Reconstructing Chola Agrarian History
Parvathi Menon Interviews Y. Subbarayalu
*Foundation for Agrarian Studies, firstname.lastname@example.org
†Retired Professor and Head of the Department of Epigraphy, Thanjavur University, email@example.com
In the interview, Subbarayalu draws on inscriptions, the primary source material of this period, to discuss aspects of the agrarian history of Tamil Nadu, and particularly of the Cauvery delta, between 900 and 1279 CE. A great deal of what we know of modern Tamil Nadu’s ancient and medieval past, including its economic and agrarian history, lies inscribed on the walls of the numerous temples of South India. Subbarayalu discusses inscriptions as a historical source, the difficulties encountered in deciphering them and in drawing evidence-based conclusions from what they offer. The Epigraphy section of the Archeological Survey of India in Mysore holds nearly 50,000 inscriptions in the four south Indian languages. Of these, the Tamil inscriptions number 27,000, out of which only 11,000 have been published in properly edited volumes. Much history remains to be written from these, particularly the history of the Vijayanagar period.
The interview makes the following points. The first is that there was a significant growth and consolidation of private property in land (a change from earlier forms of joint ownership) over the 400 years of Chola rule (900 CE to about 1279 CE), first in Brahman settlements, and then within non-Brahman settlements. Secondly, the complex mechanism of land measurement, land gradation, taxation, and tax collection, plus the evidence of a standing army, would suggest that the Chola state was a centralised one, and not a “segmentary” state, as proposed by some historians. Thirdly, with the growth of private property, society became increasingly stratified, with landowners, tenant cultivators, agricultural labour, and artisanal groups becoming clearly identifiable components of a social structure that was also stratified by caste. Lastly, the inscriptions provide undeniable evidence of social protests by direct producer castes against the Chola state and landowning classes, Brahman and non-Brahman.
Significant gaps remain in our knowledge of the agrarian world of that period, Subbarayalu points out in the interview. Agricultural labourers, for example, are relatively “invisible” in the inscriptions, as are women, particularly those from the working classes and castes. The interview provides a summary of the current state of knowledge and scholarship on the agrarian history of ancient and medieval Tamil Nadu, and signposts paths for future research.
Besides that, I am collaborating with a team from the French Institute, Pondicherry, on a study on the history of science and technology over the last three or four hundred years. This also involves data from inscriptions, on which I am helping them. In the course of this study, I have come across other issues of possible future research.
My question to you is about the source base for the study of economic and social history in ancient and medieval south India. Tamil Nadu has a unique corpus of secular literature for the ancient period. But it is largely inscriptions that have been used for reconstructing the history of the medieval period. Could you discuss the sources of the period, what they offer, and how they complement and supplement each other?
The Sangam period is coterminous with the Satavahana period, and had somewhat similar state systems. These were not very developed states.
This is the early stage, that is, up to the third and fourth centuries CE. Then there is the period of the ethical literature, followed by the so-called Kalabhra interregnum, a term used by Nilakanta Sastri to refer to the period 400-600 CE.
The earliest inscriptions are from between 100 BCE and 300 CE, and largely found in rock shelters. These are called label inscriptions, because they each contain only one or two sentences. The main corpus of inscriptions begins from 600 CE. However, there is one inscription of note prior to this. This is dated to the fifth century and is called the Pulankurichi inscription. It is important because it offers the earliest inscriptional reference to land relations. Inscribed on a large rock face in Pulankurichi (Ramanathapuram district), it mentions some Brahmadeya villages [a Brahmadeya village is one in which the landowners are Brahmans] with references to the agrarian. I have mentioned it in my book, and Raghava Variar and I have published an article on it. For the first time we read of a king, Chendan Kuttran (who orders the inscription), and also of his son, Chendan. I suggested they could be Kalabhra rulers, although I now believe they must have been Pandya kings. Chendan claims a wide area as his territory. The inscription records a donation by an army commander of Chendan Kuttran to three temples. One of them is a devakulam (Siva or Vishnu temple), which is located on the small hillock above the rock. The second temple (called Taapatappalli, perhaps a Jain place of worship) is in Madurai, and the third (another devakulam) is at Vilamar on the Ramanathapuram coast. The inscription however only gives details about the first devakulam. The donor bought land from some Brahmadeya villages and donated the state’s share of the land tax for the upkeep of the temple. The king names a group of people that include the temple priests and some soldiers, and asks them to protect the temple and its property. He also states that only the tenants (kudumbi) assigned specifically for the purpose by a previous ruler continue as tenants after the donation. It is interesting that he uses the term kudumbi (the Sanskrit term for the Tamil kudi or tenant). We find here the existence of landowners (the temple) and cultivators as separate entities. The land was purchased from two or three Brahmadeyas, though the only village names that remain unmutilated on the inscription are Chittur and Kadayavayal (which still exist near Pudukkottai).
Compare the Pulankurichi inscription with the Pallava copper plate inscriptions in Prakrit of the third century and the Sanskrit copper plates of the fourth to sixth centuries. We come across the earliest creation of Brahmadeya villages in the Prakrit Pallava copper plate inscriptions.
In the inscriptions of the Pallava and Pandya kings of the seventh and eighth century, there are more land grant inscriptions than before. Even though these are small inscriptions, they refer to land grants either to temples or Brahmans. We have some bilingual (Tamil and Sanskrit) Pandya copper plates also in this period. After this, without a gap, we see an abundance of inscriptions that are carved on the walls of temples until the sixteenth century. In fact, their numbers increase and their content gets ever more detailed. Most of them are found on temple walls, and we get information related to society, land, and other related issues.
The Emergence of a New Social Formation in the Cauvery Delta
However, we see a lot of social and economic change over this period.
As far as society is concerned, the Pallavas started the process of bringing Brahmans from the Andhra area to create Brahmadeyas or Brahman-owned villages. This was a slow process of migration into the northern parts of Tamil Nadu and the Cauvery delta. New settlements were established for the Brahmans. The early settlements were small, comprising only a few families. By late 800 CE, the settlements became bigger, with some of them having a hundred households each. Thus, Brahmadeyas were created throughout the northern parts of the territory, or what was called Tondainadu, as well as in the delta region, or Cholanadu. The Cholas encouraged the migration of Brahmans even more vigorously than the Pallavas. By the time of Rajaraja, there were already about 250 Brahman settlements (constituting fifty per cent of the total known to exist by the thirteenth century) in the Cauvery delta. This was the first change in the agrarian regime. As Karashima argues, it represented the breakup of communal ownership and its replacement by private ownership, the Brahmadeya settlement acting as one of the catalysts for the growth of private ownership by individual Brahmans. When a settlement was created, the village was divided into shares (pangu), say of 100 shares or so, and one share was given to each Brahman. They possessed this individually, even though they may have cultivated the land communally. They also sold land. For example, there is evidence of outsiders from faraway villages purchasing land in a particular village and donating it to the local temple.
Karashima argues that such transactions encouraged private ownership in the non-Brahman villages as well. The increase of private ownership in non-Brahman settlements happened gradually, but by 1100 CE it had become fairly widespread. However, it was in the Brahmadeya villages that many transactions, such as the sale of land to outsiders, took place. Non-Brahmans did buy land in Brahmadeya villages, and in time, non-Brahman landowners even started to live in Brahmadeya villages, a practice that emerged very clearly in the twelfth century.
Caste and the Structure of the Village Community
Brahman villages had a few streets for wealthy non-Brahmans such as merchants and others. Cultivators lived in the village, too, but separately. In Uttaramerur, for example, two streets were set aside as commercial streets. Later, Vellalar (a non-Brahman “upper” caste) landowners are mentioned as living in a Brahman village, although they too lived separately from the Brahmans. Once a Brahman village was established, so too were the hamlets attached to it, but at some distance away from the main village. These hamlets were called pidagai. In the pidagai lived tenant cultivators, artisans, and other constituents of the village community.
Karashima has written about two donative inscriptions dating to 1014 CE from the temple that Rajaraja built in Thanjavur. These inscriptions record that Rajaraja assigned 40 villages as temple villages (devadana), and diverted government revenue from these villages towards the upkeep of the temple. The ownership as such did not change here; but the king’s share of revenues was henceforth paid to the temple.
The descriptions of the 40 villages are interesting. These were non-Brahman villages, not Brahmadeyas. There appears to have been a main village and some outside extensions called cheri. The landholders, in this case maybe Vellalar, lived in the main village, and the other castes in the cheri. The Dalit connotation of cheri came much later, although in these inscriptions a reference to some “untouchable” castes is also found. The inscriptions record castes and occupational groups such as agricultural labourers or paraiyar, toddy tappers or izhava, artisans or kammala, and four or five other categories. A few villages were commercial, and they were called nagaram. The revenues of the 40 villages were given to the temple. We know from other inscriptions that such caste-segregation was in place in Brahmadeya villages too at this time. However, it is obvious from the inscriptions on the 40 villages that by 1014 CE, or the last years of Rajaraja, such segregation was commonplace in non-Brahman villages as well.
Landowners and Tenant Cultivators
We do not have much information about the relations between landowners and cultivators in the eleventh century. The two classes are mentioned, but inscriptions portray a somewhat peaceful picture of the relationship. But when we enter the twelfth century, a clear picture emerges of growing antagonisms in the relationship between landowners and tenants. Tenant cultivators become increasingly prominent in the inscriptions.
The Chola kings tried to increase agricultural production because they needed to enhance revenues. They therefore required that land was cultivated well. Cultivation on temple lands appears not to have been always well managed, and the relationship between temples and the intermediaries, tenants and sub-tenants, became antagonistic, according to inscriptions from the reign of Rajendra, Rajaraja’s son. Temple lands were being deserted (perhaps due to intense warfare). We see royal orders from Rajendra to several temples in the northern part of the kingdom, with the same inscription appearing in several temples, warning those who had abandoned the land that their land would be given to other cultivators if they did not return within three years.
There are several inscriptions dated to the time of Kulottunga I [1070-1122 CE] and after, in the early twelfth century, that record the inability of tenants to cultivate the land, leading to landowners requesting the government to reduce taxes. (The reasons why tenants could not cultivate are not stated, but it could perhaps have been because of conditions of drought.) The state refused to give the owners any concessions, and landowners had to borrow from the temples. Unable to pay back the loans, the landowners had to sell their land to the temples.
Taxes, Rents, and Labour Services
The amount that the tenant had to pay as rent to the landlord was also referred to melvaram or kadamai. The same terms also refer to the land tax paid to the government by the landowners. It is only by the inscriptional context that we can tell whether the term melvaram/kadamai refers to rent or tax. Landowners, whether Brahman or Vellalar, had to pay the major tax, the melvaram/kadamai to the state. Cultivators had to pay rent (also called melvaram/kadamai) to the landowners, and, in addition, to perform kudimai services, or a bundle of labour services, such as irrigation management and other village duties, performed in lieu of tax.
The kudimai became more prominent in later Chola inscriptions. Early grants of the period usually refer to kudimai in the context of land donated to temples (devadanam). In such grants, the donor stated that the temple did not have to pay the tax (melvaram) on the donated land to the state. Who then paid this tax? The donor provided for such taxes by giving a sum of money, usually in the form of gold, to the local sabha (in the case of Brahmadeya villages) or the ur (in the case of non-Brahman villages), and it was from this sum that the local village committee paid land tax (melvaram/kadamai) on temple land to the state. We also find inscriptions that state that temple tenants were exempted from performing labour services (vetti/kudimai) on such donated land. This latter concession was made, perhaps, to encourage cultivation.
The relations between Brahman landowners and their tenant cultivators worsened because the state demanded more tax from landowners, and they, in turn, transferred that burden to tenant cultivators. Tenant cultivators then complained of the burden and of their inability to cultivate. From the twelfth century onwards, kings became increasingly ruthless in collecting taxes. There is another reason for the increase in the tax burden. From Kulottunga’s time [1170 -1122 CE], Chola territorial power decreased and there was a parallel growth of the power of officers who become local magnates in the areas north and south of the delta. Chola kings now depended on the delta for their revenue and they therefore intensified tax collection in the delta. There is inscriptional evidence of the increase in tax mainly in Brahmadeya villages. In big Brahmadeya villages we come across inscriptions in which, for the first time, landowners – Brahmans, but others, too -- complained to the king about the tax burden.
Service tenure lands (jivitams) given to officials, and military tenure lands (padaipattru), given to soldiers, also increased over time. (These officials were not paid in cash but were assigned land from which they could collect revenue.)
In the twelfth century we find landowners complaining that their tenants were being subject to increasing rents levied by service-tenure and military-tenure holders. The tenants complained about being unable to cultivate their lands because of exactions by officials and military officers. The king, in response to the complaints, ordered the abolition of military tenures as shown by inscriptions from the second half of the twelfth century. We do not know how long this order was in operation, but there is evidence of such an order. We can say, therefore, that there was increasing pressure to extract taxes on land during the twelfth century.
Agricultural labourers become more visible only during the Vijayanagara period, from c. 1400 onwards. A particular genre of literature called Pallu (relating to a Dalit community of labourers) that appears in the late Vijayanagara period, sixteenth century and after, throws some light on agrarian labour. The poetry describes the Pallu labourers and their relationship to the Andai, or lord. The authors of the poems are themselves Pallu men, and the stories include references to their wives as well. This poetry has been edited and discussed. I earlier mentioned that in the Periyapuranam, the author Chekkizhar writes about the Saivite Nandanaar, a paraiyar and leather worker by birth, and his early life. Nandanaar, in fact, was an ur paraiyar (a village paraiyar as opposed to a paraiyar agricultural worker) and was assigned the work of distributing footwear in the village, and of providing some services to the temple. He was given a small piece of service tenure land.
The first stage of conflict, mostly reflected in complaints by landowners to the king, also mentions tenants protesting against the burden of heavy rents. Such conflict is mentioned in several inscriptions in the decade between 1170 and 1180 CE. Tenant-cultivators refused to cultivate in several places, with landowners, Brahman and Vellalar, negotiating and trying to pacify them by saying that they would reduce rents. Tenants also refused to transport the king’s share of the tax in paddy from the threshing floor to the places it was required to be sent (the exact locations to which paddy had to be transported is not mentioned, but we can assume it was to granaries in local towns). The inscriptions describe the landlords’ insistence that it was the duty of tenants to transport grain from the threshing floor, and the tenants’ reply that it was not their duty but that of the landlords. The inscriptions describe these negotiations. There was also conflict over the sharing of labour expenses or aal-kuli for agricultural labour. Aal-kuli did not come under the kudimai labour services.
The process of sharing the gross produce took place as follows. The entire paddy crop was heaped on the threshing floor and measured. The portions for the state, landowners, cultivators, and village servants were then measured out. There were disagreements over who was to pay the aal-kuli. Was the payment to be made from the share of the landowners or from the share of the tenants? The landowners tried to bypass the original agreement and pass the payment of aal-kuli on to the tenants, while the tenants refused and told them to take the aal-kuli from their own share of the produce.
I described the first stage of social protests as shown in the inscriptions between 1170 and 1180 CE. The second stage emerged around 30 years later and was centred on the issue of labour services or kudimai. The inscriptions provide some interesting details on this. Differences between landowners and cultivators regarding kudimai flared up between 1230 and 1235 CE. These occurred mainly in the delta area or Cholamandalam (by this time the rulers had only nominal control over Tondaimandalam).
There were three or four categories of labour service. One was for the upkeep of irrigation resources. A second was the provision of labour for temple services during festivals. A third category included tasks such as the upkeep of the roofs of the common halls of the village (for which the workers had to supply thatch). A fourth category comprised labour services in the palace.
Temple services were important and included labour services for local temples as well as for big temples at Chidambaram and Thiruvavur, temples very dear to the Chola kings. The temples were often situated very far from the villages from which people and material had to be sent. Negotiations in this regard took place between landowners and tenant cultivators. Tenants complained about the burdens of such services, and stated that they could not send as many labourers to the temples as had been demanded. The protests resulted in landowners and the state agreeing to restrict the numbers of labourers sent to temples and to the palace to perform vetti. In respect of palace vetti services, the number of people to be sent by a tenant cultivator depended on the number of units of land he cultivated. Some inscriptions mention the numbers of people and the number of days of work allocated for particular tasks.
There are examples of such protests over labour services from inscriptions in several places on the east coast and up to Kumbakonam. There may be more on this subject from inscriptions that remain unpublished.
From 1300 CE until early 1400 CE, other groups, such as artisans, joined the idangai group.
Information on the composition of the idangai people is given in a series of Vijayanagara inscriptions, dated between 1429 and 1430 CE. The idangai people were now joined by artisanal groups such as kammalar and kaikkolar. The inscription refers to the idangai 98 and valangai 98 (the number 98 did not necessarily refer to the actual number of castes in each category). In the inscription, the spokespersons of the idangai 98 and valangai 98 state that they had come together to oppose three groups that oppressed them, namely, the state (the Vijayanagar ruler they refer to is Devaraya II), landowners (Brahman and Vellalar), and military officials. This particular solidarity inscription is replicated in around 12 or 13 temples, all of which lie between the Pennar and Cauvery rivers. They span the Tamil year Saumya (roughly from December 1429 to December 1430 CE). The inscriptions trace the progression of the struggle over one year, starting in the Tamil month of Chithirai. The earliest inscriptions in this series register the complaints of these (largely) direct producer castes against the state, landowners, and military personnel, and their tone is sharply confrontational. The tone becomes more conciliatory as the year passes, and finally records them celebrating the concessions they have won. In gratitude, the idangai-valangai people contributed to the temple on the walls of which the inscription is carved.
Also of great interest is the fact that the many jati in the two idangai-valangai groups are listed individually, although the group to which each jati belonged to is not recorded. Some are landowning castes, some are commercial people, some are artisans, and some are labourers. Paraiyar or agricultural labourers are mentioned in two categories, namely artisan paraiyar and cultivating paraiyar.
About what did those who issued the inscription complain? The arbitrary demand of land tax was one complaint, and the use of improper tools for land measurement was another. There must have been irregularities in such tools, as the complainants carved a sample of a measuring rod on the temple and urged the authorities to use rods of the specified length for future measurements.
According to a recent official estimate, the Epigraphy office (in Mysore) is in possession of 70,000 inscriptions, of which nearly 50,000 are in south Indian languages. As for Tamil, it has a collection of 27,000, out of which only 11,000 have been published in properly edited volumes. The case of Kannada is a little better, because the old Mysore State published around 10,000 inscriptional texts in the Epigraphia Carnatica series. Over the last few decades, the Mysore University has revised and published some of these in enlarged volumes.
One reason for the slow progress in the publication of inscriptions by the Epigraphy section of ASI in Mysore is the relative neglect over the years of this important task by ASI administrators in Delhi. When the Archaeological Survey of India (ASI) was expanded after Independence, and centres established elsewhere, the Epigraphy wing in Mysore did not get the attention and resources it deserved. Posts allotted to the Mysore wing were not filled up for one or other reason, and today it is languishing without senior epigraphists to lead and guide young epigraphists. In the last few years, the Epigraphy wing in Mysore has sought the help of some retired epigraphists to edit and publish a few volumes. This cannot go for long, as the number of senior retired epigraphists is fast decreasing.
If the urgent task of editing and publishing the huge collection is to be completed, young scholars must be trained by experienced epigraphists.
The Tamil Nadu State Archaeology Department has, from the 1960s, published nearly 7000 inscriptions. Even so, it has a big publication backlog. Here again, most epigraphists of the older generation have retired, and there are very few trained epigraphists to replace them. The urgent task of editing and publication of the huge collection requires a new generation of trained epigraphists.