without friendly tribes to
help them (as there may have been in their campaign against Harsha). These
Durjayas must have had their own charismatic chiefs, their own complex
religious rites, their own sense of identity and time and place nestled away
in mist-wreathed hills and forests bubbling with the sounds of streams,
animals and distant drums. A history of India from their perspective and
that of other tribes like them, confronted by aggressive agrarian kingdoms,
might look very different.
Of course, the relationship between tribal peoples and sedentary states
was not always either amicable or antagonistic, but lay on a spectrum.
Interactions were guided by many factors, including the relative balance of
power between both sides. In some areas, Durjaya chiefs accepted land
grants from the Chalukyas, as did members of other tribes such as the
Matsyas, and a group that still exists today: the Boyas. Indeed, there is
evidence that some Boya priests were granted Brahmin status and received
land grants from Chalukya hands,25 hinting that royal Hinduism in the
seventh century was highly effective at appropriating or incorporating local
religious ideas in practice, and vice versa. Boya chieftains could use the
prestige and legitimacy that Hinduism conferred to solidify their authority,
and may thus have deliberately integrated with this new complex of ideas
that the Chalukyas patronized – just as the Chalukyas themselves had
performed sacrifices and built temples to strengthen their grip on power in
the Malaprabha river valley a few generations ago. As we can see, the
interplay of religious and cultural legitimacy, political calculations and
ownership of land in medieval India helped Puranic Hinduism spread far
and wide as the premier religion of kings and lords.
Vishnuvardhana also began to build his own, independent base in Vengi,
consisting of other ambitious local warriors, imported Brahmins and
Deccani warlords, and elements of the old ruling classes directly loyal to
him. All these diverse agents were rewarded with revenue rights to villages
and other financial and symbolic rewards in return for their military,
cultural and political support to his new regime.26 He, like his elder brother
Pulakeshin II, also tapped into the other strategies available to Indian lords
of means and discernment, hiring publicists and panegyrists to establish his
own personal connection to the gods.
In addition to Vishnuvardhana’s campaigns, denizens of Vengi had
previously been subjected to brutal Chalukya aggression at the hands of
Pulakeshin II during the bloody battle of the Kolleru lake. But obviously
not a hint of this comes across in the younger Chalukya’s inscriptions.
Vishnuvardhana instead portrayed himself as virtuously rescuing the world
from the darkness of unjust kings and immoral people, which according to
mythology was an ever-present feature of the Kali age that had begun after
the Mahabharata war. He, like his brother Pulakeshin II and uncle
Mangalesha, intended to be remembered as a royal saviour, which meant
that Vishnu legends were trotted out and liberally applied to him as well.
For example, Kubja-Vishnuvardhana could also be interpreted as ‘Increaser
of the Glory of Dwarf-Vishnu’ – the avatar of Vishnu called Vamana, the
dwarf who rescued the Earth from the anti-gods.
Vamana was not the only deity pressed into the service of this new
Chalukya’s royal brand. Vishnuvardhana seems to have had a fairly active
sex life, and took for himself the titles of Kamadeva – the god of love –
while adopting as his personal emblem Kama’s symbol, the makara, a
fantastic creature that was a synthesis of an elephant, a crocodile, a snake
and a peacock; he thereby equated himself with the embodiment of male
attractiveness and sexuality.27 For someone with a physical handicap to
make this claim is indicative, one would like to believe, of a certain dry
humour.
So far we have seen the medieval India propaganda machine in action at
the hands of many of the early Chalukya lords: Pulakeshin I, his son
Mangalesha and Mangalesha’s nephews Pulakeshin II and Vishnuvardhana.
This may have given the impression that Chalukya propaganda was always
effective and lapped up by the people they dominated – but that’s primarily
because most sources surviving from the period are Chalukya land grants
and architecture, a great example of how history ends up being written by
the victors.
 But just as not all modern voters are foolhardy enough to take political
ads at face value, the denizens of Vengi had a similarly complicated
relationship with Chalukya proclamations. In Vengi – which, as mentioned,
was once a major Buddhist centre – these voices came from the sangha
itself, providing us with a rare perspective from the ‘losing side’, those
medieval Indians who found themselves on the wrong side of new
equations of royal power, religion and economics. This is a voice of
privilege to be sure, but nevertheless hints at a far more turbulent process of
conquest and acculturation than kings might like us to believe.
According to Chandrakirti, an elderly monk from Vengi, contemporary
kings (probably the Chalukyas) were little more than ‘trained monkeys and
dogs that must look to their [Brahmin] masters for instruction before they
act’.28 What about the idea that these kings were associates of the divine just
because they built a temple or two? Nonsense – any king who did not use
his position to benefit the people and reduce the suffering of living things,
but was instead driven by lust for wealth and women, was more likely to be
reborn as a donkey or a worm29 than rise to heaven.30 What’s more, argued
the monk, these new rulers couldn’t even be trusted to ensure law and order.
After all, the exemplar of all thieves was the looting, conquering king who
dug holes in the walls of cities and forts just as thieves dug holes in the
walls of houses.31
But, vocal as Vengi’s Buddhists were, neither of the Chalukya brothers
paid them any attention. They didn’t need to: a Chinese monk who visited
Andhra in the seventh century remarked on the fact that barely twenty
Buddhist monasteries remained, whereas there were hundreds of flourishing
Shaivite temples,32 crowded with devotees.
The presence of all these temples offers us another chance to question the
centrality of kings in our understanding of medieval India. We’ve seen how
temple building was a way for royals to ingratiate themselves with new
religious communities or create new ones. But kings were not omnipotent:
none of their activities could have seriously changed the religious landscape
of India if not for the fact that other, far more powerful forces were at play,
remaking the religious world of thousands of people. The Chalukyas and
their ilk were merely benefiting from and amplifying a great tsunami of
religious change that was beginning to engulf southern India. The temples
of kings are the most visible remains of this period, but they are merely
reflections of the small and conceited thing that is royal power, often
responding to or taking advantage of something far vaster than itself.
This was the explosive flowering of bhakti – an intense, almost mad
devotion to a god. In particular, devotion to the god of the matted hair and
the tiger skin: Shiva, Destroyer, Auspicious One.

Although it might seem that Indians have always been mostly Hindus, and
that Hindu kings are the most important characters in the story of the
subcontinent, that is not so evident in the early medieval south. Indeed, the
rise in the popularity of Hindu gods, temples and kings in south India, and
specifically in Tamilakam, the Tamil country, was a sudden and somewhat
improbable phenomenon. History is shaped not just by individuals but by
fundamental transformations in how societies of hundreds of thousands of
people see and structure themselves. This dynamic can be seen through
first-hand accounts from Tamilakam in the seventh century.
While the Deccan and Andhra both had significant Buddhist and Jain
presences, as we’ve seen above, Tamilakam, with its prosperous kingdoms
and ports, was the real bastion of their power. Many of that land’s great
epics had been written by influential Buddhists and Jains – both ordained
monks and members of the lay community. Buddhist and Jain monasteries
competed with each other and with sects dedicated to ‘Hindu’ gods for
social and royal largesse. By the sixth and seventh centuries, Buddhists and
Jains in the region participated in surprisingly similar forms of worship,
digging cave temples into cliff sides and using idols as a focus of
devotion;33 they freely borrowed ideas from each other in an attempt to
remain as popular and relevant as possible.34
This balance was upset with the rise of poet-saints proposing a radical
new theology focused on absolute devotion to Shiva – one that,
coincidentally, complemented the new temple-based Hinduism that royals
were promoting. In doing so, they became among the most pivotal figures
in Indian and global religious history. Their theology, bhakti, presented
people with an alternative and viable path to spiritual salvation and social
status that did not require the expense of Vedic sacrifices, or the extremities
of worldly renunciation that the Buddhists and Jains demanded. It was a
path open to all, ascetic or householder. All that was required was absolute
love and surrender to Shiva, the Auspicious One. The lack of emphasis on
renunciation or expensive rituals, coupled with an easy-to-understand
emotional philosophy, made bhakti wildly popular in comparison to Vedic
Hinduism.
The bhakti saints wrote beautiful, memorable music and led crowds of
devotees in song, travelling from village to village proclaiming the power
and love of Shiva. From holy site to holy site they went, followed by
cheering and singing crowds, converting householders and rival monks
alike.
‘How shines before my eyes his brilliant trident!’ sang the crowds. ‘How
gleams the crescent moon atop his lengthy locks! How heavy the scent of
konrai garland! How bright the glow of conch-shell and thodu! Like
thunderbolt the sound when elephant skin he tears – Look how he wraps it
round like a mantle!’35
We can scarcely imagine the profound spiritual experience it must have
been, in a world torn by war and famine and misery, to join these dancing,
ecstatic crowds, surrounded by bodies throbbing to the beats of the damaru
drum, hearing praises of Shiva’s power, Shiva’s love, Shiva’s mercy. It
seems, understandably, to have inspired almost fanatical devotion. You
could ask Shiva for anything – one saint even asked him to rescue him from
the predicament of having cheated on his wife – and you could say anything
to him.36 You didn’t have to know the rituals, the chants. You could refuse
to go on pilgrimages, ignore the flashy new temples that kings were
building, search within. Or you could do those too, if you wanted. Shiva,
according to the saints, accepted all, and was merciful to all.
‘We are slaves to no man!’ exulted one saint. ‘… we belong forever to
Shankara [Shiva] … The Lord of the warlike bull has taken us. Why need
we listen to the words of men who parade themselves in silk and gold?’37
Buddhists and Jains refused to give up silently when faced with a deadly
threat to their lives and livelihoods. One bhakti saint, a former Jain, led a
vitriolic campaign against his former colleagues, calling them ‘arrogant and
fat … stinking and debased … lacking in both virtue and clothing’, and
singing of how Shiva drove them away, ‘bearing as weapons fire and a
white axe’.38 Jains responded in kind, denouncing Hindu deities as dissolute
characters unworthy of worship,39 who would only lead their followers
further away from enlightenment and into the entanglements of the mortal
world – a reference to the bhaktas’ claim that one could be saved by Shiva’s
grace even as a householder. Buddhists and Jains held a dominant position
in many towns and courts, challenging the bhaktas to perform public
miracles to prove that Shiva actually did intervene in the world to help his
devotees. Of course, according to later Shaiva legend, Shiva did exactly
that, granting the saints the power to light lamps filled with water instead of
oil, cure kings of disease, and survive lime kilns and mad war elephants
with aplomb.40
 Though Shaiva saints might have had a good deal of contempt for kings –
‘men who parade[d] themselves in silk and gold’ – their poetic activities
were profoundly useful to royals. New temples soon emerged, capitalizing
on the devotional fervour kindled by the saints. Surely, royal propaganda
implied, if people could have absolute faith in the gods, then they should
have absolute faith in kings, the beloveds of the gods. The Chalukyas
watched attentively as this wave began to spread into the Deccan from the
south, alongside the Shaiva cults of the northern Deccan already prominent
in their territories, and the flourishing new religious communities now
gathering around their own Shiva temples in the Malaprabha valley. They
had astutely chosen the winner from among the whirlpool of India’s
religions. Buddhism would never again rise to the position it had once held
in southern India. Despite the searing critiques Buddhism made of the
Chalukyas, it seems to have been ignored by royals and commoners alike,
and we hear less and less of resistance against the power of temples and
kings. Gradually, the people of Vengi and the Deccan, their lands and minds
conquered, seemed to have reconciled themselves to the rule of their
warlike new overlords and the gods who supposedly supported them.

The eventual fate of Pulakeshin II was decided by a Tamil king who had
been directly touched by the message of Shiva’s saints. This man is one of
the most uniquely complex of all Indian kings. A keen observer of the
society over which he ruled, his sardonic wit and sharp sense of self make
this aesthete, playwright and poet real like few other medieval kings –
especially in comparison to Pulakeshin, whose inscriptions, stage-managed
by court poets, present him as more of an archetype of the perfect Indian
conqueror than an actual human being with flaws, misgivings or even a
point of view. In contrast, this poet-king made it a point to compose his own
inscriptions, and even wrote multiple plays, of which two have survived.41
His name was Mahendra-Varman, and he belonged to an old dynasty
called the Pallavas, rulers of the northernmost part of Tamilakam, bordering
both Vengi and the territories of the Gangas in the southern Deccan.42 By
the 600s, the Pallavas had catapulted themselves to a position of pre-
eminence in Tamilakam by campaigning against the Buddhist/Jain-affiliated
dynasties from their capital city of Kanchipuram, which still thrives today.
Despite this, their approach towards the cut-throat religious competition
within their territory had generally been fairly balanced. But Mahendra-
Varman, supposedly because of the teachings of one of the seniormost of
the bhakti saints, explicitly converted from Jainism to Shaivism. While
personal convictions were certainly part of this decision, its political
usefulness must have also played a role.
This complexity of motivation is on clear display in one of Mahendra’s
surviving plays, Matta-Vilasa Prahasana (Farce of the Drunkard’s Games).
The Matta-Vilasa in the title, meaning ‘Drunken Sporter’, is a reference to
Mahendra himself, who made no secret of his love for alcohol, and even
used the epithet in cave temples he commissioned.43
The farce is about an alcoholic kapalin (skull bearer), an ascetic member
of a Shaiva cult whose activities involve wandering around Kanchi with a
skull that doubles as a begging bowl for offerings of meat and liquor. One
day, the hungover kapalin and his equally hungover girlfriend/apprentice
discover they have lost their skull bowl. Their suspicion alights on a smug-
looking Buddhist monk who is hurrying about, quite satisfied with the rich
food and lodgings afforded to his order and secretly wishing that his
monastic vows did not prevent him from indulging in wine and women (an
interesting contrast to the kapalin who lives an otherwise austere life with,
however, plenty of alcohol and sex). The two drunkards attempt
unsuccessfully to grab him and rough him up, and end up flailing around
and making fools of themselves.
Their scuffle is then enlivened by a Pashupata, member of yet another
Shaivite school, who has a crush on the kapalin’s girlfriend and attempts,
poorly, to adjudicate the dispute. This, of course, fails. Someone suggests
they go to the judges of Kanchi, but the rest reject it because of how corrupt
they are – a candid admission of the royal author’s limitations as Kanchi’s
ruler, and a notable departure from other medieval kings who are generally
quite self-satisfied in their boasts of providing perfect law and order. The
absurdity is unexpectedly resolved by a madman – possibly a metaphor for
Shiva himself in his form as Bhola, the simple-minded44 – who has found
the skull bowl in the possession of a stray dog.
There is much tomfoolery and physical humour in the farce, but the
dialogue is razor-sharp and showcases Mahendra’s keen observation of the
dogmas and absurdities of each of these schools. His characters might not
be particularly deep, but they reveal a great deal about how a royal Shaivite
viewed the religious innovation, plurality and absurdity of the times. As a
Jain who had converted to Shaivism, he pokes only gentle fun at his
original faith and reserves most of his wit for the ‘heretic’ Shaivite schools,
and of course the Buddhists, all of whom he evidently disapproves of.45 The
farce appears to have been enough of a hit with the literati that it continued
to be performed on and off for centuries, copied out again and again to
come down to us today and give us a rare glimpse into the mind of a
medieval Indian king.
In the 620s, Chalukya campaigns in Vengi were reaching a crucial stage,
and Palakeshin II and Vishnuvardhana could countenance no threats from
the south. The Pallavas had been battling the rulers of Andhra for control of
this region for decades,46 and the only way for the Chalukyas to stop their
ambitions of northward expansion was through force. Vengi, fractured
between many petty dynasties and fading Buddhist monasteries, had been
easy for the Chalukyas to swallow whole. But nothing of the sort could be
achieved in the religiously and politically vibrant Tamil country. Pulakeshin
II seems to have instead aimed to secure Mahendra Pallava’s submission47
and inflict a decisive enough defeat on the Tamils to ensure his grip over
Vengi.
At the battle of Pollilur48 – close to where, twelve centuries later, Tipu
Sultan and Hyder Ali would inaugurate modern military rocketry by
smashing a British army with gunpowder – the Chalukyas under their boar
banner crushed the Pallava forces and drove them behind the walls of
Kanchi.49 The Pallava guards must have watched with apprehension as the
dust clouds of the approaching Chalukya army, hot on their heels, blotted
out the sun. Hundreds of colourful flags and umbrellas dotted the vast force
as the sun glinted off thousands of spears. On an especially large elephant
near the centre was Pulakeshin II, wearing a garment of silk and a golden
coronet, glistening with sweat in the sweltering heat despite the great
imperial parasol an attendant held over him.50 The fearsome boar banner
was held high on a standard nearby, waving in the breeze. Around the
emperor were his highest generals and nobles, also mounted on elephants,
surrounded by elite hereditary troops holding the fluttering banners of
aristocratic households.
Hearing the roars and drums of the Chalukya army, the trumpeting of
elephants, and quite possibly the taunts of Pulakeshin, Mahendra apparently
realized his kingdom was not, at this time, ready for the challenge. And so
he did what every classical Indian text on statecraft would recommend:
follow a policy of appeasement.51 In return for tribute and perhaps a
promise that Mahendra would not contest Chalukya control over Vengi,
Pulakeshin was convinced to leave Kanchi and its denizens untouched. And
so he returned to Vatapi, confident in his status as the dominant ruler of
southern India.

The Pallava Mahendra-Varman had every intention of trashing his

agreement as soon as he could. He, personally, had neither the inclination
nor the ability to invade either the Deccan or Vengi. But he was no pacifist,
and understood that the geopolitics of medieval India were a wild jungle,
governed by the principle of matsya-nyaya, the Law of the Fishes, where
the big ate the small, where one either ate or was eaten. The peace with
Pulakeshin was only a lull. Though his middle age seems to have been
spent exploring the arts, Mahendra had overseen many major Pallava
victories over his Tamil neighbours in his early career as king.52 He now
spent his remaining years bolstering Pallava ties with the other Tamil
kingdoms as well as strengthening his forces and preparing them for
revenge.
Despite its recent humiliation at Chalukya hands, the Pallava army was
still led by some of the subcontinent’s most feared commanders. Among
them was Mahendra’s son and heir Narasimha-Varman, who would prove to
be the Chalukyas’ nemesis. Narasimha came to the throne around 630,
when Chalukya campaigns in Andhra were coming, at last, to a delicate
conclusion.53
In 631, Vishnuvardhana Chalukya’s new regime in Vengi had finally
secured control over the blood-soaked coastal plains. His power stretched
from the modern-day Visakhapatnam district in the north to Guntur in the
south.54 Pulakeshin II himself was invited to visit on a royal tour, and served
as a witness to a land grant made by his younger brother. In a tacit
acknowledgement of Vishnuvardhana’s growing might, the grant declared
that his descendants, not Pulakeshin’s, would inherit the kingdom of
Vengi.55 Pulakeshin probably saw these new Chalukyas of Vengi as vassals
– which was convenient for his ambitious younger brother, the hunchback
who was now,
against all odds, a king. The Pallavas thus had powerful Chalukya-affiliated
neighbours to their north (Vishnuvardhana) and west
(the Gangas).
It is unclear precisely what happened next: perhaps Pulakeshin sought to
put down any hopes of Pallava resurgence; perhaps he was encouraged to
do so by the Gangas, his in-laws, who bore a particular grudge against the
Pallavas,56 or perhaps Narasimha, confident in his preparations, somehow
goaded the elderly Chalukya emperor into
an attack.
On Pulakeshin’s previous campaign in Tamilakam, he had managed to
ensure the neutrality of the other Tamil kingdoms, possibly through his
time-honoured strategy of bribery. By now, however, Narasimha appears to
have managed to convince the other Tamil kings that Pulakeshin was a far
greater threat to them than the Pallavas in the near run.57 A coalition of
three kings – perhaps the Chola, Pandya and Pallava – now awaited his
entry into their lands. Pulakeshin, oblivious and overconfident, marched
from the arid plateau of the Deccan into the rice fields of Tamilakam.58
Within a matter of weeks, he was 20 miles from Kanchi, at a village called
Manimangala, which still survives.
And there, in a massive upset, Narasimha defeated him.
Pulakeshin retreated, licking his wounds. He gave battle again. And
Narasimha defeated him again.
How could Pulakeshin II, Defeater of Harsha, not prevail? Surely all that
was required was another attempt.
And, again, Narasimha defeated him. The Chalukya army, under the
hitherto victorious boar crest of Vishnu, fled the billowing Pallava bull
banner, the resplendent white mount of Shiva. The Deccanis were driven
from the Pallava lands. It was a disastrous blow to the prestige of
Pulakeshin and the Chalukyas.
Pulakeshin’s victories of the 610s and 620s had been nothing but one
brilliant success after another. He may very well have believed in his own
invincibility after all those years of watching the Malaprabha river valley
grow prosperous, jeering at fleeing foes, and luxuriating in his new-found
wealth. And now, fortune had deserted him. In slow motion, Pulakeshin’s
domains descended into turbulence and chaos in the 630s, as the chiefs of
the Deccan began to once again question Chalukya authority.
Pulakeshin had boasted endlessly of his victory over Harsha in 618 and
had gone about building an image of invulnerability through his brutal wars
to keep his vassals in line and cow his neighbours into submission. Now the
Pallava Narasimha gleefully gave this man, the hero of the Deccan, a taste
of his own medicine and milked his victories for his own propaganda. A
later Pallava inscription describes him as: ‘Narasimhavarman … who wrote
the syllables (of the word) victory … on Pulakeshin’s (fleeing) back, which
he saw in the battles of Pariyala, Manimangala, Suramara, etc.’59
With his position and prestige so badly shaken, Pulakeshin must have cast
about desperately to reassure his vassals and to search for allies. One of
them, perhaps, was his own brother – who, rather than come to Pulakeshin’s
aid, declared himself Maharaja Vishnuvardhana, Great King, an
independent royal title, not that of a vassal, in
641 CE.60 Perhaps this was a betrayal, or perhaps it was a prearranged deal,
struck to ensure that Vishnuvardhana did not contest his nephews’ right to
succeed to the throne of the ageing, exhausted Pulakeshin.61 That would
prove to be moot, because that throne faced a much more dangerous threat
than fratricide: the very next year, 642 CE, Narasimha’s elephants were at
the gates of Vatapi.62 It was time for the Pallavas’ revenge, time to put an
end to these upstart Chalukyas.
How did Pulakeshin II, once Ereya, the climber of rocks, once the rebel
who murdered his uncle, once a city-seizing emperor, face his death? Did
he sally forth from the walls knowing he would not return alive? Did he
wait quietly in his palace, watching the enemy elephants break through the
gates of the citadel, their soldiers pouring through the city? Was he forced
to watch his great victory pillar, established to commemorate the victories
over Harsha and Vengi, being uprooted by a team of elephants, to be carried
like a tree trunk to Kanchi?63 Or perhaps he had died a sad death well before
Narasimha Pallava’s arrival.64 As with so many things about early medieval
India, we don’t know for sure, and we never will. His own descendants
never mention the exact circumstances of his death, as though it was taboo
to speak of such a great conqueror like he was a mortal.
Though the defeater of Harsha was himself defeated, that legend would
never die. For the rest of the Chalukyas the bloody cycle of vengeance had
only just been set in motion.
Bakker 2014, 112.
Karen C. Lang, ‘Candrakirti on the Medieval Military Culture of South India’, in Buddhism in the
Krishna River Valley of Andhra, eds. Sree Padma Holt and A.W. Barber (SUNY Press, 2008), 130.
The Vishnukundins of Vengi had to face multiple incursions from the Pallavas of Kanchi at
roughly the same time that the Chalukyas under Pulakeshin I first began to establish their kingdom.
Dikshit 1980, 92.
Ibid., 91.
Buchanan 1985, 100.
Dikshit 1980, 92.
The ensuing discussion, it should be noted, is by necessity a simplistic explanation of a much more
complex process of religious change. The rise of Puranic Hinduism and the fading away of south
Indian Buddhism is still a matter of considerable scholarly debate.
Himanshu Prabha Ray, ‘Creating Religious Identity: Archaeology of Early Temples in the
Malaprabha Valley’, in Archaeology and Text: The Temple in South Asia, ed. Himanshu Prabha
Ray (Oxford University Press, 2010), 15–37.
See John Michael McKnight, ‘The Gupta Temple Movement: A Study of the Political Aspects of the
Early Hindu Temple’, PhD dissertation (McMaster University, 1973); Michael D. Willis, The
Archaeology of Hindu Ritual: Temples and the Establishment of the Gods (Cambridge University
Press, 2009); Nicholas B. Dirks, ‘Political Authority and Structural Change in Early South Indian
History’, Indian Economic and Social History Review 13, no. 2 (1976): 125–57,
https://doi.org/10.1177/001946468101300201.
Buchanan 1985, 26–27.
Ibid., 22.
Ibid., 22.
Buchanan 1985, 32.
See Part 2, ‘Kalyana and the Chalukya Legacy’, in Richard M. Eaton and Phillip B. Wagoner, Power,
Memory, Architecture: Contested Sites on India’s Deccan Plateau, 1300–1600 (Oxford University
 Press, 2014), 77–161.
Buchanan 1985, 109.
EHD I, 205.
Dikshit 1980, 22.
Ibid.
Ibid.
See Vidya Dehejia, The Body Adorned: Sacred and Profane in Indian Art (Columbia University
Press, 2009), specifically Chapter 1.
EHD I, 216.
Ibid.
EHD II, 472.
Aruna 2000, 101.
Ibid., 102.
Ibid.
EHD II, 472.
Lang 2008, 135.
Ibid.
Ibid., 140–42.
Ibid., 138.
Ibid., 128.
Karen Pechilis Prentiss, The Embodiment of Bhakti (Oxford University Press, 1999), 69.
Ibid., 71.
Vidya Dehejia, Slaves of the Lord: The Path of the Tamil Saints (Munshiram Manoharlal, 2002), 33.
Ibid., 56.
Prentiss 1999, 62–63.
Ibid., 72.
Dandin, Tales of the Ten Princes, trans. A.N.D. Haksar (Penguin India, 2007), Kindle Edition,
Location 1156.
Dehejia 2002, 37.
Marilyn Hirsh, ‘Mahendravarman I Pallava: Artist and Patron of Māmallapuram’, Artibus Asiae 48,
no. 1 (1987): 109–30.
Chakrabarti 2011, Location 1669.
Mahendravikramavarman, ‘Matta-Vilāsa: A Farce’, trans. L.D. Barnett, Bulletin of the School of
Oriental and African Studies 5, no. 4 (1930): 697–717.
Hirsch 1987, 118.
Wendy Doniger, The Hindus: An Alternative History (Penguin, 2009), 363.
Lang 2008, 143.
Dikshit 1980, 94.
Ibid.
EHD I, 216.
The ensuing description is based on Chalukya inscriptions and information from medieval courtly
sources, as studied in Ali 2006.
The Arthashastra recommends that when the enemy possesses a strategic advantage, samsraya or
submission is the best policy.
Chakrabarti 2011, Location 1672.
Dikshit 1980, 95.
Chakrabarti 2011, Location 1659.
N. Venkataramanayya, The Eastern Cāl.ukyas of Vēngi (Vedam Venkataraya Sastry & Bros, 1950),
17.
Dikshit 1980, 77.
Ibid., 94.
EHD I, 217.
South Indian Inscriptions, Vol. I, 152. Hereafter SII. It should be noted that the Pallavas might have
been conjuring great victories out of indecisive clashes to bolster their military reputation.
Dikshit 1980, 95.
EHD I, 217.
Ibid., 219.
SII II, no. 98: Velurpalaiyam Plates of Vijaya-Nandivarman (III).
Buchanan (1985) argues it is more likely that the Pallavas invaded just after Pulakeshin’s death,
while other authors have presented his death more dramatically.
 3
Vikramaditya’s Revenge
Summer, 642 CE

The Pallava army gorged themselves on the city of Vatapi, fat with the
wealth of generations of Chalukyas. The palace was sacked, the men killed
and women seized,1 and its temples looted. According to a later oral legend,
the ‘Vatapi Ganapati’ idol, today the subject of one of the most famous
hymns in Carnatic music,2 was among the spoils, carried away to the
Uthrapathiswaraswamy temple near modern-day Nagapattinam3 in Pallava
territory.
The survivors of the sack of Vatapi – all those immigrant preachers,
architects, traders, officials, cooks, sweepers and peddlers – must have fled
with what little they could carry, and would not return for years. The
Pallava king ordered a message to be left for any Chalukya prince who
dared return. In a temple in the capital, he had engraved in the
characteristically beautiful, almost floral Pallava script that his occupation
of his enemy’s capital was a ‘fierce retribution’ that befell Pulakeshin II for
having twice threatened Kanchi.4 On a boulder near the city, he issued
another large inscription, later defaced but still preserving the words
‘Vatapi’ and ‘Maha-Malla’ (Great Wrestler), one of Narasimha’s many
titles.5 The Pallava king also adopted the title of Vatapi-Konda, Conqueror
of Vatapi, which would pass into his dynasty’s legends, just as ‘Defeater of
Harsha’ had passed into the legends of the Chalukyas.
Through the rest of his reign, Narasimha would never let the Chalukyas
forget their humiliation at his hands. He chose to do this in a way that
shows us that the Pallavas too were masters of using myth to support
propaganda. In both the Ramayana and Mahabharata, the sage Agastya is
tricked into eating a demon called Vatapi, who would magically transform
himself into a meaty curry to entice weary travellers to eat him, after which
he would turn himself back into his demon form, blowing them apart from
inside. Agastya, however, immediately digested Vatapi and expelled the
remains as gas. This amusing story was recognizable to well-educated
courtly audiences, and Narasimha made the most of it. He boasted that he
was the ‘equal of Agastya, the crusher of Vatapi, who frequently conquered
Vallabha-Raja (Pulakeshin II)’. 6
With the central hub of the Chalukya political network annihilated, and
without the wealth and income of the city of Vatapi to sustain them, the
court’s factions, vassals and families now splintered and went their own
ways. Pulakeshin II had married women from many of the Deccan’s most
prominent dynasties to bind them to him, but that now meant that a host of
his sons saw themselves as rightful successors to their father’s throne.7
Though the inscriptional record is unclear, it appears that a number of them
wielded varying degrees of influence in different parts of the Deccan, their
home bases probably selected with reference to the bastions of their
mothers’ families. Further complicating the picture were the machinations
of the other lords of the Deccan, who had little interest in bowing once
again to a powerful Chalukya ruler. Pulakeshin II’s empire building, it
would seem, was for naught: all the propaganda in the world could not
engender lasting loyalty.
Indeed, former Chalukya vassals were wasting no time in proclaiming
their independence, with varying degrees of confidence. The viceroy of
southern Gujarat adopted the title of ‘raja’ but did not claim to be a
sovereign ruler until he could be sure no victorious Chalukya would punish
him for his insolence. So too did the Sendrakas, Pulakeshin’s mother’s
family, who seem to have thought they might as well take their chances
with temporary independence.8 It would appear to have been quite a
sensible move: the eldest of Pulakeshin’s sons set up base near Kurnool,
Andhra Pradesh, and declared himself the Chalukya emperor, Sri-Prithivi-
Vallabha, only to die within a year,9 either assassinated or killed on the
battlefield. His son and successor was no luckier.10 Neither was a younger
son of Pulakeshin’s, whom we know only through the inscriptions of his
wife, the first of a series of formidable Deccan queens that we will meet: a
lady known as Vijaya-Mahadevi, who stepped in as regent for their infant
son. Vijaya-Mahadevi may have originally been a courtesan, and she was
not shy about it. The name ‘Boddivoddi’, with the ‘standard suffix’
(voddi/poti) for a courtesan, appears in a land grant that the queen issued,
and it has been argued that the construction of her sentences suggests that
Vijaya-Mahadevi and Boddivoddi are one and the same.11
Chalukya queens – as Mangalesha’s mother, who encouraged him to
attack her birth family, the Kalachuris, shows – did play a role in
determining their clan’s policies. But Vijaya-Mahadevi’s position as queen
is especially remarkable. Even aside from her origins as a courtesan, this
queen, as a regent for her son, issued land grants, hinting at something
unique at play: a woman rising to an unprecedented degree of prominence
and control in seventh-century Deccan. Earlier generations of queens were
generally only allowed to make grants and donations with the approval of,
and in the name of, their sons or husbands, and would appear in public only
on ceremonial occasions, usually in a subservient role to the men of the
family. But Vijaya-Mahadevi issued land grants in her own name. This
implies she was appearing before crowds, attended by some of the insignia
of kingship, indulging in behaviours that were usually within the domain of
royal men alone. She was claiming the divine right to redistribute land that
had hitherto been the exclusive preserve of the Chalukya kings.
 In contrast, Vijaya-Mahadevi’s rivals and predecessors among Chalukya
royal women had been confined to the court, though it should be noted that
they had considerable agency within the court itself.12 Royal women were in
many ways as educated and qualified as their male counterparts, learning
everything from politics to poetry, warfare to administration. Though often
treated as tokens of exchange to seal political relationships,13 it would be
unjust to these remarkable women to see them as only devoted mothers and
wives. As the men of the clan spilled blood on battlefields, they too fought
for favour and prestige and glory in the corridors and courtyards of the
antahpuram, the Inner City,14 the women’s quarters of the palace.
Princesses in medieval India brought with them their own dowries and
personal retinues, were often well connected both within and outside the
court, and could receive stipends from the state treasury.15
Within their Inner City were hierarchies, ranks and etiquette.16 Seniority
and rank among queens were determined by politics ‘directly continuous’
with those of the main assembly halls17 where the ruler received the
submissions of his vassals: their ranks were generally dependent on those of
their husbands, but the politics of the women in the Inner City could also
determine royal favours, appointments and titles, and thus indirectly
influence the outside world. A princess who successfully rose through the
ranks of the Inner City could, and very often did, serve as her birth family’s
primary lobbyist at court. Women’s interests were intimately tied to those of
their marital families as well, and in the event of a military disaster or
unfavourable succession, their bodies and freedom were even more at risk
of capture and assault than those of chiefs fleeing on war elephants. Only a
truly foolhardy king would dare ignore what the royal women thought of
his political or strategic decisions. Sanskrit literature is replete with
instances of these ladies conspiring to brutally murder, mutilate or exile
male members of the family.
The circumstances under which Vijaya-Mahadevi came to meet a
Chalukya prince, married him and was declared chief queen are unknown,
but it would certainly seem that she was a person of great ambition, ability
and charm.18 As a courtesan, in stark contrast to royal ladies who usually
appeared in public spaces only during ceremonial events, she would have
often been seen in theatres and music and poetry recitals, traditionally the
domain of men.
Whatever the circumstances under which her career began, Vijaya-
Mahadevi took to her role as chief queen with aplomb over the course of
the 640s. As the widow of a king, she had to manage and manipulate both
the men of the court and the women of the Inner City, in addition to
adhering to the ritual and social expectations that came with royal power.
As a former courtesan appearing in public as a queen, she was transgressing
orthodox expectations of women in public spaces. Her success in doing so
indicates that seventh-century Deccan societies had attitudes different from
those espoused by Sanskrit literature, and were less hostile to powerful
women than contemporary north Indian courts were.
One member of Vijaya-Mahadevi’s new circle was her brother-in-law
Vikramaditya,19 ‘Sun of Power’. He was Pulakeshin II’s son by a Ganga
princess, and he had been waiting in the wings for a long time. It is unclear
what he was up to in the immediate aftermath of his father’s death, but later
in the 640s he appears to have not been important enough to seize the
throne by himself, and thus supported Vijaya-Mahadevi and her son, the
emperor. But by the 650s, the death of his elder brothers meant he was the
most prominent adult male Chalukya still standing. And unlike Vijaya-
Mahadevi, he had a powerful dynastic backer: his mother’s family, the
Gangas of southern Karnataka. The Gangas now found themselves in the
happy position where the next Chalukya emperor might be half-Ganga.
Vikramaditya also married a Ganga princess,20 further cementing the
alliance. (Dravidian kinship allows for marriage between cousins.) With the
support21 of the Gangas and the friendship of Vijaya-Mahadevi, the regent
for the infant Vallabha, Vikramaditya set out to restore the Chalukya
political network, and began the bloody process of once again subjugating
recalcitrant chiefs to re-establish Chalukya supremacy – ostensibly in the
name of the infant emperor.
What Vijaya-Mahadevi thought of her brother-in-law’s growing power is
unknown. Perhaps she had plans and contingencies to make sure he did not
threaten her son’s (and thus her) hold on the throne. Meanwhile,
Vikramaditya’s new wife, the highborn Ganga-Mahadevi, must have had no
intention of playing second fiddle to a former courtesan in the Chalukya
Inner City. For now, though, their interests were in ostensible alignment.
The Chalukya territories were slowly reclaimed and Vatapi recaptured. The
inscription that Narasimha Pallava left on a boulder outside, boasting of his
humiliation of the Chalukyas, was erased almost a decade after it had been
made.
And so this round of wars between the Deccan and Tamilakam came to an
exhausting end. But there would be many more to come. The geopolitics of
south India practically mandated a rivalry between the Deccan and
Tamilakam – with Vengi in modern-day coastal Andhra often the prize at
stake, for its capture could give the Deccan access to the eastern seaports
and the Tamil land a near monopoly over eastern trade.22
Vikramaditya Chalukya, his reputation made, with his family’s former
vassals once again swearing their loyalty, now sought revenge against his
father’s murderer, Narasimha Pallava. But he had to first bloody his hands
with the lives of his kin.23 For when the middle-aged prince triumphantly
declared himself Vikramaditya I, Sri-Prithivi-Vallabha, to a circle of vassals
in 655 CE, no mention was made of Vijaya-Mahadevi and her infant son.
They are never heard of again – but it will not be the last time we hear from
a powerful Chalukya queen.

As the shattered Chalukyas regrouped, Vikramaditya I’s court decided that

two major changes were in order, setting in motion the last steps in the
Chalukyas’ transition from local chiefs to kings in the classical Indian
mould.
First, Chalukya power and political discourse needed to be set on a
standard ‘historical’ narrative to make sure their genealogy and legitimacy
were unquestionable. Vikramaditya I’s father (Pulakeshin II), grandfather
(Kirti-Varman I) and grand-uncle (Mangalesha) had already made the initial
moves in that direction, with the Varaha iconography and the appropriation
of Kadamba land grant formulae featuring Karttikeya and the Seven
Mothers. However, as noted, a number of different legends of their origin
were already in circulation during Pulakeshin II’s time. This problem was
exacerbated by the gossip and myth-making that inevitably accreted around
such larger-than-life royal characters. Now, any stories about Chalukya civil
wars and fratricide – of which there had been plenty in the last two
generations – needed to be erased from official records. The dynasty needed
to be presented as the embodiment of the virtues of kingship to bolster its
legitimacy and right to rule.
To address this problem, Vikramaditya embraced one of the most
distinctive literary forms of the early medieval period: the prashasti.
Prashastis are royal eulogies which seek to connect living, breathing
dynasties to divine and heroic ancestors, and in the process articulate
official ‘histories’ of royal lives, enumerating kings’ military exploits,
valour, beauty and sexual prowess. They were used to preface land grants,
which were integral to the devolution of royal power, and were widely
distributed and read aloud by kings’ agents and vassals. As such,
standardizing them was crucial to ensuring that a ruler’s subordinates,
successors and subjects all received the same loud, confident political
message from his court. This was vital for the Chalukyas in the turbulent
years of the mid-650s, as Vikramaditya I
set out to redistribute the lands of the Deccan among his loyalists.
While the majority of the Chalukyas’ subjects spoke – and would
continue to speak – a wide variety of regional dialects of Old Kannada,
Vikramaditya’s prashastis and land grants were in Sanskrit.24 Part of this
was due to Sanskrit’s association with legend and divinity, which it had
possessed since the days of the Vedas. But in the early centuries CE,
politicians and poets had also begun to experiment with its copious
grammars and lexicons, inaugurating a flourishing literary culture. By the
seventh century, new myths, epics, religious stories, dramas and poems
bloomed across the subcontinent, and Sanskrit had become established as
the premier language of intellectualism, literature and power, creating a
universalizing, cosmopolitan world for the upper classes. This cosmopolitan
use of Sanskrit had been introduced to the Malaprabha valley by the early
Chalukya kings, and we have seen how they used Sanskrit titles and terms,
as well as myths and religious ideas expressed in Sanskrit, to justify their
rule.
It was to this pan-subcontinental ‘Sanskrit Cosmopolis’, as Sheldon
Pollock calls it, that Vikramaditya’s prashastis were primarily addressed,
not to his subjects. Narasimha Pallava, through his Vatapi analogy, had
humiliated the Chalukyas before this elite audience; Vikramaditya now
sought to restore his reputation before them. Through their new prashastis,
the Chalukyas declared they were warlike and virtuous heroes in search of
fame and glory like their ancestors had supposedly been from time
immemorial;25 they thus granted themselves the hallowed sanction of
dharma, myth and history – which, in medieval India, were often
inseparable. Chalukya prashastis were declaimed or sung in the appropriate
heroic metres; the land grants they prefaced were accepted with folded
hands and touched to the forehead of the grantee. The audience – officials,
courtly layabouts, queens, princes, priests, and administrators across the
Deccan – would henceforth all hear the same standard history, be reminded
of the same ideas, and all increasingly think of the Chalukyas as a
permanent Deccan institution.
For more than half a millennium, those who claimed the Chalukya name
would all begin their prashastis with the same splendid formula. This now-
forgotten sound was declaimed for centuries from the top to the bottom of
the Deccan, etched into the minds of generations of its poets and
panegyrists and priests and officials and kings. It must have been one of
medieval India’s most recognizable jingles:
jayatyāvis.kr.tam. vis.n.orvarāham. ks.obhitārn.avam.
daks.in.onnatadam.s.t.rāgraviśrāntabhuvanam. vapuh.
Victory to the Boar Incarnation of Vishnu, shaking the ocean as it comes into view
 With the Earth resting at peace on the tip of its upraised right tusk26
This image vividly embodies not only the Chalukya identification with
Varaha, but also the idea of Indian kingship as fundamentally leading to the
salvation of the king’s subjects.27 Despite their brutal wars and perfumed
courts distanced the toiling masses, the Chalukya kings wanted to be
thought of as saviours. The Earth was ‘resting at peace’ because of their
successful performance of royal dharma, commencing with their imaginary
ancestors from the very beginning of time; there is no longer any hint of
their humble origins. The prashasti continues, detailing a rewritten
Chalukya genealogy in extravagant Sanskrit. The new official account
mentions Pulakeshin I, performer of Vedic sacrifices, learned in dharmic
texts; praises his elder son, the father of Pulakeshin II and the conqueror of
the Kadambas; sings the praises of Pulakeshin II, defeater of Harsha; and
then describes Vikramaditya himself, Sri-Prithivi-Vallabha, the son of
Ganga and Chalukya. Of all the ideas the Chalukyas had played around
with in their previous propaganda, and of all their acrimonious family
history, this was all that made the final cut. No mention is made of
Mangalesha, Vijaya-Mahadevi or Vikramaditya I’s elder brothers.
Prashastis were written in an ornate and impressive language, showcasing
how deeply the Chalukyas were now involved in the subcontinent-spanning
world of Sanskrit. But a prashasti also erased much of the complicated lives
and motivations of the people it mentioned. Unfortunately, prashastis make
up the vast majority of texts that survive from medieval India, giving us a
lopsided view of a possibly far more complex and interesting past.
Five years after his coronation, on 30 April 660 CE,28 Vikramaditya I
made the second of his two great changes to Chalukya kingship, reorienting
the family’s religious support base towards Shaivism, better reflecting the
reality of Deccan power equations. The influence of Shiva-worshipping
teachers and institutions had been spreading across the northern Deccan
over the last centuries, shifting large sections of the population away from
their ancestor- and nature-worshipping ways. In 660, Vikramaditya
underwent a formal Shaiva religious ceremony at Marruru, near Alampur, a
flourishing pilgrimage centre at the holy spot of the confluence of the
Krishna and Tungabhadra rivers29 – the very spot where his father
Pulakeshin II had paused in the early 620s to divide the spoils of Vengi
among his followers. The guru who performed this shiva-mandala-diksha
(Initiation into Shiva’s Circle) was granted the revenue of an entire village
for his services. Hinting at how systematically Shaivas worked to expand
their footprint, this priest then distributed plots of land in the village to
other Shaiva priests.30
Thanks to activities like these, the age of Shiva had well and truly begun
in the Deccan by the late seventh century. Once again, we see how the
activities of royals were often only responses to much grander forces, and
how courts constituted only one among many powerful organizations that
shaped India’s history. Dozens of cults dedicated to Shiva flourished across
the Deccan plateau. But whereas Tamil Shaivism was assisted by the spread
of the bhakti movement, in the Deccan, this transformation was due to the
efforts of organized sects such as the Pashupatas – one of whom had
appeared in Mahendra Pallava’s Farce of the Drunkard’s Games – and the
Kalamukhas or ‘Black Faces’, who smeared their foreheads with a black
streak.31 These oddball ascetics delighted in breaking taboos and flouting
the orthodox views of the Dharmashastra. Pashupatas, for example,
engaged in ‘various forms of antisocial behaviour’ including going about
‘making lewd gestures to young women’, especially those of good families,
which would earn them abuses and beatings.32 This act would magically
transfer their sins to the abusers, and transfer to them whatever religious
merit the abusers happened to have accrued through more traditional means
– such as following their dharma,33 going on pilgrimages, making charitable
donations and so on. Pashupatas also made a habit of hanging around
cremation grounds and smearing their bodies with ash – which signified the
‘death of the body and of sexual desires’.34 They were often major
presences in some towns, living off petty charity and monastic institutions
attached to temples and paid for by royal endowments.
The Pashupatas had spread over centuries from western India into the
Gangetic plains and the northern Deccan, establishing massive monasteries
overseen by charismatic teachers preaching their doctrine, supposedly first
recited by a corpse reanimated by Shiva. So active and energetic were they
in spreading the word of Shiva that there were Pashupata preachers in
Cambodia in the sixth century CE,35 travelling alongside merchants and
Buddhist monks to East Asia and challenging the modern notion that
‘Indian’ religions were not interested in proselytizing. Medieval monks, like
kings, were not averse to wealth, fame and power.
But Pashupatas were only one of many flourishing, squabbling Shaiva
cults that spread into the Deccan in the early centuries of the first
millennium. The Kadambas, those early nemeses of the Chalukyas, had
donated extensively to monasteries, temples and ascetic groups that
worshipped Shiva in myriad forms and ways.36 These organized Shaiva
sects had a very different approach to their god compared to the bhakti
saints: they were far more comfortable with royal patronage for one, and
were big believers in scripture37 and ritual. They also used the revenues
granted to them to set up feeding houses,38 conduct religious processions
and intercede with the gods on behalf of the laity for humdrum matters such
as harvests and births. Such intercessions may have been considered
abhorrent by the bhakti saints, but were apparently popular with
worshippers. As their influence grew, the most prominent Shaiva monastic
gurus would command the attention of elites and lay worshippers across the
borders of kingdoms, and control resources that would put most petty kings
to shame.39 Some were even posthumously worshipped in the form of a
Shiva linga.40 The institutions that these gurus led would eventually rise to
dominate south India’s daily lived practice of Shaivism.
Early Chalukya kings, reflecting the societies they ruled over, had
incorporated many subtle Shaiva influences. This is why they had claimed
in their inscriptions to be supported by Shiva’s son, the war god
Skanda/Karttikeya, while simultaneously portraying themselves as
incarnations of Vishnu. In 660, Vikramaditya I decided that a direct royal
association with Shaiva institutions offered a better means of strengthening
Chalukya authority in a time of crisis, never mind that the Pallavas were
also Shaivas. Restoring the Chalukya political network meant that
accommodations had to be reached with powerful players – and who were
more powerful than the Shaiva sects?
Soon after his initiation at Marruru, Vikramaditya likely underwent a
second royal consecration, perhaps in a field in Vatapi. This time, the
ceremony probably took place under the auspices of a Pashupata guru or
representative of some other Shaiva group, rather than the Vedic Brahmin
chaplains who oversaw the royal consecration of his great-grandfather
Pulakeshin I – another sign of how much the Deccan had been transformed.
Pashupata rituals were also different from the bloody Ashvamedha that had
consecrated Pulakeshin I. One version of their consecration ritual required
that the king and his consort be seated on ‘a platform covered with the skins
of a fighting bull and a cat’,41 after which they had to ‘offer worship to
Shiva, the Fire, the [royal] weapons, the [royal] banner’.42 Next, they were
to consume the ‘five products of the cow’43 – a fermented mixture of milk,
ghee, curds and a little dung and urine – along with ‘rice porridge [prepared
on the sacred fire]’44 and clean their teeth with a twig. Rather than seeking
to reorder the universe through the sacrifice, as Vedic religion claimed to
do, the Pashupata ritual was based on tantric ideas that magical formulae
and ritual actions infused the practitioner directly with power, an idea that
kings must have found quite gratifying. It reflects the growing influence of
tantrism in Indian religions.
The next day, after the recitation of sixteen sacred mantras, consecrated
water was poured on Vikramaditya’s head – this was the rajyabhisheka
ritual that would continue to be practised by Indian royals for centuries
after. Water may also have been sprinkled on the front rows of his elite
troops, waiting in the light of the early morning, topknots oiled, swords
sharpened, shields of wood and bamboo freshly painted with boars and
lions and elephants and flowers.45 Preceded by the resplendent Chalukya
boar banner, now ‘blessed’ by Shiva, the king may have led a ‘full military
parade’,46 undoubtedly mounted on his beloved piebald horse Chitrakantha,
imported at great expense through a port on the west coast that the
Chalukyas controlled once again. Vikramaditya must have wound through
the streets of the city where he had grown up in the shadow of his dead
brothers and father, accompanied by the sounds of drums and the
triumphant roars of his subjects as they showered him with parched rice
from the roofs of their mansions. Artisans and poets were returning to
Vatapi, vassal lords came once again to pay obeisance to Fortune’s
Favourite. But Vikramaditya would not forget why they had stopped their
tribute in the first place.
From his citadel atop the sandstone cliffs, the Vallabha looked south, to
the Tamil country.

Narasimha Pallava had not been idle the last few decades. He was, like his
father, a man of great aesthetic sense, and the treasure he had seized from
Vatapi was poured into an ambitious construction project: a new Pallava
royal city. Though its construction may have been begun by his father
Mahendra, it would be most enduringly associated with Narasimha, and is
remembered by one of his titles: Maha-Malla-Puram, ‘City of the Great
Wrestler’ (modern-day Mamallapuram). It was built on the shore to connect
the Pallava court directly to the profits of the Indian Ocean trade, and it
would be decorated by dozens of unique monolithic shrines carved from
entire boulders, many of which still survive today. This was a natural step
upwards from carving temples out of caves, and may have been intended to
outdo Pulakeshin II’s freestanding temples at Vatapi, assembled from
blocks of cut
stone. This ‘architectural arms race’ is an indication of political struggle:
temples were (and are) not built purely from outpourings of devotion.
 What a city Mahamallapuram was. Narasimha’s treasury there, cut into
the bedrock and sectioned off with bricks for different categories of
valuables, is still visible today. So is a proud monolithic lion-shaped throne
gazing balefully to the north, where his Chalukya rivals lived. On it,
Narasimha Pallava lolled about, a silk cap covering his oiled hair gathered
in a knot,47 taking in the ‘cooling off-shore breezes from the East … or a
commanding view of irrigated paddy and orchards, stretching off … beyond
to [his capital] Kancipuram’.48
In Mahamallapuram, Narasimha Pallava also embarked on a remarkable
creative experiment, which shows us once again the close interaction of art,
politics and myth in the framework of medieval Indian power. This
experiment was the massive narrative sculpture known as the Great
Penance (or The Descent of the Ganga), carved from a single block of
sandstone and profusely bedecked with animals, gods, sages, serpents and
sophisticated visual double entendres that put even the Chalukya kings,
with their sculptural and poetic allusions to Vishnu, to shame. Indeed,
putting the Chalukyas to shame was probably one of Narasimha’s
motivations in commissioning this masterpiece, though his primary goal
was, of course, his own glorification and, by extension, the glorification of
the Pallavas. For a project on this scale, there was, luckily, a medieval
Tamil Michelangelo at hand: the great artisan Mandhatar, ‘who surpassed
the Greeks in his artistry’.49
At the centre of this great sculpture, an emaciated sage, every muscle and
bone standing out, balances on one foot with his sunken eyes fixed on the
heavens. Next to him, the merciful god Shiva extends his hands in blessing,
attended by portly laughing dwarves. Half-bird, half-human beings called
kinnaras sound a great din with horns, drums and cymbals. Animals join
the host: deer, lions, geese, monkeys, cats, mice, elephants. Each of them is
carved with remarkable realism and attention to detail, and almost all of
them gaze with great wonder at the centre of the sculpture. All around them,
a great host of gods mobilizes to attend this event, the ones further away
looking intently and benevolently at the audience, gesturing towards the
sage.

There is an immense cleft at the centre of the rock, which once flowed
with crystal-clear water through which shimmered the carved nagas – water
serpents, fertility divinities with the upper half of a human and the lower
half of a great snake, believed to live in rivers. And next to this vertical
river is that starving sage, whom the audience will know is Bhagiratha, the
sage who prayed to Shiva that the Heavenly Ganga be allowed to descend
upon his densely matted hair and thus come to Earth.50 Of course, as the
bhakti saints and the Shaiva sects promised, the merciful Shiva had
appeared to him and granted his boon, and as proof, the holy Ganga still
flowed through the lands of the plains over 2,000 kilometres to the north.
But to a sophisticated audience trained in art, mythology and literature,
many other interpretations of the sculpture were evident. The starving
ascetic, for example, could be not only Bhagiratha, but also Arjuna, a great
hero of the Mahabharata. The ascetic is receiving from Lord Shiva the
mighty weapon known as Pashupata, symbolized by a dwarf with a
demonic face painted on his belly. Arjuna was famed as a great wrestler, but
so was King Narasimha-Varman, the Maha-Malla. The sculpture can thus
also be read as a eulogy to the Pallava king, a declaration that he, like this
great hero, had earned Shiva’s special favour.
Moving away from the centre, the huge composition is full of delightful
nuggets for interpretation. Near the bottom of the panel, where the waterfall
once spread into a pool, stand two figures, grinning at the water; one of
them holds a pot, a kumbha. Now it just so happens that the sage Agastya –
who destroyed Vatapi – was born from a pot.
(The story of Agastya’s birth: The gods were performing a great sacrifice.
Among the chief officiants were Mitra and Varuna. The gorgeous apsara –
divine nymph – called Urvasi joined the crowd. So alluring was her walk,
so tempting her smoky eyes and luscious curves, that these two gods were
not only aroused but actually ejaculated on the spot. Their semen poured
into a pot, from which was born Agastya, henceforth known as kumbha-
yoni, water pot womb.51)
Who else had destroyed a Vatapi as Agastya had? Narasimha, king of the
Pallavas.
And at the bottom of the sculpture, three sages were carved, allusions to
King Narasimha, his father King Mahendra, and his grandfather King
Simhavishnu,52 but also to Asvatthama, an ancestor of the Pallavas
according to their own imaginary prashasti genealogies, his teacher-father
Drona (one of the commanders of the Kaurava army in the Mahabharata),
and his teacher Agastya the Pitcher-Born.53 These sages, representing three
generations of Pallava kings of great importance to the dynasty’s self-
image, are without heads today. The reason for this will soon become
apparent.
Such an elaborate programme of connection between a living dynasty and
the myths of heroes and sages had hitherto only appeared in Sanskrit
prashastis. The Great Penance, brought to life at Mahamallapuram by
Narasimha Pallava and the master artist Mandhatar, is thus a sophisticated
prashasti in stone. In the race to gain prestige by patronizing the greatest
artists and to express political ideas through religion and art, the Pallavas
had outdone the Chalukyas – and they had built this after defeating and
looting their capital. It was a marvellous embodiment of the dynasty’s
humiliation and the glory of the Pallavas.
Vikramaditya Chalukya must have heard of this, sooner or later. Traders
constantly moved between India’s east and west coasts as well as overland,
as did poets, artists and intellectuals. He learned of the monolithic
sandstone boar that Narasimha Pallava had had carved and immersed in a
temple pond, symbolizing his devotion to Varaha, and also his capture of
the Chalukya boar standard.54 Perhaps architects and sculptors spoke with
admiration of the wave of new temples that Narasimha was having made –
freestanding monoliths carved out of great boulders, almost as if he was
thumbing his nose at Pulakeshin II’s temples, assembled from many smaller
pieces of sandstone at Vatapi. Carved with marvellous reliefs with dozens
of characters – representing the manifestations of the Great Goddess,
Vishnu, Shiva or the wily Mahendra Pallava – their domes adorned with
beautifully spaced and carved miniature temples, they were like nothing
south India had yet seen.
Artistic masterpiece or not, Vikramaditya decided it was time for the
Chalukyas to have their revenge on Mahamallapuram.
Vikramaditya I had come to the throne in his middle age, and he would
bring no less than three generations of royal Chalukyas to the project of
crushing the Pallava enemy.
During the tragic regency of Vijaya-Mahadevi, this Chalukya had already
built up a reputation as a king killer. Her inscriptions refer to Vikramaditya,
her heroic brother-in-law, destroying kings in many battles, achieving
victories in all directions, and restoring the ‘glory and fortune’ of his
ancestors.55 Even if this is exaggeration, meant for Vijaya-Mahadevi to gain
some support by flattering her increasingly powerful and ambitious relative,
Vikramaditya certainly seems to have had a long, brutal military career; in
his prashastis he refers to himself as possessing armour into which ‘many
blows had been plunged’.56 In his first wars against rebellious chiefs in
erstwhile Chalukya territory, he built up the experience he would need to
take on Narasimha Pallava.
This endeavour was marked by the same patience that characterized his
early years. Vikramaditya was nothing if not dogged and willing to pay the
price for the pursuit of his ambitions. He had waited nearly a decade for his
brothers to finish each other off before making his bid for power, allying
with Vijaya-Mahadevi and disposing of her and her son when he no longer
needed them. He now waited for more than a decade after his coronation,
working out the political and religious bases of his power, before
challenging the Pallavas again. In addition to his experiments with
prashastis and supporting the Shaivas, he, like his father Pulakeshin II,
spent time consolidating Chalukya interactions with the rich revenue
streams of the western coast, also importing Arab horses at great expense to
maintain elite cavalry squadrons.57 In Lata in southern Gujarat,
Vikramaditya, again like his father, allowed a trusted younger brother to set
up a cadet branch58 (though these Chalukyas of Lata would eventually turn
out to be only marginally more loyal than the Chalukyas of Vengi,59 who
had so memorably declared themselves independent just when Pulakeshin II
desperately needed their assistance against the Pallavas in 641/2). In the
meantime, the Pallava king Narasimha, his new city built but his success
perhaps soured by a restored Chalukya kingdom always just beyond his
reach, died. His successor Mahendra-Varman II attempted to attack the
Gangas of southern Karnataka as a prelude to attacking the Chalukyas, but
was defeated and killed.60
In 674, Vikramaditya finally decided his time had come61 and led a
marauding Deccan army once more into Tamilakam. Over the next few
years, he set out to brutally avenge Narasimha’s conquest of Vatapi,
claiming for himself the title of Ranarasika, ‘Lustful for Battle’, and
Rajamalla, ‘Wrestler of Kings, because he has destroyed the family of the
Mahamalla [Narasimha Pallava]’.62 In this campaign, the king was also
likely attended by Shaiva priests, who offered religious services including
pre-battle consecrations to ensure victory.63

After a gap of nearly half a century, the Chalukya boar banner once again
arrived at Kanchi, forcing the young Pallava king – Narasimha’s grandson –
to flee, abandoning his people to Vikramaditya’s mercy. Not that the
Chalukya ruler was in a particularly forgiving mood. His inscriptions
explain that Kanchi’s ‘unfathomable’ moat was filled in, its
‘insurmountable’ rampart scaled by his troops, and the city brutally sacked.
He refers to the event in language that is shockingly different in tenor to the
almost comedic way in which the Pallavas treated their seizure of Vatapi.
For Vikramaditya, drawing on the Chalukya tradition of referring to kings
as ‘lovers’ of the Earth, describes with almost macabre glee how he had
‘forcibly wooed the lady of the Southern Quarter and taken possession of
Kanchi, the city which was her girdle’.64 Success in war was a sign of
virility and masculinity for medieval Indian kings.
Next in his sights was Mahamallapuram, that ode to Pallava power and a
lasting homage to their humiliation of the Chalukyas. Once the city had
been looted and its population either cowed or scattered, we can imagine
the Chalukya king riding calmly down its wide avenues, mounted on
Chitrakantha, his famous piebald steed. He was probably surrounded by a
detachment of elite bodyguards on similarly splendid imported horses. If
Chalukya sculptures are any indication, the king himself wore a lower
garment of fine silk, and was bare-chested and bedecked in only a few
necklaces to take in the warm sea breeze.
 How did Vikramaditya react as he took in the sight of all the temples built
with stolen Chalukya gold? All we know for certain is that he visited
Narasimha Pallava’s temple complex, where he was shown the particular
images which referred to the Pallava kings Narasimha, Mahendra and
Simhavishnu. Perhaps a grin of ghoulish satisfaction crossed his face as he
ordered the images beheaded.65

Strange as it may seem to us today, the effacement of Pallava art does

indeed seem to have been one of the objectives of the Chalukyas in their
campaign in Tamilakam in the 670s. It shows us that for medieval Indian
rulers, art was a profound religious and political statement, worth investing
in architects, sculptors and stonemasons to create, but also worth the
expense and risk of gathering warriors, fighting battles and seizing cities to
destroy. Today, the monolithic temples at Mamallapuram still bear the scars
of Vikramaditya I’s raid – by one count, 264 finials were chiselled off.66 The
Varaha that Narasimha had immersed in a tank to commemorate his capture
of the Chalukya boar standard was shattered by the blows of heavy
hammers.67 The Chalukya king may have planned to engage in similar
iconoclasm on the rest of the Great Penance relief and other portraits of
Pallava kings in the Mahamallapuram temple complex. Luckily for
posterity, he seems to have been called away before he was able to do so.
Far from being cowed by this onslaught, not only was the young Pallava
king gathering an army to repulse the Chalukyas, but his allies – the other
kings of Tamilakam – were rallying in support. Vikramaditya would not
make the mistake his father Pulakeshin II had made in ignoring the other
Tamil kingdoms, and embarked hurriedly on an expedition against them.68
As Vikramaditya rampaged through Tamilakam, the Chalukya political
network in the Deccan was back to full bloom, aided by the new
prominence of the Gangas. At home, his son Vinayaditya and grandson
Vijayaditya consolidated the Chalukya royal line through marriage ties and
war, ‘rooting out all thorns from the body politic’.69 Vikramaditya’s brother-
in-law and cousin, the Ganga king, also chalked up a series of successes
against the Pallavas.70
Eventually, old age, the vast and indomitable Tamil country, and the
sandstone heads of dead Pallava kings convinced Vikramaditya he had
satisfied the call of ‘honour’, and he retired back to Vatapi. His decision
was probably spurred by a sudden reversal in Peruvalallanur, where his
army of ‘several lakhs’71 was supposedly put to flight by a smaller Pallava
force.72 It seems more likely, given the way the rest of the campaign
unfolded, that the Pallavas managed to inflict some casualties on the
Chalukya rearguard as it withdrew from their lands. Whatever the truth of
the matter, it was now clear to all of south India that the Chalukyas were
back, that they were not to be crossed, and that they were not going to lose
their hard-won supremacy any time soon.
Malini Adiga, ‘Of Concubines and Devadasis: Prostitution in Early Medieval Kartataka’,
Proceedings of the Indian History Congress 73 (2012).
R.H. Kulkarni, ‘Vatapi Ganapati, an art historical revisit’, New Indian Express, 22 August 2020,
https://www.newindianexpress.com/opinions/2020/aug/22/vatapi-ganapati-an-art-historical-revisit-
2186817.html
The man responsible for this, the Pallava general Paranjothi, supposedly went on to become a Shaiva
saint. It should be noted that there is no contemporary evidence supporting this oral legend and it
 may date to the twelfth century, when the canon of Shaiva saints was being developed. However, it
was not unheard of for kings to seize and relocate idols that were politically important to their
rivals. See Davis 1993.
EHD I, 219.
SII XI Part I, 1.
Ibid., As noted in Chapter 1, Vallabharaja was an abbreviation of the title Sri Prithivi-Vallabha-
Maharajadhiraja, the full Chalukya imperial title. The fact that their inveterate enemies, the
Pallavas, used it to refer to them is a sign of just how widely the title was identified with the
Chalukyas.
Ali 2006, 55.
EHD I, 219.
Dikshit 1980, 116.
Ibid.
Adiga 2012.
Kumkum Roy, The Power of Gender and the Gender of Power: Explorations in Early Indian History
(Oxford University Press, 2010), 212.
Ali 2006, 51.
This book hereafter uses these terms interchangeably; the term ‘harem’ has Orientalist connotations
today and tends to strip these women of their agency.
Arthashastra 5.3.3, 5.3.7, in Patrick Olivelle, King, Governance, and Law in Ancient India:
Kautilya’s Arthaśāstra (Oxford University Press, 2013), 261.
Ali 2006, 52.
Ibid., 54.
Malini Adiga, ‘Dharmasastras, the Dravidian Kinship System and Female Inheritance in Karnataka in
the Early Medieval Period (400 A.D. to 1300 A.D.)’, Proceedings of the Indian History Congress
64 (2003).
Dikshit 1980, 121.
This is based on the assumption that the ‘Ganga-Mahadevi’ who requested Vikramaditya’s land grant
at Gadval in 674 was indeed his ‘favourite queen’, as portrayed in G.R. Rangaswamaiah,
‘Bhuvikrama’s Role in the Chalukya–Pallava Conflict (635–679)’, Proceedings of the Indian
History Congress 35 (1974), http://www.jstor.org/stable/44138760.
Dikshit 1980, 123.
See Ranabir Chakravarti, ‘The Pull Towards the Coast: Politics and Polity in India (c. 600–1300
CE)’, Presidential Address (Monograph), Indian History Congress, 72nd Session (2011); Dilip
K. Chakrabarti, The Geopolitical Orbits of Ancient India: The Geographical Frames of the Ancient
Indian Dynasties (Oxford University Press, 2010).
Dikshit 1980, 122.
Older Chalukya inscriptions also appear in Sanskrit, but it was Vikramaditya who established a
standard prashasti format and added the classic ‘jayatyāvis.kr.tam. vis.n.orvarāham.’ verse to the
beginning.
Pollock 2006, 145–46.
Pollock 2006, 152.
Ibid.
EI XXXII, 175–84.
Ibid.
Alexis Sanderson, ‘The śaiva Age’, in Genesis and Development of Early Tantrism, ed. Shingo Einoo
(University of Tokyo, 2009), 266.
Nandi 1973, 85.
Flood 2004, 124.
Ibid.
Thomas J. Hopkins, The Hindu Religious Tradition (Wadsworth Publishing Company, 1971), 98.
Cited in Cummings 2006.
David Chandler, A History of Cambodia (Taylor and Francis, 2018), Kindle Edition. It would seem
that it was easier for Pashupatas to use maritime trade networks to reach Cambodia than it was to
brave the warlike peoples of the Deccan to reach the Malaprabha river valley – at least until the
rise of the Chalukyas made the region much more amenable to Shaiva monastic power.
George M. Moraes, The Kadamba Kula (B.X. Furtado & Sons, 1931), 250.
Nandi 1973, 83.
Ibid., 107.
Sanderson 2009, 267.
Nandi 1973, 104.
Sanderson 2009, 256. This is the Pashupata version of the consecration ritual. We cannot be sure this
is what Vikramaditya’s consecration looked like but it is likely to have been so, considering that
the Pashupatas were already quite influential in northern Deccan.
Ibid., note 595.
Ibid.
Ibid.
These details may be inferred from Deccan hero stones from the period.
Sanderson 2009, 257.
Representations of this silk cap, which somewhat resembles a fez without a tassel, can be seen in
sculptures of Pallava kings at Mahabalipuram.
Michael D. Rabe, ‘The Māmallapuram Praśasti: A Panegyric in Figures’, Artibus Asiae 57, no. 3
(1997): 189. This book follows Doniger (2009) in her conclusion that Rabe’s analysis is the most
likely explanation for the structure of the work.
Ibid., 200.
This story is recounted in the Vana Parva of the Mahabharata.
A reference to Agastya’s two fathers and his slaying of Vatapi can be found in the Sabha and Vana
Parvas of the Mahabharata.
Rabe 1997.
Ibid.
Ibid.
Dikshit 1980, 124.
EI X, 105.
This is heavily indicated by the appearance of Vikramaditya’s horse Chitrakantha in his prashastis.
The horse is most likely meant to be a sign of wealth, status and martial valour, but it could only
have been militarily useful if accompanied by other horses in a cavalry unit.
EHD I, 222.
Romila Thapar, The Penguin History of Early India: From the Origins to AD 1300 (Penguin UK,
2015), 331.
K.A. Nilakanta Sastri, A History of South India from Prehistoric Times to the Fall of Vijayanagar
 (Oxford University Press, 1997), 138. The Gangas would claim quite a record of killing Pallava
kings in battle.
He was already in Tamilakam by 671, but by 674 he had apparently gained a decisive edge over the
Pallavas.
Dikshit 1980, 124.
Sanderson 2009, 258. The purpose of these multiple consecrations was to repeatedly invest the kings
with sacred authority to complement their martial authority, establishing their position at the top of
the social hierarchy.
Dikshit 1980, 124. Emphasis added.
Rabe 1997, 230.
Ibid.
Ibid.
EHD I, 222.
Ibid., 224.
Dikshit 1980, 132.
SII I, 154.
The young Pallava king would claim a tremendous victory (ibid.), and later sources would claim that
he had sacked Vatapi, which Sastri (1997) and Thapar (2015) both take at face value. This is
debatable given that the Chalukyas make no mention of such a dramatic event. After all, the
second sacking of Vatapi within a century would certainly have been worthy of note and have left
some material traces. At best, the Pallavas may have succeeded in raiding Chalukya territory
before being repelled by Vinayaditya and Vijayaditya.
 4
Great Goddess
Winter, 682 CE

By the time Vikramaditya I, that bloodthirsty restorer of the family fortunes,

died in 682 CE,1 the Chalukyas had come a long, long way from their
humble beginnings as local chiefs living in terror of the Kadambas and
Kalachuris. Now, the once mighty Kadamba capital province was merely
one component of a vast pan-Deccan political network dominated by their
rivals, the Chalukyas. This Chalukya ‘circle’ of kings and sub-kings
stretched from the seashores of southern Gujarat across the dry plateaus of
Maharashtra, the thick thorny forests of the Deccan plateau, and the rushing
rivers of Karnataka all the way to the very borders of the Tamil country.
Slowly growing towns and cities interacted with each other in gradually
denser networks. And the largest hub of that network, the sun of this solar
system, was the court of the new Chalukya Vallabha, Vinayaditya, moving
around the Deccan as Harsha had once moved around northern India – at
the heart of an immense military encampment that served as a mobile
capital, always ready to deal with the slightest hint of insubordination. The
Deccan was beginning to grow into the status of a true superpower
comparable to the ancient north Indian empires.
The Chalukyas were certainly the dominant power of the Deccan, but the
land was simply too expansive, turbulent and multi-centric, requiring the
constant mobilization of economic, political and religious resources to
control. Contrary to our notions of vast and powerful centralized empires
that paint swathes of modern maps, in reality, a map of medieval India was
more like a patchwork of petty chiefdoms separated by wide stretches of
sparsely inhabited wilderness, with a few major dynasties exerting various
degrees of control over geopolitically important areas. In particular, the
Chalukya vassal chiefs of the Deccan were well capable of going their own
way or just ignoring Chalukya commands in practice while paying them lip
service in public. By the time Vikramaditya’s aged son Vinayaditya was
finally crowned as Fortune’s Favourite, wars and rebellions were already
erupting to his east and north.2 Other means had to be found to widen and
deepen Chalukya connections to elites beyond the landed aristocrats with
their private armies. The solution lay in expanding royal influence through
the most iconic sites of the medieval period, which to this day are so
integral to India’s self-image: places of pilgrimage.
To ensure that the peoples of the Deccan continued to acknowledge his
family’s supremacy, the roving Chalukya Vallabha visited sacred river
fords3 or tirthas. At these sacred sites where one bank neared the other, the
world of the gods neared the world of mortals, and devotees could cleanse
themselves of their sins. Once there, the Vallabha – often at the urging of a
lady of the antahpuram, an indication once again of the intertangled politics
of the court – would donate lands to a Brahmin or religious institution, who
would then be expected to serve Chalukya interests. Tirthas always
attracted large crowds, especially during seasonal festivals, when local
cultivators and pastoralists might gather to participate in community
ceremonies, watch religious orders parade their regalia, dance, listen to
chanting monks, and bathe in the sacred waters. As their religious, political
and economic weight increased due to royal endowments such as
Vinayaditya’s, many of these pilgrimage centres gradually took on a life of
their own. In 689/90, the Vallabha issued a copperplate inscription ‘from his
victorious camp at Pampa tirtha on the Tungabhadra river’.4 This seventh-
century sacred site of the fierce goddess Pampa would grow into
Hampi/Vijayanagara by the sixteenth century, one of the largest cities on
Earth and the seat of a gunpowder-wielding Deccan empire whose
aristocrats used a modified version of the Chalukya boar crest. We will soon
visit another such tirtha, Ellora, in the eighth century.
From the Chalukya perspective, in the seventh century, their temples in
the sacred sites of the Malaprabha river valley reminded their direct
subjects of Chalukya power and devotion to popular gods. Similarly, having
a Chalukya loyalist at the Pampa tirtha to remind the crowds of their distant
obligation to the Chalukya monarch, irrespective of the actual degree of
Chalukya control over the tirtha, simply made good political sense. This
simple politico-religious calculation, made repeatedly over the centuries by
the Chalukyas and their vassals and rivals, helped create the web of
pilgrimage sites and sacred cities that are so integral to modern Hinduism,
particularly in the Deccan.
Towards the end of the seventh century, with the Pallavas to the south
beaten into a simmering peace, and with territories in modern Karnataka
returning to their fold, the Chalukyas needed to once again secure their grip
on Maharashtra and the northern Deccan. Unfortunately, recruits for their
endless wars were increasingly hard to come by. Their subjects in the
Malaprabha heartland were introduced to a new tax on childless couples5
(which, the scholar Carol Bolon argues may have led to a surge in the
popularity of fertility cults dedicated to the lotus-headed Lajja Gauri, who
still remains popular in the area). But there were still no recruits to be had,
which led to the Chalukya ruler mobilizing military power by arranging one
of the most significant marital alliances of the early medieval period –
which will introduce us to an extraordinary Deccan queen.
Since Mangalesha Chalukya’s raids in the late sixth century, the Kalachuri
family had re-emerged as a major power in the northern Deccan. These later
Kalachuris called themselves Haihayas after a legendary family from the
Mahabharata, and their relationship to those of the sixth century is
somewhat unclear. After campaigning to secure their submission, the new
Chalukya ruler decided not to extend the Chalukya network there as directly
as his grandfather Pulakeshin II had. He decided, instead, to bind his
fortune with the Kalachuris by marrying his grandson, the young prince
Vikramaditya, to two of the family’s princesses.6
Imagine what sights these girls,7 the sisters Loka-Mahadevi and
Trailokya-Mahadevi, saw as their gorgeously bedecked palanquins were
carried from the northern Deccan to the Malaprabha valley. The babble of
Maharashtri Prakrit they had grown up with was replaced by Old Kannada
and the accented Sanskrit of north Karnataka. They saw the Chalukya boar
banner fluttering over the mud walls of Vatapi, and entered the halls of the
palace to the sound of drums and trumpets. What did they see then, as they
were escorted to the mural-decorated Inner City whose corridors the
courtesan-queen Vijaya-Mahadevi had once walked, where they would die
some day, far from the place they had called home? What mix of trepidation
and anticipation did they feel when they were welcomed by a crowd of
royal women with eyes as cold and glittering as their jewels? How did it
feel as they were slowly subsumed into the politics of this strange court in a
strange land? We cannot know, but it will not be long before we hear from
them again.
The Vallabha Vinayaditya, meanwhile, was busy with grand designs.
Apparently mobilizing the military support of his new Kalachuri in-laws, he
sent the crown prince Vijayaditya (the father of his grandson Prince
Vikramaditya, thus the father-in-law of the two Kalachuri princesses) to
lead a raid on north India in the 690s.8 The region had been in utter chaos
after the death of Harsha, raided and attacked by powerful new kingdoms
based in Kashmir, Tibet and Nepal. The mighty Gujarati ruler Siladitya III,
who was on the verge of dominating the Gangetic plains, was defeated by
Vijayaditya.9 The vanquished king had somewhat prematurely adopted
imperial standards depicting the Ganga and Yamuna rivers as a symbol of
dominance over that ancient and prestigious heartland; these were seized10
and brought to the Deccan.
 This stupendous military achievement was the first attested instance of a
Deccan power crossing the Vindhyas in half a millennium11 – it may even
have been the first – but it would not be the last. It would initiate a series of
Deccani expeditions that pre-dated the Maratha campaigns in north India by
over a thousand years. This military upset may have been as shocking in the
eyes of the urbane and the political Indians of the seventh–eighth centuries,
as the Marathas’ defiance of the Mughal emperor Aurangzeb Alamgir was
in the eighteenth century. This Chalukya incursion into north India was the
first sign of a centuries-long pattern of Deccan geopolitical dominance over
the rest of the subcontinent.
The events of the Mahabharata and Ramayana, the earliest kingdoms and
the most sacred pilgrimage centres and tirthas of the subcontinent, were all
based in north India. Look at us, the Chalukyas seemed to say. We are the
captains now. We are making a world of our own, we are remaking the
‘imperial topography of the subcontinent’.12 We have our own tirthas, our
own courts, our own temples, as splendid as any in the world, they said.
The people of the Deccan watched in awe as sculptures depicting the
goddesses Ganga and Yamuna, which had hitherto been used by north
Indian polities to signify dominance over the Gangetic plains, henceforth
adorned the gateways of new Chalukya temples.13 As the Deccan plateau’s
many kingdoms and dynasties competed and innovated, as its monasteries
grew and pilgrimage hubs began to swell, it was now rapidly becoming the
subcontinent’s new political and cultural centre of gravity.14 And at its
forefront stood the Chalukya emperor: the Vallabha, Fortune’s Favourite.
No longer would a simple boar banner suffice: henceforth Chalukya
emperors were preceded by the pali-dhvaja, a ‘compound flagstaff’ or ‘flag
in rows’15 that bound to the Varaha standard the banners of defeated and
vassalized kings, a powerful visual reminder of the vast, diffused political
network they controlled.

Over the next two generations, with no other Deccan power left to
challenge Chalukya military supremacy, tribute and agricultural revenues
poured into the Malaprabha valley, the Chalukya home base. The soil of the
valley was poor and best used for millet farming,16 with other agricultural
produce brought in from other Chalukya territories or traded along well-
established routes.17 At one end, protecting the entrance to the valley, was
the citadel of Vatapi, commanding views for dozens of kilometres from its
fortified heights. Within the valley were the towns of Mahakuta,
Aryapura/Ayyavole (modern-day Aihole) and Kisuvolal (modern-day
Pattadakal), which had been centres of religious significance for centuries,
thanks to their position near rivers or springs.18 They were now increasingly
decorated with temples dedicated to Puranic deities. New money and people
poured into the Chalukya heartland, drawn by the military pre-eminence of
the court, and the valley was further transformed by these forces. This
brings us to another institution that shaped medieval India: the merchant
guild.
Perhaps the most important medieval Indian merchant guilds were the
Ainurruvar, the Five Hundred Swamis (Lords) of the great agrahara at
Aihole, a group of five hundred affluent upper-caste men. They were
chaturvedi (learned in the four Vedas) Brahmins, probably with close ties to
the court, and in the seventh–eighth centuries, they worked together to take
over commercial activity in the Malaprabha river valley.19 It is significant
that it was not the merchant setti castes but the better-connected,
landowning Brahmin castes that found themselves best positioned to exploit
the new economic opportunities emerging as the Malaprabha valley’s cities
and towns grew and the demand for food and luxury items swelled.
Over the decades, the Five Hundred Lords gradually brought under their
control disparate wandering and local merchants and traders, allocating
‘brokerage and monopoly to individuals or groups of traders on certain
items’.20 As the association grew, their moniker soon lost its literality and
became the go-to name for them. The Five Hundred brought together many
locally powerful groups.21 As the guild expanded from the Malaprabha
valley through the fertile lands between the Krishna and Tungabhadra
rivers, ‘small and workable federations’ were established under its general
aegis, sometimes over commercial units spread over as many as four
modern districts.22 Within the overarching ‘umbrella’ that was the Chalukya
political network, more complex economic arrangements than had hitherto
existed in the Deccan now began to emerge.
The Five Hundred and their like were powerful agents of the Deccan’s
transformation, and their activities unfolded alongside another significant
trend. Much has been made of the role of kings in shaping India’s religious
landscape, but just as much credit is due to the networks of wealthy people
connected to the court. As courts and the people associated with them –
aristocrats, poets, merchants, priests, administrators, military men – became
wealthier, a second spurt of temple-building activity23 exploded across the
Malaprabha river valley. By the early eighth century, elite Chalukya
subjects could call upon the talents of many guilds of professional sthapatis
who were drawn there to develop temples dedicated to a wider array of
popular deities. Just as with the projects of the early Chalukya monarchs,
these elites were not paying for these expensive undertakings purely out of
piety. Rather, they competed to display their wealth, sophistication and
proximity to the Chalukya court through temple patronage. The more one
was able to donate, the higher one’s status in the pecking order. Inscriptions
(usually prefaced by flowery praises of the Vallabha) proclaimed how much
they had donated, making it clear to all who could read – or hear the
inscriptions read out – who the great magnates of the court were.
The fortunes that were lavished on these buildings attracted workers from
all over the Deccan. These included a surprising number of people from the
Telugu country,24 who apparently found the Chalukyas of Vatapi to be far
more generous patrons than the Chalukyas of Vengi. As they worked,
climbing up scaffolds to quarry the best-quality stone for temples, these
people whom history would otherwise have forgotten carved their names
into the cliffs, and marked out the number of days they worked so as to be
paid on time. Names like Ayicchasvami, Kottimanchi, Buru, Bithamarga
and Dronamma appear,25 unmistakably Indian but so different from modern
Indian names. The sandstone cliffs of the Malaprabha valley, which stand
silent now, must have buzzed with activity in the day. We can guess that
their lives were far more difficult and full of hard physical labour than those
of their aristocratic employers, but little more. It is a shame that we know so
little about the lives of those not wealthy enough to leave behind a
monument.
What we can see, though, are the great works of art they created on behalf
of the elites who ruled over them.
Temples, and the hierarchies they articulated, were designed to be visible
to the people of the Malaprabha valley, positioned in pleasant environs to
attract a cosmopolitan crowd. Though the earliest temples had been set up
on pre-existing sacred sites, they now began to appear on arterial roads and
highways on the outskirts of cities.26 Here, more land was available and
they would be accessible to city folk, as well as to people from the villages
who grew food for them. They were surrounded by large gardens and
orchards, or near lakes and reservoirs, to draw crowds and impress upon
them the generosity and piety of the temple’s patrons.27 These crowds, in
turn, attracted shopkeepers selling everything from foodstuff to textiles,
creating commercial hubs directly or indirectly regulated by merchants or
by the king, thus becoming integrated into the new economic and political
hierarchies of the Malaprabha valley.
We take all of this – the wealthy patronizing temples, shrines acting as
social and commercial centres – for granted in the practice of modern
Hinduism. But it is worth reiterating that in the medieval Deccan, this was
all overwhelmingly new. To a significant extent, it is through the activities
of the elites who rose to power alongside the Chalukya clan that temples
became as ubiquitous as they are today. But even as their patronage of
temples in the Malaprabha valley proclaimed their loyalty to the Chalukyas,
these lords and ladies were gradually growing in confidence, power and
popularity. It would not be long before they mounted serious challenges to
the Deccan’s premier dynasty.

When Prince Vikramaditya Chalukya, husband of the Kalachuri sister-

wives, was appointed Yuvaraja sometime around 710 CE, it seemed as
though Chalukya power had reached a glorious apogee. His father
Vijayaditya, his grandfather Vinayaditya, and his namesake and great-
grandfather Vikramaditya I had expanded the dynasty’s control and
enforced relationships weighed in favour of the Chalukyas across most of
the Deccan. And the Pallavas had made no serious attempts to renew their
ancestral conflict.
At long last, the Malaprabha river valley, after generations of its people
had been sacrificed to the ravenous flames of empire, was beginning to
recover – indeed flourish. Vikramaditya’s father, the Vallabha Vijayaditya,
returned to his home after a life spent mostly on the battlefield. He then
proceeded to fall head over heels in love with a young courtesan,28 and
spent his last decades relatively at peace. New temple spires were rising
around him, urban centres were growing. Tours of the territory were
undertaken to visit the dozens of in-laws and married-off adult daughters
that made up the courts and antahpurams of the emperor and his circle. It
appears to have been a relaxed period after the chaos of the preceding years.
At least that is how things seemed on the surface. Just as the world
outside the court had many other power centres, such as temples and
merchants, the court itself was a viper pit of politics and intrigue with many
competing centres of power.
The Chalukya imperial formation, with its unfurling economic and
religious networks and hubs, had catapulted many more people into the
leagues of the powerful. This was both their ultimate success and the seed
of their own destruction. New actors – merchants, ascetics, landlords,
aristocrats – were now competing with the Chalukyas for primacy in many
new battlefields: temple and artistic patronage, diplomacy, wealth. This
created fresh political pressures, requiring the royal family to invest in
increasingly lavish displays of public power in order to maintain their
primacy in all these fields. Consider even their grandest royal temples, such
as the towering Lokapaleshvara temple (now called the Galaganatha) at
Pattadakal.29 It is built in the north Indian style, with an imposing
curvilinear spire, rather than the tiered south Indian style. The architecture
is apparently meant to commemorate Vijayaditya’s raid of north India, and
the god’s name – ‘Lord of the Directions’ – was evidently meant to signify
that the Chalukyas dominated territory in all the cardinal directions. It was a
project of such scale that it practically established Pattadakal as the
dynasty’s ‘new religious centre’, or at least as a ‘major temple site’.30 But
why did it have to be built on such a scale and at such expense? Because
this sandstone superstructure, confidently proclaiming the might of the
Chalukyas, is actually meant to conceal uncertainty and fear: the silent rock
hints at a need for public validation and display of imperial power.31
 There might have been less fighting and raiding going on by the early
eighth century, but there were other challenges the court had to face. The
Chalukya political network had grown both wider and deeper over the last
four generations. Take, for example, the Harischandra family of the
Konkan, with their power concentrated near Puri (modern Elephanta near
Mumbai, not Puri in Odisha), once conquered by Pulakeshin II.
Svamichandra, the founder of the line, was established as the hereditary
samanta – vassal king32 – of that region by Vikramaditya I. Two generations
down the line, his grandson Bhogashakti was still ruling it under
Vikramaditya’s grandson Vijayaditya,33 the current emperor. But the term
‘under’ is somewhat misleading: vassals of this rank had the right to
appoint their own candidates to rule over districts and groups of villages,34
which meant that in practice they enjoyed considerable autonomy.
Bhogashakti was essentially a king in his own right, and is known to have
set up a temple to a god named after himself and ordered local merchants to
administer it on his behalf, making a land grant of his own to do so.35 This
situation of de facto independence grew over the generations, despite ties
through marriage and tribute to the imperial family.
However, these samantas were just one rank among a vast hierarchy of
local rulers and minor and major lords, with direct or indirect ties to the
Chalukya court. This included village chieftains, lords of districts,
governors of provinces. And alongside them, of course, were the nouveau
riche, the merchants and bankers and monks and courtesans who might not
have ruled over land themselves, but nevertheless wielded enormous clout
in political affairs. Keeping them all in line required constant alertness and
activity from the Chalukya emperor and his court. Titles, honours,
marriages, appointments, gifts and submissions had to be negotiated and
arranged. Agreements needed to be upheld, challenges dealt with,
opportunities exploited. And the court, paralleling the rest of the imperial
formation, was also made up of talented and ambitious individuals with
hereditary ties to the Chalukya clan. The Punyavallabha family, for
example, served Vikramaditya I and his descendants as maha-sandhi-
vigrahakas,36 Great Controllers of Peace and War (similar to foreign
ministers). They went wherever the Chalukya emperor went in his moving
court/military encampment, personally composing the land grants and
official documents he issued.37 And just as the Harischandras of the Konkan
sat near the top of the hierarchy of semi-autonomous local rulers, the
Punyavallabhas were the most prominent in the hierarchy of court offices
that many powerful families of the Malaprabha heartland of the Chalukyas
vied for.
Supremely influential among these lobbies were the women of the
antahpurams, whom we have previously seen outdone by the courtesan-
queen Vijaya-Mahadevi. As opportunities for advancement, profit, glory,
rivalry and vengeance proliferated, so too did activity in the Inner Cities of
all these powerful men. As has been mentioned, hierarchies in the
antahpuram paralleled those of the court. All these positions flowed from
politicking: the question of who would occupy the position of chief queen
was a decision based on talent, public perception and private negotiations,
not on who happened to marry the king first.38 The wives and daughters of
the rajas, maharajas, samantas, mahasandhivigrahakas, chief priests, senior
generals and so on also occupied ranks to which they had to adhere. These
elite women were expected to attend the ‘life-cycle rites of the imperial
family’, and served as attendants of the seniormost queens.39 Within the
antahpuram, the women’s ‘fertile periods’ were supposed to be
meticulously tracked by female attendants reporting to the king’s officials,
as were their sexual interactions with him.40 As much power and influence
as a queen could wield, it should be remembered that (in theory) she was
little more than the king’s appendage. Only the most important, connected
and qualified women would be allowed to gain influence by becoming the
mothers of the next generation of potential kings and queens. Royal
succession, like so much else in medieval India, was a complex affair.41
Of course, not all women who entered the antahpuram did so with even
this degree of autonomy. References to queens and princesses kidnapped in
war and forced to live in their abductor’s antahpuram are commonplace in
historical texts, and they are unlikely to have been well treated either by the
king or the senior queens. Violent sex – whether consensual or not – was
also not uncommon, being described in the Kamasutra, a manual on love
and relationships meant for a courtly and urbane audience, and probably
known to the elites of the Malaprabha valley.42 Indeed, the Kamasutra goes
so far as to say that ‘Ferocity and roughness are in a man’s nature, it is said;
for woman it is lack of power, suffering and giving up, and a sense of
weakness …’43 Soon after, it provides examples of kings accidentally
killing their wives or dancing girls during sexual intercourse by beating
them for pleasure. Though it condemns that extreme as ‘hurtful, barbaric,
and unworthy of respect’,44 the fact that the practice was brought up in the
first place – given it was meant for such an elite audience – is telling.
Moving beyond aristocratic women, the antahpuram was also home to a
crowd of attendants to provide the former with everything from cosmetic
services to musical accompaniment, from dance performances to
confidential advice. There were also of course cooks, cleaners, launderers,
guards and administrators to run the Inner City and keep potentially
seductive or influential male partners well away from the king’s women.
A medieval Indian king’s visits to his antahpuram must have been fraught
with political tension. The king was expected to visit the Inner City on a
regular basis, where the women would be lined up to meet him. During this
interaction, he would confirm or change their hierarchies through a
ritualized process of gift giving. Queens would gift him ‘garlands, perfumes
and unguents’, and he would return them ‘as forms of grace’.45 Carefully
done up, his body washed and perfumed and adorned with jewellery, he was
then supposed to talk and joke with the assembled women ‘in keeping with
their status and the time they have been in the harem’.46 He would then
repeat the process with ‘remarried women’ (which is to say, queens
abducted in war), as well as the courtesans and dancing girls who resided in
the antahpuram.47
As such, the declaration of Prince Vikramaditya Chalukya as Yuvaraja,
circa 720 CE – coming decades after his marriage to the Kalachuri
princesses – suggests that he was just one of many half-brothers, sons of the
Vallabha Vijayaditya by an array of queens, who were all lobbying for
succession. He and his two sister-wives must have had to navigate their
own personal dynamics, as well as their relations with the other
aforementioned power centres of the Malaprabha valley, to ensure that he
was declared the heir.

Once this was done, both princesses, now wives of a middle-aged

Yuvaraja under an ageing Vallabha, found their positions immeasurably
strengthened. They now had precedence over the wives of other Chalukya
princes, as well as most other women of the court with the exception of
their mothers-in-law, the Vallabha’s wives. Up to this point, the royal
women of the Deccan (with the exception of Vijaya-Mahadevi) had played
a relatively muted role in the patronage of religious and architectural
projects, often undertaking them with the financial or political support of
their male relatives.48 No longer.
When Trailokya-Mahadevi, the younger of Vikramaditya’s two sister-
wives, gave birth to his son and heir apparent in the early 720s, she
embarked on a project of astonishing scale. She ordered the construction of
a Shiva temple at Pattadakal, so ambitious in scale that it sucked Chalukya-
employed artisans away from their other building projects in the
Malaprabha valley for no less than fourteen years.49 Pattadakal was the
coronation site of Chalukya monarchs, one of the central hubs of their
power in the Malaprabha valley, and was also emerging as a major religious
centre thanks to the Vallabha Vijayaditya’s previous temple projects. At a
time when elites were increasingly using temples to establish their position
in the valley’s pecking order, for a princess to sponsor such a major project,
evidently with the Vallabha’s support, indicates an extraordinary degree of
self-assurance; evidently by this point this foreign princess did not lack for
connections and influence of her own in the valley.
Trailokya-Mahadevi, unfortunately, did not live to see the completion of
her temple. When her son was still in his teens, she died suddenly of
unknown causes, and all the resources and workers gathered for her temple
were transferred to her father-in-law’s projects, though the old man was
apparently touched by her vision and drive.50 But now her elder sister,
Loka-Mahadevi, rose to the apex of the Chalukya antahpuram, taking her
younger sister’s son and her husband under her wing. It is interesting to
wonder what the relationships between these people were like: did Loka-
Mahadevi spoil the young prince as the son she never had, a memory of a
beloved sister, or emotionally torture the child for his status as a rival to her
power? We will never know.
Loka-Mahadevi was perhaps one of the most powerful of all medieval
Indian queens.51 Soon after her husband Vikramaditya’s appointment as
crown prince, despite being lesser in rank than her sister Trailokya-
Mahadevi, the mother of the heir, there were already hints of her
independence and influence. According to her land grants, she administered
two regions in her own right, successfully built a large temple that has not
survived to today, and assigned the revenue of many villages to it.52 If so
much evidence of her deeds survives, it stands to reason that her career
must have involved even more activity than we are able to see today. Even
her sister had never had such influence. For a royal lady to do all this, and
do it publicly, was exceptional; the only other one to have done so53 was
Queen Vijaya-Mahadevi. Unlike that former courtesan who had risen to
power in a moment of chaos, Loka-Mahadevi flaunted her authority in full
view of the Chalukya imperial network, apparently with the full backing of
the antahpuram, her husband the crown prince, and her father-in-law the
Vallabha himself.
By distributing land in her own right instead of asking for the emperor to
do so ‘at her request’ as most other royal ladies did,54 Loka-Mahadevi was
making a bold statement about her importance to the Chalukyas. She
seemed almost to imply that if her father-in-law and husband were Vishnu,
the ideal king, then she was the female equivalent: the female energy from
which Vishnu’s power sprung – in a sense the fertile, generative power
behind the Chalukya dynasty.55 She was Fortune, Loka-Mahadevi, the Great
Goddess of Earth, and Vikramaditya was her favourite.
There can be little question that Loka-Mahadevi’s importance was
recognized by her contemporaries. Later Chalukya inscriptions would
declare that her husband Vikramaditya obtained ‘great energy’ (mahotsaha)
from his union with her56 – a reflection of the idealized relationships of
gods and their shakti, their female counterparts, but also a striking centring
of a queen in otherwise king-centric prashastis.
As Loka-Mahadevi took the Chalukya antahpuram and her viceregal
territories in hand over the decades, her husband, Crown Prince
Vikramaditya, proved himself a talented and well-liked administrator and
diplomat, displaying little military inclination. He and his father travelled
far and wide to maintain the Chalukya imperial network in a reasonably
healthy state, constantly negotiating and solidifying ties with local elites.
They could be surprisingly pragmatic about this, granting considerable local
autonomy in return for an acceptance of their sovereignty and adherence to
their overall geopolitical goals.
One particularly famous inscription from Vikramaditya’s early career
features a ‘constitution’ he established in a town called Porigere (modern
Lakshmeshwar, Karnataka), in agreement with its foremost citizens and
representatives of various social groups.57 According to this deal, royal
officers would be responsible for enforcing the Vallabha’s proclamations
and the terms of land grants in Porigere. However, the income from taxes
on various types of households and guilds, as well as fines for theft and
other delinquencies, were to be paid not to royal officers, but to the
merchant guild. The collection was to be made in the month of Karttika,58
the start of the campaigning season – a hint that this revenue right was
granted to the merchants in return for assistance with maintaining the
Chalukya armies. Clearly, the Five Hundred and other such guilds were
already more influential in some areas than the imperial court, and the latter
understood how to turn these new politico-economic equations to their
advantage.
Vikramaditya’s father, the Vallabha Vijayaditya, would reign for nearly
forty years – quite exceptional for the times, though for the last decade or so
of this period it was Vikramaditya and Loka-Mahadevi who ran the affairs
of the court, especially after Vikramaditya’s installation as crown prince c.
720. Through all these years, the Chalukyas had remained relatively at
peace, focusing on political competition, temple construction and the like.
But as the Vallabha’s health continued to deteriorate, Vikramaditya and
Loka-Mahadevi seem to have realized they needed a more grandiose
display of their power. Perhaps other ambitious claimants were raising their
heads, or perhaps the couple wished to establish they were as capable in the
arts of war as they were in the arts of peace. After all, every generation of
Chalukya kings before them had been extraordinarily militarily successful,
maintaining and extending their authority through war. Vikramaditya, living
under their shadow, had to publicly establish himself as worthy of ruling in
the eyes of the immense pan-Deccan network that watched the court
attentively for any sign of faltering or weakness.
And not just Deccanis, either. In 708, a distant superpower, the Umayyad
Caliphate, headquartered in Syria, had begun a campaign of conquest and
raiding in India that would continue into the 730s and leave much of
western and central India reeling.59 There must have been much fear and
uncertainty in the air.
The easiest way for the royal couple to consolidate their position was
probably to summon the allies and vassals they had made over their career
and unleash them on a common enemy. But, apparently, Vikramaditya did
not feel secure enough about his political position or military ability to
tackle the terrible threat that was emerging to his north. Instead, he searched
around for a softer target. And c. 731, just such a target was available to
him to his south.60

Over the past three decades, the Pallavas had managed to substantially
recover from the devastation of the Chalukya king Vikramaditya I’s
campaigns of the 670s. The new Pallava king, Narasimha-Varman II, was
an artistically inclined type, like his great-grandfather Mahendra, the author
of the Farce of the Drunkard’s Games. Part of this may have been due to the
influence of his court poet and acharya (teacher), Dandin (Stick Wielder).
 Dandin is an excellent example of the cosmopolitan and multi-
generational character of medieval Indian courts, and his life offers us a
glimpse of the surprisingly human stories of some of these elites. His great-
grandfather had left western Maharashtra to seek easier opportunities down
south at the Pallava court.61 The Pallavas were, like the Chalukyas,
perpetually on the hunt for educated Brahmins to build up social, human
and religious capital. Dandin’s grandfather and father held comfortable
courtly jobs, and may have been involved in the projects of Mahendra and
Narasimha I. All that came to a disastrous end with the Chalukya
Vikramaditya I’s expedition of revenge, which left Kanchi sacked, the
Pallava court in disarray, and the countryside in flames. The orphaned
young Dandin had wandered about for twelve years62 until he finally
regained the hereditary position of court poet. But his time on the streets
had changed him, tempering his erudition with a deep cynicism of the world
in general and of kings in particular.
Dandin was and is one of the most remarkable figures of the Sanskrit
literary canon, a virtuoso whose words were described as a ‘treasure of pure
nectar, their expanse a jewelled mirror for the sport of Sarasvati’,63 and
references to him are ‘widespread, from ninth-century Kashmir to tenth-
century Sri Lanka and thirteenth-century Tibet’.64 He was a stylist, as well
as a critic and a theorist of style,65 authoring a manual called the
Kavyadarsha (Looking-Glass of Poetry) – a book that would be of great
significance to the Deccan centuries later, as we shall see.
One of Dandin’s lost works is an epic poem simultaneously narrating the
stories of the Ramayana and Mahabharata by relying on the remarkable
capabilities of Sanskrit grammar and word conjugations.66 In another, one of
the greatest surviving works of Sanskrit prose, he narrates the stories of ten
adventurous princes seeking to restore a lost kingdom. This was perhaps
designed to impart to his young ward, the Pallava king, lessons on kingship,
the morality of dharma, and perhaps also the things he had seen as a
wandering orphan.67 It reveals a great deal about the ethical world and
political calculations of the elites of early medieval India.
The heroes of the work, the Dasakumaracharitam (Tale of the Ten
Princes) freely deploy seduction, riot-mongering, thievery and murder; they
impersonate sages, bankrupt rivals and overthrow kings. But, arguably, all
the kings they overthrow are unrighteous, do not follow their dharma, and
are most often unwary. The villains of the Dasakumaracharitam are caught
up in personal vendettas, have made themselves unpopular with their
subjects, or are slaves to their passions, and all these flaws allow the heroic
princes the opportunity to overthrow them, usually after a careful
calculation of whether their actions will help uphold dharma or not. One
prince, for example, holds himself back from sleeping with a queen, another
man’s wife, for fear of hurting dharma. He then calculates that ‘the
compilers of the scriptures permit this if both artha (material success) and
kama (sexual pleasure) are also attained at the same time68… That should
neutralise any sin, and may also reward me with some fraction of
dharma.’69 He then spends the night in illicit pleasure, leaving the queen
covered in love bites and scratches, savouring her moans and the faint line
of sweat on her brow as they make love.70
The princes’ calculations, and the strategies they devise to achieve their
goals, are warnings to an unrighteous and unwary king. They also highlight
the overriding importance of dharma as a moral code that Indian kings were
expected to follow, or appear to follow. While playing with these complex
themes in the Charitam, Dandin displays his literary gifts: one prince, his
lips bruised by love bites, narrates his entire tale without employing sounds
that require the usage of one’s lips: pa, pha, ba, bha, ma).
The new Pallava monarch would follow these lessons carefully. He
recognized there was little that could be done to undermine Chalukya
dominance, and there were no further attempts at revenge and wars in the
Deccan. Instead, the Pallavas looked to trade, across the seas, to a world
utterly transformed.
In West Asia, the Prophet Muhammad had died in 632 CE. In the
aftermath, the tribes of the Arabian peninsula, now united under the banner
of Islam, had embarked on a programme of conquest hitherto unrivalled in
human history. ‘Alexander was upstaged, Caesar overshadowed.’71 The
remarkably mobile and capable Arab armies, though they only numbered in
the thousands, managed to crush the Romans and Persians, the superpowers
of West Asia, who were exhausted and demoralized after decades of war.
They quickly secured Iraq, Syria, Egypt and some of the southern
Mediterranean coastline. They were driven by a combination of religious
conviction and tribal affinity, as well as the more usual medieval thirst for
plunder and booty.72
By 644, the Arabs were settling into garrison cities to administer this vast
empire.73 In 656, the year after Vikramaditya I ended the Chalukya civil war
and declared himself Fortune’s Favourite, the Arabs fought their first civil
war. This eventually culminated in
the foundation of the Syria-based Umayyad Caliphate by 661. Soon after,
the entire southern Mediterranean coast fell to Muslim armies; Spain was
conquered and used as a base to raid France; Muslim armies began to wrest
control of the wealthy oases of the ‘Silk Road’ trade routes from the Turkic
nomads of Central Asia. Even the mighty Chinese were forced to send
expeditions against the Arabs.
In 708, the armies of the Umayyad Caliphate were unleashed on the
Indian subcontinent. An Arab governor of the Caliphate conquered Sind on
India’s western periphery. Major trading cities were completely sacked, and
the governor sent back to the caliph a treasury of 60 million dirhams –
equal to billions of dollars today.74 The Umayyad capital city of Damascus
rapidly grew into one of the greatest urban centres of the world as traders,
adventurers and scholars flocked to it from across the Mediterranean.
 North India was, meanwhile, also being raided by the newly founded
Tibetan empire, which was also causing no end of headaches to the
neighbouring Chinese Tang dynasty. The Arab conquest of the entirety of
Persia – and, soon after, Sind – must have sent economic shockwaves
through north India, which had once traded closely with Persia. It was also
almost certainly a cause for great consternation in the Deccan, which had a
keen interest in trade in the western Indian Ocean. Deep down south, in
Tamilakam, the Pallavas also watched with interest. They were, of course,
separated from all this turmoil by thousands of kilometres as well as many
kingdoms, but saw an opportunity for their own advancement in it all.
Embassies were sent by the Pallavas to the imperial Chinese court bearing
gifts of leopard fur and a colourful talking parakeet,75 applying for
‘permission’ to attack the Arabs and a request that the Chinese emperor
provide a name for this army.76 The delighted emperor, unaware that the
Pallava kingdom had neither the ability nor the inclination for such
conflicts, ‘praised it greatly’ and named the force ‘the Army Which
Cherished Virtue’.77 Soon after, a Chinese ambassador was sent to confer
upon Narasimha-Varman II the pompous title of ‘King of the Kingdom of
South India’, which the Pallava, who had already outdone all his
contemporaries with a total of 250 known titles,78 happily accepted along
with his real goal: luxury goods and hefty trade concessions from the
delighted Chinese.
Now, ‘maritime trade flourished’79 from the great port of
Mahamallapuram, apparently completely recovered from Vikramaditya I’s
attack in the 670s. Trading fleets traversed the Bay of Bengal into Southeast
Asia, where powerful new kingdoms were emerging. Southeast Asian elites,
like the Chalukyas had a century ago, used the ritual prestige that came with
participation in the Sanskrit Cosmopolis to elevate themselves above their
subjects. This allowed them to undertake wide-ranging reorganizations of
societies and economies in their own interests. Southeast Asia participated
in the flourishing Sanskrit Cosmopolis, deriving prestige and power from
re-adapting its forms and norms to local contexts just as south Indians did,80
while retaining their indigenous traditions to a significant extent.
 In the 720s, while north India burned and the Chinese lost sleep over the
marauding Tibetans and Arabs, south Indian and Southeast Asian courts
prospered and grew wealthy. The Chalukyas and Pallavas had not
completely stopped their competition, except now it unfolded in the field of
temple building as opposed to the battlefield. The famous ‘Shore temple’ at
Mahamallapuram was constructed at this time. A grand new temple was
also dedicated to Shiva in Kanchi, where the chief Pallava queen’s
donations were commemorated on walls.81 All these developments may
have been keenly observed by her contemporary, Loka-Mahadevi.82
When the Pallava king Narasimha-Varman II died, the Chalukya crown
prince and princess, Vikramaditya and Loka-Mahadevi, seized their
opportunity before his successor Parameshvara-Varman II could consolidate
his power. The elderly couple had, by 731, essentially controlled the
Chalukya kingdom for over a decade, with the Vallabha Vijayaditya almost
certainly ailing83 (he would die in 733). Attacking the Pallavas and seizing
their wealth would have served to reinforce the couple’s domestic position
by establishing Prince Vikramaditya’s military credentials after a long
period of peace,84 and also help finance more temple building, thus
signalling to other vassals the religious and martial primacy of the
Chalukyas. It was, apparently, the perfect strategy.
And so, over forty years of relative peace came to a close. Vikramaditya
and his commanders brought the boar banner to Tamilakam, leading a
combined army with the Pallavas’ other nemeses, the Gangas,85 as well as
contingents of troops from the other lords of the Chalukya imperial
formation.86 The young Pallava king was defeated, humiliated and forced to
pay a massive ransom to the much older Chalukya crown prince to keep
him from marching on Kanchi.87 Crown Prince Vikramaditya then returned
to the Malaprabha valley in triumph. We can imagine him striding into the
palace at the head of his elite troops, bowing to his frail old father and
laying the symbols of the Pallava royalty – banners, crowns, jewels and
musical instruments – at his feet.
The humiliated and infuriated Pallava king then attacked the Gangas,
hoping to catch them unawares and recoup his losses. This move was a
disaster: he was killed on the battlefield, throwing the Pallava kingdom into
crisis as factions at court jostled over which royal child should be appointed
the next ruler.88
From Prince Vikramaditya’s perspective, things were going marvellously.
Dharma was a very important part of the medieval Indian worldview, as
alluded to in the ravishing tales of the Pallava court-poet Dandin. Now
Vikramaditya declared that he had defended his family’s honour and upheld
the royal dharma by defeating their ancestral enemy. Further improving his
position was the fact that his young Pallava rival, who could have been a
thorn in the Chalukyas’ side for decades, had disposed of himself through a
rash attack. Vikramaditya’s expedition to Kanchi was thus a resounding
success, establishing his credentials for the throne beyond any dispute.
And so, when his father finally died in 733 CE, he came to the throne as
Fortune’s Favourite, Earth’s Beloved, Vikramaditya II – apparently the most
powerful Chalukya monarch yet. But external events were moving faster
than he could have ever expected. The world was changing, and
Vikramaditya’s priorities would soon be called into question.

In the early eighth century, as we have seen, the Umayyad Caliph decided
to expand into the vast, wealthy lands of India, potentially securing rich
new provinces to fund the increasingly desperate internal and external wars
he was entangled in. Over the next few years, Umayyad commanders
exploited the political chaos of northern India to the fullest, freely allying
with any power that would enable them to enrich themselves. In the 720s,
the ambitious Shiva-worshipping kings of Kashmir helped89 these Tajikas,90
as the Indians called them, burn and sack the Kangra valley.91 Indian
seafarers and pirates also helped them attack the wealthy cities of Gujarat
and Rajasthan, sacking Vallamandala (modern-day Barmer) and Marumala
(modern-day Jaisalmer). Kaccha (modern-day Kacch), ‘al-Mandal’ near
modern-day Ahmedabad, Saurashtra and Bharuch were conquered
outright,92 and even the outskirts of the ancient fortified city of Ujjain in
Malwa were raided for loot and captives.
As the Tajikas went from success to success, many more equally
ambitious and unscrupulous Indians saw an opportunity to profit from the
loot and slaughter. Thousands of innocents were enslaved and shipped back
to Iraq and Syria.93 In a few years, the Umayyads are estimated to have
secured booty worth as much as 250 million dirhams – many, many billions
of dollars today.94
But almost immediately after the initial devastation, the Caliph’s new
Indian ‘subjects’, who now understood how the Tajikas fought, struck
back.95 Gujarat, Rajasthan and even the Umayyad’s main foothold in the
subcontinent, Sind, rose up in revolt. So bloody and costly was the uprising
that Umayyad troops outright refused to fight any more in the vast and
hostile subcontinent, and ‘insisted on leaving the province, refusing to ever
go back’.96 And yet, somehow, convinced that victory was only a matter of
time and throwing men at the problem, the Caliphate persisted.97
In 731, when Vikramaditya II was attempting to establish his credentials
by attacking Kanchi, Umayyad reinforcements reached Sind. Despite a
series of setbacks at the hands of ambitious new Indian princelings, they
managed to claw back parts of Gujarat, and then decided to attack rich new
territories that had hitherto never seen Tajikas before. As they began
moving deeper south along the coast of Gujarat, they began to approach the
domains of the Chalukyas, or, to be specific, the Chalukyas of Lata, who
had been settled there generations ago by Vikramaditya I.
These Chalukyas of Lata had likely been observing Tajika tactics and
strategy for the previous decade. They had carefully learned from the
defeats of their northern neighbours. While Crown Prince Vikramaditya
was busy bullying the Pallavas, the Chalukyas of Lata had been left
basically unassisted98 to deal with this terrible threat. The Tajikas were
coming for them, ‘reportedly with the intention of subduing all the southern
kingdoms of India’.99
And the kingdoms were ready and waiting.
The Lata Chalukya force was commanded by one Pulakeshi-Raja
(perhaps named after his grandfather Pulakeshin II). This man was a distant
grand-uncle100 of Vikramaditya II. In 737,101 Tajika forces entered his
territory, intending to sack his capital city, Navasarika (modern-day Navsari
in Gujarat), a trading port on the west coast. Pulakeshi knew roughly which
route the Tajikas would take, whereas they may have been in the dark about
his strategy.
What exactly this strategy was is open to debate. The only primary source
from this encounter is a commemorative land grant that Pulakeshi-Raja
made soon after. It seems to have been a long and hard-fought battle.
Pulakeshi-Raja praises the Tajikas as those who ‘had not previously been
vanquished even by numerous eminent chiefs among hosts of kings’;102 says
that ‘they were great warriors and had their sharp swords reddened by the
blood flowing from the torn loins and trunks of hostile elephants’;103 and
describes a great cloud of dust rising towards the skies as the Arab horses
galloped. But after hours of brutal slaughter, the legendary Deccani infantry
finally bested their enemies, their armour ‘reddened by the streams of blood
gushing from intestines spilling out of bellies impaled by spear-heads’.104
The Chalukyas – or, to be precise, a Chalukya ‘vassal’ who was also a
Chalukya – had defeated one of the mightiest military forces of the day. The
Caliphate had been dealt a blow in India from which it would never
recover.105 From the ashes of once-sacked cities in Gujarat and Rajasthan
now rose new dynasties and warlords, determined to drive the marauding
Tajikas out once and for all.106 Henceforth the Arabs107 would leave the
subcontinent in peace. Its treasures were simply not worth the trouble.

Pulakeshi-Raja’s astonishing victory over the Tajikas was a critical moment

in South Asian, if not global, history. It was one among a series of serious
setbacks suffered by the Umayyad Caliphate across the world in the 730s,
including defeats at the hands of the Turkic tribes in Persia/Central Asia and
the Franks in France in 732. Combined with a series of domestic revolts and
a pyrrhic war in the Caucasus, the dynasty would soon collapse108 and be
replaced by a radically different Muslim empire, the Abbasid Caliphate.
However, Pulakeshi-Raja’s victory, while it saved the Deccan from
devastation, must have rocked Vikramaditya II’s position to its foundations.
Now that the Tajika threat, which had terrified much of South Asia for the
last decade, had been unexpectedly dealt with, the emperor could hardly
claim glory for this victory. It certainly seems that Pulakeshi acted on his
own initiative, though some scholars have claimed, based on fragmentary
evidence, that Vikramaditya II had ordered other vassals to go to his aid.109
In a matter of days, all the military prestige Vikramaditya II had obtained
through his raid on the Pallavas before his coronation had been completely
eclipsed. It seemed that a vassal of his could better perform the dharma of
protecting his dominions and subjects than the emperor himself!
In an attempt to conciliate the man and keep him within the Chalukya
fold, Vikramaditya II had to grant Pulakeshi four very significant titles.
These were Dakshinapatha-sadhara (Pillar of the Deccan), Challuki-kula-
alamkara (Ornament of the Chalukya Family, a not-so-subtle reminder of
their family connection), Anivartaka-anivartayitri (Repeller of the
Unrepellable), and, tellingly,
Prithivi-Vallabha – the Chalukya imperial title itself.110
Pulakeshi, however, had had a taste of glory now. Once again, we see the
elements of medieval Indian power coming together. He had inscriptions
issued glorifying his own branch of the family tree, post facto granting
himself, his elder brother and father imperial titles, and announced that the
Goddess of Royal Fortune had chosen him as ‘her lord’.111 Next, just as the
Vatapi Chalukyas had brought Brahmins from prestigious, foreign north
India to legitimize their rule in Karnataka, Pulakeshi-Raja now imported
Brahmins from prestigious, foreign Karnataka to validate his rule in
Gujarat.112 And like the earlier Chalukya kings, he soon set out to expand
his kingdom, chasing out the retreating Arabs and attempting to fill the
power vacuum they left in Gujarat. Perhaps an attempt on the Chalukya
throne in Vatapi itself was in the works.
The Chalukya emperor could not leave this challenge unanswered, though
he could do nothing to directly attack his grand-uncle who, it should be
noted, still referred to him with the utmost (if perhaps sarcastic) respect in
public (‘the illustrious Vallabha who is fond of heroism’).113 Instead,
Vikramaditya II adopted a three-pronged strategy to shore up the influence
of the Vatapi Chalukyas. First, he nurtured another ambitious vassal in the
northern Deccan to nibble away at Pulakeshi’s heels.114 The best candidate
for this was Pulakeshi’s neighbour and grand-nephew from another branch
of the family, who may have previously joined in or supported his
campaigns against the Arabs. This young man was the leader of a clan
called the Rashtrakutas, whose core territories were in the interior of
Maharashtra around modern Elichpur,115 near the Burhanpur gap in the
Vindhya hills. He ruled an area of great strategic importance, commanding
one of the main routes of the great Dakshinapatha, the Southern Way
connecting the trade of north and south India. And his rather unique name,
Danti-Durga, ‘Elephant Fort’, was an almost prophetic allusion to the future
that lay in store for him.
In the early 740s, Danti-Durga, unlike his grand-uncle Pulakeshi, made
his loyalty and subservience to the Vallabha clear as day, claiming the rank
of maha-samanta-adhipati, ‘High Lord Among Tributaries’.116 As his
military and political career slowly advanced under Vikramaditya II’s
mentorship and overlordship, his loyalty would be rewarded with the titles
of Khadgavaloka (Sword sight) and, surprisingly, Sri-Prithivi-Vallabha. The
award of this supremely significant title to two of the Chalukyas’
northernmost vassals, so distant from Vikramaditya II’s base in Vatapi, hints
at just how desperate the court was to keep the region under its control.
Second, with Lata already pulling out of the Chalukya orbit, Vikramaditya
II led a second attack on the Pallavas, to at least shore up his position with
his southern vassals.117 It is possible that Danti-Durga participated in this
campaign,118 perhaps meeting Vikramaditya’s heir and promising the old
Vallabha that he would be the Chalukyas’ most reliable vassal in their
northern territories.
It was the perfect time for an attack. The factions at the Pallava court had
selected a child from a collateral branch as the next king, and civil war had
broken out almost immediately.119 Soon after the wars resolved in favour of
the child king, now a teenager, Vikramaditya decided to invade and
establish his dominance, offering us a glimpse of the symbolic texture of
medieval Indian wars and the strange ‘chivalry’ that kings sometimes
showed their foes.
After swatting aside the Pallava forces, Vikramaditya II seized the mighty
musical instruments which were used to command their troops and drive
them into martial frenzies: the drum called ‘Roar of the Sea’, and the
massive trumpet named ‘Harsh Sounding’. Also captured were many
‘renowned’ war elephants and ‘a heap of rubies’.120 He then advanced
ominously to Kanchi, where the populace was, no doubt, shivering at the
thought of another Vikramaditya Chalukya sacking it, as if the first had not
been bad enough. Vikramaditya II, however, had something else to prove.
What could be more glorious than sacking? Not sacking. Once in Kanchi,
the Chalukya emperor confounded all expectations by donating ‘heaps of
gold’ to the great temple of Narasimha-Varman II, the Pallava king who had
sent embassies to the Chinese. Having ‘inspected’ the riches belonging to
the god, he returned them.121 A half-Sanskrit and half-Old Kannada
inscription122 was made on the walls of this Pallava royal shrine, composed
by Vikramaditya’s Great Minister for Peace and War, Anivarita
Punyavallabha. In it, the Chalukya emperor demanded that none destroy
this testament to his charity,123 a commandment that seems to have been
obeyed by the stunned populace of this city, given that it survives to this
day.
What Vikramaditya was attempting to show his vassals with this
generosity was that he had so much wealth to throw around that he did not
need the petty wealth of campaigns of conquest; furthermore, he had
indulged in the campaign not out of any need for money, but rather to
restore the ‘splendour [of] the previous kings born of his race’.124 He left his
ancestral enemy alive, their royal shrine untouched, because keeping them
around as a testimony to his power, benevolence and adherence to dharma
was far more useful. He set out to portray himself as the ‘righteous’
conqueror, loyal to the legacy of his fathers, who restored to his foes after
taking from them the most priceless treasure of any medieval Indian king:
glory. For now, at least, this decisive display of Chalukya power and
prestige was enough to stave off any immediate moves by their vassals.
Queen Loka-Mahadevi may also have been present at this momentous
occasion; it was not uncommon for queens, even elderly ones, to
accompany the king on his campaigns. If so, she would have seen in this
temple an inscription likening one of the Pallava queens to Parvati,
commemorating the queen’s construction of a tiny sub-shrine in the
complex. This was a pitiful achievement compared to those of the Chalukya
queen, for Loka-Mahadevi was responsible for a crucial pillar that upheld
the trembling edifice of Vikramaditya II’s power: a grand temple in the
Malaprabha river valley commissioned by her a decade before and which
remains, to this day, the most powerful testament to the Chalukya dynasty
and to the brilliance of its queens.
As the Malaprabha meanders through its valley, gently flowing towards the
Bay of Bengal, it temporarily flows directly to the north, as if pointing to
the sacred Ganga, to the Himalayas125 where the great god Shiva lived. At
this sacred site is the town of Pattadakal, the ‘Stone of Anointing’, the
coronation site of the Chalukya monarchs, where we have previously seen
the temples of the Chalukya emperor Vijayaditya and his daughter-in-law
Trailokya-Mahadevi.
At this most holy place, in the early 730s,126 Queen Loka-Mahadevi
summoned127 a senior architect by the name of Gundan, master (acharya) of
a guild of architects called Sarva-Siddhi (Universal Success). At the time,
the Sarva-Siddhis seem to have been at the centre of a major scandal which
was the talk of the entire Malaprabha valley.128 Some very influential people
were calling for the entire guild to be excommunicated from their caste,
potentially rendering them unemployable129 overnight.
Precisely what the Sarva-Siddhis had done to cause such a scandal is not
clear. Legal texts from the period hint at some possibilities, such as
embezzlement.130 Either way, Gundan was probably in a state of high
tension as the queen explained to him that the Sarva-Siddhis were in a great
deal of trouble, and the only way to save their business was through her
intervention.131 That Loka-Mahadevi was able to promise this was a sign of
the extraordinary influence she wielded in her own right at court. Perhaps
this is how Gundan came to be commissioned to build the grandest and
largest132 of all Chalukya temples: a temple named not for a king or a place,
but for a queen.133 The Lokeshvara temple – for Queen Loka-Mahadevi.
This great project was commissioned soon after Vikramaditya II was
declared Fortune’s Favourite in 733, and construction would continue for
the better part of a decade,134 ending soon after Pulakeshi-Raja’s victory
over the Tajikas and the Vallabha’s attacks on Kanchi. Built to the southeast
of the temple of the queen’s dead sister Trailokya-Mahadevi, the
Lokeshvara temple employed so many craftsmen that progress on most
other large royal temples came to a halt: yet another indication of its
importance to the Chalukyas and especially to the queen.135 On the walls of
her new temple, we can still see tantalizing hints of her power and
influence. She announced she was confirming the privileges that her
deceased father-in-law, the Vallabha Vijayaditya, had bestowed on a guild
of local singers136 and assigned to the temple the revenues of a district of
fifty villages.137 Most importantly, she asserted that Gundan, by building it,
had averted the excommunication of the Sarva-Siddhi guild,138 conferring
upon him the ‘fillet of honour’ called mume-perjerepu and the title of
Tribhuvanacharya, ‘Master of the Three Worlds’, and ‘the (most eminent)
Sutradhari [chief architect] of the southern country’.139
As we can see, this medieval Indian queen realigned economic and social
networks to make her Lokeshvara temple in Pattadakal, now called the
Virupaksha temple and a UNESCO World Heritage Site. She created a new
engine of religious and social activity, as any medieval Indian king would
have. She handed out titles that transformed the social prominence of the
Sarva-Siddhis, raising them up to the highest ranks of courtly society.
Why did she do all this? On an inscription on the eastern gateway, issued
after the temple’s completion, she calls it the ‘temple of Lokeshvara of
Lokamahadevi, (the queen) of Vikramaditya, who three times conquered
Kanchi’.140 (We have seen two of these expeditions so far, with the third to
follow in the next chapter.) This suggests that it was meant to inflate her
husband’s martial credentials, but the temple was actually commissioned
before the second and third expeditions had even taken place, at a time
when Loka-Mahadevi and Vikramaditya had just solidified their dominant
political position through a successful campaign to Kanchi followed by
their coronation. And though Loka-Mahadevi was ostensibly building her
temple to support Chalukya power and signal the authority of the Vallabha
and his successful performance of his royal dharma, it was an authority in
which her own rule, power and performance of dharma were also
intertwined. This structure is a temple to Lokeshvara, Loka’s Lord – thus to
both Vikramaditya and Shiva.141 But the Lokeshvara is also an
unprecedented association of a queen’s identity with a temple she made
primarily for herself. In contrast, her younger sister’s construction of a
grand temple – a project that came to an abrupt end with her death – may
have been intended to commemorate the birth of the heir and to support her
position as a royal mother; Loka-Mahadevi’s was unquestionably about her
position as Vikramaditya’s co-ruler.

Visitors to the Lokeshvara temple paddled their boats along the

Malaprabha to alight at ghats of mud which lined the waterfront at
Pattadakal, and made their way through bathing devotees and wandering
ascetics142 as they headed to this grand edifice. Stepping through the ornate
eastern gateway, they would have immediately come across an elevated
mandapa of warm sandstone which shrined an image of Shiva’s bull, Nandi,
carved out of cool black stone. This mandapa was decorated with niches
featuring sculptures of beautiful women, representations of the Mother
Goddess143 and perhaps modelled after powerful members of the queen’s
own retinue. On its pillars are carved sexually intertwined couples,
auspicious symbols of creation and generation.
Today, the carvings on the walls of the Lokeshvara/Virupaksha temple at
Pattadakal reveal a great deal about the politics of the queen who
commissioned it. Loka-Mahadevi herself can be seen depicted on a pillar;
though elderly when the temple was made, she is depicted as a beautiful
young woman standing on a throne upheld by three lions,144 symbols of the
warrior-goddess Durga. Her arms are decorated with bands of gold, her hair
piled up and fastened with a large jewel. In her hand she holds a standard
crowned by an elephant – a symbol of the royalty that she figuratively and
literally upheld. She looks down at viewers with a demure expression, even
as the fact that this entire building is hers tells a different story. Elsewhere
on the mandapa, the queen is again alluded to by a large, noble lion under
which a little boar takes shelter – a statement that she is Durga, the lion-
riding aspect of the Divine Mother that shelters the Chalukya family with
her bravery and ferocity.145 Close to this lion and boar is a slender young
woman, her arm draped around a tree branch (a symbol of fertility), with a
small boy standing at her feet as she stares off to the northwest. This,
perhaps, is her dead sister Trailokya-Mahadevi, mother of the heir, who
stands in her sister’s temple looking at the temple that she did not live to see
completed.146 One is tempted to see a flash of personal rivalry in this: Loka-
Mahadevi never ordered her sister’s temple completed through all her years
as queen, and yet chose to have her portrayed as a minor sculpture on her
own temple and had her positioned in this way. Whatever the truth of this
matter, the mandapa of the Lokeshvara, with its repeated imagery of the
goddess and the generative powers of the feminine, is a clear symbol of
Queen Loka-Mahadevi’s position as the mother of the Chalukyas. ‘By her
erection of the Lokesvara temple the Queen upholds her duties to both god
and husband and thus upholds dharma.’147
The Lokeshvara temple is also is a powerful testament to Vikramaditya II.
A single look at this enormous building makes it clear how far the Deccan
in general, and the Chalukyas in particular, had come.

The Chalukya king Mangalesha had once donated a single village’s

revenue to his family deity, Shiva Makuteshvara, in gratitude for his
military success against the Kalachuris. Now the Kalachuri wife of his
distant descendant, Vikramaditya II, could donate the revenues of fifty
villages and mobilize the resources of hundreds of settlements to build a
temple in her own name. The Chalukya king Pulakeshin II’s temples had
been simple, austere structures with their architectural decoration barely
visible. Every part of the Lokeshvara temple, in contrast, is beautifully
formed and adorned with mini-temples. The dome topping the
superstructure seems to cascade down along the cardinal directions, the
rhythm expanding at every tier with larger and larger shala (barrel-shaped)
and kuta (dome-shaped) miniature shrines. This much larger building also
has a more complex floor plan: no longer a simple square, but a square with
additional squares projecting out of its axes of symmetry. This evokes a
sense that the rhythm of the superstructure is carried all the way down to
the base, and creates recesses and niches that impart dynamism to the
otherwise still walls – a uniquely Deccan architectural innovation.
On this increased surface area of the temple are placed carefully selected
sculptures with clear political meanings that would have been obvious to
Vikramaditya II’s vassals. Shiva the Ganga Bearer, for example, stands
there as a reminder of Shiva in Narasimha I Pallava’s Great Penance relief
that had been defaced by Vikramaditya I; this sculpture also commemorates
the conquests of Vikramaditya II’s father Vijayaditya, who brought to the
south the Ganga and Yamuna banners. Mythological stories abound on the
Lokeshvara’s walls, images of wrestling warriors symbolizing the long
Chalukya–Pallava struggle as well as the divine wrestling match between
Arjuna and Shiva. Rama, the ideal hero-king, is depicted, hinting at the
supposedly ideal Vikramaditya II. Also featured is his rival the demon-king
Ravana, a metaphorical representation of Narasimha Pallava, shaking the
great mountain Kailasha on which Shiva lived, only to have his arms
crushed by a single tap of the god’s foot. This sculpture seems to say that
Narasimha Pallava was little more than a demon-king. And now
Vikramaditya II, like Shiva, had humiliated and embarrassed the Pallavas
with his own mercy.
 The Lokeshvara temple has many messages sequestered in its walls by the
queen and the sthapati. But what appears to dominate is the Chalukya
victory over the Pallavas, a message that the royal couple thought best
established them as upholders of dharma, and thus worthy to rule and
continue to rule.
Perhaps, as many historians have seen it, the Lokeshvara temple is indeed
the pinnacle of Chalukya power. Or, perhaps, it is politics and religion
turned into high art, a story, a drama, a statement of power that ignored a
present that was far more complex, unfavourable and uncertain than the
story it tells. Whatever its true meaning, the resources that had been poured
into this enormous publicity stunt would soon be desperately needed
elsewhere. For in the north, the young man whom Vikramaditya II had
encouraged in his early career was about to transform the Deccan. Danti-
Durga Rashtrakuta had big plans.
Dikshit 1980, 133.
Ibid., 148.
On Vinayaditya’s visits to river fords, see Subhashini Kaligotla, ‘Shiva’s Waterfront Temples:
Reimagining the Sacred Architecture of India’s Deccan Region’, PhD dissertation (Columbia
University, 2015), https://academiccommons.columbia.edu/doi/10.7916/D83N2B1V/download,
230.
Ibid., 231–32.
Hemanth Kadambi, ‘Sacred Landscapes in Early Medieval South India: The Chalukya State and
Society (ca. AD 550–750)’, PhD dissertation (University of Michigan, 2011), 230.
Dikshit 1980, 148.
The princesses were likely still children; the younger of them would give birth to Vikramaditya’s son
and heir only around 720 ce. They may have had many daughters up to that point, whose names
have not survived.
Davis 1993, 37.
Dikshit 1980, 153.
Ibid., 152.
The Satavahana king Gautamiputra Satakarni is believed to have done something similar to seize
relics of the Buddha, but this is difficult to verify as the only sources claiming this were
commissioned by his immediate relatives.
Davis 1993, 40.
Ibid., 40–41.
Inden 2000, 258.
Ibid.
Kadambi 2011, 88.
Himanshu Prabha Ray, ‘Creating Religious Identity: Archaeology of Early Temples in the
Malaprabha Valley’, in Archaeology and Text: The Temple in South Asia, ed. Himanshu Prabha
Ray (Oxford University Press, 2010), 15–37.
Ibid.
R. Champakalakshmi, Trade, Ideology and Urbanisation: South India, 300 BC to AD 1300 (Oxford
University Press, 1996), 313.
Ibid., 520.
Meera Abraham, Two Medieval Merchant Guilds of South India (Manohar, 1988), 47.
Ibid., 74.
Kadambi 2011, 155.
Srinivas V. Padigar, ‘Craftsmen’s Inscriptions from Badami: Their Significance’, in Ellora Caves:
Sculpture and Architecture, eds. Ratan Parimoo, Deepak Kannal and Shivaji Pannikar (Aprant,
2018), Revised Edition, 313.
Ibid., 316.
Kaligotla 2015, 184–85.
Ibid., 183.
Indian Antiquary, Vol. X, 103. Hereafter IA.
Buchanan 1985, 344.
Ibid.
Kaligotla 2015, 174.
Ali 2006, 35.
Dikshit 1980, 220–21.
Ibid., 221.
Nandi 1973, 16–17.
Ibid., 213.
Ibid., 212.
Ali 2006, 52.
Ibid.
Ibid.
Ali 2006, 54.
Early Chalukya texts make it fairly clear that they were familiar with the vocabulary of texts finalized
in the early centuries ce and associated with the ‘mortal goals’ of artha and dharma, such as the
Arthashastra and Manavadharmashastra, thus making it highly likely that they were also aware of
the Kamasutra, dedicated to the goal of kama and dating from slightly later than these two. In his
seminal work on court culture, Daud Ali (2006) uses it alongside a later Chalukya text, the
Manasollasa, to develop an understanding of the attitudes of the courtly world.
Vatsyayana, Kama Sutra: A Guide to the Art of Pleasure, trans. A.N.D. Haksar (Penguin, 2011), 57.
Ibid., 58.
Ali 2006, 54.
Vatsyayana 2011, 110.
Ibid.
An example is Prince Vikramaditya’s aunt Kumkumadevi. See EI XXXII, 317–25. Mangalesha’s
mother also did something similar in her donations to the god Makuteshvara.
Buchanan 1985, 352.
Ibid.
Arguably the most powerful and prominent was the ninth-century Odia queen Tribhuvana-Mahadevi.
EHD I, 235; SII XI, 2.
Cummings 2006, 99.
Ibid., 97.
This idea is explored at length in the second chapter of Cummings 2006.
Ibid., 101.
EI XIV, 189.
Ibid.
Khalid Yahya Blankinship, The End of the Jihad State: The Reign of Hisham Ibn’ Abd al-Malik and
the Collapse of the Umayyads (SUNY Press, 1994), 131.
Dikshit 1980, 160.
Robert DeCaroli, ‘An Analysis of Dan.d.in’s Daśakumāracarita and Its Implications for Both the
Vākāṭaka and Pallava Courts’, Journal of the American Oriental Society 115, no. 4 (1995).
Ibid.
‘Introduction’, in Dandin 2007. For ease of reading, Sarasvati is substituted for ‘Vedhas’ spouse’.
Ibid.
Ibid.
Ibid.
DeCaroli 1995.
Dandin 2007, Location 1549.
Ibid., Location 1551.
Ibid., Location 1569.
John Keay, India: A History. Revised and Updated (Grove/Atlantic 2011), 180.
Hugh Kennedy, The Armies of the Caliphs: Military and Society in the Early Islamic State
(Routledge, 2013), 6.
Ibid., 7.
Peter Frankopan, The Silk Roads: A New History of the World (Bloomsbury, 2015), 43.
Hans Bielenstein, Diplomacy and Trade in the Chinese World, 589–1276 (Brill, 2005), 75.
K.A. Nilakanta Sastri, Foreign Notices of South India: From Megasthenes to Ma Huan (University of
Madras, 1939), 116.
Ibid.
Keay 2011, 174.
Sastri 1997, 139.
Keay 2011, 176.
SII I, 10.
Cummings 2006, 101, Note 167.
Assuming Vijayaditya was at least in his twenties in 674 (since Vikramaditya I mentions him
commanding armies), he would have been in his eighties by the 630s.
Why attack in 731 and not in 728? It may be because the Arab position in western India had suffered
a temporary setback and they were already being attacked by other Indian states, so Vikramaditya
could signal his strength by leading an attack in another direction. In any case, if Vikramaditya was
going to get involved with the Arabs, he would certainly have done it during Junaid’s first set of
campaigns in Gujarat in the early 720s. That he chose to attack the Pallavas instead and rekindle a
conflict that had been at rest for at least forty years suggests that he had only been waiting for the
right opportunity. Moreover, the possibility that Paramesvara-Varman II’s position was weaker in
731 than in 728 due to domestic factors we know nothing about – thus prompting Vikramaditya to
attack him – cannot be ruled out. The fact that he surrendered a ransom apparently without offering
serious resistance certainly points to the fact that his position was weaker than one would expect.
Adiga 2006, 102.
We will see an example of such a lord, the Rashtrakuta king Danti-Durga, in Chapter 5.
Dikshit 1980, 160; Sastri 1997, 141.
Sastri 1997, 141.
Blankinship 1994, 134.
Arabs are generally referred to as Tajikas or Parasikas in Indian sources, which according to Jayashri
Mishra (1992, 171) is derived from ‘Taji’, ‘an Arabic word meaning Arab country and also Arab
horse’. Meanwhile, Brajadulal Chattopadhyaya indicates that the term was ‘derived from Arabic
tribal or tribal confederation of the Tayy’, and, further, that ‘an old Parthian formation of the name
which by the third century must have been Tāzīg may be envisaged’. See Brajadulal
Chattopadhyaya, Representing the Other: Sanskrit Sources and the Muslims, Eighth to Fourteenth
Century (Primus Books, 1998), Kindle Edition. This book will use the term ‘Tajika’ to refer to the
Umayyad armies in this chapter, and call them Arabs and Persians elsewhere, in order to bring out
the disparate meanings with which the terms are imbued in medieval and modern contexts.
Blankinship 1994, 133.
Ibid.
It should be noted that the Arabs were not exceptional in their involvement in the slave trade: African
 and Indian rulers also participated in it to various extents through the medieval period.
Blankinship 1994, 134.
Ibid., 148.
Ibid.
Ibid., 146.
R.C. Majumdar claims that the Rashtrakuta king Danti-Durga was commanded by Vikramaditya II to
assist against the Tajikas. See note 109 below. R.C. Majumdar, The Age of Imperial Kannauj, The
History and Culture of the Indian People, Vol. 4, Second Edition (Bharatiya Vidya Bhavan, 1959).
There is no evidence provided to support this assertion, and this book will depict Danti-Durga’s
relationship with the Chalukyas somewhat differently.
Ibid., 187.
Pulakeshi-Raja was the paternal cousin of Vikramaditya II’s grandfather Vinayaditya. Strictly
speaking this makes him his first cousin twice removed, but ‘distant grand-uncle’ has been used as
a more intuitive term.
Dikshit 1980, 167.
Corpus Inscriptionum Indicarum IV, 144. Hereafter CII.
Ibid. Paraphrased to simplify the language.
Ibid. Paraphrased to simplify the language.
Blankinship 1994, 188
Ibid.
It is worth mentioning that though the Arabs hereafter adopted more peaceful policies, they did not
speak for the entire Muslim world, and the subcontinent would eventually be invaded by Turks,
Afghans, Turco-Mongols and Persians.
Blankinship 1994.
Majumdar 1959, 18, Note 11: ‘This para is based on the inference suggested by the Navsari Plates of
Pulakeshin and Ellora Plates of Dantidurga.’ In the Navsari Plates, Pulakeshin (Pulakeshi-Raja in
the text) mentions no other allies in his battle, and the Rashtrakuta king Danti-Durga in his Ellora
Plates similarly makes no claims to this effect. Dr Majumdar may be basing this on a misreading of
the Ellora Plates suggesting that Danti-Durga conquered Sind, but this has been dismissed by other
scholars. As the editor of the Ellora Plates of Dantidurga puts it: ‘there is no question of
Dantidurga conquering Sind.’ See EI XXV, 29.
CII IV, 143.
Ibid.
Ibid. Pulakeshi was only the second generation of Chalukyas to have a strong position in Lata and
may very well have been seen as a foreigner by his subjects.
Ibid.
Given the fragmentary nature of the evidence, this is proposed as a much better explanation for
Danti-Durga’s sudden rise to prominence than Majumdar’s (1959) proposal.
Chakrabarti 2011, Location 1756.
Inden 2000, 244.
In later inscriptions of Vikramaditya II, he implies that he went to war with the Pallavas immediately
after his accession, which does not make sense since he makes no reference to them in inscriptions
dedicated in the second year of his reign. Dikshit (1980) argues that this second campaign occurred
between 735 and 740 but does not connect it to domestic politics owing to the assumption that
Pulakeshi-Raja was acting on Vikramaditya’s orders. Though Dikshit mentions Pulakeshi’s use of
 the title Parama-Bhattaraka, he does not connect it to the Pallava campaigns, and seems uncertain
about Pulakeshi’s intentions. The perspective adopted here is that all these occurrences are
connected, and that the second campaign may have been prompted by a need to respond to
Pulakeshi-Raja’s activities in the northern part of the Chalukya domains.
He would later claim to have sacked Kanchi, which, given that his power was concentrated in
northern Maharasthra, seems implausible unless he had done it under the authority of the
Chalukyas early in his career.
Sastri 1997, 139–40.
Dikshit 1980, 167.
EI III, 360.
Ibid., 359.
Ibid., 360.
SII I, 146.
Kaligotla 2015, 181.
Buchanan 1985, 400.
It is possible to make this statement with confidence as it is hardly possible that the queen would not
have interacted with the chief architect of her own temple, given that both of them are explicitly
mentioned in inscriptions on it. Given that Loka-Mahadevi also administered territory in her own
right, it is also very likely that she was used to speaking and commanding both female and male
subordinates and likely had her own court.
IA X, 164–65. In this inscription, Loka-Mahadevi announces that Gundan had averted the
excommunication of his caste by building the temple. As such, it seems reasonable to assume that
any issue which led to such an important guild almost being excommunicated would certainly have
caused a huge scandal.
It is, according to the Dharmasutras, illegal to provide any sustenance to one who has been
excommunicated from caste. We cannot be certain this was followed uniformly across the
subcontinent or even to what degree it was followed if it was.
Gautama Dharmasutra 21.1, in Patrick Olivelle, Dharmasutras: The Law Codes of Ancient India
(Oxford University Press, 1999), 114.
Cummings 2006, 95.
Buchanan 1985, 399.
Cummings 2006, 102.
Buchanan 1985, 399.
Ibid., 400.
IA X, 166.
Ibid., 167.
Ibid.
M.K. Dhavalikar, ‘Sutradhara’, Annals of the Bhandarkar Oriental Research Institute 52, no. 1
(1971): 218.
Cummings 2006, 102.
Ibid., 91.
Kaligotla 2015, 177.
Cummings 2006, 120.
Ibid., 125.
Ibid., 126–27.
Ibid., 127–28.
Ibid., 138.
 Part II
Apogee: The World of the
Rashtrakutas
 5
The Elephant Fort
Monsoon, 737 CE

Let us leave the sandstone cliffs of the Malaprabha river valley and fly
north, to meet the dynasty that would bring doom upon the Chalukyas.
What might we see below us, as we traverse these untamed medieval
Indian landscapes? It would mostly be the craggy, unrelenting natural
beauty of the Deccan. Monsoon clouds break over the land having shed
their waters over the Western Ghats. Below us are whorls of red and black
soil interspersed with dry vegetation, the remnants of primeval volcanic
activity; rain-fed rivers cutting through landscapes of boulders; jungles
teeming with wildlife and flocks of thousands of colourful birds. Here and
there, we might see tiny villages. Sometimes, along rivers, we might even
see small towns, usually along trade routes or at a river ford or confluence.
For all the changes we have witnessed over the last two centuries, northern
Karnataka was far less urbanized than it would be in its heyday as the seat
of the great empires of Vijayanagara and the Deccan Sultanates in the early
modern period.
But as we fly over Maharashtra, we see a somewhat different landscape.
As we enter the great valley of the Godavari river – one of the
subcontinent’s largest, third in size to the valleys of the Ganga and Indus –
we might see villages more frequently. We might see cities of considerable
antiquity, such as Pratishthana (modern-day Paithan) and Nasikya (modern-
day Nashik), where the Chalukya king Pulakeshin II had resided to
consolidate his power in the region a century ago. We come across people
who speak Maharashtri Prakrit, an Indo-Aryan language very different from
the Dravidian Old Kannada spoken by the Chalukyas. We see ancient
Buddhist cave temples and trade routes that traverse the Western Ghats,
leading to bustling port towns. We see the growing pilgrimage centre of
Ilapura (present-day Ellora), not too far from the Dakshinapatha trade route
between northern and southern India. Swooping down into misty, rolling
hills covered in forests, we would also see many Adivasi peoples, whose
history, culture and languages barely survive today. But at the time, they
were major power brokers. Ellora had originally been one of their sacred
sites. Royals who had settled in this general region, such the Lata
Chalukyas, made marriage alliances with them. And they controlled
important natural resources, such as luxury woods and elephants.
Moving northwest from Ellora, we would see, in the distance, the
Narmada river, where Pulakeshin II defeated Harsha. On its northern bank
is Mahishmati, once the seat of power of the Kalachuris, the birth family of
the Chalukya queen Loka-Mahadevi. Our view of this town is obscured by
the ancient Vindhya hills, but there is a gap between them, through which
traders, monks, artists, priests and armies have moved for many centuries –
the Burhanpur gap. Off to the southeast of the gap near the Tapti river, close
to the trade route and with an eye to the dense forests of central India, is
Achalapura (modern-day Elichpur). Here, according to some scholars, the
family known as the Rashtrakutas (literally ‘state headman’),1 so distant
from Chalukya power at Vatapi, dreamed of empire.
The origins of the Rashtrakutas, as with the Chalukyas, are obscure. Some
historians have attempted to prove, based on flimsy evidence, that they
were Marathas or at least Marathis. Some even claim that they were Rajputs
– though neither of these groups or categories would exist for centuries
after. These anachronistic claims are driven by a need to project modern
political identities onto a much more complex past: many forgotten
ethnicities and groups have risen to and fallen from power in the vast
history of the subcontinent, and the Rashtrakutas present a particularly
unique case of this. It is most likely that they were Kannada-speaking
military aristocrats settled at a strategic point in modern-day Maharasthra
by the Chalukyas or some other powerful group, perhaps to keep an eye on
trade routes and various tribal peoples. Their ‘original’ caste and ethnicity
is impossible to retrieve; given their journey to imperial status, and the
complex intermarriages and new self-presentations this involved –
processes we have seen with the Chalukyas – it is not even relevant to their
future story.
Here at the northernmost frontier of the Chalukya political network, the
Rashtrakutas and the Chalukyas of Lata trod a thin line between obeying
the emperor at Vatapi and defiantly pursuing their own relentless ambitions
of conquest. By the time of the Chalukya Vikramaditya II’s accession, the
two families had already clashed once. Bhavanaga, the niece of the Lata
Chalukya king Pulakeshi-Raja, defeater of the Umayyads, was married to a
Rashtrakuta king. But this does not seem to have been a voluntary marriage
according to some: later Rashtrakuta sources claim2 they kidnapped her on
the very day of her wedding to some other nobleman. This might just be
propaganda to show them as being more ‘virile’ and martial than the Lata
Chalukyas. And it seems the Lata Chalukyas cared little for what became of
their daughter. Either way, this lady represented a major step up, maritally
speaking, for the Rashtrakutas, then petty provincial warlords. Bhavanaga’s
father was a Chalukya, thus granting her descent from Pulakeshin II,
Defeater of Harsha. Her mother may have been the daughter of an
important Adivasi tribal chief.3
 Whatever the circumstances of her marriage, Bhavanaga, in public at
least, was treated with the utmost respect by her Rashtrakuta in-laws, for
she had (involuntarily) ennobled their line through a direct tie to the
imperial clan.4 And she seems to have reclaimed her agency over the
decades as a Rashtrakuta queen – by adopting, as the Kalachuri princess
Loka-Mahadevi had, the ambitions of her marital family. Unlike Loka-
Mahadevi, though, Bhavanaga seems to have brought up her son, for
whatever reason, to take revenge on her Lata Chalukya brothers and then
her Vatapi Chalukya relations. This young man, Danti-Durga Rashtrakuta,
who was probably raised on tales of his own descent from Pulakeshin II,
and of the vacillation and distraction of Vikramaditya II, had in his mother
Bhavanaga a powerful ally and adviser.5 Danti-Durga was a brilliant and
canny politician, receiving the title of Prithivi-Vallabha while, at least
outwardly, a loyal vassal of Vikramaditya II; he was also conceited and
decisive, and was a master of art and propaganda. He led one of the most
astonishing lives of eighth-century India, as short as it was transformative,
and he would leave the Deccan profoundly changed.
This Rashtrakuta king and his mother watched as Vikramaditya II and
Loka-Mahadevi obsessed over the Pallavas and the Lokeshvara temple
while the Tajikas overran Sind and Kathiawar and raided Malwa, spreading
chaos and confusion. They watched as Pulakeshi-Raja, puffed up with his
hard-fought victory over the Tajikas, alienated his overlords, leaving his
rear exposed to them. Carefully they planned their move.

In the first part of this book we’ve observed first-hand how a medieval
Indian kingdom was created where none had existed before, and how
Sanskrit court culture was adopted by former agriculturists to aid their
transformation into sophisticated aristocrats. The Rashtrakutas, starting with
their eccentric young king Danti-Durga, will allow us to see in detail the
inner workings of medieval Indian kingdoms and aristocrats: their savage
military power, their astounding global ambitions, and their sometimes
surprising humanity. To do this, we will have to peer beyond the flowery
language of prashastis and the elaborate symbolism of art, and imagine the
living, breathing people beyond the staid, upstanding, ‘dharmic’ image they
try to portray.
When the young Danti-Durga came to the throne in the early eighth
century, he and his mother Bhavanaga appear to have first set up a solid
military base with which to contest Chalukya supremacy. They seem to
have done this by securing the grandest symbol of Indian kingship, the tank
of the premodern world, Danti-Durga’s namesake: the elephant, danti.
By the time of Danti-Durga’s accession to the Rashtrakuta throne, the
Indian tradition of elephant warfare (gaja-shastra) was at least two
thousand years old, and Indian mahouts were the envy of the world.6
Indeed, the use of these animals for war was almost certainly an Indian
invention.7 Given the position of the Rashtrakuta kingdom – at the northern
edge of the Deccan plateau, close to the dense forests of central India –
Danti-Durga was perfectly positioned to capture these vital military
resources.8 One of the earliest targets of Rashtrakuta expansion seems to
have been southern Chhattisgarh (then called Kosala), famed for its
elephants.9 Its rulers, probably mostly Adivasi chiefs or minor kings, were
defeated and forced to pay tribute in elephants. A portion of them may have
been proffered to Danti-Durga’s relative and overlord, the old Chalukya
emperor Vikramaditya II, to reassure him of Danti-Durga’s good intentions.
It is tempting to imagine a queue of the great beasts trudging the hundreds
of kilometres to Vatapi, their squeals and trumpets interspersed with the
yells of mahouts, the clinking of chains.
A captured elephant – a sensitive, intelligent and tortured creature – is a
sorry sight. But a fully trained male war elephant in heat – musth, a state of
heightened aggression and sexual energy – must have been magnificent and
terrifying to gaze upon. As with those of many powerful medieval kings,
Danti-Durga’s elephant corps likely consisted of dozens of the animals.
Cinnamon-scented fluid leaking from their temporal glands, eyes glowing,,
barely restrained by the sharp goad (ankusha)10 that their mahouts dug into
their flesh to guide and control their aggression, flicking their man-killing
trunks – all this military power gathered in one spot must have been a hair-
raising sight. Massive and muscular, painted with designs and decked with
garlands, long, gleaming tusks fitted with spikes, a clanging necklace of
bells around the neck,11 an elephant on the battlefield called for a very brave
man (or a fool) to stand up to it.
Medieval India, of course, lacked for neither fools nor brave men to fight
Danti-Durga’s beloved12 elephant corps. One such individual was his
relative, Pulakeshi-Raja of the Lata Chalukyas.
Almost immediately after his victory over the Tajikas in 737, Pulakeshi-
Raja was already positioning himself for a takeover of Gujarat, as his
propaganda recycled elements from the prashastis of older ruling dynasties.
He adopted for himself and his forebears the imperial title of Parama-
Bhattaraka – without fully realizing how his actions would be perceived in
Vatapi. Soon after, Danti-Durga, his ambitious relation, appears to have
served in the Chalukya emperor Vikramaditya II’s second expedition to
Kanchi. The emperor seems to have encouraged the young man’s ambitions
of expansion13 as a counterbalance to the increasingly independent
Pulakeshi-Raja. Given what we know of Deccani women in antahpurams, it
is easy to imagine that the Chalukya queen Loka-Mahadevi and the
resentful Rashtrakuta queen mother Bhavanaga played a role in these
decisions.
But as much as Danti-Durga may have loved his elephants, the real source
of this Rashtrakuta’s power was masterful politics and propaganda, which
enabled the long-term marshalling of economic resources. The northern
Deccan, especially in the regions corresponding to modern western
Maharasthra, had been urbanized and irrigated for many centuries, and its
trade and agricultural networks were vast and ancient – much more so than
the Malaprabha river valley. This was why earlier generations of Chalukya
kings had devoted considerable attention to keeping their northern frontier
under their control with conquest and marriage. Old Pulakeshin II had spent
so much time at the city of Nashik that a contemporary Chinese pilgrim
believed it to be his capital. Vikramaditya II’s father, Vijayaditya, had
visited the pilgrimage site of Ilapura/Ellora during one of his northern
campaigns, and handed out land there.14 But now, it seems, the Vatapi
emperor no longer had sufficient resources or attention to dedicate to the
northern Deccan, relying on his vassals to do the hard work while he
focused on easier targets in the south. This allowed Danti-Durga
Rashtrakuta to expand his influence nearly 400 kilometres beyond his
family’s ancestral holdings to one of the most ancient and prestigious
religious sites in the entire Deccan. In 742 CE, Danti-Durga came to the
tirtha of Ellora. Here, we shall see how pluralistic the religious practices of
medieval India really were, and observe the brilliance of its artists,
propagandists, kings and monks in Ellora – a wondrous and tragically
ignored centre of religious, artistic and political flowering.

Millions of years ago, the Deccan was shaped out of lava flows that
engulfed the Indian subcontinent, then an enormous island, as it moved
towards its collision with Asia. At Ellora, cliffs of black basalt formed out
of the lava. These cliffs would eventually become an important site in the
medieval period, and many generations would leave traces of the religious
diversity, innovations and struggles of their world in the shrines, temples
and sculptures carved here. They also illustrate the continuity of religious
practices before and after the coming of kings, Brahmins, and organized
temple-based and monastic cults. Though royals, through their inscriptions,
portray themselves as the prime movers of all history, the truth of the
matter, as we have seen, is rather more complicated.
High up above these basalt cliffs, in hills and forests, the river
Ela/Aparnika flows south through sacred groves and pools, cascading into a
spectacular waterfall15 at Ellora, and thence flowing down to the river Shiva
and the Godavari.16 Aparna, ‘Leafless’, was the goddess of the ancient
forest-dwelling peoples17 who had called this place home for hundreds of
years before the coming of Shaivas and other organized religions.
Bhavanaga, Danti-Durga’s mother, was possibly a descendant of these
peoples.
The worship of the river and the goddess had been central to the religious
practice of these Adivasis. Indeed, the worship of a river goddess continues
there to this day, though Aparna’s original worshippers have long since
been incorporated into India’s teeming, diverse masses.18 The dominance of
tribal forms of worship at Ellora, however, did not last forever. As a holy
site that drew worshippers from the neighbourhood as well as itinerant
merchants and preachers, it was only a matter of time before kings and
chiefs, Buddhists and Brahmins, became interested in Ellora. Indeed, the
earliest definite traces of Buddhist activity there can be discerned by the
fourth century CE.
As Puranic, temple-based Hinduism began to spread across the
subcontinent in the fifth century CE, a new religion arrived at Ellora.
According to legend, a chieftain ‘marked with the emblems of Shiva’19
came to the area to find worshippers of Vishnu and the Great Goddess,
Jains, assorted ascetics and gurus, and sects called Sauras and Maladharas
there already. Evidently, irrespective of its tribal origins, Ellora held a
magical allure for all the major sects of the medieval period, and they
coexisted without too much political interference. This fifth-century Shaiva
chieftain, however, is supposed to have begun the process of claiming the
land for Shiva by settling Brahmins in ten villages there.20 It would be a
couple of centuries before Shaivas would dominate the site, however:
Buddhists and Jains wielded considerable influence at Ellora. Cave temples
carved into the basalt cliffs, often commissioned by communities of non-
royal patrons, show that all these cults were served by the same architects
and sculptors. They freely exchanged iconographic elements and perhaps
even ritual practices,21 contributing to ‘Ellora’s fame as a powerful and
extraordinary place of worship’.22
Patrons, both royal and non-royal, paid for thriving monasteries to be cut
into the primordial rock, resounding with the drone-like chants of saffron-
and white-clad monks. Epigraphic evidence from the sixth century CE
suggests that the early Kalachuris of Mahishmati – Loka-Mahadevi’s
ancestors – were involved in excavating some of the structures there. At
roughly the same time, the Vatapi Chalukyas had taken over the ancient
spring at Mahakuta in the Malaprabha valley and remade it in the image of
their new family deity, Shiva Makuteshvara, the Crowned Lord. But as
Ellora shows, royals are not always needed for the process of religious
transition: medieval merchants, pilgrims and preachers were capable of
profoundly transforming landscapes by themselves.
As generation after generation of sthapatis and patrons came and went,
cave temples teeming with sculpture were carved into the primordial Earth,
shallow and stiff and small in the first generations, slowly growing to titanic
proportions. These elaborate compositions supported some of the most
beautiful sculptures the subcontinent had yet seen. By the eighth century,
Ellora’s sculptors23 had perfected a sculptural style that depicted stately
deities brimming with calm energy. Ellora had many other features that set
it apart from other tirthas, increasing its appeal for any ambitious king. It is
one of the few sites in early medieval India where the Shiva linga –
believed to be a symbol of primordial masculine energy – is attested to
being actively worshipped by women.24 According to a story that emerged
around this time, Shiva had brought back to life the murdered son of a
Brahmin woman who worshipped him at the sacred waterfall, and female
worshippers continued to come to Ellora in the hope of receiving similar
blessings.25
 Through the gradual accretion of such legends, and with the growing
influence of Shaivism across the subcontinent, the tirtha at Ellora came to
be associated primarily with Shiva.26 The ten tiny villages of the fifth
century had grown into a great pilgrimage hub. Hordes of preachers and
pilgrims, and the crowds of hawkers and merchants needed to feed them,
were drawn there, as well as powerful kingdoms seeking to advertise their
wealth and devotion.
It is little wonder, then, that when Danti-Durga set out to solidify
Rashtrakuta power, he chose Ellora, associated with so many gods and
peoples, as the site to do so. His land grant of the year 742 was given to
Brahmins from Navasarika, the capital of the Lata Chalukyas, after bathing
in the waters of the tirtha27 – this is a waterfall that can still be seen in
Ellora today during the monsoon. Just as his Vatapi Chalukya ancestors had
imported Brahmins from prestigious northern India, and his Lata Chalukya
grand-uncle Pulakeshi-Raja had brought them from prestigious Karnataka,
so Danti-Durga imported Brahmins from the Lata Chalukya capital28 to
legitimize himself and establish himself as a royal patron.

This new kingdom of Danti-Durga’s – his activities and influence now

stretching from Elichpur to Ellora – spanned an area of hundreds of square
kilometres, a respectable kingdom by any standard, already equivalent to
those of the early Chalukya kings. But the young Rashtrakuta had goals
much larger than those of the impatient Pulakeshi-Raja of the Lata
Chalukyas. For what Danti-Durga was implying with his performance of
the dharma of kings at this ancient site, granting land and flaunting his titles
– including Prithivi-Vallabha, granted him by Vikramaditya II – was that his
descent from a Lata Chalukya mother entitled him to a claim on the Vatapi
Chalukya throne itself. He was declaring himself as martial, righteous,
generous, a supporter of Brahmins and of religion before the eyes of the
entire Deccan.
Vikramaditya II, meanwhile, was convinced he was doing the same thing
– by repeatedly attacking the hapless young Pallava king and funding
temples in his home territories. There can be little explanation for his
complacency regarding Danti-Durga’s activities and ambitions other than
that he nurtured a bit of a blind spot towards the young man he saw as an
ally against the Lata Chalukyas. Perhaps he was thrilled to hear the
Rashtrakuta king had one-upped the headstrong Pulakeshi-Raja by inviting
Brahmins from Navasarika to receive land at Ellora.
The year after Danti-Durga made his grant, 743 CE, construction of the
Lokeshvara temple reached its final stages. Vikramaditya II sent a third and
final expedition to attack Kanchi,29 this time under the command of his son,
Yuvaraja Kirti-Varman, attempting to establish the crown prince’s
credentials just as he and Loka-Mahadevi had established their own a little
more than a decade earlier by attacking the Pallavas. It does not seem this
expedition was anywhere near as kind to Kanchi as the one Vikramaditya
himself had led a few years earlier. Danti-Durga, ‘loyal’ vassal that he was,
was almost certainly also present in this campaign (as he may have been in
Vikramaditya II’s second campaign against the Pallavas), and helped
himself to a fair share of the booty (he would claim, in his later years, that
he had conquered Kanchi all by himself).30 The old Vallabha Vikramaditya
II, now having triumphed three times against this ancestral enemy, and
having lived a most exemplary life by the standards of medieval Indian
royalty, died soon after, in 745.31 His son Kirti-Varman II was crowned soon
after.
The death of Vikramaditya II was exactly what Danti-Durga had been
waiting for. His years as a Chalukya vassal had given him a keen
understanding of how their political network worked, and acquainted him
with most of the major players in the region. He also understood
Vikramaditya II’s enormous political miscalculations: not directly attacking
the Umayyads; refusing to make his presence felt in the northern Deccan;
and repeatedly attacking the already defeated Pallavas to shore up his
reputation. While the ambitious Rashtrakuta king had decided not to test his
luck against a well-established Vallabha with a solid support base in the
southern Deccan, there was no reason not to do so with the successor. By
745, Danti-Durga had been on the throne for more than a decade. As his
activities in Ellora show, he was far more confident in his power than the
newly crowned Chalukya emperor Kirti-Varman II. He immediately moved
to set himself up as a serious competitor to the Vatapi throne in the eyes of
the rest of the imperial network.
The most spectacular way to do so was through an ostentatious display of
military prowess,32 a public declaration of his right and ability to reorder the
world and uphold the dharma of kings. And so Danti-Durga’s elephants
headed across the Vindhyas to the ancient city of Ujjain. Ujjain had
successfully withstood the raids of the Umayyads little more than a decade
earlier. But by the late 740s, the region was torn by strife between two
branches of the ruling clan,33 the Pratiharas.
What the Tajikas had been unable to achieve, Danti-Durga now
accomplished. His war elephants served him well in this campaign,
battering down fortifications, charging down the streets, clearing the way
for his infantry to secure Ujjain. As his prashastis tell us, ‘the turreted
fortresses of his enemies fell down, together with their hearts’.34
Now, with the eyes of half the subcontinent on him, in a city immortalized
in poetry, plays and legend – a ‘hallowed piece of heaven’35 as the
celebrated poet Kalidasa once called it – Danti-Durga performed the great
sacrifice known as the Hiranyagarbha,36 the ‘Golden Womb’. This would
establish him as an individual capable of restoring cosmic order37 through
his connection to the gods and upholding of dharma. He was doing exactly
what his distant ancestor, Pulakeshin I, founder of the Chalukyas, had done
almost two hundred years before with his performance of the
Ashvamedha.38 He was declaring himself a truly independent, world-
ordering sovereign.
Like all the most important sacrifices, the Hiranyagarbha was a spectacle,
meant to awe visitors with its performance and impress upon them the
interconnected cosmic and political changes that were being enacted.
Around the sacrificial pavilion, doorways of fine wood must have been set
up, one in each cardinal direction, where the aristocrats of Ujjain defeated
by Danti-Durga39 were forced to stand as ceremonial gatekeepers. Ritual
manuals demanded that gold coins and gems be handed out to ‘the poor, the
helpless, the good, the worthy, and the Brahmanas’,40 amounting to as much
as 90–100 kilograms. We can imagine crowds thronging and cheering the
young king as the wealth was flaunted and distributed.
The sacrificial pavilion itself probably had fire pits in each of the cardinal
directions; it was most likely decorated with brightly coloured cloth banners
in red, yellow, green and blue,41 in addition to bells and the Rashtrakuta
imperial emblem, which was an eagle. Brahmins dressed in white silk,
wearing gold ornaments, droned on while Danti-Durga waited at the centre
of the pavilion in a great golden jar emblazoned with a sun, representing the
Cosmic Egg from which all Creation had sprung. The Egg used in the
sacrifice that day was fashioned from ‘gold in great quantity, dazzling in
splendour … strung with pearls and studded with rubies’.42 At long last, as
the chanting grew to a crescendo, there was a great fanfare of trumpets and
drums, and the triumphant young king emerged to the roars of crowds of
onlookers. Danti-Durga would henceforth be regarded as having obtained a
‘celestial body’ befitting an emperor, after his rebirth from the ‘Earthly
body’ that his mother had given him.43 A number of ceremonies, such as the
jatakarman, were then ‘celebrated as if he was a newly born child’,44 with
Danti-Durga’s wife playing the ceremonial role of a mother.
The entire Hiranyagarbha ceremony was carefully planned to convince
the performer and his witnesses of his rebirth and the new-found power and
authority that came with it. The exhilarated Rashtrakuta distributed ‘gifts of
sandals, shoes, umbrellas, seats, utensils, villages, countries’45 to his loyal
vassals, to the hosts of defeated aristocrats major and minor, and the
Brahmins who had gathered at Ujjain to bless him and seek his patronage,
as well as the crowds who gathered to watch and cheer this spectacular
royal ritual. The message was clear: the rulers of the Deccan now had not
one but two potential Vallabhas to whom they could offer their loyalties.
And Danti-Durga was very clearly wealthier, more powerful and more
generous than the new Chalukya emperor on the throne of Vatapi.
The Rashtrakuta and his Chalukya mother Bhavanaga wasted no time in
capitalizing on their position, as vassals began to flock to his eagle banner,
the Garuda-dhvaja of Vishnu. Whatever was left of his grand-uncle
Pulakeshi-Raja’s rule in Lata was now overrun by Danti-Durga’s rampaging
forces,46 and he boasted that ‘mankind gazed intently upon the … rending
asunder of the high banks of the great river Mahi and of the Reva, [the core
of the Lata Chalukya kingdom] accomplished by his victorious elephants’.47
Navasarika, once Bhavanaga’s home, was seized, and a Rashtrakuta
princeling installed as governor. And yet the new Chalukya emperor, Kirti-
Varman II, did not respond. Meanwhile, correctly interpreting Danti-
Durga’s triumphant Hiranyagarbha as a moment of humiliation and
dispersal for the moribund Chalukya imperial network, one of the Tamil
kings led an attack on the most loyal of the Chalukyas’ southern partners,
the Gangas. The Chalukya emperor headed south to confront him – and was
disastrously defeated in the battle of Venbai in 748.48 His performance of
royal duties now in question, he seems to have realized the only way left to
stabilize his position was by defeating Danti-Durga.
By 749,49 he was heading north, attempting to give land to loyalists: but
the further he went, the less support he found. Danti-Durga was now
handing out grants to his loyalists in villages and towns all across the
enormous basin of the Godavari.50 He even claimed that his mother
Bhavanaga Chalukya had made land grants in 400,000 villages51 – certainly
an exaggeration, but an indication that local power brokers, Brahmins and
village communities were much enthused by Rashtrakuta promises of a
more just and attentive rule. The Godavari river basin, under Rashtrakuta
domination, offered a much better base to a would-be Deccan empire than
the Chalukya-controlled Malaprabha river valley. Growing in confidence,
Danti-Durga expanded his footprint on the western coast by seizing parts of
the Konkan from the Vatapi Chalukyas in 751.52 An inscription
commissioned by merchants in this region shows that they acknowledged
Danti-Durga’s overlordship but were not yet confident enough to call him
their ultimate sovereign,53 waiting until the struggle was resolved one way
or another.
By 754, that resolution seemed imminent. The Rashtrakuta imperial
network, through feverish political activity, embraced most of modern
Maharashtra. Danti-Durga boasted that he had ‘straightaway conquered
Vallabha with a spike of wild rice that served him as a mace’,54 an indicator
of how crucial control over agrarian resources was. Significantly, this claim
was written in Nagari script – a distant ancestor of the modern Devanagari
script used mostly in northern India at the time, a contrast to the Old
Kannada script generally used by the Vatapi Chalukyas. This is the earliest
known instance of this script being used south of the Narmada, and its use
may have been meant to align Danti-Durga with the prestige of the Sanskrit
Cosmopolis, imbuing him with what the scholar Whitney Cox calls
‘imperial charisma’.55 This reminds us, once again, of how symbols of
prestige played a significant role in medieval Indian power struggles.
Danti-Durga may also have scored a few decisive victories against the
Chalukyas on the battlefield, though the inscriptional evidence is hazy. As
the balance of power shifted, the Rashtrakuta king set out on more
propaganda campaigns to garner support. And so we return once again to
Ellora for a final crowning display of Danti-Durga’s imperial might. The
Rashtrakuta king decided to build a more splendid edifice than had ever
existed at the site, taking over a Buddhist monastery that had only just
begun to be cut into the cliffs.56
Danti-Durga now set out, in the footsteps of the Pallava king Narasimha I,
to put the Chalukyas to shame through art and architecture. He ordered the
excavation of a cave temple that to this day is one of the largest in the
subcontinent. Here, the artists of Ellora would give free rein to their talent
with the unprecedented resources of a Deccan imperial formation – rather
than those of small communities of worshippers – at their disposal. Their
iconographic, compositional and sculptural talents were unleashed by this
ambitious young ruler with little patience for doing things the old way.
This new temple of Danti-Durga’s, which posterity would call the
Dashavatara cave at Ellora, has forty-four elegant pillars57 on its first floor.
The sthapatis understood that there was only one major light source in the
cave temple: the sun, filtering in from the façade. As you enter the temple
and look deeper into the cave, your perspective shifts and refracts along a
seemingly infinite expanse of pillars. As the sculptor and photographer
Carmel Berkson puts it:
like Indra’s gem which reflects light from a million cut planes … No single view can reveal the
composite totality. The shifting scenes seem to be limitless. With each step … the reality of the
cave begins to change, as the participant is plummeted into different relationships with the
elements in the cave, the columns, the space and the relief panels.58
 Your eyes are drawn to what lies straight ahead of you in the first pillared
corridor. Sunlight bathes the front row of pillars with their auspicious
overflowing pots, floral patterns, serpents and dwarves. As though drawn
through elegant gateways, you get closer and closer to the work of a true
master sculptor, of a calibre that puts the artistry of the Chalukyas’
Lokeshvara temple to shame. In a panel taller than the average man, a deep
cavity has been cut into the rock, around the sculpture, making it almost
pop out into the open space where one stands. It is a sculpture of
Narasimha, Vishnu the Man-Lion, locked in a titanic struggle with the
demon Hiranyakashipu (the brother of Hiranyaksha, the demon who stole
the Earth and was killed by Varaha). The stillness, the calm energy, you see
in older caves at Ellora is absent here.59 Instead, the sculptors of the
Dashavatara cave have created beings in motion, the energy and thrust of
their movement boldly positioned along diagonal axes.60 The strictly
delineated ideal proportions of orthodox shastras are dispensed with,
imbuing a raw, primal character to the sculptures’ heads, hands and feet.
Though one leg of each of its protagonists has long since fallen away, at one
point their feet were planted powerfully on the floor, directing their strength
and energy towards the clash at the centre. The multi-armed god radiates
power, the effect amplified by his many outstretched arms; his demonic
rival holds a shield and a thick straight sword, leaning slightly backwards
for a powerful slash. Evidently the two have been captured the very second
before the god disembowels the demon. The god and the demon are likely
meant to be analogies for Danti-Durga and his Chalukya rival. And this is
just one among the many priceless sculptures at the Dashavatara cave.
(Despite its name, it is dedicated not to Vishnu but to Shiva.)
As you approach the linga, an ancient silence descends upon you, close to
the heart of the mountain.61 In the dim light, Shiva slaughters demons,
Vishnu strides over the three worlds, and a powerful Varaha bears the Earth
upwards. How could anyone doubt that the person who created such
masterpieces was any less worthy to rule over the Earth than Vishnu
himself? Danti-Durga claimed for himself the full complement of Chalukya
imperial titles: Sri-Prithivi-Vallabha (Fortune’s Favourite, Earth’s Beloved),
Maharajadhiraja (Great King of Kings), Parameshvara (Supreme Lord),
Parama-Bhattaraka (Foremost Revered One), Khadgavaloka (Sword Sight),
Sri Danti-Durga-Raja-Deva. And perhaps he deserved them.
 Danti-Durga was preparing for a final diplomatic or military offensive by
756, aiming to capture the support of the most long-standing and loyal
Chalukya partners, such as the Sendrakas and the Gangas – who, being
bound to them by marriage, were prepared to defend their coalition with
bloody warfare. A great inscription was issued on the walls of the
Dashavatara, a full-fledged Rashtrakuta imperial prashasti. But when the
thirtieth verse was being carved, mere months before the army was due to
set out,62 disaster struck. What exactly happened, we do not know. We are
only told that word of the heroic Danti-Durga had reached the heavens, and
he was bombarded with the ‘pressing requests of the heavenly damsels’,63 to
which he understandably acquiesced.
One of the most remarkable individuals of the eighth century –
administrator, publicist, diplomat, aesthete – was dead.64 His beloved
elephant corps may have lined up solemnly as his corpse, so young that it
had left no heir,65 was carried to the grand funeral pyre. The overlordship of
the Deccan was not yet settled.

The Chalukya Vallabha Kirti-Varman II, his morale restored by this

apparent sign of divine favour, tried to bounce back. In fact, the man had
never really lost hope despite Danti-Durga’s activities over the 750s. While
Danti-Durga had claimed control over Maharashtra by 754, Kirti-Varman
had indulged in counter-propaganda, erecting a great pillar to remind his
remaining vassals in the south of his credentials. The pillar was prominently
positioned outside the Lokeshvara temple so that visitors would be sure to
see it and, hopefully, be reminded of the prestige and achievements of his
clan.
The dedicatory inscription began with an ode to the union of Shiva (Hara)
and his wife Uma (Gauri), ‘in which the face and breasts of the goddess are
passionately kissed [caressed] by the left arm of the god’,66 drawing a
parallel between Shiva and Uma and his own parents. He declared that the
Lokeshvara temple was built by Loka-Mahadevi, ‘who, like the divine
goddess Uma, was the very mother of mankind’, and that she had been the
wife of Vikramaditya II, Fortune’s Favourite, Earth’s Beloved, ‘the bruiser
of the town of Kanchi’. He also restarted construction on his mother
Trailokya-Mahadevi’s temple, which had been in abeyance for many years
after her death. Finally, the pillar attests that the emperor gave land to a
learned Brahmin brought at great expense from the north bank of the
Ganga,67 and gave him responsibility over the temple built by Kirti-
Varman’s grandfather Vijayaditya: the Vijayeshvara, Lord of Victory. He
was evidently attempting to reassure his vassals with a display of his reach,
devotion and influence. It was only appropriate that some effort be put into
pleasing the Lord of Victory, the lord of Vijayaditya, the Vallabha who had
raided north India. Interestingly, this inscription of Kirti-Varman’s – like
those of his rival Danti-Durga – was in the Nagari script, suggesting that he,
too, was now vying for legitimacy in the eyes of cosmopolitan elites.68
The year after the Rashtrakuta king’s death, Kirti-Varman II finally
gathered his forces and moved north, perhaps hoping to replicate his
grandfather’s achievements.69
But the Rashtrakuta family was not as fragmented as the Chalukyas might
have hoped. One of their foremost military commanders – Danti-Durga’s
paternal uncle Krishna – seized the throne for himself. Krishna may already
have fought the Chalukya emperor as a general in the early 750s, while
Danti-Durga occupied the throne. Now, in 757, the long-awaited Chalukya–
Rashtrakuta confrontation had arrived: and the Chalukyas found that
Krishna’s might could not be matched. The Rashtrakuta network was
simply too vast, and its economic and military resources beyond anything
the Chalukyas could hope to muster at this point. The Great Boar, according
to later Rashtrakuta propaganda, ‘turned into a she-deer’,70 and the last
emperor of the Vatapi Chalukyas and all his sons were killed.71 No trace
remains of how, when and where that battle was fought, but it was
undoubtedly a military disaster for the Chalukyas from which there was no
recovery. Krishna, we are told, ‘forcibly wrested away … on the battle-
field’ the Goddess of Royal Fortune of the Vatapi Chalukyas,72 ‘wearing the
garland of the fluttering banner in rows’,73 composed of the boar banner and
all the dozens of other standards that the Chalukyas had seized from their
defeated rivals over centuries. The goddess, supposedly overcome by
Krishna’s martial virility to the point where she was ‘listless’, acceded to
his affections.74 One wonders what Bhavanaga, the former Chalukya
princess and mother of the dead Danti-Durga Rashtrakuta, thought of all
this. At long last, she had finally had her revenge on her family.
The final defeat of the Vatapi Chalukyas sent political shockwaves
through Karnataka. All of a sudden, this dynasty, which had practically
become an institution over the past two hundred years, was gone – and
there was almost no hope of it returning. Local rulers and strongmen –
ranging from the priests of important shrines to wealthy merchants to vassal
kings – had to reconsider their position and decide whether to support a
Chalukya claimant or submit to this new superpower of the Deccan.
The Rashtrakutas moved quickly. Krishna Rashtrakuta, the new Vallabha,
took on the majestic titles of Subhatunga (Foremost in Fortune),
Akalavarsha (Rainer of Unexpected Favours),75 and, most tellingly, Sri-
Pralaya-Maha-Varaha76 – Great Boar of the Apocalypse. For in the deluge
that ended one age in an apocalyptic flood, Varaha the Great Boar had
rescued the Earth from the waters, allowing for new life and the beginning
of a new age. In Rashtrakuta propaganda, as Danti-Durga’s successful
conquests, sacrifices and temple building proved, the Chalukyas’ conduct
had not been righteous or strong enough. We are expected to believe that
old Krishna Rashtrakuta, ‘of fierce disposition towards the fierce, a mighty
repository of generosity towards the poor, most dear to women’,77 was
merely saving the Earth from the Chalukyas by taking it into his own hands
and regenerating it with his rule. It was a strange parallel to the Chalukyas’
own vicious overthrow of the Kadambas.
While there is little information of the tribulations that followed for the
Chalukya family and the Malaprabha valley, we know a good deal about the
horrors that the Rashtrakutas inflicted on other Deccan powers, specifically
the Gangas. The Ganga rulers had been personal friends and relations by
marriage of the Chalukya monarchs for generations. They were too
involved with the old structure of the network, especially given their own
primacy within it, to consent to a sudden shift in the political centre of
gravity to the northern Deccan – which would reduce their strategic
importance and thus their status in the long run. Nor did they have any
reason to think they would be better off submitting to a new emperor when
they themselves were such an ancient, prestigious and successful clan.
And so, war came once again to southern Karnataka, leaving its footprints
in the form of families torn apart, blood spilled, homes burned and silent
memorials to the dead. The Ganga ruler was a wiry and experienced old
man, and, despite his defeat by the Tamils a few years earlier and the death
of the Chalukya emperor, his nephew by marriage,78 he decided to fight it
out to the last breath. He assumed the title of Sri-Prithivi-Vallabha79 and
ordered the vassal kings Mutta, Nagata, Jadiya, Kitta, Pulikkada and
Siyagella to marshal their forces and rally to fight the invaders.80 They, and
many of the greatest heroes of the Ganga polity – distinguished in raids
against the Pallavas, celebrated warriors all – fought and died in desperate
battles against the Rashtrakutas in Pinchanur, Kagemogeyur, Ogaballi and
Baygeur.81 After each one, the devastated Gangas set up hero stones to
commemorate valiant dead warriors and give their deaths some sort of
meaning in the chaos. Sons, brothers, uncles, commanders, subordinates –
all were equal before the bloody tusks of the Rashtrakuta elephants, those
gleaming swords which cut through the Ganga armies like the flames of a
forest fire.
On these hero stones, the valiant dead were celebrated for their loyalty to
their overlords and adherence to the dharma.82 Decorated with flowers and
cloth,83 worshipped with blood, rice and ritual self-mutilation,84 these
memorials became rallying points for the heartbroken men who had
outlived their comrades. To the sound of throbbing drums and chants, these
warriors fell into trances, as they were supposedly possessed by the
vengeful dead,85 and went again and again to battle the Rashtrakutas.
But all their desperate resistance was to no avail. The hero stones of
Gangavadi, the Ganga kingdom, sprang up like meadows of flowers
blooming in the blood-soaked ground as Krishna Rashtrakuta
systematically devastated that country, unleashing his vassals and armies to
loot, sack and kill any who dared resist his authority as Vallabha, the sole
superpower of the Deccan. Sacrificed to the clash were men at every level –
ranging from nameless soldiers, who served only as background imagery in
their commander’s hero stone, to one of the Ganga king’s own sons.86 A
warrior called Kittayya, who has recently become a minor celebrity after his
hero stone was rediscovered, was killed by Rashtrakuta forces where the
modern city of Bengaluru87 would eventually emerge.
 In 768, eleven years after the death of the last Chalukya Vallabha, the new
Rashtrakuta Vallabha finally occupied the Ganga capital, Manyapura
(Manne in the modern-day district of Bengaluru Rural). Krishna ordered the
donation of land to Brahmins there88 to establish his right to distribute
Ganga territory as he saw fit, signalling his status as the overlord of their
lands. All other imperial claims except his own now lay in bloody tatters.
Like all those who had claimed the paramount sovereignty of the Deccan
before him, he had proved his ‘right’ to seize and hold the region with
violence. He had earned with blood and fire the titles of Sri-Prithivi-
Vallabha Maharajadhiraja, Fortune’s Favourite, Earth’s Beloved, Great
King of Kings.
 And so the Chalukyas of Vatapi, those sackers of Kanchi, those devotees
of Shiva, those builders of mighty temples, blood-soaked, vengeful,
tenacious, passed at last into history. They would raise their heads no more.
All that are left of them are stories, legends, the silent testimony of the
monuments they erected, the bustling of towns and markets they helped
create.
But the Deccan had been transformed. The process that the Chalukyas had
begun two centuries ago, gradually merging the ancient networks of the
northern Deccan with those of the south through war, diplomacy and
patronage, was reaching a culmination. If we look deeper, beyond the
turbulent lives of all these vicious and glamorous royals – Pulakeshin I and
II, Vikramaditya I and II, Loka-Mahadevi, Bhavanaga, Pulakeshi-Raja,
Danti-Durga, Krishna Rashtrakuta – we can see a profound change in the
ambitions and worldview of the peoples of the Deccan. A vast imperial
formation now stretched from the Narmada to the borders of the Tamil
country – a tapestry weaving together rich market towns, multicultural
ports, thriving cities, dense forests, expanding irrigation networks, growing
temples and intensifying political, social and cultural complexity – under
the firm grasp of ruthless and powerful Kannada-speaking emperors.
It was now time for the Deccan to terrify and astonish the subcontinent
and the world.
EHD I, 249.
The only source directly attesting to this event dates to more than a hundred years after it may have
happened, at a time when the Rashtrakutas had an unpleasant relationship with the Chalukyas of
Vengi. Its depiction of the marriage as a ‘rakshasa’ marriage, involving the abduction of the bride,
could have been a fabrication or a literary device to reinforce Rashtrakuta virility – especially
given that is followed soon after by a verse describing how one of them ‘forcefully wrested away
the Fortune of the Chalukya family’ on the battlefield. See EI XVIII, 252.
Michaela Soar, ‘The Tīrtha at Ellora’, in Parimoo, Kannal and Pannikar, eds. 2018, 64.
EI XIV, 127.
There can be little doubt about Danti-Durga’s close relationship with his Chalukya mother (though of
course his apparent closeness to her may be a propaganda device to establish that he was half
Chalukya) considering he would later claim that she donated land in each of the 400,000 villages
he ruled over.
Chakravarti 1941, 56.
Ibid.
Arthashastra 2.2.15–16, quoted in Thomas R. Trautmann, Arthashastra: The Science of Wealth
(Penguin, 2016). ‘Those from the east, Cedi and Karusha, from the Dasharnas and Aparantas are
considered of medium quality among elephants’ (80).
In EHD I, Kosala is depicted as being the last of Danti-Durga’s conquests, but no justification is
given for this supposition.
Thomas R. Trautmann, Elephants and Kings: An Environmental History (University of Chicago
Press, 2015), 65.
Chakravarti 1941, 54.
A man with a regnal name meaning ‘Elephant Fort’ probably made them an integral part of his
strategy. In ancient Indian writing, the fort, durga, was one of the seven constituent elements of the
state. Elephant sculptures are present in large numbers in the great monolithic Krishneshvara
temple built by his uncle and successor, and it is difficult not to see a nod to Danti-Durga here.
EHD I, 254. The authors argue that it is likely Danti-Durga gained the title Prithivi-Vallabha for
helping Pulakeshi-Raja against the Tajikas (though he concedes that this narrative is purely
conjectural). This book presents an alternative hypothesis that better fits the evidence we have. To
reiterate: it is suggested that Pulakeshi-Raja attempted to seize independence, that Vikramaditya’s
second expedition to Kanchi was a response to this, and that Danti-Durga helped Vikramaditya
bring Pulakeshi-Raja into line, being rewarded with the title Prithivi-Vallabha for his services as
well as for his conquests in central India and his Chalukya descent. Note that this does not
necessarily imply Danti-Durga attacked Pulakeshi-Raja immediately – merely that he had
generally won significant victories and may have been acknowledged as an important partner by
Vikramaditya. A decade later, Danti-Durga claimed to have completed the conquest of Lata.
Soar 2018, 64.
Ibid., 64.
Ibid., 66.
Ibid., 65.
Ibid.
Ibid., 63.
Ibid., 64.
Lisa Nadine. Owen, ‘Art, Architecture and Devotional Practice at Ellora’, in Living Rock: Buddhist,
Hindu and Jain Cave Temples in the Western Deccan, ed. Pia Brancaccio (Marg Foundation,
2013), 136.
Ibid.
Though the term ‘Kalachuri artists’ has been used previously to refer to artisans from here who made
some of the early Chalukya temples in Vatapi, perhaps ‘Ellora artists’ would be a better way to do
justice to this influential hub of Deccani sculptural traditions.
Soar 2018, 73.
Ibid.
Ibid., 77.
Lisa Nadine Owen, Carving Devotion in the Jain Caves at Ellora (Brill, 2012), 132.
EI XXV, 29.
Dikshit 1980, 184.
EHD I, 254, and footnote 2.
Dikshit 1980, 170.
There is much debate as to the chronology of Danti-Durga’s conquests and some writers prefer to
place these early in his reign. I think that performing the Hiranyagarbha before the death of
Vikramaditya II would have been a remarkably bold statement for Danti-Durga to make especially
against a relatively well-established overlord – and thus I place it after the accession of Kirti-
Varman II.
Anant Sadashiv Altekar, The Rashtrakutas and Their Times; being a political, administrative,
religious, social, economic and literary history of the Deccan during C. 750 AD to C. 1000 AD
(Oriental Book Agency, 1934), 40.
IA XI, 114.
Kalidasa, Meghadutam, trans. Srinivas Reddy (Penguin Random House, 2015), 1:31.
EHD I, 255.
Inden 2000, 247.
Ibid., 248. Inden argues that this was meant to replace the Ashvamedha.
D.C. Ganguly, ‘The “Gurjaras” in the Rastrakuta Inscriptions’, Proceedings of the Indian History
Congress 3, vol. 3 (1939).
Matsya Puranam 274.65, in B.D. Basu, ed. The Sacred Books of the Hindus Translated by Various
Sanskrit Scholars (The Indian Press, 1917).
Matsya Puranam 274.30–32. ‘Banners of various colours resembling the complexion of the four
Lokapalas should be made.’
Inden 2000, 247.
Ibid.
Bhattacharyya 1975, 72.
Matsya Puranam 275.25.
This reconstruction is conjectural, as are other narratives of the period. However, it seems reasonable
to assume that Danti-Durga’s attack on Lata was after the accession of Kirti-Varman II, as it was a
very direct move against the Chalukyas, no matter Pulakeshi-Raja’s disloyalties. Such an attack
could have forced Vikramaditya II to take the field to defend his own status as a powerful overlord.
IA XI, 114.
Dikshit 1980, 185.
Ibid.
Soar 2018, 67.
IA XI, 114.
V.V. Mirashi, ‘Manor Plates of Rāshṭrakūta Dantidurga: Śaka Year 671’, Proceedings of the Indian
History Congress 21 (1958).
This is, to my mind, the simplest explanation for why the merchant corporation of the Manor plates
(ibid.) would acknowledge a Rashtrakuta governor, called Danti-Durga Prithivi-Vallabha, but not
acknowledge him as maharajadhiraja, paramabhattaraka or the equivalent.
Mirashi 1958.
Whitney M. Cox, ‘Scribe and Script in the Cāl.ukya West Deccan’, The Indian Economic and Social
History Review 47, no. 1 (2010): 19.
Owen 2012, 135.
M.K. Dhavalikar, Ellora: Monumental Legacy (Oxford University Press, 2003), 38.
Carmel Berkson, ‘Daśāvatāra Cave: Its Importance in the History of World Art’, in Parimoo, Kannal
and Pannikar eds. 2018, 167.
Ibid., 172.
Ibid.
Cummings 2006, 214.
Owen 2012, 133, footnote 8.
EHD I, 256, quoting a later Rashtrakuta inscription.
It cannot be ruled out that some of Danti-Durga’s family members, including his much older uncle
Krishna and his ambitious sons Dhruva and Govinda, had something to do with this, as they had
plenty to gain from this sudden tragic death.
According to some sources Danti-Durga had two young sons, but in the absence of a protector they
 appear to have been unceremoniously murdered and are never mentioned again.
EI III, 6.
Ibid., 7.
Cox 2010, 19.
Dikshit 1980, 189.
Ibid.
EHD I, 259.
This may be an allusion to a practice of abducting or raping women who accompanied a defeated
army. EI XVIII, 252.
Inden 2000, 249.
Indeed, this may be the actual conclusion that we are meant to draw from the later Rashtrakuta
inscription, which first mentions the abduction over a hundred years after the actual fact. Perhaps
this is an assertion of virility or adherence to the Kshatriya dharma this later Rashtrakuta,
Amoghavarsha I, wishes to claim on behalf of his ancestors. We cannot even be sure if Bhavanaga
was abducted, given the paucity of evidence. All we can say for certain is that she was, indeed,
Danti-Durga’s mother.
EHD I, 258.
EI XIV, 123.
Ibid., 128.
He was married to a Chalukya princess by the name of Vijaya-Mahadevi, thus making him
Vikramaditya II’s brother-in-law and the uncle of Kirti-Varman II.
G.R. Rangaswamaiah, ‘The Rashtrakuta Relations with the Gangas of Talkad’, in The Rashtrakutas
of Malkhed: Studies in Their History and Culture, ed. B.R. Gopal (The Mythic Society/Geetha
Book House, 1994).
Adiga 2006, 117.
Shivanna 1977.
Rohitha Eswar, Vrushab Mahesh and K. Krishnan, ‘“Iha” to “Para”: A Hero’ s Transitional Journey
to the Other World from This World – Reading Veeragals of Hassan’, Heritage: Journal of
Multidisciplinary Studies in Archaeology 2 (2014).
Adiga 2006, 325.
Ibid., 327–28.
Ibid., 325.
Shivanna, ‘Rashtrakuta Krishna I and Gangavadi (758–774 A.D.)’, Proceedings of the Indian History
Congress 38 (1977); EHD I, 259.
Divya Shekhar, ‘Inscription Stone discovered in Hebbal could be Bengaluru’s oldest’, The Economic
Times, 23 June 2018, https://economictimes.indiatimes.com/magazines/panache/inscription-stone-
discovered-in-hebbal-could-be-bengalurus-oldest/articleshow/ 64708042.cms, accessed 17
December 2019.
EI XIII, 282.
 6
To Kailasha
Winter, 769 CE

With Gangavadi in flames, Vatapi cowed, and Chalukya blood flowing in

the Malaprabha river, Krishna Rashtrakuta returned in triumph to Ellora,
increasingly a centre of Rashtrakuta power.1
We cannot know what may have occupied his mind, as the great army of
the Rashtrakutas and their vassals returned to their homes to enjoy their
new wealth and prominence. But we do know of one idea he had. Krishna
Rashtrakuta, Fortune’s Favourite, needed to prove his newly imperial
dynasty’s authority and dominance in all directions, in all ways. He thus
decided to commission at Ellora a temple of such stupendous scale and
craftsmanship that it would humble all that had come before. This temple
dedicated to Krishna Rashtrakuta’s lord, Shiva Krishneshvara, needed to be
twice as massive as the Lokeshvara at Pattadakal, eclipsing the crowning
glory of the Vatapi Chalukyas. Its iconography would showcase the latest
Shaiva innovations and ideas, thus making it a great engine for the
development of Shaivism, now the Deccan’s royal religion par excellence.
The Krishneshvara would be a radical departure from its predecessors. It
would not be assembled painstakingly from carved and jointed panels.
Rather, it would be excavated straight down a colossal basalt hill at Ellora.
A gigantic mass of rock was to be isolated and carved to make the
Krishneshvara – a building that was really a sculpture, a cave temple of
hitherto unimagined proportions and complexity. In comparison, the cave
temples of the early Vatapi Chalukyas would be damp, unimaginative
grottos, and Narasimha Pallava’s temple monoliths at Mahamallapuram
mere pebbles. But we will return to the Krishneshvara temple soon. For the
moment, suffice it to say that by leaving the architectural, iconographic and
religious ‘arms races’ of the Chalukyas and Pallavas in the dust, it was
intended to proclaim to the subcontinent that the Rashtrakutas were the
wealthiest, most martial, most devoted to Shiva of all south Indian royal
clans; the most deserving to rule the peoples of the Deccan and to claim the
supreme authority and ornate regalia of sovereign emperors.
But the Rashtrakutas were not the only Indian superpowers interested in
making claims to unprecedented supremacy at this time. In Rajasthan,
Malwa and Gujarat, a clan calling themselves the Pratiharas, Keepers of the
Gateways, managed to consolidate power. They previously had been forced
by Danti-Durga Rashtrakuta to serve as his ‘doorkeepers’ in the great
Hiranyagarbha sacrifice that he performed in Ujjain. (‘Doorkeeper’ was a
cruelly appropriate pun on their title.) But now, by slowly expanding their
power towards the fertile Gangetic plains, the Pratiharas were becoming
one of the most successful polities that north India had seen since Harsha’s
death nearly two hundred years earlier; they, like the Rashtrakutas, claimed
the title of Maharajadhiraja.
Indeed, Harsha’s ghost still loomed large over the turbulent north. His
capital, Kannauj, like Delhi a thousand years later, had become the ultimate
prize, the crown jewel that any would-be emperor of northern India had to
control. Commanding a geopolitical crossroads between the fertile lands of
Punjab and the Gangetic plains, armies and adventurers from Kashmir to
Malwa had flocked to Kannauj in the centuries since Harsha’s death. Even
the Tajikas had made an abortive attempt to seize the city before being
routed by one such adventurer.2 But by the eighth century, most rulers of
Kannauj tended to control little more than the vast city itself. They must
have watched with trepidation the growth of powers like the Pratiharas and
others, such as the Karkotas of Kashmir, and far to the east in Bengal, the
Palas. Unlike the Rashtrakuta and Pratihara clans, who came from fairly
well-established military aristocracies, the Palas were almost certainly
commoners. Near-contemporary sources describe them as either belonging
to the ‘menial’ castes, or as being administrators and scribes. Their birth
was no barrier to the growth of their power, however, as the Palas
successfully consolidated the fertile and populous eastern Ganga valley,
declared themselves emperors and began to look beyond. From across these
vast, lush, and urbanised territories, they began to wrestle with the
Pratiharas for influence and control. Kannauj was crucial to both polities:
controlling it would help secure the Pratiharas’ northwest frontier and offer
a useful staging ground for campaigns into the Gangetic plains; to the Palas,
it would secure their control over the vast river system while offering a
route into Central Asian markets.
By the late eighth century, these once-regional kingdoms were, like the
Deccan, emerging as subcontinental superpowers. Their rulers were no
longer interested in merely being acknowledged by the petty rajas and
maharajas of their immediate neighbourhood. Now, their superiority needed
to be recognized by the defeat and conquest of their equals and rivals, the
maharajadhirajas. This, they imagined, would grant them primacy over all
the dynastic polities that constituted their world. In the late eighth century,
as geopolitical tension crackled in the air, these dominant polities of eastern
and western India circled Harsha’s old capital and dreamed of undisputed
supremacy. And, meanwhile, south of the Narmada, the lord of the Deccan
watched them both carefully and made his plans.
Every few centuries or so, at least in the premodern period, a constellation
of factors comes together and an explosion of art follows. We’ve seen this
before with the artistic competition of the Chalukyas and Pallavas, and met
the talented masters Mandhatar of Mahamallapuram and Gundan of
Pattadakal. It is a tragedy that when one of southern India’s most
extraordinary such confluences came together, we do not know the name of
the person or persons who designed the greatest marvel of the time: the
Krishneshvara temple at Ellora, known today as the Kailashanatha temple
or Cave 16.
Let us briefly dwell on this enigmatic building, which to this day inspires
awe and wonder in sightseers, emotions we certainly share with its long-
forgotten eighth-century visitors. Why did this Nameless Sthapati, as we
shall call him, shape the Krishneshvara this way?
This man was most likely a veteran of Ellora’s sculpture and architecture
industry, and may even have worked with Danti-Durga on his innovative
Dashavatara cave temple.3 Ellora before the Rashtrakutas was a small, if
bustling, sacred centre; after the Rashtrakutas commissioned the
Krishneshvara and began to frequent the site, the glamour and wealth of this
medieval imperial court must have reshaped the contours of the town. What
must it have been like for the Nameless Sthapati to meet his new
colleagues, the guilds of northern Karnataka and the Tamil country, that
came to Ellora attracted by Rashtrakuta blood money? What were the
streets of Ellora like when full of Rashtrakuta warriors, rowdy and drunk, or
crowded with impoverished labourers in huts of mud and thatch? Large-
scale excavations are desperately needed across the Ellora UNESCO World
Heritage Site for more information.
Whatever his experiences with Krishna and the Rashtrakutas, the
Nameless Sthapati struck upon a genius design to meet the Vallabha’s
vision. Had he simply doubled the size of the Lokeshvara at Pattadakal and
carved it into a hill, the bottom part of the temple would have been damp
and in perpetual darkness, ruining the spatial experience of
circumambulation and rendering it ineffective. The Nameless Sthapati
solved this by excavating in the rock hill a truly gigantic cavity, leaving
there a temple of humongous proportions. It would stand on a much higher
plinth than ever used in either northern or southern temples,4 an entire
storey by itself, boosting the main structure into the open space of the
excavated cliff and exposing both the ground floor and the first floor to
sunlight. The first floor would carry the main shrine, hall, secondary shrines
and the primary spire – a tiered Dravidian superstructure, the farthest north
this design had ever appeared. The elevated bottom storey, plinth included,
would be decorated with imposing sculptures almost twice the height of a
man, creating a dramatic aesthetic juxtaposing the raw cliff side, towering
monolith and bright sky. The scale of the excavation needed to execute this
plan was unprecedented in the subcontinent: 2 million cubic feet of rock
had to be removed,5 leaving behind a temple that would be one of
the largest constructions of the eighth-century world, and certainly the
largest free-standing structure in the Indian subcontinent at the time.
An undertaking such as the Krishneshvara could only have been imagined
by an individual who had imbibed the knowledge of generations of cave
temple makers, with logistical, sculptural and iconographic solutions for
diverse patrons and religions. It could only have been paid for by the
resources of a vast imperial formation with something to prove. And it
could only have been made with the help of teams of talented priests,
monks and sculptors, coming from rich artistic and religious traditions
across southern India. All of these, which we have observed evolve over the
last few chapters, had congregated in Ellora by the late eighth century.
 But a temple is not merely the work of architect, patron, sculptor and
labourer. As a gigantic curation of religious art, it also requires the services
of the priest. A temple like the Krishneshvara offers hundreds if not
thousands of surfaces for decoration, and there must have been teams of
sthapatis and priests who worked together to curate and assign elements to
each surface. These were layered in striking ways: the roof of the temple,
for example, features four roaring lions facing the ordinal directions
(northeast, southeast, southwest, northwest), their powerfully muscled paws
gripping a large lotus. This was probably meant to reflect the Rashtrakutas’
military supremacy over the rest of their world, and may have had other
meanings that have since been lost. The priests who worked with the
sthapatis were most likely members of the school or monastery that was on
the ascendant at the Rashtrakuta court at the time, and were plugged into
the latest ritual, scriptural and mythological developments of Deccan
Shaivism. The Krishneshvara’s sculptures thus also offer us a snapshot of
Hinduism in evolution in the eighth century, a topic we will return to.
Finally, how did the Nameless Sthapati actually carve out this
architectural sculpture? Contemporary architects tend to think he
deliberately chose a part of the Ellora hills with a relatively gentle incline
and turned it into a ramp. At or near its top, a football field-sized rectangle
was carved into the rock so that it paralleled the ground, creating a ‘floor’
for sculptors to work on, hundreds of feet above ground level. This plane,
including the top of the ramp, was then meticulously carved downwards
towards the ground foot by foot, leaving, at every level, a large block of the
correct size and shape in the middle for sculptors to carve. The slope was
retained till the very end, to dispose of the debris of the carving.6 One can
imagine that extreme care needed to be taken with measurement, alignment,
demarcation and carving to ensure that the design of the temple remained
intact throughout the process. This hints at a powerful grasp of
mathematical, architectural and engineering principles.

The actual execution of the project was a logistical and managerial

challenge unprecedented in the subcontinent, if one does not
count military campaigns. The Krishneshvara’s main designers and
executors were almost certainly a multi-ethnic, multilingual, highly
experienced and self-confident group of individuals. They had to
figuratively dive downwards into the black rock, and trust that they could
collectively deal with the challenges that were sure to arise7 in the years to
come as they fashioned this 150-foot-tall monolith. And so sometime in the
late 760s, this group of people set out to make a wonder of the world.
The excavation of the temple must have been an impressive sight. As the
rough surfaces of the temple emerged, level by level, foot by foot, ever
larger teams of master sculptors were brought in to work on more surfaces
simultaneously. Meanwhile, toiling labourers removed chunks of rock
around them in layers, rolling them down the gentle incline of the hill. No
scaffolding was needed for the sculptors or the debris removal8 as the
temple gradually took shape, every level being finished before the next was
excavated. We can imagine that arguments and fights broke out as
tempestuous master sculptors declared that they would do things their way
and no other. Or, perhaps, new friendships and loves blossomed as they
delighted in the creativity of their peers, and Ellora resounded with the
sound of laughter and feasts and weddings between clans who would never
otherwise have met.
By the time the Krishneshvara’s sculptors finally reached the bottom
floor, the project had gathered together the greatest sculptors of the
subcontinent and kept them rich and well fed and busy for around twelve
years, allowing them to learn from and compete with each other. Together,
they unleashed a flourishing of artistic ability9 that is among the most
extraordinary legacies of the medieval Deccan. We will return to see this,
alongside the Rashtrakuta kings who commissioned it, towards the end of
this chapter.

Ellora’s architects were not the only Deccanis busy with grand and world-
shaping deeds in the late eighth century. In 769, soon after brutally forcing
the Gangas to submit, Krishna Rashtrakuta made sure that the Chalukyas of
Vengi, the only royal Chalukyas left standing, could not make their own bid
for sovereignty in the Deccan. His sons were ordered to lead an army into
the fertile coastal lands of Andhra.10 This pair, Govinda and Dhruva
Rashtrakuta, Danti-Durga’s cousins, had probably been partners in the
project of overthrowing the Vatapi Chalukyas from its very inception. But
though the Rashtrakutas’ sovereignty was very much a family enterprise,
like most medieval polities tended to be, they were not directly in the line of
succession in Danti-Durga’s lifetime. At best, they might have expected a
life as generals or provincial governors under Danti-Durga and his heirs.
The young Rashtrakuta king’s sudden death, and his uncle Krishna’s
successful bid for the throne, had led to an abrupt change of fortune.
By 769, the brothers almost certainly had quite a reputation in south
Indian circles thanks to the sudden shock of their defeat of the Vatapi
Chalukyas. The younger, Dhruva, was a ruthless warlord, who preferred to
be called Kali-Vallabha, Strife’s Beloved.11 The Vengi Chalukyas, despite
their vaunted ancestral name, very astutely decided not to test him and
instead threw themselves on the mercies of Dhruva’s gentler elder brother,
Govinda.12 Soon after,13 Dhruva, who already had two sons14 by an earlier
wife, was married to the Vengi Chalukya princess Shila-Mahadevi. She was
almost two decades his junior, being in her late teens.
The Vallabha Krishna Rashtrakuta would not live to see the completion of
his magnificent temple.15 His death in 773, followed by the accession of
Govinda II, gave these newlyweds an opportunity to seize power. Dhruva
was appointed governor of Nashik and Khandesh,16 two critical territories.
However, it seems he began to put out word that Govinda had given himself
to a life of pleasure and debauchery, and was getting too friendly with the
vassals of the Rashtrakutas, instead of devoting himself to his royal dharma
and strengthening the kingdom through war.17 It does seem that the Gangas
were already preparing a challenge to Rashtrakuta power, or that the
political situation was otherwise sensitive enough that many grandees saw
Dhruva as offering better solutions than his elder brother. After some deft
politicking, Dhruva attempted to issue his own independent land grants,
without referring to the authority of Govinda.18 This challenge provoked an
immediate response from Govinda, who stripped his younger brother of his
titles. Now claiming to be the aggrieved party, Dhruva rebelled and led a
coup d’état against the Vallabha. Govinda II, like so many who had held the
throne of the Deccan before him, vanished into darkness.
In 780, Dhruva Rashtrakuta and his wife Shila-Mahadevi were ritually
crowned.19 Shila-Mahadevi, the young Chalukya princess from Vengi
whose Vatapi relations had been overthrown by the Rashtrakutas, seems
nevertheless to have convinced Dhruva to allow her a degree of prominence
similar to that of Loka-Mahadevi, her distant Kalachuri predecessor as the
Deccan’s most powerful queen. Soon after his coronation, she was probably
also invested with a magnificent coronet called the patta-bandha, thus
attaining the enviable status of patta-mahadevi, the highest possible status
for any Indian queen still associated with a king20 (queens who ruled in their
own right – of which we have examples from medieval Odisha and
Kashmir – preferred to simply use the male title of maharaja or
maharajadhiraja). A patta-bandha, possibly adapted from earlier Persian
styles,21 was a ‘fillet or band-like piece of silken cloth or gold which was
fixed around the head and displayed … on the forehead’.22 This investiture
was part of a consecration ritual of its own, which conferred on the queen
‘either the gift of land, sovereignty or other powers’.23 By awarding Shila
such a status, Dhruva was essentially proclaiming her as his equal partner in
the cosmic powers he had attained through various coronation rituals.24
The new Rashtrakuta emperor now set out to perform a digvijaya, the
Conquest of the Directions or Quarters, the greatest possible achievement
for any warlike dharma-upholding crowned monarch. It was the act by
which he could reconstitute the circle of kings around him, the raja-
mandala, the network of medieval India’s many violent and competing
polities, into ‘an imperial formation, a single polity’.25 His armies would
subjugate the Four Quarters, and loot and conquer and reorder them as his
position as Fortune’s Favourite demanded. Over the next decade, Dhruva
would begin the process of expanding the Deccan’s influence to proportions
it would not reach again until the rise of the Marathas, a thousand years
later. In order to do this, however, Rashtrakuta power needed to be secured
once again against any challengers. It is difficult to reconstruct this process
in detail, but it seems that by 786/87, Rashtrakuta forces had completed
raids into southern Karnataka and Tamilakam, brutalizing what remained of
the unfortunate Pallavas. One Rashtrakuta poet had a picturesque way of
putting it: the Pallava king surrendered as he saw he was trapped between
the two oceans that were the Rashtrakuta army and the Bay of Bengal.26
On 27 September 786, the day after a total solar eclipse, the empress
Shila-Mahadevi, having bathed and purified herself, ordered her personal
officers to execute a land grant to some Brahmins. She, like Dhruva, seems
to have travelled around with her own personal court, her ladies-in-waiting
– the wives or daughters of vassal kings – as well as administrators, tax
collectors, messengers, scribes and panegyrists.27 On this brisk September
day, she ordered the engraving of a powerful prashasti, allowing us a
glimpse of this eighth-century queen’s relationship with her husband and
the marvellous possibilities of the Sanskrit language which aristocrats like
herself used. The empress announced:
As (Dhruva) subjected forts which were the essence of the three worlds, augmented his fame by
stopping the flow of the Ganga, and made his own the prosperity of the [Pallava] ruler exalted
by the bull-banner, he alone in this world displayed the quality of Parameshvara [Paramount
Lordship] clearly and powerfully.28
Though the Sanskrit double meanings have been lost in translation, the
subtle parallel between Dhruva and the god Shiva is apparent. The ‘flow of
the Ganga’, the river entwined in Shiva’s locks, is an allusion to how
Dhruva had entrapped and stopped the Ganga family. The defeat of the
Pallavas, who used the bull banner, added to his glory.29 Only her world-
conquering husband was worthy of being a true king, worthy of
paramountcy, said the young empress. It would seem that Shila was not
exaggerating. By 788, the Rashtrakuta royal pair finally crushed the Gangas
and imprisoned the young Ganga king.30 Dhruva’s eldest surviving son was
appointed the ruler of Gangavadi. His father-in-law ruled in Vengi, securing
his eastern flank. The roaring waves of the Arabian Sea secured his west.
The Indian subcontinent south of the Narmada river was now pacified: the
Gangas, Cholas, Cheras, Pandyas and Vengi Chalukyas had all submitted,
as had hosts of minor kings and chiefs. Dhruva had come very close indeed
to a successful Conquest of the Directions, one of the duties of the ‘ideal’
ruler. But this was hardly enough for this ambitious and power-hungry
warlord, perpetually on the hunt for opportunities to expand and aggrandise
himself.

In the late 780s,31 when Dhruva Rashtrakuta’s position in the Deccan was
finally secure, a tiny event set off a chain reaction leading to an explosive
confrontation between medieval India’s superpowers. The kingdom of
Kashmir – a pugilistic and culturally sophisticated Himalayan power –
raided Kannauj, imprisoning its king. As the Kashmiris attempted to
ransom him, the Pratihara emperor Vatsaraja advanced to the city and set up
his own candidate on the throne; he thus announced that the Pratiharas were
claiming the sovereignty of north India, and that they saw the fertile
Gangetic plains as their rightful sphere of influence.
The Palas immediately retaliated by choosing their own candidate to the
throne of Kannauj. The Pala emperor, Dharmapala, moved west to install
him at the head of an army of his own, but Vatsaraja Pratihara was ready
and waiting. The Palas were defeated – though not conclusively – and
driven away. The two white imperial umbrellas which were symbols of Pala
sovereignty were gleefully seized by Vatsaraja, whose claim to north Indian
supremacy now seemed indisputable.
Meanwhile, Dhruva Rashtrakuta had been carefully observing these
events unfolding hundreds of kilometres away, and saw an opening. The
overconfident Vatsaraja had left his southern flank, Malwa, completely
exposed – a critical strategic error. We do not know the precise details of
what followed. Did Dhruva initially set out only to conquer Malwa, or was
he always planning on securing a symbolic submission from the two
arrogant north Indian monarchs? How much did he know of what had
happened between Vatsaraja and Dharmapala, and how did he plan his
logistics and route of advance to confront them? Did his army consist of
Rashtrakuta forces alone, or did vassal kings – such as his father-in-law, the
Chalukya king of Vengi – join in?
All we know for certain is that the Rashtrakuta Vallabha did indeed march
to war. Rituals were performed beforehand, astrologers consulted, plans
made, and a great army of infantry, elephantry and cavalry was gathered
and ordered to march. Along with the troops followed thousands of
labourers, cooks, physicians, dancers, musicians, courtesans, poets,
panegyrists, priests and preachers,32 the army stretching out for many
kilometres, shaking the Earth, advancing like an immense, noisy, colourful
serpent. Harsha, Lord of the North, had so fatefully crossed the Narmada
into the Deccan almost exactly one hundred and eighty years earlier. Now
Dhruva, Lord of the South, crossed the same Narmada from the Deccan into
north India. How the tables had turned: the time had come for the Deccan,
which once trembled at north India’s might, to terrify that ancient seat of
empires.
It would not have been long before the Pratihara emperor Vatsaraja,
concerned about the loss of his core territories, rushed to confront Dhruva.
The precise location and date of their collision are unclear, but it is likely to
have occurred somewhere in northern Malwa, perhaps near Vidisha or
Jhansi, before Dhruva could erupt into the Gangetic plains. It was probably
one of the most spectacular set-piece battles of the medieval period: a head-
on collision between two almost equally brash and untested Indian imperial
powers, two equally power-hungry warlords.
What might this climactic battlefield have looked like? We might see two
immense armies arrayed to face each other as huge drums are beaten,
trumpets and conch shells blown in a ‘tumultuous uproar’.33 Perhaps the
might of southern India was gathered there that day: men from the many
kingdoms and provinces of Maharashtra, Karnataka, the Konkan, Andhra
and Tamilakam. The vast majority were certainly on foot, while some were
mounted on elephants from dense jungles, others on horses from Arabia or
Persia across the ocean. Mercenaries and troops of elite aristocratic
families, in clean, organized lines, may have rubbed shoulders with crowds
of levied farmers, pastoralists and artisans. All must have been gathered
there that day with banners and standards and round shields painted with
animals and flowers, long hair tied into topknots and side knots, loins
girded with bright cotton cloth.
 A description of what may be the ferocious infantry maintained by the
Rashtrakuta imperial household survives today: their wavy black hair was
fastened with vibrant cloth bands tied around their foreheads, their bare
chests draped in ‘triple necklaces made of many-coloured beads’,34 and
their forearms, up to the elbows, covered in thick iron bracelets. Their legs
were covered in loincloth tucked around their knees, a band around the
waist held a dagger, and their hands clasped their weapon of choice –
anything from a mace to a sword. Rashtrakuta murals from the eighth and
ninth centuries also depict light cavalry: sword-wielding riders wearing
Persian caps and armour made of overlapping scales, riding expensive
imported steeds festooned with ribbons.35 Such warriors were organized in
complex, overlapping hierarchies – language, region, overlord, caste and
many others – culminating in Dhruva himself, a stocky, muscular old man
with a hairy chest36 seated on a mighty elephant. The aristocrats and
generals surrounding Dhruva must have watched their opponents silently
amidst a host of banners, including the imperial pali-dhvaja, the Banner in
Rows, featuring the Rashtrakuta Garuda and other glittering emblems.
Eventually, the pataha and dhakka37 drums broke into a roar like the ocean
in a raging storm, while embroidered and painted flags and banners snapped
and waved.
Studies of medieval warfare suggest that the two vast armies then
advanced slowly, watching each other for the slightest gap or vulnerability.
When they came within range of each other, flocks of barbed arrows were
launched, and then javelins and stones, continuing for a few hours to
attempt to wear down the enemy and score a few easy kills. The first
screams of pain and the first cries for help were heard, the first spatters of
blood appeared. The infantry then advanced steadily to clash, building up
momentum like a huge wave, attacking each other in compact rectangular
or square38 formations, pushing and shoving at each other, stabbing and
chopping at any exposed individuals or body parts.39 This lasted for about
20 minutes,40 after which both sides withdrew for a brief rest, screaming
insults at the other, flinging missiles, and dragging back their dead and
wounded – as well as any well-adorned enemy corpses that could be looted.
They then charged again when sufficiently prepared. This cycle played out
over many hours: battles were long-drawn-out affairs.
Here and there, we might see small elite squadrons of war elephants or
cavalry charging at each other, opening up into ‘alleys’ so that their riders
could duel. In other areas, bloodthirsty elephants may have smashed into
and trampled down crowds of men, routing them and leaving them to be cut
down by Rashtrakuta light cavalry. Among these daring and deadly warriors
may have been Dhruva and Shila-Mahadevi’s son, the heir apparent
Govinda, who would later rise to the Rashtrakuta throne as Govinda III and
repeat his father’s military exploits.
The screams of maddened elephants, of horses whose bellies had been
punctured by sharp spears, and of dying men missing limbs or drowning in
pools of their own blood, filled the skies. Every now and then, the
drumbeats might change tempo, a horn would be heard through the
madness, and one unit would withdraw, only to be replaced by another. The
average foot soldier knew little more than chaos, dust, blood, pain, terror
and the thrill of a kill, of survival, of booty. Tactics and strategy were
reserved for the men sending them to their deaths sitting on elephants well
behind the front line.
As the battle wore on, the neat formations seen at dawn probably turned
into a bloody, confused mess. Huge clouds of dust suffocated the
increasingly exhausted warriors and animals as Dhruva and Vatsaraja sent
them to their deaths in their contest for supremacy. But more than either
emperor, the outcome of any battle was determined by the force of crowd
psychology. Which side could convince people on the other that they would
be slaughtered if they didn’t run away? On that day, the answer to that was
the Rashtrakuta side: the collapse of the Pratihara forces appears to have
been sudden, not a planned retreat in any way (even if it started as such).41
The ferocious Deccan army pressed the advantage as panic spread through
the north Indian army, and a fatal rout set in. For Dhruva to have a free hand
in Malwa or a secure road onward to Kannauj, he could leave absolutely no
Pratihara forces intact, and there seems to have been a brutal slaughter.
Vatsaraja was chased away, leaving behind him the dust of the battlefield,
the dust of his ambitions, and the two precious imperial umbrellas he had so
gleefully seized from the Pala emperor.42 The Pratihara fled deep into the
deserts of Rajasthan,43 where he was safe but also too distant to be anything
more than a nuisance to the victorious Deccanis until his shattered forces
were able to rest, recover and regroup. Dhruva, however, had no time to
lose. Perhaps it only now dawned on him that such a decisive victory
offered him a way to claim dominance over the entire subcontinent.
When the Rashtrakuta army, victors over the Deccan, victors over the
south, victors over the west, finally burst on to the Gangetic plains – where
the subcontinent’s mightiest empires had risen and fallen for a thousand
years already – they found one last challenger waiting. The Pala emperor
Dharmapala may not have expected Dhruva to defeat Vatsaraja so
decisively, and seems to have planned to seize Kannauj while the two were
at each other’s throats. He may even have entertained ambitions of
reclaiming his lost umbrellas, which Dhruva had now brought with him
after seizing them from
Vatsaraja; perhaps he thought the southern Indian army would be too tired
to mount any sort of effective resistance. He was in for a rude shock: not
only did Dhruva successfully drive him off, but he also relieved the Pala of
the two new umbrellas he had brought to the battlefield!44
The capture of all these umbrellas might seem trivial to us, but it was
hugely significant in the symbol- and status-obsessed worlds of medieval
Indian courts. An imperial parasol, gleaming white, jewel-encrusted, was as
representative of a ruler’s sovereignty as his crown. Dhruva’s victories and
capture of the imperial regalia of the Pratiharas and the Palas signified that
he was now their overlord – which technically made him paramount
sovereign of all India. In the space of barely a year, this second son, this
king’s cousin, Fortune’s Favourite, Strife’s Beloved, had emerged as the
most powerful ruler in the entire Indian subcontinent. In doing so, a Deccan
ruler had now become one of the medieval world’s most powerful
individuals.
 And so the Rashtrakuta Vallabha, his Conquest of the Directions
complete, his dominance over the subcontinent symbolically and violently
established, returned to the Deccan to celebrate his triumph. With him went
the treasuries of the Pratihara and Pala emperors and many new additions to
the Rashtrakuta pali-dhvaja. To the high central flagstaff mounted with the
image of ‘Garuda, the eagle, Vishnu’s emblem’45 were tied in each cardinal
direction, at heights corresponding to their rank, rows of banners of the
kings of the subjugated directions of the Rashtrakuta world. The Pandya
fish, the Pallava bull, the Chola tiger, the Chalukya boar and the Chera bow
of the south now shared space with the Pratihara Lakshmana (brother of the
god-king Rama), and the Pala Tara (a Buddhist goddess),46 representing the
extraordinary prowess and military dominance of Fortune’s Favourite. But
now all these were shifted to the side in favour of two resplendent new
additions. Henceforth the Rashtrakuta pali-dhvaja would also carry a
banner symbolizing the Ganga and the Yamuna rivers, a reminder, a boast,
of overlordship over the whole of north India.

In the year 792 the Rashtrakuta imperial family may have held a festival at
the Krishneshvara.
The monumental temple, by now finished, was oriented almost perfectly
to the west.47 Anyone standing in front of it in the darkness of the early
morning saw the sun rise from behind the magnificent black cliffs like the
temple’s halo, ascending to a great din of clanging bells and throbbing
drums. Soon, crowds may have started to appear, thronging into the dozens
of cave temples at Ellora, staring at the murals, silently praying, giving gifts
to their gods, asking for blessings. As the sun climbed higher, Ellora’s cliffs
must have grown noisier and noisier. In rowdy dirt arenas, animals fought
and wrestlers wrangled as bets were made and audiences cheered; there
could have been plays, dances and musical performances as well.48 Princes
and ministers wearing patta fillets and coronets came with their retinues,
preceded by the pancha-maha-shabdas – the Five Great Instruments, the
horn, conch, drum, victory bell and something called a ‘tammata’ (whose
meaning is unclear).49 These were granted exclusively to the highest ranks
of the Rashtrakuta imperial network, and were loudly sounded to warn the
crowds to stay out of the way of these lords of the great houses.50 If they
were important enough, they would have had a small coloured parasol of
their own, and perhaps one or two attendants with a chauri, the fly whisk
made of a yak’s tail.51 Sometime during the day, the emperor Dhruva
himself may have put in an appearance, in the company of vassal kings,
ministers, chamberlains and attendants. His splendour eclipsed all his
subordinates, marking him out as many times wealthier and more
prominent. Dhruva might have looked in many ways like the classical
Indian king we imagine; in other ways, he certainly was not.
Above his head was held aloft a huge parasol of white silk, adorned with
gold and precious gems; on his head was a tall, heavy gold crown, perhaps
encrusted with rubies and carved with fantastic creatures; his chest was
covered with thick necklaces of pearls, gold, jewels, and fragrant garlands
of rare flowers; his mouth was probably red with exotic betel.52 Behind him
was a crowd of gorgeous female attendants in the finest of clothes, fanning
him with chauris, fluttering around his shoulders like birds.53 Shila-
Mahadevi may have been present, attended by the wives of Dhruva’s
vassals,54 and her sons, the heirs,55 attended by the sons of Dhruva’s vassals.
The Nameless Sthapati, his head covered with a turban and fillets of
honour, was in all likelihood also part of this parade. All these people were
probably resplendent in festive finery, with elaborate garlands and coiffures
and make-up.
The crowds must have cheered, clapped and gawked at the procession,
drinking in the sight of the tall, well-fed men and women, the handful of
elite, sophisticated, ruthless families that ruled them now and would do so
for generations after. They must have almost seemed to have descended
from a heavenly world, perfumed water sprinkled before them, fans waving,
music playing and women dancing before their splendid parade in the
sweltering heat. Though the masses were unlikely to have been allowed into
the priceless Krishneshvara, worth hundreds of millions of dollars in
today’s currency, Dhruva’s retinue were welcomed obsequiously into its
hallowed portal. They would immediately have been struck by the
remarkable experience, the tangible presence of divinity, that the Nameless
Sthapati’s design unfolded before their eyes.56
 Let us join them as they step through the shadowed gateway of the
monolithic temple. They see a sunlit panel welcoming them in, a relief of
that same Royal Fortune whom Krishna Rashtrakuta had supposedly seized
on the battlefield: the goddess Sri or Lakshmi, the emblem of Indian
kingship, accompanied by royal elephants showering her with water –
signifying the ablution that accompanied the royal consecration. She is
seated on a field of lotus leaves, which the sculptor cunningly carved to
give the illusion of perspective and depth to the relief: the pads at the
bottom are larger than the ones higher up, thus drawing the viewer’s eyes
towards the goddess. From darkness to light, the goddess seems to say.
From the mundane world to the divine world.57 Dhruva would then have
taken a left into the sun-drenched courtyard. There stands an immense
statue of an elephant, carved by Tamil sculptor guilds into a figure of
unmatched ‘grace, volume, and swiftness of movement’58 to delineate the
space59 and act as a nod to the creator of the Rashtrakuta family’s fortunes:
Danti-Durga, the Elephant Fort. On Dhruva’s left were stark cliff faces
decorated with caves for worship, vanishing into the darkness of the living
rock, decorated with torches and garlands. Above and around him were
thousands of tonnes of sacred stone, suspended as if defying the laws of
gravity, and above that the sapphire-blue sky. And we can imagine that
Dhruva couldn’t tear his eyes off what was emerging on his right: the
spectacular temple proper, rising like the peak of the sacred mountain
Kailasha, painted white and decorated with vibrant colours. Here was the
snowy abode of Shiva and his raucous retinue, recreated in all its glory, as
though the Rashtrakutas had brought the heavens themselves to the
Deccan,60 along with the Ganga and Yamuna banners.

Here the retinue could have stopped to admire a relief of Durga slaying
the buffalo-demon Mahisha, a representation of the indomitable, terrifying
energies of the Divine Female61 with her weapons arrayed around her like a
halo of blades. There they may have shuddered at the titanic carving of
Shiva in his horrific form as the Slayer of the Elephant Demon. Everywhere
they saw potent combinations of political and religious iconography,
entirely new forms which are present only here at the Krishneshvara. For
example, the plinth is made up of dozens upon dozens of powerful
elephants
and lions in combat, evidently made by sculptors who had seen the beasts in
action. They leap out of the monolith at the worshipper, bearing the weight
of the imposing artificial mountain above, lifting it into the air and light.
They could also be interpreted as an indicator of the importance of the royal
war elephant to the Rashtrakutas,62 and another nod to Danti-Durga.

In the cliff sides on the ground floor were galleries cut under the line of
the rock, which artificially increased the space of the courtyard and created
a pillared colonnade through which light filtered into tall, framed
sculptures. Here, the royal retinue may have admired a sculpture depicting
Ravana sacrificing his ten heads to Shiva for unbounded power. Its quality
may not have been comparable with some of the others in the temple, but it
was probably interesting to any initiated Shaivas in the group because what
it depicted was actually fairly new at the time.63
 In literary works completed centuries earlier, such as the Ramayana and
Raghuvamsa, Ravana is depicted as sacrificing his heads to Brahma, the
Creator, in return for magical powers.64 Yet in literature composed a few
centuries after the Krishneshvara was excavated, such as the Shiva Purana
of eleventh–twelfth-century Varanasi, they narrated a tale of Ravana
sacrificing his heads to obtain powers from Shiva instead, making him out
to be an ideal Shaiva devotee – just as the panel in the Krishneshvara
does.65 What this suggests is that we are seeing Shaiva theology and myth in
motion, captured on the rock of the Krishneshvara. Similarly, scholars have
noted that the Krishneshvara’s depictions of the lives of the hero-gods
Rama66 and Krishna67 are not what we see in the classical texts, but seem
instead to reflect contemporary south Indian narratives of the myths, which
Dhruva Rashtrakuta and his retinue were probably most familiar with. The
Krishneshvara temple is thus not only a political or artistic achievement: it
is an invaluable historical artefact which could tell us a great deal about the
evolution of Shaivism and Indian religions in the Deccan. It challenges our
stereotype of unchanging Indian rituals and myths with a history where
priests, kings and communities instead actively participated in making and
remaking them.
Moving away from this gallery, perhaps Dhruva’s retinue next climbed up
the stairs into the main hall, seeing its murals illuminated by flickering oil
lamps, as gold leaf and semi-precious stones glittered in the dark. And there
in the heart of the temple, in a cell blazing with light, was the splendid
manifestation of Shiva Krishneshvara himself, a linga decked with ‘rubies,
gold, and all other precious things’.68 The victorious Dhruva must have
heaped it with jewels and prostrated himself to the chanting of priests and
clanging of bells, decorating with the bloody spoils of war this
representation of the awesome power of human creativity and
determination.
At sunset an observer on the other side of Ilapura’s cliffs could have
watched the sun’s red glow vanish behind the cliffs to the west, only to be
replaced by the faint glow of the temple’s lamps, as if the sun ‘had
descended of its own accord’.69 Day and night, it blazed like the
subcontinent’s crown jewel, ‘billowing smoke and incense … courtyards
sprinkled with pure scented water’70 – a heady, intoxicating, overwhelming
religious experience. According to later inscriptions, gods flying above it
stopped to stare in astonishment and concluded that a creation of such
beauty could not have been the artifice of mere human hands but was a self-
generated manifestation of Shiva himself.71
What remains now of Krishna and Dhruva Rashtrakuta except the dry
testimonies of prashasti propaganda, that leave no trace of their humanity
and complexity? What remains of their conquests, of their boasts, of their
politics and their betrayals? Where is the immortality and glory they were
supposed to have earned by killing thousands?
But what the Nameless Sthapati created – that will stand till the end of
time.
Where the capital of the early Rashtrakutas was is a matter of considerable debate. Though
inscriptions show us that later Rashtrakuta kings ruled from Manyakheta as well as a host of other
royal cities, there is less evidence to show where their power was concentrated immediately after
the overthrow of the Vatapi Chalukyas. Considering the scale of the investments made in Cave 16,
the Krishneshvara, it is fair to assume that Krishna Rashtrakuta frequently visited Ellora once he
had secured his grip over the Deccan.
This would be Yashovarman of Kannauj, c. 725–52 ce.
Deepak Kannal, ‘The Regional Lineages and Possible Masters at Kailāśanātha Temple of Ellora’, in
Parimoo, Kannal and Pannikar eds. 2018, 205.
Kannal 2018, 204.
M.K. Dhavalikar, ‘Kailasa: The Stylistic Development and Chronology’, in Gopal ed. 1994.
Kannal 2018, 203. Professor Kannal discussed his ideas with the author in Bengaluru in 2019. While
the book presents a further developed version, the credit for the reconstruction should go to him.
Kannal 2018, 203.
H. Goetz, ‘The Kailasa of Ellora and the Chronology of Rashtrakuta Art’, Artibus Asiae 15, no. 1/2
(1952).
Kannal 2018, 206.
EI VI, 212.
EI X, 84. Some translators present this as ‘Beloved of Warriors’.
EI VI, 212. Though Govinda’s inscriptions record that he had led a ‘victorious army’ into Vengi, they
do not detail a military confrontation. It seems likely that if there had been such a confrontation,
the crown prince would hardly have neglected to mention it.
Many historians put the date of the marriage to after Dhruva’s accession and supposed attack on
Vishnuvardhana IV in 780. This is questionable because it doesn’t seem likely the Chalukya
princess Shila-Mahadevi would be accepted as Dhruva’s chief queen and equal partner after he had
just defeated her father.
These would be Karkka and Stambha. Stambha was old enough to rule as a viceroy in Gangavadi
very soon after Dhruva’s bid for the throne in 780, which means he would have been a child – at
least – in 770. His elder brother, Karkka, predeceased Dhruva.
Kannal 2018, 206. Professor Kannal argues that there is no evidence supporting the idea that the
entire temple was excavated during the reign of Krishna I. This book therefore depicts it as having
been completed sometime during the reign of Dhruva.
Shivanna, ‘An Estimate of Dhruva’, in Gopal ed. 1994. This is an excessively rosy estimate of the
man given how little evidence of his personality actually survives – what does come across are
ambition and ruthlessness, which is how this book portrays him.
EI IX, 194–97: ‘His younger brother … who, on perceiving him to be self-conceited ... and even
devoid of policy ... assumed the royal authority ... in order that sovereignty might not deviate from
the family.’ It is difficult to take this claim at face value given that it was issued by a descendant of
Dhruva’s, who can hardly be expected to be honest about how his line rose to power. It certainly
seems, however, that Dhruva’s public explanation for his usurpation was his brother’s debauchery.
EI VI, 212. See also EI XXII, 180.
EI XXII, 99.
Ali 2006, 119, footnote 57.
Ibid., footnote 56.
Ibid., 118–19.
Ibid., 119.
This may also be a recognition of the role she played in manoeuvring him on to the throne in the first
place, perhaps with the support of her father Vishnuvardhana IV of Vengi.
Inden 2000, 261.
EHD I, 263.
EI XXII, 101.
Abridged from EI XXII, 104.
Ibid., footnote 2.
Shivanna 1994.
In this matter the book follows the narrative of Altekar (1934). A later date makes sense since: (a) it
would have made no strategic sense for Dhruva to head north, far away from his domains, unless
he was confident there would be no deadly attacks from the Gangas;
(b) Shila-Mahadevi would hardly have neglected to mention something this important in her 786
inscription; and (c) it was only in 788 that Dhruva imprisoned the Ganga king Sivamara II, thus
 freeing up resources for a northern campaign.
Chakravarti 1941, 93–103.
Ibid., 122.
K.K. Handiqui, Yaśastilaka and Indian Culture (Jaina Samskriti Samrakhsana Sangha, 1949), 59.
S.K. Bose, Boots, Hooves, and Wheels: And the Social Dynamics Behind South Asian Warfare (Vij
Books, 2015), Electronic Edition, ‘Warfare of the Late First Milennium’.
EI IV, 347. While the description of Dhruva is likely just poetic licence, its details are nevertheless
interesting.
EI IV, 334–37.
Both art historical evidence (in the form of hero stones) and literary evidence (in sources such as the
Arthashastra and Nitisara) depict highly organized formations, though literary sources describe
somewhat impractical arrangements of troops.
For a study of the experience of battle in the front lines, see John Keegan, The Face of Battle: A
Study of Agincourt, Waterloo and the Somme (Random House, 2011), and Philip Sabin, ‘The Face
of Roman Battle’, The Journal of Roman Studies 90 (2000).
See Sabin 2000.
If it was a planned retreat, Vatsaraja would not have lost the Pala umbrellas to Dhruva.
Davidson 2002, 53.
EHD I, 264.
Davidson 2002, 53.
Inden 2000, 250. This is a description of Dhruva’s son Govinda III’s pali-dhvaja, but it should be
fairly similar to Dhruva’s considering they fought the same enemies.
Ibid., 251.
M.N. Vahia, A.P. Jamkhedkar and Parag Mahajani, ‘Astronomical Orientation of Caves in Ajantha
and Ellora’, Journal of Multidisciplinary Studies in Archaeology 5 (2017).
Mishra 1992, 87.
Altekar 1934, 263.
Ali 2006, 118.
Ibid.
Betel, as a marker of courtly sophistication, would grow much more important in Deccan in the
following centuries, but scattered references to it can be found from earlier.
EI IV, 334–37.
Ali 2006, 116.
While it is unclear whether the future emperor Govinda III was Shila-Mahadevi’s son or not, it is
highly likely. Govinda’s selection as heir apparent may have been because he was the grandson and
nephew of the present and future kings of Vengi, and thus most likely to be able to control the vast
Rashtrakuta political network.
Roger Vogler, The Kailas at Ellora: A New View of a Misunderstood Masterwork (INTACH
Aurangabad and Mapin Publishing 2015), 18.
Ibid., 27.
Kannal 2018, 208.
Vogler 2015, 31.
This will be explored in detail later. The ideas come from Inden (2000) and Diana L. Eck, India: A
Sacred Geography (Harmony, 2012).
Sudhir Kakar, Intimate Relations: Exploring Indian Sexuality (University of Chicago Press, 1989),
 138. Quoted in Vogler 2015, 42–43.
K.V. Soundara Rajan, The Ellora Monoliths (Gian, 1988), 90. Quoted in Vogler 2015, 31.
Benjamin J. Fleming, ‘Ellora Cave 16 and the Cult of the Twelve Jyotirlingas’, in Living Rock:
Buddhist, Hindu and Jain Cave Temples in the Western Deccan, ed. Pia Brancaccio (Marg
Foundation, 2013), 118.
Ibid., 119–20.
Ibid., 120–21.
Stephen Markel, ‘The “Rāmāyaṇa” Cycle on the Kailāsanātha Temple at Ellora’, Ars Orientalis 30,
Supplement 1 (2000).
John Stratton Hawley, ‘Scenes from the Childhood of Kṛṣṇa on the Kailāsanātha Temple, Ellora’,
Archives of Asian Art 34 (1981).
IA XII, 228–30.
EI IV, 334–37.
Vogler 2015, 47.
IA XIII, 228–30.
 7
King of Poets
Summer, 814 CE

It is barely thirty years since Dhruva Rashtrakuta’s splendid military

triumph in the Gangetic plains and the completion of the Krishneshvara
temple. But the Rashtrakuta imperial network is already at risk of
collapsing.
The lords of the sixteen1 huge principalities and dozens of smaller states
which together made up the Rashtrakuta polity had gathered in a temporary
town2 built to accommodate them on the outskirts of the still new
Rashtrakuta capital Manyakheta,3 modern Malkhed,4 near Gulbarga. Some
of these mighty aristocrats came from other Rashtrakuta lines. Many more
came from established local dynasties with their own unruly vassals,
ambitions and claims. These included the Gangas and the Vengi Chalukyas.
All of them knew very well what magnificent treasures Manyakheta
contained, and they knew these treasures were, all of a sudden, within their
grasp. After all, they were there to witness a fourteen-year-old, a mere boy,
being crowned their lord and master.
His name was Sarva Rashtrakuta, and he had once seemed a child of
destiny. He had been born in the town of Sri-Bhavana, nestled within the
ancient Vindhya hills, during the monsoon season of the year 800. His
parents were Dhruva’s son, the dashing young5 Vallabha Govinda III
Jagattunga, ‘Pinnacle of the World’, and his queen Asagavva. The prince
was born in the immediate aftermath of a series of brilliant Rashtrakuta
campaigns, almost as if he were a reward from the gods for the dynasty’s
upholding of the royal dharma. His father Govinda III had crushed a
coalition of twelve south Indian kings including those of Vengi and the
Ganga country, executed a lightning expedition to north India, and forced
the lords of central India to pay him tribute; he may even have forayed into
Odisha. Though some of these regions had already been subdued by
Dhruva, medieval Indian emperors could only be sure of the loyalties of the
core territories their families ruled directly. They were less certain of
territories ruled by dynasties tied closely to their own through bonds of
family or military alliance, as we have seen with the Vatapi Chalukyas and
their fractious relations in Lata. Regions ruled over by dynasties with their
own storied legacies – especially those brought into the fold recently, as
Vengi had been into the Rashtrakuta political formation – often stopped
paying tribute upon the accession of a new ruler, or whenever an established
overlord’s grip was seen to be faltering.
Early in Govinda III’s reign, his military successes had established that he
was an overlord best not crossed. Indeed, it seems that the minor rulers of
the subcontinent had been reduced to a state of abject terror at Govinda III’s
display of Rashtrakuta military might. One of them, Govinda’s mother’s
relation, the Chalukya king of Vengi, had scrambled to show his devotion
by helping build sections of the wall for Govinda’s grand new city of
Manyakheta, which the delighted imperial court claimed ‘touched the
summit of the sky’.6 Another, the king of Sri Lanka, sent Govinda two idols
of the Buddha, ‘his own personal lord and the highest lord of the Sinhala
polity’.7 These were set up, according to the Rashtrakutas, ‘like pillars of
fame’ in a Shiva temple in Manyakheta, yet another endorsement (no matter
if only symbolic) of the Deccan’s dominance8 over the entire subcontinent.
After such a violently triumphant start, it might have seemed that Govinda
III was on the verge of a long and successful reign – and yet in 814, barely
middle-aged, the Vallabha was dead. Prince Sarva Rashtrakuta was now left
to control a vast and multi-centric empire, surrounded by the relentlessly
ambitious lords of the Deccan. While Govinda had been able to find the
resources and inclination to repeatedly crush them into subordination or
overawe them with propaganda and public ritual, it certainly seemed that
Sarva would not be able to do so.
 One can only imagine what the boy went through at his coronation,
feeling the heat of the sacrificial fire as the chanting of verses was almost
drowned out by the roars of the Manyakheta crowds. As his rivals and
relations and vassal kings watched with hungry eyes, the waters of the royal
ablution were poured over his head, transforming him. No more would he
be the boy Sarva: from this day to his death, he was Fortune’s Favourite,
Earth’s Beloved, Paramount Lord, the Great King of Kings: Amoghavarsha-
Raja-Deva, the Unfailing Rainer of Gifts.9 A long period of deadly peril lay
on the horizon for the young Vallabha.
Nobody in the coronation pavilion that day – probably not even the newly
minted Amoghavarsha himself – could have expected that he would rule
with charm, refinement and violence for sixty-four years, preserving the
Deccan as one of the subcontinent’s dominant powers for most of the 800s,
and profoundly impacting the cultural history of southern India. He would
leave a legacy that would be sung of by Arab and Kannadiga alike for
centuries. In observing his career, we’ll embark on an odyssey from the
intensely personal to the global, and see how the Deccan under the
Rashtrakutas participated in the emergence of the global system of trade
and interaction that still exists.

Deccan kingship had come a long way from its early days under the
Chalukyas, who had initiated the institution’s gradual integration into the
Sanskrit Cosmopolis. By the ninth century, we can infer from inscriptions
that rulers like Amoghavarsha were exposed to the gamut of knowledge and
attitudes circulated across this vast geocultural formation. The young
Vallabha learned of the Three Goals of human life: following and upholding
dharma, the acquisition and protection of wealth (artha), and the pursuit
and enjoyment of pleasure (kama).10 Parts of encyclopaedic shastras,
explaining the correct means to achieve these goals,11 were probably drilled
into his head until he could recall them from memory. These could include
everything from ‘logic, law, royal policy, composition and metrics …
stories and poetry, itihasa (epics) and the Puranas’, to ‘musical instruments,
dancing, painting, leaf-cutting, gambling, knowledge of omens and
astronomy, architecture, carpentry, gem-testing’ and so on.12 He seems to
have learned how to use force and wealth to win others to his side,13 and
been taught how to regain and steward his kingdom and his family’s glory.
But most importantly, he learned – or was taught – how to manage the
people of his court.
As a young prince, before his accession, Amoghavarsha probably imbibed
the complex gestural and verbal protocols he would need to navigate
courtly hierarchies: to rise from his seat; to touch the feet of his elders; to
fold his hands above his head, in front of his face, or in front of his chest
according to the rank of the addressee; to bend at the head or waist or touch
various parts of his body to the ground as a prostration.14 He needed to
cultivate a low, pleasant voice15 and get used to formal modes of address
even for those closest to him: higher-ups could not be referred to by their
names unless permission was explicitly granted; the term deva (Your
Highness) used for vassal kings and queens, arya or bhadra (noble one/sir)
for learned Brahmins and accomplished artists and so on.16 The boy would
not have interacted with his father, the Vallabha Govinda III, very much,
appearing before him mostly on very formal occasions, during which he
had to adhere to a strict verbal and gestural protocol.
After his coronation in 814, as suggested by normative texts as well as
royal inscriptions from across the Sanskrit Cosmopolis, Amoghavarsha had
to learn how to carefully control his every facial movement. In medieval
Indian courts, even the eyes, like the rest of the body, were considered to be
capable of making gestures.17 A single miscalculated smile might be
interpreted by the palace crowd as a bestowal of royal favour and lead them
all to recalculate their opinion of the king and the lucky recipient.18 A
furrowed eyebrow or a harsh word might be taken as a sign of a terrible
falling out.19 A sidelong glance might be interpreted as one of great
intimacy.20 It must have been a stressful life for a child to lead.
The Rashtrakutas had come to power by overthrowing their Chalukya
overlords when they were distracted and vulnerable. In 814, in the
aftermath of Govinda III’s death and Amoghavarsha’s accession, things had
come full circle, as their vassals plotted against them in this moment of
vulnerability.21 The political crisis came to a head in 815/16, when
Amoghavarsha had been on the throne for little more than a year. The
Chalukyas of Vengi – cowed down by Govinda III and Dhruva Rashtrakuta,
and tied by blood to the Rashtrakutas through the empress Shila-Mahadevi
– sought their revenge. Her brother, the king of Vengi22 and
Amoghavarsha’s grand-uncle, allied himself with the Gangas23 and led an
army through the Eastern Ghats and Telangana, appearing straight in the
heart of Vidarbha (eastern Maharashtra).24
It was a bold move. Without a leader of Dhruva’s or Govinda III’s
charisma and military ability, the Rashtrakuta armies and vassals seem to
have been unable or unwilling to resist determined opponents. A scholarly
adolescent, Amoghavarsha must not have cut a very inspirational figure to
the battle-hardened veterans who had served under his brilliant and brutal
father and grandfather. The armies of the Vengi Chalukyas hacked and
burned their way deep into the Rashtrakuta heartland with little serious
resistance. Simultaneously, in the south, the Gangas raised their elephant
banners and defeated the armies the boy-emperor sent to punish them.25
Parts of the Deccan sank into anarchy,26 casting, we are told, ‘the glory of
the Rashtrakuta house into the Chalukya ocean’.27
The imperial court was forced to flee, seeking the assistance of the dead
emperor Govinda III’s general and nephew, Karkka Rashtrakuta: a man
bearing the awe-inspiring title of Patala-Malla, Hell Wrestler.28 As a
member of a cadet lineage of a Deccan imperial house, like the Lata
Chalukya king Pulakeshi-Raja before him, Karrka ruled in Lata in southern
Gujarat with the rank of maharaja, ensuring that the Rashtrakuta imperial
centre retained its crucial connections to Indian Ocean trade.
We do not know how the loss of his throne changed Amoghavarsha, what
he must have seen, and what it must have taught him about the world and
his place in it. It is difficult not to sympathize with the boy’s situation.
Meeting Karkka Rashtrakuta, the Hell Wrestler, and convincing him to
throw his lot in with the imperial court – perhaps through tools we know
other medieval kings used, such as expensive and prestigious gifts, praises
of his martial valour, and calls to familial loyalty and personal virtue – must
have been one of the early successes of the young Vallabha’s political and
diplomatic career. It would not be the last.
By May 821,29 the Hell Wrestler had worked out an arrangement with the
Vengi Chalukyas by offering his sister to the Chalukya crown prince in
marriage.30 The Gangas seem to have settled for an uneasy truce with the
Rashtrakutas as well, though both dynasties continued to fear and distrust
the other. Amoghavarsha, by now probably a rather world-weary and
determined twenty-one-year-old, could at last return to take the throne of
Manyakheta. The first deadly challenge to his rule had, for now, been dealt
with. As he paraded into the city, the imperial Garuda banner was raised on
a towering flagstaff to the blasts of trumpets, signifying that the Vallabha
had finally returned, that the King of Kings had once again assumed his
rightful place at the axis of the four quarters,31 that order was at last restored
to the world.
Amoghavarsha had left Manyakheta as a boy, now he returned as a man.
Self-control, astuteness in diplomacy and an unusually introspective turn of
character would characterize his public
image. ‘What does one fear?’ he would write, decades later. ‘Death. Who is
in even worse plight than the blind man? The passionate man. What is hell?
Dependence upon another.32 What is truth? The welfare of the beings.’33
This attitude informs many of his later inscriptions, and he would one day
famously sacrifice his little finger to the goddess Mahalakshmi,34 ostensibly
to save his subjects from a famine or outbreak of disease, an almost
unheard-of act for any contemporary ruler.
It is doubtful that most kings in the early ninth century were anywhere
near this contemplative.35 Many of them were raised as though they were
gods on Earth (with all the suffocating rituals and duties that entailed).
Rulers, as we know from historical records across millennia, could often be
spoiled, selfish, arrogant and intolerant of those below them, insulated from
the consequences of their actions by layers upon layers of security and
protocol.
But just as life today is not completely defined by presidents and prime
ministers, medieval life was not entirely defined by kings. Amoghavarsha
might have been a great force in the Rashtrakuta palace, but outside those
pleasant, perfumed environs, his capital, Manyakheta, was a diverse,
crowded melting pot where the Vallabha’s activities were generally
peripheral to the humdrum daily lives of the people of the Deccan.36 We
know very little about what the city actually looked like, thanks to centuries
of war, neglect and the lack of archaeological study of its probable site. But
if the worldview of Rashtrakuta courtly literature is any indication, its
residents came from all over the Deccan and beyond, from Bengal, the
Gangetic plains, Rajashtan, Malwa, Chhattisgarh, Gujarat, Maharashtra,
Andhra, the Konkan, Karnataka and so on – a true microcosm of the
Deccan’s far-flung politico-economic networks. Sanskrit literature on cities
describes occupations ranging from ‘ministers of the king, judges of the
court of justice, and ācāryas or professors, to scribes, physicians,
merchants, grammarians, guards, bards, courtesans, vit.as [libertines],
dhūrtas [knaves], pīthamardas [companions], vīnā players, dance masters,
drummers, master painters, and sabhikas (club keepers), down to male and
female servants … carpenters, gardeners, elephant riders, cart drivers…
cowherds, horse keepers, masseurs’,37 and butchers, cobblers, carpenters,
doctors, monks, bureaucrats, generals, sthapatis and more.
This diversity in occupation was matched by religious plurality and
proliferation. All the major Shaiva sects probably ensured they were
represented in the Deccan’s most important city, from the Pashupatas to the
Kapalins to the Kalamukhas. So too did the Jains, whose expansive network
of monasteries ensured they were still one of western and southern India’s
dominant religious forces, the fulmination of Tamil bhakti poets
notwithstanding. Today, though Jainism has almost vanished in the south,
retaining a foothold only in the balmy shores of coastal Karnataka, it
continues to be a major force in Gujarat and Rajasthan.
But one would have been laughed out of Manyakheta in the ninth century
if one had claimed that south Indian Jainism would some day vanish. In the
centuries since the Chalukya king Vikramaditya I’s initiation into Shaivism
in the seventh century, Jains in the Deccan had proven every bit as
innovative and responsive to new religious and political trends as the
Shaivas. The idea of the jina, the ascetic who conquered the senses, was
creatively presented as analogous to the warrior who conquered his
enemies,38 thus making this once-peaceful religion (which theoretically
could not even stand for an insect to be harmed) most appealing to some
warlike south Indian aristocrats. Indeed, ninth-century Jains offer a striking
insight into the continuing development of religions in the subcontinent.
 South Indian Jain sects, such as the Yapaniyas, allowed women to be
ordained as nuns, and made active attempts to entice householders into the
fold.39 The Yapaniyas also adopted practices that are commonly identified
with Hinduism today, including the idea of caste,40 permanent monasteries
dependent on agriculture, the worship of goddesses such as Jvalamalini,
Bearer of a Fire Garland, and the propitiation of astrological deities on
behalf of laypeople.41 Many Jain schools even ran full-fledged temple
establishments complete with ritual bathing of the deity, festivals,
processions and musical soirees.42 Like contemporary Shaiva monastic
orders,43 they were also wealthy enough to build temples in their own right,
or at least influential enough to request it as a form of royal grace. These
were functionally almost indistinguishable from Shaiva temples, acting as
religious, social and political centres just as the Lokeshvara and
Krishneshvara temples had, hiring the same sthapati guilds and drawing
patrons from similar classes. (Jain temples, however, tended to be
somewhat more austere with their sculptural decorations than Shaiva
shrines.) All these innovations, which were shrewd responses to extant
socio-economic trends, powered Jainism to a nearly 800-year-long reign,44
alongside Shaivism, over south India’s religious landscape. The striking
similarity between the two religions (in practice, if not in theology) also
challenges us to think of developments in Indian religions as long-term
overall trends that evidently cut across sectarian boundaries. From this
perspective, Indian religion in general is perhaps best understood as a
response to, and driver of, evolving social, economic and political patterns.
The Gangas, those implacable rivals of the Rashtrakutas, were great
patrons of Jains. Now, in the aftermath of a devastating war, sitting
precariously on his throne, Amoghavarsha saw an opportunity. Just as the
old Chalukya Vallabha Vikramaditya I had cultivated closer ties with the
Shaivas in a moment of crisis, the Rashtrakuta Vallabha Amoghavarsha
encouraged Jain literature and patronized that religion’s powerful network
of monasteries and temples.45 Indeed, he and his successors would construct
a number of Jain shrines and cave temples at Badami and Pattadakal: the
Malaprabha river valley seems to have continued to be an invaluable
economic centre to the lords of the Deccan even after the fall of the
Chalukyas of Vatapi, now a fading memory.
This is not to say definitively that Amoghavarsha was a Jain: these
boundaries are difficult to establish in medieval India. Religiosity at the
time was fluid,46 and monarchs liked to spread their devotions and
patronage as widely as possible to appeal to the broadest possible cross-
section of society. For example, Amoghavarsha would refer to himself in
his own inscriptions as Vira-Narayana, the Valiant Vishnu, and as
mentioned, sacrificed his little finger to the goddess Mahalakshmi.47 The
Jains were only one of the many squabbling and ambitious social groups of
Manyakheta amongst whom the Rashtrakuta Vallabha spread his largesse.
Another favoured group – and perhaps far more significant both to
Amoghavarsha and to us – were poets. The young Vallabha was a master of
aesthetic theory and poetry, especially in the world of Sanskrit literature,
and had rather strong opinions about what comprised good poetry and what
did not.48 The Deccan had come a long way from the days when the
Chalukya Vallabhas had slowly, tentatively, begun to popularize the use of
Sanskrit for its prestige and political utility. Now centuries later,
Manyakheta, the Deccan’s new imperial centre, thrived with poets as the
great lordly houses competed with the Rashtrakuta Vallabha to attract poetic
talent. Poets were well respected and highly specialized professionals with
considerable presence in aristocratic daily lives. For example, the
contemporary Deccan text Yashastilaka, a Jain epic, describes royal
officials selecting an elephant and a horse for a king. An elephant named
Udayagiri, sent by the king of Kalinga (southern Odisha)49 as part of his
annual tribute, was presented to him with
an elaborate report on the characteristics of the chosen animal, full of technical details.
Meanwhile, a bard named Karikalabha recited a number of verses in praise of elephants.
Similarly, a notable white horse of Kamboja [most likely Central Asia in this context] was
selected by a committee of experts … after which a bard named Vajivinodamakaranda recited
some verses in praise of horses.50
Like all the Vallabhas who had come before him, Amoghavarsha also
toured the dozens of cities and towns of which he was the hereditary lord,
as well as the temples and monasteries and cities of his many vassals.
During these travels, he appears to have noticed the gulf between the
urbane, intellectual, Sanskrit-speaking world and the overwhelmingly
Kannada-speaking world of orally transmitted legends and stories – the
world inhabited by the lower ranks of his vassals, who were much more in
awe of him than those in the court of Manyakheta.
Perhaps, as he heard Jain myths and fantastical legends of Rashtrakuta
and Chalukya kings, local heroes and deities, and regional takes on stories
from Sanskrit epics and Puranas, as he judged and rewarded bright young
Kannada poets in Pattadakal and Koppana and Porigere and Okkunda,51 the
outlines of a magnum opus began to take shape. But it would take
considerable resources, time and talent to bring it to life. For now, the
Vallabha had other things on his mind.
Constantly pressed for resources and goods to reward his loyalists and to
maintain his pre-eminent status within the Deccan, and lacking the military
capabilities of his father and grandfather, Amoghavarsha was in desperate
need of wealth and luxury goods. By the ninth century, an unprecedented
constellation of global and local factors were beginning to come together.
Far to his west and east, two vast political formations – the Abbasid
Caliphate in West Asia, and the later Tang dynasty of China – were growing
increasingly interested in the profits of global trade. As the young Deccan
emperor toured the dusty plateau and its noisy cities, interactions in the
Indian Ocean were growing to unprecedented levels. This would offer the
young Vallabha a means to maintain the splendour of the Rashtrakuta court
without having to rely primarily on ceaseless war and tribute, as his
predecessors had.

Let us pause and look at the enormous world of which Amoghavarsha

Rashtrakuta and other medieval Indian kings were an integral part.
As anyone who has traversed modern India by road will attest to, the
subcontinent is almost incomprehensibly vast. And yet it is dwarfed by the
size of the Indian Ocean and the continents that surround it. By the ninth
century, the subcontinent, as a result of its geographical location in the
centre of the Indian Ocean, the sophistication of its craftsmanship, and its
bounteous natural resources, had emerged as the central link in the
exchange chain between Europe, Africa and much of Asia. The scale and
depth of this interconnectivity is exemplified by stunning archaeological
discoveries dating to the sixth–eleventh centuries on the Swedish island of
Helgo, not far from Stockholm. These include a sixth-century bronze
Buddha from Kashmir and a bronze ladle from North Africa, as well as
‘Arabic coins, Frankish glass and metal-work from across western
Europe’.52
 By the ninth century, great tides of people were in motion across the Afro-
Eurasian landmass, interacting with new goods, languages and ethnicities.
Moving from west to east across this landmass, we would see
Scandinavians beginning to move into the British Isles as well as down the
Volga river into Ukraine. We would fly over Frankish kingdoms – the
ancestors of the later kingdoms of France – preserving Latin knowledge in
monasteries and manufacturing glass and metalwork, sending embassies to
the courts of Iraq, and receiving elephants in return: the first time the
animals had set foot in France since the Roman Empire centuries ago. We
might pause in awe to behold the glittering palaces, libraries and ports of
Constantinople on the Bosporous, the direct medieval descendant of Rome,
overflowing with perfumes, silks, spices and gold. We would traverse the
vast Abbasid Caliphate, which had overthrown the Umayyads, connected
the markets of Egypt, Iraq and Persia, and established the metropolis of
Baghdad as the world’s premier centre of trade, education and scientific
research. Moving south into Africa, we would see vanishing into the
distance the wealthy city-states of the Swahili coast, a region that has
historically been deeply connected to India, exporting gold, ‘ambergris,
ivory, leopard skins and tortoise shell’.53 Leaping across the ocean, we
would traverse the dusty plateaus of Iran with their underground canals and
rich cities, pass the thriving oasis towns of Central Asia, and reach China,
where the superpower that was the Tang dynasty matched the ‘gravitational
pull’ exerted on the global economy by the remnants of Rome, the Islamic
world and the Indian subcontinent.54 (For context, the early Tang emperors
had been rough contemporaries of Harsha, Pulakeshin and the Prophet
Muhammad.) By the ninth century, the Chinese too had developed an
interest in international trade.55 Across Europe, East Africa, West Asia and
East Asia, the ninth century thus saw unprecedented attention to the profit
and prestige to be gained through global interactions.
 But this continent-spanning network of interactions depended on ninth-
century South Asia’s thriving ports, merchant organizations, avaricious
aristocrats and toiling agrarian masses. In a war-torn time, where military
success and control over land had made fortunes for many dynasties,
merchant organizations such as the Five Hundred Lords of Aihole (whose
emergence we witnessed in Chapter 4) helped connect producers and
markets, and moved goods across vast geographical extents. Such
organizations could mobilize large amounts of capital; insulate members
against the risk of banditry, piracy or natural disaster; and hire mercenaries
to protect their depots and caravans. They could also, as we will see, secure
consignments of rare goods or strategic resources for warlords, and use
their political networks to connect international traders and Indian markets.
Meanwhile, for rulers such as Amoghavarsha, control over important ports
was critical because it allowed a degree of taxation, and because
international trade allowed them to obtain luxury goods that were crucial to
maintaining their status and prestige over competing lords. However, the
exact degree to which Indian royals and their agents succeeded in taxing or
directly profiting from international trade is a matter of considerable debate.
To get a better sense of the subcontinent’s importance in ninth-century
global trade, we will further explore systems of production and exchange
within the subcontinent. We will then try to catch some glimpses of
Amoghavarsha’s own activities through the testimony of Arab merchants
and travellers, as well as evidence from archaeology and inscriptions in the
great Rashtrakuta port of Sanjan.
Over the last few chapters, we have seen medieval Indian kings making
land grants to local elites to create regional power bases. But these actions
were also tied to the development and encouragement of commerce. By
granting land and giving grandees the right to pay less tax in return for
irrigation and rainwater harvesting works, kings sought to deepen and
widen networks of agrarian production, supporting further economic
specialization. The surplus agricultural produce that new landowning elites
(temples, monasteries, Brahmin landlords and local warlords) collected
were moved to local exchange centres which began to emerge as
commercial hubs: we have seen this process earlier with the emergence of
the Five Hundred Lords of Aihole in the Malaprabha river valley. Such
exchange centres supplied the growing cities and towns of the Deccan with
food and crafts, and also established routes for the exchange of other goods,
both imported luxury items and local manufactures, especially cloth.
However, the commercial economy was not yet mature enough to support
large-scale artisanal specialization. Many local artisans used their craft as a
way to supplement agricultural and raiding activity, with true specialists
being confined to cities, which had both the economic surplus and constant
consumption required to sustain them.56
The evolution of a complex and stratified economic structure is always a
complicated process – and often tragic. The transformation of the dry lands
of the Deccan into networks of irrigated, commercially productive hubs was
often enforced at sword-point, and inscriptions speak of defiant villages
slaughtered, of women raped, even of children attacked.57 However, for
those with the social and fiscal capital to manoeuvre themselves, all this
was a great opportunity for profit.
None were better positioned to do so in the ninth century than the Five
Hundred Lords of Aihole. A century earlier, as the Vatapi Chalukya
Vallabha Vikramaditya II raided the Pallavas and his queen Loka-Mahadevi
erected her temple, the Five Hundred had rapidly expanded their influence
across the Deccan. By the late 800s, when the Rashtrakutas had seized
paramountcy in the Deccan, the Five Hundred had managed to establish a
franchise of sorts in Pudukkottai, in the Palar river valley of the Tamil
country.58 This location was crucial for trade between the two regions.59
This sudden leap in their influence – from dominating a hub connecting
manufacturers and traders in the northern Deccan to establishing a presence
in another hub, straddling trade routes into the Tamil country – may have
been connected to Rashtrakuta aggression in the region under Dhruva and
Govinda III, and helped establish them as one of southern India’s most
powerful merchant associations.
Meanwhile, on the subcontinent’s west coast, Zoroastrian and Christian
merchants from Persia visited every year and settled down. Some moved
inland, some intermarried with locals, and some formed their own diverse
guilds, such as the Anjuvannam, a name possibly derived from the Persian
anjuman, ‘association’.60 Alongside them, in Malabar, were the
Manigramam, an indigenous merchant group which worked closely61 with
pepper-cultivating indigenous Christians62 organized under the aegis of
native churches. These two organizations seemed to have worked together,
monopolizing the purchase of pepper and its sale to buyers across the
western Indian Ocean. The Manigramam ensured that local chiefs turned a
benign eye to the trade63 and probably focused on buying and sorting
pepper, while the Anjuvannam used its far-flung international networks to
find buyers and secure capital for shipments. During the eighth–twelfth
centuries, similar merchant networks stretched all the way from India to
Iran, Iraq, Central Asia, the Mediterranean world, North Africa and even
Spain.64
Inscriptions of both the Anjuvannam and Manigramam are often found in
association with the much vaster organization that was the Five Hundred. It
therefore seems that the expansion of the Five Hundred by the ninth century
helped connect newly productive agrarian hinterlands controlled by Deccan
warlords – as well as ravenous urban markets such as Manyakheta – to
international trading networks through these coastal merchant
organizations. In response to blossoming forces of supply and demand
within India, Arab ships increasingly began to make journeys through the
year instead of primarily during the monsoon season, as earlier. These ships
usually moved in a circuit from Siraf on the Persian Gulf to ports in Gujarat
and the northern Konkan, making fortunes for their owners in the process.
All this was also aided by the actions of Indian rulers great and small with a
presence on the west coast, eager for a slice of the extremely lucrative
trade65 through tariffs and the purchase and sale of luxury goods and horses.
In the Konkan, for example, a dynasty called the Shilaharas managed to
conquer and unite the three great emporia of Gopakapattinam (now Old
Goa), Balipattana and Chandrapura (which Arabs called Sindabur) into a
single ‘regional economic unit’ that would emerge as one of the major
trading centres of the west coast,66 acting as an international and regional
hub for coastal trade.67 This was accomplished in the same decade that
Amoghavarsha returned to the throne of Manyakheta, c. 830.
Other ports that emerged as thriving trading hubs for Arabs include
‘Kanbaya (Cambay), Sandan (Sanjan) [both in Gujarat], Tana (Thana),
Subara (Sopara), Saimur (Chaul) [all in coastal Maharashtra], and Manibar
or Mulaybar (Malabar)’.68 All of these, with the exception of Malabar and
probably Cambay, were within the Rashtrakuta sphere of influence in the
ninth century. Arab traders were aware that their local partners owed
ultimate allegiance to a distant, splendid monarch far off in the interior.
Over the next two hundred years, stories of the ‘Balhara king’, the
Vallabha-Raja, would attain almost legendary proportions, and Arab
merchants in the ninth century would insist ‘there is in fact no [Indian] king
who has greater affection for the Arabs than the Balhara’.69 This almost
certainly refers to Amoghavarsha, suggesting that the Vallabha was well
aware of the profits to be made from trade and sought to systematically
encourage it.
West Asians – Muslim Persians and Arabs, as well as Zoroastrians, Jews
and Christians of diverse ethnicities – were soon settling down in these
major ports, establishing permanent bases for trade. The modern ideals of
chauvinistic religious and linguistic nationalism scarcely existed in the
medieval Indian Ocean world, despite the violent conquests of the
Umayyad Caliphate a few generations before this time. It appears that
occasional elite-directed violence was not allowed to get in the way of
much more frequent people-to-people interactions. The Rashtrakutas, for
example, were pleased to allow these wealthy immigrants the right to build
mosques: Amoghavarsha and his successors even appointed Muslim
officers to administer personal law70 to traders. In Chaul, Maharashtra, there
is evidence of a group called the Bayasira – born to West Asian fathers and
Indian mothers.71
These immigrants also integrated themselves with gusto into the
cosmopolitan culture of the times. One fascinating example is a Persian
Muslim known to us as Madhumati, son of Sahiyarahara (a Sanskritized
form of Muhammad, son of Shahryar). In the tenth century, Madhumati was
appointed the Rashtrakuta governor of the entire region of Sanjan, including
much of the northern Konkan, a region crucial to the Rashtrakutas’ global
connections. As with any contemporary Indian king, Madhumati’s prashasti
declares that he had ‘conquered the chiefs of all the harbours’, indicating
his appointment as governor was due to his naval expertise as a seaman. He
claims to have spent on public charity, establishing two free ferries and a
feeding house which served ‘rice, curries and ghee’ free of cost.72
Madhumati worked closely with a diverse city assembly which included
both Indian and West Asian merchants,73 and was served by a loyal Indian
minister – whose friend, a well-connected Brahmin aristocrat, set up a
‘monastery or temple’ in the city with Madhumati’s approval.74 West Asians
would also serve as witnesses and important community members when
land grants were made to shrines by their Indian colleagues. Indeed,
Madhumati himself made a land grant to a Durga temple in the region.75
It should be noted that such activities were sometimes met with orthodox
disapproval, and varied significantly by ethnicity and region. Documentary
evidence suggests that Arab merchants at times had to obtain special legal
dispensations to trade with ostensible ‘infidels’,76 whereas some Indian
merchants sometimes insisted on separate quarters for eating for reasons of
caste purity. With all of this, they nevertheless seemed to have maintained
warm business and interpersonal relationships. All this suggests that the
Rashtrakutas were highly involved in the multi-ethnic, multilingual,
cosmopolitan and ‘globalizing’ world of the ninth century.
But why did the Rashtrakutas, an inland Deccan empire, put so much
effort into encouraging trade on the subcontinent’s coasts? The benefits to
them were strategic, political and economic.
Acquiring horses was a priority for Deccan kings from the earliest times:
recall Chitrakantha, the famous steed of the Chalukya ruler Vikramaditya I.
Coins of Govinda III, Amoghavarsha’s father, constitute another example:
gold dinaras have been found of him as a turban-clad shirtless warrior
sitting on an Arab stallion, one arm holding a sword with a wicked,
decapitating hook on top, the other raised. These and other Rashtrakuta
coins are marked with decorative patterns inspired by the Arabic script,
which Indian courts appear to have found a beautiful curiosity; they also
feature the auspicious symbols of the Garuda eagle, lotus, and conch shell.
They declare Govinda to be Apratihata, ‘Invincible’.77
On these small pieces of portable propaganda, Govinda III seems to be
depicting himself as a military man adapting to the latest global trends of
his time: a stark contrast to the stereotype of medieval Indian monarchs as
disinterested in new technologies and the benefits of international trade. It
is also a most striking parallel to the similarly innovative emperors of
fifteenth- and sixteenth-century Vijayanagara: Deccan rulers who relied on
Portuguese and Arab traders to secure cavalry for their armies.
 As the Deccan grew more connected to global trade during the reign of
Govinda III’s son, Amoghavarsha, the scale of its exchange with the world
in general, and the Islamic world in particular, expanded further. These
exchanges were complex and multidimensional, taking advantage of
developments from across continents. Medieval Indians exported sugar
produced by cutting-edge Indian technology,78 which would eventually be
adopted by the Chinese;79 salt produced from seaside pans, based on
Chinese technology learned by the Arabs from interactions in Central
Asia;80 mangoes and citrus fruits grown in vast estates,81 and jewels and
semi-precious stones set in ‘bowls, knife-handles, paper-weights, beads and
bangles’,82 among many other items. Also on the list of Indian exports were
extremely fine cotton cloth ‘which could pass through a small ring’83 and
indigo (correspondence from the period even gives some West Asian indigo
merchants the name nili, blue, a term derived from Sanskrit84). Sanjan in
Gujarat, where the Persian Madhumati ruled on behalf of the Rashtrakutas,
was a major centre for the indigo trade,85 hinting at intense commercial
cultivation around the region.
Excavations conducted by the Indian Council for Historical Research at
Sanjan in the early 2000s found the remains of a ‘prosperous and
flourishing city’,86 including glass beads, bangles and semi-precious stones.
Most importantly, large quantities of West Asian glass87 and Chinese
ceramics88 were discovered, signs of Sanjan’s globe-spanning import and
export networks. In another indication of the complex and intertwined
nature of history, Sanjan appears to have been the site where Zoroastrian
refugees from Persia, fleeing Umayyad persecution, made their landfall in
India in the eighth century. Little more than a century later, descendants of
these refugees were conducting a brisk international trade under oversight
of Madhumati, the Muslim Persian governor appointed by Amoghavarsha’s
successors, who were Kannada-speaking kings who patronized Hinduism
and Jainism.
 Indians also exported perfumes, cloves, nutmeg, pepper, mace and
cardamom, alongside the odd peacock, rhinoceros or even elephant89 for
particularly rich menageries in Baghdad. Arabs are recorded to have been
aware of ‘thousands of such [Indian] commodities’,90 which poured into the
expensive kitchens and fashionable salons of Baghdad. In the high-stakes
world of Abbasid politics, Indian poisons, swords and jewellery were well
regarded.91 So too was Indian knowledge: the Abbasids brought together
scholars and texts from the East and the West to develop their
understanding of ‘philosophy, spiritualism, cosmology, geography, geology,
history, botany, mineralogy, physics, mathematics, astronomy, astrology,
medicine, pharmacology, anatomy, optics, agriculture, irrigation, [and]
zoology’.92 Indian mathematicians, physicians and philosophers lived in
Baghdad, and some were employed in the courts of the Abbasid caliphs Al-
Mansur and Harun Al-Rashid, contemporaries of the early Rashtrakutas.93
Indian scholars are attested to have worked alongside Abbasid scholars in
the great library known as the ‘House of Wisdom’, and Indian numerals,
arithmetic, geometry and natural philosophy had a profound impact on the
development of the sciences in the Islamic world.94 Though the evidence is
extremely fragmentary, there are some suggestions in Rashtrakuta texts that
Abbasid innovations – especially in the realm of ‘fine technology’,
producing sophisticated mechanical devices such as water clocks, fountains
and automata – were reaching the Deccan even as Indian knowledge moved
West.95 At a more intimate level, there is evidence that some Arab
merchants were so enamoured with India that they are recorded to have
named their daughters ‘Hinda’.96

Caches of letters from the period reveal not just good professional
relationships between Indians and their Indian Ocean business partners, but
strikingly human tales of friendship, greed, adventure, loss and the sorrow
of parting.97 A story from medieval Kerala describes merchants from
Karnataka, Malwa, Gujarat, Andhra, Odisha, Greece (or Anatolia), China
and Arabia (or Central Asia) sitting on black carpets, counting money and
chatting.
A senior merchant among them tried to impress his juniors in a bragging
tone:
If I sell a jonakuttira (Arab horse) in the Cōl.a [Tamil] country, I will immediately get two
 thousand anayaccu [coins] … in cash. For my elephant I will get eight thousand … If I go to
Kollam (Quilon) and Kollapuram (Kolhapur) I can sell quickly all the good karpuram (camphor)
…98
Evidently, as the historian Ranabir Chakravarti points out, the merchant is
attempting to impress his audience with the fact that he is trading in high-
value commodities, the only buyers of which could have been landowning
military elites and religious institutions. This is no common merchant, it
seems. He then continues his boast, in a telling revelation of the closeness
of mercantile and political interests in early medieval India: ‘I have to get a
hundred thousand accu [coins] by way of interest on the loan I gave to
Vallabha.’99
Meanwhile, as the ports of the west coast buzzed with activity, some
particularly enterprising Arabs, seeking to write travelogues and on the hunt
for commercial opportunity, entered the Deccan, reaching the Rashtrakuta
heartland itself. Their writings portray India as a strange land of great
splendour and prosperity. Such depictions were intended to instil wonder
among their readers – a major concern in Arabic poetics – rather than being
factual representations of travels. Nevertheless, it is fair to say that India’s
‘wonders’ were an important component of Arabic travel literature, and a
crucial part of their imagination of the world.
One Arab writer was particularly taken by the ascetics who wandered the
jungles and peaks of the Western Ghats. Some of them went about naked
with a ring on their penises ‘to prevent all sexual relationships with
women’, while others wore leopard skins.100 With their matted hair, wild
eyes and pungent odour, they must have made quite an impression. One of
them stood facing the sun somewhere in the Konkan, where this Arab
traveller first saw him. He was apparently still there sixteen years later,
when the Arab visited again and was stupefied by the man’s discipline and
by the fact that his eyes had apparently not been burned into empty husks
over all those years.101
The Rashtrakuta capital, Manyakheta, was also fertile ground for Arab
travellers’ imaginations. Manyakheta, now a hub of subcontinent-spanning
political and economic networks, may have rivalled Baghdad itself (though
large-scale archaeology is needed to say for certain). Through the ninth
century, as the Rashtrakuta imperial network and the merchants it
patronized flourished, the city seems to have been remade through the
profits of war and trade; indeed, inscriptions often claim that
Amoghavarsha ‘made’ the city. In the absence of major military adventures
during his reign, this may be read as a hint of the wealth that global trade
brought the Rashtrakutas. Decked with temples, markets and mansions,
surrounded by rivers on three sides,102 Manyakheta had a great moat103 and
rampart fluttering with the pali-dhvaja of the imperial dynasty and banners
of its most powerful lords, its gods and its guilds. One Arab visitor, clearly
awed by what he saw and seeking to impress a cosmopolitan audience in
Iraq, wrote (rather hyperbolically):
‘In that city there are for the ordinary people one million elephants which carry the merchandise
… In this temple there are about twenty thousand idols made of a variety of precious metals …
and various carved stones mounted with shaped and artistically worked precious jewels … In
that house is an idol whose height is twelve cubits and is placed on a throne of gold in the centre
of a golden cupola, the whole of which is set with jewels like white pearl, ruby, sapphire, blue
and emerald stone.104
Amoghavarsha Rashtrakuta, Vallabha of the Deccan, had, it seems, made
a wise choice in relying on trade rather than the tribute and ceaseless
warfare used by his predecessors to fund their politico-military machines.
As his coffers swelled, the sophisticated young emperor finally had the
resources for a different kind of magnum opus than we have seen up to this
point.
After the setbacks he had faced early in his reign due to his inability to
personally command and motivate his armies, Amoghavarsha had
developed a flair for diplomacy and politics, an eye for talent, and a knack
for promoting and inculcating loyalty. A master of literature and aesthetics,
he now brought together an unprecedented concentration of poets and
teachers to decorate his court at Manyakheta. He could have merely set
them to the banal task of composing prashastis and dramas, but
Amoghavarsha did something else, of exceptional vision, that would echo
down to the very languages that Indians speak today.
Amoghavarsha’s court poet Srivijaya, working on guidelines laid down by
the Vallabha,105 was given the task of composing a great manual of courtly
Kannada grammar – the first the language had ever seen. This Kavi-raja-
margam (The Way of the King of Poets) was the first text in the world to
‘self-consciously’ theorize the relationship between a vernacular language
and the cosmopolitan Sanskrit, the domain of urban and courtly elites.106
This work reflects the extraordinary sophistication that the science of
linguistics and poetics had reached in the subcontinent by this time; there is
nothing comparable to it across the Afro-Eurasian world.
Amoghavarsha’s intention in propounding The Way of the King of Poets
was to create a register of Kannada with aesthetic and poetic qualities
comparable to the elite literature of the Sanskrit Cosmopolis.107 For this, the
structure of older works on Sanskrit
poetics – especially the Kavyadarsha of the old Pallava court poet Dandin108
– was systematically reworked to establish the grammatical, metrical and
aesthetic rules this courtly Kannada would require.
Dandin’s text had marked out the limits of the Sanskrit-speaking universe,
discussing the extent of the Indian subcontinent and the varieties of regional
Sanskrit styles within it; now Amoghavarsha and Srivijaya did the same for
the world of Kannada, declaring it to be the land between the Godavari and
Kaveri rivers109 and classifying a ‘northern’ and ‘southern’ variety of
Kannada poetic styles.110 ‘Sanskrit and Prakrit are well-established
languages, with their characteristic features and examples,’111 said the
Rashtrakuta Vallabha and his poet.
‘It is difficult to create poetry in Kannada …112 With the native language having so many
variations, even Vasuki [the thousand-headed king of serpents, the paragon of scholarship]
would become frustrated, unable to identify and fix the faults in the usage of Kannada
dialects.’113
These ‘faults’ included unpalatable and harsh sounds, incorrect meanings
and difficulty in comprehension114 – all of which were regarded with horror
in Dandin’s Kavyadarsha. (It is debatable whether local Kannada poets,
storytellers and singers, who had been plying their craft well before
Amoghavarsha set out on this project, thought so harshly of their language,
but their opinions have not survived.)
In their newly developed courtly Kannada, the Vallabha and his poet now
set out their own definition of Kannada literature, the qualities to be sought
in its two varieties (poetry and prose), and the assorted blemishes they
thought made Kannada poetry less than perfect. They also provided nearly
two hundred examples115 – taken from older Sanskrit manuals but reworked
into their new vision of Kannada literature – of the proper rules of
composition and mixing of Sanskrit, Prakrit and Kannada sounds and
grammatical rules. These were developed from deep scholarly engagement
with well-established Sanskrit and Prakrit grammars, linguistics, aesthetics
and rhetoric, as well as a detailed survey of extant Kannada poetry, prose
and metres.116 The purpose Amoghavarsha had in mind was clear: now, at
last, Kannada could match the aesthetic and expressive capabilities of
Sanskrit. At last, it could be a language of power and prestige and appear in
the prashastis of kings; at last, it could be a language of beauty and be
admitted into and celebrated in courts. It could compete with that ancient
language of prestige on its own terms.
This new ‘courtly’ Kannada struck enormously fertile ground.
Amoghavarsha’s hundreds of vassals and rivals, and their thousands of
poets and panegyrists, now unleashed a new wave of literary production in
their own language, a wave of literature that was more intelligible than
Sanskrit was to their peers and subjects, more open to the participation of
new social groups, and thus far more politically instrumental. The
transformation was immediate and dramatic: ‘the proportion of records in
Sanskrit shrank from about 80 percent in the period 741–819 (the
approximate level of the Bādāmi [Vatapi] Cāḷukyas) to 15 percent in the
period 819–974’, and to a negligible 5 per cent by 996.117 Henceforth in the
Deccan, Sanskrit, the once-dominant language of Indian literary,
philosophical and scientific works, would increasingly be used with
Kannada compounds, and in many cases be replaced by it altogether –
except in some very significant cases, as we will see later in this book.
This was a landmark in the history of world literature, a moment of
transformation. In addition to the emperor Amoghavarsha Rashtrakuta, the
illustrious litterateurs involved in this project included Parama-Srivijaya
(Paramount Victory of Fortune), Kavishvara (Lord of Poets),
Panditachandra (Moon Among Pandits), and Lokapala (Guardian of the
World).118 These titles were all honours granted to them by the emperor as
rewards for their work. For all future Kannada poets – down to this very
day, when there are more speakers of Kannada than the entire population of
some countries – this moment would stand out as ‘a rupture in time, a
moment of discontinuity, when something new began’.119 It was the moment
when predominantly Kannada-speaking polities began to shift their gaze
from the Sanskrit Cosmopolis and its connections to subcontinent-wide
cosmopolitan literary cultures, to audiences closer to home, more directly
relevant to their political networks.120
But Kannada had existed before Amoghavarsha, and it had certainly been
used by those toiling at the bottom of the political and religious hierarchy to
tell their stories and sing their songs for centuries before, irrespective of the
sophisticated aesthetics of the Sanskrit Cosmopolis. From this perspective,
Amoghavarsha’s establishment of ‘courtly’ Kannada could also be
interpreted as an appropriation of an already thriving vernacular culture to
suit the aesthetic of power to which he and his vassals were accustomed.
What might the average Deccani have thought of this ‘aestheticization’ and
elite transformation of a language they had spoken for centuries?
All these intellectual activities did not make Amoghavarsha a pacifist by
any stretch of the imagination. His sheer will to power could put any
Mughal emperor to shame. His relation, Dhruva of Lata, the son of Karkka
Patala-Malla (the man who had helped Amoghavarsha retain his crown
when he was a boy-king) followed in the tradition of Deccan cadet branches
based in the wealthy coastal region, and rebelled against the Vallabha in the
840s. Amoghavarsha’s armies would end up killing the young man,
spawning a family vendetta lasting almost twenty years.121 Meanwhile,
Ganga prashastis inform us, their kings were also occupied with
‘illuminating the sky of their own kingdoms, which had been overcast by
the darkness of the night of the Rashtrakuta’.122
By the mid-850s, Amoghavarsha attempted to crush the Gangas once and
for all. He was now at the peak of his power, with awestruck Arab visitors
acclaiming him as one of the Four Great Kings of the World, his only other
rivals being the Byzantine Emperor, the Emperor of China and the Abbasid
Caliph himself.123 His general Bankeya, the commander of the fearsome
Rashtrakuta elite hereditary guard corps, was ordered to ‘extirpate that lofty
forest of fig-trees – Gangavadi, difficult to be cut down’.124
Bankeya’s story is interesting: his meteoric ascent from the son of a
village chieftain125 was accompanied by increasingly elaborate titles and
governorships, from Bankesha (Lord Bankeya) to Bankeyaraja (King
Bankeya). It appears that as a member of the Kannada-speaking peasant
elite, Bankeya was able to take advantage of Amoghavarsha’s new courtly
Kannada culture. Through a display of the courtly ideals of valour and
military brilliance, he had risen far beyond his station and catapulted
himself into the innermost circles of aristocrats. The man would even
establish a city named after himself, Bankapura,126 which would become an
important centre of Deccan Jainism; it will feature again in this book.
The campaign was derailed, however, when one of Amoghavarsha’s sons
rose in rebellion, joining up with his relatives in Lata and attempting to
overthrow the elderly Vallabha. But the loyalty Amoghavarsha was able to
inspire among his subordinates saved the day. Returning to Manyakheta
post-haste, Bankeya promised he would either end the rebellion in three
months and make the emperor ‘drink milk’127 to calm his mind, or he would
immolate himself. He then unceremoniously killed the disloyal prince and
scattered the Vallabha’s enemies, though precisely how this was done is
unknown. After the crisis had passed, the delighted Amoghavarsha issued
an edict which ‘till the world’s end proclaims him [Bankeya] a hero’,
eulogized the general’s loyalty, fame and fury, granted him additional lands
and titles, and confirmed and expanded Bankeya’s donations to a Jain
monastery.128
However, more invasions and rebellions would follow as Amoghavarsha
continued to age. The Pratiharas of western India, now on the ascendant,
conquered Malwa and parts of Gujarat; meanwhile, a new king of Vengi
began to raid Maharashtra,129 though the Rashtrakutas claimed a great
victory against him, having ‘offered the Chalukyas [of Vengi] like a
sacrifice to Yama’, the god of death, at the battle of Vingavalli.130 Nearing
seventy, Amoghavarsha finally sought a diplomatic resolution to the Ganga
problem: two of Amoghavarsha’s daughters were married to Ganga princes,
and the Ganga king, who had successfully brought the subcontinent’s
greatest power to terms, was only too happy to establish a military alliance
with the Rashtrakutas131 after generations of bloodshed. This would last
until the collapse of both houses in the late tenth century. From the
Rashtrakuta perspective, this was at least if not more significant than the
north Indian adventures which captivate the modern imagination.
Gangavadi’s terrain was extremely hilly, with an average elevation
comparable to the Vindhya mountains; it thus offered them a secure
southern frontier and staging ground for campaigns in the Tamil country.
We will witness this later in the book.
Ageing rapidly, Amoghavarsha finally appointed one of his younger sons
as crown prince. This man would eventually rise to the throne as Krishna
II132 and face (with somewhat less success) the challenges of the vast,
unwieldy empire. His life had been one of high drama and low betrayal, of
creative brilliance and hard administrative work, of decades of peace ended
by a decade of war. Nevertheless, by the late 870s, Amoghavarsha had
dominated the Deccan for over sixty years, a feat no other monarch before
him had managed, one that none after him – not even the emperors of
Vijayanagara and Delhi – would be able to repeat. His encouragement of
trade and development of courtly Kannada would long outlive him.
Fortune, it is said, favours the brave, the audacious, but she also favours
the tenacious. Well could Amoghavarsha claim to be Vallabha-Narendra,
the Beloved Lord of Men, the Beloved Indra of Mortals, and Nripatunga,
Pinnacle of Kings, and have his poets declare that ‘the seals of all kings he
has broken with his Garuda seal’.133
What might the people of Manyakheta have thought about this man when
they saw him during seasonal festivals and public rituals? Generations had
come of age under his reign, knowing no Vallabha other than him. His
wrinkled face was perhaps arranged into the calm, benevolent expression he
had learned so many decades ago, dispensing nods and glances and smiles
graciously.
The Vallabha Amoghavarsha, Sarva Rashtrakuta, once a boy who lost his
father, once terrified of the intrigues of older men, had outlasted them all.
To his contemporaries in the subcontinent and the world, he was the
Vallabha, perhaps the most iconic of all Deccan emperors – though his
importance as a South Asian historical figure has faded since.
Amoghavarsha had held the Deccan together at all costs. Would those who
followed be able to do the same?
EI XVIII, 254.
This is based on a description from the Yashastilaka, a near-contemporary Jain text composed under
the patronage of a Rashtrakuta vassal.
Govinda III’s son Amoghavarsha is generally given credit for completing Manyakheta, but this does
not mean he initiated its construction. This decision was probably made by Govinda. See Mishra
1992, 205. Both ‘Manya’ and ‘Kheta’ can mean different things, and the name translates roughly
to ‘Sacred Shield’, ‘Auspicious Shield’ or ‘Venerable Shield’ or even ‘Honoured Weapon’ or
‘Honoured Field’.
While Malkhed is commonly identified with Manyakheta, precious little archaeological work has
been done to verify this.
Assuming that Govinda was Shila-Mahadevi’s son by Dhruva, the earliest possible date for his birth
(as depicted in this book) is 772, making him about thirty at this point. EHD I instead suggests that
Govinda was born after Dhruva’s accession in 780, fought alongside him in north India before he
was ten, led a series of campaigns and had a son at the age of barely twenty in 800, and yet died of
old age in 814.
EI VI, 250.
Davis 1993, 37.
Ibid.
EHD I, 274.
Ali 2006, 70.
Ibid., 71.
Ibid., 87–88.
Ibid., 73.
Ibid., 124.
Ibid., 132.
Ibid.
Ibid., 133.
Ibid., 132.
Ibid., 133.
Ibid., 134.
EI XVIII, 254.
This was Vijayaditya II, son of Vishnuvardhana IV and brother of Shila-Mahadevi. He was an
inveterate enemy of the Rashtrakutas, having earlier been dethroned by Govinda III and replaced
by his younger brother before managing to crown himself.
Though Vijayaditya is recorded as having frequently battled the Gangas, these probably date to the
earlier part of his career while he was Govinda III’s vassal, since they both had a common enemy
in the Rashtrakutas after Govinda’s death and it seems unlikely they would fight each other at that
point. Of course, hostilities may have broken out again after the Vengi Chalukyas and Rashtrakutas
reached an arrangement in 821.
Chakrabarti 2011, Location 1797.
G.R. Rangaswamaiah, ‘The Rashtrakuta Relations with Gangas of Talkad’, in Gopal ed. 1994, 206.
EI XVIII, 254–55.
Ibid.
Karkka was the son of Indra Rashtrakuta, Govinda’s younger brother and right-hand man. He is often
portrayed as having been Amoghavarsha’s regent, but there is little to no evidence attesting to this.
See V.V. Mirashi, ‘The Javakheḍa Plates of Amoghavarsha I: Śaka 742: A Critical Examination of
Some Problems’, Proceedings of the Indian History Congress 23 (1960): 31–36,
http://www.jstor.org/stable/44304005.
EHD I, 274.
Sastri 1994, 173.
Inden 1981, 105.
It may be worth pointing out the irony in an individual who depended on the direct and indirect
labour of hundreds, if not thousands, for his day-to-day activities and consumption saying this.
These quotes are from the Prasnōttararatnamālika, ‘The Jewel-Garland of Questions and Answers’,
attributed to Amoghavarsha. See Mishra 1992, 117–18.
EI XVIII, 254.
Was Amoghavarsha himself truly as introspective as presented here? His inscriptions certainly seem
to have somewhat different priorities than those of his contemporaries, detailing the control of the
senses, the vagaries and duties of kingship, repeated betrayals and abdications, and so on. It is
certainly probable that the various difficulties of his early life had an impact on his psyche.
This is the general conclusion reached by Professor Shonaleeka Kaul in Imagining the Urban:
Sanskrit and the City in Early India (Seagull Books, 2011).
Ibid., 206.
Paul Dundas, The Jains, Second Edition (Routledge, 2002), 119.
Nandi 1973, 47.
Mukund Lath, ‘Somadeva Suri and the Question of Jain Identity’, in The Assembly of Listeners:
Jains in Society, eds. Michael Carrithers and Caroline Humphrey (Cambridge University Press,
1991), 19–30, 29.
Nandi 1973, 59.
Ibid., 61.
Ibid., 2.
Ibid., 52.
EHD I, 279.
Dundas 2002, 120.
Though Mahalakshmi is also worshipped in Jainism, the sacrifice of a finger suggests that
Amoghavarsha was approaching this from the ritual practice of left-hand tantra, generally
associated with Shaivism and Shaktism at this time.
R.V.S. Sundaram, Śrīvijaya’s Kavirājāmārgam., trans. Deven M. Patel (Manohar, 2017), 18.
Kalinga is generally thought of as a name for premodern Odisha, but the area corresponding to
 Odisha actually consisted of a number of smaller kingdoms, including Kalinga, Kongoda,
Dakshina Kosala, and Uttara and Dakshina Tosali.
Handiqui 1949, 26–27. ‘Charger’ has been changed to ‘horse’ for readability.
These are mentioned as important towns in his Kavirajamargam. See Sundaram 2017, 29.
‘The Helgo Treasure: A Viking Age Buddha’, Irish Archaeology (28 December 2013),
http://irisharchaeology.ie/2013/12/the-helgo-treasure-a-viking-age-buddha/. The article cites peer-
reviewed chapters and archaeological surveys.
Edward Pollard and Okeny Charles Kinyera, ‘The Swahili Coast and the Indian Ocean Trade Patterns
in the 7th–10th Centuries ce,’ Journal of Southern African Studies 43, no. 5 (2017): 1–21, 7.
S.A.M. Adshead, T’ang China: The Rise of the East in World History (Palgrave Macmillan, 2004),
68.
Tansen Sen, Buddhism, Diplomacy and Trade: The Realignment of Sino-Indian Relations, 600–1400
(University of Hawai’i Press, 2004), 152.
Ramendra Nath Nandi, State Formation, Agrarian Growth and Social Change in Feudal South India
c. AD 600–1200 (Manohar, 2000), 114.
Ibid., 126.
Mishra 1992, 45.
Ibid., 50.
Ibid., 25.
Ibid., 26.
Malekandathil 2010, 9.
This can be seen in the Quilon copper plates of Sthanu Ravi, where the Syrian Mar Sapir Iso is
mentioned as having founded a town. Witnesses include members of the Manigramam as well as
the chief of Venad and a local militia group. See Y. Subbarayalu, South India Under the Cholas
(Oxford University Press, 2012), 177–78.
Amira K. Bennison, The Great Caliphs: The Golden Age of the ʿAbbasid Empire (Yale University
Press, 2009), 140.
Ibid., 140.
Malekandathil 2010, 21.
Chakravarti 2012, 84.
Malekandathil 2010, 21.
Sastri 1939, 124.
Elizabeth Lambourn, ‘Describing a Lost Camel: Clues for West Asian Mercantile Networks in South
Asian Maritime Trade (Tenth–Twelfth Centuries ce),’ in Ports of the Ancient Indian Ocean:
Proceedings of the Kolkata Colloquium 2011 (Median Project), eds. M.-F. Boussac, J.-F. Salles
and J.-.B. Yon (Primus Books, 2016), 366.
Ibid.
EI XXXII, 47.
Lambourn 2016, 368.
EI XXXII, 47.
Ranabir Chakravarti, ‘Merchants, Merchandise and Merchantmen in the Western Seaboard of India:
A Maritime Profile (c. 500 bce–1500 ce),’ in The Trading World of the Indian Ocean, 1500–1800,
eds. Om Prakash and D.P. Chattopadhyaya (Pearson Education, 2012), 99.
For a discussion on how this was circumvented, see Lambourn 2016.
Gautam Jantakal, Bhushan Kapadia and Prakash Jinjuvadiya, History and Coinage of the
 Rashtrakutas (IIRNS Publications LLP, 2019),
34–35.
This mostly involved various small-scale boiling and refinement.
Sen 2004, 150.
Adshead 2004, 81.
Mishra 1992, 147.
Ibid., 152.
Ibid., 170.
Chakravarti 2012, 97.
Ibid., 86.
Kurush F. Dalal, ‘Sanjan: Digging Deep Into History’, Live History India, 22 September 2019,
https://www.livehistoryindia.com/story/cover-story/sanjan-digging-deep-into-history/.
Rhea Mitra and Kurush F. Dalal, ‘A Report on the Glass Vessels from Sanjan, 2002’, Journal of
Indian Ocean Archaeology 2 (2005).
S.P. Gupta et al., ‘Preliminary Report on the Excavations at Sanjan (2002)’, Puratattva 32 (n.d.).
Mishra 1992, 172.
Bennison 2009, 145.
Mishra 1992, 170.
Burjor Avari, Islamic Civilization in South Asia: A History of Muslim Power and Presence in the
Indian Subcontinent (Routledge, 2019), 30.
Alberuni, Alberuni’s India, trans. Edward C. Sachau (Kegan Paul, Trench, Trubner & Co., 1910),
xxxi–xxxii. See also Avari 2019, 32.
Gopal Stavig, ‘Congruencies Between Indian and Islamic Philosophy’, Annals of the Bhandarkar
Oriental Research Institute 81, no. 1/4 (2000), http://www.jstor.org/stable/41694615.
Devices known as ‘yantras’ are mentioned in the Yashastilaka of Somadeva, a Jain ascetic writing in
the tenth century who appears to have modelled his description of a royal court on that of
Manyakheta. In the eleventh century, these ‘yantras’, corresponding to much better-attested
automata in the courts of Constantinople and Baghdad, are mentioned in literature produced by the
Malwan ruler Bhoja of Dhara. All these hint at Indian courts importing Abbasid technology. See
Daud Ali, ‘Bhoja’s Mechanical Garden: Translating Wonder across the Indian Ocean, circa 800–
1100 CE’, History of Religions 55, no. 4 (2016).
Ibid., 171.
Chakravarti 2012, 99.
Ibid., 92–93.
Ibid.
Sastri 1939, 125.
Ibid.
Mishra 1992, 205.
Ibid., 206.
Mishra 1992, 190–91.
Pollock 2006, 343.
Ibid., 330.
Ibid., 343.
Ibid., 344.
Sundaram 2017, 28.
Pollock 2006, 349.
Sundaram 2017, 31.
Ibid., 32.
Ibid., 35.
Ibid., 45.
Pollock 2006, 344.
Ibid., 342.
Ibid., 332.
Ibid., 339.
Ibid., 341.
Ibid., 337.
Suryanath U. Kamath, ‘Amoghavarsha I: An Appraisal,’ in Gopal
ed., 84.
Rangaswamaiah in Gopal ed. 1994, 206.
Sastri 1939, 123. This claim was based on a misunderstanding of Rashtrakuta territories as extending
all the way to China, so it should be taken with a pinch of salt. Interestingly, the Arabs claim that
this is a classification that originated in India and China.
EI VI, 35.
Ibid., 35.
Mishra 1992, 107–08.
EI VI, 35–36.
Ibid.
Vedam Venkataraya Sastry, ‘The Rashtrakutas and Vengi and Cultural Contacts’, in Gopal 1994, 173.
Ibid. It is quite possible that given Vijayaditya III’s later activities he was nowhere near as crushed by
this Rashtrakuta victory as the Rashtrakutas make him out to be.
Rangaswamaiah in Gopal ed. 1994, 207.
EHD I, 260.
EI VI, 35.
 Part III
Twilight: The Kalyana Chalukyas and
the Chola Empire
 8
King of Kings
Spring, 959 CE

We return to the medieval Deccan nearly a century after the career of

Amoghavarsha I.
Through this intervening period, South Asia saw an intensification of the
trends we have witnessed so far. Courtly Kannada literature, especially by
south Indian Jain writers, continued to be produced in ever-larger quantities.
The Rashtrakuta imperial family involved themselves deeply in the politics
of central India through intermarriage with the Kalachuri family of Tripuri,
based in modern Madhya Pradesh and Chhattisgarh. The Gangas of
southern Karnataka also gradually became the Rashtrakutas’ closest allies
as a result of generations of intermarriage, a trend Amoghavarsha I had
inaugurated.
The Rashtrakutas also gradually regained an overwhelming influence over
the politics of Vengi: many Vallabhas would sack the region, often
encouraging bloody dynastic conflict among the Chalukyas of Vengi to
ensure their unquestioned dominance. However, the aristocratic houses of
the Deccan grew more powerful, often involving themselves in these
political struggles, joining factions opposed to their Rashtrakuta overlords,
and on occasion successfully intervening in succession crises within the
imperial family.
To their north, the influence of the Pratiharas initially rose rapidly under a
series of energetic rulers. Kannauj, Harsha’s erstwhile capital commanding
the eastern Gangetic plains, became one of the major centres of their power.
However, by the early tenth century, the Rashtrakuta emperor Indra III once
again led a Deccan army north and sacked the city and its environs, dealing
a mortal blow to Pratihara power. Their ensuing decline led to the
establishment of smaller regional kingdoms in western India, some of
which, such as the Paramaras of Malwa, were loosely integrated into the
Rashtrakuta imperial network through ties of tribute. Further east, the Palas
of Bengal also lost their predominance over the eastern Gangetic plains,
with their ostensible vassals having far more influence and power in the
region.
To the south of the Rashtrakuta imperial formation, in Tamilakam, the
Tamil country, Pallava power had significantly eroded. Once the region’s
predominant dynasty, it was now under threat from innovative new polities:
the Cheras, trading with the western Indian Ocean from their base of
Mahodayapuram (modern Kodungallur) in what is now Kerala; the Pandyas
of Madurai; and an ambitious dynasty based in the Kaveri river valley,
calling themselves the Cholas.
By the late tenth century, all the great imperial houses of the early
medieval period were in terminal decline, waiting only for an ambitious
vassal to overthrow them at last and create new regional imperial states of
their own. The Rashtrakutas appeared to be the sole exception to the rule.
The powerful Vallabha Krishna III, succeeding a string of weak and
incompetent rulers, had managed to convince the aristocratic houses of the
Deccan to join him in a series of campaigns north and south, propelling the
Rashtrakuta imperial formation to its territorial apogee. As it would turn
out, the empire was nowhere near as strong as it appeared: Krishna III’s
vassals, the aristocratic houses of the Deccan, were merely biding their time
and waiting for an opportunity to seek their own fortunes on the battlefield.
As we witness the final collapse of this imperial dynasty, we will see how
the actions of the Rashtrakuta Vallabhas inadvertently created the political
conditions needed for these new regional kingdoms to rise in south and
central India. And as it fades in confusion and chaos, we will see how, in a
world very different from that in which the earliest Chalukyas had
established their power in the Deccan, a new dynasty calling themselves the
Chalukyas would be established as an innovative, resilient, new Deccan
superpower.

In March 959, Amoghavarsha’s descendant, the Rashtrakuta emperor

Krishna III, Gandamartandaditya (Sun among Sun-like Warriors), built a
victory-pillar1 and two temples to gods named after himself –
Krishneshvara and Gandamartandadityeshvara – on the southernmost tip of
the subcontinent.2 Rampaging over the Tamil lands, his armies had brutally
slain the crown prince of the up-and-coming Cholas, a local Tamil dynasty,
at the battle of Takkolam in 949. In the ensuing decade, Krishna III outdid
all previous Vallabhas: whereas they had been content with subduing and
extracting tribute from the Tamil kings, Krishna divided the Tamil lands
among his vassals and followers, perhaps with the intention of creating a
new power base loyal to him and preventing any new imperial power from
rising in the region. This conquest was commemorated through inscriptions
in Kannada,3 not Sanskrit, previously the premier language of power in the
south.
A few years later, the Rashtrakuta emperor led an expedition into Malwa
alongside his relation,4 the Ganga king Marasimha II. As a reward,
Marasimha – descended from a family once implacably opposed to the
Rashtrakutas – had proudly accepted from Krishna the title of
Gurjaradhiraja, Supreme King of the Gurjaras, claimed the honour of
personally organizing and protecting the emperor’s camp, and awarded two
of his generals the title of Ujjenibhujanga, Conqueror of Ujjain. This attack
of Krishna’s would be responsible for the spread of Kannada inscriptions to
the furthest north they would ever reach, near Jabalpur in modern-day
Madhya Pradesh. Evidently, a century after Amoghavarsha Rashtrakuta, the
inscriptions of south Indian kings now sought to speak to their vassals in
their very own language of prestige, no longer bothered with the distant,
hallowed world of the Sanskrit Cosmopolis.5
Krishna III’s title, Gandamartandaditya, Sun among Sun-Like Warriors,
was a compound of Kannada and Sanskrit, only allowed by the
grammarians – following the lead established by his ancestor
Amoghavarsha I – in royal titles such as this.6 By his death in 967, it
seemed that no other king had ever been the overlord of the subcontinent’s
entire southern peninsula as completely as Krishna III Rashtrakuta.7 And
yet, in 972, five years after his death,8 the great Rashtrakuta capital of
Manyakheta, which had awed south Indian kings and Arab travellers for
nearly two centuries and sent out armies to ravage much of the
subcontinent, was sacked for the first time.
On the banks of the Narmada, at the ford called Kalighatta9 near modern
Thalghat10 – three and a half centuries after the Chalukya Vallabha
Pulakeshin II had humbled Harsha, lord of the north, on the same shores – a
new Harsha faced the armies of the Deccan. He was called Siyaka ‘Harsha’
Paramara, king of Malwa in Madhya Pradesh, of a dynasty of military
aristocrats once subservient to the Pratiharas of northern India. The
warlords of the Deccan had sacked his capital at Ujjain barely a decade
earlier11 and forced him to prostrate before Krishna III. Now that Krishna
was dead, succeeded by his elderly and incompetent brother, Siyaka’s time
had come.
At Kalighatta, the Rashtrakuta armies tried to hold off Siyaka’s bands of
boat-riding troops, killing one of his generals in hails of javelins and
flaming projectiles.12 But the Paramara king, coming up with
reinforcements, outmanoeuvred the Deccanis and inflicted a devastating
defeat on them.13 This defeat would slowly snowball into one of the most
unprecedented military disasters in the subcontinent’s entire history.
 A man bearing the title of ‘Harsha’ had finally crossed the Narmada and
entered the Deccan.
The subcontinent must have reeled in shock as news of the disaster began
to spread. As we have seen, the tenth century had been one of war and
tumult in northern India as the Palas and Pratiharas declined and new
regional kingdoms emerged. Now chaos was coming to the Deccan as well.
Perhaps the Malwans rushed through the gates like a river through a
broken embankment, pouring down the royal avenue.14 They stormed past
whitewashed mansions with balconies and terraces from where aristocrats
had watched parades.15 They saw the splendid elephant racecourses of the
city, where the Rashtrakuta emperors, preceded by troupes of dancers, came
to enjoy the cinnamon scent of the tuskers’ musth, the cheers of their
subjects.16 Perhaps they paused before sacking the dozens of temples that
decorated Manyakheta. What would they have thought of the beautiful
tiered southern architecture unfamiliar to their eyes?
Everywhere, like other armies across the medieval world, and as implied
by Indian inscriptions, they probably ran about in gangs in the
neighbourhoods, slaughtering, raping, plundering. They may have pried out
the gold panels, made piles of loot out of family heirlooms, and assaulted
their captives, gloating over their power. ‘Do not the hand devoid of
wristlets, the breast devoid of necklace, the eyes deprived of collyrium
[kohl], the ear without the ear-ornament, the waist bare of girdle and the
tender-leaf-like feet wanting the anklet, of the wives of his enemies …
bespeak the heroism, the overpowering capacity, and the prosperity of the
king …?’ asks a medieval Deccan poet, offering us an insight into the
attitudes of the warlike aristocrats of the period.17
It must have been surreal for King Siyaka Paramara to enter the imperial
palace, outside whose gates crowds of vassal lords had once gathered to the
beat of the Rashtrakuta drum. He strode past its intricate series of
courtyards, accompanied by a platoon of guards and senior generals,
including his son Vakpati Paramara, Lord of Speech, a poet and capable
military leader. In those courtyards, Siyaka had once waited in humiliation
with other subjugated kings as the Vallabha’s officers displayed captured
horses and elephants before them,18 waiting for the palace chamberlains to
strike the floor with their golden staves and admit them to the next
courtyard and the next. This continued until they finally reached the
imperial court where they had to gaze upon their Rashtrakuta overlord with
compulsory adoration,19 and should they be called to do so, touch their
crowned heads to his feet.20
Siyaka inspected the offices and sub-offices of the Manyakheta palace,
and had their records, carved on copper plates, seized.21 What might he
have seen as he explored and ransacked the palace? Rashtrakuta literature
speaks of exquisite murals and carvings and lush gardens: flowers,
waterworks, artificial hillocks, and trees carefully manicured and
arranged.22 There were pleasure pavilions and bathhouses with exquisite
fountains and tanks, water-powered automata ingeniously crafted into the
shapes of lotuses, animals and women: hydraulic marvels23 inspired by
developments in Baghdad and Constantinople. Perhaps the Malwan king sat
on Amoghavarsha Rashtrakuta’s lion throne, gloating in his new-found
power and the woe of his vanquished enemies.
Within weeks, Ganga troops, led by Krishna III’s loyal general
Marasimha II, had chased Siyaka out of the Deccan.24 But it was too late.
Fortune had finally abandoned the Rashtrakutas, though there would still be
Rashtrakutas claiming the title of Sri-Prithivi-Vallabha for years after.25
However, with neither an army nor the great capital and its wealthy
surrounding territories to support them and enforce their authority over the
Deccan’s kingdoms, nobody paid them any attention.
The sudden evaporation of the Rashtrakuta dynasty might appear
surprising given that it seemed to have reached its apogee under Krishna III
less than a decade ago – but it also reveals how military and political
successes are an inadequate lens through which to study historical change.
A nuanced understanding of the power relations between the Rashtrakuta
court and its vassals, situated within the changing political economy of the
tenth century, awaits the discovery of more primary sources and
archaeological evidence from the period. Until then, the collapse of the
Rashtrakuta Vallabhas will remain a dramatic tale of hubris, imperial
overreach and the fragility of even the mightiest dynasties in the face of the
tides of history.
In the aftermath of the sack of Manyakheta and the collapse of the
Rashtrakuta imperial house, a man called Taila emerges onto the historical
scene. As for who exactly he was and where he came from, we have only
fragmentary evidence and the unreliable testimony of his descendants. He
was a high-ranking officer in Krishna III’s army and did very well for
himself through the emperor’s reign, rising to the rank of raja, and assigned
the status of guard over a hefty thousand-village district in the vicinity of
Manyakheta.26 He was also connected to a minor Rashtrakuta line through
his wife,27 a marital alliance that may well have been arranged by the
Vallabha to honour this loyal subordinate of moderate means and prestige.
Taila would later claim to be descended from one of the many Chalukya
families that still survived in the Deccan, some of them minor branches of
the Vatapi line, others that seem to have used the name mostly because of
its prestige and allure. Whether or not his ancestors were actually one of
those long-forgotten kings, it is difficult to say, but as one of the
Rashtrakuta emperor’s military officers, Taila was likely well connected in
the Manyakheta court, understood how the cut-throat world of high politics
worked, and personally knew many of the warlords who would now emerge
to stake their claim to be Fortune’s Favourite.
Soon after Siyaka’s departure from Manyakheta, Taila seems to have
managed to seize the city, thus guaranteeing his own eminence in the new
political order that would inevitably emerge.28 As war whittled away other
claimants and his confidence grew, a large faction gradually began to gather
around him, claiming that Taila was a Chalukya. However, all concrete
memories of that ancient line – their fantastic prashasti claims of being
nourished by the breasts of the Seven Mothers29 and their Sanskrit verses on
each king’s reign and descent – seem to have faded by this time. Only the
Chalukya name, the boar banner and the iconic ‘Victorious is Vishnu made
manifest as the Boar!’ invocation still remained in popular memory, ‘like
dimly remembered formulas of a lost heroic language’.30 In medieval south
India, though, remembered histories were a powerful and potent political
force, and it was something that Taila’s new ‘Chalukyas’ would exploit with
great success.
 Though his descendants’ inscriptions (and some modern historians) have
construed Taila’s rise as inevitable, this was by no means guaranteed in the
late tenth century. He was far from the largest fish in this pond – or even the
most powerful contender of ostensibly Chalukya descent. That distinction
belonged to Rajaditya Chalukya, apparently an actual descendant of that
ancient family, who was married to one Ganga and one Rashtrakuta
princess,31 and the foremost leader of the forces arrayed against a
Rashtrakuta restoration. Opposing him was the Ganga king Marasimha II,
leader of the pro-Rashtrakuta faction represented by a grandson of Krishna
III.
Meanwhile, Siyaka’s son and heir Vakpati Paramara, sacker of
Manyakheta, master of Malwa, claimed the titles of Maharajadhiraja, Sri-
Prithivi-Vallabha and Amoghavarsha,32 comparing his own poetic talents to
those of the celebrated Deccan monarch of the ninth century. He would
cross the Narmada multiple times to raid the Deccan,33 attempting to carve
out an empire that spanned the great river. Rounding out the list of
contenders to the overlordship of the Deccan was yet another Rashtrakuta
princeling, propped up by two powerful hereditary ministers.34 And in every
part of the Deccan, there were now practically independent regional
kingdoms that were hardly enthusiastic about returning to an imperial
network with distant Manyakheta at the centre.
 Precisely how Taila went about consolidating his position is unclear. As
with most events of the time, all that survives are flashes of complex
warfare and politicking that lasted till the very end of the tenth century.
Taila appears to have drawn to himself similarly upstart, ambitious men of
aristocratic backgrounds, often provincial governors or landed magnates35
who were richly rewarded with titles and wealth in return for their personal
loyalty to him and the efforts they undertook on his behalf. The upper
echelons of Taila’s faction were highly militarily competent: one of them
bore the title of Giri-Durga-Malla, Wrestler of Hill Forts.36 They worked
together to cajole or otherwise convince remnants of the Rashtrakuta
imperial formation to accede to a reworked ‘Chalukya’ imperial formation.
As one might imagine, the creation of this new polity was a bloody process,
and this is attested to in many of Taila’s inscriptions. The picture that they
suggest is that Taila’s faction astutely chose its moment to strike, waiting
until after the biggest contenders had eliminated themselves.
This moment came in 974. At the beginning of that year, the Ganga king
Marasimha II seemed to be at the peak of his power. His territories had
vastly expanded due to lands granted to him by Krishna III in gratitude37 for
his participation in imperial campaigns in the north and south, and even
securing the submission of some Tamil kings. Had Marasimha prevailed, he
might have reinstalled a puppet Rashtrakuta emperor at Manyakheta – but it
was not to be. He and the most powerful Chalukya claimant, Rajaditya,38
fought a gruelling, exhausting campaign around the great citadel of
Ucchangi (near modern-day Davanagere, Karnataka). The horrors inflicted
by both sides seem to have eliminated Rajaditya from contention and
convinced this great Ganga king to give up his ambitions. An ordained Jain,
Marasimha gave up his crown and spent a year moving across his territories
before ritually starving himself to death at the feet of his preceptor
Ajitasena in a three-day fast in Bankapura39 near modern-day Hubli-
Dharwad. This was the very same city founded by Amoghavarsha
Rashtrakuta’s general Bankeya. ‘Aho! Chola king, quiet down by gently
rubbing thy palpitating heart! O Pandya, give up weeping! O Pallava, run
not away in fear; O retreat not from thy territory, but remain …! The Ganga
chieftain … has gone in triumph to the abode of the gods!’40 wailed a poet.
Soon after, Taila and his army arrived on the borders of Gangavadi proper,41
and subdued it.

By 980, Taila was the acknowledged Chalukya overlord of much of

northern and southern Karnataka, and his armies had quelled the families
that controlled the Western Ghats.42 Shortly thereafter, his generals attacked
Goa and forced its rulers to surrender. With
most of the southern and western Deccan now subservient to the Chalukya
Varaha, Taila finally turned to deal with the last powerful remaining
claimant to the lordship of the Deccan, a man whose sacking of
Manyakheta had caused the anarchy pervading the plateau: Vakpati
Paramara, the north Indian who dared to claim the title of Prithivi-Vallabha.
Realizing that Taila had emerged as the dominant force south of the
Narmada, Vakpati, sometime around 994–996, decided to gamble it all in an
open confrontation.43 Raids were no longer enough to maintain his claims to
Deccan overlordship. Taila had to be defeated, and Manyakheta, which
Vakpati’s father Siyaka had sacked in 972, had to be permanently occupied.
This would lead to a showdown so memorable that it would pass into oral
legend and eventually be included in a Sanskrit storybook by the Jain monk
Merutungacharya in fourteenth-century Gujarat. According to this work,44
when Vakpati decided to mount a full-scale invasion of the Deccan, his
minister45 warned him that disaster was waiting for him across the
Godavari. The king ignored the advice and set out, and the distraught
minister immolated himself.46 Should this be even a dim memory of real
events, it seems that Vakpati was betting he would be able to beat Taila
decisively deep in the Deccan. But his dead minister was right: this was a
suicidal mistake.
As the Malwan armies marched across the Narmada, the Vindhyas and the
Godavari, into the heart of the Deccan, a trap was sprung. The Seuna
Yadavas of northern Deccan, who had accepted a position as Taila’s vassal
kings, swung in and cut off Vakpati’s supply lines and channels of
communication to his core territories in Malwa.47 The Malwan king was left
at Taila’s mercy. Vakpati’s army was annihilated and the Paramara king
captured.
And so one of the biggest military blunders in medieval Indian history
came to a close. The supremacy of Taila’s ‘Chalukyas’, and their claim to
the title of Sri-Prithivi-Vallabha Maharajadhiraja, was indisputable: the
Gangas, Rashtrakutas, Paramaras and other Chalukyas had all been crushed,
and the Silaharas, Seuna Yadavas, minor Rashtrakuta lines and dozens, if
not hundreds, of other political groups and clans had all yielded.
Krishna I Rashtrakuta, nearly two centuries earlier, had described his
defeat of the Vatapi Chalukyas as his having ‘forcibly wrested away’ on the
‘battlefield that was the auspicious hall of marriage’ the Goddess of the
Royal Fortune.48 The Seuna Yadavas, vassals of Taila’s new Chalukyas,
now described how they gave Fortune ‘a sound thrashing’49 for associating
with Vakpati and ‘forced her to take to the life of an obedient house-wife’50
in Taila’s palace. (These, incidentally, were the same Seuna Yadavas who
would be defeated by Alauddin Khilji centuries later and have their capital,
Devagiri/Daulatabad, taken over by the Tughlaq Sultans of Delhi.)
Vakpati came to an unpleasant end. After his capture, legend tells us that
he was stripped, put in a cage, paraded through the streets of Manyakheta,
and forced to go from household to household begging for food.51 If this is
true, he must have been bound and chained and carefully watched by guards
through this humiliation. It is unlikely that anyone was so forgiving as to
feed the son of Siyaka Paramara, the man who had sacked their city.
And so, after a miserable and humiliating few days, the unrecognizable
Malwan monarch, his once-glorious moustache torn and yanked, his fair
skin probably splattered with mud and excrement and blistering in the harsh
Deccan sun, was dragged to a public execution site on the city outskirts.
Here he was brutally impaled on a stake52 to the roar of cheering crowds
before being beheaded. His rotting head, covered with ‘thick sour milk’,53
was put on display for visitors to the new court of Manyakheta, leaving
them in no doubt as to the power and intentions of the ‘Chalukyas’ who had
returned to claim the mantle of Fortune’s Favourite.

These had been momentous decades for the Deccan, but also for the world.
Even as regional kingdoms had emerged across northern India over the
tenth century, the Abbasid Caliphate headquartered in Baghdad had faded,
and governors of various parts of the far-flung Abbasid empire turned their
provinces into power bases of their own.54 In 969, Egypt became the seat of
a new Caliphate:55 the Fatimids, supposedly descended from the daughter of
the Prophet Muhammad, Fatima, and her husband, the Imam Ali.
The Fatimids soon founded a new capital city that would rival
Manyakheta and Baghdad in splendour. This was al-Qahira,56 Cairo – which
has certainly fared better than these two rival cities in its thousand-year
history. The Fatimid Caliph took up court there in 973,57 the year after the
Paramaras’ sack of Manyakheta. The Fatimids nurtured Egypt’s fertile
agrarian tracts and encouraged international trade. Cairo’s markets boomed.
The ancient trade routes of the Red Sea, where fleets of hundreds of Roman
ships had once gathered in the early centuries CE to be blown to India by the
gale-force winds of the southwest monsoon,58 once again roared back to
life.
However, it was in many ways China that emerged as the ‘world axis’ of
the time, and its interaction with the rest of Afro-Eurasia would accelerate
the formation of a ‘thickening world system’.59 A new dynasty, the Song,
rose to power around 960, inaugurating a period of artistic, intellectual and
technological achievements – including, for example, large-scale iron
manufacturing verging on an ‘industrial revolution’,60 and the invention of
gunpowder.61 In addition, ‘new varieties of crops were introduced into
China from Southeast Asia, improved irrigation machines and techniques
spread rapidly, and the Chinese population started migrating towards the
fertile southern region of the country’.62 China’s population would grow
‘almost fourfold’ from 32 million to 121 million during 961–1109,
urbanizing all the while.63
To feed these ravenous new cities and provide them with the good things
of life – ‘frankincense, sandalwood, black pepper and cloves’64 – and luxury
goods such as rose water, coral, glass, printed textiles and carved ivories –
private enterprise was encouraged,65 with spectacular results. Merchants
from across the world flocked to the coasts of China in the tenth century.66
Goods from far-flung regions arrived here in unprecedented quantities.
The share of international commerce in China’s state revenue exploded to
nearly twenty times67 what it had been under earlier dynasties. In 987, the
Song court sent out four missions ‘invested with imperial authority … to
foreign countries … to come more frequently to the Chinese ports on the
promise of special facilities and import licences’.68 These missions travelled
west by sea, first visiting the archipelagos of Southeast Asia, dominated by
the Srivijaya confederacy: a loose-knit alliance of trading cities that
stretched from Sumatra through the Malay Peninsula, led by Srivijaya
(modern-day Palembang in Indonesia), which gave the confederacy its
name. The confederacy’s many allied fleets dominated the Straits of
Malacca, the gateway to the Indian Ocean.
After crossing the straits, the next stop for the Song ambassadors was
southern India. The arrival of these missions in the region, coinciding with
the rise of the Fatimids and the Chalukyas, would inadvertently assist the
emergence of a polity that profoundly impacted the history of the Indian
Ocean.

In Tamilakam, the Rashtrakuta Vallabha Krishna III’s killing of the Chola

crown prince in 949 had consequences which rapidly took on a life of their
own. Krishna III had, as mentioned, redistributed lands in the Tamil country
to his Deccan vassals, significantly eroding the power of minor local
dynasties and the Pallavas alike. After the death of the Vallabha in 967, his
vassals retreated from the Tamil country to compete in the Deccan political
arena; and the remaining Tamil dynasties rapidly expanded into the political
vacuum. The Cholas, now led by the energetic younger brothers of the dead
crown prince, rapidly regained much of their lost territories in the Kaveri
river valley, and expanded into the erstwhile Pallava heartland. To their
east, the Chera dynasty of Mahodayapuram (modern-day Kodungallur)
controlled the thriving ports of the region corresponding to Kerala today,
while to their south, the Pandya dynasty of Madurai dominated the
southeast tip of the subcontinent. All these medieval dynasties identified
with ancient patrons of Tamil Sangam poetry from the late centuries
BCE/early centuries CE, thus connecting themselves to a long-vanished
‘classical’ age.69 (As with Taila’s ‘Chalukyas’, there is almost no way to
verify these unlikely claims.)
Any would-be Tamil superpower also had to grapple with the autonomy
of hundreds of smaller polities, village assemblies, local magnates and
Brahmin settlements70 that had emerged over the centuries. Tamilakam’s
multicentric political system may well have continued for centuries, just as
the Deccan could well have shifted to a pattern of subregional kingdoms
after the Rashtrakuta collapse. But in both cases, ruthless competition led to
the emergence of leaders of extraordinary political and military ability.
In July 985, when the Cholas had finally gained the upper hand over the
Pandyas and emerged as the primary power in the northern and eastern parts
of the Tamil country, a prince by the name of Arulmozhivarman acceded to
the throne, adopting the regnal title of Rajaraja, King of Kings. The
screaming of trumpets, the clanging of bells and chanting of priests at the
ceremony may have all seemed to portend a subcontinent – and a world –
that would be changed forever.
Rajaraja Chola understood that domination of lucrative trade routes was a
sure way to distinguish himself and his court from the other fragmented
polities of the Tamil country. But he soon learned that rival polities to his
east were suddenly receiving more and more traders from across the seas.
This was because traders from prosperous Fatimid Egypt were now
reaching the Malabar coast, under the control of the Chera dynasty. Within
a few years of his accession, the young71 Chola king moved to seize the
riches of the region by attacking the great port of Kandalur,72 ruled by a
Chera vassal. This happened around the same time as the arrival of the first
embassies from Song China to the Indian Ocean region, and its timing
cannot have been a coincidence. What was possibly south India’s largest
collection of ships at the time was gathered at Kandalur port. The fleet
probably numbered at least a few hundred, and would primarily have been
composed of vessels owned by merchant corporations. Rajaraja appears to
have caught it at its most vulnerable – in harbour, perhaps preparing to set
out for that year’s trading season – and ordered it burned.
It must have been a spectacular and horrifying sight. Medieval Indian
texts describe ships painted white, red, yellow and blue; with prows shaped
like lions, buffaloes, snakes, elephants, tigers, birds and people; and with
sails of white, red, yellow and black.73 As the burning cloth snapped in the
sea breeze, scattering sparks and smoke, the unbearable heat spread from
one ship to another. Masts must have collapsed, teakwood cracked and
slipped under the roiling waves, probably to the cheers of thousands of
Chola soldiers as they ransacked the populace and held back weeping
merchants at spear-point. Rajaraja had seized a colossal loot and violently
established himself and the Cholas as one of the rising powers of the
southern tip of the subcontinent. Trade would only be allowed to flow if
merchants reached an accommodation with the Cholas, it seemed.
It was a sign of things to come. Over the next decade, Rajaraja Chola
would display an extraordinary grasp of political and military strategy,
tactics and operations. If an opportunity presented itself, Rajaraja Chola
could be relied upon to exploit it – and to use it to send a message. By the
990s, inscriptions suggest he had overrun much of the erstwhile Pandya
territories and appointed his own governors there.
Soon after, a merchant informed him of a mutiny of mercenaries from
Karnataka and Kerala, who served as the private army of the Sri Lankan
king. Lanka at the time was quite deeply integrated into south Indian
movements of goods, labour and capital. Rajaraja immediately moved to
take advantage of the chaos in the island. A Chola army penetrated deep
inland and completely sacked the nearly thousand-year-old Sinhala capital,
Anuradhapura. The chief queen74 was captured and sent to Rajaraja, as were
the royal jewels, crown and sword.75 Northern Lanka was then savagely
looted: stupas were broken apart and golden images, reliquaries and
offerings within stolen.76 A temple to Shiva was set up after the capture of
the city of Polonnaruwa.77 The ransacking of the great Buddhist viharas –
some of which were home to hundreds of monks and supported by the
produce of villages and bonded labour,78 just like temples in India – was so
comprehensive that later Sinhala chroniclers would describe the Chola
troops as ‘blood-sucking yakkhas [capricious nature-deities]’.79 Lanka, like
the Pandya country, would now also become home to a permanent Chola
presence as Rajaraja built a number of Shiva temples and established a
garrison at Polonnaruwa.
After the conquest of Lanka, the Cholas dominated not only the southern
tip of India, but also the Palk Strait, the gateway between the eastern and
western Indian Ocean. Control of this crucial geopolitical crossroads
allowed Rajaraja Chola, and merchant guilds connected to him, to extract
profits from the movement of goods between East, West and South Asia.
The death of the Ganga king Marasimha II in 974 provided Rajaraja
another opportunity to expand. He now made an attempt to seize control of
the trade routes between the southern Deccan and Tamilakam. His generals
first conquered the many hill forts that dotted Kongu, the drylands where
the rocky plateau transitioned into the lush green fields of Tamilakam. In
991, Rajaraja himself was in Gangavadi.80 Encountering no opposition, he
headed further north, only to find Taila Chalukya, by now the Deccan’s
dominant ruler, heading towards him at speed at the head of an army. In this
first major clash between Chalukya and Chola, the Chola was thrashed,
apparently losing one hundred and fifty war elephants81 to the older and
more experienced Chalukya. This was a serious setback, especially given
the close ties between a warlord’s prestige and the size of his elephant
corps. But Rajaraja was given a reprieve: before Taila
could decisively deal with the brash young Tamil king, he was forced to
turn north to repel Vakpati Paramara, thus leaving Gangavadi in Rajaraja’s
hands. Nevertheless the defeat humiliated Rajaraja: he supposedly vowed
he would never hunt again until he had sacked Manyakheta.82
Taila Chalukya died in 997, having regained some of the Deccan empire’s
lost territory and founded a new imperial dynasty. He was succeeded by his
son Satyashraya – a title apparently meant to connect him to his illustrious
purported ancestor, Pulakeshin II, who had used Satyashraya, Refuge of
Truth, as one of his titles. By this point, the Cholas had successfully overrun
most of Gangavadi,83 which had so heroically resisted generations of
Deccan Vallabhas; they had also wiped out the Pallavas, dominated the
Cheras and Pandyas, and conquered much of Sri Lanka. The hill chiefs and
kings of the southern Deccan were gradually being integrated into the
burgeoning Chola state; indeed, one can still find Chola temples to Shiva in
modern-day Bengaluru. Never before had the Deccan faced such a
dangerous foe. And the Cholas were only just getting started.

Temporarily stymied by Chalukya power to his northwest, Rajaraja Chola

cast about for easier targets. In Vengi – crucial for Chola dominance of east
coast trade – he found one, taking advantage of the mess the Rashtrakutas
had left behind. By persistently interfering in the politics of Vengi for nearly
two hundred years, the Rashtrakutas had created a situation of perpetual
dynastic strife within the Chalukya dynasty of Vengi – the only truly royal
descendants of the old Chalukyas of Vatapi. In the aftermath of a civil war,
two Vengi Chalukya princes fled south to the Chola court,84 and Rajaraja
was only too pleased to intervene on their behalf. He invaded Vengi and
installed the elder prince as king. The younger remained in the Chola court
and was married to Rajaraja’s daughter.85 Their children’s story, as we’ll
see, would be tied to the ultimate doom of Rajaraja’s male line.
By 1000 CE, Chola military and political successes had responded to and
influenced crucial long-term trends in global history. The achievements of
Rajaraja Chola, this singularly ambitious and capable man, would have
ramifications for centuries. Thanks in part to his career, powerful, stable
and commercially inclined polities now controlled each of the nodes and
access routes to the thriving trade networks of the Indian Ocean: the Red
Sea by the Fatimids; the Palk Strait by the Cholas; the Malacca Strait by
Srivijaya; and China by the Song dynasty. International trade boomed and
Chola coffers swelled.
The wealth this dynasty had gained from their military success and the
control of trade is readily apparent in the gigantic Rajarajeshvaram
(Rajaraja’s Lord) temple, today known as the Brihadishvara or ‘Big’ temple
in Thanjavur, which Rajaraja endowed in 1003.86 This building was so huge
that it contained about forty times as much stone as the average Chola
temple,87 and its construction is a testament to the scale of the resources
Rajaraja was able to mobilize. Over the next decade, until his death, the
Rajarajeshvaram would be showered with the wealth of Chola conquests
north, south and west. It was, as the Krishneshvara at Ellora was for the
Rashtrakutas, a potent statement of the abilities and vision of this dynasty.
The gifts that the king, his elder sister, his vassals, his army regiments and
the temple women dedicated to the god added up to thousands of kilograms
of gold and silver, alongside hundreds of valuable gemstones88 and dozens
of
trumpets, parasols and other royal and military paraphernalia.89 These had
been seized from the newly arisen Chalukyas in the Deccan, the Buddhist
monasteries and stupas of Sri Lanka, the ports of Kerala, and the sacked hill
forts and towns of recalcitrant south Indian chiefs and kings. Luxurious and
exotic gifts to Rajaraja’s Lord – incense, camphor, musk – were sourced
from merchants increasingly connected to globe-spanning trade networks.
The temple itself was built by importing thousands of tonnes of granite
from higher up the Kaveri river valley at great expense,90 and maintained
with produce from villages across Chola dominions, even as far away as Sri
Lanka.91 All these were deliberately arranged to reflect the immense reach
of Chola royal power.

The Rajarajeshvaram’s colossal pyramidal vimana (temple spire) rose to a

height of 190 feet in fourteen storeys, dwarfing even the Krishneshvara, its
only competitor in southern India at the time for sheer scale. It was topped
by an 80-tonne granite shikhara (a dome atop the temple spire) – probably
raised by teams of dozens of elephants and, according to oral legend, a
stupendous earthen ramp 4 kilometres long. The Rajarajeshvaram towers
over a vast open courtyard a quarter kilometre long and an eighth of a
kilometre wide, behind imposing gateways flanked by colossal dvarapalas
(door guardians). A human is just barely as tall as its plinth. At the time of
its construction, it was one of the largest free-standing structures anywhere
in the world.
Some inscriptions refer to this extraordinary edifice as Dakshinameru,92
the World Mountain of the South, a counterpart to Shiva’s residence, Mount
Kailasha, the northern World Mountain. ‘It is a measure of the ambition
embodied in this imperial act that Rajaraja could portray himself as having
offered Siva a new home in the south, equal to Siva’s Himalayan abode,’
writes Shaivism scholar Richard H. Davis.93 The Cholas had outdone the
wondrous monolithic residence the Rashtrakutas had once made for the god
in the Deccan, and brought Shiva to watch as they conquered all in his
name.

While Rajaraja’s fortune rose, the Vallabhas of the Deccan had not been
idle. It was all well and good to claim they were Chalukyas and seize the
throne of Manyakheta by force. But something more concrete had to be
done if anyone were to take their claim of descent from those half-forgotten
Pulakeshins and Vikramadityas seriously, when the shadow of the
Rashtrakuta Amoghavarshas and Krishnas still stretched even into the heart
of Malwa – and when actual Chalukyas, descended from Pulakeshin I,
performer of the Ashvamedha sacrifice, still ruled in Vengi. And so, they
turned, as previous Vallabhas had, to masterful literary propaganda. This
will allow us a glimpse of one of the most vibrant periods of the
development of courtly Kannada, and see how these new Chalukyas
legitimized their power in their chaotic world.
The first tentative steps towards the reinvention of Taila’s new Chalukyas
were made by the brilliant poet Ranna, who had risen from the courts of
petty chieftains to those of Rashtrakuta vassals and finally received the title
of Kavi-Chakravarti, Emperor among Poets, from the Vallabha Taila
himself.94
In the 980s, soon after Taila’s forces had subdued former Rashtrakuta
vassals in the Konkan, Ranna was set the task of turning this bloody
campaign, fought on red hillsides and lush coastal jungles, into a work of
glory and legend. To do so, he adapted a story from that grandest of all
Indian epics, the Mahabharata: specifically the deadly duel between the
cousins Bhima, strongest of the heroic Pandava brothers, and Duryodhana,
eldest of the wicked Kaurava brothers. The pair develop a memorable
rivalry in both the Mahabharata and in Ranna’s retelling. After Duryodhana
and his brothers attempt to strip the Pandavas’ wife Draupadi in full view of
the court, Bhima promises to one day wash her hair with Duryodhana’s
blood. He eventually does so after shattering Duryodhana’s thighs with his
mace towards the very end of the devastating familial conflict.
Ranna’s magnum opus, the Gadayuddham (Duel of the Maces), also
called the Sahasa-Bhima-Vijayam (The Victory of the Bold Bhima), mixed
verse and prose, as well as Kannada and Sanskrit, as standardized by
Amoghavarsha Rashtrakuta a century ago.95 It was designed to be read out
in the gamaka tradition still extant in Karnataka today, with musical
accompaniment and pauses to allow audiences to fully experience the
complicated array of emotions96 of the protagonists. The Gadayuddham is
filled with martial imagery, including, for example, a depiction of a war
elephant strike corps in action: ‘His cardinal elephant stomps to the fore
against the famous Gurjaras on their elephant army. The princes’ elephants
follow the leading elephant … so the thread follows the needle, and princes
follow [the then crown prince] Satyashraya’s elephants.’97 This was, no
doubt, meant to flatter Taila’s successor Satyashraya ‘Irivabedanga’ (A
Wonder among Those Who Pierce in Battle), an Old Kannada title.
All these, combined with the depiction of Duryodhana as an unusually
compelling antihero, made the work an immediate hit; indeed, it is
considered one of the classics of the Kannada courtly tradition to this day. It
was a perfectly timed meditation on the human costs of the brutal wars that
had been fought to restore order to the Deccan after the anarchy and horror
of the previous decades. It was a reminder that the Earth would renew itself
just as it had after the devastation of the Mahabharata war, especially under
the stewardship of virile, sophisticated and martial Chalukya kings.
In the Gadayuddham, Ranna also initiated a process of directly
connecting these new Chalukyas to the old. He had dug into extant oral
legends to uncover Pulakeshin I’s glorious horse sacrifice all those centuries
ago and then inserted in his composition a reference to it.98 By the time
Taila’s grandsons99 came to the throne in the early 1000s, this process of
creating links to the old Vatapi Chalukyas was further systematized and
expanded.100
Scholars were dispatched to Aihole, 200 kilometres away from
Manyakheta, to examine a great prashasti composed for Pulakeshin II.101
Soon after, we see its opening verses replicated in the land grants of these
new Chalukyas, using chaste Sanskrit, not courtly Kannada as was then in
fashion.102 All those marvellous formulations, which had not been heard in
the Deccan for two hundred years, reappeared:103 from the image of Varaha,
the Ocean, shaking as he raised Earth on his right tusk, to the Chalukya’s
nourishment at the breasts of the Seven Mothers, to their ‘uninterrupted’
dominance over other kings granted by Skanda, to their acquisition of the
ever-victorious boar banner at the hands of Vishnu himself.104 This was not
some mindless replication but a self-aware one, designed to tell a new story
of new Chalukyas suited to drastically different political circumstances.
In this version, Pulakeshin II never rebelled against his uncle Mangalesha.
Instead, Mangalesha merely acted as the boy’s dutiful regent and handed
over the throne to him as soon as he was mature, for as the prashasti’s
rewriter puts it, ‘what member of the Chalukya dynasty would ever stray
from the path of the dharma?’105 This must have dovetailed beautifully with
new grants to Brahmins and temples across the Chalukya imperial
formation, as elites celebrated the new Chalukya Vallabha’s restoration of
the supposed dharma of his ancestors, a dharma that we have seen always
existed more in the elite political imagination than in actual fact for most of
their subjects.
Definitive family connections to the Vallabhas from Vatapi were also
dreamed up and included in the prashasti, ‘setting right once and for all the
historical relationship’106 between the two families. According to these new
Chalukyas, Taila was actually the heir of an obscure son of Vikramaditya II.
This individual’s descendants, according to the new genealogy, had
continued to rule some unspecified territories while the treacherous
Rashtrakutas reigned over the Deccan – just as Pulakeshin II’s sons had
survived the anarchy of the Pallava invasion.107
All this self-conscious moral and historical positioning was enormously
useful in legitimizing this new dynasty. The new Chalukya history, with the
inferred prestige it conferred on the Vallabha, was likely circulated and
carefully analysed by maha-sandhi-vigrahakas (Great Controllers of Peace
and War, chancellors-cum-foreign ministers) and kings within and outside
their vassal network to assess the martial, territorial and ritual claims it was
attempting to make.108 A connection to the Deccan’s seventh-century
superpower certainly helped consolidate the ‘Chalukya’ family’s claims to
primacy within this turbulent region in the eleventh century.
This connection also served external geopolitical objectives. In particular,
by establishing these new Chalukyas as directly tied to Pulakeshin II, it
established a connection to his younger brother Kubja-Vishnuvardhana, and
thence to Vengi, which Rajaraja Chola had practically taken over only a few
years earlier. This fertile coastal belt, well connected to the trading
networks of the eastern Indian Ocean, had also recently emerged as a hub of
Kannada poetry.109 The Chalukya Vallabha was searching for ways to take
Rajaraja Chola down a notch after the endowment of the splendid
Rajarajeshvaram temple, and the Chola king’s success in incorporating the
Vengi Chalukyas into his own lineage. And so the lords of the Deccan once
again plotted to conquer Vengi, that blood-drenched and fertile land
between the Krishna and Godavari deltas.

The Chalukyas attacked Vengi in 1006, looting and burning the ancient
Buddhist site of Dhanyakataka/Amaravati110 before being chased out by
Chola forces.
The next year, the Tamil chiefs and viceroys and generals gathered to the
beating of drums and the blowing of trumpets, and retaliated with a
devastating raid of the Deccan. They were led by Rajaraja’s equally violent
son, Rajendra Chola (Lord of Kings). This sort of an attack was both a
political and economic activity.111 It was intended to do much more than
slaughter the enemy: a successful raid was the ultimate symbol of a king’s
ability to subordinate his rivals and reward his vassals with the wealth of
their enemies.
According to the Chalukyas, Rajendra, ‘the constant joy of Rajaraja, the
ornament of the Chola family’,112 attacked with ‘a host of nine hundred
thousand’ (more realistically a tenth, if that, of that number) and ‘was
ravaging the whole country, perpetrating murders of women, children, and
Brahmans’.113 Their inscription is also quite emphatic that the invaders
‘caught hold of girls and destroyed their caste’.114 Which is to say,
Rajendra’s army committed mass rape and plunder. Though the scale of this
activity is unclear, this testimony cannot be dismissed as unreliable out of
some misguided desire to portray the Cholas as chivalric heroes. The
capture of women in raids and war in medieval India is extremely well
documented,115 and we have repeatedly seen how violence against enemy
women was associated with martial success and royal virility. On the other
hand, ostentatious respect for the highborn wives and daughters of one’s
subjects was considered a mark of chivalry; it is rare to find a medieval
king claiming to respect women as people.
Chalukya propaganda thus sought to portray the Cholas as violating
dharma through these atrocities, just as the Chola attacks served to
humiliate the Chalukyas, and portrayed them as unable to follow the
dharma of protecting and enriching their subjects and loyal followers.
When the expedition returned to Tamilakam, the delighted Rajaraja Chola
had golden flowers crafted out of his son’s loot, and ‘worshipped the feet of
the god’ in the altar of the Rajarajeshvaram.116 Attended by all the new
warlords of the Tamil country and their wives and children in glittering
jewels and Chinese silks, and accompanied by the clanging of cymbals and
bells and the pungent odour of wet flowers and Arabian incense, this must
have been a sight to behold.
And so began the Chola–Chalukya wars: a series of devastating annual
attacks and counter-attacks that would destroy cities and towns across
southern India and leave thousands, if not tens of thousands, dead and
displaced. The stakes of the conflict rose higher every year as the two
imperial dynasties sought to outdo each other, leading to ever-escalating
campaigns to secure or seize glory and wealth. This vicious pattern had
emerged in full force by the time of Rajaraja’s death in 1014: in 1010, the
Chalukya Vallabha Jayasimha II would describe himself as ‘the Lion to that
Elephant, Rajendra-Chola’117 and award one of his generals the title
Tigularamari, Death to the Tamils.118
But the Cholas were not the only deadly threat the Deccan now had to
contend with. The Malwans, under a new king, Bhoja Paramara, known
also as Bhoja of Dhara, were out for revenge for Taila’s brutal slaughter of
Vakpati. Bhoja was one of the most brilliant kings of the medieval world,
eclipsing his contemporaries with his encyclopaedic knowledge – which
spanned disciplines from aesthetics to architecture – and extraordinary
poetic talent. ‘In the courtyard of his tongue,’ we are told, ‘the goddess
Sarasvati used to dance in ecstasy.’119
 His artistic inclinations notwithstanding, Bhoja’s armies crossed the
Narmada in the mid-1010s, reaching southern Gujarat, and headed south,
along the west coast, to the lucrative emporia of the Konkan.120 The
Vallabha Jayasimha II was forced to move to repel them. Rajendra, now the
Chola emperor and on the lookout for any opportunity to loot and seize
glory, saw a gaping patch in the Deccan’s defences, and did not hesitate to
take advantage of it.
 Around 1016121 CE, Manyakheta, once the glittering crown jewel of the
Deccan’s peoples, would be sacked for the last time.

The invading Chola army seems to have encountered almost no serious

resistance as it punched past the Krishna river and headed for the city. A
huge exodus probably began as news of the invading Tamils spread,
panicking denizens clogging the bejewelled gates under which
Amoghavarsha Rashtrakuta once paraded. The Chola army camped in
orchards of tall sal trees on the Manyakheta outskirts as they prepared to
storm the walls.122 One can only imagine what the remaining denizens of
Manyakheta thought as they heard the drums and chants of the Chola
regiments in the dark of the night. What rumours may have spread, what
desperate prayers for a salvation that would never come, what terror and
weeping.
In the morning the invaders marched towards the great city, which on its
final day, according to the Cholas, was ‘shining like the expanse of the
Earth, surrounded by the Lokaloka mountain’.123
Manyakheta’s glory days were well behind it, and it must have been
dilapidated indeed after the sacking it had faced at Paramara hands in 972.
But its prestige as the ancient capital of the Deccan, the home of the
Vallabhas, was still intact, and it was probably slowly limping to a fresh
lease of life as the new imperial centre of the Chalukyas. Destroying it once
and for all would send a painful and devastating message. And Rajendra
Chola had ordered124 his generals to do exactly that. Manyakheta’s great
teakwood gates were pushed open for the last time. The city was once again
systematically looted, and Chola sources tell us it was set afire by thousands
of torches thrown by Rajendra’s army.125 ‘The women running through the
smoke in the terraces of the bejewelled mansions looked like lightning
flashing through groups of clouds,’ a Chola poet would say.126
As the flames leapt higher, ash and screams clogged the air. We can
scarcely imagine what was destroyed that day: utensils, tiny idols,
jewellery, trinkets, manuscripts carefully inscribed and painted by centuries
of Deccan Shaivas and Jains, the knowledge of generations of poets,
administrators, generals and connoisseurs. Spectacular palaces, carefully
built over generations, crumbled. Hovels of mud and thatch, homes to
migrating manual labourers, were incinerated. And, as Chola sources tell
us, many of Manyakheta’s inhabitants must have suffered agonizing deaths
through burning or suffocation. ‘The gods, abandoning their palaces set
alight by the terrible fire burning aloft from that city, suddenly fled away
out of fear, suspecting it to be the fire of the apocalypse,’ they claim.127
Manyakheta – where generations of merchants and kings and poets and
courtesans and queens, fisherfolk and sculptors and priests had built
communities and families and livelihoods, where Arabs had been bedazzled
and emperors had poured scented water and milk over idols of gods and
rested their feet on their vassals’ crowned heads – was dead. Today, at
modern-day Malkhed in Karnataka, the likely site of that city, nothing
remains of its glorious past. It has been erased, consigned to oblivion, and
there has been little archaeological excavation conducted to precisely locate
and study the site.
We do not know how the news of Manyakheta’s destruction reached the
Chalukya ruler Jayasimha II, Taila’s grandson. Thanks to the Paramara king
Bhoja, he was already on the back foot in the Konkan. We do not know how
he reacted to the disaster when he heard of it. All we know is that the man
himself did not give up the struggle, and continued to fight on the Deccan’s
northern and southern fronts. And unlike after Siyaka’s sack of Manyakheta
in 972, this time vassals stuck with the Chalukya emperor. This Chalukya
court was only three generations old. Its vassals and generals as a group
were much more cohesive and invested in the survival of their imperial
formation than the huge, unwieldy Rashtrakuta court had been, which was
nine rulers and many worlds away from its initial founding members. The
Chalukyas even had in their service men (and women128) who remembered
Taila’s precarious rise, and who had fought and defeated Tamils, perhaps
even under the command of Krishna III before the Rashtrakuta collapse.
Of course, any thoughts of revenge had to wait. The destruction of
Manyakheta, when the Chalukyas were already so vulnerable, must have
been a devastating blow to their wealth, prestige and morale. It annihilated
a major hub of agrarian, trade and religious networks, which had been
crucial to the prestige and power of the Vallabha. But not all was lost. In
reality, the Chalukya capital was wherever the Vallabha’s court was, and the
court at this time was in a moving military encampment in the Konkan,
fighting the forces of the Paramaras and their allies. It seems to have moved
quickly to the smoking ruins of the city, leaving the Malwans to
prematurely celebrate their conquest of the Konkan in 1020.129 Meanwhile,
an old city on the route from Ellora to Manyakheta – close enough to
Manyakheta that the refugees from the scorched city could settle there –
was chosen to be a new capital. This city would give these later Chalukyas
the name by which they are still remembered, to set them apart from their
putative ancestors from Vatapi and their fractious relations in Vengi:
Kalyana.

Rajendra Chola cared little for what these ‘Kalyana Chalukyas’ did or did
not do. The loot of Manyakheta was paraded and used in public festivals
and in the Chola capital, where Rajendra appeared on a horse surrounded by
similarly mounted princes, ‘a hero in the midst of the cavalry’,130 as
inscriptions put it. His cheering subjects were informed that through this
act, his father Rajaraja’s vow to destroy that city had been fulfilled.
Rajendra now sent his armies south, where Chola control over the former
Pandya territory and in Lanka remained contested and unpopular, and
looted the Sinhalas again in a brutal two-year campaign.131 The Cholas
could not afford to lose their dominant position in the Palk Strait, the gap
between India and Lanka, the gateway between the eastern and western
Indian Ocean. Ships conveying goods between Egypt, Arabia, Persia,
Indonesia, Malaysia and China needed to pass through this crossroads,
offering enormous business potential to Tamil merchant guilds close to the
Cholas, as well as an opportunity to the Cholas to levy transit fees. The Palk
Strait offered an unavoidable geopolitical choke point where maritime
traffic was concentrated: there was no escaping a polity which controlled
both of its shores. In comparison, the geography of the Konkan coast
allowed daring captains and merchants to avoid hostile ports, which was
why it was so important for the Paramaras and Chalukyas to control it in its
entirety. But to medieval Indian rulers, all these risky and expensive
military campaigns were well worth it, because there were real fortunes
being made in the Indian Ocean trade. Pottery from China has been found
all across the coast of Tamil Nadu in recent archaeological excavations,
hinting at an explosion in the scale and value of maritime interactions
across the eastern Indian Ocean in the eleventh century.132
The best evidence we have of the scale of the Indian Ocean trade in the
ninth–twelfth centuries comes from a shipwreck from 830 CE, discovered
near Belitung, an island off the east coast of Sumatra. The ship was an Arab
dhow on a return voyage from China, carrying nearly 60,000 ceramic items,
attesting to industrial-level mass production for global markets, specifically
for buyers in West Asia.133 Its cargo also included extremely fine gold and
silver items apparently intended as gifts from the imperial Chinese court to
the Sailendra dynasty of Java.134 Other objects recovered from the wreck
included a paperweight, ‘a re-soldered bracelet sized for a woman’s wrist’,
‘a ceramic whistle shaped like a fat bird, and a small, charming
ceramic dog’135 – all remnants of lives just as human as ours, a diplomat’s
writing equipment perhaps, a gift for a beloved, souvenirs for a child.
The ceramics found in the wreck had been produced in their tens of
thousands in Changsha in south-central China, and then packed and shipped
to the embarkation port of Guangzhou. Here there were large communities
of international traders: Persians, Arabs, and various groups from South and
Southeast Asia.136 The consignment consisted of bulk orders placed by
merchants based in West Asia, as revealed by the decorative motifs used on
the ceramics, which were similar to examples from Iraq and the Persian
Gulf. The ceramics were mass-produced according to consistent templates.
As Professor Geraldine Heng elegantly puts it, they and the other objects in
the ship ‘are summaries of the socioeconomic relations that propelled
international commerce; a shorthand for deciphering political and
diplomatic initiatives that were taking place in the world; and a
dramatization of the artistic exchanges that were crisscrossing the world’s
creative pathways as early as the ninth century’.137 This is all the more
striking because it is so reminiscent of our modern, globalized world.
 If not for the roaring trade of the medieval Indian Ocean world, both
India’s coastline and the great cities in its interior would have looked
profoundly different. Rajaraja Chola, his son Rajendra and their
contemporaries worked closely with the growing merchant corporations of
southern India to reshape the region’s economic and social landscape.
By this point, the Five Hundred Lords of Aihole had evolved into a vast
organization called the ‘Five Hundred of the Thousand Directions, known
in every direction in all the Eighteen Lands’138 and had brought together
almost ‘all possible specialist merchant groups, itinerant and sedentary,
local and foreign’.139 Including its commercial partners, suppliers, artisans,
guard groups and so on, tens of thousands of individuals were probably
associated with it, making it one of the most powerful groups in medieval
India, far more so than most petty kings. The Five Hundred’s influence
expanded in the wake of the politics of military aristocrats – entering the
Tamil country alongside Rashtrakuta attacks, and now expanding through
Gangavadi, Lanka and Malabar alongside the Cholas.140
To defend their diverse and spread-out interests, merchant groups like the
‘Five Hundred’ hobnobbed not only with emperors, but with other medieval
power centres as well: local kings, administrators of minor cities, and
temples. The last two often went hand in hand – in the eleventh and twelfth
centuries, temple building helped catalyse a great wave of urbanization
along the east coast, often described as India’s ‘third urbanization’, after the
urbanization of the Indus and Ganga river valleys. Merchants, artisans,
landowners and city administrators worked in a symbiotic and mutually
beneficial relationship that led to an explosion in economic specialization
and an array of goods produced in newly prominent southern Indian
cities.141 These ranged from low-cost, easily available staples such as grains,
straw and beans to more expensive pepper, areca nuts and cinnamon; from
imported horses, camels and pedigree cattle to exotic and sophisticated
goods such as conch shells, silk, musk, garlands, incense, sandal paste,
camphor and iron ingots.142
The amount of economic activity that merchants could generate by linking
up local economies to global supply and demand, and the clout they
wielded made working with them a smart move. Merchants could secure all
sorts of concessions from rulers. One group in the Malabar coast, for
example, obtained from a local king the right not to pay customs duties ‘if
they feel wronged (by the officials)’, to try their own members for crimes
they had committed, and permission to ‘carry on elephants the purification
water for their rituals’.143 Another, led by a Jewish man by the name of
Issuppu Irappan (Joseph Rabban), was even allowed the free use of boats
and vehicles and exempted from paying duties entirely.144 The race to make
money from global, regional and local trade would shape the dynamics of
societies and states across southern India for centuries thereafter.
As Rajendra Chola exulted in his conquests in the Deccan and Sri Lanka,
global trade opportunities and the military interests of south India’s new
superpower now aligned to give rise to one of the most extraordinary events
of the eleventh-century world.

From the 990s onwards, diplomatic missions from across the world to the
Song court had drastically expanded in frequency and opulence as people
from across the Indian Ocean smelled fortunes to be made. Those
embassies that made a favourable impression in the Song court would be
rewarded with higher diplomatic status and thus preferential access to
Chinese markets145 – something that both kings and merchants, especially
the Five Hundred and their partners, were eager for.
The first Chola embassy to the Song court, sent by Rajendra’s father
Rajaraja, was received on 17 October 1015, arriving roughly a year after his
death.146 The gifts from the Chola to the Song court included ‘one robe and
one cap adorned with real pearls, 21,000 ounces of real pearls, 60 elephant
tusks, 60 catties [half-kilos] of frankincense, jade, glass, and cotton
fabrics’.147 The ambassadors – wealthy members perhaps of the Five
Hundred, or powerful port guilds seeking trade concessions for themselves
– ‘presented on their own 6,600 ounces of pearls and 3,300 catties of
aromatic drugs’.148
The Song were highly impressed by the embassy and the dignified
conduct of these supposed barbarians (medieval Chinese considered all
people barbarians by default, unless convinced otherwise by overwhelming
wealth and sophisticated mannerisms). The emperor ordered his officials to
treat them with ‘great regard’,149 and they were invited to participate in the
emperor’s birthday festivities in early 1016. They departed soon after,
bearing an ‘imperial edict and rich gifts’150 and perhaps some trade
concessions. That was not all they carried with them, for they had learned
something very worrisome about the activities of Srivijaya, the maritime
confederation that ruled the Straits of Malacca. The historian Tansen Sen
argues that until the Chola embassy had arrived in the Song court,
merchants from Srivijaya had been informing the Chinese that the Cholas –
who were actually south India’s undisputed superpower – were mere
vassals of Srivijaya.151 This had secured preferential trade licences for them,
as representatives of an ostensibly greater kingdom.
Rajendra Chola was neither going to accept such a humiliation, nor pass
up the opportunity to make money. A seaborne raid was immediately sent to
punish Kadaha/Kadaram (modern-day Kedah, a state in Malaysia), a major
city of the Srivijayan confederation that controlled the entry point to the
Malacca Strait. Somehow, the undoubtedly rather seasick Chola expedition
appears to have taken those doughty Srivijayan traders and sea lords by
surprise, or so Rajendra’s poet would claim: ‘It is no wonder that the fire of
his [Rajendra’s] prowess consumed the great race of Taila [the Chalukya –
an insulting comparison of his name to oil/tailu] … But it is a wonder that
having crossed the ocean [that which is sealed], it burnt Kadaha
[Kedah/kadai, a cooking vessel].’152
The Srivijayans seem to have dismissed this raid as a one-off. In 1018, the
king of Kedah sent a large gift of gold for a temple and its undoubtedly
well-connected Brahmins in Nagapattinam, the premier Chola port on the
Indian Ocean, perhaps in thanks for their assistance in smoothing things
over with Rajendra’s court. Perhaps he thought that was atonement
enough.153 In any case, the many other wealthy Srivijayan ports that studded
the island of Sumatra and the coast of the Malay Peninsula had been totally
unaffected by Rajendra’s raid. They were, for the time being at least, left to
their own devices, as the Chola emperor turned his attention to other
projects.

By the mid- to late 1020s, the Chalukya Vallabha Jayasimha II finally

managed to uproot most of the Paramara presence in the Konkan.154 He then
set out to weaken Chola control over the Deccan and ensure there was no
repeat of the disaster at Manyakheta, pushing his influence incrementally
further south towards the Krishna river.155 His son and crown prince,
Someshvara (Lord of the Moon), a shrewd politician, also involved himself
in Vengi to try and create a new front to engage the Cholas.
Familial strife was back on the boil in Vengi because of the royal family’s
exodus to the Chola court a generation earlier. Two half-brothers – one
based in Vengi, the other probably at the Chola court – now laid claim to
the Vengi throne. One of them, Vijayaditya, was the son of a lady belonging
to a powerful Telugu clan. The other, Rajaraja-Narendra, was the son of
Rajaraja Chola’s daughter Kundavai, and thus Rajendra Chola’s brother-in-
law. Searching for allies, Vijayaditya went to Kalyana, where he was
befriended by Someshvara Chalukya, who helped him gather troops for an
expedition to Vengi. This was (temporarily) successful, and the prince
declared himself Vijayaditya VII of Vengi.
 Someshvara, however, was playing a double game: it appears that
Vijayaditya VII’s attack on Vengi was meant only to divert Rajendra
Chola’s attention there as Chalukya armies attempted to secure the Raichur
Doab, the fertile territory between the Krishna and Tungabhadra rivers.
Vijayaditya VII was soon defeated by Chola forces in Vengi and forced to
flee156 to the court of Kalyana, where, as we shall see, he would have a most
remarkable career. His half-brother, Rajaraja-Narendra, was crowned thanks
to the efforts of Chola generals.157
But from the Chola perspective, Rajendra Chola had not contested the
Raichur Doab because Vengi was a move towards achieving a much
grander goal. Rajendra needed to send a message to the proud lords of the
subcontinent. The merchant corporations of India’s east coast – particularly
in Bengal, where the Palas still sat atop routes that controlled access to the
‘great North Indian market’158 – also needed to learn that a far more
powerful imperial system was interested in taking over trade in the Indian
Ocean.
For years, a great army had been trained and amassed in the Tamil
country. Now, with India’s east coast up to Vengi under Chola domination,
Rajendra had the secure base he needed to repeat the feats of the ancient
Rashtrakuta Vallabhas of the Deccan. Once again, the drums and trumpets
of a south Indian army would resound along the banks of the Ganga, and
shatter the peace of northern India.
In 1022–23, an expedition was dispatched under a veteran Chola
commander, bearing (among others) the titles of Araiyan Rajarajan
(Rajaraja’s Noble) and Jayasinga-Kula-Kala (God of Death to the Chalukya
Clan). Moving along the subcontinent’s east coast over a two-year
campaign, the expedition seized forts and destroyed and captured towns in
the regions corresponding to modern-day Odisha, Chhattisgarh and Bengal.
Chola sources claim the cities of Tandabutti, Takkanaladam and
Uttiraladam were sacked, and Madura-mandalam, Namanaikkonam, and
Panchapalli destroyed (their precise locations are debated to this day). No
less than three kings who attempted to fight off the marauders in Bengal –
vassals of the Palas159 – were defeated and their elephants, women and
treasure seized.160 The size of the list of raided cities and regions that
appears in Rajendra Chola’s propaganda, the geopolitical awareness it
presumes in its original readers, and the strategic, tactical and operational
abilities needed to execute such long-distance operations are all astounding.
As Professor Nilakanta Sastri puts it, even if Chola accounts of it ‘gloss
over reverses and exaggerate successes’, the campaign was an extraordinary
‘exhibition of the power of the Chola empire and a demonstration of its
strength’161 to other Indian kings and to Rajendra’s own subjects. At its end,
waters of the Ganga were transported south, perhaps on elephant-back, in
huge containers made from the mud of Bengal. The Rashtrakutas were
totally outdone. Loot flowed with the Ganga’s waters to Rajendra’s new
capital city, named after himself: Gangai-Konda-Chola-Puram, the City of
the Chola who Conquered the Ganga.

What the Chalukya Vallabha Jayasimha II and his son Someshvara must
have thought of all this can scarcely be imagined. Before Rajendra Chola’s
north India expedition, there must have been celebration in the Deccan
early in the 1020s. Though there was much more to do, the Paramaras, their
deadly enemies from Madhya Pradesh, had been removed from the Konkan,
and the ports of the west coast were once again beginning to acknowledge
the Chalukya Vallabha, Jayasimha II. Vengi had been gained and lost, but
that was a small price to pay for conquering the fertile lands between the
Krishna and Tungabhadra rivers. Crown Prince Someshvara was celebrated
in his father’s great moving military camp, showered with flowers by the
women of the court, praises composed to his political astuteness in Kannada
and Sanskrit, jewels and titles heaped on him by delighted generals and
ministers. It may have seemed that, at long last, Fortune had granted a
respite to the tenacious Chalukyas. Perhaps campaigns were being planned
to reconquer Gangavadi, to contest Vengi, to fully assimilate the Konkan, or
raid Malwa.
And then, in 1022–23, the years of the Cholas’ north India expeditions,
the Chalukya dreams must have begun to unravel as the true might of the
Cholas began to dawn on them, and dismaying reports from its network of
informants, clients and vassals poured in. In an attempt to sabotage and
humiliate their Chola rivals as they carried out their audacious campaign,
the Chalukyas even appear to have indirectly aided the kings of Odisha and
Chhattisgarh,162 but in vain. The Chola armies – which also included
significant numbers of Deccani fighters and commanders,163 reminding us
that medieval India was far more heterogeneous than modern nationalism
would have us believe – overpowered all. There was no coalition of Indian
kings, merchants and religious institutions that could challenge the Chola
imperial formation. Submission was the only option in the face of such
might.
Chalukya court factions squabbled over their course of action, and it was
not long before the Vallabha’s leadership was being questioned. The men of
the court, who spent ‘large portions of their youth’ in arduous military
training, seem to have clamoured for an opportunity to perform in the great
theatres of war where they could display their prowess before their peers
and superiors.164 There were probably many young nobles, born in the
generation after the collapse of the Rashtrakutas, steeped in the propaganda
of the Chalukyas, who now demanded that the Vallabha take the field for
glory and fortune.
Jayasimha II, the Chalukya Vallabha who called himself ‘The Lion to that
Elephant, Rajendra Chola’, would come very close to losing his throne in
these fraught years. For obvious reasons his records are tight-lipped about
the matter, but it appears that he was just barely rescued from a coup
attempt by the prompt actions of the senior general Kalidasa;165 we know
little else of the conspiracy. The general claims only to have ‘maintained
with vigour the burden of government which had become dissolute when
those generals and vassals proved false to King Simha’.166 Yet the worst
days for the Deccan still lay well in the future.
Someshvara, the Chalukya crown prince, learned from this shock, just as
he learned from the Cholas’ successes and atrocities. This young man had
already developed a taste for politics: we have seen how he encouraged his
Chalukya ‘relation’ from Vengi, Vijayaditya VII, in his attempt to seize the
throne. Over the course of the 1020s, the two developed a close, almost
familial relationship, with Vijayaditya serving as Someshvara’s right-hand
man through many of the hardships that followed. We can imagine the pair
riding next to each other on garlanded elephants, exulting in the momentum
of the great beasts, or perhaps chasing and shooting down fleeing animals
from their imported steeds in royal hunts, surrounded by the baying of
pedigree hounds,167 meant to exhibit Someshvara’s martial prowess and
suitability to succeed his father as emperor.
The Cholas were an enemy unlike any that the Deccan had faced. The
Paramaras could be defeated; the vassal lords of the Deccan guarded
against, crushed, cajoled, flattered. But the Cholas didn’t play by the rules –
not even by the violent and anarchic standards of medieval India. To an
educated prince like Someshvara Chalukya, violence (at least in theory if
not in practice) was ‘ritualised and honourable’, to be ‘exercised with
restraint, to be combined and complemented with the virtues of
compassion, kindness and gentility’.168 To accept the submissions of
defeated highborn enemies and restore them to power – as the Chalukya
Vallabha Vikramaditya II had done with the Pallavas, as the Rashtrakuta
Vallabhas Govinda III and Dhruva had done with the Vengi Chalukyas,
Palas and Pratiharas – was to publicly display one’s adherence to the
correct, orderly rules of this world.
But the Cholas did nothing of the sort. Driven by hard-nosed geopolitical
realism, they paid no heed to the rules of the game – and in return they had
received victory after victory. Rajaraja and Rajendra Chola conquered
outright the territories they deemed important, garrisoned them with Chola
armies and Chola viceroys of royal blood, and renamed them after
themselves. Lanka, that ancient Buddhist kingdom which had sent its idols
as submission to the Deccan emperor Govinda III, was now to be called
Mummudi-Chola-Mandalam, The Circle (territorial division) of the Chola
Who Wears Three Crowns. The former Pallava heartland in northern
Tamilakam, which so many Deccan Vallabhas had sacked but left otherwise
untouched, was now to be called Jayan-Gonda-Chola-Mandalam, the Circle
of the Victorious Chola.169 The Kaveri river valley and the coastal area
around it were renamed Cholamandalam, the Chola Circle – a name that
survives to this day as ‘Coromandel’.
The Chola machine was kept going with incessant warfare, looting,
temple building and urbanization at an unprecedented scale. The fortunes of
the Tamil country burned bright, its cities teemed with people, its palaces
were full of captured women from the many kingdoms of the Deccan,
Kerala, Lanka, Andhra, Odisha, Bengal and Chhattisgarh, forced to be the
servants of Chola queens and princesses, the concubines of the royal guards
of the princes.170 Someshvara could do nothing but survive against such a
terrible enemy.
But Rajendra Chola’s ambitions were still far from fulfilled. Now, in the
eleventh century, Rajendra Chola had the resources and the inclination
needed to perform the most spectacular of all medieval Indian military
expeditions.
Srivijaya would learn its lesson.

In a daring and extremely risky expedition – not replicated by any South

Asian power until the British Raj – Rajendra Chola, having conquered the
north, south and west, sent an army east, across the thundering sea, to raid
as many Srivijayan cities as it could. It was intended as punishment for their
audacious attempt to claim primacy over the Cholas in the rich Song court.
Though how the Chola armada was organized and how it executed its
objectives are unknown, it appears that by January/February 1026, it
returned to the Coromandel coast laden with booty from those cities across
the seas.171 The huge, colourful fleet returning to its home shores must have
been a sight to behold.
This Chola campaign was spectacular like no other Indian undertaking in
the medieval period. The Krishneshvara at Ellora was outdone, the
campaigns of Dhruva Rashtrakuta eclipsed. Even Rajaraja Chola’s
conquests and enormous temple at Thanjavur could not match his son
Rajendra’s crossing of the ocean itself to plunder Srivijaya.
In Gangai-Konda-Chola-Puram, Rajendra made a display of the wealth he
had seized, making a particular example of Kedah, the city on which he had
ordered an attack in 1017–18. In 1024–25, Rajendra’s armies had
supposedly ‘caught’ the ‘King of Kadaram (the Chola name for Kedah),
along with (his) rutting elephants, [which were as impetuous as] the sea in
fighting’172 and ‘(took) the larger heap of treasures, which he had rightfully
accumulated’.173 The great Vidyadhara gate of Kedah had been torn down
and was now put on display in the Chola capital. So, too, were a ‘jewel-
gate, adorned with great splendour’, and another ‘gate of large jewels’.174
These gates, transported back to Thanjavur or Gangai-Konda-Chola-Puram,
were likely put on display for Rajendra’s subjects as a testament to this
extraordinary feat.
Also on display, perhaps, were treasures from Srivijaya (modern-day
Palembang175) itself, the eponymous capital of the Srivijaya confederacy.
This was a commercial hub of such density that an Arab trader reported that
in one street alone he had counted eight hundred money changers.176
Palembang could not have been sacked with anywhere near as much
aplomb as its ally, Kedah; it was guarded by two straits and estuarine terrain
that the Cholas would have found extremely difficult to navigate in the
limited time they had. This, of course, did not stop Rajendra from claiming
to have captured it anyway, but details of the defeat of the city’s king and its
supposed sacking are, unsurprisingly, lacking.
In his propaganda, Rajendra also claimed, on the east coast of Sumatra,
the capture of ‘Pannai’ (modern-day Panei), ‘Malaiyur’ (modern-day
Jambi) and ‘Ilamuridesam’177 (modern-day Lamuri). The list of the cities he
claimed to have captured is even more extensive along the Malay Peninsula
– either because the Chola army had based itself around Kedah, or because
there were simply more cities there. Here, his forces supposedly captured
‘Yirudingam’, ‘Pappalam’, ‘Ilibangam’, ‘Valaippanduru’, ‘Takkolam’ and
‘Tamalingam’,178 all accompanied with descriptions, including brief praise
of their defences and martial valour, which apparently did them no good
against his armies. To round out this maritime ‘Conquest of the Directions’,
Rajendra also informs us that his forces had supposedly attacked the
Nicobar Islands – ‘the great Nakkavaram, whose flower-gardens
(resembled) the girdle (of the nymph) of the southern sea’.179
These were all places the people who saw Rajendra’s parades and displays
were probably familiar with to some degree. Word of the success of these
raids spread like wildfire across trade and pilgrimage routes, and Rajendra’s
own status as a celebrity king rose to a level none of his contemporaries
could hope to challenge. In medieval India, neighbouring kingdoms were
seen as foreign land. The further away the enemy kingdom, the more
prestige gained by the king who successfully plundered it: thus the
Rashtrakuta expeditions to north India and Tamilakam, and the Chola
expedition to Bengal. But the sack of Srivijaya, involving an audacious
crossing of the seas, put it in a different category altogether. Through this,
Rajendra Chola almost completely eclipsed the Chalukya emperor
Jayasimha II and his heir apparent, Someshvara, whose military
encampment was even then roving around the Deccan,180 still attempting to
piece together their shattered political and economic networks.
Today, we tend to think of what Rajendra Chola did as something all
South Asians should be proud of. But in the medieval Deccan, there must
have been a sense of doom at this man’s meteoric rise, his display of
dominance, and the fact that this destroyer of cities and despoiler of lands
seemed utterly unassailable.
But Rajendra Chola’s success was more superficial than it seemed, despite
the awe-inspiring spectacles of paraded loot from distant Bengal and
Indonesia. Many Srivijayan cities had been bankrupted by the devastation
caused by the Chola army, but the confederacy was very much still in
business.181 Rajendra’s raid captured booty, sent a message and played to his
domestic audience, but it was by no means a ‘conquest’ or ‘colonization’.
Srivijaya didn’t even take that long to return to the China trade: by 1028,
merchants from there were once again thronging the docks of
Guangzhou.182 Rajendra himself soon lost interest in Srivijaya now that it
had been put in its place, though he seems to have arranged some
intermarriages with prominent local dynasties. His eyes were on the big
picture, the profits of trade which could fuel his wars in the Indian
subcontinent. The Chola emperor sent another embassy to the Song court in
1033, with a letter written in gold leaf, and his ambassador scattered pearls
from a silver bowl before the emperor, receiving in return the honorary
titles of Grand Master of the Palace with Golden Seal and Purple Ribbon,
and Civilizing General.183

But the cultural and economic opportunities that the Chola attack on
Srivijaya created would lead to deep and persistent changes in world
history. Through the 1024–25 campaign to Srivijiaya, the Chola state had,
without intending to do so, made an immense leap towards the ‘emergence
of a world market’184 by clearing the way for deeper, direct connections
between southern India and eastern Asia. A ‘new wave of cultural
influence’185 was now borne deep into Sumatra by the expanding Tamil
merchant guilds, who were granted trading outposts and ports.186 Art styles
fused, interacted, bloomed.187 Sculptures, spices, handicrafts and Indian
textiles – pleated, painted, patterned with geometry, petals, stars, suns and
flowers – were exported from south India to East Asia.188 Foreign kings and
nobles luxuriated in them. From the eleventh century onwards, we see the
footprint of Tamil merchants in China rapidly expanding;189 in the port of
Quanzhou, for example, there is evidence of Shaiva temples and Tamil
inscriptions left by these traders. Remains of Tamil architectural elements
and a sculpture of an Indian goddess depicted by what seem to be Sino-
Indian sthapatis are visible today in the Quanzhou Maritime Museum.
These remind us, once again, that globalization and syncreticism are by no
means unique to the modern world. Though competition with the
Srivijayans and Chinese would continue in the ensuing centuries, it appears
that the biggest winners of the Chola expedition were, at the end of the day,
Tamil merchants.

Rajendra Chola might not have agreed with such an assessment. In 1034,
when his embassy returned laden with gifts from the Song court of China, it
had been barely half a century since his father’s coronation in 985. And
now the Chola dynasty was without question the richest, most powerful and
most famous in the entire subcontinent, without any indication the situation
would change any time soon. In his great palace at Gangai-Konda-Chola-
Puram, Rajendra lounged on silks imported from China, inhaling exotic
incenses from Ethiopia and Arabia burning in bronzes cast by the skilled
artisans of Tamilakam, surrounded by gorgeous attendants from Bengal,
Odisha, Andhra, Karnataka and Sri Lanka waving fly whisks made of the
tails of Himalayan yaks. All around him were the towering spires of the
gleaming temples he had built and the cities he had enriched. His new
‘Kailasa-like imperial Shiva temple’, the Gangaikondacholeshvaram,
rivalled his father’s Dakshinameru in size, and its great temple reservoir
was purified with the waters of the distant Ganga.190
Gangai-Konda-Chola-Puram was full of trophies, the symbols he had
seized from all over the world and brought here to signify his power over
all. ‘From the [Kalyana] Chalukyas a sun-pedestal, several images of
Durga, and a Ganesa-image’ had been seized and placed in shrines attached
to his own massive temples. The city was also decorated with other images:
‘from the Eastern [Vengi] Chalukyas, a resting Nandi, Shiva’s bull mount;
from the Kalingas of Odisha, three large stone images of Bhairava and
Bhairavi and an awesome eight-armed Kali image; from the Palas of
Bengal, a bronze image of Shiva dancing on Nandi’s back’,191 and many
others. At the four towering gates of the city, images of Durga – one stolen
from the Chalukya emperor – guarded his subjects.192
Rajendra had three adult sons who would now have to rule the empire
their father had raised to unprecedented heights. It was time for a transition.
How many gates and walls would they add, he may have asked himself.
How high would their temples ascend?
Certainly he would never have imagined that these three – who would one
day bear the titles ‘The Victorious Rajendra’, ‘The Divine Rajendra’ and
‘The Heroic Rajendra’ – would be the last of his direct male descendants to
sit on his throne.
EI IV, 289.
This was at Rameshvaram, not Kanyakumari.
EI IV, 290.
Marasimha was the son of the Ganga king Butuga II, Krishna III’s brother-in-law. He was Butuga’s
son by another wife, but retained a close bond to the Rashtrakuta emperor.
Pollock 2006, 335.
Ibid., 334.
Paraphrasing EHD I, 296.
Altekar 1934, 122.
EI XXII, 43–44.
Chakrabarti 2011, Location 2050.
Altekar 1934, 121.
EI XXII, 43–44.
EHD I, 297.
Archaeological and literary evidence from medieval India strongly suggests that most royal cities,
such as Dvarasamudra in the southern Deccan, were organized along a central axis.
This is based on descriptions from Somadeva’s Yashastilaka, completed in 959 in the court of a
Rashtrakuta vassal, filled with details apparently inspired by the composer’s time in Manyakheta.
See Mishra 1992, 83.
Ibid., 89.
The Inscriptions of Nagai, Hyderabad Archaeological Series, No. 8 (The Nizam’s Government,
1928), 18. This refers to the Chalukya ruler Jayasimha II, but is an example of medieval attitudes
to captured women more generally – similar examples can be found across India at this time.
Altekar 1934, 155.
Ibid.
Early medieval prashastis, Rashtrakuta prashastis specifically, frequently mention this; for a more
detailed discussion. See Ali 2006.
Inscriptions of Siyaka’s successor Vakpati, made on the blank sides of copper plates stored in
Manyakheta’s land record offices, have been found. EI XXIII, 101.
This is based on descriptions from Somadeva’s Yashastilaka, as well as the work of Daud Ali. See
Ali 2006.
Somadeva mentions these in his Yashastilaka. See Mishra 1992, 89.
Marasimha II defeated some unnamed enemy at Manyakheta. See EI V, 177–80. This was almost
certainly Siyaka, as there is no other enemy he would have had to fight there.
Altekar 1934, 131–32.
EHD I, 319.
Ibid., 320. The authors of EHD I suggest that this wife was from the imperial line, but her father,
Bhammaha Ratta, is not found in other Rashtrakuta inscriptions, suggesting they were from a
minor branch.
We know that Marasimha II recaptured Manyakheta from Siyaka, and given that Taila’s landholdings
were close to the city, it seems reasonable to assume that the man played some sort of role in this
incident. Soon after, Marasimha returned to Gangavadi. In later years, Taila is known to have been
active in the Konkan and the northern and southern Deccan but never claims conquest of
Manyakheta, suggesting it was already under his control.
EHD I, 205.
Pollock 2006, 154.
Epigraphia Carnatica, Vol. II, lxxxiv. Hereafter EC.
Arvind K. Singh, ‘Interpreting the History of the Paramāras’, Journal of the Royal Asiatic Society 22,
no. 1 (2012). This may have been a deliberate move either to claim the mantle of Amoghavarsha
Rashtrakuta or to establish himself as rewarding loyal followers, or both.
Merutungācārya, The Prabandhacintāman.i, or Wishing-Stone of Narratives, trans. C.H. Tawney (The
Asiatic Society, 1901). These raids are also depicted as originating in the Deccan, with Vakpati
being a righteous monarch who only defended his territory. Given the titles he claimed and his
father’s activities in the Deccan, this depiction is clearly disingenuous or based on a
misremembering of historical events.
This was Karkka II, EHD I, 299.
EHD I, 321.
Ibid.
V.R. Deoras, ‘Fresh Light on the Southern Campaigns of the Rāshṭrakūṭa Emperor Kṛishṇa III’,
Proceedings of the Indian History Congress 20 (1957).
This ‘futile war’ is referenced in EHD I, 320 and in Marasimha II’s inscription. See EI V, 179–80.
EI V, 180.
Ibid. Verse abridged for readability.
EHD I, 321. ‘Gangavadi’ by this time also included much of Banavasi as a result of Krishna’s
gratitude towards his powerful brother-in-law.
EHD I, 321.
Taila died in 997/98, and Vakpati was still alive and spending his loot on patronizing poets in 993/94
going by dedications to him. This leaves a date of 994–96 for his death. See Merutungācārya 1901,
ix.
Merutungācārya 1901, 30–36.
Who, in the story, is named after an actual historical figure attested to in inscriptions: Vakpati’s
minister Rudraditya, whose story the poet respun with all the appropriate compositional elements,
turning him into an archetypal ideal minister.
Patron–client relationships in medieval India could range from intense bonds, with death being sworn
by the subordinate party if they failed their duties to manipulative, cynical, extortionate equations
not dissimilar to those in modern boardrooms.
EHD I, 322.
EI XVIII, 252.
EHD I, 322, citing EI II, 215.
Ibid.
This, according to Merutunga, happened after an ill-fated seduction of Taila’s sister Mrinalavati who
was apparently deputed to serve food to Vakpati, but this can safely be dismissed given what we
know of courtly protocol and other aspects of the chronology.
At least this is the fate of Vakpati in Merutunga’s tale. Merutungācārya 1901, 35.
Ibid., 36.
Hugh Kennedy, The Prophet and the Age of the Caliphates, Second Edition (Pearson Longman,
2004), 313.
Ibid., 316.
Ibid., 317.
Ibid., 319.
For a detailed discussion, see Raoul McLaughlin, Rome and the Distant East: Trade Routes to the
Ancient Lands of Arabia, India and China (Continuum, 2010).
S.A.M. Adshead and Reshmi Dutta-Flanders, China in World History (Palgrave Macmillan, 1988),
109.
Ibid.
See Tonio Andrade, The Gunpowder Age: China, Military Innovation, and the Rise of the West in
World History (Princeton University Press, 2017).
Tansen Sen, ‘The Military Campaigns of Rajendra Chola and the Chola-Srivijaya-China Triangle’, in
Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia,
eds. Hermann Kulke, K. Kesavapany and Vijay Sakhuja (Manohar, 2019) Reprint.
Ibid.
Ibid.
Ibid., 62.
Ibid., 63.
Sen 2019, 63.
Hermann Kulke, ‘The Naval Expeditions of the Cholas in the Context of Asian History’, in Kulke,
Kesavapany and Sakhuja, eds. 2019, 5.
Whitney Cox, Politics, Kingship, and Poetry in Medieval South India: Moonset on Sunrise Mountain
(Cambridge University Press, 2016), 42.
Ibid., 42–44.
SII III, 29. Rajaraja was supposed to have been a ‘tender youth’ when he ordered the burning of the
Chera fleet.
Robert Sewell, Historical Inscriptions of Southern India (Collected till 1923), ed. Krishnaswami S.
Aiyangar (Madras University, 1932), 55. There is some debate over whether Kandalur was a port
or a military school; the evidence seems to indicate the former.
Mamata Chaudhuri, ‘Ship-Building in the Yuktikalpataru and Samarangana Sutradhara’, Indian
Journal of the History of Science 11 (1975): 143.
Wilhelm Geiger and Mabel C. Rickmers trans., Cūlavam.sa: Being the More Recent Part of the
Mahāvam.sa (Asian Educational Services, 1992), 186.
Ibid., 187.
Ibid., 188.
K.A. Nilakanta Sastri, The Cōl.as (University of Madras, 1955), 173.
George W. Spencer, ‘The Politics of Plunder: The Cholas in Eleventh-Century Ceylon’, The Journal
of Asian Studies 35, no. 3 (1976).
Geiger and Rickmers 1992, 188.
Sastri 1955, 174.
EHD I, 323.
K.G. Krishnan, Karandai Tamil Sangam Plates of Rajendrachola I (Archaeological Survey of India,
1984), 198–99.
Usha R. Vijailakshmi, ‘Tamilian Migration into Karnataka (The Period of Chola Conquest of
Southern Karnataka and the Consolidation of Power from 850–1279 ad)’, Proceedings of the
Indian History Congress 66 (2006).
J. Durga Prasad, ‘The Chalukyas of Vengi’, in Comprehensive History and Culture of Andhra
Pradesh, Volume III: Early Medieval Andhra Pradesh ad 624–1000, ed. B. Rajendra Prasad
(Tulika, 2009).
Cox 2016, 36.
Geeta Vasudevan, The Royal Temple of Rajaraja: An Instrument of Chola Imperial Power (Abhinav,
2003), 44.
Adam Hardy, Theory and Practice of Temple Architecture in Medieval India: Bhoja’s
Samarān.gan.asūtradhāra and the Bhojpur Line Drawings (Indira Gandhi National Centre for the
Arts, 2015), 4.
Vasudevan 2003, 44.
These are all enumerated at great length in SII II.
Vasudevan 2003, 44.
Ibid., 46.
See, for example, SII XIV.
Richard H. Davis, Ritual in an Oscillating Universe: Worshipping Śiva in Medieval India (Princeton
University Press, 2014), 5.
Ranna, Gadāyuddham: The Duel of the Maces, ed. Akkamahadevi, trans. R.V.S. Sundaram and
Ammel Sharon (Manohar, 2019), 19.
Ibid., xvi.
Ibid., xvii. The technical term is rasa, which means ‘essence’.
Ibid., 11.
Pollock 2006, 155.
This process was started by Vikramaditya V.
Pollock 2006, 159.
Ibid.
Ibid., 156.
Ibid.
IA XVI, 21.
Pollock 2006, 158.
Ibid., 156.
Ibid., 157.
Ali 2006, 82.
Pollock 2006, 160.
EHD I, 324. The grand old stupa itself was not damaged.
An extensive discussion can be found in Spencer 1976.
EHD I, 324.
EI XVI, 75.
Sastri 1955, 176.
Death while defending ‘women’s honour’ is quite frequently celebrated in hero stones. See S. Settar
and Gunther D. Sontheimer, eds. Memorial Stones (Institute of Indian Art, 1982).
SII II, 13.
EHD I, 327. This was Jayasimha II.
Ibid., 325.
Pollock 2006, 179, citing a Paramara period source.
EHD I, 326. If Bhoja Paramara was celebrating the conquest of the Konkan with Brahmins from
Manyakheta and Vatapi in 1020, it must have taken him some time to establish enough of a
permanent presence to do this. It may be that he took advantage of Rajendra’s invasion, or that
Rajendra took advantage of his, or even that these attacks were coordinated in some way. See note
below.
Professor Sastri is of the opinion that this event happened before 1008, because in Rajendra’s first
claim to have captured ‘Mannaikkadakam’ (Manyakheta) he mentions Satyasraya as the Chalukya
king, and this man died in 1008. See also SII III, 18. However, the Karandai plates of Rajendra I
(Krishnan 1984), issued in his eighth regnal year (1021–22), explicitly say that Rajendra was only
able to capture Manyakheta after his father’s death, which was in 1014. In addition, Jayasimha II,
Satyasraya’s son, is known to have used Kalyana as his capital in 1033. The destruction of
Manyakheta must thus have happened during 1014–33. In the 1020s, Rajendra was occupied with
organizing raids along the eastern coast and the second set of raids on Kedah. (He sent out one raid
in 1017.) In 1018–19, Chola armies were busy in Sri Lanka, as per the Mahāvam.sa and
Cūlavam.sa. At a strategic level, the likeliest date for the burning of Manyakheta is therefore in the
late 1010s, when Jayasimha was preoccupied with Bhoja’s invasion attempt to his north. This leads
us to the conclusion that Rajendra’s earlier claim of having captured Manyakheta in 1008 was
intended to portray his early career in Karnataka as successful to his domestic audience; it was
most likely a limited raid on its suburbs that was repelled by Chalukya armies.
Krishnan 1984, 199. Of course, the inscription only describes Manyakheta, and does not specify that
the army actually encamped in the sal orchards, but it seems likely.
Ibid.
Ibid. The poet is quite clear about the fact that Rajendra himself was at his capital when this act
happened.
Ibid.
Ibid. The poet’s name was Narayana, son of Samkararya and a resident of Parshvagrama. I have
simplified the language for readability.
Ibid. I have simplified the language for readability.
This would be Akkadevi, the sister of Jayasimha II, the formidable aunt of Someshvara I, who will
appear in the next chapter.
IA VI, 54 and EI XVIII, 321 via EHD I, 326.
Krishnan 1984, 199.
Spencer 1976.
See Noboru Karashima, ‘Medieval Commercial Activities in the Indian Ocean Region as Revealed
from Chinese Ceramic-Sherds and South Indian and Sri Lankan Inscriptions’, in Kulke,
Kesavapany and Sakhuja eds. 2019.
Geraldine Heng, ‘An Ordinary Ship and Its Stories of Early Globalism’, Journal of Medieval Worlds
1, no. 1 (2019): 13, https://doi.org/10.1525/jmw.2019.100003.
Ibid., 42–43.
Ibid., 49–50.
Sen 2019, 63.
Ibid., 17.
Subbarayalu 2012, 40.
Ibid., 190.
R. Champakalakshmi, Trade, Ideology and Urbanisation: South India 300 BC to AD 1300 (Oxford
University Press, 1996), 317. See also Noboru Karashima, ‘South Indian Merchant Guilds in the
Indian Ocean and Southeast Asia’, in Kulke, Kesavapany and Sakhuja eds. 2019, 151.
Sen 2004, 156–58.
Karashima 2019, 141.
Subbarayalu 2012, 177–78.
Ibid., 178.
Sen 2019, 69.
Noboru Karashima and Tansen Sen, ‘Chinese Texts Describing or Referring to the Chola Kingdom as
Zhu-nian’, in Kulke, Kesavapany and Sakhuja eds. 2019, 297–98.
Hans Bielenstein, Diplomacy and Trade in the Chinese World, 589–1276 (Brill, 2005), 77.
Ibid. Emphasis added.
Karashima and Sen 2019, 299.
Ibid., 77–78.
For a detailed discussion, see Tansen Sen, ‘The Military Campaigns of Rajendra Chola and the
Chola-Srivijaya-China Triangle’, in Kulke, Kesavapany and Sakhuja eds. 2019.
Krishnan 1984, 200.
Kulke 2019, 6.
EHD I, 327.
This ended at the inconclusive battle of Maski (Musangi). See EHD I, 327–28.
Sastri 1955, 205–06.
EHD I, 328.
Hermann Kulke, ‘Śrīvijaya Revisited: Reflections on State Formation of a Southeast Asian
Thalassocracy’, Bulletin de l’Ecole Française d’Extrême-Orient 102, no. 1 (2016): 62–63.
Nitish Sengupta, Land of Two Rivers: A History of Bengal from the Mahabharata to Mujib (Penguin,
 2011).
EI IX, 233.
Sastri 1955, 209–10.
The lords of the Deccan played a significant role in the politics of east-central India: both the
Kalyana Chalukyas and Rashtrakutas intermarried with the Kalachuris of Chedi, and in the later
part of the Chola–Chalukya wars, the Kalyana Chalukyas even attempted to seize forts such as
Chakrakuta in Chhattisgarh. See Sastri 1955, 206.
Ibid., 255.
Ali 2006, 99.
The Inscriptions of Nagai, 19.
Ibid.
Nalini Sadhale and Y.L. Nene, ‘Sarameyavinōda in Mānasollāsa: Dogs for Recreation and Hunting’,
Asian Agri-History 14, no. 3 (2010),
https://www.asianagrihistory.org/pdf/articles/8_Sarameyavinoda.pdf.
Ali 2006, 100.
Subbarayulu 2012, 214.
Daud Ali, ‘The Service Retinues of the Chola Court: A Study of the Term Veḷam in Tamil
Inscriptions’, Bulletin of the School of Oriental and African Studies 70, no. 3 (2007).
B. Arunachalam, Chola Navigation Package (Maritime History Society, 2004), 53.
Noboru Karashima and Y. Subbarayulu, ‘Ancient and Medieval Tamil Sanskrit Inscriptions Relating
to Southeast Asia and China’, in Kulke, Kesavapany and Sakhuja eds. 2019, 279–80.
SII II, 28.
Karashima and Subbarayulu 2019.
SII II, 28.
Kulke 2019, 64.
Ibid.
Ibid.
SII II, 28.
EHD I, 330. Jayasimha II does not seem to have issued inscriptions from Kalyana.
Kulke 2016, 68.
Ibid.
Karashima and Sen 2019, 299–300.
Sen 2004, 156.
Kulke 2019, 10. See also Karashima and Subbarayulu 2019.
Kulke 2019, 43.
See P. Shanmugam, ‘India and Southeast Asia: South Indian Cultural Links with Indonesia’, in
Kulke, Kesavapany and Sakhuja eds. 2019.
See Hema Devare, ‘Cultural Implications of the Chola Maritime Fabric Trade with Southeast Asia’,
in Kulke, Kesavapany, and Sakhuja eds. 2019.
Risha Lee, ‘Rethinking Community: The Indic Carvings of Quanzhou’, in Kulke, Kesavapany and
Sakhuja eds. 2019.
Davis 1993, 42.
Ibid.
Ibid.
 9
Lord of the Moon
Summer, 1042 CE

Pattadakal, the ancient coronation site of the Chalukya Vallabhas of the

Deccan.1
Here, on the shores of the Malaprabha river where generations of Vatapi
Chalukyas had built their temples and made their boasts of conquest, a new
Vallabha, Someshvara I of the Kalyana Chalukyas, was about to be
crowned. The ancient royal consecration of Fortune’s Favourite was
enacted once again, as it had been for centuries. His family’s ancient crown
– probably very similar to the crown on the cover of this book2 – was
lowered onto his head. The dharma of kings was to conquer, to protect, to
seek fame. Perhaps Someshvara believed it was his duty to carry that
burden against the might of the Chola juggernaut.
It would take all of the new Vallabha’s political, diplomatic and strategic
guile to do so. The Chola imperium was now led by the ruthless Rajadhiraja
Chola: Rajendra Chola’s eldest son and co-ruler, the deadliest enemy the
Deccan had yet seen. Those present at Someshvara’s coronation in 1042
must have expected that horrific destruction lay on the horizon. They knew
that the subcontinent’s great superpowers, the Chalukyas and Cholas, were
preparing for a new round of brutal competition across the vast geopolitical
chessboard of south India. Aristocrats and merchants, poets and priests
from the Konkan to Vengi, from Dhara to Kanchipuram, waited with bated
breath for the emperors to make their move.

Rajadhiraja Chola had burst onto the geopolitical chessboard of southern

India in the 1030s. Less than a decade after his father Rajendra’s
expeditions to the Ganga and Srivijaya, the old aristocracy of Tamilakam
proved that they were far from overawed. Other royal south Indian
dynasties whom the Cholas had supplanted, such as the Cheras and
Pandyas,3 rose in revolt. Warriors flocked to their ancient banners: the
Chera bow and the Pandya fish.
Rajadhiraja Chola now set out to punish the kings who had dared to
challenge the Conqueror of the Ganga. One royal rebel, we are told,
possessed a handsome head ‘(adorned with) large jewels, inseparable from
the golden crown’.4 Rajadhiraja had the rebel’s head cut off, crown and all.
Another heroic upstart, whose thick legs were normally adorned with gold
ankle rings, was captured in battle. Rajadhiraja had the ankle rings replaced
by fetters, after which he was ‘pleased to get him trampled’5 to death by his
elephant, Attivarana. Another aristocrat fled desperately from the battlefield
into exile. In Kerala, a somewhat braver group of minor kings were
summarily ‘sent … to the country of heaven’.6 Finally, replicating his
grandfather Rajaraja Chola’s feats, a fleet of ships at Kandalur was burned
to ashes, after Rajadhiraja had forced the ruler of the city to flee and hide in
the thick jungles of the Western Ghats.7
While the Chola armies were thus occupied in the 1030s, the Chalukyas,
under Someshvara’s father Jayasimha II, had crossed the Tungabhadra.
They subdued the kings of the south-eastern Deccan8 and even expanded
into southern Chhattisgarh. The citadel of Chakrakuta (modern-day
Chitrakoot in Bastar district), previously seized by the Cholas during
Rajendra’s 1022–1023 Ganga expedition, was captured9 to act as a base for
Chalukya diplomatic and military efforts along the east coast. The Kalyana
Chalukyas planned to use the lords of Chhattisgarh and Odisha to hem in
Chola expansion in that direction. This was accompanied by another
Chalukya attempt to seize Vengi and install their close ally Vijayaditya VII
of the Vengi Chalukyas as king. This attempt was ultimately repelled by
Chola forces, but by the late 1030s, it was clear to all the rulers of the
subcontinent that the apparently invincible Chola tiger had been bloodied.
It is possible to discern some broad patterns in Someshvara’s geopolitical
manoeuvres against the Cholas after he came to the throne in 1042. He
seems to have understood that the Kaveri river valley alone would not be
able to provide the agrarian revenue and military manpower to sustain the
Cholas against the many challenges they faced. This was why the Cholas
had attempted, time and again, to seize the Raichur doab, between the
Tungabhadra and Krishna rivers. This ‘Idutarainadu’, as the Cholas termed
it, was so important to them that Rajendra Chola often mentioned it first
among his lists of conquests in his prashastis.10 Despite early Chola
successes in the region, Someshvara’s father Jayasimha II had managed to
retain Chalukya influence in Raichur, or Edudorevishaya (as the Chalukyas
called it). Vengi, between the Krishna and Godavari rivers, offered another
such fertile base for Chola power, and unlike Raichur, it was firmly under
the Chola sphere of influence. Meanwhile, the lucrative Konkan coast had
only recently been regained by the Chalukyas from the Paramaras. As
Vallabha, Someshvara would thus need to better integrate the Raichur doab
and the Konkan into his domains. Furthermore, he had to ensure that the
Cholas could no longer draw sustenance from Vengi.
And so, when Rajendra Chola died in 1044, Someshvara Chalukya
immediately moved to take advantage. Vijayaditya VII was once again sent
to Vengi with a Deccan army to overthrow his half-Chola half-brother
Rajaraja-Narendra. In this third and most serious attempt, so far, to claim
his birthright, Vijayaditya had the assistance of a new generation of Deccan
generals, who had earned their titles in decades of campaigns against the
Cholas. The Chalukyas also appear to have increased the size and capability
of their armies in the nearly twenty years since Rajendra Chola’s Bengal
and Srivijaya campaigns. The beleaguered Narendra desperately requested
Chola assistance, and an ‘equally matched’11 army commanded by three of
Rajendra Chola’s top generals set out to defend him.
But the toll of the Cholas’ endless raids and wars was becoming apparent.
The flower of the Chola army, the units and officers that had successfully
raided across the seas, were ageing. They had also taken severe losses over
thirty years of campaigning. Though constantly swelled by recruits from
hill forts and rice fields, the effectiveness of the Chola military machine
seems to have declined somewhat. All three commanders of this relief army
were killed by the Chalukyas in a pitched battle at Kalidindi in the modern-
day Krishna district. The Cholas were forced to temporarily retreat south
beyond the Krishna river, leaving the unfortunate Narendra to build temples
to Shiva over their generals’ ashes.12
Rajadhiraja Chola, who had by now been crowned emperor, took the
field. A challenge so early in his reign, in such a geopolitically crucial
region, could not go unpunished. With him, he probably brought substantial
reinforcements and new officers from his campaigns against the rebellious
Pandya and Chera kings. Successfully rallying the Chola forces, he
managed to blunt the Chalukya advance at Amaravati in 1045. ‘Vikki
[perhaps a disparaging reference to a general with the title of Vikramaditya]
and Vijayadityan [Vijayaditya VII of Vengi] … retreated like cowards!’13
declared Rajadhiraja’s eulogist.
The Chalukyas retreated, following the Godavari back west towards
Kalyana. At this point, two different versions of what happened emerge.
The Chalukyas claimed to have successfully fought off the Cholas at
Kollipakam (modern-day Kollipaka in Telangana): Someshvara Chalukya
awarded one of his vassals the title ‘Protector of Kollipaka’ that year.14
Rajadhiraja Chola, on the other hand, claimed in his inscriptions to have
burned Kollipakam.15
Though the Chalukyas had been unable to install their candidate in Vengi,
news of the Cholas’ struggles in this war spread rapidly through the
subcontinent. One after another, subjugated powers spluttered back to life.
Lanka rose in rebellion, with adventurers from across the subcontinent
seeking their fortunes in the chaos.16 The Chola emperor withdrew from his
conflict with Someshvara and moved to quell the uprisings to his south.
The Vallabha Someshvara, however, did not attempt to seize Vengi again
at this point of time. He had other scores to settle.

Over the past century, we have seen how Malwa – once little more than the
Deccan’s forward base for north Indian campaigns – had taken a deadly
revenge on its former overlords. First, Siyaka ‘Harsha’ Paramara had
sacked Manyakheta in 972. Next, Vakpati Paramara, the poet-king, had
attempted to claim the titles of Amoghavarsha and Sri-Prithivi-Vallabha,
Fortune’s Favourite, before being brutally killed by Taila, the Chalukya
ruler of the Deccan, c. 994. Bhoja Paramara, arguably the most acerbic and
brilliant of the Paramara family, had then wrested the Konkan from Taila’s
grandson Jayasimha II, c. 1020, and Chalukya authority in the region
remained weak despite later attempts to reclaim it. Now, Taila’s great-
grandson Someshvara would settle the score once and for all.
Rajendra Chola, after decades of war, had become the Indian
subcontinent’s great military celebrity king. But Bhoja Paramara had
embraced a more complex route to royal superstardom. Bhoja was the
greatest of royal Sanskrit litterateurs, acknowledged and celebrated by
critics and connoisseurs across South Asia. He was among the greatest
medieval patrons of Sanskrit literature, eclipsing Amoghavarsha
Rashtrakuta in a world still obsessed with that ancient language’s
marvellous sounds. (Indeed, as far as Bhoja was concerned, vernacular
languages, including that of his own subjects, were too unrefined to even be
considered worthy of literature.17)
Though it is generally believed Sanskrit had its golden age during the
Gupta period in northern India in the fourth to fifth centuries, it was really
now, in the eleventh century, that Sanskrit literature began to reach its
apogee. Courts across South Asia presided over waves of prashastis and
poetry and dramatic productions: this literary golden age of Sanskrit
produced hundreds if not thousands of texts across the vast subcontinent.
From Kashmir to Kannauj, from Varanasi to Vengi, from Gauda to Gujarat,
from Kalyana to the Kaveri river valley, salons and courts resounded with
Sanskrit literature – much of which is now lost. And none exemplified this
zenith of Sanskrit literature better than Bhoja.
With interests spanning a range of subjects from grammar to yoga to
astronomy to architecture,18 this Paramara king was a remarkable polymath,
seeking to compose, versify, patronize and write his way into eternal fame.
Encyclopaedic shastras, bringing together centuries of theoretical and
practical knowledge transmitted in Sanskrit, were composed under his
direct supervision. Poets from all directions converged upon his capital,
Dhara, where one of the Indian subcontinent’s greatest libraries of texts was
being amassed. The Paramara court resounded with urbane conversation,
poetry, drama and erudite debate on everything from aesthetics to
metaphysics. Bhoja himself oversaw it all. So qualified was he, so
appreciated for his generous patronage and marvellous insights, that his
salons soon became part of the subcontinent’s tapestry of legends. Twelfth-
century Kashmiri poets would bemoan never being able to meet him and
receive his gifts; the sixteenth-century Deccan emperor Krishna Raya of
Vijayanagara would declare himself Abhinava-Bhoja, the ‘New Bhoja’.19
Bhoja was also a builder of vision. On the Betwa river, he constructed a
dam of extraordinary proportions. On the shores of its reservoir, he began
the construction of a temple – the Bhojeshvara – intended to be nearly one
and a half times the size of the Rajarajeshvaram temple at Thanjavur, with a
10-foot-tall Shiva linga at its heart, still visible today. Referring to this
colossal project, architect Adam Hardy writes: ‘Mirrored in its lake it would
have doubled up like an expanding
śivalin.ga. Would Bhoja have declared anything less?’20
This Bhojeshvara was, unfortunately, to remain the greatest temple never
built, as the Deccan Vallabha Someshvara finally wreaked his vengeance on
the flamboyant, brilliant old Paramara king.
In 1046–47, four senior generals of the Kalyana Chalukyas – Nagadeva,
Gundamaya, Jomarasa and Madhuva21 – crossed the Narmada. The first
would declare himself a ‘Garuda to the serpent Bhoja’, the second, ‘a royal
swan strolling on both banks of the Narmada, an evil comet to the
Malavas’, the third, ‘the flame of doom to Bhoja’.22 Crossing the Narmada,
the citadel of Mandu was captured and burned by the Deccani armies, and
then Dhara itself besieged. Bhoja, pulled from his poetry and his building
projects, surrendered23 in a futile attempt to salvage the situation, but to no
avail. The Chalukya Vallabha was not interested in maintaining a vassal in
such a dangerous territory. Unlike his Rashtrakuta predecessors,
Someshvara had no ambitions for expeditions to northern India. Malwa was
important only insofar as it had to be reduced to the status of a secure
northern flank, which meant Paramara power had to be smashed once and
for all.
 And so Dhara, with its spires and pinnacles and poetry halls and temples,
was captured. Chalukya generals tore apart Bhoja’s palace, named after one
of his grammatical treatises,24 Sarasvati-kantha-abharana (The Necklace of
the Goddess of Knowledge), and seized its jewels and treasures. (To his
credit, Someshvara seems to have ordered that Bhoja’s library be left intact;
it was carted away by a Gujarati warlord soon after.25)The city was then
comprehensively sacked and set on fire. So too was the ancient city of
Ujjain, where Danti-Durga Rashtrakuta had once performed his Sacrifice of
the Golden Egg.
The kohl of the weeping widows of Malwa, dragged back along with the
victorious Deccan armies, supposedly turned the waters of the Narmada as
black as those of the Yamuna.26 So memorable was this event for Deccan
armies that one of the Chalukyas’ vassal dynasties – the famous temple
builders called the Hoysalas, who would later rule their own kingdom in
southern Karnataka – would integrate tales of their participation in this
campaign into their prashastis as evidence of their martial prowess. Indeed,
they would even replicate the architectural forms of Malwa in their great
twelfth-century imperial temple at Belur, 200 kilometres away from
modern-day Bengaluru.
As his generals returned in triumph, Malwa at last reduced to a state from
which it would never again rise to challenge the Deccan, Someshvara
himself attacked the Konkan, captured the last of its recalcitrant lords, and
‘wrung his neck’.27 Marriage alliances were arranged between
Someshvara’s family and local dynasties.28 His aunt, the fearsome warrior-
queen Akkadevi, ‘whose lotus-like feet are touched by the diadems of
opponent kings’,29 was positioned to guard the passes of the Western Ghats
as the Vallabha withdrew in triumph. His armies laden with loot,
Someshvara headed to Kalyana, preparing for a glorious celebration.
Unfortunately, Someshvara Chalukya was not the only emperor headed to
Kalyana that year.

Over the past two years, Rajadhiraja Chola had savagely defeated all
challengers to his power. In Lanka alone, the Chola emperor had seized no
less than four splendid jewel-studded royal crowns, capturing and
mutilating kings, queens and their mothers alike.30 A fragile peace had
fallen upon the brutalized island.
Then, in 1047, Rajadhiraja decided he could do to the Deccan what the
Deccan had done to Malwa. He would burn cities and temples, and unleash
his armies to rape and sack and slaughter. He would at last integrate the
Raichur doab into the Chola dominions, and he would secure it by seizing
tirthas, setting up victory pillars, and destroying cities. In Kalyana’s ashes,
Rajadhiraja Chola planned to perform a ritual known as the virabhisheka,
the Hero’s Consecration, with the Chalukya court as hapless witnesses and
attendees. He would thus destroy every shred of glory and divine favour
that the Chalukya Vallabha could claim. He would make sure that no lord of
Kalyana or Manyakheta would dare threaten the Cholas again.
As Rajadhiraja Chola entered the Deccan, he was immediately attacked
by a swarm of smaller armies. The minor kings Gandaraditya, Naranan,
Ganapati and Madhusudhanan ‘of the fragrant garland’31 were swatted
away. Rajadhiraja then headed to the city of Kampili, ‘whose gardens
diffuse fragrance’,32 near the future site of Hampi/Vijayanagara on the
Tungabhadra river. Here, in what was likely the southern base of the
Chalukya emperors, Someshvara’s palace was captured, looted of all its
precious treasures – imported vases and plush clothing and jewellery, idols
and banners and gateways – and burned to the ground. Someshvara quickly
dispatched an army to intercept the invaders. The commander was one
Viccaya, a Telugu speaker,33 suggesting that by this time Someshvara had
begun to cultivate the chiefs of the Telugu country as well.
 The two armies met at Pundur, on the south bank of the Krishna. Here
Viccaya was defeated, his parents, officers and elephants captured, and
paraded before Rajadhiraja in the Chola camp.34 The Chalukya emperor
Someshvara ‘Ahavamalla’, Great in Battle, sued for peace immediately:
Pundur was only a few weeks’ march away from Kalyana itself. But
Rajadhiraja Chola was not having it. The Chalukya embassy was humiliated
and thrown out of his camp, forced to wear banners around their necks
declaring that ‘Ahavamalla is a Despicable Coward’,35 to the jeers of the
Chola army. The unfortunate city of Pundur was now witness to the first of
Rajadhiraja’s atrocities in the Deccan: it ‘was razed to the ground, its site
being ploughed by asses and sown with … coarse millet’.36 Refugees must
have fled the invading Chola in desperation, families breaking their feet on
the rocky Deccan soil, exhausted elders left to fend for themselves,
traumatized children wailing at night.
The bloodthirsty Chola king now rampaged around the Raichur doab,
displaying his might to ruined towns and crowds of fleeing villagers. This
fertile region, as noted above, had already been the site of much warring
between the two dynasties during the reign of Rajadhiraja’s father Rajendra.
Seizing it again seems to have been one of the Chola emperor’s primary
objectives in this 1047 campaign. Rajadhiraja then headed slowly north,
capturing sacred river fords to deprive the Chalukyas of their spiritual and
material benefits, and bathing his herds of elephants in them37 in a
breathtaking display of contempt and confidence.
Someshvara, meanwhile, seems to have dispatched army after army to
stall Rajadhiraja’s advance. As effective as the Chalukya forces had been in
Vengi in 1044, it appears that they could not match a full-fledged Chola
imperial army. Rajadhiraja, in his prashastis, would claim to have driven
away many senior Chalukya generals.38 But every Chalukya defeat bought
time. As Rajadhiraja advanced inexorably towards Kalyana, Someshvara
was moving out of his city, transporting his court and treasury to safer
territory in the east, where the Cholas could not follow without stretching
their lines of supply and communication to breaking point.39
Meanwhile, Rajadhiraja crossed the Krishna and planted a victor pillar at
Yadgir, emblazoned with the crest of the Cholas, the leaping tiger.40 Now,
his camp decorated with fluttering flags, he received yet more ambassadors
from Someshvara. These, too, were humiliated: one of them was forced into
a sari and dubbed ‘Ahavamalli’. Another’s head was shaved badly, leaving
cuts and scrapes and five tufts, and was declared ‘Ahavamalla’.41 Then,
covered with bruises, their status stripped away, they were thrown out of the
Chola camp. The message was clear: no peace; only complete defeat.
Rajadhiraja then continued onwards to Kalyana. The capital of the Deccan’s
foremost dynasty was captured and methodically sacked and burned, like
Manyakheta before it.
But Rajadhiraja’s arch-enemy Someshvara was nowhere to be found.
Though he had wished to humiliate the Chalukya emperor by forcing him to
witness his Hero’s Consecration, the virabhisheka, Rajadhiraja could not
‘force his rival to take the role of ceremonial doorkeeper in person’,42 as
Danti-Durga Rashtrakuta had once done to the Pratihara kings of Malwa
during his Golden Womb sacrifice in the captured city of Ujjain, c. 750.
Instead, a statue of a door guardian43 was seized from a Chalukya temple to
‘witness’ Rajadhiraja’s reconsecration, his rebirth as ‘Vijaya-Rajendra’, the
Victorious Rajendra, Lord of Kings, in an elaborately staged spectacle at
which loot and captives were paraded.44 The Deccan, it seemed, lay broken
at the Chola emperor’s feet.

The Vallabha still had some contingencies he could call upon in this
moment of apparent Chola apotheosis. Moving east, into Telangana, what
remained of the Chalukyas’ armies were joined by the forces of an
ambitious local chiefdom, the Kakatiyas. Telangana at the time was similar
to the ancient homeland of the Chalukyas, the Malaprabha river valley: it
was a dry, war-torn inland region watered by rain-fed rivers, home to
itinerant pastoral groups. The Kakatiyas would one day transform it with an
imperial network of vassals and temples and tanks, akin to that of the rest of
the Deccan. That day was, however, far in the future.
In the sweltering, dry heat of Telangana in 1048, fanned by attendants and
sipping on buttermilk45 cooled in earthen pots, Someshvara and his generals
probably argued over how best to dislodge Rajadhiraja Chola. The Kakatiya
ruler and his son were almost certainly present with them.46 It is tempting to
imagine what these planning sessions looked like: it is unknown whether
medieval Indian strategists even used maps, but we can guess that military
officers and aristocrats of various ethnic, religious and linguistic groups,
clad in everything from simple cotton to luxurious silks, presented their
ideas to the Vallabha.
Rajadhiraja, now back in the Raichur doab, was gradually spreading his
devastation west.47 It was only a matter of time before the grandees of the
region began to swear their loyalty to Rajadhiraja out of sheer desperation.
The Chalukya court settled on a risky solution: they would take advantage
of Rajadhiraja’s preoccupation with the Deccan to raid Tamilakam itself,
forcing him to turn back and defend his core territories. It was the only way
to ensure that Someshvara’s imperial network would not unwind and
collapse. Kakatiya reinforcements – probably light infantry used to raiding
in the drylands of Telangana – were gathered. A Chalukya general with the
title of Pulakeshin was appointed to lead the force alongside the son of the
Kakatiya chief. This Pulakeshin was ordered to raid and set fire48 to what
was, for the Chalukyas, the most symbolically important of all the Tamil
cities: Kanchi, the ancient capital of their dead rivals the Pallavas,
previously seized by two Vikramaditya Chalukyas.
Precisely how this raid was put together and sent on its way is unclear, but
it must have been a race against time: should the Cholas get wind of it too
soon, Rajadhiraja would certainly move to stop this Chalukya force before
it got anywhere near Tamilakam. The Chola emperor could not afford to be
humiliated in Kanchi.
For just as the old Pandya and Chera heartlands had proved recalcitrant to
Chola authority, so had the Pallava core region around Kanchi. Here, in
drier territory quite unlike the fertile valley of the Kaveri, the Cholas had
only poorly integrated the patchwork of powerful warrior gentry,
agriculturist groups and local ‘independent zones’ around Brahmin
settlements established by the Pallavas centuries ago.49 Nevertheless, the
territory had been renamed soon after Rajaraja Chola conquered it in the
late tenth century, and it bore the name of Jayan-Gonda-Chola-Mandalam –
one of Rajadhiraja Chola’s titles.50 It was integral to the prestige of the
Chola emperor and the polity that his grandfather had designed.
General Pulakeshin and his Kakatiya forces were more successful than his
namesake, the long-dead Vatapi Chalukya Vallabha Pulakeshin II, more
than four hundred years ago. Kanchi was captured, its gates torn down,51 its
suburbs ransacked and burned, and fires lit to cover the raiders’ escape.
This sacking – made all the more humiliating because of the fact that it was
a ‘Pulakeshin’ doing it – forced Rajadhiraja Chola to finally withdraw from
the Deccan with whatever loot he had seized.
This entire series of events – Rajadhiraja’s repeated humiliation of
Chalukya ambassadors, the erection of victory pillars, his use of the door
guardian at his virabhisheka at Kalyana, and Pulakeshin’s raid of Kanchi –
remind us once again that for medieval Indian kings, military campaigns
were as much about stating and reinforcing power through symbols as they
were about capturing territory and treasure.
By August 1049, after the horrible deaths of thousands and incalculable
destruction, Rajadhiraja Chola had failed to secure his grip over the Raichur
doab. It would be years before he managed to gather enough forces to
attempt another invasion. Someshvara and his queens made land grants in
the southeast Deccan,52 not too far from Chola territory, a defiant show of
normalcy and Chalukya suzerainty. Despite all the destruction and
humiliation that Rajadhiraja Chola had rained down upon the Chalukyas,
attempting to reduce them to political subjection53 as he had the rest of the
south, Someshvara Ahavamalla had maintained his grip over the Deccan.54
The Cholas had, for now, been wrestled to a standstill, and Raichur still
remained under Chalukya influence, though it had taken a desperate gamble
and the loss of Kalyana itself to do so.

Now, Kalyana would be remade. At a strategic site straddling the crossroads

of the emerging regional language zones of Kannada, Marathi and Telugu,55
Someshvara Chalukya would build a new Kalyana, a capital of such
magnificence that it would dominate the Deccan imagination till well into
the sixteenth century.
Someshvara and his architects chose an area close to the ruins of the old
Kalyana that Rajadhiraja Chola had destroyed, or perhaps cleared the rubble
and ash to create a new layout.56 According to architectural texts from
medieval southern India, such as the Manasara Shilpashastram, a wide area
at the city centre, or slightly away from it, would first have been measured
and cordoned off for the Vallabha’s palace57 and temples associated with the
royal family. This would act as the city’s sacred, political and administrative
centre. A road would then have been demarcated around it, and from there,
arterial roads laid out aligned to the cardinal directions. In theory, these
would act as the basic anchor points of an enormous geometric grid,
positioned so that its ‘energy’58 radiated from Someshvara’s own residence,
creating a complex pattern of plots intersected by streets and lanes, with
areas demarcated for castes. Brahmins, courtiers and generals were
supposed to be closest to the palace; merchants and traders next to arterial
roads; and the productive ‘lower’ castes – tailors, shoemakers, butchers,
cowherds, weavers, servants, blacksmiths, oil pressers and so on – were to
be segregated into streets and neighbourhoods59 well away from the main
ritual hubs.
This, at least, is how the shastras describe it. However, archaeological
evidence from a major medieval Deccan city, Dvarasamudra (modern-day
Halebidu), built by the Hoysalas, vassals and successors of the Chalukyas,
complicates the picture. If Dvarasamudra is any indication, caste and class
divisions in the layout of medieval Deccan cities were rather more fluid60 in
practice. Here, the most prominent occupational groups (not just of high
caste) were clustered around the city’s main arterial road as well as the
royal palace and temples.61 The city was encircled by a wall in turn
surrounded by a wide ditch, with enormous gateways where the arterial
roads met it at the cardinal directions.62 Dvarasamudra was not particularly
large by modern standards: the walled city was about as big as a modern
residential neighbourhood, and home to only a few thousand people.
Someshvara’s Kalyana was most likely of a much larger scale. We know
from literature that Kalyana had suburbs, temple complexes, and lanes
populated by goldsmiths, silversmiths, jewellers and merchants of exotic
imported goods to serve the tastes of its elites. As inscriptional depictions of
other Deccan towns suggest, Kalyana probably blurred the lines between a
‘rural’ and ‘urban’ area, with agricultural land, gardens and orchards both
within and outside the city walls. The Chalukyas certainly planned it to be a
most imposing place; as one Sanskrit text puts it, ‘There is not, was not, nor
will ever be a city on this Earth like Kalyana.’63 It may have been one of the
densest urban centres in the Deccan at the time. Further archaeological
work at its site – modern Basavakalyan – is needed in order to learn more
about it.
The Chalukya court had defiantly returned to the very site of its greatest
defeat – where, not so long ago, a Chola emperor had smeared himself with
its ashes and performed his Hero’s Consecration. Generals and ministers
hobnobbed with courtesans and merchants in glittering two- to three-storey
edifices with luxuriously appointed salons with gilded wooden furniture,
residential suites decked with sumptuous Chinese silks and elegant
murals.64 But outside these rarefied circles, drawn by all variety of social,
political, spiritual and economic opportunity, the people of the Deccan
flocked to Kalyana. Unfortunately, the archaeological record is too
fragmentary to build an understanding of the struggles and joys of the lives
of this vast majority of the Deccan’s peoples.

A fragile peace now fell across the Deccan. Nearly three centuries ago, the
Vatapi Chalukya prince Vikramaditya II and his father Vijayaditya had
toured the Deccan, establishing the rights and duties through which cities
and guilds were bound to the royal family. So too would Someshvara and
his family.
However, by this time, the social and political context within which these
Deccani royals were operating was quite different. The ambitions and scale
of this Deccan imperial polity were vast. There were now dozens more
major cities that required the direct attention of the court, and far more
powerful local aristocrats needed to be won over through titles, grants,
alliances and privileges. Inscriptions
mention a number of ‘royal cities’, probably directly linked with the
imperial family: Annigere, Mulugunda, Nadapura, Kohalli, Mandaligere,
Belgali, Banavasi, Karividi, Navile, Nandavadige and Peruru.65 The
inscriptions of the long-faded Vatapi Chalukyas, in comparison, rarely
mention anything approaching this degree of urban complexity.
In comparison to the Vatapi Chalukyas and Rashtrakutas, many more of
the imperial ladies played important roles in the consolidation of Kalyana
Chalukya power.66 This was, perhaps, a tacit acknowledgement of the
increasing power of local Deccan aristocracy, the birth families of the
Vallabha’s wives, in relation to the imperial court. These families, once
petty warlords leading raids from the security of mud forts, had gradually
transformed into kings, the owners of immense plantations toiled over by
bonded agricultural labourers. Their hill forts were now urban and
commercial hubs67 that funded retinues, armies and courts steeped in both
Sanskritic and Kannadiga courtly culture. The Hoysalas of southern
Karnataka, tribespeople from the Western Ghats who had risen to the status
of rulers, are an excellent example: they had achieved considerable wealth
and status as feudatories and generals of the Chalukyas, as we have seen
earlier. One of Someshvara’s queens, Hoysaladevi, came from that family.68
Another queen of Someshvara’s, Ketaladevi, managed a major Brahmin
agrahara. A senior queen, Mailaladevi, governed the Banavasi 12,000 in
1053.69 (The number in medieval Deccan district names establishes their
‘rank’ in comparison to other territories, and might refer either to the actual
number of villages within them or merely be a designation of status.)
Banavasi was once the seat of the ancient Kadambas whom the Chalukyas
of Vatapi had overthrown centuries ago, and thus the intention behind
Mailaladevi’s appointment seems to have been to declare the imperial
court’s personal commitments to these prestigious and long-standing hubs
of social and political power. The governorship of the Banavasi 12,000
would later be held by Someshvara’s cousin, the son of his aunt Akkadevi
by a princeling descended from the Kadambas,70 and by Someshvara’s
second son, Prince Vikramaditya.71
 This queen, Akkadevi, is worth briefly dwelling on. She is described as ‘a
very Bhairavi in battle and in destroying hostile kings’,72 ruling over the
districts of the Kisukadu 70, Roragare 60 and Masavadi 140,73 which
suggests that she controlled a total of two hundred and seventy villages.
This formidable queen thus seems to have commanded armies in her own
right – generally an exclusively male prerogative – and had even been
involved in Someshvara’s subjugation of the Konkan in 1047. By 1050, she
was invested with the authority to send out ‘seven royal ministers and other
administrative officials’ who granted to ‘the eight Settis [senior merchants]
and eighty households’ of the town of Sudi ‘a renewal of their corporate
constitution which had partly broken down in the stress of the war with the
Cholas’.74 Sudi was an important commercial and urban centre, a
transregional trading hub with the authority to mint its own coins to
facilitate exchange – and thus crucial to Chalukya power.75
Someshvara, meanwhile, seems to have set in motion a serious attempt to
connect the Deccan’s many power centres directly to the imperial court,
expanding the reach of the Chalukya chancellery and an institution called
the ‘Office of Records’. In the late tenth–early eleventh centuries, when
Taila Chalukya had seized the title of Fortune’s Favourite in anarchy, and
his sons and grandsons struggled to fend off invasions from two fronts, the
process through which the court interacted with the Deccan’s many local
power centres was somewhat ad hoc. High-ranking military officers ruled
over them under the direct authority of the Vallabha, with awe-inspiring
titles such as Maha-Prachanda-Danda-Nayaka, Great Furious Commanders
of the Forces. These grandees conducted the administration of justice, tax
collection, land grant composition and so on, through members of their
personal retinues.76
But by Someshvara’s reign, ‘a process of regularisation or rationalisation’
had set in.77 An ‘imperial chancellery’ of sorts was beginning to emerge:
some varieties of local administrative functions, especially related to the
composition and recording of land grants, were permanently associated
with official positions tied directly to the imperial authority.78 These
officials now executed their duties at the authorization, rather than the
command, of whatever aristocrat happened to be shuffled around to oversee
a region by the imperial court. This does not seem a professional
bureaucracy as we would know it, but rather a new, non-military elite class,
often highly mobile Brahmins: pandits from Kashmir, especially, were
drawn to lucrative careers in the Deccan. As scribes, poets and
administrators without local roots and thus loyal only to the imperial court
at Kalyana, they appear to have been integral to Someshvara’s attempts to
counterbalance the influence of powerful aristocrats.79 These individuals
helped standardize another policy of the Chalukyas of Kalyana: the
resurgence of Sanskrit in imperial land grants.
We have seen how, after the composition of Amoghavarsha Rashtrakuta’s
landmark Kavirajamarga, Kannada had very rapidly begun to replace
Sanskrit in royal land grants. The Rashtrakuta emperors had even begun to
adopt titles of mixed Sanskrit and Kannada. By the time of the Kalyana
Chalukya grants, while the vast majority of stone inscriptions – usually
commissioned by local grandees declaring fealty to the imperial court –
were in Kannada, the land grants of the Chalukya emperors themselves,
usually made on copper plates roughly the size of modern A4 sheets, were
all in Sanskrit. Apparently, the ambitions and worldviews of the Vallabhas
had shifted again: Someshvara’s family were seeking to associate
themselves once again with the prestige, transregional appeal and eternal
allure of the subcontinent’s grandest language of power. The imperial
chancellery seems to have ensured a remarkably consistent standard
Sanskrit script across the reign of the Kalyana Chalukyas: Nagari (rather
than, say, Old Kannada letters, which was more frequently used to write
Sanskrit words in the region). Nagari was the script used in the inscriptions
of the Vatapi Chalukya king Kirti-Varman II and his Rashtrakuta rival
Danti-Durga, when they had struggled with each other for imperial
dominance of the Deccan in the eighth century.80 It had an extraordinary
currency and usage across the Indian subcontinent’s elites in the eleventh
century. It seems that Nagari witnessed a resurgence in eleventh-century
Deccan, and Sanskrit began to reappear in Chalukya land grants, in order to
reflect the imperial and transregional ambitions of Someshvara’s court. As
the land grants put it, they were meant to be maintained by all dynasties that
came after: how else should the Chalukyas command future generations
than through Sanskrit in the prestigious Nagari script? We shall return to
this Nagari script, related to the modern Devanagari script used for Hindi,
in the next chapter.
Let us now move beyond the court and see how far the Deccan has come
from when we first saw it: a dry, war-torn land with few cities and
dynasties, and an economy primarily based on loot and barter.
Half a millennium after our story began, we would see a society and
economy of vast scale. Perhaps the most significant change we would see is
the preponderance of currency: coins were now being minted by private
workshops (under royal supervision) in ever-larger numbers,81 not just gold
but lower denomination coins of copper as well. These coins were being
used with increasing regularity for everything from purchasing goods to
paying construction workers.82 The importance of this transformation
cannot be understated. Money is far more fungible than bushels of grain. It
can be moved around faster, and used to buy more varied and complex
goods and services. The increased availability of coinage helped establish
financial networks to extend credit, exchange currencies and pay interest on
deposits.83 The biggest owners of fixed assets – temples and landlords – had
also leapt into the game, lured by exorbitant interest rates ranging from 30
to 40 per cent per annum. The money supply exploded, and commercial
activity with it.84

Everything from ‘drugs, clothes, spices, perfumes, oil-seeds, jaggery,

white sugar, gems of every description and also grain, areca-nuts and betel
leaves’85 was available in prosperous market villages as also in large urban
centres. People even outside of cities no longer had to split their time
between cultivating and manufacturing as they had done for centuries. They
could simply manufacture and sell goods, and use the cash to buy food from
great agricultural estates which used bonded labour to undercut each other
in the markets. Lords and peasant groups went out of their way to invite
merchants and producers to their towns and markets, giving them free land
or concessionary tax rates in return for their aid in developing commercial
centres.86 Some towns were even administered by merchant princes with the
title of Pattanasvami, Lord of the City, working in close conjunction with
minor kings and landlords.87
This diverse, polycentric ruling elite, coming from an array of caste
backgrounds, occupying ever-shifting positions in relation to each other and
to various glittering courts shifting around Someshvara’s, all competed for
social status and for material possessions, kindling a ravenous appetite for
exotic produce from the farthest corners of the world.88 As we compare all
this to the relatively simple, pastoral, and barely urbanized landscapes of
northern Karnataka in the seventh and eighth centuries, the picture that
emerges is of a land that has been totally transformed through centuries of
warfare, agrarian and religious expansion, and growing political
complexity. Much of the Indian subcontinent went through similar
processes through the early medieval period, c. 600–1200, setting the
foundation for many of the social, economic, religious and political
structures that survive in some form today.
Medieval India had great cities, tournaments, forts, lords and ladies, poets
and priests, peasants and craftsmen, monasteries and temples. Its kings and
queens wore costumes of great diversity and sophisticated style, its lordly
houses wielded insignia granted by emperors, and even followed a code of
honour, as we’ve seen. Yet, today, the word ‘medieval’ in the global context
generally conjures up either an image of a European knight in shining
armour, or perhaps a sultan in the Middle East. This image of a medieval
Indian past that was as complicated as our present seems to have totally
vanished from our imagining of our own history, not to mention that of the
world.

Such a profusion of wealthy, status-conscious people, with the means to

hire flourishing sthapati guilds from the rapidly expanding market cities of
the Deccan, naturally led to a profusion of those most iconic of medieval
buildings: temples.
The eleventh century was a flourishing age of temple construction, just as
it was for Sanskrit literature. The Paramara king Bhoja of Malwa had been
able to dream of creating such a colossus as the Bhojeshvara because the
means and expertise to create it finally existed. The visions of great pillared
halls and temple spires that Bhoja probably saw in his mind as he composed
or revised great architectural tomes such as the Samaranganasutradhara,
could now be brought to life. The scale of the projects that we see in the
eleventh century, including Rajendra Chola’s Gangaikondacholeshvara and
the ‘Rani ki Vav’ stepwell in Patan, Gujarat (both UNESCO World Heritage
Sites today), prove that sthapatis could create load-bearing structures of
enormous size and weight, and adapt them to a plurality of local conditions
and design requirements. Their existence is a testament to the considerable
resources and engineering capabilities that states and sthapatis had
managed to accumulate since the fledgling stone temples of the sixth–
seventh centuries.

In the eleventh century, Deccani architects unleashed a vast wave of

temple building that has left us with a truly ‘glorious body of monuments’89
in Karnataka. Whereas in the seventh–eighth centuries royals were the most
important patrons of temples, the vast majority of eleventh-century temples
were built by feudatories and subordinates of the imperial court.
Unfortunately, many of these masterpieces are relatively unknown,
subsumed in our imagination of south Indian architecture by the thriving
pilgrimage sites and imposing temple gateways of Tamil Nadu. They can be
seen at Sudi: the Mallikarjuna and Joda Kalasa temples; Annigeri: the
Amriteshvara temple; Ittagi: the Mahadeva temple; and Badami, the old
home of the Vatapi Chalukyas, right in front of the great reservoir that still
exists today, the Yellamma temple.
The floor plans of the Lokeshvara and Krishneshvara temples built by the
early Chalukyas and Rashtrakutas had resembled squares projecting from
squares, with walls decorated with niches for sculpture. Their roofs rose up
in tiers, studded with carvings of miniature buildings. Made of hard rock
such as sandstone and basalt, these architectural decorations had a certain
thickness and weight to them; Gerard Foekema describes them as ‘chubby’.
But now, by the eleventh century, Deccan architects had arrived at floor
plans of greater complexity: squares projecting out of squares projecting out
of squares projecting out of squares projecting out of squares. They even
introduced a floor plan inspired by designs current in Malwa and northern
India, embedding rotated squares into other squares, creating stellate
designs – stars with anywhere from sixteen to thirty-two points. The Dodda
Basappa temple at Dambal, perhaps the most remarkable example of a
stellate temple in existence, is based on a twenty-four-point star.90
 With exponentially more surfaces to ornament, sthapatis incorporated
new materials to meet this design requirement. In the Dharwad region,
south of the Krishna river, soapstone and potstone91 were worked into
astonishingly florid architectural decorations depicting all sorts of fantastic
creatures and foliage.92 Pillars inside the temple took on polished, rounded
shapes almost reminiscent of lathe-turned carvings.93 The temple spires
maintained the tiered logic of Dravidian temples,94 but with many more
projections and repetitions of the main spire along the central axes,
following the floor plan composed of squares projecting from squares. The
emanating walls of these temples were decorated with carved pillar forms
bearing miniature depictions of dozens of different temple styles, reflecting
the vast variety of regional architectural idioms that its cosmopolitan
sthapatis were now familiar with – a far cry from the relatively simple
architectural depictions seen in early Deccan temples in Vatapi. The result
is temples that seem truly alive, almost dancing. The Ittagi Mahadeva
temple can be considered an example of this type.
North of the Krishna river, closer to the heart of Chalukya power, the
preference seems to have been for harder varieties of stone.95 The
architectural style also reflects north Indian influences, with tall plinths
taking up most of the bottom half of the wall, and tall, curvilinear brick
towers on the superstructure.96 Pillars in the halls generally look as though
assembled from stacked square or octagonal forms decorated with abstract
geometric shapes.97 Instead of pillar forms decorated with miniature
temples, the walls are adorned with sculptures of dancing women contrasted
against elegant flat surfaces – the sophisticated ‘Metropolitan’ style of
Kalyana and its environs.98 The Dattatreya temple at Chattarki, in the
modern Vijayapura district, exemplifies this style. The importance of these
two new Deccan architectural lineages will become apparent when we
discuss the successors of the Chalukyas in the next chapter.
Deccan sthapatis in the eleventh century competed to create the most
innovative floor plans, the most imposing temples, the most elaborate
sculptural ornamentations. The greatest shrines could be declared, just like
the greatest poets, ‘emperors among temples’,99 admired by rivals and
inspire imitations.

The thriving economic and cultural structures of the eleventh-century

Deccan could not have existed without a stable and resilient polity of vast
scale to enable them. The temples, coinage and land grants that have
survived to this period establish that such a polity did indeed exist, and that
unlike Sri Lanka or the kingdoms of India’s east coast, it was able to
successfully withstand the might of the Cholas on multiple occasions. And
the credit for this should go to the grit, determination and political abilities
of the court of Someshvara, Fortune’s Favourite, the Vallabha. Though his
career is eclipsed in popular memory by those of his contemporaries,
especially Rajaraja Chola and his sons, he deserves to be recognized for
manifesting and anchoring an ambitious and powerful imperial polity in the
Deccan, and for successfully holding his own against the forces arrayed
against him. Someshvara Kalyana Chalukya is one of the subcontinent’s
unsung political masterminds. His history, like that of the Deccan, is
unfortunately eclipsed by the stories of kings who fit better into nationalist
and militaristic representations of our past.
Someshvara’s consolidation of the Chalukya imperial network came not a
minute too soon. In 1054, Rajadhiraja Chola, having recovered from the
exertions of his 1048 campaign, once again invaded the Deccan.100 For six
years, he had trumpeted his successes and flaunted the door guardian he
seized from the ashes of Kalyana, inscribed prominently with the label ‘The
door-keeper brought by Sri-Vijaya-Rajendra-Deva after burning
Kalyanapuram’,101 a stand-in for the supposedly subjugated Vallabha,102 still
visible today at a museum in Gangai-Konda-Chola-Puram. Now, the
Deccan was ready to show the Cholas just how wrong their king’s claim of
victory was.
Rajadhiraja ‘Vijaya-Rajendra’ Chola, Victorious Lord of Kings, does not
seem to have anticipated what was in store for him. While he was busy
boasting of his victories, the ‘proud and furious’103 Someshvara had steadily
strengthened his own positions, dispatching an army led by a number of
veteran commanders to Koppam104 on the north bank of the Tungabhadra.
Koppam protected a crucial ford that Rajadhiraja would have to cross in
order to invade the Deccan again.
Rajadhiraja began his campaign with the same moves he always adopted.
Moving from the southern Deccan towards the Raichur doab, he ‘began to
ravage the country by defiling its rivers and demolishing its towns and
villages’.105 It would appear that his strategy was, once again, to parade his
forces around the region, loot it and perhaps capture the rebuilt Kalyana, in
which Someshvara had invested so much treasure. However, with a
Chalukya force at Koppam now prepared to challenge Chola attempts to
move beyond the Tungabhadra, he would have to deal with it first.
The commanders of the Chola and Chalukya armies would probably have
arrayed their troops with the river protecting one side of their formations,
and advanced towards each other.
There were probably vast regiments of infantry bearing spears, swords,
maces and everything in between. According to a late tenth-century source,
Tamil soldiers wore short-cropped hair and thick beards, with gold rings
swinging from their ears;106 it is possible that many continued to be
costumed as such in the mid-eleventh century. Their bodies were painted
with turmeric, their shields covered in blood-red lead.107 Evidence from
travellers’ accounts and Chola inscriptions suggest that on this day,
Rajadhiraja Chola, standing on a splendid howdah on the back of an
enormous elephant alongside his bodyguards, advanced near the front.108 He
was surrounded by a dense formation of heavily armed elephantry painted
with frightful designs; the lordly beasts, like Rajadhiraja and his officers,
bore titles awarded to commemorate their valour and prowess on the
battlefield.109
Arrayed against them were the senior generals of the Deccan, also on
elephants. Present on that day, supposedly, was one of Someshvara’s
younger brothers, as well as General Pulakeshin, the burner of Kanchi.110
They carried bamboo bows, coated with red arsenic, vermilion and lac,
decorated with jewels and tied to their hands with leather cords.111 These
men were trained to shoot at high speed, supposedly having an arrow ready
to launch as soon as one had been dispatched towards its target.112
After a period of skirmishing, when both sides traded projectiles, the
order to attack was given. Rank after rank of infantry clashed with an
enormous roar. As desperate foot soldiers yelled at each other, moving back
and forth across the battlefield, attempting to break their enemy’s cohesion
and stay alive in the chaos, the elephant-riding men of rank sought each
other out113 to display their combat abilities and bravery. ‘The battlefield
was to the warrior what the court or salon was to the poet.’114 The poor foot
soldiers were there to kill and be killed. But, as we saw in the previous
chapter, the aristocrats were there to perform, to display their prowess to
their followers and the Vallabha. Studying medieval hero stones and
inscriptions, and comparing with warfare in other parts of the world, we can
deduce that these warlords were drawn to the front lines by the banners of
enemy commanders, their approach accompanied by the cheering of their
troops, with poets announcing their lineages and achievements, and the
blowing of conches and trumpets. Some might even fling themselves into
knots of heavy fighting, inspiring their troops to fight ever harder; many
Deccan hero stones seem to commemorate precisely these kinds of deaths,
as do depictions in literature. For example:
A good fighter … holding with his left hand his restless head which, half-cut by an arrow,
remained on his neck, made everyone praise (him) … Some one, holding his mouth into which
the enemy-elephant thrust its small tusk … appeared as if sportively drinking the juice of the
nectar of fame with a lotus-stalk.115
On this day in 1054, Rajadhiraja Chola’s leaping red tiger banner was
sought out by just such a team of braves, riding crack elephants, probably
with a light infantry escort, led by the generals Mararasa and Bachiraja;116
the latter descended from the general Kalidasa who had rescued the
Chalukya emperor Jayasimha II, Someshvara’s father, from a coup attempt
in the 1020s. For whatever reason, the Chola emperor himself seems to
have ended up in the thick of combat – perhaps he was lured into a weak
point in the Chalukya line and attacked it without enough support, perhaps
he was overconfident about his abilities and led a cohort of elephants
directly into the thick of the fighting. Either way, in the chaos of the battle,
Mararasa found him, vulnerable, and attacked.
Rajadhiraja’s elephant probably came under a hail of Chalukya arrows,117
dispatched from Mararasa’s squadron on foot and from their elephant-back
howdahs. The Chola’s bodyguards must have desperately attempted to
cover him with heavy shields of cow and horse leather.118 Sharp elephant
lances, the heads twenty-one fingers across, with a curved hook and sharp
backward-facing triangular blades below,119 would have been stabbed by
Chalukya riders into any exposed patches of the Rajadhiraja’s elephant’s
skin that they could find – its eyes, trunk, genitals, belly and mouth,
shredding its flesh as the weapons were yanked out, forcing the screaming
animal to collapse.
An emperor of Rajadhiraja’s rank may well have been given the chance to
surrender, but that was not the Chola’s style. One is tempted to imagine the
Chalukya generals calling down to him in Sanskrit, only to be met with a
barrage of curses in Tamil. And so the ‘Victorious Rajendra’ was brutally
put to death on the battlefield, dying as he had lived, his head – perhaps still
anointed with the perfumed oil of his morning ablution – grabbed by a
bloodstained hand and hacked off.
The death of Rajadhiraja was a crushing blow to the Chola army’s morale,
but his younger brother Rajendra-deva was able to bring up the reserves and
drive away the Chalukya elephant squadrons before they could take
advantage.120 It was, even according to usually chest-thumping Chola
accounts, a desperate and closely fought affair, though of course they still
claimed victory. Though many senior Chalukya officers were killed at
Koppam – including General Pulakeshin,121 burner of Kanchi – and the
Cholas were left in possession of the battlefield, their ‘victory’ was entirely
hollow. Rajadhiraja Chola, the Victorious Rajendra, had gone ‘up into the
sky, welcomed by the women of the world of Indra’,122 and was eulogized
as Anaimerrunjina Devar, the King Who Died on the Back of an
Elephant.123 His younger brother crowned himself Rajendra II on the blood-
soaked ground the next day,124 and quickly retreated to forestall any
succession struggles or rebellions at home. No matter how much the Cholas
attempted to portray themselves as victors of the battle of Koppam, it was
evident to the entire subcontinent that they had been struck a devastating
blow. Lanka once again smouldered into rebellion,125 drawing away Chola
forces. The Deccan was saved.
Celebrations must have broken out across Someshvara’s cities, thanks
given to Shiva to the sound of bells, drums and conch shells. ‘The wicked
Chola, that deadly sinner, failing in his [royal] duty, abandoning the ancient
dharma of his family, invaded … he burned down a multitude of temples …
his guilt bore an immediate harvest in his hand and he gave his live head to
Trailokyamalla [Someshvara],’126 declared the triumphant Chalukyas. The
Chola emperor’s death was a resounding reminder, trumpeted by the court
to all who would hear it, that Someshvara was, indeed, Fortune’s beloved.
The general Mararasa was awarded the unique title of Rajadhiraja-Chola-
Gonda, the Destroyer of Rajadhiraja Chola, and declared Queen
Mailaladevi’s Champion,127 while Bachiraja’s family boasted of how he had
delivered the ‘freshly decapitated head’ of the ‘savage enemy’ to
Someshvara.128 This grisly trophy must have been the target of no small
amount of merriment in Kalyana, the object of parades and humiliation,
payback for the burning, rape, murder and looting it had once inflicted
when attached to a body. After all the struggles of the previous decades, it
must have appeared that Someshvara had, at last, saved the Deccan and
established it as the premier power of the south, as it had once been.
Pattadakal became the site of the imperial coronation once again after the establishment of the
Kalyana Chalukyas. See EHD I, 382.
The illustration was based on Prasanna Kumar Acharya, Architecture of Mānasāra: Illustrations of
Architectural and Sculptural Objects with a Synopsis, Mānasāra Series, Vol. 5 (Oxford University
Press, 1934), Plate No. 138. The Mānasāra is a medieval Deccan text, and the illustration is based
on a Chalukya sculpture.
Sastri 1955, 221.
SII III, 56. I have tweaked it to improve readability.
Ibid.
Ibid.
Ibid.
EHD I, 333.
The Chalukya general Nagavarma was already claiming this in 1047, and it seems to make sense that
this had happened by the events of the 1045 campaign. See also Shrinivas Ritti and G.C. Shelke
eds. Inscriptions of Nanded District (Yashwant Mahavidyalaya, 1968), xxiii.
Cox 2010, 5, footnote 9.
EHD I, 334.
EI XXIX, 62–63.
EHD I, 335. The authors suggest this was Prince Vikramaditya, Someshvara’s second son. This
seems a little doubtful as he died in 1126. Assuming this Vikki is indeed Vikramaditya, he must
have been nearing twenty if he was a commander, thus making him over a hundred years old at the
time of his death.
EHD I, 335.
Ibid.
Sastri 1955, 249–51.
Pollock 2006, 109.
Michael Willis, ‘Dhār, Bhoja and Sarasvatī: From Indology to Political Mythology and Back’,
 Journal of the Royal Asiatic Society 22, no. 1 (2012): 130.
Srinivas Reddy, Raya: Krishnadevaraya of Vijayanagara (Juggernaut Books, 2020).
Ibid.
K.N. Seth, The Growth of the Paramara Power in Malwa (Progress Publishers, 1978), 153.
Ibid., 154.
Arvind K. Singh, ‘Interpreting the History of the Paramāras’, Journal of the Royal Asiatic Society 22,
no. 1 (2012): 21.
Pollock 2006, 180.
Ibid.
Seth 1978, 152.
Venkataraya V. Sastri, ‘The Latter Days of Bhoja “The Great”’, Proceedings of the Indian History
Congress 10 (1947): 263.
Ritti and Shelke eds. 1968, xxiii–xxiv.
EI XVII, 123.
SII III, 56.
EHD I, 336.
SII III, 57.
EHD I, 336.
Ibid.
Ibid.
Ibid.
Ibid.
These might be exaggerations.
The extent of this evacuation is unknown, but it seems inconceivable that Someshvara would sit
around and allow Kalyana and its people to burn, when Rajadhiraja’s burning of Kampili and
Pundur had given him so much advance warning. It also helps explain how the Chalukyas were
able to bounce back from the destruction of their capital so quickly.
EHD I, 336.
Ibid.
Davis 1993, 34.
Ibid., 32.
Though we do not know the details of these parades, Rajadhiraja’s youngest brother Vira-Rajendra
would perform similar spectacles with loot from the Chalukyas. See Davis 1993.
This is speculation, but buttermilk was a prescribed drink for medieval Deccan kings. See G.K.
Shrigondekar, ed. Mānasollāsa of King Someśvara, Vol. II (Oriental Institute, 1939), 22.
The Kakatiyas were present at the sacking of Kanchi that followed soon after. See EHD I, 333.
In 1050, Sudi, further west from Rajadhiraja’s 1047 campaign trail, was granted ‘a renewal of their
corporate constitution which had partly broken down in the stress of the war with the Cholas’. See
SII XI-I, 84.
EC VIII, 56.
Ibid.
SII III, 57.
EHD I, 337.
SII XI-I, 75.
Ibid., 256.
Sastri 1955, 255.
Richard M. Eaton, A Social History of the Deccan, 1300–1761: Eight Indian Lives, The New
Cambridge History of India (Cambridge University Press, 2005), 99.
EHD I (330) suggests that ‘Someshvara ‘made’ the city ‘not in the sense of having founded it for the
first time, but in that of having adopted it as capital of his empire, adorned it with many new
buildings, and added to the amenities of life available there’. Of course Kalyana was already an
ancient site, but the authors may have neglected to consider the alternative presented here: namely
that Someshvara rebuilt Kalyana after Rajadhiraja’s burning.
Prasanna Kumar Acharya, Architecture of Mānasāra: Translated from Original Sanskrit, Mānasāra
Series, Vol. 4 (Munshiram Manoharlal, 1934), 65, 83.
This, at least, is how vastu experts see it (and thus perhaps how Someshvara’s sthapatis saw it as
well). See Sashikala Ananth, The Penguin Guide to Vaastu: The Classical Indian Science of
Architecture and Design (Penguin Books, 1998), 115.
Acharya 1934, 70–71.
Katherine E. Kasdorf, ‘Forming Dōrasamudra: Temples of the Hoysaḷa Capital in Context’, PhD
dissertation (Columbia University,
2013), 73.
Ibid., 74.
Acharya 1934 (Vol. 4), 79.
Eaton and Wagoner 2014, 7.
These details can be inferred from texts composed in Kalyana a little later, such as the
Vikramankadevacharitam, as well as Chalukya prashastis and sculptures.
SII XI-I, 100.
Adiga 2003, 178.
Nandi 2000, 118.
EHD I, 340.
Ibid.
See Gazetteer of the Bombay Presidency, Volume I Part II: History of the Konkan Dakhan and
Southern Maratha Country (Government Central Press, 1896), 435, 439.
Ibid.
IA XVIII, 274.
SII XI, 84, 87.
Ibid.
Om Prakash Prasad, ‘Trade in the Growth of Towns: A Case Study of Karnataka (c. 600–1200
A.D.)’, Proceedings of the Indian History Congress 41 (1980).
Cox 2010, 9.
Ibid., 10.
Ibid., 11.
Ibid., 10.
Ibid., 19.
Nandi 2000, 111.
Ibid., 108.
Ibid., 109.
Ibid., 109–10.
Ibid., 113.
Ibid., 117.
Ibid.
James McHugh, ‘The Incense Trees of the Land of Emeralds: The Exotic Material Culture of
Kāmaśāstra’, Journal of Indian Philosophy 39, no. 1 (2011).
Gerard Foekema, Architecture Decorated with Architecture: Late Medieval Temples of Karnataka,
1000–1300 (Munshiram Manoharlal, 2003), 49.
Adam Hardy, ‘Tradition and Transformation Continuity and Ingenuity in the Temples of Karnataka’,
Journal of the Society of Architectural Historians 60, no. 2 (2001).
Foekema 2003, 52.
Ibid.
Ibid., 82.
Eaton and Wagoner 2014, 80.
Ibid., 79.
Ibid., 80–81.
Ibid., 82.
Ibid., 79.
Hardy 2001, 194.
EHD I, 338.
Ibid., 336.
Davis 1993, 34.
SII III, 63.
Ibid.
EHD I, 338.
Handiqui 1949, 60. This might not be an accurate description of all the Tamil soldiers at Koppam in
1054, but it likely holds true for at least a proportion of them.
Ibid.
Karashima and Sen 2019, 296. Chinese sources describe elephants advancing at the forefront of the
Chola army.
Ibid., 303. The titles can be seen in vols I–III of SII.
SII III, 63.
Shrigondekar 1939, 27.
Ibid.
Ali 2000, 264. Medieval courtly texts such as the Manasollasa recommend this for men of the court.
Ibid.
Bilhana, Vikramankadeva Caritam: Glimpses of the History of the Cāl.ukyas of Kalyana, trans. Sures
Chandra Banerji (Sambodhi Publications, 1965), 266.
EHD I, 339. It is not explicitly stated that these two set out to kill the emperor; that is, however, what
the evidence suggests.
SII III, 63. Someshvara’s generals get the credit for this, but given the lack of concrete detail about
the battle, it is equally possible that Rajadhiraja died ignominiously at the hands of anonymous
archers. That said, his death is depicted as coming at the hands of Someshvara’s generals since that
is what the fragmentary sources attest to.
Shrigondekar 1939, 27.
Ibid., 30.
SII III, 63.
Ibid.
EHD I, 338.
Sastri 1955, 258.
Ibid., 257.
SII III, 63.
The translation provided in EHD I, 339 has been merged with that provided in EI XV, 345.
SII XI-I, 92.
EHD I, 339, and The Inscriptions of Nagai, 20.
 10
Fortune’s Favourites
Winter, 1054 CE

After the death of the emperor Rajadhiraja in 1054, as the Cholas turned
inwards, the Kalyana Chalukyas turned outwards. It seemed that
Someshvara’s career had reached a splendid apogee: Raichur was firmly
held by the Chalukyas, the Konkan had submitted, Malwa had been
crushed, and Vengi too would soon be captured.
Rajaraja-Narendra, the Chalukya king of Vengi, finally realized that there
could be no more resistance to the influence of the Vallabha of the Deccan.
His entire career had been spent chasing his half-brother Vijayaditya VII,
Someshvara’s dear ally, out of Vengi as their ‘political and martial fortunes
seesawed, intensified by their alliances with their powerful, self-interested
neighbours to the west and south’.1 As the growing weight of the Kalyana
court tipped south India’s centre of gravity once more to the west, it seemed
that Vijayaditya’s moment had come. But Narendra was not ousted and
replaced. Instead, the Vallabha Someshvara made a deal with him and
allowed him to keep his throne as the vassal king of Vengi. The Vallabha’s
eldest son, also named Someshvara, was appointed the ‘Lord of the City of
Vengi’,2 suggesting that the Deccan emperor intended Vengi to be ruled
directly from Kalyana after Narendra’s death.
Vijayaditya VII must have been aggrieved. He was compensated with
splendid governorships and titles elsewhere in the Deccan.3 But to a man
with his ‘incisive political mind,’4 this deal with his half-brother Narendra
must have been a humiliating betrayal by Someshvara. Vijayaditya kept his
peace for now, but he seems to have found a powerful ally, also sidelined,
also ignored: Someshvara’s second son, Prince Vikramaditya of the
Kalyana Chalukyas, who had fought in his father’s armies alongside
Vijayaditya for years. And so he began to plot with this man, as masterful a
politician as the old Vallabha himself.
Narendra, meanwhile, subsided into a sorrowful retirement under
Someshvara’s shadow, far away from his Chola relations. Narendra was the
son of a Vengi Chalukya prince and the Chola princess Kundavai, daughter
of Rajaraja I; he was married to Ammangai, daughter of Rajendra I and his
first cousin. Their son, named Rajiga5 for his Chola grandfather Rajendra,
was now far away in Tamilakam. This young man seems to have spent
much of his early life enamoured by the opportunities his maternal uncles,
the Cholas, could provide him – rather than what his father could offer. We
will hear from Rajiga again soon.
Apparently seeking a project to keep himself occupied, and perhaps
inspired by old Amoghavarsha Rashtrakuta’s success at creating a new
courtly Kannada, Narendra now commissioned one Nannayya, a Brahmin
from Tamilakam,6 to compose the first great poem in a new, courtly Telugu
– a classicized register of the language spoken in Vengi and its environs.
Narendra had a simple request for Nannayya, which also reveals to us how
seriously kings took the claims of divine descent that their ancestors
included in their prashastis, and the complex motivations which led them to
patronize poets:
 My lineage begins with the moon, and then proceeds through Puru, Bharata, Kuru, and King
Pandu. The stories of Pandu’s famous sons [the Pandavas], virtuous and beyond blame, are ever
close to my heart … My mind inclines day and night to those stories. With all your learning,
please compose in Telugu a book that makes clear … the proven meaning bound to the
Mahabharata text.7
The Mahabharata, one of the longest poems ever composed, was now
slowly reborn in Telugu, in what Professors V.N. Rao and David Shulman
call a ‘lyrical, laconic and precise’ form ‘combining long Sanskrit
compounds and Dravidian-based Telugu words and adapting a variety of
Sanskrit and regional meters’.8 It was a distant parallel to the birth of
courtly Kannada poetry under the rigorous Sanskrit-inspired grammars of
Amoghavarsha’s court. Nannayya would never finish this great
undertaking: that was a task left to the generations of master poets who
came after him. But courtly Telugu, which would one day grow in
splendour to the point where it practically eclipsed the literature of the
Kannada and Tamil regions that had so cruelly attacked and exploited the
fertile coast, had been born. It was the founding moment of a glorious
literary tradition. Courtly Telugu would reach its apogee in sixteenth-
century Vijayanagara, the mighty empire of the southern Deccan, and
continue to flourish in the Qutb Shahi court of Golconda and the Nayaka
states of Tamil Nadu in the seventeenth and eighteenth centuries.9
Returning to the eleventh century: as poetry flourished in courts across the
subcontinent, in 1055, the greatest of its royal patrons – Bhoja Paramara –
died under a cloud of shame and sorrow. The brilliant old king had never
recovered from Someshvara’s 1046 sack of Dhara. His beloved capital was
sacked again soon after by the rising power of the Chaulukyas of
neighbouring Gujarat (not to be confused with the Chalukyas of Kalyana),
and then the Kalachuris of central India. A later story tells us that Bhoja,
attacked by these two kings, ‘lost his pride, as a snake, overcome with a
charm, loses its poison’.10 The Paramaras fell into anarchy and dynastic
squabbling: one candidate sought the help of his family’s enemies, the
Kalachuris and the Kalyana Chalukyas.11 The Vallabha Someshvara seems
to have realized this could potentially help settle his northern frontier. His
second son, Prince Vikramaditya (who, as noted earlier, was already
searching for allies of his own) was sent to ensure that the ambitious
Paramara prince succeeded to Bhoja’s vacant throne.12
Bhoja had been the most extraordinary of the medieval Paramara line of
poet-kings. Despite the brilliance and tragedy of his life and almost total
absence from narratives of world history, Bhoja finally did become
immortal in Indian storytelling and literature: he would be remembered for
centuries after as the ‘ultimate arbiter of grammatical correctness, rhetorical
propriety, and literary good taste’,13 rewarding courtiers and litterateurs for
witty or unique turns of phrase.
By 1061, Prince Vikramaditya had returned from Dhara to the south,
where the Cholas were once again seeking to wrest back the Raichur doab.
Rajadhiraja’s brutal death on the battlefield was hardly conducive to future
peace, and the Cholas sought revenge for this humiliation. The Tamil
kingdom was also facing some political churn. Rajendra II’s succession to
the throne in a moment of deadly crisis had not been disputed by his elder
brother Rajadhiraja’s sons. However, he now appointed his son
Rajamahendra as crown prince, sidelining the claims of his ambitious
younger brother – the youngest of Rajendra I’s three sons, Vira-Rajendra.
To help secure the heir’s future, Rajiga Vengi Chalukya, the son of their
sister Ammangai and Rajaraja-Narendra of Vengi, was married to Rajendra
II’s daughter Madhurantaki (Rajiga’s first cousin). The emperor also seems
to have promised that Rajiga would be seated on his father’s throne, and
that Vengi would be freed of the influence of the Kalyana Chalukyas.
But Prince Vikramaditya of the Kalyana Chalukyas moved on the Cholas
before they had a chance to attack – apparently while they were still
gathering their forces in the former Ganga territories in the southern
Deccan. Vikramaditya’s attempt was not successful. Indeed, the new Chola
crown prince, Rajamahendra, dramatically claimed to have forced the
Chalukyas back with a single war elephant at an inconclusive battle in 1061
at Kudala-Sangama,14 where the Tunga and Bhadra rivers merge to become
the Tungabhadra.
But before the war could unfold any further, all plots and plans and
preparations were thrown out of the window by a new crisis. The fuse was
lit in Vengi, where, in 1061, Rajaraja-Narendra, part Chola, part Chalukya,
died. The first on the scene was his half-brother and deadliest rival,
Vijayaditya VII, who entered Vengi, at last, in triumph. The goal that had
driven him for nearly half a century was at last achieved. It seems that by
this point Someshvara Chalukya had modified his plans for the kingdom
and agreed to Vijayaditya’s reclamation of his ancestral throne. However,
Vijayaditya would not forget his old ally’s earlier betrayal so easily, as we
shall see.
Vijayaditya VII, his life’s ambition apparently fulfilled, now crowned his
son as the new king of Vengi. But the Chola royal family had already
realized what was happening and were moving to turn the situation to their
advantage. Rajiga Vengi Chalukya, son of the dead Narendra, nephew and
son-in-law of Chola emperors, had just as strong a claim to Vengi as his
paternal uncle Vijayaditya VII. An army was dispatched under the
command of Vira-Rajendra Chola, younger brother of the emperor Rajendra
II.
In 1062, the Chalukya and Chola armies, with all their pent-up enmity and
hatred, clashed in battle. Vijayaditya’s son, the new king of Vengi, was
killed, though his side apparently managed to drag his body away before the
Cholas could seize it. Another senior Chalukya commander who was killed
was not so lucky: as with Rajadhiraja, his body was beheaded. The
commander’s daughter, ‘who resembled a peacock in beauty’, was
captured, and the elderly Chola prince Vira-Rajendra cruelly ordered her
nose to be cut off.15 It was a chilling reminder of how women’s bodies were
the most frequent targets of aristocrats who wished to humiliate and
underline their dominance over their enemies.16
Vijayaditya was devastated.17 According to his inscriptions, it was not
until ‘hereditary well-wishers’18 managed to get through to him that he
agreed to seat himself on the sorrowful throne of Vengi. The twisted
humour with which Fortune had seen it fit to grant his dearest wish, only to
tear it away and return it to him, seems particularly cruel.
A year later, in 1063, the Chola emperor Rajendra II died. Rather than
continue to conduct the war in Vengi, his younger brother Vira-Rajendra
returned to Gangai-Konda-Chola-Puram, hell-bent on seizing the throne.
Rajendra II’s son, the crown prince Rajamahendra, disappeared in
somewhat murky circumstances. It is not entirely clear whether or not he
predeceased his father. Vira-Rajendra had abandoned Vengi to Vijayaditya
VII. The claims of his nephew Rajiga were not to be enforced. Rajiga had
lost his patrimony.19 This young man’s allies and his branch of the Chola–
Chalukya family now began to cast about for opportunities of their own.
The new Chola emperor Vira-Rajendra, meanwhile, had much bigger
challenges facing him. A half-century of incessant war, the loss of crucial
spheres of influence to the Kalyana Chalukyas, and the growing power and
autonomy of Tamil lords, merchants, cities and chiefs needed to be urgently
addressed. The military successes that had fuelled the careers of the last
four Chola emperors were increasingly untenable in the face of constant
rebellions, dwindling military resources and agricultural surplus, and the
ascendant power of the Chalukyas.
Sitting on his throne in Gangai-Konda-Chola-Puram, twenty years after
his father Rajendra Chola’s death, Vira-Rajendra had neither the resources
nor the power with which his predecessors had pursued their ambitions.
Continuous rebellions in Lanka and requests for intervention in the politics
of the Srivijaya federation20 – which Prince Rajiga was dispatched to attend
to with a small force, c. 1068 – were straining his resources enormously.
More importantly, he was already advanced in age, and was determined to
ensure that his beloved son Adhirajendra would be able to succeed him and
maintain Chola power. He would need to do something radical, and he was
not the only south Indian ruler who faced this imperative. Vijayaditya VII
of Vengi, despite the tragic loss of his son, was once again back to his
political manoeuvring, alongside his partner-in-arms, Prince Vikramaditya
of the Kalyana Chalukyas.
By 1067, the Kalyana Chalukya Vallabha Someshvara was seized by a
mysterious disease, a ‘malignant fever’ according to a Chalukya poet.21
Priests and physicians were anxiously attempting to cure him. His heir,
Crown Prince Someshvara, was increasingly taking charge of affairs in
Kalyana, while his most capable son, Prince Vikramaditya was elsewhere,
busy with his own machinations. Under these circumstances, a Chola raid
led to the capture of some Chalukya officers, whose heads were nailed to
the walls of Gangai-Konda-Chola-Puram.22 A ritual humiliation was also
performed, whereby a Chalukya diplomat was garlanded with a banner
declaring that Someshvara was a coward23 (which appears to have been an
occupational hazard for any Chalukya ambassador to the Cholas). Around
this poorly attested incident, the Chola court poets fabricated a feverish
fantasy claiming that the humiliated Someshvara had demanded a rematch
of the battle of Kudala-Sangama, providing a date and place for the contest;
apparently Vira-Rajendra adhered to it, but Someshvara, that ‘liar’, did not
show up and instead ‘ran away until his legs became sore, and hid himself
in the western ocean’, leaving the Chola emperor to conquer and burn the
entire Deccan!24
Nothing of the sort had happened. This was all for show, for the
amusement of the crowds of his subjects, demanding war and loot and
entertainment. Vira-Rajendra had subtler plans in motion.
Around 1067, as his old ally Someshvara I of the Kalyana Chalukyas lay
dying, Vijayaditya VII of Vengi, unseated and humiliated by two
generations of Chola emperors, who had lost his own son to Vira-
Rajendra’s armies, humbly submitted to the Chola emperor. He received in
return a position as an honoured vassal.25 The Chola emperor now declared
that his dead brother Rajendra II’s vow regarding Vengi had been fulfilled.26
It was yet another blow for the young Rajiga Vengi Chalukya,27 who must
have been infuriated. However, apparently learning from his paternal uncle
Vijayaditya VII that patience might eventually bring him what he sought, he
contented himself for now with the position of a minor lord in Chhattisgarh,
protecting Chola control over Vengi. Here he butted heads with Prince
Vikramaditya28 of the Kalyana Chalukyas, the beginning of a lifelong
enmity. In return for his apparently calm resignation to his fate, Rajiga
Vengi Chalukya, this ‘seemingly loyal and indebted near-relation’, was also
‘compensated by being allowed to exercise a certain authority … in the area
linking his two dynastic heartlands’, the ancient Pallava domains around
Kanchi,29 between Vengi and the Chola core territories. However, as the
scholar Whitney Cox puts it, Vira-Rajendra had fatally underestimated this
young man’s own ambitions and abilities.30
As much as he might have been frustrated with his new-found struggles
with Rajiga Vengi Chalukya, Prince Vikramaditya of the Kalyana
Chalukyas was probably not too discomfited by Vijayaditya VII’s betrayal.
He may, in fact, have been responsible for it in the first place. By
facilitating Chola control over Vengi, he hoped to gain Vira-Rajendra’s
support for his own bid for the Chalukya throne. By bringing together the
Kalyana Chalukyas and the Cholas, Vikramaditya could finally end the
perpetually escalating conflict that was consuming both families.31

At this point, it is worth discussing what sort of man this Prince

Vikramaditya of the Kalyana Chalukyas really was. Of course, as with so
many things in medieval India, all we have are snapshots of the man’s
furiously active life. Uniquely, we also have access to a mahakavya called
the Vikramankadevacharitam, an epic poem written in Sanskrit by Bilhana,
a Kashmiri poet who joined the court of Kalyana, rewriting Vikramaditya’s
personal history and depicting him as an ideal king, another Rama. In truth,
Vikramaditya, for all his undoubted capabilities, was far removed from that
god-king of legend.
 This prince was born in the generation after the destruction of
Manyakheta by the Cholas32 – which made him and his siblings among the
first generation of truly ‘Kalyana’ Chalukyas, some of the first members of
the dynasty to see that city as their home. Growing up in the palace at
Kalyana, he had been surrounded by the women of the Inner City. He
received instruction from senior Brahmins while his father and grandfather
– Someshvara and Jayasimha II – were constantly away battling the many
deadly rivals of the Chalukyas. Vikramaditya’s court poet claimed that he
showed signs of bravery and military potential very early on: ‘He loved to
chase the royal swans and to tease the lion-whelps in their cages.’33
(Whether true or not, this interesting nugget hints to us that perhaps the
Chalukya palace at Kalyana maintained a zoo or menagerie.)
Vikramaditya learned the skills of sexual pleasure and connoisseurship of
art, poetry, architecture and adornment; he learned to fight with dagger,
sword, shield and spear, exhibiting his abilities before the court in an arena
attached to the palace.34 He went about in expensive patterned silks, his hair
oiled, shoulders and chest marked with moon-shaped symbols of sandal
paste, neck adorned with expensive jewellery.35 He personally knew the
Chalukyas’ great generals and vassal kings, who visited the city on
ceremonial occasions to pay obeisance to his father and grandfather, and
exchanged gifts and refined conversation with them in the salons of
Kalyana. It did not take long for a faction to form around him, supporting
his claim to the throne.
Vikramaditya’s father Someshvara, who had become Fortune’s Favourite
in 1042, could do little about this sort of politicking. His sons had been
lectured in filial piety from legends and shastra texts, but that artifice was
no guarantee of loyalty to their father. Someshvara nevertheless allowed
Vikramaditya a position of prominence: he needed his son’s talents
desperately to meet the challenges the Deccan was facing from all
directions. The prince was married to Chandralekha, the daughter of an
important Konkan family, probably soon after Someshvara’s campaign
there in 1046–47. As has been mentioned earlier, soon after, Vikramaditya
was commanding armies against the Cholas, and served as his father’s
representative in Malwa in 1055. By this point, Vikramaditya’s elder
brother Someshvara already appears to have been the preferred candidate
for succession, despite his relative lack of achievements. Having played an
important role in the installation of the ruler of Malwa, Vikramaditya
returned to the south by 1061, where he fought the Cholas at Kudala-
Sangama, before the death of Rajaraja-Narendra in Vengi upended the
political equations of south India. He then moved to Chhattisgarh, where,
alongside the forces of the ambitious and ever-eager Hoysala chiefs of
southern Karnataka, he battled Rajiga Vengi Chalukya for control over the
strategic fort of Chakrakuta. All these activities indicate that Vikramaditya
had diplomatic and military abilities of a very high order. He was, therefore,
understandably averse to letting his elder brother succeed their father to the
throne of Kalyana.
By 1067, as Someshvara Kalyana Chalukya, Great King of Kings,
Fortune’s Favourite, lay on his deathbed, Vikramaditya was in the province
of Banavasi – the ancient capital of the Kadamba dynasty, where his
supposed ancestor Pulakeshin II had first made his name by sacking their
capital. Here, he and his younger brother Jayasimha plotted treason against
their father and elder brother. The Kadamba lords of Goa as well as the
Alupas, a lordly dynasty of southern Karnataka, swore their loyalty to
Vikramaditya. They seem to have arranged for a direct meeting between
him and the Chola emperor36 Vira-Rajendra. Supposedly, Vikramaditya
shared a seat with this deadly enemy of his family, exchanging vows of
friendship. Vira-Rajendra appears to have promised to help Vikramaditya
seize the throne of Kalyana, in return for Vengi remaining under Chola
control.
Soon after, like his long-forgotten namesake, the Vatapi Chalukya
emperor Vikramaditya II, this ambitious new Vikramaditya also entered
Kanchi. But instead of the Pallava royal shrine, it was a palace that he
visited, and instead of being preceded by an invading army, he was
preceded by the beating drums of a wedding party. Vikramaditya Kalyana
Chalukya married the daughter of Vira-Rajendra Chola,37 thus declaring that
he would support Vira-Rajendra’s heir Adhirajendra in return for the Chola
emperor’s help in seizing Kalyana.
Someshvara, in a delirious fever, may not have known that any of this was
happening. Or perhaps he saw the ominous machinations taking place as
clear as day. We will never know. But in 1068, as it became increasingly
clear that he would not recover from his illness, the emperor Someshvara
informed the court that his career was at an end.
The 29th of March 1068 was an important and terrible day in the history
of the Deccan.38 That morning, the cool waters of the Tungabhadra swirled
around Someshvara Kalyana Chalukya’s feet, as his court stood around him
solemnly.39 He had already said his last words, his farewells.40
 Someshvara stepped forward. Eddies swirled around his calves, his knees,
his chest, his neck and finally the Vallabha submerged himself. A Chalukya
poet writes a few decades later: ‘…and so it was then that, to the crashing
accompaniment/of the river’s surging wave/he went to the city of moon-
crested Shiva.’41
Thus ended the life of the Someshvara Kalyana Chalukya, Fortune’s
Favourite, Earth’s Beloved, Great King of Kings.

On 11 April 1068, less than two weeks after their father’s suicide,
Vikramaditya’s elder brother rose to the throne of Kalyana as Fortune’s
Favourite, Someshvara II ‘Bhuvanaikamalla’, Sole Wrestler of the Earth – a
significant title, hinting at the fact that this unfortunate prince understood
his younger brothers’ intrigues very well. He was, nevertheless, forced to
grant them formal recognition as the ‘governors’ of various provinces in
southern Karnataka. He surrounded himself with loyal generals, hoping to
strengthen his grip on power and eventually humble his treacherous
siblings.
Meanwhile, events were working out in Vikramaditya’s favour. His Chola
father-in-law, Vira-Rajendra, mounted a limited invasion of the Deccan,
subduing a cursory attempt at defence by Someshvara II. Vikramaditya then
declared himself the sovereign ruler of Gangavadi, southern Karnataka,
along the Chola border,42 and adopted imperial titles. As Vikramaditya and
Vira-Rajendra now gathered their forces, it would seem that a Chola ruler
was on the verge of a full-scale intervention in a Chalukya succession
crisis.43
And then, in 1070, Vira-Rajendra suddenly fell ill and died. Vikramaditya
was stymied. Had the old Chola emperor lived a little longer and guaranteed
his accession to the imperial throne of the Deccan, Vikramaditya had
planned to ensure the succession to the Chola throne of his brother-in-law,
Vira-Rajendra’s son Adhirajendra. But now, he had other matters to attend
to: there was little he could do but wait and attempt to overthrow his brother
Someshvara II by himself. He could only hope that his brother-in-law was
capable of establishing himself as the next Chola emperor in his own right.
Someshvara II, meanwhile, was encouraged by the news of Vira-Rajendra’s
death. That same year, 1070, he gathered his armies together and mounted
an expedition to Malwa, where his brother Vikramaditya had so carefully
appointed a loyal successor to Bhoja in 1055–56, and dethroned him.
Meanwhile, the political situation in the Tamil country had become far
more complicated after Vira-Rajendra’s death. His son, Adhirajendra Chola,
had no martial accomplishments to his name – especially in comparison to a
new claimant to the Chola throne, who had emerged almost out of the blue.
This was none other than Vikramaditya Kalyana Chalukya’s rival, Rajiga
Vengi Chalukya, ruler of the territories around Kanchi, now calling himself
Rajendra Chola. In his inscriptions and public documents, this Rajiga-
Rajendra’s court masterfully recounted his successes in Chhattisgarh, where
he had supposedly distinguished himself against Vikramaditya Chalukya
with behaviour becoming of a king.44 In a manner unprecedented for any
Chola before him, his inscriptions drew on the mythology of the Puranas to
emphasize his descent from both the Chola and Chalukya families, the
mythical dynasties of the Sun and the Moon supposedly united in his
blood.45 The Chalukya boar, which generations of Tamils had excoriated,
was now presented as the gentle restorer of Earth, analogous to this new
‘Rajendra’ himself.46
His rival, Vikramaditya’s young brother-in-law Adhirajendra, had ‘his
father’s courtiers and the resources of the Kāveri delta nominally at his
command’.47 Perhaps, given time, he could have emphasized his legitimacy
as a patrilineal descendant of the great conqueror Rajendra I Chola and
eventually unseated this Rajiga-Rajendra, his Chalukya–Chola cousin. But
it was not to be: in late 1071, the young king followed his father to the
grave, possibly dying of gangrene.48
The courtiers of Rajiga-Rajendra now triumphantly fanned out across the
Pallava heartland, which the Cholas had, up to this point, failed to truly
integrate into their court society, confirming the privileges of magnates and
agrarian assemblies. As Whitney Cox puts it, a new, decentralized network
was being born in the Tamil country, a new realpolitik was at play, there
was a new series of political equations.49 By 1074, Rajiga-Rajendra, who
could, in the male line, draw his ancestry hundreds of years back to
Pulakeshin I, the chieftain-turned-king who performed the horse sacrifice in
a distant time and place, rose to the glittering throne of the City of the
Chola Who Conquered the Ganga. He adopted the title of Kulottunga
Chola, Pinnacle of His Race. Henceforth he will be referred to as Rajiga-
Kulottunga in this book.
With his brother Someshvara II now far more secure on his throne to his
north and his rival Rajiga-Kulottunga established to his south, Vikramaditya
Kalyana Chalukya was under enormous pressure. Luckily for him, his
successful military career seems to have convinced most of the Deccan that
he was a sounder choice as emperor than his brother was, and his superb
self-confidence,50 at least in public, had already led to a virtual partition of
the Chalukya political network.51 Someshvara II stuck close to Kalyana,
where his father’s generals still remained loyal to him, and toured the
Deccan making his own land grants and attempting to rally the support of
new power centres.52 But he was running out of time. South of the
Tungabhadra, Vikramaditya had adopted the imperial title of
Tribhuvanamalla, Wrestler of the Three Worlds.53 Vassal kings from the
Konkan to the Narmada had taken sides in a civil war which Vikramaditya
was sure to win: the only question was when it would begin.
In 1075, Rajiga-Kulottunga Chola finally asserted his claims to Vengi
with the entire resources of the Chola imperial network at his disposal. He
was not content with leaving his paternal uncle Vijayaditya VII in control,
given their personal history. The Kalyana Chalukyas were too divided to
contest this move, which would finally grant the Cholas undisputed control
over this fertile territory. Vijayaditya, his machinations and betrayals over
so many decades having come to naught, fled Vengi and joined
Vikramaditya Chalukya in Gangavadi.54 The next year, Someshvara II
attempted to ally with Rajiga-Kulottunga, who sent an expedition to fight
his old rival Vikramaditya, who managed to fend it off. Claiming
(apparently with no trace of irony) that his elder brother had betrayed the
Chalukya family by allying with their ancestral rival,55 he then advanced to
Kalyana, his forces swelling larger and larger with the armies of his
adherents.
After some desultory resistance, the elder brother that Vikramaditya had
perhaps hated for being his father’s favourite despite all his hard work over
the years, was captured. Like so many others who had attempted to seize
the glittering, bloody throne of the Deccan for themselves over the centuries
– Mangalesha Chalukya, the courtesan queen Vijaya-Mahadevi, the
Rashtrakuta kings Danti-Durga and Govinda II, the Paramara ruler Vakpati
– Someshvara Chalukya’s eldest son vanishes from history.
The father’s favourite was dead.56 Long live Fortune’s Favourite,
Vikramaditya VI.

Vikramaditya VI would reign for fifty years, casting himself as a millennial

sovereign, the greatest of all Deccan emperors. Coming to the throne in
roughly the year 999 of the Shaka era (1076 CE), one of the most
widespread year-counting systems in the Indian subcontinent, he would
inaugurate a new era, the Vikrama, used in Chalukya inscriptions through
his rule and after.57 He would reign over the vast, warlike and increasingly
prosperous region with the same furious energy he had displayed as a
prince, diplomat and general under his father: his inscriptions are found in
Kalyana, Banavasi, Ponuguppe, Etagiri, Pottalakere, Appayanadakuppa,
Ballakunde, Manyakere, Kollipake, Jananathapuram and Bijapur58 – a wide
swathe of territory extending through the heart of the Deccan, from present-
day north Karnataka to the southwest. His feudatories are known from
Nagpur in the north to Gangavadi in south Karnataka and Draksharama in
coastal Andhra,59 though the degree of allegiance to the Chalukya emperor
varied across this vast landmass. Following in his father Someshvara I’s
footsteps, he ensured the production of large amounts of Sanskrit
documents and texts through his reign, contributing significantly to
Kalyana’s status, prestige and legend. ‘Subsequent generations, in particular
those of the sixteenth century, would recall his long and prosperous reign
with admiration and view the Chalukya house generally as having
embodied the epitome of imperial glory and moral righteousness.’60

Under Vikramaditya VI, the glamorous court of Kalyana attracted a

number of talented poets and religious leaders – including, as mentioned
before, Bilhana, a Brahmin who emigrated from distant Kashmir and
composed the aforementioned Vikramankadevacharitam, a magnificent
Sanskrit text rewriting Vikramaditya’s personal history as that of an ideal
prince and son. (Despite such claims, Vikramaditya does not seem to have
been a very successful family man: his younger brother Jayasimha rose
against him just as he himself had risen against their elder brother
Someshvara, but unsucessfully.) Other supremely influential texts
composed in the twelfth-century Chalukya court include the Mitakshara, a
concise distillation of an older Brahminical law book. This was the only
Dharmashastra text to ever be translated into Persian, as well as vernacular
languages such as Tamil and Telugu.61 Another example is the Manasollasa,
a text that is part encyclopaedia and part ‘Mirror for Princes’ (dedicated to
the cultivation and enjoyment of courtly pleasures), attributed to
Vikramaditya’s son and successor Someshvara III. But even as the
Chalukya court moved around the heart of the Deccan, setting itself above
its subjects with the conspicuous consumption of art and poetry, elephant
races, wrestling matches, hunts and the like,62 the elites of the Deccan
continued to grow increasingly affluent and powerful. A number of
influential Brahmin families served the Chalukyas across generations as
ministers and generals,63 while the royal feudatories of the Chalukyas grew
more powerful in their territories. This would cause a disaster late in
Vikramaditya’s reign, something we will return to shortly.
Vikramaditya may have dreamed of dominating the Chola country and
bringing to an end the wars that had devastated both empires. The accession
of the capable Rajiga-Kulottunga to the Chola throne had prevented that,
but Rajiga-Kulottunga was a very different man from his predecessors.
During his competition with his cousin Adhirajendra before his accession,
Sri Lanka had finally evicted Chola power, c. 1070–73,64 and he made no
serious attempt to reconquer it. Instead, Rajiga-Kulottunga’s former base,
the old Pallava heartland, was more deeply integrated into the Chola state.
Vengi became a Chola viceroyalty, a proving ground for his sons, and here
he would go by the title of Vishnuvardhana65 – that ancient name once
borne by Pulakeshin II’s hunchback brother Kubja-Vishnu-Vardhana, the
first Chalukya king of Vengi. North of Vengi, near the lands of modern-day
Odisha, Rajiga bound himself to local dynasties through marriage and
shows of military might.66 But the terrible Deccan and oceanic expeditions
of previous Cholas were not repeated during his reign. Through
considerable political pragmatism, personal charm and deep engagement
with varied local power centres, Rajiga-Kulottunga successfully (and
perhaps more realistically) recast the Chola empire as a ‘purely Tamil
power’, as Professor K.A. Nilakanta Sastri puts it.67
Rajiga-Kulottunga and his great rival Vikramaditya VI might not have
fought as relentlessly as their predecessors, but their personal enmity was
never truly forgotten. As the Chola aged, the Chalukya emperor
successfully conquered swathes of northern Vengi; he may have done more
if not for the fact that an uprising by a coalition of his vassals – led by the
increasingly powerful Hoysalas of southern Karnataka, in alliance with the
Kadambas of Goa – shook his throne to its foundations.68 These dynasties
had, ironically, served Vikramaditya closely in his early career and in his
rise to the throne. Indeed, Gangavadi, the core territory of the Hoysalas,
was Vikramaditya’s temporary base before he overthrew his brother
Someshvara II. All this lays bare the ruthlessness of power-obsessed
medieval aristocrats.
Though this uprising was savagely defeated and Gopakapattinam, the
Kadamba capital, sacked, the Hoysalas remained unbroken. Their king,
Hoysala Vishnuvardhana, would successfully drive the Cholas out of the
former Ganga kingdom in southern Karnataka and seize the ancient Ganga
capital of Talakkad for himself by 1118.69 Indeed, Professor Nilakanta
Sastri suggests that the Hoysalas raided parts of Tamilakam, broke open
temples and seized images within.70 Meanwhile, Vishnuvardhana embarked
on an ambitious programme of temple building, agricultural expansion and
political consolidation, repeating the pattern of medieval Indian kingship
that we have seen the emperors of the Deccan deploy before in northern
Karnataka. After Vikramaditya VI’s death in 1126, Hoysala
Vishnuvardhana would adopt imperial titles. He and his successors, as well
as other former vassals of the Kalyana Chalukyas such as the Seuna
Yadavas and Kakatiyas, would reduce both the Chalukyas and the Cholas to
insignificance.
With Kalyana unable to enforce any sort of control over them, the
Hoysalas set out to create an empire of their own, conquering parts of the
upper Kaveri valley from the Cholas. Capitalizing on this, the Pandya
dynasty, centred around Madurai, rose up and sacked Rajendra Chola’s
great capital of Gangai-Konda-Chola-Puram,71 bringing Chola power to an
end by the thirteenth century.72

This might seem like a repetition of the same endless pattern of dynastic
rise and fall that we have seen so often through the course of five hundred
years. But when we look at the symbolic texture and political geography of
these events, we catch a glimpse of something grander. In the beginning of
this book we saw how the Chalukyas of Vatapi had been cultivators or
pastoralists. Out of nothing, out of a political vacuum, they built an empire
in northern Karnataka. We take terms like ‘king’, ‘empire’, ‘aristocrat’,
‘governor’, ‘vassal’, ‘monastery’, ‘temple’, ‘patronage’ and so on for
granted today. But the Chalukyas had applied these concepts to
geographical and political spheres that had never seen these things. They
had created the first Deccan empire, based in north Karnataka, by adapting
to this region the Sanskritic imperial rituals and systems that had originated
elsewhere, but now became the framework of a uniquely Deccan polity.
We saw this imperial system grow to a vast scale in the centuries that
followed. It fused with the ancient agrarian centres of the Godavari river
basin in Maharashtra, and the flourishing trade ports of the west coast, and
reached an apogee under the Rashtrakutas. Courtly Kannada then added
further depth and reach to the Deccan imperial system, as the Rashtrakutas
began to use it as a courtly language of power adopted by the warlike
aristocrats of the Deccan. After the Rashtrakuta collapse, we saw the
Chalukyas of Kalyana lead the region to a remarkable period of cultural
confidence and military dominance, using both Kannada and Sanskrit in a
variety of inscriptions. The Kalyana Chalukyas formulated new hierarchies
of power as their rulers moved across a vast geopolitical chessboard to fight
off another equally ambitious, powerful and transregional imperial Indian
polity: the Chola empire. All through this, the Deccan’s connections to the
trading world of the Indian Ocean continued to expand and thrive.
And now we see the logical continuation of these patterns. Within the
broader context of the Indian subcontinent, the Chalukyas had created a
framework of imperial power that expanded in both breadth and depth over
centuries through the Deccan. They had created the architecture of a
Deccan empire, the vocabulary of political and religious power, the
backbone of agrarian control, the networks of commercial production and
interaction. And now, within the context of the Deccan, the feudatories,
rivals and sub-imperial dynasties of the Chalukyas would re-adapt these
frameworks of imperial power to their own territories. What the Chalukyas
had done in northern Karnataka, their vassals would do in Telangana,
southern Karnataka and Maharashtra. These regions had ascended to royal
status and begun to participate in pan-Indian networks of knowledge and
exchange as feudatories of the Vallabhas; now, as their ruling dynasties
challenged the Vallabhas, they would become new imperial centres in the
Deccan’s micro regions.
Urbanization, military expansion, temple building, artistic and literary
innovation in regional languages: the Chalukyas and Rashtrakutas had
established an exemplar of successful southern Indian kingship and a
repertoire of cultural and material paraphernalia to build and negotiate ties
with a dizzying array of potential power centres. This template would be
imitated by many of the dynasties we have seen in previous chapters to
successfully consolidate their regions under new imperial formations. The
Seuna Yadavas, who had assisted the Vallabha Taila in defeating the
Paramara king Vakpati in 994, gradually transitioned from Kannada-
speaking warlords to powerful patrons of Marathi and Sanskrit, ruling over
much of Maharashtra by the thirteenth century, centred around the citadel of
Devagiri. The Hoysalas, whom we have seen serve the Vallabha
Someshvara I in campaigns in Malwa and fight alongside Prince
Vikramaditya in Chakrakuta in Chhattisgarh, established a mighty kingdom
in southern Karnataka, stretching at times deep into the Kaveri river valley.
The Kakatiyas, who had fought alongside the forces of Someshvara I at
Kanchi in 1047–48, created a powerful state in Telangana, using many of
the concepts and methods of the Chalukya kings while establishing their
independence in the late twelfth century. The Kakatiyas would also
successfully expand into Vengi, devastated over the centuries by the
depredations of Kannada- and Tamil-speaking polities, and thus create the
first empire to rule all of the Telugu-speaking lands together. This initiated a
vast movement of Telugu peasant-warriors from the fertile coast into the
dry upland Deccan and through much of southern India in the centuries to
come.
These three dynasties – the Kakatiyas, Seuna Yadavas and Hoysalas –
would fight many savage wars with each other over the supreme
overlordship of the Deccan. They would also compete architecturally, as
previous Deccan dynasties had: the Kakatiya Ramappa temple at Palampet
(one of India’s newest UNESCO World Heritage Sites), the Hoysala
Vijayanarayana/Chennakeshava temple at Belur, and the Seuna Yadava
Gondeshvara temple at Sinnar were descended from the stellate and
staggered-square architectural forms established by patrons and sthapatis in
the eleventh-century Chalukya imperium.73 As these new regional dynamics
emerged, Kalyana, the imperial centre which once dominated the Kannada,
Marathi and Telugu lands, began to fade. The once-glittering Chalukya
Vallabhas of Kalyana were overthrown by a Kalachuri dynast called Bijjala
in the twelfth century, bringing to a rather unceremonious end the name that
had awed the Deccan on and off for half a millennium.74
 But beyond this political turmoil, the Chalukyas had already set in motion
religious and political trends of extreme importance. The constant
expansion of elite temples, the increasing concentration of power and
wealth among the upper-caste aristocracy, and the relatively low status of
women and of the bevy of other castes and professions led to the emergence
of a new bhakti movement in the late twelfth century: the Virashaivism.
These Heroic Shaivas, as the name suggests, bravely defied the hierarchy of
political and religious power upon which the imperial structure of the
Deccan depended. Singing of the accessibility of Shiva to all in a vast hall
in Kalyana called the Anubhava Mandapa, they rejected caste and gender
discrimination, the authority of the Vedas, and even the idea that temples
could lead to an experience of Shiva. Their poetry also reveals a deep
discontent against the wealthy landed aristocrats who were so important to
the Chalukya court.
The Virashaivas’ activities were actively encouraged by the religious
leader Basavanna, a Kalachuri minister,75 and popularized by hundreds of
remarkable male and female poets, from assorted social backgrounds,
singing in Kannada. This group of people includes the relatively well-
known Akka Mahadevi, who according to legend was married to a minor
king; Allama Prabhu, who may have been a temple musician; Basavanna,
who was a Brahmin; cobblers such as Madara Channayya, cowherds such
as Siddharama, a sex worker by the name of Sankavve, and a thread maker,
Remmavve, among many others.76 Their activities, over just a few decades,
shook the balance of power at Kalyana to its foundations. According to the
Basava Purana, composed about a century after these events, this led to a
violent response from the lords of Kalyana that seems to have escalated into
widespread destruction, the assassination of Bijjala Kalachuri, the
annihilation of much of the great city of Kalyana, and the scattering of the
Virashaivas across the Deccan.77 Kettittu kalyana – ‘Kalyana is wrecked’ –
would become a common phrase in Kannada literature and folk narratives
in the centuries after.78 There is a strange romance in devotees of Shiva –
who, in the eighth century, had been crucial to renewing the power of the
Vatapi Chalukyas – bringing to an end the last remnant of Kalyana
Chalukya power in the twelfth century. Today, a town at the site of Kalyana
is known as Basavakalyan, after Basavanna, considered one of the most
important saints of the Virashaiva and closely related Lingayat traditions.
Kannada literature and poetry would be profoundly shaped by the poetry of
the Virashaivas. Indeed, Professor H.S. Shivaprakash argues that in terms of
literature, the Virashaiva movement and its successors were more
influential than the courtly Kannada epics composed by powerful Deccan
dynasties up to this point.79
Even grander changes were afoot in the rest of the world. The rest of the
subcontinent had seen similar trends as the Deccan, with new polities
forming in Madhya Pradesh, the northwest and the Gangetic plains, such as
the Chandellas (known for their construction of the temples of Khajuraho)
and the Chahamanas (more famously remembered as the Chauhans). From
the tenth century onwards, these polities had to contend with waves of
Central Asian nomads who began to expand into South Asia as well as West
and East Asia. By the late twelfth century, one of these waves – led by
Turkic warriors – had overrun a number of north Indian dynasties and
founded the Sultanate of Delhi, just decades after the collapse of Kalyana in
the Deccan. Partly in order to fund wars against other Central Asian peoples
such as the Mongols, the armies of Delhi would lead the collective military
might of much of northern India to raid the Deccan in the late thirteenth
century. By the early fourteenth century, the Hoysalas, Seuna Yadavas and
Kakatiyas had been uprooted by Delhi’s armies, and the Sultanate
established garrison towns across the Deccan.

Attempts to integrate the Deccan into a north India–centric imperial

network failed spectacularly, however. From the remnants of the Chalukya
successor states, radically new south Indian polities, deeply integrated into
the broader ‘Persianate’ world, emerged. North of the Krishna river, a
powerful new kingdom, the Bahmani Sultanate based in Gulbarga in the
fifteenth century, had given way to five Deccan Sultanates of varying size
and power by the sixteenth century. South of the Tungabhadra, the mighty
empire of Vijayanagara dominated southern Karnataka and the Tamil
country that had defied so many Deccan emperors.
Vijayanagara and the Deccan Sultanates, especially the Sultanate of
Bijapur (centred in a former Chalukya royal city), were among the most
populous and vibrant polities of the sixteenth-century world. Both states
substantially expanded the Deccan’s integration into global networks of
trade, a process the Vallabhas encouraged, as we have seen. But they would
seek to attract not just goods but also people from the wider world,
especially Persians and Turks, and even the Portuguese. The innovative
nature of these polities is indicated in a letter written by Alfonso de
Albuquerque, Portuguese viceroy of Goa – based, incidentally, in the city
that was once Gopakapattinam, the capital of the Kadambas of Goa – to
King Manuel I of Portugal. According to Albuquerque, Bijapuri gunsmiths
from Goa were as skilled as those of Germany.80 Vijayanagara, meanwhile,
dedicated enormous resources to obtaining horses from across the Indian
Ocean,81 resurrecting the ancient geopolitical strategies of the Rashtrakuta
kings.
 Both Vijayanagara and Bijapur also competed to claim the imperial
mantle of the Chalukyas of Kalyana. Elements from Chalukya temples
south of the Krishna – with their lush soapstone decorations and ‘lathe-
turned’ circular pillars – would be reused in temples and marketplaces in
Vijayanagara. Indeed, these can be seen today in the sixteenth-century
Bhuvaneshvari shrine in Hampi’s Virupaksha temple,82 and in one of the
pillared hallways in the vast Vijayanagara period bazaar outside.83 A
Kalyana Chalukya tank was also relocated stone by stone to the royal
compound of the city,84 where it is still visible today. The powerful
sixteenth-century Vijayanagara ruler Rama Raya used the Chalukya boar
banner and he would explicitly call himself Chalukya-Chakravarti and
‘Lord of the Excellent City of Kalyana’:85 a major goal of his foreign policy
seems to have been to capture the city of Kalyana from the rival Deccan
Sultanates to add additional symbolic support to his status.
Bijapur, meanwhile, seamlessly integrated the grand halls, pillars and
elegant architectural ornamentation of Chalukya temples north of the
Krishna into their mosques, forts and palaces. Quite contrary to modern
stereotypes of Muslim intolerance towards ‘Hindus’, the curiosity and
respect that many rulers of Bijapur (such as Ali Adil Shah I in the late
sixteenth century, and his successor Ibrahim Adil Shah II) had towards
Deccani religions and cultural norms is well attested. Indeed, in the mid-
sixteenth century, it was Bijapur that was expelling Persian speakers and
encouraging the use of Kannada and Marathi, while Vijayanagara sought to
attract Persian talent.86
Professors Richard Eaton and Philip Wagoner suggest that Bijapur’s
assimilation of Chalukya architecture should be seen as an attempt to
directly associate Bijapuri kingship with the glory of the Chalukyas.
Providing the example of the great hall of the Aravattukhambada temple in
Bankapur87 (incidentally the same Bankapur founded by Bankeya, the
general of the long-forgotten ninth-century Vallabha Amoghavarsha),
converted by Ali Adil Shah I into the Jami mosque, they compare it to the
Ottoman empire’s conversion of the Byzantine church of St Sophia to the
Hagia Sophia mosque. ‘In cases such as these,’ they write, ‘agents of
change used their power not to vandalise or annihilate, but to preserve, and
in this way connect themselves with a society and a culture whose
architectural achievements they manifestly admired.’88 An inscription of
Someshvara II, placed by a Brahmin in service of the Bijapuri court by the
name of Baid Panditji, decorates one of the gates of the city to this day,89 as
do inscriptions of Someshvara II and Vikramaditya VI at the main entrance
to Bijapur, set up by Sultan Ibrahim I in the mid-sixteenth century. Ibrahim
even created a dramatic colonnade at this entrance, recycling Chalukya
pillars in the more geometric metropolitan Kalyana style from an
abandoned Narasimha temple of the city. These signify an almost unheard-
of position of prominence granted by a Persianate polity to its predecessors,
an attempt as clear as Vijayanagara’s to present itself as the legitimate
successors of the Chalukyas.
Bijapur’s sense of wonder at the inherited imperial sovereignty of the
Deccan is illustrated by a text from the reign of Ali Adil Shah I, dating to
the late sixteenth century. Bijapur was as removed in time from the Kalyana
Chalukyas as we are from Bijapur, and its people were well aware of the
fact that it was only the latest of many empires that ruled the region. The
Bijapuri historian Rafi al’din Shirazi, for example, was thoroughly
impressed by the enormous Krishneshvara monolith at Ellora, carved out
during the reign of the Rashtrakutas in the eighth century. Reflecting the
attitudes of the Persianate world, he described it as being a depiction of a
king in his palace, rather than an idol of Shiva. But Shirazi’s wonder at this
extraordinary edifice shines through:
The skill of each workshop [that worked on the temple] is cut into rock to such a degree that the
human mind cannot imagine it ... One should spend several days at the palace if one wishes to
see them all, and to understand them fully a long lifetime would be needed … So many beautiful
and well-wrought things are in those buildings and courtyards that, if one wished to explain
them all, he would fail to reach the goal. The listener should prepare for fatigue of the brain!90
For Shirazi and many other Muslims, whether Deccani or settled in the
Deccan, the magnificent temples of the medieval period were to be admired
for their beauty and protected as God’s works.91
The resurgence of north Indian power under the Mughal empire in the late
sixteenth century, and the arrival of armies led by Rajput generals and
Mughal princes in the Deccan, signalled the final end of the direct legacy of
the Vallabhas. The Deccan Sultanates, worn down by generations of
warfare, were subsumed within the subcontinent-spanning Mughal empire.
The emperor Aurangzeb Alamgir, generally thought of as a compulsive
iconoclast, has left us a surprisingly humane testimony of his own visit to
Ellora, describing the Kailashanatha monolith in a letter as ‘one of the
wonders of the true transcendent Artisan [God]’.92 This fabulous temple, a
remnant of an ancient Deccan sovereignty, was left untouched by this
mighty north Indian gunpowder empire.
Though the Chalukyas and Rashtrakutas were gradually forgotten, the
Deccan at large continued to remain one of the subcontinent’s dominant
geopolitical regions: the power of the Marathas and Mysore, for example,
would prove to be one of the greatest challenges to conquerors, from the
Mughals to the British. The Vallabhas’ stories would lie dormant until
nineteenth- and twentieth-century British efforts to translate and reconstruct
images of an ‘authentic’ Hindu past before the advent of Muslim-led
polities in the subcontinent. This reconstructed past would soon be
appropriated into regional nationalist narratives while being generally
neglected by postcolonial histories of the subcontinent, which were fixated
on ‘imperial moments’ centred on the Gangetic plains and northern India.
Indeed, despite its importance in shaping the subcontinent, the medieval
Deccan was barely represented in the National Museum in New Delhi up to
2020, whereas the Guptas of fifth-century north India occupied an entire
gallery. Today the Chalukyas and Rashtrakutas, and their successors and
rivals, are remembered in modern southern Indian states, though their
memory pales in comparison to the Marathas and Vijayanagara, and the
dynamics they shaped are generally poorly understood.
Politically expedient reclamations are commonplace: twentieth-century
Kannada cinema frequently depicted the Vatapi Chalukyas, who in reality
pioneered the use of Sanskrit, as proud Kannada speakers. More recently,
the region of Hyderabad–Karnataka has been renamed ‘Kalyana-
Karnataka’, though modern Basavakalyan is a relatively minor town and
Kalyana Chalukya temples are relatively neglected in the area. Meanwhile,
Bijapur (officially renamed Vijayapura) has lost much of its Chalukya
legacy to neglect, especially in comparison to Hampi/Vijayanagara. The
temples of the Vatapi Chalukyas in Pattadakal are a UNESCO World
Heritage Site, and those of the Hoysalas in southern Karnataka are major
tourist attractions. The Kakatiyas are revered in Telangana state, serving as
the namesake of a major irrigation project and the source of some material
for Tollywood films, while Kakatiya architecture features on the state
emblem. The Rashtrakutas, however, are barely remembered except for
their Gangetic expeditions, while the Seuna Yadavas are almost totally
forgotten outside of their entanglements with the armies of Delhi in the
thirteenth and fourteenth centuries. In comparison, the Cholas have become
one of the most important dynasties in modern Tamil nationalism, with their
seafaring expeditions often presented as evidence of Indian ‘colonization’
of Southeast Asia, and they are the subject of vast quantities of historical
fiction as well as a major upcoming film.93
Despite all these reclamations and rewritings, the Vallabhas’ actual legacy
continues in sometimes surprising and subtle ways. The impact they had on
urbanism, caste, courtly languages and the religious landscape of southern
India continues to shape the way we live today. Many of the sites
mentioned in this book – not merely the great capitals, but even minor
towns such as Bankapura, Lokkigundi and Sudi – have been continuously
inhabited to the present day. The Lingayats, descended from a religious
tradition tied to the history of the great city of Kalyana, are among the most
politically influential groups in modern-day Karnataka. Jain monasteries
which trace their ancestry to the Vallabhas or their vassals continue to thrive
in coastal Karnataka. The site of Shravanabelagola – once patronized by the
Gangas, allies of the Rashtrakuta Vallabhas – continues to be a major site of
Jain pilgrimage not far from modern Bengaluru. The Devanagari script,
which is used across India for Hindi and Sanskrit today, is descended from
the same family of scripts as the Nagari script, used by the chancellery of
the Kalyana Chalukyas from the eleventh to twelfth centuries.94
 One final example of the Chalukyas’ pervasive and surprising influence
lies in the Hindu laws of inheritance used during the British Raj. These laws
were based on a codification of the Mitakshara of Vijnaneshvara, composed
during the reign of Vikramaditya VI, and were widely used through the
Indian subcontinent (with the exception of Bengal) by British judges.95 To
the British, the conciseness of the text, its popularity with Brahmins across
their domains, and its association with ‘classical’, upper-caste Indian
kingship led to the assumption that it laid down the ‘traditional’ rules by
which ‘Hindu’ society – in reality an inaccurate colonial label applied to a
far more heterogeneous populace – should be governed. Thus the
application of this single Sanskrit text – written by an elite twelfth-century
Deccani Brahmin – to the vast, complex and diverse societies of the
nineteenth and twentieth centuries had significant and often damaging
effects on the vast majority of South Asians, who had probably never even
heard of the Mitakshara and instead dealt with inheritance using their own
laws and traditions. It was only in 1956, with the introduction of the Hindu
Code Bills, that the Mitakshara’s reign over India’s peoples finally came to
an end.96 But by then it had significantly influenced the application of
inheritance law and distribution of property for many generations. Indeed,
many of the readers of this book may have grandparents or great-
grandparents whose lives were directly impacted by this text composed in
the Vallabhas’ glittering court nearly a thousand years ago.
This exemplifies the interconnected nature of the history of the
subcontinent’s peoples, despite all the polities and dynasties that came after,
despite all the appropriation and erasures of the history of the Vallabhas. We
might wish to forget the complex and sometimes uncomfortable ways in
which every region of South Asia has shaped the present we live in today,
but the legacies of history are not easily erased.
The story of the Vallabhas who ruled the Deccan, the story of the men
who called themselves Fortune’s Favourite and Earth’s Beloved, is one of
constant savagery, brilliant ambition, and horrific loss and injustice. This
book has tried to tell it in a way that reveals something of the complicated
individuals they were, and allows a glimpse of the teeming, forgotten
peoples – our distant ancestors – whom they ruled over. But all this is only
one story, one narrative, one region, one period, one group of elite families.
The immensity of medieval south India still looms over us, demanding that
we continue to tell its stories, and learn from its horrors and glories.
Cox 2016, 36.
SII XI-I, 77.
EHD I, 341.
Cox 2016, 36.
He is referred to as ‘Rajiga’ in Chalukya sources, likely an insulting diminutive comparable to
‘Vikki’ (see EHD I, 335). His original name was likely Rajendra; he would use that name later in
his career, as we will see. I refer to him as ‘Rajiga’ in this formative period, following the
precedent of Cox 2016, and for ease of reading; there are otherwise too many Rajendras in this
royal drama.
Cox 2016, 38.
Velcheru Narayana Rao and David Shulman, Classical Telugu Poetry: An Anthology (University of
California Press, 2002), 59.
Ibid., 56.
See Velcheru Narayana Rao and David Shulman, Symbols of Substance: Court and State in Nāyāka
Period Tamil Nadu (Oxford University Press, 1992).
Merutungācārya 1901, 30–36.
Singh 2012, 21.
Ibid.
Pollock 2006, 179–80.
EHD I, 341–42. The authors claim that the campaigns in Vengi were a prelude to this battle, but this
book presents an alternative explanation of the evidence – namely that the battle of Kudala-
Sangama was a minor skirmish embellished by later Chola propaganda (following Cox 2016), and
that the main thrust of the Chola–Chalukya wars of 1061–62 was Vengi, which led to the death of
Vijayaditya’s son Shaktivarman II.
EHD I, 342.
For a detailed discussion, see Malini Adiga, ‘Rape in Early Medieval Karnataka: Inter-Village
Warfare and Opportunities for Male Heroism’, Proceedings of the Indian History Congress 78
(2017).
Ryali plates of Vijayaditya VII, quoted in N. Venkataramanayya, The Eastern Cāl.ukyas of Vengi
(Vedam Venkataraya Sastry & Bros, 1950), 39–40.
Ibid.
Cox 2016, 42.
Sastri 1955, 271–72.
Bilhana, The Vikramânkadevacharita, A Life of King Vikramâditya-Tribhuvanamalla of Kalyana, ed.
Georg Bühler (Central Book Depot, 1875), 32.
SII III, 68.
Ibid.
Ibid.
Sastri 1955, 290.
Cox 2016, 67.
Ibid., 64.
Ibid., 74.
Cox 2016, 67.
Ibid., 68.
This book uses Professor Sastri’s depiction of this ‘diplomatic revolution’. See Sastri 1955, 290.
As mentioned in the previous chapter, the authors of EHD think he was present in the 1044 campaign
in Vengi. At the very least he would have had to be approaching his twenties if he was one of the
Chalukya commanders that Rajadhiraja Chola mentions in his inscriptions. This means he was
probably born sometime in the late 1020s. However, Vikramaditya VI only died in 1126 – which
would have made him an improbable hundred and ten years of age at the time of his death.
Bilhana 1875, 30.
Shrigondekar 1939, 25–26. This was not specifically said about Vikramaditya but is prescribed as
standard for Chalukya kings by his son Someshvara III, making it quite likely that it was
something that Vikramaditya did
Ibid.
EHD I, 347.
Whitney Cox, ‘Sharing a Single Seat: The Poetics and Politics of Male Intimacy in the
Vikramān.kakāvya,’ Journal of Indian Philosophy 38, no. 5 (2010).
Sastri 1955, 269.
Bilhana’s Vikramankadevacharitam makes it clear that the court accompanied Someshvara and that
he walked into the waters. See Cox 2016, 126.
This is also from Bilhana. See ibid.
Cox 2016, 127.
EHD I, 349.
Vira-Rajendra did actually fight a limited war with Someshvara II, which may or may not have been
responsible for Vikramaditya’s appointment as governor of Gangavadi. See EHD I, 350.
Cox 2016, 73.
Ibid.
Ibid., 75.
Ibid., 94.
Ibid.
Ibid., 114.
EHD I, 351.
Ibid., 351–52.
Ibid., 353.
Ibid., 352.
Ibid., 354.
This, at least, is the version depicted in the Vikramankadevacharitam.
EHD I, 355.
Ibid.
Ibid., 366.
Ibid., 364–65.
Eaton and Wagoner 2014, 6.
Donald R. Davis and David Brick, ‘Social and Literary History of Dharmaśāstra: Commentaries and
Legal Digests’, in Patrick Olivelle and Donald R. Davis, Jr., eds. The Oxford History of Hinduism:
 Hindu Law: A New History of Dharmaśāstra (Oxford University Press, 2018), 40.
These are all described as aspects of the emperor’s schedule in the Manasollasa of Someshvara III.
EHD I, 367.
K.A. Nilakanta Sastri, The Cōl.as, Volume II (Part I) (University of Madras, 1937), 15–16.
Ibid., 49.
Ibid., 36–37. According to Professor Sastri, the powerful Anantavarman Codaganga, ruler of much of
twelfth-century Kalinga, Utkala and Tosali (corresponding to modern-day Odisha), was the son of
Rajiga-Kulottunga’s daughter.
Ibid., 48.
For a brief account, see EHD I, 359–64. While this claims that the ‘Chalukya empire managed to
fairly easily weather the storm’, this explanation seems to fit the evidence better: Hoysala
Vishnuvardhana very clearly had imperial ambitions and all but declared them openly with his
enormous Vira-Narayana (now the Chennakeshava) temple at Belur.
Sastri 1937, 42.
Ibid., 44.
Ibid., 179. The course of events is much more complex than the brief summary presented here.
Indeed, on multiple occasions, the Hoysalas actually came to the aid of the Cholas to maintain the
balance of power against the Pandyas.
Ibid., 208.
It is worth pointing out that the Vijayanarayana and the Gondeshvara display architectural influences
from North India, probably due to the pan-subcontinental imperial ambitions of their ruling
dynasties, and their historical involvement with Malwa. The Ramappa is more Deccano-Dravidian.
There would be one last Chalukya ruler, Someshvara IV, after Bijjala, but for all intents and purposes
the dynasty’s power came to an end with Bijjala’s coup.
H.S. Shivaprakash, I Keep Vigil of Rudra: The Vachanas (Penguin Random House, 2010).
O.L. Nagabhushana Swamy, ed. The Sign: Vachanas of 12th Century (Prasaranga Kannada
University, 2007).
Eaton and Wagoner 2014, 14.
Prithvi Datta Chandra Shobhi, ‘Kalyāṇa is Wrecked: The Remaking of a Medieval Capital in Popular
Imagination’, South Asian Studies 32, no. 1 (2016).
Shivaprakash 2010, xii.
A detailed discussion may be found in Richard M. Eaton and Phillip B. Wagoner, ‘Warfare on the
Deccan Plateau, 1450–1600: A Military Revolution in Early Modern India?’ Journal of World
History 25, no. 1 (2014).
Srinivas Reddy, ‘Stallions of the Indian Ocean’, in Exploring Materiality and Connectivity in
Anthropology and Beyond, eds. Philipp Schorch, Martin Saxer and Marlen Elders (UCL Press,
2020).
Eaton and Wagoner 2014, 98–106.
Ibid., 96–97.
Ibid., 106–13.
Ibid., 89.
Ibid., 128.
Ibid., 139–45.
Ibid., 156.
Ibid., 137–38.
Carl W. Ernst, ‘Admiring the Works of the Ancients: The Ellora Temples as Viewed by Indo-Muslim
Authors’, in David Gilmartin and Bruce B. Lawrence, eds. Beyond Turk and Hindu: Rethinking
Religious Identities in Islamicate South Asia (University of Florida Press, 2000), 108.
Ibid., 107.
Ibid., 109.
This would be Ponniyin Selvan.
This is a claim made by Whitney Cox in the 2010 paper ‘Scribe and Script in the Cāl.ukya West
Deccan’ (p. 19) and repeated by Eaton and Wagoner in their 2014 book, Power, Memory,
Architecture (p. 8).
Ludo Rocher, ‘Inheritance: Dāyabhāga’, in Olivelle and Davis, Jr., 2018.
Ibid., 178.
 Bibliography

List of Abbreviations
CII Corpus Inscriptionum Indicarum
EC Epigraphia Carnatica
EHD Early History of the Deccan
EI Epigraphia Indica
IA Indian Antiquary
SII South Indian Inscriptions
Primary Sources
Alberuni. Alberuni’s India. Edited and translated by Edward C. Sachau. Kegan Paul, Trench, Trubner
& Co, 1910.
al-Sirafi, Abu Zayd. Two Arabic Travel Books: Accounts of India and China. Edited and translated by
Tim Mackintosh-Smith. New York University Press, 2014.
Bilhana. The Vikramânkadevacharita, a Life of King Vikramâditya-Tribhuvanamalla of Kalyân.a.
Edited by George Bühler. Bombay: Central Book Depot, 1875.
Bilhana. Vikramānkadeva Caritam: Glimpses of the History of the Cāl.ukyas of Kalyān.a. Translated
by Sures Chandra Banerji. Sambodhi Publications, 1965.
Cūlavam.sa: Being the More Recent Part of the Mahāvam.sa. Geiger, Wilhelm, and Mabel C.
Rickmers. Asian Educational Services, 1992.
Dandin. Tales of the Ten Princes. Translated by A.N.D. Haksar. Penguin India, 2007. Kindle Edition.
Gazetteer of the Bombay Presidency, Volume I Part II: History of the Konkan Dakhan and Southern
Maratha Country. Government Central Press, 1896.
Hsüan-tsang. Buddhist Records of the Western World. Translated by Samuel Beal. 2 volumes.
Trübner, 1884.
Kalidasa. Meghadutam. Translated by Srinivas Reddy. Penguin Random House India, 2015.
Karashima, Noboru, and Tansen Sen. ‘Chinese Texts Describing or Referring to the Chola Kingdom
as Zhu-Nian’. In Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to
Southeast Asia, edited by Hermann Kulke, K. Kesavapany and Vijay Sakhuja,
292–315. Manohar, 2019.
Karashima, Noboru, and Y. Subbarayulu. ‘Ancient and Medieval Tamil Sanskrit Inscriptions Relating
to Southeast Asia and China’. In Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval
Expeditions to Southeast Asia, edited by Hermann Kulke, K. Kesavapany and Vijay Sakhuja, 271–
91. Manohar, 2019.
Krishnan, K.G. Karandai Tamil Sangam Plates of Rajendrachola I. Archaeological Survey of India,
1984.
Mackintosh-Smith, Tim, James E. Montgomery, Philip F. Kennedy and Shawkat M. Toorawa, eds.
Two Arabic Travel Books: Accounts of China and India and Mission to the Volga. New York
 University Press, 2014.
Mahendravikramavarman. ‘Matta-vilāsa: A Farce’. Translated by L.D. Barnett. Bulletin of the School
of Oriental Studies, 5, no. 4 (1930): 697–717.
Matsya Puranam. The Sacred Books of the Hindus Translated by Various Sanskrit Scholars, edited
by B.D. Basu. Allahabad: The Panini Office, 1917.
Merutungācārya. The Prabandhacintāman.i, or Wishing-Stone of Narratives. Translated by C.H.
Tawney. The Asiatic Society, 1901.
Mirashi, Vasudev Vishnu. Corpus Inscriptionum Indicarum. Vol. IV, part I. Government Epigraphist
for India, 1955.
Olivelle, Patrick. Dharmasutras: The Law Codes of Ancient India. Oxford University Press, 1999.
Olivelle, Patrick, trans. King, Governance, and Law in Ancient India: Kautilya’s Arthaśāstra. Oxford
University Press, 2013.
Ranna. Gadāyuddham: The Duel of the Maces. Edited by Akkamahadevi. Translated by R.V.S.
Sundaram and Ammel Sharon. Manohar, 2019.
Sewell, Robert. Historical Inscriptions of Southern India (Collected till 1923). Edited by
Krishnaswami S. Aiyangar. Madras University, 1932.
Shrigondekar, G.K., ed. Mānasollāsa of King Someśvara. Vol. 2. Oriental Institute, 1939.
Shrinivas, Ritti, and G.C. Shelke, eds. Inscriptions of Nanded District. Yashwant Mahavidyalaya,
1968.
Sundaram, R.V.S., and Deven M. Patel, trans. Śrīvijaya’s Kavirājāmārgam.. Manohar, 2017.
The Inscriptions of Nagai. Hyderabad Archaeological Series, No. 8. The Nizam’s Government, 1928.
Various Editors. Epigraphia Carnatica. Vols II, VIII. Mysore Archaeological Department, University
of Mysore.
Various Editors. Epigraphia Indica. Vols III, IV, V, VI, X, XIII, XIV, XV, XVI, XVIII, XIX, XXV,
XXIX, XXXII. Archaeological Survey of India.
Various Editors. Indian Antiquary. Vols VI, X, XI, XII. Swati Publications.
Various Editors. South Indian Inscriptions. Vols I, II, III. Archaeological Survey of India.
Vatsyayana. Kama Sutra: A Guide to the Art of Pleasure. Translated by A.N.D. Haksar. New Delhi:
Penguin Books, 2011.
Secondary Sources
Abraham, Meera. Two Medieval Merchant Guilds of South India. Manohar, 1988.
Acharya, Prasanna Kumar. Architecture of Mānasāra: Translated from the Original Sanskrit.
Mānasāra Series, Vol. 4. Munshiram Manoharlal, 1934.
———. Architecture of Mānasāra: Illustrations of Architectural and Sculptural Objects with a
Synopsis. Mānasāra Series, Vol. 5. Oxford University Press, 1934.
Adiga, Malini. ‘Dharmasastras, the Dravidian Kinship System and Female Inheritance in Karantaka
in the Early Medieval Period (400 A.D. to 1300 A.D.)’. Proceedings of the Indian History
Congress 64 (2003):
173–84.
———. The Making of Southern Karnataka: Society, Polity and Culture in Early Medieval Period,
AD 400-1030. Orient BlackSwan, 2006.
———. ‘Of Concubines and Devadasis: Prostitution in Early Medieval Kartataka’. Proceedings of
the Indian History Congress 73 (2012): 144–54.
———. ‘Rape in Early Medieval Karnataka: Inter-Village Warfare and Opportunities for Male
Heroism’. Proceedings of the Indian History Congress 78 (2017): 207–19.
Adshead, S.A.M. T’ang China: The Rise of the East in World History. Palgrave Macmillan, 2004.
Adshead, S.A.M., and Reshmi Dutta-Flanders. China in World History. Palgrave Macmillan, 1988.
Agrawala, Vasudeva S. The Deeds of Harsha: Being a Cultural Study of Bana’s Harshacharita.
Prithvi Prakashan, 1969.
Alcock, Susan E., Terence N. D’Altroy, Kathleen D. Morrison and Carla M. Sinopoli, eds. Empires:
Perspectives from Archaeology and History. Cambridge University Press, 2001.
Ali, Daud. ‘Violence, Gastronomy and the Meanings of War in Medieval South India.’ The Medieval
History Journal 3, no. 2 (2000): 261–89. https://doi.org/10.1177/097194580000300204.
———. Courtly Culture and Political Life in Early Medieval India. Cambridge University Press,
2006.
———. ‘The Service Retinues of the Chola Court: A Study of the Term Veḷam in Tamil
Inscriptions’. Bulletin of School of Oriental and African Studies 70, no. 3 (2007): 487–509.
———. ‘Aristocratic Body Techniques in Early Medieval India’. In Rethinking a Millennium –
Perspectives on Indian History from the Eighth to the Eighteenth Century: Essays for Harbans
Mukhia, edited by Rajat Datta, 69–93. Aakar, 2008.
———. ‘Bhoja’s Mechanical Garden: Translating Wonder across the Indian Ocean, circa 800–1100
CE’. History of Religions 55, no. 4 (2016): 460–93.
Altekar, Anant Sadashiv. The Rashtrakutas and Their Times; Being a Political, Administrative,
Religious, Social, Economic and Literary History of the Deccan during C. 750 AD to C. 1000 AD.
Oriental Book Agency, 1934.
Andrade, Tonio. The Gunpowder Age: China, Military Innovation, and the Rise of the West in World
History. Princeton University Press, 2017.
Ariel, Glucklich. The Strides of Vishnu: Hindu Culture in Historical Perspective. Oxford University
Press, 2008.
Aruna, A. State Formation in the Eastern Deccan (7th Century AD–13th Century AD). Bhartiya Kala
Prakashan, 2000.
Arunachalam, B. Chola Navigation Package. Maritime History Society, 2004.
Bahulkar, S.S., and Shilpa Sumant, eds. ‘The Bijapur–Mumbai Copperplate Grant of the Cālukya
Ruler Pulakeśin II, Dated April 04, 619 CE, Mentioning His Triumph over Emperor
Harṣavardhana’. Annals of the Bhandarkar Oriental Research Institute XCII (2012): 205–09.
Bakker, Hans. The World of the Skandapurān.a: Northern India in the Sixth and Seventh Centuries.
Groningen Oriental Studies, Vol. 4, Supplement. Brill Academic Publishers, 2014.
Barua, Pradeep P. The State at War in South Asia. University of Nebraska Press, 2009.
Bennison, Amira K. The Great Caliphs: The Golden Age of the ʿAbbasid Empire. Yale University
Press, 2009.
Berkson, Carmel. Ellora: Concept and Style. Second Edition. Abhinav Publications, 2004.
———. ‘Daśāvatāra Cave: Its Importance in the History of World Art’. In Ellora Caves: Sculpture
and Architecture, edited by Ratna Parimoo, Deepak Kannal and Shivaji Panniker, Revised Edition,
166–82. Aprant, 2018.
Bhattacharyya, Narendra Nath. Ancient Indian Rituals and Their Social Contents. Manohar, 1975.
Bielenstein, Hans. Diplomacy and Trade in the Chinese World, 589–1276. Brill, 2005.
Blankinship, Khalid Yahya. The End of the Jihad State: The Reign of Hisham Ibn’Abd al-Malik and
the Collapse of the Umayyads. SUNY Press, 1994.
Bolon, Carol Radcliffe. Forms of the Goddess Lajja Gauri in Indian Art. Pennsylvania State
University Press, 1992.
Bose, S.K. Boot, Hooves, and Wheels: And the Social Dynamics Behind South Asian Warfare. Vij
Books India, 2015.
Buchanan, Susan Locher. ‘Cāl.ukya Temples: History and Iconography’. PhD dissertation. Ohio State
University, 1985.
Burjor, Avari. Islamic Civilization in South Asia: A History of Muslim Power and Presence in the
Indian Subcontinent. Routledge, 2019.
Chakrabarti, Dilip K. The Geopolitical Orbits of Ancient India: The Geographical Frames of the
Ancient Indian Dynasties. Oxford University Press, 2011.
Chakravarti, Prithwis Chandra. The Art of War in Ancient India. University of Dacca, 1941.
Chakravarti, Ranabir. ‘The Pull Towards the Coast: Politics and Polity in India (c. 600–1300 CE)’.
Presidential Address Indian History Congress, 2011, 1–48.
———. ‘Merchants, Merchandise and Merchantmen in the Western Seaboard of India: A Maritime
Profile (500 BCE–1500 CE)’. In The Trading World of Indian Ocean, 1500–1800, edited by Om
Prakash and D.P. Chattopadhyaya, 53–116. Pearson Education, 2012.
Champakalakshmi, R. Trade, Ideology and Urbanisation: South India 300 BC to AD 1300. Oxford
University Press, 1996.
Chandler, David. A History of Cambodia. Taylor and Francis, 2018.
Chattopadhyaya, Brajadulal. Representing the Other: Sanskrit Sources and the Muslims, Eighth to
Fourteenth Century. Primus Books, 1998.
Chaudhuri, Mamata. ‘Ship-Building in the Yuktikalpataru and Samarangana Sutradhara’. Indian
Journal of the History of Science (1975): 137–47.
Cousens, Henry. The Chalukya Architecture of the Kanarese Districts. Central Government of India,
1926.
Cox, Whitney. ‘Scribe and Script in the Cāl.ukya West Deccan’. The Indian Economic and Social
History Review 47, no. 1 (2010): 1–28.
———. ‘Sharing a Single Seat: The Poetics and Politics of Male Intimacy in the Vikrāman.kakāvya.’
Journal of Indian Philosophy 38, no. 5 (2010): 485–501.
———. ‘Politics, Kingship, and Poetry in Medieval South India: Moonset on Sunrise Mountain.
Cambridge University Press, 2016.
Cummings, Cathleen Ann. ‘A Study of the Iconographic Program of the Lokesvara (Virupaksa)
Temple, Pattadakal’. PhD dissertation. Ohio State University, 2006.
Dalal, Kurush. ‘Sanjan: Digging Deep Into History’. Live History India, 22 September 2019.
https://www.livehistoryindia.com/story/cover-story/sanjan-digging-deep-into-history/.
Davidson, Ronald M. Indian Esoteric Buddhism: A Social History of the Tantric Movement.
Columbia University Press, 2002.
Davis, Jr., Donald R., and David Brick. ‘Social and Literary History of Dharmaśāstra: Commentaries
and Legal Digests’. In The Oxford History of Hinduism: Hindu Law: A New History of
Dharmaśāstra, edited by Patrick Olivelle and Donald R. Davis, Jr., 30–48. Oxford University
Press, 2018.
Davis, Richard H. ‘Indian Art Objects as Loot’. The Journal of Asian Studies 52, no. 1 (1993): 22–
48.
———. Ritual in an Oscillating Universe: Worshipping Śiva in Medieval India. Princeton University
Press, 2014.
DeCaroli, Robert. ‘An Analysis of Dan.d.in’s Daśakumāracarita and Its Implications for Both the
Vākāṭaka and Pallava Courts.’ Journal of the American Oriental Society 115, no. 4 (1995): 671–
 78.
Dehejia, Vidya. Slaves of the Lord: The Path of the Tamil Saints. Munshiram Manoharlal, 2002.
———. The Body Adorned: Sacred and Profane in Indian Art. Columbia University Press, 2009.
Deoras, V.R. ‘Fresh Light on the Southern Campaigns of the Rāshṭrakūṭa Emperor Kr.ishn.a III.’
Proceedings of the Indian History Congress 20 (1957): 133–40.
Devare, Hema. ‘Cultural Implications of the Chola Maritime Fabric Trade with Southeast Asia.’ In
Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia,
edited by Hermann Kulke, K. Kesavapany and Vijay Sakhuja, 178–92. Manohar, 2019.
Dhavalikar, M.K. ‘Kailasa: The Stylistic Development and Chronology’. In The Rashtrakutas of
Malkhed: Studies in Their History and Culture, edited by B.R. Gopal, 254–64. The Mythic
Society/Geetha Book House, 1994.
———. Ellora: Monumental Legacy. Oxford University Press, 2003.
———. ‘Sutradhar’. Annals of the Bhandarkar Oriental Research Institute, no. 52 (1971): 215–20.
Dikshit, Durga Prasad. Political History of the Chalukyas of Badami. Abhinav Publications, 1980.
Dirks, Nicholas B. ‘Political Authority and Structural Change in Early South Indian History’. Indian
Economic and Social History Review 13, no. 2 (1976): 125–57.
https://doi.org/10.1177/001946468101300201.
Doniger, Wendy. The Hindus: An Alternative History. Penguin, 2009.
Dundas, Paul. The Jains. Second Edition. Routledge, 2002.
Durga Prasad, J. ‘The Chalukyas of Vengi’. In Comprehensive History and Culture of Andhra
Pradesh, Vol. 3: Early Medieval Andhra Pradesh AD 624–1000, edited by B. Rajendra Prasad, 32–
46. Tulika Books, 2009.
Eaton, Richard M. A Social History of the Deccan, 1300–1761: Eight Indian Lives. The New
Cambridge History of India. Cambridge University Press, 2005.
Eaton, Richard M., and Phillip B. Wagoner. Power, Memory, Architecture: Contested Sites on India’s
Deccan Plateau, 1300–1600. Oxford University Press, 2014.
———. ‘Warfare on the Deccan Plateau, 1450–1600: A Military Revolution in Early Modern India?’
Journal of World History 25, no. 1 (2014): 5–50.
Eck, Diana L. India: A Sacred Geography. Harmony, 2012.
Ernst, Carl W. ‘Admiring the Works of the Ancients: The Ellora Temples as Viewed by Indo-Muslim
Authors’. In Beyond Turk and Hindu: Rethinking Religious Identities in Islamicate South Asia,
edited by David Gilmartin and Bruce B. Lawrence, 98–120. University of Florida Press, 2000.
Eswar, Rohitha, Vrushab Mahesh and K. Krishnan. ‘“Iha” to “Para”: A Hero’s Transitional Journey
to the Other World from This World – Reading Veeragals of Hassan’. Heritage: Journal of
Multidisciplinary Studies in Archaeology 2 (2014): 108–24.
Fleming, Benjamin J. ‘Ellora Cave 16 and the Cult of the Twelve Jyotirlingas.’ In Living Rock:
Buddhist, Hindu and Jain Cave Temples in the Western Deccan, edited by Pia Brancaccio. Marg
Foundation, 2013.
Flood, Gavin. The Blackwell Companion to Hinduism. Wiley, 2003.
Foekema, Gerard. Architecture Decorated with Architecture: Late Medieval Temples of Karnataka,
1000–1300. Munshiram Manoharlal, 2003.
Frankopan, Peter. The Silk Roads: A New History of the World. Bloomsbury, 2015.
Ganguly, D.C. ‘The “Gurjaras” in the Rastrakuta Inscriptions’. Proceedings of the Indian History
Congress 3 (1939): 513–15.
Goetz, H. ‘The Kailasa of Ellora and the Chronology of Rashtrakuta Art’. Artibus Asiae 15, no. 1/2
 (1952): 84–107.
Goyal, Shankar. ‘Acquisition of the Maukhari Empire by Harsha’. Proceedings of the Indian History
Congress 62 (2001): 1090–91.
Gupta, S.P., Kurush Dalal, R. Nanji and Abhijit Dandekar. ‘Preliminary Report on the Excavations at
Sanjan (2002)’. Purattatva 32 (n.d.): 182–244.
Handiqui, Krishna Kanta. Yaśastilaka and Indian Culture. Jain Samskriti Samraksha Sangha, 1949.
Hardy, Adam. ‘Tradition and Transformation: Continuity and Ingenuity in the Temples of
Karnataka’. Journal of the Society of Architectural Historians 60, no. 2 (2001): 180–99.
———. Theory and Practice of Temple Architecture in Medieval India: Bhoja’s
Samarān.gan.asūtradhāra and the Bhojpur Line Drawings. Indira Gandhi National Centre for
Arts, 2015.
Hawley, John Stratton. ‘Scenes from the Childhood of Kṛṣṇa on the Kailāsanātha Temple, Ellora’.
Archives of Asian Art 34 (1981): 74–90.
Heng, Geraldine. ‘An Ordinary Ship and Its Stories of Early Globalism’. Journal of Medieval Worlds
1, no. 1 (2019): 11–54. https://doi.org/10.1525/jmw.2019.100003.
Hirsh, Marilyn. ‘Mahendravarman I Pallava: Artist and Patron of Māmallapuram’. Artibus Asiae 48,
no. 1 (1987): 109–30.
Hopkins, Thomas J. The Hindu Religious Tradition. Wadsworth Publishing Company, 1971.
Inden, Ronald. ‘Hierarchies of Kings in Early Medieval India’. Contributions to Indian Sociology 15,
no. 1–2 (1981): 99–125. https://doi.org/10.1177/006996678101500108.
———. Imagining India. Indiana State University, 2000.
Jamison, Stephanie W. ‘Roles for Women in Vedic Srauta Ritual’. In Goddesses and Women in the
Indic Religious Tradition, edited by Arvind Sharma, 1–17. Brill’s Indological Library 24. Brill,
2005.
Jantakal, Gautam, Bhushan Kapadia and Prakash Jinjuvadiya. History and Coinage of the
Rashtrakutas. IIRNS Publications LLP, 2019.
Kadambi, Hemanth. ‘Sacred Landscapes in Early Medieval South India: The Chalukya State and
Society (ca. AD 550-750)’. PhD dissertation. University of Michigan, 2011.
Kakar, Sudhir. Intimate Relations: Exploring Indian Sexuality. University of Chicago Press, 1989.
Kaligotla, Subhashini. ‘Shiva’s Waterfront Temples: Reimagining the Sacred Architecture of India’s
Deccan Region’. PhD dissertation. Columbia University, 2015.
https:/academiccommons.columbia.edu/doi/10.7916/D83N2B1V/download.
Kamath, Suryanath U. ‘Amoghavarsha I: An Appraisal’. In The Rashtrakutas of Malkhed: Studies in
Their History and Culture, edited by B.R. Gopal, 82–88. The Mythic Society, 1994.
Kannal, Deepak. ‘The Regional Lineages and Possible Masters at Kailāśanātha Temple of Ellora’. In
Ellora Caves: Sculpture and Architecture, edited by Ratan Parimoo, Deepak Kannal and Shivaji
Panniker, Revised Edition. Aprant, 2018.
Karashima, Noboru. ‘Medieval Commercial Activities in the Indian Ocean Region as Revealed from
Chinese Ceramic-Sherds and South Indian and Sri Lankan Inscriptions’. In Nagapattinam to
Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia, edited by Hermann
Kulke, K. Kesavapany and Vijay Sakhuja, 20–60. Manohar, 2019.
———. ‘South Indian Merchant Guilds in the Indian Ocean and Southeast Asia’. In Nagapattinam to
Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia, edited by Hermann
Kulke, K. Kesavapany and Vijay Sakhuja, 135–57. Manohar, 2019.
Kasdorf, Katherine E. ‘Forming Dōrasamudra: Temples of the Hoysaḷa Capital in Context’. PhD
 dissertation. Columbia University, 2013.
Kaul, Shonaleeka. Imagining the Urban: Sanskrit and the City in Early India. Seagull Books, 2011.
Keay, John. India: A History. Revised and Updated. Grove/Atlantic Inc., 2011.
Keegan, John. The Face of Battle: A Study of Agincourt, Waterloo and the Somme. Random House,
2011.
Kennedy, Hugh. Prophet and the Age of the Caliphates. Second Edition Pearson Longman, 2004.
———. The Armies of the Caliphs: Military and Society in the Early Islamic State. Routledge, 2013.
Konstantin, Nossov, and Peter Dennis. War Elephants. Osprey Publishing, 2008.
Kulkarni, R.H. ‘Vatapi Ganapati, an art historical revisit’. New Indian Express, 22 August 2020.
https://www.newindianexpress.com/opinions/2020/aug/22/vatapi-ganapati-an-art-historical-revisit-
2186817.html
Kulke, Hermann. ‘Śrīvijaya Revisited: Reflections on State Formation of a Southeast Asian
Thalassocracy’. Bulletin de l’Ecole Française d’Extrême-Orient 102, no. 1 (2016): 45–95.
———. ‘The Naval Expeditions of the Cholas in the Context of Asian History’. In Nagapattinam to
Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia, edited by Hermann
Kulke, K. Kesavapany and Vijay Sakhuja, 1–20. Manohar, 2019.
Lambourn, Elizabeth. ‘Describing a Lost Camel: Clues for West Asian Mercantile Networks in South
Asian Maritime Trade (Tenth–Twelfth Centuries CE)’. In Ports of the Ancient Indian Ocean:
Proceedings of the Kolkata Colloquium 2011 (Median Project), edited by M.F. Boussac,
J.F. Salles and J.B. Yon, 351–407. Primus Books, 2016.
Lang, Karen C. ‘Candrakirti on the Medieval Military Culture of South India’. In Buddhism in the
Krishna River Valley of Andhra, edited by Sree Padma Holt and A.W. Barber, 127–50. SUNY
Press, 2008.
Lath, Mukund. ‘Somadeva Suri and the Question of Jain Identity’. In The Assembly of Listeners:
Jains in Society, edited by Michael Carrithers and Caroline Humphrey, 19–30. Cambridge
University Press, 1991.
Lee, Risha. ‘Rethinking Community: The Indic Carvings of Quanzhou’. In Nagapattinam to
Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia, edited by Hermann
Kulke, K. Kesavapany and Vijay Sakhuja, 240–70. Manohar, 2019.
Majumdar, R.C. The Age of Imperial Kannauj. The History and Culture of the Indian People, Vol. 4.
Second Edition. Bharatiya Vidya Bhavan, 1959.
Malekandathil, Pius. Maritime India. Primus Books, 2010.
Markel, Stephen. ‘The “Ramāyana” Cycle on the Kailāsanātha Temple at Ellora’. Ars Orientalis 30,
Supplement 1 (2000): 59–71.
McHugh, James. ‘The Incense Trees of the Land of Emeralds: The Exotic Material Culture of
Kāmaśāstra’. Journal of Indian Philosophy 39, no. 1 (2011): 63–100.
McKnight, John Michael. ‘The Gupta Temple Movement: A Study of Political Aspects of the Early
Hindu Temple’. PhD dissertation. McMaster University, 1973.
McLaughlin, Raoul. Rome and the Distant East: Trade Routes to the Ancient Lands of Arabia, India
and China. Continuum, 2010.
Menon, Srikumar M. ‘The Curious Case of the Galaganatha Dolmen: Possible Links Between
Megalithic Monuments and Early Temples at Aihole’. Heritage: Journal of Multidisciplinary
Studies in Archaeology 2 (2014): 54–73.
Mirashi, Vasudev Vishnu. ‘Manor Plates of Rāshṭrakūta Dantidurga: Śaka Year 671’. Proceedings of
the Indian History Congress 21 (1958).
———. ‘The Javakhed.a Plates of Amoghavarsha I: Śaka 742: A Critical Examination of Some
Problems’. Proceedings of the Indian History Congress 23 (1960): 31–36.
Mishra, Jayashri. Social and Economic Conditions Under the Imperial Rashtrakutas. Commonwealth
Publishers, 1992.
Mitra, Rhea, and Kurush Dalal. ‘A Report on the Glass Vessels from Sanjan, 2002’. Journal of
Indian Ocean Archaeology 2 (2005): 62–68.
Moraes, George M. The Kadamba Kula. B.X. Furtado & Sons, 1931.
Morrison, Kathleen D. ‘Trade, Urbanism, and Agricultural Expansion: Buddhist Monastic
Institutions and the State in the Early Historic Western Deccan’. World Archaeology 27, no. 2
(1995): 203–21.
Nandi, Ramendra Nath. Religious Institutions and Cults in the Deccan (c. A.D. 600–A.D. 1100).
Motilal Banarsidass, 1973.
———. State Formation, Agrarian Growth and Social Change in Feudal South India c. AD 600–
1200. Manohar, 2000.
Owen, Lisa Nadine. Carving Devotion in the Jain Caves at Ellora. Brill, 2012.
Brancaccio, Pia, ed. ‘Art, Architecture and Devotional Practice at Ellora’. In Living Rock: Buddhist,
Hindu and Jain Cave Temples in the Western Deccan, edited by Pia Brancaccio, Marg Foundation,
2013.
Padigar, Srinivas V. ‘Craftsmen’s Inscriptions from Badami: Their Significance’. In Ellora Caves:
Sculpture and Architecture, edited by Ratan Parimoo, Deepak Kannal and Shivaji Panniker,
Revised Edition, 313–19. Aprant, 2018.
Pollard, Edward, and Okeny Charles Kinyera. ‘The Swahili Coast and the Indian Ocean Trade
Patterns in the 7th–10th Centuries CE’. Journal of Southern African Studies 43, no. 5 (2017): 927–
47. https://doi.org/10.1080/03057070.2017.1345266.
Pollock, Sheldon. The Language of the Gods in the World of Men: Sanskrit, Culture and Power in
Premodern India. University of California Press, 2006.
Prasad, Om Prakash. ‘Trade in the Growth of Towns: A Case Study of Karnataka (c. 600–1200
A.D.)’. Proceedings of the Indian History Congress 41 (1980): 152–60.
Prentiss, Karen Pechilis. The Embodiment of Bhakti. Oxford University Press, 1999.
Rabe, Michael D. ‘The Māmallapuram Praśasti: A Panegyric in Figures’. Artibus Asiae 57, no. 3
(1997): 189–241.
Ramesh K.V. ‘Pulakesin II – His Career and Personality’. In The Chalukyas of Badami: Seminar
Papers, edited by M.S. Nagaraja Rao, 53–65. The Mythic Society, 1978.
Rangaswamaiah, G.R. ‘Bhuvikrama’s Role in the Chalukya–Pallava Conflict (635–679)’.
Proceedings of the Indian History Congress 35 (1974): 60–63.
http://www.jstor.org/stable/44138760.
———. ‘The Rashtrakuta Relations with the Gangas of Talkad’. In The Rashtrakutas of Malkhed:
Studies in Their History and Culture, edited by B.R. Gopal, 204–10. The Mythic Society, 1994.
Ray, Himanshu Prabha. ‘Creating Religious Identity: Archaeology of Early Temples in the
Malaprabha Valley’. In The Temple in South Asia, edited by Himanshu Prabha Ray, 15–37. Oxford
University Press, 2010.
Reddy, Srinivas. ‘Stallions of the Indian Ocean’. In Exploring Materiality and Connectivity in
Anthropology and Beyond, edited by Philipp Schorch, Martin Saxer and Marlen Elders, 98–115.
UCL Press, 2010.
———. Raya: Krishnadevaraya of Vijayanagara. Juggernaut Books, 2020.
Ritti, Shrinivas, and G.C. Shelke. Inscriptions of Nanded District. Yashwant Mahavidyalaya, 1968.
Rocher, Ludo. ‘Inheritance: Dāyabhāga’. In The Oxford History of Hinduism: Hindu Law: A New
History of Dharmaśāstra, edited by Patrick Olivelle and Donald R. Davis, Jr., 164–78. Oxford
University Press, 2018.
Roy, Kumkum. The Power of Gender and the Gender of Power: Explorations in Early Indian
History. Oxford University Press, 2010.
Sabin, Philip. ‘The Face of Roman Battle’. The Journal of Roman Studies 90 (2000): 1–17.
Sadhale, Nalini, and Y.L. Nene. ‘Sarameyavinōda in Mānasollāsa: Dogs for Recreation and Hunting.’
Asian Agrarian History 14, no. 3 (2010): 273–83.
Sanderson, Alexis. ‘The Śaiva Age: The Rise and Dominance of Śaivism During the Early Medieval
Period’. In Genesis and Development of Early Tantrism, edited by Shingo Einoo, 41–349.
University of Tokyo, 2009.
Sashikala, Ananth. The Penguin Guide to Vaastu: The Classical Indian Science of Architecture and
Design. Penguin Books, 1998.
Sastri, K.A. Nilakanta, The Cōl.as, Volume II (Part I). University of Madras, 1937.
———. Foreign Notices of South India: From Megasthenes to Ma Huan. University of Madras,
1939.
———. The Cōl.as, Volume I. University of Madras, 1955.
———. A History of South India from Prehistoric Times to the Fall of Vijayanagar. Oxford
University Press, 1997.
Sastri, V. Venkataraya. ‘The Latter Days of Bhoja “The Great”’. Proceedings of the Indian History
Congress 10 (1947): 260–68.
———. ‘The Rashtrakutas and Vengi and Cultural Contacts’. In The Rashtrakutas of Malkhed:
Studies in Their History and Culture, edited by B.R. Gopal, 171–80. The Mythic Society, 1994.
Sen, Tansen. Buddhism, Diplomacy and Trade: The Realignment of Sino-Indian Relations, 600–1400.
University of Hawai’i Press, 2004.
———. ‘The Military Campaigns of Rajendra Chola and the Chola–Srivijaya–China Triangle’. In
Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia,
edited by Hermann Kulke, K. Kesavapany and Vijay Sakhuja, 61–74. Manohar, 2019.
Sengupta, Nitish. Land of Two Rivers: A History of Bengal from the Mahabharata to Mujib. Penguin
Books, 2011.
Seth, K.N. The Growth of the Paramara Power in Malwa. Progress Publishers, 1978.
Settar, S., and Gunther D. Sontheimer, eds. Memorial Stones. Institute of Indian Art, 1982.
Shanmugam, P. ‘India and Southeast Asia: South Indian Cultural Links with Indonesia’. In
Nagapattinam to Suvarnadwipa: Reflections on the Chola Naval Expeditions to Southeast Asia,
edited by Hermann Kulke,
K. Kesavapany and Vijay Sakhuja, 208–27. Manohar, 2019.
Shekhar, Divya. ‘Inscription Stone Discovered in Hebbal Could Be Bengaluru’s Oldest’. The
Economic Times, 23 June 2018.
https://economictimes.indiatimes.com/magazines/panache/inscription-stone-discovered-in-hebbal-
could-be-bengalurus-oldest/articleshow/ 64708042.cms.
Shivanna. ‘Rashtrakuta Krishna I and Gangavadi (758–774 A.D.).’ Proceedings of the Indian History
Congress 38 (1977): 65–70.
———. ‘An Estimate of Dhruva’. In The Rashtrakutas of Malkhed: Studies in Their History and
Culture, edited by B.R. Gopal, 73–81. The Mythic Society, 1994.
Shivaprakash, H.S. I Keep Vigil of Rudra: The Vachanas. Penguin Random House India, 2010.
Shobhi, Prithvi Chandra Datta. ‘Kalyāṇa Is Wrecked: The Remaking of a Medieval Capital in
Popular Imagination’. South Asian Studies 32, no. 1 (2016): 90–98.
Singh, Arvind K. ‘Interpreting the History of the Paramāras’. Journal of the Royal Asiatic Society 22,
no. 1 (2012): 13–28.
Sinopoli, Carla M. ‘On the Edge of Empire: Form and Substance in the Satavahana Dynasty’. In
Empires: Perspectives from Archaeology and History, edited by Susan E. Alcock, Terence N.
D’Altroy, Kathleen D. Morrison and Carla M. Sinopoli, 155–78. Cambridge University Press,
2001.
Smith, David. ‘One Man and Many Women: Some Notes on the Harem in Mainly Ancient and
Medieval India from Sundry Perspectives’. Cracow Indological Studies XIV (2012): 1–16.
Soar, Michaela. ‘The Tīrtha at Ellora’. In Ellora Caves: Sculpture and Architecture, edited by Ratan
Parimoo, Deepak Kannal and Shivaji Panniker, Revised Edition, 63–79. Aprant, 2018.
Soundara Rajan, K.V. The Ellora Monoliths. Gian, 1988.
Spencer, George W. ‘The Politics of Plunder: The Cholas in Eleventh-Century Ceylon’. The Journal
of Asian Studies 35, no. 3 (1976): 405–19.
Srivastava, Prashant. ‘The Harsha–Pulakeshin II War’. In Select Battles in Indian History, edited by
K.K. Thapyal and S.N. Mishra, 126–30. Agama Kala Prakashan, 2002.
Stavig, Gopal. ‘Congruencies between Indian and Islamic Philosophy’. Annals of the Bhandarkar
Oriental Research Institute 81, no. 1/4 (1994): 213–26. http://www.jstor.org/stable/41694615.
Subbarayalu, Y. South India Under the Cholas. Oxford University Press, 2012.
Sundaram, R.V.S., ed. Śrīvijaya’s Kavirājāmārgam.. Translated by Deven M. Patel. Manohar, 2017.
Swamy, O.L. Nagabhushana, ed. The Sign: Vachanas of 12th Century. Prasaranga Kannada
University, 2007.
Tartakov, Gary Michael. ‘The Beginning of Dravidian Temple Architecture in Stone’. Artibus Asiae
42, no. 1 (1980): 39–99.
Thapar, Romila. The Penguin History of Early India: From the Origins to AD 1300. Penguin UK,
2015.
‘The Helgo Treasure: A Viking Age Buddha’. Irish Archaeology, 28 December 2013.
http://irisharchaeology.ie/2013/12/the-helgo-treasure-a-viking-age-buddha.
Trautmann, Thomas R. Elephants and Kings: An Environmental History. University of Chicago
Press, 2015.
———. Arthashastra: The Science of Wealth. Penguin, 2016.
Vahia, M.N., A.P. Jamkhedkar and Parag Mahajani. ‘Astronomical Orientation of Caves in Ajantha
and Ellora’. Journal of Multidisciplinary Studies in Archaeology 5 (2017): 806–12.
Vasudevan, Geeta. The Royal Temple of Rajaraja: An Instrument of Chola Imperial Power. Abhinav,
2003.
Vatsyayan, Kapila. The Square and the Circle of the Indian Arts. Second Edition. Abhinav, 1997.
Velcheru, Narayana Rao, and David Shulman. Classical Telugu Poetry: An Anthology. University of
California Press, 2002.
Venkataramanayya, N. The Eastern Cālukyas of Vēngi. Vedam Venkataraya Sastry and Bros, 1950.
Vijailakshmi, Usha R. ‘Tamilian Migration into Karnataka (The Period of Chola Conquest of
Southern Karnataka and the Consolidation of Power from 850–1279 A.D.’ Proceedings of the
Indian History Congress 66 (2006): 239–51.
Vogel, Jean Phillippe. The Goose in Indian Literature and Art. E.J. Brill, 1962.
Vogler, Roger. The Kailas at Ellora: A New View of a Misunderstood Masterwork. INTACH
Aurangabad and Mapin Publishing, 2015.
Willis, Michael D. The Archaeology of Hindu Ritual: Temples and the Establishment of the Gods.
Cambridge University Press, 2009.
———. ‘Dhār, Bhoja and Sarasvatī: From Indology to Political Mythology and Back’. Journal of the
Royal Asiatic Society 22, no. 1 (2012): 129–53.
Yazdani, Ghulam, ed. The Early History of Deccan. Vol. I. Oxford University Press, 1960.
 Acknowledgements

Lords of the Deccan has been three years in the making – three years in
which the manuscript and my understanding of the world evolved and
transformed alongside so many friends, mentors, loved ones and kind
strangers.
I do not know the names of these strangers, but this book would not have
been possible without their unthinking generosity to a young man in awe of
a history that they hold in trust for us all. Writing a book about such a
distant past is a journey in solitude, but I will never forget the people whom
I met in my travels through the Deccan. Their kindness added so much
humanity to the images that were taking shape through my research. Thank
you to the bus conductor who advised me on the best routes to take from
Pattadakal to Aihole; to the drivers and passengers of shared autos at
Aurangabad and Ellora; to the kind old gentleman at the Mythic Society in
Bengaluru, who was more than happy to look up an out-of-print publication
for me; to the brilliant and thoughtful scholars, photographers and
government officers who have made knowledge available online in
priceless archives for anyone to read. To everyone who has taken time out
to email, tweet or message me on social media and otherwise, inquiring
about the progress of my work. To the bravehearts who have defied lobbies
and interest groups to keep Library Genesis alive – Lords of the Deccan,
and hundreds of other such books, would not exist but for you. You are
truly among the unsung heroes of our generation.
To my parents and my sister, whose love has taught me so much; and to
my aunts, uncles and cousins; to Jasman Preet Randhawa, who was there
through my darkest times; to P., who entered and left my life so suddenly,
but filled it with such love in the brief time we had. To the Amazing
Humbug (S.G.), Kevin Fernandes, Sujata Shukla, Anand Ganapathy and
Aishwarya Narayanan. Your bottomless faith in me and your
encouragement to learn more, teach more and grow to be better has shaped
me profoundly. To Aditya Ramanathan and Parul Shanker, whose warm
home and enthusiasm for history taught me
so much. To Arjun Ullas, Keshav Rajendran, Karthik Nair, Krishna P. Unny,
Jajwalya Karajgikar, Mahathi G., Tejas A.P., Karthik
Malli and Ashwitha Jayakumar, who so unexpectedly became some of my
dearest friends, and whose kindness, humour and knowledge have inspired
me deeply.
To my research assistant Sarthak Sharma – you are among the brightest
and smartest young scholars I know. Thank you for your keen attention to
detail, critical mind, eclectic online presence, for your friendship and for all
your hard work. Thank you for following your deepest passion and for
living your life. To Nakshatra Soni and Shaaz Sheikh, two gifted young
artists whom I very serendipitously met on Instagram and who made such
marvelous cover art for this book. To Chiki Sarkar, Parth Mehrotra and their
team at Juggernaut. I could not have asked for a smarter group of people to
bring Lords of the Deccan to life.
To the good folk at the Takshashila Institution – Lt. Gen. Prakash Menon,
Pranay Kotasthane, Nitin Pai, Manoj Kewalramani, Shambhavi Naik,
Rohan Seth, Suyash Desai, Anupam Manur, Prateek Waghre, Sowmya
Prabhakar, Shivakumar, Sridevi and Lakshman. To the friends I made there
– Ram Ganesh Kamatham, Ganesh Chakravarthi, Madhav Chandavarkar,
Antara Krishnamurthy, Nidhi Gupta, Shibani Mehta, Hamsini Hariharan,
Manasa Venkatraman, Asawari Ghatage and Yazad Jal. I have learned so
much from all of you over the years, and I am most grateful for your
encouragement.
To everyone I have worked with at IVM Podcasts – Vinay Joshi, Jalasmi
Hathi, Tejas Shringarpure, Ekta Valecha, Alika Gupta, Abbas Momin,
Karthik Mohan, and of course, Amit Doshi. Your investment in my work
has allowed me to follow my deepest intellectual passions. It has led me to
meet so many people who have transformed me over the years.
To Dr. Deepthi Murali, Professor Daud Ali, Professor Srinivas Reddy,
Professor Cathleen Ann Cummings, Dr. Devika Rangachari, Professor Gil
Ben-Herut, Professor Kurush F. Dalal and Dr. Namita Sugandhi. As an
outsider to the academic world of history and historiography, your kind
words and selfless sharing of your thoughts has inspired me in my efforts to
bring the work of scholars such as yourselves to young people across the
world. Thanks especially to Dr. Deepthi Murali for her endless
encouragement and warm advice; to Professor Srinivas Reddy for our
discussions on Vijayanagara; to Professor Daud Ali for his kind words on
my work as a public history writer; and to Professors Cathleen Ann
Cummings and Kurush F. Dalal for generously sharing their photographs of
medieval artifacts and architecture with me.
Special thanks are due to Manu S. Pillai for making Lords of the Deccan
happen. I am grateful to him for introducing me to Parth Mehrotra at
Juggernaut; for his kind words on my many projects and for his feedback on
this manuscript. To Rana Safvi, Ira Mukhoty, Ambassador T.C.A. Raghavan
and William Dalrymple. Your gracious comments and encouragement on
this book mean the world to me. I am exhilarated to have received them,
and I earnestly hope that my work can live up to yours in the future.
And most crucially, thank you, Parth Mehrotra. As my friend and editor,
you have constantly encouraged me to make Lords of the Deccan the book
that I’ve always dreamt of writing. You have taught me so much over the
last few years, and I feel tremendously lucky to have had your frankness
and critical mind helping make this book, this labour of my heart and mind,
something that I can be proud of forever.
Anirudh Kanisetti
January 2022
 JUGGERNAUT BOOKS
C-I-128, First Floor, Sangam Vihar, Near Holi Chowk,
New Delhi 110080, India

First published by Juggernaut Books 2022

Copyright © Anirudh Kanisetti 2022
10 9 8 7 6 5 4 3 2 1
P-ISBN: 9789391165055
E-ISBN: 9789391165024
The views and opinions expressed in this book are the author’s own.
The facts contained herein were reported to be true as on the date of publication
by the author to the publishers of the book, and the publishers are not in any
way liable for their accuracy or veracity.
All rights reserved. No part of this publication may be reproduced,
transmitted, or stored in a retrieval system in any form or by any means without the written
permission of the publisher.
Typeset in Adobe Caslon Pro by R. Ajith Kumar, Noida
Printed at Thomson Press India Ltd