AGRARIAN CHANGE IN THE PEASANT ECONOMY OF A PADDY PRODUCING VILLAGE IN THE SOUTHERN DRY ZONE OF SRI LANKA: THE GAMBARAYA SYSTEM IN TRANSITION

Since the decline of the dry zone based civilization around the 13Ih century, until the late 19"' century, the area known as present day Hambantota district remained desolate and deserted with small peasant settlements few and far between.l With the beginning of the restoration of ancient irrigation works in the district by the British in the latter part of the 19th century, large tracts of land under these schemes became available for the cultivation of paddy, in a region sparsely populated by a poverty stricken peasantry. Until the 1920s, the British colonial policy in Sri Lanka in alienating 'Crown land' for agricultural purposes under the restored major irrigation schemes was based on 'recouping irrigation costs' which the government had incurred. In the absence of local cultivators affluent enough to purchase even small plots of this land, they were sold to those who could afford to buy them: 'capitalists', mostly British government servants, professionals or proprietary planters, and later 'native capitalists' or 'small capitalists', and the 'middle c l a s ~ ' . ~ With the failure of' capitalist development of agriculture, the array of landowners that emerged over time, represented a whole variety of occupations except tha t of the cultivators. In the early stages government servants such as muhandirams, mudaliyars, headmen and Kachcheri officials dominated the land sales; later came the well to do teachers and officials, professionals such as lawyers, traders, lower level government servants and even the clergy (Herring, 1977:117-124).

Most importantly, except some of the local level native government servants such a s headmen, virtually all the new land owners came from outside the area or the district; many were from nearby coastal districts as Matara and Galle and even from far away Colombo and were not interested in settling down. Thus almost all the new landowners were absentee owners, often living far away from their lands. With the commencement of the cultivation of these lands, contractors and labourers came from the wet zone areas of the district and the neighbouring wet zone districts. Others, such as petty traders, and money lenders followed suit (Herring,ibid: 121).

THE GAMBARAYA SYSTEM: A DISTINCT FORM OF LAND TENURE
The absentee land-ownership necessitated employing on the spot manager-agents to supervise the cultivation. Peasants from the area as well as those landless who moved into the area as labourers became tenants on this land. Thus a distinctive land tenure system came into being in which the direct cultivator was typically, a share tenant, who owned no paddy land; a stratum of manager agents brgambarayas who stood in between the direct cultivator and the landlord came into being. Fairly large paddy holdings were extensively cultivated with hired labour employed during the peak periods, much of wage labour migrating from the neighbouring Matara district.
Surplus was extracted from the tenants under a series of deductions as return to fixed capital or land, and working capital, a t interest rates which were usually 100 percent per annum and some times even more. Gambaraya received a proportion of the crop share due to the landlord or a share directly from the harvest. I t was generally the gambaraya who provided the working capital, seed paddy, buffaloes for ploughing and threshing and even wage labour specially brought down for t h a t purpose from outside t h e district. The timely supply of a sufficient number of buffaloes was essential for cultivation and harvesting. Many gambarayas themselves became cattle owners in order to facilitate this. In addition to providing working capital, gambarayas gave advances to the tenants in paddy or funds to meet consumption needs, all a t very high interest rates. Thus, seasonally gambarayas collected a large portion of tenant's harvest as dues, charges and interest. The peasants were left wj t h only a small portion of the harvest which was often insufficient to last the season, forcing them to borrow from the gambaraya again for consumption needs setting in motion a typical vicious cycle of indebtedness and poverty. Virtually the entire economy of the sharecropping tenant was brought under the control ofthegarnbaraya leading to total e~onomic dependency of t h e former on t h e latter.
There were other factors, which enhanced this status of economic dependency. Share tenants owed their very existence to gambarayas or landowners that appointed them. Some of them lived on highland belonging to the landlords managed by thegambarayas. The dependency was socially manifested in t h a t not only the share tenant but even their family members were under obligation to the gambarayas. Thus the latter could impose his will on the former, compelling them to attend to unpaid chores and personal services, at tl~egambaraya's household. Abuse of this situation by gambaraya.~ was not rare. Some gambarayas also resorted to the use of physical force to impose their authority on share tenants (Herring, 1977: 140).

The Post -Colonial Developments
Thus the gan~baraya system gave rise to a peasant economy riddled ivith absentee land-ownership and share-tenancy, widespread landlessness, indebtedness and poverty among the cultivators which continued throughout the colonial period well into the post-colonial period (Ministry of Labour, Industry and Commerce, cited in Herring, 1977:133-135). As in the rest of the country, none of the post-colonial land reforms did bring about any fundamental change in the district's land tenure system. Following the Paddy Lands Act of 1958, many tenants in t h e district were forcibly evicted. The regulations of the act were circumvented: gambarayas registered themselves or their kith and kin a s s h a r e t e n a n t s or else landowners registered themselves a s owner-cultivators while continuing to. keep the direct cultivators not as legal tenants but as 'atha yata' (which literally means 'under-the-hand') tenants or "tenants-at-will".This new category of tenants did not enjoy any of the rights the legal tenants have and therefore were the most exploited and poorest among the share tenants (Herring, 1977: 138-140) and embodied an extreme form of hunger renting, where the tenant had no bargaining power due to his poverty.
However, the Act also had some positive impact. The formalization of tenancy under the 1958 Act was beneficial to the share tenant to the extent that it provided a legal base for tenancy rights. Following the 1958 Act, in the Hambantota District there h a s been some improvement in the conditions of the shareLtenancy. A survey (ARTI, 1974(ARTI, ,1975 of the paddy economy in the district carried out in late 1972, revealed that 82% of the share tenants were able to pay the rent at least partially in accordance with the Paddy Lands Act which stipulated the share rent as 1/4t" of the harvest or the fixed rent which was 12 bushels per acre for the district. However, 69%) of the tenants were obliged by the gambarayas /landlords to pay 114 share of the crop: only 3% were able to exercise the option offered by the Act to pay whichever is less of 114 of the crop or fixed rent (ARTI, 1975: 16). Land Reform Law of 1972 hardly affected the land tenure in paddy production in the district. Thus despite the land reforms, historical trends in the land tenure system in the paddy sector in the district have continued in t h e post-colonial period. Absenteeism, s h a r e t e n a n c y a n d landlessness were widespread: about 70%) of the irrigated area was cultivated by tenancy; nearly 50% of the paddy cultivators did not own any paddy land; about two-thirds of the landlords lived outside the villages where production took place, while about one hurth lived outside the district (ARTI, 197411975). Dias and Wickremanayake (1977) have documented in somewhat detail, factors leading to the evolution of the gambaraya system in the post-colonial period into a more comprehensive institution with an overwhelming influence on the peasant economy in the district, which far exceeded their original role of managing agents of paddy lands for absentee landlords. The introduction of modern technological changes further strengthened the position ofgambarayas vis-a-vis the share tenants while bringing in small owner cultivators also under their influence.
The role ofgantbarayas within the land tenure system itself had undergone changes and become diffused over time with thegantharayas assuming a multiplicity of roles: lease holder of land, landlord, and share-tenant. Through the combination of several of the above roles, many gambarayas came to exercise control over large extents of paddy land. The surplus accumulated was invested in diverse economic activities such as rice mills, lorries and cars for hiring, and trade, thus bringing the peasant cultivator under his influence at many points of contact. In this process gambarayas acquired considerable wealth and influence over time and gradually extended their influence and control into rural institutions such as Cultivation Committees or Cooperative Societies intervening in government programmes which threatened their interests, thus affecting the effectiveness of such institutions in improving the lot of the peasantry. Thus, the influence of gambarayas on every aspect of rural society was all pervasive (Dias and Wickramanayake, 1977: 144-146).
Thus, in their study conducted in the early 1970s Dias and Wickramanayake (1977: 15 1-152) concluded that the gambaraya system with its all pervasive influence on the rural economy in the district had by then 'achieved a high level of stability' with a 'self-perpetuating momentum' and still thrived and showed 'no signs of withering away'.

POLGAHAWELENA: THE DECAY OF THE GAMBARAYA SYSTEM
Polgahawelena, a village near the Tissamaharama township in Hambantota, which is the study area for the present paper, came under the dominance of the ganzharaya system during the first half of the century. Polgahawelena is located in the Tissamaharama AGA division in the eastern-most part of the district traditionally known as Magam Pattu, which has been the most fertile area of the district. The village belongs to the area identified as the arid zone, with an annual rainfall of less than 35 inches and is at an elevation of less than 600 feet. The area receives rain only through the northeast monsoon from November to January for the Maha season and in March and April for the Yala season. Due to irregularities in the patterns of monsoons and the distribution of rainfall in the area, crop failures under minor irrigation and rainfed conditions are not infrequent. Therefore, it is only under major irrigation with an assured supply of sufficient water that a successful paddy cultivation can be carried out without facing the risk of crop failure even in the Maha season when rains are usually abundant. In the Yala season, rain is insufficient to undertake paddy cultivation under minor irrigation and rainfed conditions. Paddy cultivation in Polgahawelena is carried out with irrigated water provided duringboth seasons from the Weerawila Wewa, one of the five major reservoirs fed by the Kirindi Oya.
With the commencement of cultivation in the early part of this century, the land-tenure system t h a t developed in the village was typically an extension of the gambaraya system which had already developed in nearby Tissamaharama since around the 1880s. A few absentee landlords from outside the district owned the entire extent ofland in the village managed by ganabara.yas while tenants with no tenancy rights cultivated the land for a share of the crop. Of an extent of about 420 acres of private land, (both highland and lowland) in the village, two landowners shared 150 acres between them, in holdings of 100 and 50 acres. All the other holdings were below 50 acres, most of them between 10 to 20 acres. As recollected by peasants in the villages, the gambaraya system in the village in the past shared the economic, political and social characteristics of the system, which were discussed earlier in this paper.
However, as revealed during our investigations in the 1980s, the peasant economy of Polgahawelena had undergone significant changes in the post-colonial period, leading to the decay of the gambaraya system in its original form of intermediary land agents. Nevertheless the second major function of gambarayas, t h a t is, of providing cash, material and services on credit to peasants with interest and mostly to be recovered in kind, for both production and consumption purposes was continued. The other two major features of the land tenure system, the absentee landlordism and share tenancy have continued into the post-colonial period, in the latter case however, with certain modifications beneficial to the tenants. These changes were however, accompanied by a significant development in the paddy economy.
There were several factors contributing to the deterioration of the ga,m,baraya system in its original form. Firstly, changes occurred in landlordganzbarayashare-tenant relationships. Secondly, new social forces, which a r e independent of and competitive with ganzharayas, benefiting from institutional support provided by the government, appeared on the scene through a process of social differentiation. Thirdly, political and social changes ushered in during t h e post-colonial period a t national level strengthened the hands of the peasantry against the gambarayas. The operation of these factors a r e analyzed below.

Changing Land Tenure
The 1958 Paddy Lands Act notwithstanding its failure to transform the existing land tenure system in the district, marked the beginning of changing relationships between landlords, gamharayas and share tenants. The formalization of landlordshare tenant relationship and the productivity regulations stipulated in the 1958 Paddy Lands Act was t h e first step towards improving the conditions of share tenancy. Before the Act, most of the share tenants worked without having a claim to any specific parcel of land. They could be transferred from one parcel of land to another or altogether evicted a t the will ofthe gambaraya or t h e landlord. Formalization of s h a r e tenancy eventually increased the security of those share tenants who managed to stay on without being evicted. The regulations also provided the basis for reduced share rents. The 1958Act also created an interest among the landlords in their land, and share tenants who cultivated their land, in many cases for the first time, bringing the two parties into direct contact. This new interest brought to the landlords an increasing awareness of the nature of the surplus produced from their land and its distribution among t h e parties involved in the process of production. With t h e increasing profitability of paddy production following technological improvements in the post-colonial period, some absentee landlords or their descendants migrated to the area and became local residents taking over the management of their land. Others, with the realization that their ga~nbarayas were fraudulently depriving them of their due crop share, started directly dealing with their tenants as against through the gambaarayas. Some landlords who were unhappy about the way in which gambarayas accumulated surplus at the expense of both the landlords and share tenants sold some of their land to their share-tenants, recovering the costs in instalments.
The expansion of the category of small owner cultivators, the availability of new avenues of income and increasing incomes due to the rising farm gate prices for agricultural products, the productivity increases in paddy and other agricultural produce resulted in a process of social differentiation among the peasantry giving rise to a group of middle peasants. Thus some owner cultivators or even share tenants cultivating relatively bigger extents of paddy land became independent and accumulated some surplus contributing to undermining the dominance of gambarayas.
The category of independent small owner cultivators gradually expanded over time. Apart from those peasants who bought over land they cultivated under share tenancy, there were others who bought paddy land in government auctions later.4 Despite the prevention by gambarayas of share tenants buying such land, some share tenants benefited from such transaction^.^ Developing land under abandoned small tanks in the jungle, renovating them in groups with government assistance enabled other peasants to join the ranks of owner cultivator^.^ Distribution of paddy lands acquired by the government under the 1972 Land Reform Act also contributed in a minute way t o enable some share tenants to become landlords.
In Polgahawelena, by 1980 the gambaraya who managed the largest holding of 100 acres of land had bought over the entire extent of the land from the owner and distributed it among his descendants, at least one of whom was now residing in the village. The latter was performing the roles of landowner, owner cultivator and manager of land on behalf of his family members. The descendent of the other largest holding of 50 acres, had come to reside in the village becoming the owner cultivator of most of the holding while managing some of the holding on behalf of his family members and continued to lease out some of the land on share tenancy. In both these cases managing family land was more a family affair than a mere landlord-agent relationship. There were several other instances of ganzharayas buying over land from absentee landlords, while several share tenants also had bought parcels of land they were earlier cultivating for absentee owners. As a result, on the paddy land in Polgahawelena, which originally was entirely under absentee ownership, by 1980 there were 38 owner cultivators of whom 33 were resident in the village. While the total acreage owned by the latter group of cultivators was 77 acres, the five cultivators residing outside the village together accounted for 15 acres. Five of the 'owner cultivators' cultivated 10.5 acres of 'Crown land' in total, which they considered their own. There were some that bought over land i n the adjoining village of Pannagamuwa in a government auction and became owner cultivators. Others cultivated small paddy holdings under renovated small tanks. Some others asweddumized small portions of their own highland. Under the Land Reform of 1972, four share tenants managed to buy over the land they had been cultivatirig as share tenants.
As already observed, there have been improvements in the conditions of share tenancy enabling peasants to retain a larger portion of the surplus they produced. In 1980, none of the tenants in Polgahawelena paid a rent higher than that stipulated by the 1958 Act. Of 23 share tenants in our sample of 50 paddy cultivators surveyed in 1980, 15 (or 65%) paid a fixed share rent of 12 bushels per acre while five tenants paid one fourth of the h a r~e s t .~T h e incentive of having a lower fixed minimum rent would have been an added incentive for the peasants to increase the output from their land.

Noii-Paddy Avenues of Income
Avenues of income other than paddy, both in agriculture and nonagriculture, expanded during the post-colonial period. Paddy cultivation continued to be the main occupation for many in the village. There were 14 landlords, 16 owner cultivators, 29 share tenants, 39 agricultural workers, and 15 tenants-at-will. By 1980, the occupational categories among the residents in Polgahawelena represented several sources of income both agricultural and non-agricultural other t h a n paddy cultivation. Of the 200 people employed in the village a t the time, 113 depended directly on paddy cultivation in the categories of landowner, owner cultivator, s h a r e -t e n a n t , agricultural worker, and tenant-at-will. The rest were employed in non-paddy sources of income. Among them were 11 hena cultivators, 10 non-agricultural workers, 11 engaged in businessltrade, 15 in teachinglclerical work or security work, and 13 in skilled crafts or technical categories such as carpentry and masonry or mechanical work. For 17 others main sources of income were activities such as milk production, hiring tractors, and fish vending among others.
Hena or slash and burn cultivation in the district, which experienced a decline during t h e British period, was revived in t h e p o st-colonial period.' With the advance of various development projects on existing land, the possibility of practising typical slash and burn cultivation diminished. By 1980 there was a total ban on issuing new permits for hena cultivation. Thus hena cultivation has been gradually transformed into a form of settled highland cultivation. The original subsistence form of production evolved into a mixture of subsistence and cash erop production a t varying degrees with the improvements in marketing facilities and increasing demand bringing higher incomes for some of the hena cultivators. In some hena areas where some form of irrigation facility is available, highland cultivation evolved into a highly capital intensive production of cash crops using modern cultural practices.
Hena cultivation by peasants in Polgahawelena was carried out closer to the village a t a distance of a few miles at the most, mainly in the shrub jungle area on both sides of the main road from Weerawila to Cultivation in Yala was limited in extent, intensity and crop selection. In Yala, drought resistant subsistence crops were predominant. There were some hena lands, which had better water conditions due to its location near a small tank, or had some form of irrigation facility such as pumping ground water or (illegally) diverting or pumping from a small tank or a waterway within the irrigation scheme. On such land aforrn of intensive highland cultivation of cash crops in both the seasons had developed. Hena cultivation brought cash incomes in varying degrees to those peasant families involved while it also provided vegetables and other food crops to supplement the staple diet. For those who engaged in intensive highland cultivation on herza land with irrigation facilities it was a source of significant surplus production. The land was 'Crown land' and hence did not involve a rent; a t most only an annual permit fee, of Rs. 10 in 1980." Non-agricultural workers found mostly casual labour work in activities such as gemming, loading sand from the river, making bricks, in construction work and in several government agencies in the area. In 1980, the wages for non-agricultural labour varied from about Rs.12.50 to Rs.75 a-day the latter rate being for a specific amount of work in activities such as brick making and loading sand from the river.
Those who engaged in businesshrade derived their income from rice milling, rice trade, grocery stores, small tea boutiques and selling betel etc. a t the nearby small market a t Debarawewa or fairs a t Pannagamuwa and nssamaharama. In 1980, there were four rice mills operating in the village while the work on one more rice mill was being completed. Of the above five activities, the former three would have generated a surplus in varying degrees, the latter two, particularly the last activity did not generate a sufficient income to make even the bare minimum subsistence for an average family, Those who were employed in regular white collar and blue collar jobs mostly in the government sector such as teaching, cooperative stores management, clerical work and security jobs with an assured regular monthly income were entitled to other benefits which come along with their employment such as low-interest loans. Savings from employment provided them not only interest free cash for investing in paddy cultivation or other economic activities but also provided a form of security against hazards such as crop failures. This situation offers the members of this occupational group opportunities for accumulating a surplus.
Despite the lack of facilities for rearing cattle in the area, lately milk production h a s been a n important supplementary source of income for the peasantry in the district. However, until the early 1970s when t h e government set up several milk collection centres, ganzbarayas and other large-scale cattle owners dominated the production of curd, the main form in which milk was sold on a large scale. With the setting up of milk collection centres, peasants with smaller herds of cattle could earn an assured and regular monthly income from milk production. Residents of Polgahawelena have been among those who benefited from the establishment of a milk-collecting centre in nearby Debarawewa in the 1970s.
Increasing paddy yields and prices of paddy and other food crops brought higher incomes to peasants in general. Paddy yields at the district level increased from 41.5 bushels per acre in 1966 to 69.05 bushels per acre in 1980, a 66 percent increase over the period. Areas under major irrigation such as Weerawila and Yodha Kandiya reported a much higher yield of 87.5 bushels per acre in 1980. In Polgahawelena in 1980181 Maha season, even with apparent under reporting of yields the average paddy yield of a sample of 50 farmers were 66 bushels per acre.1° However, 14 out of 50 farmers reported yields higher than 70 bushels per acre, while eight of them received yields between 91-100 bushels per acre. One farmer reported a yield of 12:L bushels per acre. Guaranteed price of paddy at the national level increased from Rs.12 per bushel in 1960 to Rs.40 per bushel in 1980, while the market price of paddy in 1980 was much higher, being Rs.51 per bushel. In Polgahawelena in 1980181 Mah,a the price of paddy in the market went up to Rs-70 per bushel. The prices of most subsidiary food crops have recorded increases in the post-colonial period. In the 1970-77 period, hena cultivators in Polgahawelena also benefited from the better prices and higher demand prevailing for subsidiary crops.
The rice ration, and subsidies on other essential food items which were available up to 1979, throughout the major part of the post-colonial period and the Food Stamps Scheme made available in the 1980s have been always there for poor peasants to fall back on, while for better-off peasants they have provided additional incomes. There were altogether 49 families in the village who received Food Stamps.
Institutional support provided by the government through Cultivation Committees and other agencies was beneficial to those sections among the cultivators who were in advantageous positions. Those who were financially stable enough to continue to borrow from institutional sources and repay seasonally even in the event of a crop failure, benefited from low-interest institutional loans. Other institutional support services, which also benefited sections of cultivators, were tractor pools, and hiring of other equipment, and the provision of other inputs a t reasonable rates and low-interest institutional credit for purchasing tractors.ll By 1980, most cultivators in the area were producing their own seed paddy thus avoiding the payment of high interest rates for borrowing seed paddy. A few bought their seed paddy from the government Agrarian Services Centres.
Highland and home garden cultivation brought some additional income to peasant families mainly in the form of supplementary food crops and cash for subsistence. The crops were perennials such as coconut or mango, plantains and vegetables. However, in 1980 there was no systematic cultivation of highland crops of any significant scale in the village. Lack of irrigation, working capital and poor soil conditions appear to be some of the reasons for this situation.

Political Changes
Political changes in the latter part of the colonial period and post-colonial period ushered in a slow but expanding process of democratization which opened up some opportunities for peasants, owner cultivators in particular to improve their social status and to improve their living standards. The introduction of universal adult franchise in 1931 elevated the position of peasants into that of voters, sought after by politicians and their local level agents such as gambarayas. The political and cultural changes initiated by the post-1956 regime opened up new avenues of employment for members of peasant families in the expanding public sector. More importantly the changes elevated the social position of peasantry from t h a t of serf-like subservience to gambarayas, landlords and the other elite groups to that of an economically productive social group of political significance a t the national level. At the local level, peasants got an opportunity to further their interests by affiliating themselves with one of the political parties. While share tenants who were under the grip ofgambarayas were politically also controlled by them to a considerable extent, those political parties which took up the cause of the share tenants created an increasing political awareness of their rights among share tenants strengthening their hands against thegambarayas. In Hambantota, political support for the rights of share tenants was evident in the demonstrations organized by progressive political groups in the district in support of the Paddy Lands Act of 1958. + In addition to money lenders, traders, tractor owners, transport agents and other middle persons who provide inputs, services and capital to peasantry in competition with the traditional ganzbarayas, a group of new rich, mostly coming from outside the area or the district also appeared on the scene. Cumulative and combined effect of the above processes on the land tenure and the pattern of income, together with political and social changes have ushered in a process of social differentiation among the peasantry undermining the dominance ofgambarayas in the district in varying degrees and giving rise to a middle peasantry independent of the control of thegarnbarayas.

Social Differentiation
In 1980, based on the relations of production and exchange of the nuclear family unit, and the head of the household in particular, the following social classes could be identified in Polgahawelena: rural wage labourers, poor-peasants, middle peasants, petty bourgeoisie, middle capitalists, and rural entrepreneurs.12 Rural wageworlcers do not have possession or ownership of means of production and their livelihood is predominantly secured through the sale of their labour. Seventeen families in Polgahawelena belonged to the category of rural wage worker^.^.' Despite the daily wage rates they receive, rural wageworkers in Polgahawelena do not have sufficient opportunities for work throughout the year.14 They get work in agriculture for a maximum of about half a month to two months during the peak periods, while other types of wage employment is not frequent and widespread in the area. The payments they receive in paddy for work during the harvesting enables most of them, especially smaller families to cover at least part of their subsistence food needs. However, large families find even survival difficult with their total incomes. Therefore, the living standard of most of the agricultural workers in Polgahawelena is lower than that of the poor peasants. The mobility of agricultural workers in Polgahawelena is minimal. The maximum distance they travel in search of work is a few miles while most of thein do not go beyond neighbouring villages. Most of these families do not own even a home garden and they live in small huts, many in poor conditions, on 'squatted' Crown land.
Poor peasants own or possess some paddy land however in minute holdings insufficient to provide the bare minimum income for subsistence and hence have to sell their labour to other cultivators and for non-agricultural purposes as well. They have the expectation of expanding their area of land by some means to a level which would elevate their position to that of middle peasants. Twenty-four families in Polgahawelena belonged to the category of poor peasants. Sixteen of these families had some paddy land ownership, however mostly in minute holdings, or with poor soil conditions or holdings at tail end of the irrigation system which do not receive sufficient water to carry out a successful paddy cultivation. Many poor peasants in the village became cultivators by gradually asweddumizing part of their highland even though these lands do not have a proper supply of irrigation water, as they are located outside the existing irrigation system. It shows their keenness to own some paddy land and somehow engage in paddy cultivation. Due to the peasant nature of their economicactivity, the minute scale, dependence on family labour and meager family incomes, there were eight other non-cultivator families that would fall into the same category of poor peasants. Among them were those who sell vegetables or betel on a very small scale a t the market or the fair, a fish vendor, and a person who engages in the illicit sale of liquor. Members of some of these families themselves engaged in wage labour to supplement their incomes.
The fifty families of middle peasants were numerically the largest social class in Polgahawelena. Middle peasants are those peasants and craftspersonsl%ho engage in production relying primarily on family labour with the least exploitation of others. Owner cultivators, landowners, share-tenants, hena cultivators and milk producers, and craftspersons with middle level incomes form the bulk of this class in Polgahawelena. Unlike the landowners of the petty bourgeoisie and middle capitalists who do not work on their own land, middle peasants cultivate their own medium size land holdings or manage their herds mostly with family labour supplemented with wage labour when necessary. Those share-tenants cultivating relatively big paddy holdings and who do not sell their own labour would also fall within this category. Economically they can earn their subsistence needs or even a small surplus without being compelled to sell their labour to supplement their income. Others who would also fall into the same economic status as t h a t of t h e middle peasants are the self employed such a s bicycle repairmen, technicians, carpenters and masons. These people have to mostly find working capital and services necessary for their work from the petty bourgeoisie and middle capitalists. Some ofthem, in economic difficulties have leased out their land ta members of the petty bourgeoisie. The predominance of medium size paddy holdings in the a r e a appears to be the reason for the largeness of this group. Families belonging to this class in the absence of any other avenues of income may easily get into debt with the slightest disruption of their economic activity due to events such as a crop failure or an unexpected event in family life and be pushed down to the category of poor peasants.
There were 22 families in the village, which would fall into the category of the petty bourgeoisie in Polgahawelena. The existence of the petty bourgeoisie in Polgahawelena is based upon petty productim and exchange in agriculture and commerce. A considerable number of them combine ownership of means of production and exchange with salaried incomes. Most of t h e petty bourgeoisie families in Polgahawelena own paddy land holdings varying from 1 to 5 acres. Several of them own two-wheeled tractors which are hired to peasants. A considerable number of these families have members who a r e salaried employees. One household hired a two-wheeled tractor bought with savings its household leader earned working in the Middle East. Some are medium scale traders, shop owners or large scale cultivators of hena. Some lend money and take land on lease. Most of the salaried employees cultivate their paddy land by using wage labour or else employing a share tenant. Some who have come from outside the district for employment in the area receive income only from their employment.
Members of the middle bourgeoisie own or control relatively large paddy lands, instruments of bourgeoisie production such as farm machinery and means of exchange and transport and make a considirable surplus from their economic activities. There are eight families in Polgahawelena who would fall into this category. Four of them are rice mill owners; four of them own four-wheeled tractors; six lend money; some lease in land; some rent out their land to share tenants; some had undertaken large scale hena cultivation. None of them come from peasant origins. They reinvest their capital in various non-agricultural avenues such as means of transport and trade and gemming.16 It was seen in the 1980s that a small class of rural entrepreneurs was appearing in the Tissamaharama area. Even though the economic basis of this class lies in management and ownership of land, they were now involved in agriculture related entrepreneurial activities which bring them higher profits. With the money earned through these activities, they expand into other non-agricultural avenues of investment. Even though there was only one person belonging to this category in Polgahawelena there were several of them in nearby villages in the Tissamaharama area seen in early 1970s, who while owning and managing paddy invested in rice milling and means of transport (Dias and Wickramanayake, 1977) who can be considered as rural entrepreneurs in formation.
The rural entrepreneur in Polgahawelena controls the largest single amount of paddy and highland in t h e village totalling about 100 acres. He himself was registered as an owner cultivator for 9.5 acres, and had rented out 29 acres of paddy land on share tenancy. He is not ofpeasant origin. Heinherited the land h e owns today from his father who originally managed the same land as gambaraya and later bought it over from the owner. The father was a cattle owner who became a gum baraya. I t was with capital accumulated by functioning as a gambaraya he paid for the 100 acres. His son the present manager of the 100 acres, combines several roles: landlord, owner cultivator, and 'share-tenant', rice mill owner, contractor and supplier of rice to the government. He ow~ls a rice mill equipped with two machines, which, during the harvesting period, operates 24 hours, supplying the government with 70,000 pounds of rice milled per week during t h a t season. His main occupation is milling rice and selling them wholesale to traders from outside the district. In addition to the 1,800 bushels of paddy he gets from his own land per season (for example in 198018 1 Maha), he also buys paddy from farmers in the village and the area. He also collects a considerable amount of paddy by providing seed paddy and tractor services to peasants in t h e area. He owns a four-wheeled tractor, two two-wheeled tractors, a lorry, a van and a car, all used for hiring. His economic power has enabled him to exert a considerable influence on many in the village, the government officials and the politicians in the area. In 1980, the officially assessed value of his wealth was eight hundred thousand rupees. He cultivates most of his land and the land he manages by using wage labour alone. In 1980, he had three permanent agricultural workers to work in his paddy fields, and six casual workers who work on his land most of the time on a regular basis. In addition, during the peak periods, about 100 male and female agricultural workers werebroughtfrom areas in Matara district for work on his paddy fields. Going against the tradition in the area of providing meals and tea etc. to wage workers and paying them in kind during the harvesting period, he was one of the two people who paid all his wageworkers entirely in The "Gambaraya" Systen~ in Transition. cash. The rural entrepreneur jn Polgahawelena, by 1980 had moved on to taking contracts in irrigation works, invested in a sawmill outside the village and was investing in developing a small tourist guest house.
The class differentiation in Polgahawelena leading to the rise of a stratum of middle peasants signifies the weakening of the grip of g a n~b a r a y a system and other agents of merchant capital on the peasantry. The formation of a stratum of rural entrepreneurs in the area originating mostly from the ranks of merchant capital including g a m b a r a y a .~ indicates what could possibly be a transitory stage of merchant capital embracing the sphere of production as well. While both the middle peasants and rural entrepreneurs represent a certain level of d.ynamism within the peasant economy of Polgahawelena the presence of the latter may indicate the possibility of capitalist development of agriculture.
Thus, the gambaraya system in its original form, of' on the spot land agents on behalf of absentee landlords with a n all pervasive influence of the peasant economy, is no more dominant in the village. The area of land managed by gambarayas as well as the influence of gambarayas on the society h a s considerably diminished. This is indicated by the changes in the usage of the term garnbaraya itself during this period. While the literal meaning of gambaraya is the person in charge of the village which the ganrbaraya was, in current parlance, the term is used by the villagers to identify any wealthy person in the village who provides working capital on credit and consumption loans to peasants or even sometimes more loosely, any wealthy person. Thus the traditional forms of control over peasantry exercised by gambarayas are fast disappearing. However, the second major function of the gambarayas, that is providing credit for production and consumption and working capital for production on high interest rates still continue in the rural economy through a variety of agents including the remaining gambarayas or their descendants who do not necessarily manage land for absentee landlords any more. The other agents include moneylenders, traders, government servants, and tractor owners and transport agents.

Dynamism in the Peasant Economy
The dynamism manifested in the emergence of new social forces was paralleled in the growth of the peasant economy. As observed already, the paddy yields in the village had significantly improved. Most cultivators irrespective of their economic status practised intensive cultivation using modern cultural practices. All the peasants used N H W s of seed paddy in recommended or higher quantities.17Most cultivators except the poor used the recommended amounts of the required types of fertilizer. The use of most modern chemicals available in the market for weed and pest control was widespread, the 'effectiveness' of the chemicals leading even to their excessive and indiscriminate use. Machines have completely replaced animal draught power in cultivation under major irrigation in the area, except in transportation where animal draught power continues to be used on a smaller scale. In 1980, none of the cultivators who came under the survey reported using animal draught for cultivation. Farm machinery was used in ploughing, threshing and winnowing and spraying of chemicals. Wage labour was the predominant form of labour used in cultivation. This is due to the relatively large size of the paddy parcels in the area and the necessity to follow a crop calendar for the maximum utility of water under major irrigation. The very high demand for labour in agriculture in the area during peak periods had led to high rural wages promoting labour saving techniques of production and discouraging labour intensive operations such as t r a n s p l a n t i n g , and manual weeding. By 1980, while chemical weedicides had replaced manual weeding, transplanting had also considerably declined.
The dynamism seen in the growth of the paddy economy is also manifested in another development of recent origin in the area, leasing of paddy land on money rent. This shows the diminishing risks in paddy cultivation and its increasing profitability. The most interesting aspect of this is t h a t most of the land leased are those under share tenancy. Without any involvement of the land owner, share tenants lease out the land they cultivate. This way it is assured that the leasing of land does not lead to the concentration of land and the dispossession of the cultivator. Even when the leased out land belongs to an owner cultivator, the system of renting did not lead to the dispossession of the cultivator. A lease is for an agreed period, a season, an year or more and a t the end of the period, unless the lease is renewed, the owner cultivator takes over the cultivation. This way, the owner cultivator gets cash rent a t the prevailing rates, even though it is lower than the profit helshe would have earned as an independent cultivator. I t is mostly poor peasants and middle peasants who leased out land. The leaseholders mostly came from the petty bourgeoisie or t h e middle capitalists. In 1980181 Maha, six holdings of paddy land i n the village were under lease rent. The lease rent varied depending on the fertility of land. While the average lease-rent was about Rs.500 per acre per season, in some cases the share rent of 12 bushels per acre to be paid to the landlord was also added as part of the lease rent. The most common reason for cultivators leasing out land was cash needs for a social event such as a funeral or a wedding in the family, or to repay a loan or for consumption.
Hena cultivation also had made advances in terms of productive forces. By 1980, many hena cultivators were using chemical fertilizer, weedicides and pesticides and modern seed varieties for cash crops while some used tractors for land preparation and even for clearing the weeds. The use of modern practices was most intensive among those who were engaged in highland cultivation.

The Old O r d e r P r e v a i l s
Despite the dynamism in the peasant economy and social relations in the village, other factors representing the old order prevailed: share tenancy and absentee ownership remained predominant; a few families controlled a significant proportion of the village land; the distribution of ownership of land as well as the size of holdings among cultivators was unequal; private money lenders dominated the rural financial market; landlessness was widespread; poverty persisted among the poorer sections of the society.
In 1980, the owner cultivators still represented a minority among the paddy cultivators both in numbers and the amount of land they cultivated. There were only 38 owner cultivators (or about one third of the cultivators) cultivating 92 acres (28 percent) of paddy land in the village of whom five were residing outside the village and not genuine owner cultivators. The larger proportion of paddy land in the village totalling about 253 acres remained under share tenancy owned by 31 landlords. Twenty seven of the landlords owning a total of about 206 acres were absentee owners, residing in the western parts of the district, neighbouring wet zone district of Matara as well as further away in coastal areas of Galle and Colombo.ln For the 153 acres ofpaddy land under share-tenancy there were 77 registered share tenants of whom only 31 were resident in the village. The rest were residing in neighbouring villages.
Siblings or close relatives are found among most of the registered landowners indicating that the control of most of the land was in the hands of a few families. Two families together accounted for the control of 105 acres of land, with one family holding 71 acres, and the other holding 34 acres. Other families which dominate land ownership and control were family number 3, (18 acres), family number 4 (16 acres), family number 5 (20 acres), and seven more families with 10 acres or more each. These 12 families together controlled 229 out of 253 acres of paddy land in the village.
Distribution of ownership of paddy land located in Polgahawelena is skewed providing a basis for economic inequality among the inhabitants. The paddy land located in the village was owned by 78 owners. As illustrated in t h e Table 1 the holdings owned by close to 60 percent of the owners were three acres or below, while a considerable percentage of owners had only holdings of one acre or below.
Among the cultivators, smallholdings dominated. The largest numbers of paddy holdings were in the category of 2 to 3 acres, 43 holdings accounting for 115 acres. Next came the 30 holdings of 1 to 2 acres accounting for about 48 acres. In the other categories, there were 27 holdings in 1-acre size accounting for 27 acres, 24 holdings in the category of 3 to 5 acres accounting for 92 acres, one holding each 8.25 and 13.75 acres respectively.
Of all the paddy land, about 53 acres or (16%) are cultivated by 11 owner cultivators in holdings above 3 acres. The rest of the 27 owner cultivators cultivate 39 acres i n smaller holdings, 23 of them cultivating holdings in the range of 1 to 1.5 acres; the rest were two holdings of 2 acres, one of 3 acres and one of 0.5 acres. The rest of the paddy land is cultivated by share tenants the majority of whom had holdings above 2 acres: 36 holdings of 2 to 3 acres and 19 of 3 to 5 acres. There were only 17 holdings of 2 to 3 acres and 19 of 3 to 5 acres. There were only 17 holdings below 2-acre size. As Table 2 illustrates, t h e majority of the households in Polgahawelena did not own paddy or highland. Of the households, 62.9%) did not own any paddy I& wbde 91.7% did ant, a m -1~18 bigb1aad. Most of those who owned land had smaller size holdings: 28.8% holdings below 3 acres, 24.7 below 2 acres and 15.3 below 1 acre. The percentage of households owning any highland other than home garden was only 8.2%. Of the ten households that owned highland, five owned holdings of 0.5 acre or less. Thirty-two (or 26.4%) of the families residing in the village did not own a home garden which meant they were residing on someone else's land or 'Crown land'. Twenty-nine households in the village had their houses built on 'Crown Land' on which they were 'squatting'. Of the home gardens, 37 (30.5%) were in the category of 0.25 to 0.50 acre size, while 37 (32.2%) were in the 0.51 to 1.0 acre size. Siphoning off the surplus from the peasantry continued by, in addition to landlords, both traditional and new agents of merchant capital: a few land agents, money lenders, traders, tractor owners, middlemen who buy agricultural produce from the peasants, shopkeepers, agro-chemical dealers, and those who lease in land from the peasants.
Borrowing cash from the landlord or a moneylender, hiring tractors, and to a lesser extent borrowing seed paddy, all to be paid in kind, continued in the village with very high interest rates prevailing. Middlemen who bought paddy from peasants kept high profit margins for themselves.
Moneylenders continued to dominate the rural financial market, charging annual interest rates ranging from 120 to 240 percent. The interest rates were higher in recovering the costs in kind. For example, for seed paddy the interest rate was 50% for a 3-4 month period. The interest rates for tractor services paid in paddy increases parallel to the increases of paddy prices i n the season. In some situations interest on cash loans taken fromfamily members, relatives and friends were relatively lower as 10% for the season or irrespective of the period. However, for most of the loans of seed paddy and tractor services taken even from family members, relatives or friends the same interest rates charged by moneylenders applied. Of the 50 farmers in our sample in 1980, only 4 ?k continued to take institutional loans for cultivation where the interest rates was 12% for the season. Only some moneylenders and not all act as monopoly buyers of paddy from the cultivator on t h e basis of providing loans to t h e latter. Some moneylenders limit their involvement in cultivation to lending only. For many cultivators, next to the borrowing of seed paddy and tractor services, purchasing fertilizer and other agro-chemicals accounted for a high portion of farm expenditure.
While the retrogressive aspects of the old order prevailed in the midst of dynamism in both the peasant economy and social relations in Polgahawelena, a t the same time these changes did not appear to be leading to a transformation of the rural social relations. The process of differentiation among the peasantry in the village, which probably originated in post-1958 Paddy Lands Act period, has slowed down. The changes in the paddy land ownership if a t all take place through the addition of new sinall owner cultivators who asweddumize their highland into paddy, and not through the changes in the ownership of existing land. Differentiation takes place more through the fragmentation of share tenancy rights among the members of the share tenant families, however without affecting neither the system of tenure nor the ownership of land. Thus, despite the slow process of continuing differentiation mostly among the share tenants there was no tendency of concentration of land ownership in Polgahawelena in the early 1.980s leading to dispossession of the peasantry.
The outcome has been persistent poverty in varying degrees among the families of poor peasants, tenants-at-will and rural wage workers a total of 56 in number representing 46% of the households in the village. The families of poor peasants in the village cultivate small paddy holdings and therefore the income from paddy land alone was not sufficient even for their subsistence. Moreover, their low incomes prevented them from fully benefiting from using modern agricultural practices to improve the paddy yields. They were also, due to their poverty, more susceptible to the control of the money lenders and other agents of merchant capital, to whom they lose a considerable portion of their income from paddy. Most importantly, the poor peasant families did not have additional sources of income with sufficient earnings to supplement the income from paddy and to engage in more productive paddy cultivation. While a few poor peasant families cultivated henas, the only supplementary source of income for most was rural wage labour which was at most available for about half a month, during the peak periods in cultivation when poor peasants themselves were partially tied up with work on their own plots.
Rural wageworkers with no land to cultivate and a limited number of working days for a year were the poorest. The government sponsored Food Stamp Scheme was their only regular source of income however meagre it was. Given the conditions of poverty among at least close t o half the families in the village it is not suprising that the Janatha Vimukthi Peramuna (JVP), the radical political group which staged an anti-state insurrection in 1971 but had temporarily returned to parliamentary politics by 1980, had a following of youth in the village and was the most active political group in the village and the area. In 1980, members of the JVP held the majority power in the governing bodies of-the village Co-operative Society and-the Rural Development Society. This shows that the JVP had the approval if not the support of not only the poor peasants and the rural wage workers, but also sections of the numerically bigger middle peasants whose support was essential to come to power in the above rural bodies.

Changes Between 1980 and 1987
These conditions in the village continued with certain changes in to the mid-eighties as revealed in our investigations repeated in 1986187 Maha season: the paddy economy continued to grow 'with an impressive increase in yields; the continuing dynamism in the peasant economy was also manifested in the spread of intensive cultivation of vegetables and other food crops for the market; the class of rural entrepreneurs continued to progressively expand their economic activities, leasing out paddy land on cash rent grew, while cash rents doubled; hena cultivation has ceased to become a source of income for the peasants in the village; the category of rural workers expanded; and significantly, there is no evidence of concentration of land leading to the dispossession of peasantry.
The average paddy yield in the village had increased by 62% from 66 bushels per acre in 1980 to 107 bushels in 1987. In our sample of 51 paddy farmers surveyed in 1986187 Maha only 9 reported yields below 70 bushels per acre. Among the rest, many had yields above 100 bushels per acre, as shown in Table 3. The above average yield was much higher than the annual average yields at both the national and the district levels for 1980, which were 66 bushels per acre and 82 bushels per acre respectively. The growth in paddy production in the 1980s may be related to several reasons which are associated with the liberalization policies and infrastructural development for agriculture followed by the post 1977 regime and the coincidence of technological advances.
By 1986/87 Maha , the farmgate price of paddy in the area had risen to Rs. 77 per bushel. Rising yields and paddy prices meant higher incomes to all categories of farmers. While the independent middle peasants in particular were bound to benefit from these changes, there seems to have been some trickling down effect of such benefits even to the lower classes. For example, unlike in the 1980 survey, in 1987 many tenants-at-will who were mostly cultivating land taken on lease rent by members of t h e petty bourgeoisie or middle capitalists, reported improved incomes from their tenancy in paddy cultivation probably due to the improvements in paddy yields and producer prices. Leasing out of paddy land in the village by middle and poor peasants had increased over 50%, from six holdings under lease in 1980 to 11 holdings in 1987. The lease rent had doubled to an average of Rs. 500 in 1980 to Rs. 1000 in 1987 per acre per season for fertile land, almost all leaseholders paying the applicable share rent payable to the landlord. The continuing dynamism in the peasant economy in Polgahawelena was manifested outside paddy agriculture through the recent diversified agriculture in the village leading to intensive cultivation of cash crops using irrigation on highland, mostly small home gardens in the village. This was partly due to the better prices for vegetables offered in the market in post-liberalization period and partly prompted by the loss of income from hena cultivation, which affected some poor peasants. Wl~ile there were two cultivators i11 the village for whnrn vegetable cultivation was the main source of' income, for the other four cultivators i t provided the secorld main source of income. In 1987, altogether there were at least 30 cultivators who were engaged in vegetable cultivation as a commercial activity. They cultivated about, 12.5 acres in total. Five of the cultivators had their cultivation located outside the vi1lage.l" It was a highly intensive form of cultivation using farm machinery, modern seed varieties, fertilizer and other agro-chemicals and lift irrigation. Some cultivators received substantial annual incomes from their cultivation. The cultivation of brinjals or the egg plant with its relatively superior qualities jn terms of yield and sustainability of the crop became the most profjtable and therefore the rnost popular, with profits in the range of Rs.20,000 t,o Rs.40,000 per acre per year depending 011 the success of the cultivation. Ten out of 30 cultivators cultivated mainly hrinjals.
Ecorlornic activities of the rural entrepreneurs in the area had expanded by 1987 with the setting up of a paddy stering system in the area with the assistance ofgovernment sponsored Bank ofCeylon. Paddy stores were set up for the storing of paddy to be milled and released t o the market during shortages of rice with a view to maintain a stable price of rjce. 75% of the cost of managing the enterprise, the purchase, storing arid milling of paddy is provided by the Bank. By 1987 there were about 15 such stores in the T7ssamaharama area, the bank loans for the project totalling about Rs.7 Million. The emergence of these paddy stores in the area while manifesting the increasing paddy yields, is also indicative of provision by the government sponsored banks EI further avenue of agriculture relat,ed investment for entrepreneurs in the area. They also have slightly ilnproved the seasonal opportunities of work for rural wageworliers.
Residents of Polgallawelena nlnrlaged two such paddy stores. One of the two stores was of a smaller scale and managed by 3 person belonging to t,he group of middle capitalists In the village whose economic activity is still mostly limited to rjce milling. The paddy stcores he managed had only an investlrlerit of about Rs.200, 000. However the rural entrepreneur in Polgahawelena identified earlier in the paper, managed a paddy store of a much larger scale wit,h an i~lvest,ment of about one million rupees of which 75%) of the finance was provided by the Bank on interest.The rural entrepreneur in Polgahawelena has also continued with other economic activities on an expanding scale except in paddy cultivation. He had increased the quantity of rice he rnjlls per month duruig the harvesting season from 70,000 pounds in 1980 to about 110,000 pounds in 1986/87 Maha. Ile also maintained the small tourist guesthouse which he opened up around 1980 with a monthly income of about Rs.8000 a month.
A major change unfavorable to the peasants i r~ the village in the j~ost-1980 period was that the heno. cult,ivatiori which was the main source of income for 11 families and brought secondary incomes to additional 11 familjes ceased to be a source of income any more. This was due to the fact that all the h c~i n land was located in the area which came under the Kirindi Oya Scheme, which was implemerltcd jn the 1980s."The land was either inundated in the newly constructed L~znugamwehera reservoir or taken over for resettlement purposes under t,he scheme. Some of the herla cultivators got compensation in the form of land under the resettlement scheme. There were at least 5 families who received such 1,md in the village. Ilowever, for the rest this meant the loss of an important source of income. Sonie ofthose who previously engaged in henc~ cultivation, by 1987 had become rural wageworkers, vegetable cult,ivators or were engaged in trade or technical work. Another significant change observed i11 the post-1980 period in Polgahawelena is the growth ofthe category of agricultural wageworkers. Those who derived their income mainly from agricultural work increased fro111 39 in 1980 to 51 in 1987, a 30fh increase. This also indi~at~es the growth of opportunities fbr agricultural wage work in the area probably due to the increase in the agricultural output mainly that of paddy. The demise of h e m cultivation and the decline in the practice of tenancy-atwill would have pushed some landless poor peasants to join the ranks of agricultural workers. The drop in the nuin1)er of tenants-at-will in the village fro~n 15 in 1980 to ' 7 in 1987, also inay be indicating that with the increasing 1)rofitability of pnddy production those who earlier cultivated land using tenants-at-1vi1l were now talring over the entire management of' cultivation into their hands using wage labour.
It, is also interesting to note that despite these changes which highlight on one Band, the contirluing growt,l~ in the peasant economy in the village and on the other hand, the process of differentiation aillong the peasantry there were no signs of concentrat~on of land or dispossession of' peasantry. This is manifested in the minimal changes of paddy land ownership recorded in the Paddy Lands Register for the period 1980 and 1985. Alt,ogether there were only 33 transactions. Of these only two transactions were direct sales of land changing ownership, jnvolving a total of 6 acres and creating two new owner cult,ivators. There were two incidents of share-tenants buying over land tot,alling about four acres from the landowner. Seven of {,he transactions were incidents of subdivision of ownership of land totalling 14.5 acres as a result of dent11 of the original owner or sale of ~)ortions of land a1110ngSa1nily members. Eighteen transactions involveti changes of share tenancy without affecting tlie ownership of land: In 16 of these a c c o u~l t i~l g for a total of about 46 acres, share tenancy was voluntarily transferred nlostly within the family, creating t,hree new share tenants in the process due to sub-division oft,enancy. There were two othcr incidents of transfer of share tenancy rights however to a party other t h a n the landlord, one, transfer. of 3 acres of land t o 3 person whonl the slzar e tenant was indebted, in lieu of the repayment of the debt; the other, 8.25 acres forcibly taken over against the wish of the sl~are tenant.
Thus into the latt~ 1980s the peasant economy of Polgahawelena continued t o grow, its cfgnamism taldng new directions in increasing leasing of land for inoney rent and diversif'yjng into vegetable cultivation on horne gardens. While living conditions of the middle peasantry possibly conti~lued to improve with higher yields and better produced prices, the conditjons of the poor peasantry did not improve much due to lack oC additional sources of income. Rural wageworkers grew in nulnber mainly due to the decline in henm cultivatjon and loss of tenancy-n t,-will among 0th er reasons.

Summary
In summary, agrarian change in t h e post-colonial period in Polgahawelena, as elsewhere in the country,21 has taken the same direction of growth in the peasant econonly and differentiation of peasantry without leading to the dispossession and disintegration. While the traditional exploitative land tenure system in paddy dominated by land agents or the ganl hara,yas on behalf of the absentee owners h a s declined, the absentee landlordisi~l and share tenancy remain predominant. The rural financial marlcet, continues to be dominated by the a g e~l t~s of merchant capital who siphon off a conslderahle share of the s u r p l~~s produced by the peasantfry. The ownership of land i s concentrated in the hands of a few however, without a continuing process of' accumulatjon of land leading to dispossession of the peasantry. Amidst increasing leasing of paddy land on money rent, sharecroppers remain entrenched without any changes in the pattern o f ownership of paddy. Inequality prevails in the distribution of land ownership as well as in the size of smallholdings while landlessness is widespread. Most importantly sources of additional income a r e very limited among the peasants, the poor peasants in particular. Rural wageworkers do not have sufficient opportunities for work thro~zghout the year. The outcome has been, persistent poverty aniong a considerable section of the village population comprising of poor peasants and rural w ageworlcers. Historical developlric~lts of peasant production in the district has been dealt with in detail by EIerring (1977) in a comprehensjve discussion of the impact of state policy and ecology on the development of the land tenure syst,em in the district since the colonial period. Leonard Woolf (1951) in the novel Village in the Jungle based on his experience as the Assistant Government Agent (AGA) represent,ing the British colonial government in the district from 3.908 to 1911, has vividly captured the miserable life of' t h e peasantry in the district in the early colonial period ravaged by both the human forces of exploitation and natural forces of disease and destruction. This is not to say that no land was alienated to the landless before 1920. There was a t least one exception of resettling Malay residellts in some areas under the Walawe Left Bank Scheme (Dias and Wickrarnnnayalre, 1977: 144). Out of consideration for t h e miserable life of the peasantry which was seen to be aggravated by parasitic landlordism, land speculation and money lending, some land was set apart fiom time to time for alienat,ion to the landless. However, the total amount of land thus distributed up t o independence in 1948 was a negligible 23,000 acres, much of which was non-irrigable land for village expansion.

3
Eve11 in 1980, a t least five of the share tenants in Polgahawelena the area under investigation in this paper were made up of either a faillily member or a relative of the landlord, a means of avoiding having a legal share tenant. Five of the owner cultivators, not resident in t l~e village managed their land totalling 15 acres though agents in the village or the neighbouring area. While the scale of' operation of' such land agents is much smaller compared to that of'the earlygcrnzbaruyas, in essence their role is the same. These lands are cultivated by t,enants-at-will.
For example, in 1940 in Pa~~naganluwa in Tissamaharama AGA division, 479 acres of' paddy land was auctjolled among the peasantry in plots ranging from one to five acres at nominal rates.

5
Our investigat,ions revealed that some ga?nhar.nyns preventfed share-tena~its under then1 from buying land m these auctions even when land was offered at 25 cents per acre, prom~sing that the land the s h a r e t e n a i l t ,~ cultivated under them woulcl be transferred one day to share tenants. 111 1980, there were 225 owner cultivators in Pannagamuwa. Later, a similar scheme on n much larger scale was impleinented 111 Bandagiriya, in Ha~nbantota AGA di\li sion.

6
By 1980 there were 168 owner cultivators (and 134 share-tenants as some of t h~s land was owned by non-cult,iv;ltors as well) owning plots ranging fiorri two to five acres cult~vat,ing 681 acres under 17 village tanlis uride~. the Weerawila Agrarian Serv~ces Centre area in Tissnrnaharairia AGA division alone.

7
Three of' the share tenai~ts paid tell bushels per acre due t o special consideratjorls of a particular landlord who held progressive political views.

S
The decljne was due t o the ban imposed on rssujng perlnits for the use of 'Crown lmd' on which hcna cultivation tool< place, under the assumption that it was u~lproduct~ive, ecologically wasteli~l and mlhealtl~y, mcul assertion, the accuracy ofwhicll is open tjo doubt (Herring, 1977, 126). f.) One hena cult~vator paid a nommal fee of Rs.10 for the annual p e r i~~~t for c~~lt~vation, which he obta~iled fro111 the government, Others have enloyc?d the possession of'their hciru land for a long time witho~lt, permits, or hawng 'inl~erlted' or ' p~r c h a s~d ' the land from ~t s previous occupant, considered and the laud their own proba1,l.y for a t least two m q o r reasons: one, the reluctance to reveal their real income fearing that the figures would reach the government sources resulting in the loss o f Food Stamps; the other, t h e fear t h a t other people in the village will be jealous of' their financial status in the event the information beconies public. Logically, the paddy yield in Polgahawele~la which h a s major irrigation and a highly intensive paddy cult,i.iratjon practised by almost, all the peasants, should be higher t h a n t,lw district level yields which incorpornt,es tile lower yields from the wet zone areas of'the district as well.
Weernwila in the Tissalnaharnn~ii AGA dlvisiorl in Magalri Pattu in Har11bantot.a tllstrict was one of the five locations in the country in which t h e Department ofAgriculture set up tractor hire statlolls ill Lcsponse to an increase In the demand for farm machinery 111 the fhce of draft power sliort,age 111 m;ljor paddy producing areas, between 1940and 1951(Perel-a, 1985. In 1980 t,here were cultivators in T'issamal~araina wllo benefited from one such scfierr~e ~Cmstitutional credit for the purchase of' two-wl~eeletl tractor h launched during the 1970-77 regime and who had improved their ecorloniy rx1nillly helped by hlnng the newly acquired tractors.
15 Non-~ultivat~ors and craftspersons are included here in the category of peasants on the basis of the n a t u r e of economic activity they are engaged in, the main characteristics o f which can be identified as the simple nature of the process, low level of technology,'capital and risk involved in the production process. I am thankful to Dr. S.B.D. de Silva for clarifying this point. Also see, Raymond Firth (1946and 1966) Alalay Fi.qhel*nlcn, Th.eir Peasan,t Economy, and Teodore Shanin (ed.) (1971. P e a s a~, t s and Peasant Societies, for discussjons on the iriclusions of noncultivators in the category of peasants.
16 In Polgahawelena there was one fanner who would fall mto the category of middle capitalists in terms of economic status hut not in social relations due to his left oriented political convictions which appeared to refrain him from expanding his econolrljc activities in the village or the area to exploit the peasantry. This was despite the fact t h a t he had a t his disposal enough resources to undertake such activity. His wealth both in the village and his home town Matara in 1980 had been assessed as worth of two hundred thousand rupees. 111 this sense he is a n exception to the social relations in agriculture in the area. 18 One of the landlords was the Land Reform Commission, which had'acquired Ten acres of land under the LandReform Law 'of 1972. The share tenants who had been cultivating this land were offered the opportunity to buy them paying the cost i n installments. Some share tenants in Polgahaweleria later bought. such land.
19 The five cultivators had their cultivation located o n land they received under the Kirindi Oya Scherne in compensation for the Izencr, land they possessed in the area coming under the Scheme.
20 Under the Kirindi Oya Irrigation and Settlement Scheme, with the objective of assuring irrigation water supply to the existing paddy land under the five major tanks in the area and expanding irrigation facilities to additional agricultural land, a large reservoir was built in Lunuhamwehera to store water from Kirindi Oya. This was accompanied by a resettlement programme, which provided land for cultivation and residence to the landless from the area as well as from outside. The expectation that the scheme would promote agricultural development in the area has now come into question with the increasing frequency and scale of crop failure on land under the Scheme i n late 1980s (see, Island International, November 23, 1988, p.5, 'Lunugamwehera Folly, The Causes', Tennyson Edirisuriya).