ROMILA THAPAR

The Penguin History of

Early India
FROM THE ORIGINS
T O A D 1300

PENGUIN BOOKS
 II
The Peninsula: Establishing Authorities
and Structures
C. A D 9 0 0 - 1 3 0 0

The Politics of the Peninsula

The late first millennium AD saw a changed situation in the Indian subconti-
nent. Regional states, earlier seeking recognition, were now taking shape
and the imprint of their identities was becoming clearer. Dynasties would
change but successor kingdoms retained a relatively consistent core area.
The trends that continued included some degree of political decentralization,
an emphasis on extending agriculture, the induction of erstwhile marginal
groups as castes, the interface between Vedic Brahmanism and the Puranic
and Shamanic religions, and new cultural articulations drawing on these
tendencies. Despite the appearance of similar patterns, each region retained
its own strong identity. That this is not a contradiction lies in the nature of
these regional forms.
The scene in the peninsula was dominated by the Tamil identity, forged
under the Cholas. This period of south Indian history with its impressive
corpus of inscriptions has been widely discussed in recent years. There have
been diverse theories and many new interpretations about the nature of the
Chola state. These enable one to speak of it as yet another classical period,
should one choose to use the label. The classicism of the Chola period drew
less on political authority and more on the institutions established at this
time, together with the articulation of cultural forms. In many spheres of
cultural life, whether of social institutions, religion, or the fine arts, the
standards established during this period came to dominate the pattern of
living in the south, and to partially influence the patterns existing elsewhere
in the peninsula. There was also an active intervention in south-east Asia to
a greater degree than before, in the commerce of the region and in its cultural
forms.
In the ninth century the Pallavas succumbed to a combined attack from
their southern neighbours, the Pandyas, and those tributary to them such

363
 EARLY I N D I A

as the Cholas. An agrarian foundation to kingdoms in Tamil-nadu had

begun more systematically with Pallava land grants to brahmans. The
subsequent expansion of newly cultivated areas encompassed pastoral
regions and forests. Brahman settlements established Vedic Brahmanism,
sometimes parallel to and sometimes intersecting with the religion of the
Vaishnava and Shaiva sects.
The Cholas emerged as the dominant power in the south. The core region
of their control - Cholamandalam - was the area around Tanjavur up to
the eastern coast, the Coromandal of later times. Their early conflict was
with the declining Rashtrakutas, whose place was then taken by a revived
branch of the Chalukyas, now known as the Later Chalukyas, rising to
power in the western Deccan. During this period the Deccan was divided
into smaller kingdoms of similar status, with some of which the Cholas
were occasionally at war. Political relations involved not only the Later
Chalukyas, but also the Yadavas of Devagiri (northern Deccan in the
region of Aurangabad) and the Kakatiyas of Warangal (in Andhra Pradesh).
Further south there was competition with the Gangas (south Karnataka)
and later the Hoysalas of Dorasamudra (in Karnataka), as well as with the
Pandyas of Madurai and the Cheras in Kerala. In their last years the Cholas
were weakened by the continued incursions of the Hoysalas and the Pandyas
into Chola territory.
Mention of Chola chiefdoms goes back to the turn of the Christian era in
the Shangam poems. Towards the middle of the ninth century, a chief
claiming Chola ancestry conquered the region of Tanjavur, the heart of
Tamilaham. He declared himself the ruler of an independent state, and
sought to establish his status by claiming descent from the Suryavamsha or
solar lineage. In AD 907 the first important ruler of the Chola dynasty,
Parantaka I, came to power and ruled for almost half a century. He secured
the southern frontier of the kingdom by campaigning against the Pandyas
and capturing their capital, Madurai. This brought him into contact with
Sri Lanka with which the Pandyas had had close relations. The ensuing
hostilities lasted several decades. The later part of Parantaka's reign saw
Chola defeat at the hands of the Rashtrakutas, who occupied some of the
recently acquired northern districts of the Chola kingdom. There followed
a period of thirty years in which a succession of weak kings brought about
a decline in the power of the Cholas. However, this situation was eventually
reversed. The Rashtrakutas in the Deccan were being harassed by the
Chalukyas, who had once been subordinate to them and were to be their
future overlords. In the confusion, Chola territory earlier lost to the Rash-
trakutas was gradually recovered. Chola power was firmly established with

364
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

the accession of Rajaraja I (985-1014) and of his son and successor,

Rajendra, which allowed about half a century for the Chola kingdom to be
consolidated and stabilized.
The reigns of both father and son were filled with extensive campaigns in
almost every direction. Rajaraja began by attacking the alliance between
the Cheras in Kerala, Sri Lanka, and the Pandyas, in order to break the
monopoly of trade held by these kingdoms with west Asia. The Arabs had
established themselves as traders on the west coast of India, and some had
been integrated into local society in Malabar and the Konkan. The Cholas
would have been aware of potential Arab competition in the south-east
Asian trade, and they tried to strike at the root of this competition by
bringing Malabar under their control. At a later date, Rajaraja conducted a
naval attack on the Maldive Islands, a staging-point in the Arab trade. The
Cholas, although unable to strike directly at the Arab trade, led a campaign
in Sri Lanka devastating the existing capital, Anuradhapura, and moved to
Pollonnaruva. At the same time campaigns against the rulers of the Deccan
states continued apace. Echoes of the old Pallava-Chalukya conflict over
the rich province of Vengi were heard now in wars between the Cholas and
the Later Chalukyas over the same area.
Rajendra I ruled jointly with his father for two years, succeeding him in
1014. The policy of expansion continued with the annexation of the
southern provinces of the Chalukyas, the rich Raichur doab and Vengi.
Campaigns against Sri Lanka and Kerala were also renewed. But Rajendra's
ambitions had turned northwards. An expedition set out, marching through
Orissa to reach the banks of the Ganges. From there, it is said, holy water
from the river was carried back to the Chola capital. Bringing back the water
through conquest symbolized ascendancy over the north. But Rajendra did
not hold the northern regions for long, a situation parallel to that of Samudra
Gupta's campaign in the south almost 700 years earlier.
Even more ambitious was Rajendra's overseas campaign, involving both
the army and navy against the kingdom of Shrivijaya in south-east Asia. It
has been suggested that this major undertaking arose from a desire for an
overseas empire. Had this been so, however, the campaign would have been
followed by Indian colonization of the coastal areas and an attempt to
conquer the hinterland. Since this did not happen, the cause of the war was
more likely a desire to protect Indian commercial interests. By the tenth
century merchants in China and south India had trading relationships. Ships
passed through the seas held by the kingdom of Shrivijaya (the southern
Malay peninsula and Sumatra) which controlled the Malacca and Sunda
Straits. The rulers of Shrivijaya realized that it would be more lucrative for

365
 EARLY I N D I A

local traders if the China-India trade had to terminate in Shrivijaya, with

local middlemen taking the goods to their eventual destinations. But
when Indian merchants in Shrivijaya territory were threatened this raised
the wrath of the Chola, who may have had his own investment in this trade,
and the result was an attack on Shrivijaya. From the viewpoint of its own
mercantile interests, Shrivijaya's interference in the China-India trade was
justified, but in this case military power decided the issue. The campaign
was successful in that a number of strategic places along the Straits of
Malacca came under Chola control, and, for a while at least, Indian
shipping and commerce were safe in their passage through Shrivijaya
territory.
The successors of Rajendra I turned their attention to conflicts within the
peninsula, primarily with the Chalukyas, reviving the competition to control
the province of Vengi. The old pattern of lightning raids into each other's
territory was repeated. A Chola raid into the heart of Chalukya territory
saw the sacking of the capital at Kalyani. This was avenged in 1050 by the
Chalukya king. Rivalry was less intense during the reign of the Chola King
Kulottunga I (1070-1118), perhaps because of kinship links between the
royal families, and this introduced a new element into the relationship. The
old enemies of the far south, the Pandyas, Cheras and Sri Lanka, meanwhile
remained hostile, but Shrivijaya was peaceful, although still smarting under
its defeat by Rajendra. This permitted a steady improvement in the com-
merce of south India and better communications with the Chinese, to whom
Kulottunga sent an embassy of seventy-two merchants in 1077.
Rajaraja I and Rajendra believed that their political status was higher
than that of Amoghavarsha the Rashtrakuta ruler or Vishnuvardhana the
Hoysala. The unobtrusive titles used by the early Chola kings were replaced
with high-sounding ones, such as chakravartigal (emperor, the equivalent
of the northern chakravartin). The cult of the god-king was encouraged
through the worship of images of the deceased rulers, together with the
building of temples as monuments to dead kings. This carries echoes of
Kushana practices. The royal household was run on an elaborate scale and
royal patronage was lavish. The political role of the purohita (priest) as
known to northern Indian politics underwent a modification in the Chola
system. The raja-guru (priest of the royal family) of the Cholas became a
confidant and adviser in all matters temporal and sacred. In their support
for Shaivism, the temple at Chidambaram was an important location, the
Shiva-nataraja (Shiva as the lord of the cosmic dance) was the icon and the
Periya Puranam was the revered text.
By the latter part of the twelfth century Chola ascendancy was waning.

366
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S A N D STRUCTURES

Neighbours were annexing territories at the fringes of the kingdom. The

power of the subordinate rulers in the Deccan increased as central control
weakened. Frequent campaigns had exhausted Chola resources and,
although they finally succeeded in establishing their supremacy, it was at the
cost of their own stability. Furthermore, the eventual breaking of Chalukya
power by the Cholas was to recoil on the Cholas themselves since it removed
the controlling authority of the Chalukyas over their tributary rulers. The
latter then set up independent kingdoms and made preparations to gnaw at
the Chola kingdom, prior to more substantial attacks.
Among these the most powerful were the Yadavas, the Hoysalas and the
Kakatiyas. The Yadavas kept mainly to the Deccan, and their contribution
to the final disintegration of the Cholas was less significant. The Hoysalas
and the Kakatiyas became active from the twelfth century onwards. The
latter, having won their independence from the Chalukyas, retired to enjoy
it, except for the periods when they were in action against the Cholas.
Kakatiya power became visible when the Kakatiyas moved from the fortress
of Hanamkonda to the plains in its vicinity, establishing their capital at
Warangal. The city and the power that it encapsulated was largely the work
of the thirteenth-century King Ganapatideva and his daughter, Rudramma-
devi, who succeeded him. The city was enclosed in a series of circular
fortifications. Its core, where the roads going in cardinal directions inter-
sected, seems to have been the location of the palace and the royal temple.
This would have conformed to the theoretical plan of a capital city. The
Kakatiyas wished to annex Vengi, which would have given them a substan-
tial seaboard, for they were well aware of the profits from the ports in the
area.
The main attack on the Cholas from the west came from the Hoysalas,
whom the Cholas were able to resist. But their older enemy, the Pandya
kings of Madurai, saw this as an ideal opportunity to revive hostilities. The
Chola strength therefore had to be diverted to two fronts, the western and
the southern. The rise of the Hoysalas is in many ways representative of
some dynasties of the Deccan from this period. The family began as hill
chiefs whose main source of revenue was brigandage, an unfailing source in
the higher regions of the plateau. Owing to the political confusion during
changes of dynasty, the hill people were eager to gain protectors. Their
support to the early Hoysalas enabled the latter to move down into the
plains, from where an even more reliable source of revenue - tribute - was
given by the people of the plains to buy off the attacks from the hill chiefs.
The regular payment of tribute established the legitimacy and authority of
the receivers. Tribute sometimes led to political loyalty, and the former hill

367
 EARLY I N D I A

chiefs gradually found themselves the possessors of small kingdoms, on the

basis of which they established a dynasty. Not all such dynasties survived,
the older kingdoms annexing the territories of some.
It was Vishnuvardhana who established a kingdom for the Hoysala
family. He ruled during the first half of the twelfth century, when the
Hoysalas were still in principle subject to the Chalukyas. The core of the
kingdom was at Dorasamudra, near modern Mysore, and Vishnuvardhana
prepared the way for political independence by consolidating his strength
around his capital. Vishnuvardhana is also remembered for his interest in
the teachings of the Vaishnava philosopher, Ramanuja, said to have per-
suaded the king to forsake Jainism for Vaishnavism. The consolidation
of the Hoysala kingdom was continued by Ballala II, the grandson of
Vishnuvardhana, and resulted in the domination of the southern Deccan by
the Hoysalas.
Chola power weakened in the thirteenth century. A Hoysala king claimed
to have rescued the Chola king who had been captured by his tributary raja.
This raja had attacked the kingdom and destroyed the temples to Shiva and
places sacred to Vishnu. According to another inscription, the Hoysala army
attacked villages, cut down the forest, burnt the ports along the sea and
seized the women. To the north, however, the Hoysalas met with opposition
from the Yadavas of Devagiri, who had also expanded their kingdom at the
expense of Chalukya territory. By the thirteenth century the Yadavas had
laid claim to Gujarat, but they could not hold this for long. The Yadavas
and the Hoysalas were to last until the fourteenth century, when new arrivals
in the politics of northern India, the Turkish and Afghan sultans of Delhi,
intervened in the affairs of the Deccan. The intervention led to other dyn-
asties and different political alignments. Further south, the Pandyas had
superseded the Cholas as the dominant power in the Tamil country, and
might have maintained this position in the subsequent century had it not
been for attacks from the rulers of the Deccan. The Pandyas remained local
rulers and subject to the changing politics of the region. Marco Polo claimed
to have visited the Pandyan kingdom in 1288 and 1293, and has left a vivid
description of the richness of the land and the prosperity of its trade.
Political developments on the opposite coast, that of Kerala, were of a
quieter nature. The Chera kingdom had alternating friendly or hostile
relations with the Cholas, but the political ambitions of its rulers became
apparent only during the reign of Ravivarman Kulashekhara at the end of
the thirteenth century. Although unsuccessful, he set out to acquire a larger
kingdom, building on the ruins of the existing southern kingdoms. But there
was little economic pressure to encourage territorial conquest, the Malabar

368
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

coast being naturally rich and obtaining an adequate income from trade
with west Asia.
It was during this period that another group of people from the west came
to India. Copper-plate charters of the Cheras, generally dated to the tenth
and eleventh centuries, granted land to Jewish traders, such as Joseph
Rabban - the earliest evidence of a Jewish community settling in India,
although tradition has it that a previous settlement in Cochin dates to the
early centuries AD. This parallels the chronological uncertainty of the arrival
of the Syrian Christians. It is possible that this later group had links with
the Jewish commercial community in Alexandria, or with the later trade
between Aden and ports in the southern part of the west coast, such as
Mangalore. The settlements in India were in pursuit of trade. The Cairo
Geniza records contain many letters of a personal nature, as well as per-
taining to business, written by Jewish traders active in commerce with India
towards the end of the first millennium AD. Indians were partners, or
representatives of Jewish merchants, and some of the latter spent time in
India and married locally. The area associated with this group of traders was
Kodungallur or Cranganore, identified by some scholars with the ancient site
of Muziris, a port that had been active in the Roman trade. The frequency
with which later traders from the west located their settlements in the
vicinity of those used by earlier traders from the west is quite striking.
Doubtless it was not coincidental, being connected with safe harbours and
access to cargo. The arrival of new groups was part of the pattern of trade
connections between the west coast and other commercially active parts of
Asia. Subsequent to the persecution of the Jews in Europe in later centuries
some Jews came to Kerala, already familiar from trading contacts, and made
it a part of the Jewish diaspora.
Further north in the western Deccan, the Rashtrakutas were being gradu-
ally unseated by their samantas, the Shilaharas. The Shilaharas took the
title of 'Lords of the West', presumably referring to their control over the
trade with west Asia. So firm was this control that in one case an association
of merchants was required to pay a regular and substantial sum to the royal
family, from the revenue of a village donated to it. One inscription suggests
that the commercial hinterland of the Konkan extended as far as Rajasthan.
The Shilaharas, wanting a monopoly over trade, discontinued the employ-
ment of Arabs as part of their higher administration.
Apart from the families constituting the main dynasties, there were lesser
rulers of smaller areas in the Deccan and some of these claimed descent
from the lineages of the main dynasties. These often began as subordinate
rulers of earlier kingdoms, or as landed magnates promoting the settling of

369
 EARLY INDIA

new lands through clearing forests, extending agriculture and constructing

irrigation tanks. Possibly some were chiefs of forest tribes, converted to
cultivation and caste, and were claiming high status. The status sought was
kshatriya, and the link was either with the Chandravamsha, or lunar line,
or the Suryavamsha, or solar line. Land rights also had to be claimed and
these required relationships with the ruling dynasty. The dynasty would
bestow title and rank on those performing administrative functions. Taxes
on a variety of produce were collected and some were transferred to the
central administration. This inevitably meant that peasants were frequently
under severe pressure to provide more wealth to the landholder.
Change in the agrarian system was noticeable throughout much of the
peninsula in the late first millennium AD. The pattern of change carried
some similarities with parts of the north, although the process in Tamilaham
in south India had variations. This may be the result of the more detailed
studies of this region. The economic backing given to political power in the
south is apparent, and to that extent makes an interesting comparative study
with other parts of the peninsula.

Structures of an Agrarian System

The Chola kingdom was initially described as a centralized bureaucratic

state, the standard description of virtually every large kingdom in India.
Recently, historians have been exploring the different ways in which states
were formed and the social and economic links to changes in politics. The
Chola state has been the subject of such attention. The most extensively
discussed have been the concepts of the segmentary state and, on a broader
level but with less intensive study, forms of feudalism for states of the
peninsula. As pointed out in Chapter i, the segmentary state model is
untenable largely because of its insistence on a dual sovereignty - political
and ritual. In the three zones envisaged - central, intermediate and peripheral
- political control gradually fades out to be replaced by ritual sovereignty,
moving from a smaller to a greater degree, from the first to the third.
Inscriptions are described as expressions of ritual sovereignty rather than
political sovereignty. Yet ritual encodes symbolism and ritual symbolism
need not indicate a weakening of political sovereignty. The changes that led
to the consolidation of power in the Chola state were reflected in ways not
C o n f o r m i n g to the segmentary state, such as in official titles especially at
higher levels of administration, the tendency to reorganize administrative
units territorially, standardization of taxation, and the gradual replacement

370
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

of chiefs by high-status officers. The theory of a segmentary state also

minimizes the impact of the merchant associations, some of which func-
tioned across the administrative units, as well as the trade that developed in
both the peninsula and south-east Asia to bring in a substantial commercial
revenue. Although the spread of the Tamil devotional movement and the
subsequent major changes in religion focused on the temple and sacred
centres, temples were not a-political as is evident from their rivalries and
competition for royal patronage when they functioned as major religio-
political institutions. This would not have supported the concept of ritual
sovereignty as distinct from political sovereignty.
Another theory argues that Chola wealth was acquired through cam-
paigns that were essentially plunder raids. This also has not found support,
in view of the impressive network of revenue collection from both agricultu-
ral and commercial activities. Virtually all campaigns at one level were
plunder raids, but such raids could not have sustained a state as complex as
the Chola kingdom.
The model of a feudal state has encouraged some forays into the social
and economic history of south India, but it requires further investigation.
The theories giving priority to agrarian change are those that emanate
from various hypotheses originally relating to feudalism. Although in many
respects they are closer to the evidence, the paradigm label of 'feudalism'
for this period of Indian history is still being debated. However, these
debates have led to a more precise understanding of social and economic
change. Terms such as feudatory or feudal cannot be equated with their use
in European history, where in any case the particular part of Europe under
discussion would have to be specified. There are some similarities and some
divergences. A clarification of the specific use of such terms in the context
of the data is always helpful. The nature of the relative autonomy of local
organizations, for example, is a theme receiving far more attention than it
did when it was thought that the ruler alone exercised political control. In
the currently emerging picture of agrarian and commercial structures, their
relationship to hierarchies of authority is a central question.
Much of the argument hinges upon the changes introduced through the
granting of land to religious beneficiaries or persons rewarded for their
services to the king. The land granted was often 'wet land' which was already
under cultivation and had irrigation facilities. Sometimes wasteland was
also given in a grant. Wherever land had to be cleared before cultivation,
and irrigation tanks constructed, this was done on a fairly substantial scale
that would have changed the landscape. The density of population in the
'wet zone', for example the fertile and well-watered Kaveri Valley, indicates

402
 EARLY I N D I A

a preference. In the later Chola period, grants were made in the dry zone
when presumably the extension of irrigation was more common. The ecol-
ogy of an area was significant to its economic use. The two landowning
groups were the brahmans and the velalas, the latter ranging from the
wealthy to those less so. These two categories had superior rights over
tenants and cultivators, although the range of tenurial rights complicates
any easy descriptive label. In some cases tenants could even be dispossessed
and earlier grants of land excluded. Artisans and cultivators formed yet
another category at a lower level.
The brahmadeya, or the agrahara, introduced Sanskritic culture which
included the norms of social organization as laid down in the Dharma-
shastras. The integration of the brahmadeyas and agraharas into existing
agrarian communities led to some innovations. Negotiations were required
between the existing peasants and the new landowners. Since the latter came
with royal backing, and often with advanced knowledge of organizing
agriculture and other technologies, they had a distinct edge over the existing
peasants or those recently converted to peasant status.
Brahmadeya donations remained unchanged in pattern from those of
Pallava times, as is evident from the Chola grants, such as the Anbil grant
of Sundara Chola recording the donations of land to the brahman Anirudha
Brahmadhiraja.

We marked [the boundaries of] the land thus defined by erecting mounds of earth
and planting cactus. The several objects included in this land - such as fruit-yielding
trees, water, lands, gardens, all upgrowing trees and down-going wells, open spaces,
wastes in which calves graze, the village site, ant-hills, platforms [built around trees],
canals, hollows, rivers and their alluvial deposits, tanks, granaries, fish-ponds, clefts
with bee-hives, deep ponds included; and everything else on which the iguana runs
and the tortoise crawls; and taxes such as the income from places of justice, the taxes
on [betel] leaves, the cloths from looms, the kanam [of gold[ on carriages, . . . the
old tenants being evicted, everything that the king could take and enjoy - all these
shall be made over to this man. He shall be at liberty to erect halls and upper storeys
with burnt bricks; to dig wells, big and small; to plant southernwood and cactus; to
dig channels in accordance with watering requirements: not to waste surplus water
but to dam it for irrigation; no one shall employ small baskets [for lifting such water].
In this wise was the old order changed and the old name and old taxes removed, and
an ekabhoga brahmadeya (land granted to a single brahman) under the name of
Karunakaramangalam was constituted.
'The Anbil Plates of Sundara Chola', in K. A. Nilakanta Sastn,
The Colas (Madras, 1955), p. 577

372
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S A N D S T R U C T U R E S

Knowledge of agriculture included assessing irrigation. This involved tanks,

reservoirs with sluices, canals and wells, which were built and maintained
with local expertise available in the villages. The more impressive irrigation
works of the Cholas, consisting of dams and anicuts on rivers, would have
been directly controlled by the state. But whether under local or central
control, irrigation involved organizing labour at both the village level and
across villages. Landowners would have had enough control over the peas-
ants under their jurisdiction to demand forced labour from them. Labour
over longer periods would largely have been from the landless or from the
lowest castes. Such subservient groups had existed for a long time in many
parts of the subcontinent, and would be a short step from what some might
call bonded labour or even slave labour, and slaves - atimai - are referred
to. Even if not identical to forms of slavery in other societies, the terms of
the bond were such that generally people were unable to regain their
freedom. Bonded persons were liable to be transferred together with the
transfer of land through a grant. The system would doubtless have been
intensified in the later period of Chola rule when individual ownership of
land became more frequent than the control of land by a village.
There are few instances of peasant revolts in Indian history. Sources of the
earlier period suggest that discontented peasants migrated to a neighbouring
kingdom or an area outside the jurisdiction of the current ruler, although
such migration was not permitted. This would be a feasible protest causing
a decline in the revenue of the ruler. But its feasibility was possible only
while land was available. From the early second millennium there are
sporadic instances of revolt, generally as objections to some aspect of
administration. Possibly the availability of easily cultivable land had
declined and the number of small kingdoms increased.
Village assemblies were crucial to Chola administration. Those living in
the usual peasant villages met in an assembly called the «r, whereas those
from the brahmadeya villages used the superior title of sabha. Royal officials
were present at the meetings of the sabha but do not appear to have played a
commanding role. Their participation in village affairs was more as observers
and advisers. This permitted continuity in local growth and development
without too much interference from political changes at the upper level, and
the degree of apparent autonomy at village level deserves to be underlined.
Large villages could be divided into wards, each with a smaller assembly
representing its households. Given the layout of the village and the tendency
for people in similar occupations to be located in the same wards, the latter
came to represent professional groups, such as carpenters, potters, smiths
and so on. The assemblies controlled production through consultation with

373
 EARLY I N D I A

the heads of the peasant families, the velalas, who were their members. This
necessitated discussion on a range of matters, including those of crucial
importance, such as the setting up of irrigation facilities. Rights in land were
insisted upon and among these were the kani, or hereditary rights.
Villages were grouped together within a nadu, a defined territory.
Agrarian organizations of the brahmadeyas, temples, and the commercial
associations linked to centres of exchange, such as the nagaram, functioned
within the nadu, although some had connections traversing the nadu.
Members of the associations of the velala handling agricultural products,
such as the Chitrameli Periyanadu who were referred to from the twelfth
century onwards, traversed the area more widely. The nadu was not an
autonomous peripheral area but was under central control. This enabled
the centre to regroup the nadus into units called the valanadu and the
mandalam, especially as the nadus were not of a uniform size. Such a
rearrangement was an indicator of control over the territories.
Agricultural expansion in the valanadu became associated with brahman
settlers receiving grants of land, as in the Tamraparni Valley of the far south.
Where the land was already under cultivation there had to be agreements
between the cultivators and the grantees, obviously to the advantage of the
latter. Other units such as mandalams could also be re-aligned to determine
revenue demands, administrative controls and the needs of cities in the
region. The population of the mandalam consisted of peasants, as well as
the settlements of forest and hill peoples in their proximity.
If the brahmadeya and the temple were important players in the restructur-
ing of the economy during the Chola period, it was not merely because of
their ritual authority, but was also a result of their administrative and
functional control over productivity. Ritual is important but does not exist
in a social void and, more often than not, is also tied to social and economic
realities and ambitions. The grantees themselves were beholden to the king
for the grant, and the king's officers were required to allocate temple
resources and audit temple accounts. As was the case with Buddhist monas-
teries, the temple complex could only survive where it had some control
over resources from agriculture, or from revenue generated by the regular
fairs and festivals which became surrogate markets. This necessitated temple
control over agriculture and irrigation, together with some participation in
commercial exchange.
The working of these assemblies differed according to local conditions.
•The ur was open to male adults of the village, but in effect the older members
such as heads of households took a more prominent part, some of them
forming a small executive body for routine matters. The sabha had the same

374
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S AND STRUCTURES

system and, in addition, could constitute smaller committees of any size from
among its members for specialized work. Election to the sabha sometimes
appears to have been by lot from among those who were eligible, though
amendments to the working of the sabha could be made whenever necessary.
An inscription from the temple wall at Uttaramerur (a brahmadeya village)
gives details of how the local sabha functioned. It dates to the tenth century
and reads:

There shall be thirty wards. In these thirty wards those that live in each ward shall
assemble and shall select each person possessing the following qualifications for
inclusion for selection by lot:
He must own more than one quarter of the tax-paying land. He must live in a house
built on his own site. His age must be below seventy and above thirty-five. He must
know the mantras and Brahmanas [from the Vedic corpus].
Even if he owns only one-eighth of the land, his name shall be included provided
he has learnt one Veda and one of the four Bhashyas.
Among those possessing these qualifications only such as are well conversant with
business and are virtuous shall be taken, and one who possesses honest earnings
whose mind is pure and who has not been on any of the committees for the last three
years shall also be chosen. One who has been on any of the committees but has not
submitted his accounts, and his relations specified below, cannot have their names
written on the tickets:
The sons of the younger and elder sisters of his mother.
The sons of his paternal aunt and maternal uncle.
The uterine brother of his mother.
The uterine brother of his father.
His uterine brother.
His father-in-law.
The uterine brother of his wife.
The husband of his uterine sister.
The sons of his uterine sister.
His son-in-law.
His father.
His son.
One against whom incest or the first four of the five great sins are recorded.
[The five great sins being killing a brahman, drinking alcohol, theft, adultery, and
associating with criminals.] All his relations specified above shall not be eligible to
be chosen by lot. One who has been outcaste for association with low people shall
not have his name chosen until he performs the expiatory ceremonies.
One who is foolhardy . . .

375
 EARLY I N D I A

One who has stolen the property of others . . .

One who has taken forbidden dishes . . .
One who has committed sins and has had to perform expiatory ceremonies of
purification...
Excluding all these, names shall be written on tickets for thirty wards and each of
the wards in these twelve streets shall prepare a separate covering ticket for thirty
wards bundled separately. These packets shall be put into a pot. When the tickets
have to be drawn a full meeting of the great assembly including the young and old
members shall be convened. All the temple priests who happen to be in the village
on that day, shall, without any exception whatever, be caused to be seated in the
inner hall where the great assembly meets. In the midst of the temple priests, one of
them who happens to be the eldest shall stand up and lift that pot, looking upwards
so as to be seen by all people. One ward shall be taken out by any young boy standing
close who does not know what is inside and shall be transferred to another empty
pot and shaken loose. From this pot one ticket shall be drawn and made the
arbitrator. While taking charge of the ticket thus given to him, the arbitrator shall
receive it on the palm of his hand with the five fingers open. He shall read out the
ticket thus received. The ticket shall be read by all the priests present in the inner
hall. The name thus read shall be put down and accepted. Similarly one man shall
be chosen for each of the thirty wards.
Of the thirty men thus chosen those who had previously been on the Garden
Committee, and on the Tank Committee, those who are advanced in learning and
those who are advanced in age shall be chosen for the Annual Committee. Of the
rest, twelve shall be taken for the Garden Committee and the remaining six shall
form the Tank Committee. The great men of these three committees shall hold office
for full 360 days and then retire. Anyone on a Committee found guilty of an offence
shall be removed at once. For appointing the committees after these have retired, the
members of the Committee for Supervision of Justice in the twelve streets shall
convene an assembly with the help of the arbitrator. The Committees shall be
appointed by drawing pot-tickets . . .
For the Five-fold Committee and the Gold Committee, names shall be written for
pot-tickets in the thirty wards (and the same procedure followed). One who has
ridden an ass (i.e., been punished) or who has committed a forgery shall not be
included.
Any arbitrator who possesses honest earnings shall write the accounts of the
village. No accountant shall be appointed to that office again before he submits his
accounts to the great men of the chief committee and is declared to have been honest.
The accounts which he has been writing he shall submit himself, and no other
accountant shall be chosen to close his accounts. Thus, from this year onwards as

376
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S A N D STRUCTURES

long as the moon and the sun endure, committees shall always be appointed by
lot...
. We, the assembly of Uttaramerur chatur-vedi-mangalam made this settlement for
the prosperity of our village in order that wicked men may perish and the rest may
prosper. At the order of the great men sitting in the assembly, I, the arbitrator
Kadadipottan Shivakkuri Raja-malla-mangala-priyan, thus wrote the settlement.
Archaeological Survey of India Report ( 1 9 0 4 - 5 ) , p p . 1 3 8 ff.

Other inscriptions refer to similar general procedures, though there are

variations in qualifications and requirements and in the sanction of expendi-
ture. The making of rules through amendments from time to time, and the
attempts to ensure that factions were kept at a minimum, are striking
features of this inscription. The assembly was summoned by the beat of a
drum and generally met in the precincts of the temple. Interchange and
co-operation among village assemblies was not unknown.
The village assembly was responsible for collecting the assessment for the
government and, where stipulated, the taxes on land and produce. In some
cases it was collected as a joint assessment on the entire village. In addition,
the assembly could make a levy for a particular purpose: for example, the
construction of a water tank. Such local funds were kept separate from
the taxes collected for the state. The activities of the assembly included the
keeping of records, particularly those pertaining to charities and taxes, and
the settling of agrarian disputes such as conflicts over tenure arrangements
or irrigation rights. The larger assemblies kept a small staff of paid officers,
but in the smaller assemblies most of the work was done on a voluntary
basis. Professionals were appointed for special purposes at the nadu level as
well. The higher officials were often drawn from among the better-off
velalas, thus creating a link between state and local administration that
enlarged the possibilities of upward mobility, incidentally providing a base
for those with political ambitions. The latter part of the Chola period saw
greater activity among revenue officials, with an increase in the number of
taxes. The degree of autonomy at village level prevented shifting relations
in the upper levels of the administrative and political structure from inter-
fering to any large extent with the routine life of the village.
Property rights varied as did the method of paying taxes, depending on
whether ownership was individual or whether a collective tax from the
village was preferred. Sometimes dues were remitted, partially or totally, in
return for some service, but this was generally on a very small scale, such as
a remittance of tax to the temple in lieu of fetching the water for bathing

377
 EARLY I N D I A

the deity. In later centuries this tax included military service. Unlike temple
service, involving proximity to the deity, military service could be performed
by low-caste groups. The recipients of brahmadeya grants, as well as temples
receiving grants, were treated as regular landowners where questions of
rights were involved. There was a sharp distinction between those with land
who paid tax, and those who did not, with the agricultural labourers
working for a wage. The distinction was largely that the labourer was
not included in the village assembly and could not hold a position of
responsibility in local administration. Many were employed on temple
estates, yet being of low caste they were not permitted entry into the temple.
Reclamation of wasteland and the clearing of forests were regular work
carried out by peasants and labourers, which was encouraged by the ruler
since it increased the land under cultivation. Cattle-raising had by now
become a subsidiary occupation except in the uplands.
Tax on land and its produce, whether in cash or kind, was a substantial
source of income for the Chola state, although other taxes were also col-
lected, such as those on mines, forests, salt, professional taxes, customs dues
and tolls, judicial fines and the equivalent in forced labour (vetti). The
assessment of land-tax was based on the quality of the land and the facilities
to irrigate it. Two or even three crops of paddy per year were regarded as
normal, though the yield varied. The tax could rise to one-third, which is
high by any standards, though in exceptional circumstances remission or
commutation was permitted by the king.
Permanent assessment of land for tax purposes was known, but was
apparently not usual. Assessment required land surveys at frequent intervals.
Land-tax, together with the local dues levied by the assembly or the temple,
must have been a strain on the cultivator for whom there were few alterna-
tives to paying the tax. He could in theory either appeal to the king for
remission or reduction of the tax if conditions were difficult at a particular
time, or else move away to a different area. The latter was an extreme step,
since mobility among cultivators was restricted even if land was available.
In the case of a village assessed as a unit, the return from the non-taxable
land was deducted from the total revenue of the village. Non-taxable
land consisted of the residential area, temples, tanks, irrigation channels,
areas where the artisans and outcaste population lived, and the cremation
ground.
In the Hoysala kingdom in Karnataka, the key figures in the rural areas
were the landholders, the gavunda/gauda, and the heggade, mentioned in
inscriptions referring to land transactions in villages, the maintenance of
irrigation, the collection of taxes and the work of the village council.

378
 THE PENINSULA: ESTABLISHING AUTHORITIES A N D STRUCTURES

References to them go back to Ganga inscriptions where they seem to have

been the heads of the families of the original settlers. Their standing and
status is indicated by their prefix. For example, the praja-gavunda, literally,
the gavunda of the people, was economically and socially of a lesser status
than the wealthy prabhu-gavunda, literally, of the lord. Their numbers
varied and were sometimes determined by their dual role of being both
representatives of the village and appointees of the state. Some judicial
functions were also included, such as the raising of a militia when required.
Initially the grants were made to brahmans and Jainas, whose activities
included the clearing of forest land and the reclaiming of wasteland to
extend cultivation. This also required the building and maintenance of tanks
for irrigation. The increasing numbers of peasants were drawn both from
the forest-peoples converted to cultivation and from migrations of peasants
from the Tamil area after the campaigns of the tenth century.
The evolving of the status of landholders followed a fairly common
pattern. Intermediary landholders and samantas changed the ecology of
the neighbourhood through expanding agriculture, secured through the
construction of irrigation tanks. Sometimes the grant of land was to the
family of a hero who had died defending village property. The identity of
the landholder was further emphasized by his establishing his family deity
in the locality. This was also part of the two-way process of introducing
new deities into Puranic Hinduism and superimposing existing Hindu deities
on local ones. The installing of the new deity was not solely a religious act
and should not be mistaken for ritual sovereignty. It accompanied the grant
of land and was equally significant as an intervention of administrative
rights and identity. With the coming of brahman settlements and the creation
of agraharas, the picture changed further. Occasionally there were some
confrontations between brahman settlers and existing cultivators and the
terms of revenue collection had to be sorted out. Peasant protests in the
peninsula are heard of from the thirteenth century, occasionally with
religious undertones. The balance between state administration and the
powers of local landowners, together with the requirements of cultivation
as a support to royal power, became characteristic of kingdoms at this time.
In an area such as Warangal (Andhra Pradesh), with its less fertile tracts,
major investments in clearing wasteland and building tanks and canals were
required to establish agriculture. Deserted villages were reclaimed by order
of the Kakatiya rulers and new villages were established, involving coloniz-
ation or encroachments into forests. This was actively encouraged by the
Kakatiyas and their tributary rulers through grants of land, especially to
brahmans.

379
 EARLY I N D I A

As a contrast to this, settlement in Kerala followed the fertile 4wet lands'.

The earlier brahman settlements of the mid-first millennium AD, said to be
thirty-two in number, expanded in this period. The alliance between rulers
and brahmans was also related to a large number of temples being estab-
lished, each of which controlled substantial land. Brahman control over
temple management meant that they controlled land both as individual
holders and as managers of temples. In some cases the two categories may
have been distinct if the brahmans associated with the Vedas held land from
earlier grants, whereas those who were only priests in the temples derived
an income from their functions in the temple. The management of temples
could have included wealthy non-brahmans, such as merchants and adminis-
trators. A hierarchy of tenants cultivated the lands of the temple and parallel
to this were the artisans, also employed by the temple. Agricultural activities
at various levels were controlled by a nexus between the brahmans and
those claiming aristocratic status, pre-eminently the caste of Nairs. The
wealth of the temple was such that many temples employed experts in the
martial arts, who acted as a militia in times of trouble. Control over
both the village assembly and the council managing the temple was tight,
particularly when supervising tenants and craftsmen.
References to local militias were now more noticeable, but so too were
increasing numbers of hero-stones. Some had a frame that suggested an
imitation of a small shrine, and the sculptured panels became more pro-
fessional. Their adornment underlined the status of the hero. There were
also occasions when the memorialized hero either came from a family of
landholders or else his kinsmen aspired to that status. In the former case,
his kinsmen's aspirations to political power could be supported by both
their landholding status and their relationship to a dead hero. The cult of
the hero was further elaborated. It was believed that a hero's death would
automatically lead him to heaven, and if his wife became a sati she was
deified. The emphasis on what the individual gained by these acts also
underlined the need for a village to be self-reliant in some situations. Mem-
orials in coastal areas recorded defence against attacks from the sea or at
sea. The depiction of ships on such hero-stones points incidentally to an
improvement in shipping technology compared to their depiction in earlier
sculpture. Hero-stones of this period could therefore be seen as comments
on the structure of power at local levels, particularly where the state did not
have a readily available presence or thought it unnecessary to mark such a
presence.
The tendency to hoard wealth was not characteristic of village life. Apart

380
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

from those in markedly rich villages, most village members had little wealth
to hoard. The average holding yielded enough to feed and clothe a family
with little surplus. Food was simple, rice and vegetables in the main. But
diet could change with caste. Brahmans, who had once eaten meat, were
now generally vegetarian. A non-vegetarian diet became customary among
some higher castes, and meat of all kinds was eaten by others, provided
it was affordable. Housing was relatively cheap, since the climate did not
call for elaborate structures. Nevertheless, the wealthy lived in well-
appointed houses. The richer members of rural society kept their wealth
employed through economic advantages in schemes to reclaim land, improve
irrigation or invest in trade. Equally, there was much religious merit to be
acquired by donations towards the building of a temple or the endowing of
a matha. Wealth was used in forms that would enhance the prestige of its
owners.

Towns and Markets

The nuclei of urban centres were generally formed by exchange centres in

rural areas where local produce was brought, administrative centres, places
of pilgrimage where large numbers gathered and goods were regularly
exchanged, or the locations of recurring festivals and fairs. Rural markets
were recognized by chiefs and administrators as potentially important to
generating revenue. The military camp, often set up as part of a regular tour
of inspection, continued to be a precursor to a town, particularly if the
location was repeated year after year. Towns that had existed prior to the
tenth century often grew as a result of expanding trade during this period.
In Karnataka and Tamil-nadu, towns referred to as pattanam began as
commercial centres, often maritime. Sometimes the main occupation was
specified in the name, for instance, Banajupattana, a town of traders. But
prior to this they could have been focal points of local administration or
places of pilgrimage. Shravana Belgola, for example, began as a pilgrimage
centre in the seventh century and by the twelfth century was an important
settlement of merchants that maintained a number of Jaina establishments.
Trading associations and professional merchants often had their beginnings
in rural centres of exchange or of administration.
Parallel processes occurred in the domain of the Later Chalukyas: Mer-
chants and wealthy landowners wielded authority in the administration of
cities. Elsewhere, weekly village markets sometimes developed into towns.

381
 EARLY I N D I A

On occasion the gavundas and heggades were also associated with positions
of urban authority. A designation of wider use familiar in many parts of the
subcontinent was mahajana, literally, the big men, used sometimes for
wealthy brahmans in their secular roles as landholders, and eventually for
those involved in commerce that sometimes coincided with urban adminis-
tration. Those who worked at the senior levels of urban administration were
recompensed by permission to keep a percentage of the levies collected from
householders. Towns went through periods of vicissitudes but from about
the tenth century there was predictable growth.
The market centre, known as the nagaram, where traders gathered, was
the focus for exchange of the produce of the region and also had links with
the temple. Administered by the nagarattar, it was also taxed, particularly
on its profits from overseas trade. Merchants and guilds paid tolls and
customs duties. Inscriptions carry details of these taxes, and of the levels of
tax collection which ranged from the local to the central. Inscriptions from
Karnataka record customs taxes on the import of horses, on commodities
made of gold, on textiles and perfume, and on produce such as black pepper,
paddy, spices, betel-leaves, palm leaves, coconuts and sugar. This was pro-
duce that came from various parts of the peninsula and was more than just a
regional trade. Corporate trading enterprise was particularly noticeable from
the eleventh century onwards. Inevitably, the presence of a hierarchy in the
bureaucracy was also recorded. In the transition from rural to urban, tax
concessions were sometimes introduced by the local administration but
were replaced with tolls, taxes and dues when commerce was profitable.
Overseas trade was among the strengths of the Chola merchants. Naga-
pattinam, Mahabalipuram, Kaveripattinam, Shaliyur and Korkai, on the
east coast, had elaborate establishments controlling the south Indian trade
eastwards. The port at Vishakhapattinam was named Kulottunga-chola-
pattinam, after the King. Trade to the east received encouragement from a
Cambodian mission to the Chola court in the early eleventh century. Mari-
time trade with China reached an unprecedented volume during these
centuries. It became a state monopoly in China, the Chinese government
not wishing to lose the income from it. Sung period sources from China
refer to the presence of Indian merchants in the ports of southern China.
Apart from trade, the earlier lively interest among Chinese savants in Indian
astrology and alchemy continued, with some startlingly exaggerated stories
in Chinese texts on the transmutation of stone and metal from worthless
.matter to valuable items by visiting Indians!
With the Mongols controlling central Asia, merchandise from China was
transported by sea, particularly from southern China to western Asia and

382
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

Europe. South India exported textiles, spices, medicinal plants, jewels, ivory,
horn, ebony and camphor to China. The same commodities were also
exported to the west, to ports such as Dhofar and Aden and, in addition,
Siraf received cargoes of aloe wood, perfumes, sandalwood and condiments.
Persia, Arabia and Egypt were the destinations of those trading with the
west, with Siraf on the Persian Gulf as an entrepot, and Cairo as well as
Alexandria involved in the trade across the Arabian Sea. Quilon, on the
Malabar coast, channelled Chera overseas trade westwards.
Marco Polo, who like many others at the time claimed to have visited
India, commented on the continuing and huge trade in horses which brought
vast fortunes to both the Arabs and the merchants of south India, who
between them had organized a monopoly of the import of horses. For a
variety of reasons, India never took to breeding horses of quality, possibly
because the climate, soil conditions and natural pasturage were not suitable.
This extremely expensive commodity therefore always had to be imported.
Marco Polo wrote:

L e t m e tell y o u n e x t t h a t t h i s c o u n t r y d o e s n o t b r e e d h o r s e s . H e n c e all t h e a n n u a l
r e v e n u e , o r t h e g r e a t e r p a r t o f i t , is s p e n t i n t h e p u r c h a s e o f h o r s e s ; a n d I w i l l t e l l

y o u h o w . Y o u m a y t a k e it f o r a f a c t t h a t t h e m e r c h a n t s o f H o r m u z a n d K a i s , o f
D h o f a r a n d S h i h r a n d A d e n , all of w h i c h p r o v i n c e s p r o d u c e l a r g e n u m b e r s of b a t t l e
chargers a n d other horses, buy u p the best horses a n d load t h e m o n ships a n d export

t h e m t o this k i n g a n d his f o u r b r o t h e r kings. S o m e of t h e m a r e sold f o r as m u c h as

5 0 0 saggi o f g o l d , w h i c h a r e w o r t h m o r e t h a n 1 0 0 marks of silver. A n d I assure y o u
t h a t this king b u y s 2 , 0 0 0 of t h e m o r m o r e every year, a n d his b r o t h e r s as m a n y .
A n d , b y t h e e n d of t h e y e a r , n o t a h u n d r e d of t h e m s u r v i v e . T h e y all die t h r o u g h

ill-usage, because they have n o veterinaries a n d d o n o t k n o w h o w to treat t h e m . Y o u

m a y t a k e it f r o m m e t h a t t h e m e r c h a n t s w h o e x p o r t t h e m d o n o t s e n d o u t any
veterinaries o r a l l o w a n y t o g o , b e c a u s e they a r e o n l y t o o g l a d for m a n y of t h e m t o
d i e in t h e k i n g ' s c h a r g e .

M a r c o P o l o , Travels (Pelican edition), p. 2 3 7

Allowing for Marco Polo's usual proneness to exaggeration, there is never-

theless some truth in these remarks. However, there were veterinaries
familiar with the medical treatment of horses and elephants, both important
to the cavalry and elephant wing of the army, and Sanskrit texts were
written on the subject. If it is true that horses died in such large numbers,
the reasons have to be sought elsewhere.
The continuing interference by coastal pirates was an impediment to
maritime trade. Shangam poetry had referred to pirates attacking Yavana

383
 EARLY I N D I A

ships and similar references continue. On the west coast piracy was some-
times organized by local chiefs to plunder the merchants active in the
Rashtrakuta kingdom. Hero-stones from the Konkan carry graphic rep-
resentations of the hero warding off the pirates. Such disturbances occurred
in many coastal areas of the Arabian Sea, the Bay of Bengal and the South
China Seas. In contrast, the less disturbed ports provided a rich potential
for trade and merchant associations made large profits. The Kakatiya king
issued a special charter in the thirteenth century to protect merchants trading
at the port of Motupalli on the coast of Andhra. The inscription in Sanskrit
and Telugu stated that only a fixed duty was to be taken from merchants
and their safety was to be guaranteed. According to Marco Polo, the exports
from Motupalli were diamonds and muslin, both of the finest quality. The
larger associations maintained their own force to protect the goods and the
traders, a tradition going back to earlier times.
Most production was for local consumption but trade, particularly over-
seas, provided an additional incentive to an existing market. Elephants,
horses, spices, perfumes, precious stones and superior quality textiles were
the commodities involved in large-scale trade, with metalware, jewellery,
pottery and salt (produced in salt pans from sea water) being of lesser
importance. Merchant associations, often described as guilds, controlled
much of the trade. They may not have traded collectively although they
belonged to an association. The more frequently referred to were the Mani-
gramam, Ayyavole, Nanadeshi, Nagarattar, Anjuvannam and Valanjiyar.
They did not all deal in the same items or trade in the same regions. The
Manigramam were smaller bodies than the Ayyavole, while the Anjuvannam
was particularly active in Kerala. Long-distance contacts and activities are
evident from the fact that the Ayyavole, originating in Aihole and constituted
of brahmans and mahajanas, was active in the trade of the Deccan and the
Chola centres. Such brahmans were sometimes from agraharas and had the
wealth to speculate in trade, ignoring the prohibition on brahmans being
traders except in dire necessity, or for that matter the prohibition on cross-
ing the seas. Members of this association were known as the 500 svamis of
Ayyavole and they claimed that they also protected the customary law of
their members. Intervention in the trade within the peninsula would inevi-
tably have meant some intervention in overseas trade as well. Such associ-
ations are referred to in inscriptions in south-east Asia. The power of these
associations in the economic life of the period is indicated by their access to
any region irrespective of the boundaries of kingdoms, and by the fact that
they could finance local projects, for example, the construction of a temple,
or provide a loan to the king.

384
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S A N D S T R U C T U R E S

Local merchant associations, the nagaram, were probably affiliated with

the larger ones that acquired goods at the place of manufacture to distribute
through an elaborate network. State support was not a condition of their
ventures into overseas trade, but if necessary and where the state felt that it
could interfere effectively on their behalf, as in the case of Srivijaya, it
came to their assistance. Even on such occasions, however, the purpose of
interference was not to acquire centres of raw material or markets but to
remove the obstructions placed on the trade. Kings and higher officers of
state may have invested in this trade, or else were provided with exclusive
presents by the guilds to ensure support where required.
Given the wealth and commercial ramifications of the large merchant
associations, it is curious that they did not aspire to greater political power.
The brahman element in these guilds was probably averse to challenging
the political authority of the king, since they derived their financial capital
from the land granted to them by the king, and the possibility of revoking
the grant could have been a threat. Urban structures supporting the potential
for an independent power-base, built on the ambitions of a range of pro-
fessional organizations, seem not to have asserted greater independence. By
now the concept of monarchy as the legitimate form of political authority
was strongly rooted. Moreover, many guilds had overseas interests and were
therefore dependent in the last resort on the military and naval strength of
the Chola state. Those who bestowed legitimacy on the system would
certainly have used every effort to curb the political power of the guilds.
Merchant associations were more powerful, however, in the smaller mari-
time kingdoms which were largely dependent on mercantile prosperity.
Barter continued in rural areas as a means of exchange, the major items
in these transactions being paddy, domestic animals and cloth. Copper coins
continued to be in circulation and used in markets. Gold and silver coinage
was naturally reserved for buying and selling goods of high value, for
long-distance trade and occasionally for the paying of taxes. The Cholas
minted some silver coins from time to time, but gold remained the metal of
high-value currency. The gold kalanju, sometimes valued in terms of the
gadyana of the Deccan, was used in expensive transactions and the kashu
for lesser ones. One of the advantages of merchant guilds having bases in
various parts of the country was that, as in earlier times, the use of promiss-
ory notes encouraged long-distance trade.
The circulation of gold coin was unrestricted, although some appear to
have been debased from the eleventh century onwards. However, this is a
debatable point since the gold content of coins varied from region to region.
Differences in weights and standards compelled the gold committee of the

385
 EARLY I N D I A

assemblies, for instance, to constantly evaluate gold and gold currency. The
use of coins encouraged an income from usury, by now a generally accept-
able activity indulged in even by brahmans and the managers of temples.
Presumably it was justified in these cases as helping in the propagation and
prosperity of the religion.

The Temple as an Institution

There is a striking increase in the number of temples built at this time. This
is unlikely to have been purely the result of a greater interest in worship.
The temple would have performed other functions as well. Temples built
from royal donations were not only closely connected to the court, but were
also perceived almost as surrogate courts. As such, they could draw on
resources from anywhere within the kingdom, for example, the Tanjavur
temple which received revenue from villages in Sri Lanka. This encouraged
a greater intervention of the court in local matters, although in many cases
physical distance from the court and the immanence of local authority,
required concessions to local opinion. The hymns of the Alvars and Nayan-
ars underlined the parallels between the deity and kingship, and by endorsing
the one they endorsed the other. Temples were visualized as palaces, and
rituals marking the daily routine of the deity imitated those of the king. The
temple received offerings and tribute and the service of the worshippers.
Temples built through royal donations could be located in the capital and
intimately associated with the court, as was the case with the Rajarajeshvara
temple at Tanjavur and the Rajendreshvara at Gangaikondacholapuram,
both built in the eleventh century and celebrating the reigns of kings. The
temple at Tanjavur had a commanding location and acquired the aura of
the leading ceremonial centre. Sometimes worship came to focus on the
king, thus helping to imprint divinity onto kingship. Ritual was another
channel of authority, at times subtle and at other times obvious, particularly
as it was not in itself sufficient and required reinforcing by more tangible
sources of power. The rhetoric of ritual was not intended in a literal sense
but as reflective of symbolic power. Puranic Hinduism did not require the
king to be the patron of the Vedic sacrifice, but, being symbolic of the deity,
he was thought to participate in the grace of the deity and this encouraged
the devotion of the worshippers. The king, therefore, did not always have
to proximate deity and in some situations it was sufficient if he was just the
ideal bhakta, a devotee. Doubtless the king and the court, irrespective of
religious conviction, were not unaware of the political edge of bhakti.

38 6
 THE PENINSULA: ESTABLISHING AUTHORITIES AND STRUCTURES

As in all such structures, the assumed presence of the deity converted the
temple into a sacred space where a relationship between the deity and the
devotee could be sought. But implicit in this space and in the rituals were
questions relating to authority and the establishing of rights and duties. The
temple of Puranic Hinduism provided social and political space for the
working out of such concerns over a large social spectrum, although the
lowest social groups were excluded. The managers of temples represented
political and economic interests and religious concerns, as is apparent
from a close reading of temple inscriptions. Royal grants were a form of
distributing wealth and acquiring supporters, even if the ostensible purpose
of donations was to please the deity. There was a continuous tradition of
giving gifts and donations - dana - and, although the form of the gift
changed, the intentions were similar.
Temples built from resources other than those gifted by the royal family
had also evolved into complex institutions, related not only to religious
requirements but also to fiscal, political and cultural needs. This can be seen
more easily in the functioning of the larger temples in rural areas. Land
owned by the elite was donated or, alternatively, land was purchased and
then donated to the temple. The fiscal role of the temple became even clearer
in later Chola times when both temples and merchants were the most
frequent purchasers of land. As an institution, the temple became the loca-
tion for many kinds of routine exchange, the focal point for many pro-
fessions to whom the temple gave employment either directly or indirectly,
a symbol of authority as a landlord who could intervene in the lives of rural
people, a rural bank, a channel of various forms of legitimation and, not
least, the focus of a particular sectarian religion. In rural areas, temples
were the locations of the village assemblies and of formal education for
upper-caste boys.
Whereas the brahmadeyas were often grants to brahmans who were
specialists in Vedic studies, the temples were closely associated with the
belief and practice of Puranic religions. The juxtaposition of the two brought
the rituals into proximity, but at another level demarcated the difference
between them. New rituals and deities, incorporated through the assimi-
lation of local cults, could be given respectability if introduced into temple
ritual and the creation of new myths. Cults that were refused such admission
were generally those associated with the marginalized groups, such as
untouchables and certain lower castes, and their places of worship were
segregated.
The architecture, plan and embellishment of the temple marked a
departure from the earlier monuments. Rock-cut temples gave way to

387
 EARLY I N D I A

free-standing temples, small to begin with but eventually reaching an

immense size, as dictated by institutional functions. By the late Chola period,
large temples were laid out in an extensive area with multiple courtyards,
incorporating many lesser shrines within the parameters of the main temple.
The presence of the brahmadeyas in the locality encouraged brahman con-
trol over temples even if the temple had been financed by an independent
grant. The location was frequently at a nodal point of exchange or trade, or
an administrative centre. Where it was the former it attracted merchants,
some of whom were associated with the administration of the temple in
addition to the brahman management. Small temples in the cities were built
and maintained through the donations of guilds and merchants and would
therefore be associated with a section of the city's residents. As in the
case of Buddhist stupa complexes, the wealthier temples also had a wide
geographical reach among their patrons and this was reflected in the
donations that they received from merchants and landowners.
The income of the temple came from the wealth and land acquired as
donation, from contributions by the village assemblies, from taxes that it
was permitted to collect as part of the grant, from offerings of devotees and
from its function as a banker in rural areas. The temple maintained the
priests who performed the rituals, as well as the record-keepers, accountants
and administrators who looked after its management. There were others
who serviced the buildings and guarded them. Professional herdsmen took
care of the animals owned by the temple. And those involved at a lesser
level with the ritual and the entertainment were the cooks, garland-makers,
musicians, dancers and devadasis. Many were encouraged to offer their
services free as a form of devotion, but others were also dependent on the
temple for their livelihood. Since temples had access to surplus resources
they attracted traders and craftsmen, among whom were bronze-workers,
stone-cutters and the makers of textiles.
The process of building a large temple, over many years, altered economic
relationships within the area with the provision of building materials,
labour, skilled artisans and those employed in its long-term maintenance.
When this extensive patronage towards maintaining a temple ceased, the
temple would fall into disrepair and even be deserted. Temples could be in
ruins because of the collapse of the authority responsible for maintaining
them, not necessarily only as a result of conquest and deliberate destruction.
The decline of a temple, therefore, as of other large monumental structures,
could be caused by a variety of changes.
The maintenance of temples compares with that of any large-scale insti-
tution. The temple at Tanjavur, which took almost a decade to build and

388
 THE P E N I N S U L A : E S T A B L I S H I N G A U T H O R I T I E S A N D STRUCTURES

was possibly the richest during this period, is said to have had an average
income of 500 lb troy of gold, 250 lb troy of precious stones and 600 lb
troy of silver, which was acquired through donations, income from taxes
and the revenue from about 300 villages. It also maintained temple staff,
consisting of about 600 employees, among which were the devadasis, 212
attendants - which included treasurers, accountants, record-keepers and
watchmen, 57 musicians and readers of the texts, quite apart of course from
the craftsmen of various categories (such as carpenters, braziers, goldsmiths,
tailors) and the many hundreds of priests who also lived off the temple. A
number of these were allotted land to live permanently in the vicinity,
and inevitably an urban centre emerged alongside the temple. It became
imperative for the temple authorities to keep the income flowing in. Temples
did this in part through financing various commercial enterprises and
through acting as banker and money-lender to village assemblies and similar
bodies, loaning money at the generally accepted rate of 12 to 15 per cent
per annum. In this the temples were now following the tradition of the
wealthier Buddhist monasteries.

Caste and Sect

The brahmadeya also acted as an agency of political integration, the well-

being of the elite being a primary concern. The system introduced the varna
hierarchy into an area where it may have been new or may have become
more dominant than before. Social distinctions of earlier times that drew
on kinship connections were giving way to caste. Nevertheless, although
caste was adopted as a form of social stratification there were adjustments
of exclusion and inclusion of certain social categories that were related to
regional forms and functions. The velalas, who in varna terms were often
equated with shudras, were second in importance after the brahmans, but
the velalas spanned a large economic range. The rest of society was gradually
shuffled into a caste hierarchy.
Vedic Brahmanism had declined somewhat but was still the privileged
practice of brahmans and kings. Its insistence on exclusivity confined it to
the upper castes. The centrality of the temple grew with the growth of the
Shaiva and Vaishnava sects. Those who managed the temple, especially in
rural areas, were inevitably the brahmans and the wealthier velalas. These
identities congealed into what had begun earlier as the brahman and the
shudra, forming the two main social groups. Temples and Shaiva monas-
teries also followed the pattern of Buddhist and Jaina institutions and sought

389