The aim of this article is to discuss the dynamics of the social relationships among African peasantries through describing the changing process of labor exchange and social relations. Labor exchange as a source of social safety and good neighborliness in village life has been one of major topics in the study of 'moral economy' or 'economy of affection' of peasant societies. Peasants in rural Africa pool their labor when they need intensive work on their fields such as clearing, weeding, and harvesting. They gather people not only from their own family or kin-group but from their neighbors too. Such kinds of co-operative labor often take the form of 'exchange labor' based on various degree of reciprocity and this co-operation lays the foundation for their sense of neighborliness.[1] In many cases, as among the Sabiny of eastern Uganda - the focus of this article - it is common practice to boost this spirit of cooperation among workers by sharing meals and especially drinking beer after they have finished.[2]

Thus, as other ethnographical studies describe, locally brewed beer epitomizes the symbolic medium of social ties among many peasant societies in rural Africa. For example, in his study of the Iteso of Uganda, Karp describes how people drink beer together at various occasions of their social life such as rituals and mere gatherings.[3] Among the Iteso, he says, beer is a symbol of diffuse solidarity. The Iteso have a saying that neighbors are 'people with whom one shares beer' and 'the primary source of labor supply for large-scale tasks'.　As in the case of the Iteso, many other African peasant societies combine labor exchanges with social events like the beer party. There are two aspects of the beer party system that are important here. First, people reinforce and renew their we-feeling by drinking together at the beer party, and second, people exchange their labor by the medium of beer.　

In recent years, however, quite a few studies point out that farm labor tends to become more individualist and 'traditional' social networks based on long term reciprocity like 'beer party system' have declined the role as channels of access to labor.[4] According to Ponte, after the liberalization of economic markets and commercialization of rural life in Tanzania, people utilize hired labor based on short-term contracts instead of relying on indigenous social networks like the beer party.[5] 

Like Ponte's study, most other studies begin from the premise that as the market economy penetrates rural Africa, channels of access to labor are disembedded from social context and individuated. This is a parallel to the classical view of modernization in the social sciences. For instance, Simmel says that personal face-to-face relationships decline and become anonymous once money mediates between people.[6]  As Parry and Bloch summarize, that kind of view treats money as promoter of individualism and devastator of community solidarity.[7] In Sabiny society, there is a tendency for the beer party system to decline while other forms of labor exchange and wage labor are becoming more common.

In this article I will describe the changes of labor co-operation in the research area and examine how far the mainstream theory about the impact of economic liberalization of labor holds.[8] I begin by introducing the socio-economic background of the area. I then proceed to discuss the 'beer party system' and the reasons for its decline. In the third section of the article, I analyze present labor exchange patterns and how they have evolved since colonial days. Lastly, I will discuss the significant points about the changes of social relationships in the area.

THE SOCIO-ECONOMIC BACKGROUND OF THE AREA

The Sabiny are a Southern-Nilotic people living on the north slope of Mt. Elgon in Eastern Uganda and the population is about 110,000. They have their patrilineal, exogamous aret (clan) system and a practice of virilocal residence. The research area is M Village, with a population of 360 divided into 57 households. People live on the terrace of the mountain. They grow maize, bananas, beans and other crops for family consumption and sale in the market. 

Until the first half of 20th century, Sabiny people were primarily agro-pastoralists, keeping cattle and goats, and growing crops like sorghum, millet and yam for family consumption. Many elders say the landscape of the area at that time was covered by bush and most of the land was uncultivated, used for grazing their livestock. They also brewed beer from their harvest and had beer parties. In the research area, people introduced ox-plow and maize around 1950. Maize cultivation emerged gradually afterwards. In the last decades of the 20th century they began to grow hybrid maize and nowadays the landscape is covered by fields of maize and bananas. In recent years, maize has become the main crop. Sixty percent of the households sell their maize in 100 kilo bags. Since the Ugandan government began encouraging production of maize as a non-traditional commercial crop, the national production rose sharply after 1990.[9] Today, my research area is one of the main producing centers in eastern Uganda.

Weeding the maize fields is a labor-intensive task done by women. During the rainy season they must weed their fields twice. When I did my fieldwork in 2002, I interviewed women about this work and sometimes participated in some of the weeding while doing my interviews. I gathered a good deal of my information about past and present practices from these interviews as well as with their husbands and other male residents in the area. Festive gatherings were often good occasions for getting information. I was often invited as a guest and those present requested to be photographed.  It turned out that these gatherings were often organized by women's credit and savings associations. They are very popular in the area today and, according to elders, were not practiced several decades ago there were beer parties as major festive gatherings in the community.

THE 'BEER PARTY SYSTEM' AND ITS DECLINE

Moyket : the 'beer party system' in the colonial era

Labor exchange based on the 'beer party system' was very popular among the Sabiny in the colonial era.  People call it moyket, or isyeet ak komek (literally: work of beer). As late as the 1960, moyket was the basic means of recruiting workers for major tasks, and beer parties were the regular social gathering of the community. When a married woman wanted to organize moyket, she brewed local beer called komek. Neighbors witnessing it would ask her for the date of work, and through the word of mouth they would come on the day proposed by the woman. People would work from morning to early afternoon. Around 2 pm, workers would return to their homes, clean up, change their clothes, and gather again for beer.[10]

In this moyket system, there are two types of reciprocal exchanges. First, in the short term, there is the exchange of one-day labor for beer. Second, hosts and workers take their turns during the season so, at least in theory, in the long term the labor exchange through the medium of beer involves every household in the community. Since every woman will be host and guest, this system reinforces the mutual obligations among women.[11] Borrowing Goldschmidt's words, this moyket system socializes women's labor.[12] In the years when it was widely practiced, sharing works meant sharing a pot of beer, and people enjoyed that occasion. Moyket was not just a way of saving labor or filling a gap in the family labor. It also rendered work less irksome.

This form of labor exchange reinforced generalized reciprocity among the villagers. Moreover, after work, they would talk about anything; sometimes gossip or quarrel, and inform each other of news in the area while they were drinking together. The woman whose field they have worked on served beer in a pot, from which 10-20 participants would drink using a long tube. The 'beer party' after moyket work was always a festive event to which sometimes guests who did not participate in the work during the day, called bendyo (literally: meat) would be invited. Goldschmidt reports that there were 30 such guests out of 295 workers in beer parties he observed.[13] However, there was some restriction too. Firstly, the host would control the number of guests per pot. Secondly, people at the beer party could not welcome a person who is regarded as sorcerer or just selfish - sokoronet. Such rules of the beer party helped sustain a form of loosely generalized reciprocity.

During the colonial era there was another form of labor exchange, formed by three to five women, called yemdoy. There was also a co-operative labor arrangement called kworishet which used a special meal of sour milk and goat meat or chicken instead of beer. In rare cases, there was ad-hoc hired labor in which workers were paid by cash or in kind (usually crops). Older women among my respondents said that yemdoy and kworishet were the ways young wives gathered unmarried youth of the same age, mostly for small-scale works. But neither system would attract large numbers to participate. Regarding ad-hoc hired labor, they explained to me that it would happen when they hired a few Gisu (a neighboring ethnic group) migrant laborers who lived nearby. [14] An old women told me with laughter, 'you do not know how we were in those years. We can neither eat nor drink money! Who could do anything for it?' What she and others emphasized is that, although other ways were there, they put significance on moyket as social events. 

The decline of moyket

Moyket began to decline for two reasons. One is spread of religions like Islam and Pentecostalism, a second is the commercialization of agriculture.

In 1970s a segment of the population began converting to Islam. Later, in the late 1980s, people began joining the Pentecostal Church. Those who joined Islam or the Pentecostal Church declined to drink alcohol and refused to attend beer parties. In 1980s, Muslims used to organize their weeding work by kworishet, preparing a special meal such as sour milk, chicken, sometimes goat meat instead of beer. In the earlier years, neighbors did not welcome them because refusing to attend the beer party was the same thing as refusing neighbors. But by the 1990s, Muslims and Pentecostals had become the majority. People drinking alcohol had been reduced to approximately twenty per cent of the population. This minority was now being called 'piko ak komek (literally: 'people of beer').

For a long time, brewing was almost the only way to sell maize. Since maize became a cash crop for sale in the market, most women have abandoned beer brewing. Only older women continue the practice today. Those who still do it do so because it gives them an income. Most sell their beer as wholesalers to the kurabut (club) or a bar in a neighboring village while some confine their sales on a retail basis at their own compound. Kurabut is the beer party of the savings and credit association.

The result is that the practice of drinking beer has changed in two important respects. First of all, it is less common today because religion has marginalized it. Only elders who drink really organize moyket today. In 2002, not a single woman organized moyket for the purpose of weeding her field.  At the same time, the savings and credit association had become a new venue for socializing and drinking beer. For those still drinking, its club (kurabut) was taking the place of the old beer party.

FROM 'WORK OF BEER' TO 'WORK OF MONEY'

Two types of weeding labor

There are particular divisions of labor by sex in the area. For example, men do plowing work with oxen.[15] Weeding work belongs to women and married men never participate in it. Married women basically depend on the labor of their own children and less often on their mother or unmarried younger sisters. I call this family labor. Apart from family labor, there are two ways women organize neighbors for weeding labor of their field. One is a form of labor exchange called fakiyet, the other is hired labor called kondaras.

Fakiyet is a rotational group exchange of labor. Some people explained that the term 'fakiyet' comes from the Swahilli word '(ku)fagia', meaning 'sweep up'. Each fakiyet group has its fixed members for the season, varying in size between three and twenty five workers - bigger than yemdoy and smaller than moyket of past years. Members of the same fakiyet exchange the same amount of weeding work by measuring stints per worker, and they work in each field by rotation. Reciprocity in this labor exchange is rigidly balanced. Unlike moyket, there is no partaking of beer or special meal after work. Only sometimes do they take lunch together.

Kondaras, etymologically derived from the English word 'contract', is ad-hoc hired labor. There are no seasonal wageworkers from outside the area, so the all hired workers are women or unmarried youth from the local community. There is also no partake of beer or any meal after work. According to the kondaras system, workers weed some stints at their own discretion. After finishing their stints they take their wage and leave. Old women commented that this kondaras is the best way of bringing a number of workers together as moyket was, and they usually added, 'if one had money.' The stints of fakiyet and kondaras are called mutagara. The husband or grown-up son of the woman usually measures the mutagara in paces before weeding work. The size of one mutagara is six times twelve paces. Payment for kondaras is 300 Ugandan shillings (about 17 cents) per mutagara.[16]

Network of labor transactions

I studied the arrangements for weeding labor among all married women in M village. They were sixty five in total; only three of them didn't have access to their own maize field. Table 1 shows that how women weeded their fields in 2002. Whereas twenty six women depended only on their own family labor, thirty three out of sixty two women utilized labor from outside the family for weeding their fields. They organized their weeding work either by fakiyet or kondaras, or by a combination of both. Many also went to someone's field for weeding to be hired as kondaras, wage labor.

Women who earn their own income tend to organize kondaras labor for weeding their fields. Fiffteen women organized kondaras labor in 2002. Table 2 shows their sources of wage payment for kondaras workers. Five women with banana fields are especially well placed because harvesting bananas is something that women do. They can sell bananas any time they wish. Another five women have their own petty businesses. They trade charcoal from other villages to town, sell homemade sour milk or locally brewed beer within the area. Some also sell other seasonal crops like sweet potatoes or tomatoes that they grow in their small vegetable gardens. But not all married women have chances to join these businesses. There are many others who do not have enough banana fields or kitchen gardens to sell crops. Some women complained that they do not know how to make sour milk well; some do not have their donkeys to carry charcoal. Another five women were helped by their husbands or sons because they were sick or had recently given birth to a child - or merely refuse to weed. Husbands or sons in these cases are shop owners in the village or work for a local NGO. 

Kondaras and fakiyet are intertwined. The former is the way to access labor and to earn a wage. People in the area who do not own their business are ready to be hired as kondaras labor on the fields. Sometimes the owner of a field informs or asks neighbors in advance, at other times people who like to be hired as contract laborers come to the filed to be recruited because they have heard from someone else that the opportunity exists. In the busy farming season young women can be seen roaming the area in search of kondaras work. Some of them come as a fakiyet group and convert their group labor to work kondaras.

An example of how women divert fakiyet to kondaras labor may be instructive. Suppose three women - A, B, and C - make up the fakiyet group. They fix their stint by mutagara, for instance, each worker agreeing to weed one mutagara per day. Each woman's field is of different acreage. For example, the field of A is equal to 5 mutagara, B's equals 8 mutagara, and C's equals 12 mutagara.  In a first round, they weed each other's field. In a second round, A finishes her weeding with her own family labor because only a small portion is left to weed on her plot. A now exchanges her fakiyet labor for a wage labor contract on someone else's field in the area. She takes B and C to this field where the holder organizes weeding work by kondaras, and is allowed to get not only her own but also the wages of B and C. This means that A can get 900 Ugandan shillings because the stint of their fakiyet group is 1 mutagara per head and payment of kondaras weeding is 300 Uganda Shillings per mutagara. Afterwards, the three women weed the fields of B and C. B and C will also divert fakiyet labor for a contract after a third round, the same was as A did in the second round. 

Table 3 shows the actual process of fakiyet and kondaras are intertwined. The field of C is larger than that of A and B. After the first round, both A and B diverted their fakiyet labor into kondaras and got 7,200 Uganda Shillings in total.

Thus, women can organize their weeding work in three ways: (a) their own family labor, (b) exchange labor (fakiyet), and (c) wage labor (kondaras). These three ways complement each other.  Women have several channels of access to labor for their own fields, and also have channels of access to wage labor. 

My point is that although the 'beer party system' has declined and hired labor has become the main way of access to labor, it is still embedded in the local context. Work arrangements among women are nested in other daily social interactions. Women help each other in their respective businesses, they participate in the same religious services and they engage in conversations as they fetch water, wash clothes or just chat in their compounds. In short, there are many different ways by which they can retain social ties among themselves. Perhaps no other venue, however, in recent times has become more important than the rotating savings and credit association. As Hyden notes, this form of association is one of the most common informal institutions in both rural and urban Africa. Its role among the Sabiny women deserves more attention.[17]

ROTATING SAVINGS AND CREDIT ASSOCIATIONS

Many married women in the area have been eager to join a rotating savings and credit association.  Table 4 shows the type, size and activities of such associations in 2002. These associations are the place for regular social gathering among people just as the beer party was before. 

Each association has its fixed members. Members of each association come from both M village and other neighboring villages. People classify these associations to three types: (a) gruuput (corruption of the English word 'group'), (b) paatyit ('party'), and (c) kurabut ('club'). Each association meets once a week and pools deposits. Each member gets to take saved deposits in the form of gift or a lump sum of money on a rotational basis. It takes more than two years for large associations to give a gift or money to all its members while it takes seven months for the smaller associations to do so.

Three associations have their own name: 'Gas Tai Women's Group', 'Gas Tai Mixed Group' and 'M Village Women's Party'. The words gas tai mean 'looking forward' and have a 'development' connotation. Another characteristic of these larger associations is that they give dividend by gifts, not by cash. Among the biggest two gruuput ('group' type of associations)', gifts of dividend are usually construction items like roofing materials and timber, blankets, bed mattresses, etc.[18] Traditional thatched roof round houses are still common in the area but tinned-roof square houses have become much more common. Before the 1990s, only a small number of villagers had built tinned-roof houses. Such houses are one of the barometers of good living. Before, only a few people with a regular wage income like policemen, soldiers or watchmen and some rich men who were able to sell their coffee or cattle, could afford to build them. Gruuput is popular among villagers because it enables them to build new modern houses like their better-off neighbors already have.

These associations have a chairperson, secretary, and bookkeeper just like the Local Council under the decentralization policy of the current Ugandan Government. Some members request to receive their dividend by cash. In such a case, the person must justify why she needs that amount (use of that amount) and members would give approval. Reasons are, for example, to pay a family member as patient at the hospital, or school fees of a son in secondary school. As far as I could gather, in the smaller paatyit or kurabut types of associations that give their dividend by cash, recipient must also declare for what purpose the money will be used.

I analyzed how these associations relate to fakiyet labor groups. There were six fakiyet groups for weeding in M village in 2002. Four are sub-groups of different saving associations to which they belong, which means each member of a fakiyet group also finds fellow members in the meetings of the savings and credit association. In these meetings, members not only make their deposits but also talk about problems each member faces. In addition, they often report the weeding condition in their fields and call on members to join their weeding fakiyet group at those meetings.

In short, the savings and credit association has become a venue for conducting a range of different activities. Membership also tends to have the effect of ensuring priority to such income-earning opportunities as kondaras. A woman belonging to a paatyit type of association (D in Table 4) is a case in point. Although nine other women in the same association formed a fakiyet group in 2002, the woman was too busy to join because she had her own charcoal trading business. She has to walk to a village at the foot of the mountain with her donkeys and buy charcoal by the sack and then proceed to town to sell her bags the next day. She hired those nine members of the same association for weeding her field. Among those nine, three of them converted fakiyet group labor to kondaras and each of them got 5,400 Uganda Shillings. According to the woman, there was a kind of agreement that when a member would organize kondaras for weeding, she should hire the members of the same association. For some members, the deposit of 3,000 Uganda Shillings once every three weeks is not easy to manage because they do not have a business of their own or sufficient sources of income. The agreement to hire members of the same savings and credit association is aimed at helping those who are poor.  Because members can infer the economic condition of each other, and furthermore, any person can tell just by looking which woman has a daily income from some business, members who do not have enough income for deposit can legitimately approach those who have income to hire them for money.

Membership of a particular association reflects daily relationships and also social category. People are keen to explain what kind of persons belongs to what kind of association. Members of the two biggest gruuput type of associations are supporters of different local politicians. There are two opposing candidates for Member of Parliament from the district and the supporters of each candidate formed two associations in 2001.[19] The paatyit type is the most common in the area. Members of this type are only women. They usually have a chat over milk tea while meeting and their meeting is brief compared with that of other types of associations. Members of the kurabut type are all drinkers. They are the successor of the beer party today, however they rarely organize moyket.

In sum, people put a certain label on each type of association. For example, members of gruuput are 'people who like talking politics', members of kurabut are 'drunk', 'dirty' or 'immoral' etc. Thus these classifications between associations imply some nuance of social segments in their society though not every co-operative relationship in daily life reflects these segments. 

CONCLUSIONS

Is farm labor in rural Africa becoming more contractual as Ponte insists? He argues that with commercialization of rural life, the social negotiation over access to resources such as land, labor becomes contractual.[20] Thus, according to him, in respect to farm labor, people tend to choose short-term labor exchange rather than a long-term 'beer party system', and furthermore are likely to choose ad-hoc contract labor instead of labor exchange.

I basically agree with his analysis, because in the research area most women choose fakiyet or kondaras rather than moyket. Women, however, combine fakiyet and kondaras, and make complex networks of labor transactions. Furthermore, the solidarity of savings and credit associations is also connected to networks of labor transaction. That is, festive gatherings like beer party and labor exchanges are seemingly separated, but they are in fact linked to a certain degree. To be sure, today's network is different from the 'traditional' one that sustains and reinforces their social order.  At the same time, this network is not a mere collection of ad-hoc dyadic transactions as a perfect contract society would be.

The difference between Ponte's and my conclusions stems at least in part from our choice of research location.[21] In his research area, many seasonal laborers came from neighboring ethnic groups and were hired temporarily during the farming season.　In my research area, by contrast, women keep their networks of labor transactions partly because there are almost no seasonal laborers from outside the area. Even if there were such seasonal laborers, however, it is unlikely that the current networks would disappear because they are grounded in their long-term transactions.

Ponte also argues that farm labor increases inequality. As contractualization of farm labor progresses, people who have access to off-farm income can easily afford to hire labor.　At the same time, women who do not have enough access to off-farm income hardly hire farm labor for their fields. My research shows that the situation is more complex and varied. Even if a woman does not have an income from off-farm business for organizing kondaras and her own family labor is not enough, she can join fakiyet. Furthermore, a woman can join kondaras as a hired labor at any time in a busy farming season. 

Women embrace the idea of reciprocal help for other members of the same fakiyet or association. When a woman joins fakiyet, she can divert group labor to kondaras for her income. As long as she belongs to an association, she can report her hardships in the meetings of the association. Members will assess whether she should get income from the funded deposit. As suggested above, a member who has her own off-farm income is expected to help other members. 

Summing up, I would say that my work indicates that the penetration of the market economy does not transform peasant society to the same extent that Ponte argues. Nor does my work confirm the point that Parry and Bloch make that social relations based on long-term transactions are so deeply embedded in culture that they subordinate all short-term transactions. Instead, I conclude that the activity of associations links both long-term and short-term transactions thereby sustaining ties among members while also generating new social orders. This is evident in my research area in the way savings and credit associations popularized the tinned-roof houses and changed the meaning of it from the symbol of rich man to symbol of solidarity. These associations also created new forms of associational life that are based not just on kinship or neighborliness. My study, therefore, shows the creativity or vitality of short-term transactions. They have the potential of changing long-term transactions and even the social order itself. This unending competition between short-term and long-term transactions explains the real dynamics of transformation of the peasant economy.

Endnotes: