From: African Studies Review 40(2):59-89. September 1997.

by

Tony Waters

Department of Sociology and Social Work

California State University, Chico

Chico, CA 95929

tel: (916-268-6595 (home)

(916) 898-6874 (office)

email: tony_waters@macgate.csuchico.edu

twaters@foothill.net

February 1997

This paper is an assessment of structural adjustment on a rural village, Shunga, in Western Tanzania. The villagers in this village are subsistence farmers. In order to assess the relative effectiveness of structural adjustment, comparisons are made to other macro-economic policies which have been implemented during the 20th century. It is concluded that structural adjustment has not had significant effect on village economic life. As before, villagers are dependent on hoe-based subsistence agriculture for basic needs, and are not tied closely to the broader national or international economies.

But, the "ineffectiveness" in Shunga of structural adjustment though is a characteristic shared with other macro-economic policies of the past, including ujamaa socialism and mercantile colonialism. In other words, no macro-economic policy has been effective in significantly changing the daily economic activity of Shunga's villagers. In any assessment of structural adjustment, or other macro-economic policy, the persistence of the subsistence way of life should be a central consideration.

Depending on who you listen to, structural adjustment has led to either rapid economic growth or a decline in living standards within Africa. And while in different places both circumstances are probably true, neither scenario describes what I have seen in rural Tanzanian villages since 1984. Despite the predictions about the effects of structural adjustment, my impression is that life goes on with only marginal changes in the daily economic activity of villagers. Economically, things are a little better on some fronts: the clothing is nicer, there are a few more bicycles, etc., but the changes are not dramatic; the daily economic activity is still hoe-based agriculture. It is also unclear whether such incremental improvements would not have occurred given any other macro-economic policy.

Indeed in such villages, past macro-economic policies including Ujamaa socialism, market capitalism, or even mercantile colonialism can also be correlated with one economic improvement or the other. But as with these past cases, unravelling the relationship between correlation and causation is a tricky business. This paper is an attempt at doing this for structural adjustment policies by comparing economic activity and relationships in 1985 before structural adjustment policies were adopted, and in 1995-6 well after the policies had had a chance to take effect.

STRUCTURAL ADJUSTMENT AND RURAL TANZANIA

That the fate of rural Tanzania should be ignored in analyses of structural adjustment is in many ways odd. Tanzania was aggressive in protesting the reforms in the early 1980s, but later turned around and adopted an austerity program, particularly devaluation, with vigor: between 1985 and 1995 the Shilling dropped from 17 per dollar to over 520 per dollar. (1) The two sides in the debate were not shy about predictions what would happen to the rural farmer if these policies were implemented. The World Bank of course predicted that the rural farmer would benefit from higher commodity prices. Opponents of structural adjustment countered that rural people would be unable to compete on world markets, and would suffer from declines in the quality of public services. My impression, though, is that neither scenario has emerged. Production has not noticeably increased, and services continue to be delivered in a rickety fashion that is not that different from the past.

But why have I concluded that neither predicted scenario happened? Why haven't Tanzania's farmers expanded production and vigorously entered world markets? Or, as skeptics claim, why haven't austerity measures resulted in the deterioration of health and education services by a government strapped for cash?

In my search for answers to these questions, I adopt an unorthodox approach. First, I will describe heuristically the social and historical context the one village in Tanzania with which I am most familiar. This heuristic approach emphasizes analysis across time, and asks what has happened there as a consequence of past macro-economic policy such as mercantile colonialism, and ujamaa socialism, and compares it to what happened during structural adjustment. This analysis provides the basis for understanding why I answer my "How was structural adjustment important?" question with the rhetorical reply: "Was structural adjustment really all that big of a deal in rural Tanzania?"

At the end of the paper then is then found an assessment of how widely read theoretically-grounded assessments of Africa miss out on the consequences of structural adjustment in places like Shunga.

Shunga Village is located in a remote area of western Tanzania. Currently (late 1996), the nearest market center, the District Headquarters at Kasulu, is about 1/2 days walk/bicycle ride from the village. The village itself continues to be surrounded by scrubby forest, which is diminished by firewood collection, cattle grazing, and hoe-based swidden agriculture. Fields are located both among the homesteads, in clearings further away, and recently, in valley bottoms. Primary crops are maize and beans of which there are normally 2 crops per year. Bananas are grown near the homesteads for beer brewing. Minor crops include peanuts, cassava, oranges, pineapple, mango, sweet potatoes, sorghum, tomatoes, onions, and greens.

As in the case of many villages in Tanzania, the current form of the village dates from the "Ujamaa" villagization of the 1970s. During this time, the Tanzanian army collected villagers from remote areas, and concentrated them in what is now Shunga Village. Before that, Shunga was the site of a rural Anglican Mission Hospital, church, and primary school serving a core of households, but particularly the scattered homesteads in the broader area.

The village is located on the side of a hill, and is about 3.5 kms. long, and perhaps 2 kms. wide. A small (50 hut) satellite village, Shunga Mpya (New Shunga), was established in 1993 0.5 kms. from the village proper.

Contact with the broader world occurred late, and gradually. Before the 1900, what is now Shunga was located between the Bashingo and Heru Kingdoms of the Waha. Peasants paid tribute to the Tutsi chiefs. As today, the economy was subsistence-focused. A difference was that the few trade items were locally produced by blacksmiths, coppersmiths, and makers of local bark cloth. The area also exported cattle and goats to Kigoma where they were exchanged for salt from the Uvinza Salt Mines 75 kilometers to the South. As with Burundi and Rwanda, Buha resisted attempts by Ngoni, Nyamwezi, and Arab invaders of the 19th century. How this was done is not clear. However, Tippu Tib in his autobiography reports that to assault the fortified wooden palisades of a village in southern Buha in the 1880s a wooden siege engine was used. This implies well-organized military defenses. After the arrival of Arab traders in Ujiji in about 1830, cloth and other manufactures probably entered Buha. By and large, though, the economy was isolated from that of the world until 1900.

Buha was occupied by the Germans between 1900 and 1910. The German movement into the area was probably facilitated by the decline of Buha's cattle-based economy following the rinderpest epidemic with consequent famine. At about the same time, Arab traders based in Urundi had bases at "Mbirira" a since abandoned site near Shunga. These bases were apparently used for cattle trading/rustling, and as a station for the purchase of slaves for shipment to Ujiji. It was also at that time that maize replaced cattle products as the main source of subsistence, presumably as a consequence of the cattle epidemics and famines (2). The Germans introduced a head tax in order to coax the area into some form of agricultural production. But this, according to Iliffe, by and large failed. Iliffe writes that after killing 548 Waha, the Germans gave up on their plan to tax Buha.

In 1916, Belgians invaded from the Congo and occupied Buha. Memories of the Belgian occupation are vague, and focus on attempts to take labor forcibly to the Belgian Congo. In 1922 the area was occupied by the British as part of the League of Nations mandate for Tanganyika.

In the 1930s, began the first attempts to pull Shunga into the world economy. First the Anglican church founded a clinic and church near an old marketplace in what was to become Shunga. As part of the program to support the mission, the missionaries initiated coffee plantations. Second, recruiters from coastal sisal plantations began hiring. Both impacts were trivial at the time, but were to have long-term consequences on Shunga. Labor recruitment, meant that many of the efforts of men were spent away from the area. The gender ratio within the village has remained disproportionately female beginning in the he 1930s (3). Finally, in the 1930s the British government initiated the first "villagization" program. Villagers were forced to leave the forest clearings due to the sleeping sickness epidemic in the area. Shunga emerged as a result..

A dirt road past Shunga was cleared in the 1950s using corvee labor conscripted by the chiefs to facilitate colonial tax collection This, as well as higher world coffee prices, led to a brief flourishing in the coffee market. Indeed, there was even enough trade to support the Indian and Greek traders who replaced the Arabs in the 1940s and early 1950s. The isolated ruins of the shops are still visible between Shunga and the neighboring village of Buhoro at "Kibirizi" where there are 7-8 concrete foundations. Nevertheless, these traders departed by the late 1950s, probably due to drops in coffee prices. The individuals and their shops are locally remembered, but little else; current residents do not know why traders left in the first place, or where they went. In any event, the once substantial plantings of coffee have disappeared.

The Germans Missionaries were interned by the British in 1939, and the mission station remained abandoned during World War II. Australian missionaries re-established the station in the late 1940s, and built the first school. At this time, too, tastes in dress began to shift from the traditional bark cloth to manufactured items from international markets. Local iron smelting and blacksmithing for hoe production began to wither as manufactured hoes became widely available. Thus, for the first time, a modicum of cash was required for the subsistence economy on which Shunga continued to be primarily based. These were incremental changes though, and did not result in any major discontinuities in the subsistence lifestyle. Thus, at the time of Tanganyikan Independence in 1961, and thereafter, all Africans in the Shunga area continued to obtain their basic food and shelter from subsistence agriculture without mediation by the marketplace.

Tanzania's policy of Ujamaa villagization changed the character of Shunga. As with the earlier British villagization policies, the majority of "villagers" living in scattered homesteads were relocated--often forcibly--into planned areas near the mission. The results of the relocation can even be seen in 1996, over 20 years later. The former homesites are identifiable by the mango trees seen in remote abandoned areas.

Ujamaa policies strengthened local government. The Village Council and Ministry of Education received control over the village primary school which was nationalized in the 1970s, and mass education began. The village also purchased and operated a grain mill, and opened a "village store." A Swedish aid program installed a water system. Vaccinations programs initiated in the 1960s continued. The government began also to participate in the funding and coordination of health services at the hospital, although control remained with the Anglican Diocese. Public health measures, in particular vaccinations, were initiated. To fund the development initiatives, villagers in the late 1970s and early 1980s were forced by local militia to cultivate cotton for sale to parastatals. In support of the cotton cultivation programs, a USAID funded go-down was constructed in Shunga. The coercive cultivation methods were however ineffective, and the cotton never reached harvest stage due to sabotage by hoe-wielding villagers during the weeding (4)

Each political and economic initiative had different effects which had accumulated by the 1980s. Roads were left form the colonial era, schools from the missions, but shops from no one.

By late 1984 when I first arrived in Shunga, the most recent initiative, ujamaa, was showing their age. Mass education programs continued, and larger numbers of children were literate; however, funding for teacher's salaries, books, and school materials were erratic. School attendance was also erratic. The material for the children’s' uniforms was unavailable. The village council's retail store periodically received small shipments of sugar and other commodities for sale at artificially low government prices, but the amounts were not sufficient even for the needs of the village's peasant 'elite' who typically sold the entire meager supply to each other. Laws against "economic sabotage" were enforced. Building materials (nails and iron sheets), grain milling services, new and used clothing, soap, cement, diesel, and other simple products were not available for sale on the open market in Shunga or Kasulu, and the black market did not compensate. Trading in such items was illegal, and maintaining supplies put people at risk for arrest for "hoarding." Both Kasulu and Shunga were at the end of the "Regional Trading Cooperative" pipeline which started in Dar Es Salaam; the few goods provided at the fixed prices were typically sold off at long-before reaching the distant west. Smuggled goods from nearby Burundi, were occasionally available, but were purchased only by risking arrest for "economic sabotage." But, not everything was grim. The absence of manufactures did not effect the food supply or housing, both of which continued to be rooted in subsistence production. On the positive side, a measles epidemic bypassed Shunga in 1985 because government coordinated vaccination programs proved successful. The village government's grain mill operated off and on. But the water system did not work, even sporadically.

The efforts at forming cooperatives for the marketing of coffee, cotton, and maize failed. Transport was also a critical related issue in 1985. Government controls on diesel effectively limited vehicle operation. Indeed, the only vehicles to regularly leave the village were owned by the German nurses operating the hospital. Unusual for the area, one of the villagers, the Baptist pastor owned an old lorry. However, at least in the mid-1980s, this lorry did not operate due to lack of spare parts and fuel in the government controlled markets, or for that matter in the black market.

Government Land Rovers made it to Shunga less frequently, maybe once every month or two. Most popular were the visits of the District Game Officer who came to shoot the lions that occasionally entered the village to take dogs and goats, most recently in 1993. Lorries from the parastatals also arrived once or twice per year to purchase whatever grain surplus happened to be generated. However, it was a hit or miss proposition for whether they arrived to make their purchases before the crops spoiled. Caravans operated by ivory smugglers passed too, going through Shunga on their way to Burundi as recently as 1987.

Salaried workers in the village included 40-50 people associated with the hospital or school. Salaries were minimal in 1985, about 1,000 shillings per month (i.e. $58 at official rates, and $10-12 at black market rates). Due to the continued provision of free medicine from the government, and subsidy of medicine from the German Missionary Society which sold medicine at "official" exchange rates, medical care was relatively cheap, and available. Schooling was free, although, the irregular payment of teacher salaries meant that teachers were often absent, and school materials (paper, pencil, chalk, etc.) were usually unavailable. Attendance rates were probably about 90%, and children were though at least literate at minimal levels.

Despite the grim prognoses in 1985, though, life in Shunga continued. Adequate supplies of food continued to be available in the subsistence sector as it had been in the past. Unlike other areas of Tanzania, there was no famine in the 1980s caused by crop failures, although many children suffered from chronic protein deficiency. As they had since colonial days, men continued to leave the village in order to find work in the more distant towns as unskilled laborers, with the result that the gender imbalance persisted.

The major social occurrence during the last ten years has been the influx of refugees from Burundi in late 1993 and early 1994. In the short-term, this created a crisis at the hospital which was the only medical facility to directly care for several tens of thousands of refugees in late 1993. Mortality rates for the refugees were high (about 50 in two months), and villagers depleted their grain reserves to accommodate the refugees. However, for the villagers themselves, supplies proved adequate to cope. More significant, though, is the establishment of a small refugee camp of about 8,000, Ntabila, on the border of the village. At the insistence of the UNHCR and central government, the boundaries of this camp were formally surveyed. This has meant that there is now a boundary to the village other than the forest. Thus, for the first time there is a formal limit to the expansion of the fields. One result has been friction between the villagers of Shunga and the refugees.

Surveys have also begun to replace traditional consensus-based allocation of plots within the village. The village itself is divided into four sections since 1992. Title to house plot are now sold back to the owners for $13.33 for the survey of a numbered 20 x 30 meter house plot. The householder is given a receipt a receipt for this transaction, and there is no hint yet of corruption. This receipt provides a guarantee against claims by families which may have had prior usage rights under more informal systems of "land registration." More distant forest and river-bottom land though remains available for the clearing, without reference to survey.

Economic stimulus, too, has resulted from the refugee camp. Jobs with the Non-governmental agencies, primarily the International Rescue Committee (IRC) and Medicines sans Frontieres/France (MSF/F), operating the camp are now available for small numbers of villagers. Wages for unskilled jobs are subsidized with international refugee assistance money (in 1995, the lowest refugee camp wages were $65 per month, i.e. about double that which Shunga Hospital and the government offered for similar positions). This affected the ability of Shunga Hospital which pays such workers with patient fees to recruit semi-skilled staff in particular. However, only 2-3 nurses at the hospital have left to take jobs with MSF/F and the IRC.

Another economic effect of the camp for the local area has been a sudden increase in the demand for surplus maize. The refugees continue to be on World Food Program-based food rations, and as a consequence for the first time there is a significant local demand for maize outside the subsistence sector, i.e. by WFP. Local transports have responded by traveling from village to village in order to collect what surpluses are available. It is not apparent that local farmers have begun to plant for surplus, though. What seems to have happened is that farmers respond by planting more cassava, the preferred food of the Burundian refugees, i.e. a sign of a differentiated market-based production.

Shunga has been effected by the AIDS epidemic which has effected this part of Africa. Most effected are the "modernizing" portions of the village, i.e. those with systematic contacts beyond the village: nurses, teachers, and the wives of traveling laborers. The majority, engaged as they are in subsistence agriculture within the immediate area are unaffected. Rates are undoubtedly lower than more urban areas, even if later they are likely to expand in places like Shunga, as epidemiologists predict (5). As a result, the village has continued to expand in population significantly. In the late 1980s, a new area was cleared for settlement across the ridge. Farmlands have also extended further, and the forest fringe has also receded significantly.

Thus, the subsistence economy has remained persistent during the last ten years. Villagers continue to eat what they grow, and there has been no obvious specialization of production in order to serve world markets. All Africans in Shunga continue to farm, as they have in the past.

So, how is Shunga's economy different in 1996 than in 1985? How have the years of structural adjustment affected a remote village like Shunga? Have the effects of structural adjustment changed economic life in Shunga any more than other circumstances? What was more important, structural adjustment, or the chance creation of a refugee camp on the village border? Has structural adjustment affected social life in Shunga any more than previous macro-economic policies? Below is a brief summary of economic changes observed during the last ten years in Shunga. Included are activities in both the subsistence and market economies.

a) Shunga Mpya and continuing population expansion despite the age of AIDS

Rural Tanzania continues to have a high birth rate; in 1992, the growth rate was reported at 2.8% for Kigoma Region. . Most obvious is the establishment of the satellite village "Shunga Mpya" (New Shunga) where about 5-10% of the villagers live. The rest of the village has grown noticeably since 1985, with the village fringe now 100-200 meters beyond where it was previously.

Migration by young men for labor continues to be common, with destinations including Mwanza, Dar Es Salaam, and tobacco farms at Ngaruka (a village between Kigoma and Tabora). A number of these young men have returned to Shunga to die of AIDS. The toll that AIDS has taken in Shunga is unknown as many continue to die at home. Death and the keeping of medical statistics still is a taboo subject.

What this means is that birth rates continue to exceed the combined out-migration or mortality rates. Migration into the village has at best, been insignificant, although it is known that a few Burundi refugee women have come to the village without bride price payments as second or third wives.

b) Trees: more and less

Continued population growth, land-extensive agricultural practices, a new population of refugees, and continued dependence on wood for fuel caused a highly visible retreat in the forest fringe. In many respects, this is the most obvious change in economic resources during the last ten years, effecting as it does land, i.e. the primary means of production. Deforestation varies in intensity, with the areas around the village for perhaps 500 meters now grassland. For another 5 kilometers, the forest is noticeably "shorter" than it was in 1985; the full-grown trees are gone, and all that remains is a scrub no more than about 1.3 meter high.

The diminishing forest has been noticed. Government tree planting programs have been initiated, and peasant farmers plant eucalyptus and fruit trees. This interest is a change since 1985 when coercion was necessary to induce planting. In 1996, eucalyptus poles are necessary for roof construction, as forest sources diminish. Straight fast growing eucalyptus poles have acquired a market value as a result.

c) Bricks, Iron Sheets, Doors, and Furniture: Investing in the homestead

In 1985, well over 90% of the houses were of a mud and wattle with grass roof construction. The vast majority of buildings in Shunga continue to be of this construction; however the number of improved houses, i.e. burnt brick with or without corrugated iron roofs has increased noticeably. Improved houses now include approximately 20% of the structures in the village, with a majority found near the older established homesteads at the village center. In contrast, new buildings at the village fringe are almost uniformly of the mud and grass roof construction. Of the fifty buildings in Shunga Mpya, only one is of bricks and corrugated iron..

Construction of these houses has occurred in a combination of the subsistence and cash economies. Burnt bricks are all produced in Shunga. No bricks are imported from commercial clay works. Kiln construction requires extensive labor in the preparation of mud, kiln preparation, and firewood for firing. Bricks are available on the cash market, but only "on order," no inventories are maintained. Skilled masons are available for house construction on a contract basis, particularly during the dry season when there are no agricultural activities.

Another popular item, wooden doors, is more firmly in the cash economy. During the last ten years, most of the grass and corrugated iron doors have been replaced with hardwood doors, even in mud houses. Wooden doors and windows are an important status symbol, as well as a protection against burglary, and are made by village carpenters. In 1995, about 90% of the houses in Shunga had the doors which are typically made by local carpenters, one of the few specialized crafts in the village. There are no power tools in Shunga. As a result, rough boards continue to be planed by hand, a time-consuming and highly skilled process. The carpenters though do not live from the proceeds of their handiwork. All also maintain fields.

There is also a demand for chairs, beds, and cupboards. The main carpentry workshop in the village remains in the Mission compound, and is capitalized by the expatriate nurses living there; they arrange for the purchase and transport of the timber stocks Items are produced primarily for the use of the mission and its hospitals. Students from the mission's workshop have though established independent workshops producing chairs, doors, window frames, tables, and furniture.

Home improvements are also rooted in the broader District market. Manufactured nails, corrugated iron sheets, hinges, and other needs for building are purchased in Kasulu. There is no regular stock of building materials (except nails) in village markets. Corrugated iron sheets in particular, are used as a form of savings and are often hidden in the house. The corrugated iron has several advantages; corrugated iron is safe from shilling depreciation, and the heavy bundles of sheeting are difficult to steal. Corrugated iron is also safe from fire.

d) Where are the Coffee Trees? Still Looking for a Cash Crop

Coffee is periodically promoted by the central government as a cash crop, however never seriously enough to establish reliable extension services, credit facilities, nurseries, or purchasing facilities. Burundi was a traditional but illegal market for the small amounts produced in the past; war in Burundi since 1993 has however meant that that market has shrunk. The net result is that coffee production remains as minor as it has been since the 1950s. Despite soil and weather conditions suitable for coffee production, structural adjustment has not been adequate for the re-emergence of production for the export sector, legal or illegal.

Cotton was planted in newly cleared forest clearings on the northern side of Shunga for the first time in over ten years in 1996. Villagers expect the harvest to be purchased by Mwanza-based traders. There has yet to be a harvest.

e) Subsidies from the International Refugee Regime: wage labor and grain sales

The refugee camp in nearby Ntabila has effected the economy of Shunga since it was opened in early 1994 in two ways. First, a few of the nurses have been hired away at wages which are about 2-4 times what they were paid in the hospital serving Tanzanians (i.e. US$100-120 per month). Such employment opportunities are though available to only a handful of the villagers. The net effect is minor for the village as a whole, while the individuals concerned of course do relatively well. This in turn exacerbates the ability of the hospital to keep staff.

Second, the World Food Program and UNHCR have arranged for local maize purchases in area villages. For example, a large sale was made in late 1995 which generated much of the cash which went to the purchase of building supplies. This was done in anticipation of a bean harvest in late December/early January; unfortunately though, expected rains did not arrive, and the bean crop failed. At this writing (Feb. 1996), it is expected that villagers will be entering the cash markets for food for what is perhaps the first time. This will be necessary if they are to survive until the bean harvest in April-May.

(May 1996: the bean harvest failed, but villagers have not entered cash markets despite earlier worries. Indeed the villagers complain that maize is not found in Shunga's markets, despite substantial stocks at nearby villages within in bicycling distance. I do not know why enterprising traders do not make this trip, despite claims that villagers are "hungry" for maize. Instead, villagers turn to cassava, cooking bananas, and sweet potatoes as staples. Oddly, the cooking bananas and cassava are the preferred staple of the Burundian refugees, while maize is the preferred staple of the Tanzanians of Shunga. As a consequence, there is some bartering back and forth. Despite general village grumbling about "hunger" for maize, the hospital has not seen any change in malnutrition rates. Towards the end of May, Petro, who is described below, took delivery of 4 tons of maize which was stored in his house)

f) Of eight grain mills and a broken water system

In 1985, there was one grain mill in Shunga which was operated by the Village government. There was also a water system. Of the two, the grain mill operated sporadically, while the water system not at all. Both sectors were the exclusive prerogative of the village government under socialist laws of the time.

Structural readjustment opened grain milling to private individuals. The result is that now there are eight private grain mills in the village. (The village's grain mill was privatized.) As a result, the "thump-thump" of the diesel-driven engines is now a common background to village sounds. Thankfully for the women, the onerous task of milling maize manually has disappeared. Milling costs $0.04 per "bakuli" (a local measure equivalent to approximately one liter). Cassava costs $0.03 per bakuli to mill.

All acknowledge that the villages grain milling needs are saturated; the operation of eight grain mills is more than the village can support. The decision to purchase a machine seems to be as much in the interest of status-achievement as capital investment.

The village water system remains in-operational. Excuses given are the same as ten years ago; diesel and spares are lacking. Oxfam repaired Shunga's system in 1996 following complaints by villagers that the refugees received free piped water from an Oxfam-installed system, while they did not. Oxfam agreed to install three spring protection if the villagers would assist with manual labor. In two places, an uncovered spring protection system was installed. This was a small dam with a piece of plastic pipe sticking out of it. Villagers failed to provide the necessary volunteer labor for the third site.

g) The New Capitalists?

The increased availability of cash means that a few, very few, individuals established businesses based in the investment of capital and the accumulation of profits. The owners of the grain mills are of course among these "capitalists" as are those establishing small retail shops. The shops are hard to recognize. Unlike the many small "dukas" along main Tanzanian roads, the shops of Shunga have not invested in the signage to attract strangers. Despite structural adjustment, the ubiquitous Cafenol cough drops and Coca-Cola signs found in more traveled areas of Tanzania are missing from Shunga. Indeed, missing also is the popular soft drink itself.

Briefly, I will describe the business of the two leading "capitalists" of Shunga.

--Petro, the capitalist who charges his parents

Petro has the nicest house compound in the village. He owns two lorries, one of which operates a transport business from Shunga to Kasulu, with trips going every Tuesday. He also stocks retail items for sale in secure 20 foot containers near his house.

Petro's father was the pastor of the Baptist Church, which received strong backing from the American Southern Baptist Convention during the last 15 years. Indeed, exploiting this connection gave Petro's father preferential access to imported items during the early 1980s when importation was closely controlled. At that time, Petro's father acquired two grain mills, which he subsequently lost when such assets were "nationalized" under ujamaa policies.

Petro himself took advantage of trade liberalization to establish a transport business. He transports both goods and passengers, and has established a retail business in a 20 foot container. Half of the container is dedicated to bicycle spare parts, while the other half has a mixture of women's khangas, cheap tennis shoes, razor blades, salt, and other consumer goods. He has also purchased two grain mills, one of which is operated in Ntabila refugee camp, and the other in a neighboring village. He has not entered the crowded grain milling market of Shunga. Among the villagers, Petro has a reputation for business-minded ruthlessness: he is criticized because he charges his own parents a fare for going to Kasulu.

Of the "entrepreneurs" in the village, Petro is probably the one who could separate himself from farming and live on profits. Or more likely, perhaps, he could move to town and operate his transport enterprise from there. He professes to prefer Shunga though, which is familiar, and dislikes travelling to Mwanza.

Despite his relative wealth, Petro continues to have family members farm in the subsistence sector; he himself does not purchase grain for his own consumption. At his homestead, he also grows oranges, and other "innovative" crops recently introduced to the area.

--Mathayo: the video from the back of a bicycle

Mathayo is a young man with a capitalist vision: he bought a video and small generator. These are mounted on the back of two bicycles which he brings to sites within the village where he charges 50 shillings ($0.11) to watch. He bragged once that this charge was adequate to pay for his capital investment within a few months. Among other business enterprises, he has organized a hardwood timber cutting operation in a distant forest, and capitalized a retail shop which has a prime location across from the mission hospital, and a carpentry workshop. The shop sells imported consumer items including khangas, thongs, underpants, soap, sugar, etc. Payment in cash or maize is accepted. Mathayo is hesitant about extending credit.

Mathayo did not extend his video enterprise to neighboring villages because he says this would involve moving the expensive equipment along remote trails subject to banditry. Others say that he does not do this because he cannot find employees who will not pocket the proceeds.

Structural adjustment made the enterprises of both Petro and Mathayo possible. Mathayo was not able to operate a lorry from Shunga before diesel and spare parts became more available in the early 1990s after import restrictions relaxed. Likewise, under socialist laws governing importation in the past, Mathayo would not have purchased a video machine. Whether the enterprises of Petro and Mathayo are beneficial or not is open to a value judgment. However, clearly, both enterprises are possible due to structural adjustment.

Trade liberalization has also permitted individuals to open small tea shops, and small retail shops. Shops opened have far exceeded demand, with the result that, except for the shops of Mathayo and Petro, they open and close. Many people within the village undoubtedly have their unsold inventories of soap, sugar, bubble gum, razor blades, and aspirin tucked away in their houses where they are slowly consumed. Oddly, following the failure of the bean crop in April 1996, 5 or 6 small shops were opened by young men no longer pre-occupied with agricultural activities.

Walking or driving around the paths of Shunga, the most visible evidence of involvement in wider markets are in the quality of clothing worn by the villagers, and the larger number of Chinese-made bicycles darting about. Both items while available in 1985 were rare, but most young men, and some of the salaried young women working in the hospital now own a bicycle. The bicycles typically cost between 1-2 months salary.

In 1995, the clothing was noticeably brighter in colors, and less ragged. Substantial numbers of school age children now have school uniforms, although they are well-worn. Women have more colorful khangas, while especially the young men have new clothes. In other words, some of the prosperity of the last ten years was spent on clothing.

By local standards, their has been a marked increase in trading since 1985 when trading was severely restricted by Ujamaa era laws. It is worth re-emphasizing that in 1985 even the simplest consumer items like soap, sugar, bubble gum were lacking in large part due to restrictive importation laws, and laws against "economic sabotage" which discouraged accumulation of retail stocks. While it is true that such items are expensive, particularly when considered in the context of the low wages, it is also true that items like bicycles are slowly making their way into even the poorest households.

h) The Commoditization of Land?

In 1985, usage of land was assigned by the village at no cost to the villagers. This was done in a wide if poorly defined area. Since then, though, land has become commoditized. This is particularly true for homestead plots near the center of the village which demand a premium. Plots further out, for example those in Shunga Mpya have lower prices. The borders of the village have also become more carefully delineated by the District authorities when the survey for the refugee camp was done in 1994-5.

Forest land "free" for the clearing is still to be found some ten to fifteen kilometers away from Shunga along the Burundi border. This is being taken advantage of by Burundian refugees who continue to settle there surreptitiously. Shunga's Tanzanians are also known to take more remote lands. In particular, they have begun cultivating bottom land vegetables something that was not done in 1985.

Land boundaries continue to be the result of verbal agreements enforced by the village government; there is no central depository or formal survey of plots. Land disputes continue to be common. The delineation of field boundaries is the subject of many long-term disputes. Rumors of murder by poisoning feature in some cases. The most recent example of such unexplained deaths by land disputants was in 1993 and 1994. These cases have not been followed up by the police or District authorities, but do become part of village lore.

i) The Middle Class? Nurses and teachers at $30/month

There are in 1996 about 100 employed in Shunga for a wage that averages $30-50 per month in 1995. These include teachers and nurses who are in effect the "middle class" of Shunga. Their salaries give them a regular source of cash which most others do not have. This means that they buy nicer clothes, corrugated iron sheets, and other consumer items; little though goes towards food or subsistence items. All "eat from the farm;" the salary is inadequate for the purchase of the food for subsistence. This has not changed from 1985 to 1995.

Devaluation policies mean purchasing power of salaries did not improve substantially from 1986-96. On the other hand, the wider availability of goods means that there are more choices about how what little is earned is spent. In real terms (i.e. 1985 black market rates versus 1995 structural adjustment rates, unadjusted for dollar inflation) there is a US$10-15 per month increase in salaries(not adjusted for inflation). In other words, the gains of the salaried "middle class" of Shunga have at best been trivial, despite structural adjustment. As before, wage earners engage in subsistence farming, and make no purchases of food in the marketplace. Meanwhile, "middle class expenses" such as increased school fees, commoditized land, loss of access to goods at artificially low prices, etc. have probably caused the overall purchasing power of salaries to decline.

j) Continuing Limitations in the Village Marketplace

The village marketplace has expanded during the last ten years. In 1985, there were few vegetables and fruit available, e.g. tomatoes, oranges, and papaya. Sugar and other goods available through the regional trading company rarely made it to the village, and then were only sold at artificially low government prices to the village leaders.

Since trade liberalization, the village market has expanded to small amounts of manufacturers, including nails (sold individually), hinges, used clothing, soap, used clothing, shoes, khanga wraps for women, candy, sugar, cosmetics, and razor blades. Dried fish from Lake Tanganyika is also usually available. Major items such as bicycles, iron sheets, are not available in Shunga, and must be purchased in Kasulu. Livestock (cattle and goats) are not available. Cattle in particular, remain outside the cash market and continue to be important only for the payment of bride prices. Only extremely minor amounts of staple foods, chickens, or meat are available in the village markets. However, purchase can be negotiated outside the market.

Saturday, as in most Tanzanian villages in the area, continues to be the main market day when the most items are available. The market area (approximately 22 x 22 meters) itself is informal: vendors must bring mats to spread on the ground to display goods.

k) Public services: AIDS and continuing subsidies in medical care

Part of structural readjustment was the discontinuation of free medical care at government hospitals, and the discontinuation of the free basic medicines distributed to private hospitals such as Shunga. The net result is a slight price increase in the fee-for-service consultations.

The sharpest price increases were in the price of medicines ordered by the German Mission from Europe. Before structural adjustment, the mission society serving Shunga Hospital purchased drugs in Europe using Deutsche Marks. The cost of the drugs were then re-imbursed by patients at the artificially low government rate. Devaluation, though, has effectively eliminated this subsidy, and is a new cost passed to patients. This problem though been compensated for by improvements in the supply of drugs via the government-run drug warehouse in Tabora which, since 1993, has improved drug availability for Shunga.

Treatment of indigents has continued. This was a particular problem during the Burundian refugee influx of late 1993 when there was an epidemic of bloody diarrhea. Re-imbursement for the cost of medical services was promised by the UNHCR, however the agreement was never put in writing, and the process for re-payment never completed. A consequence was a temporary shortages of drugs and supplies for Burundian and Tanzanian patients alike.

Whether due to the treatment of indigents, supply problems, or increased demand by refugees, drug shortages remain common throughout the region, particularly in village-level government dispensaries.

AIDS is a major health problem in Shunga as it is in the rest of Tanzania. Rates are undoubtedly lower than in urban areas; the continued travel of men to the cities and tobacco fields though means that the death of both young men and women is more frequent than before. Putting figures on the increase is not possible, though, as their is no systematic testing is permitted. HIV infection is a strongly stigmatized condition.

l) Public services: continuing problems in the school

School attendance has declined throughout Kasulu District since 1990. Government statistics indicate that in Kasulu District, attendance has fallen from 85% of those required to attend, to 65%. Direct causation cannot be demonstrated, but this fall in attendance correlates with the introduction of fees for school children (in Shunga 800/= per child), requirements for the wearing of school uniforms produced in Mwanza (1500-3000/= per uniform), and an assortment of other fees associated with school attendance. The most recent campaign is to require students to wear shoes to school; in May 1996, only about 10% of the school children complied with this rule.

It is not clear whether there has been an improvement in the quality of education or not. Attendance by teachers continues to be problematic. (In fairness to the teachers, it should also be noted that the prompt arrival of salaries remains problematic). Few if any students from Shunga are able pass the Standard 7 exam each year (6).

m) Public services: CCM party politics and Village Government

In 1985, Tanzanian was a one party state, and the village government was naturally the monopoly of Chama cha Mapinduzi (CCM) members. Due to socialist policies, this meant that party members controlled the few economic activities of the Ujamaa system, e.g. crop marketing, retail sales, grain milling, etc. Under this system, the chairman of CCM was also the Village government leader. By 1995 a new system of "multi-partyism" had been introduced in which the CCM chair and the head of the government, and the village chair, had been separated even though in the case of Shunga, the positions were held by the same man. The position "Katibu Mtendaji" (Village Executive Officer) was introduced with multi-partyism to represent the central government, and was appointed from Kasulu.

No announcement about which party won the first multi-party parliamentary/presidential election in Shunga in October 1996 has been made. However, as a rural sector of a District where CCM won, it can be assumed that CCM also swept the vote in Shunga. There was some activity in Shunga on the part of the opposition party NCCR-Mageuzi by Mathayo and other young men, however this was attractive only to a small number of young people. There was a great deal of discussion about the election, and national newspapers appeared in Shunga for the first time. These were read carefully as they were passed from villager to villager.

Particularly in the era of multi-partyism, village politics has became a surrogate for how the villagers deal with questions of modernization and development. The official village leadership is aligned with the central government and national policies of expansion of education, health, roads, agricultural extension, etc. In 1985, this role involved fighting for a share of the resources monopolized by the central government in Kasulu, Kigoma, and Dar Es Salaam. Taxes consisted of an annual "development tax" which was a head tax levied on every adult. In 1985, this tax was considered something of an annual nuisance; it was small enough that it was promptly paid in villages like Shunga.

Structural adjustment, shifted more financial responsibility for schools, hospitals, and other government services back to the village. The Katibu Mtendaji, who is appointed by the Kasulu District Commissioner, organized taxes for the operation of public facilities, in addition to the head tax. Two strong factions, those who support the Katibu Mtendaji's attempts to finance development programs, and those who do not, emerged. The Katibu Mtendaji focused on the visible structures of development, i.e. the schools, etc., while those opposed claim that the village leadership is hopelessly corrupt in their handling of revenues. The more conservative opposition was led by the Chair of CCM/Village.

The severity of this cleavage was demonstrated in late December 1995 when the Katibu Mtendaji's opponents attempted to burn him alive by torching the grass roof of his house in the middle of the night. This act was done following the failure of the bean harvest due to lack of rain. Ostensibly the house was torched because the Katibu Mtendaji refused permission to the village militia to beat up three diviners who were blackmailing the village by stopping the rains (7). He also did not approve of the beating of an old single woman by the village militia accused of using witchcraft to stop the rains. The Katibu Mtendaji and his family escaped the fire, and the following day, the grass and eucalyptus pole roof was quickly replaced by the villagers who supported his position. Nevertheless, the Katibu Mtendaji was transferred back to Kasulu because all agreed that his life was in danger if he remained in Shunga. The District Commissioner fired the Village/CCM Chair for corruption.

Crimes are a continual cause for villager concern both within and beyond the political sphere. Robbery on the remote roads has long been a fear of villagers. Bandits laid in wait along remote stretches of trail in 1985 and 1995; at both times crime was the focus of villager concern. Nevertheless, little changed in law enforcement. The central justice system is still not capable of dealing with village level crime. Police and courts are still too far away to provide effective justice, and are viewed by villagers as susceptible to bribery. As a consequence, criminal offenses are dealt with haphazardly. Bicycle thieves from the next village who lie in wait for villagers are, it is said, protected by their well-known Tanzanian families. There is fear of revenge in the event of a police report, and no official sanctions against offenders are levied as a consequence.

In contrast, Burundian refugee bandits were handed over to the police; their relatives are unknown, so it was reasoned that police action is possible. Other thieves caught red-handed have been subject to arbitrary justice; one in another neighboring village was lynched with a burning tire 'necklace.' A Shunga man caught stealing a grain mill in a neighboring village in 1996 escaped public burning (villagers had begun to gather grass and firewood) only after his family intervened, and the village chair backed off. Less public burning of roofs at night, witchcraft accusations, and poisoning remain a popular, and more common form of revenge.

While the factual details of such incidents of mob justice cannot easily be confirmed, such tales shape social relations, including economic activity. The sometimes arbitrary terror involved in necklacing, robbery, poisoning, witchcraft accusations, and mob justice effect the decision-making of all, not just those committing crimes. The widespread perception that criminals are released after paying bribes also contributes to the perceived arbitrariness in the justice system.

The point of this essay is to ask what role structural adjustment played in remote Shunga Village. Throughout the last hundred years, macro-economic and political policy formulated in remote capitals has had an impact on Shunga, shaping life in this remote corner to a greater or lesser extent. What are some generalizations that can be made about the effects of structural adjustment in Shunga? I would suggest the following answers:

The economy of Shunga has not shed its subsistence base; those who have become dependent on the market for daily subsistence (i.e. food) are not present in Shunga. In Shunga, the primary means of economic subsistence remains the individual with a hoe and land. There is no evidence that the Shunga economy is systematically supporting any traders or "petty bourgeoisie." As a consequence, structural adjustment focusing on market-based activities, has had only a limited impact.

But, both market activity and ethics are emerging. Differentiated market-like activity is apparent in Shunga in the building trades in particular. Specialized skills and materials are valued for the construction of houses. These materials and skills have been developed in Shunga, and include timber, poles, burnt bricks, etc. A number of manufactured items are also desired, including cement, corrugated roofing sheets, nails, hinges, etc. Notably, this is different than it was seventy years ago when houses were made from grass. This has not though led to what proponents of structural adjustment would expect. There has been no consequent surge in production for national or international markets. There has also been little specialization of labor.

The demand for construction skills and building material is though stimulated by the increased circulation of cash following structural adjustment. But structural adjustment by itself was not enough to generate even these minor economic shifts. An argument could also be made that without the concentration of villagers under Ujamaa around the mission station, the development of skills necessary for such construction would not be in Shunga today. Indeed, there is every evidence that such skills were developed in earlier days; masonry and carpentry are both skills used in the area since the 1930s. In other words, it is likely that improvements in building skills and technology would have occurred without structural adjustment. However, it is also likely that structural adjustment sped up construction activity, particularly that which involved the acquisition of manufactured items such as corrugated iron sheets.

There is also substantial evidence that economic conditions have improved in Shunga during the last ten years. Some of these improvements, e.g. the availability of corrugated iron, bicycles, and other consumer items may be very well attributed to policies of trade liberalization. However, there is no evidence that structural adjustment has had any more profound effect on economic life in Shunga than earlier policies. Indeed, in terms of economic production, the combination of high coffee prices in the 1950s and coercive British colonial practices was the only effective economic policy stimulating production for world markets. In recent years, neither structural adjustment, nor Ujamaa socialism matched this record. (But then, in even earlier years, neither did earlier German policies of forcible monetization, or mass murder work either.) This again points to the resiliency of the subsistence mode of production in rural western Tanzania, whatever macro-economic policy may be the political fashion.

Social change in Shunga was more often the unintended result of broader political policies rather than economic policy. Forcible villagization in both the 1930s and 1970s had a more profound effect on social routines and economic life Shunga than the macro-economic policy adjustments of other eras. Neither program effected though the availability or land, or subsistence agriculture as being a legitimate activity from which social status is derived. The most significant economic fact in Shunga is that subsistence agriculture continues to dominate economic activity in 1996 as it did in 1900. This is how villagers spend their time and insure survival and they are likely to do so as long as

1) land remains available for the clearing, as it has in the last ten years in Shunga, and

2) Subsistence farming remains a socially legitimated economic activity.

There is also no evidence that macro-economic policy will tie Shunga into the broader worldwide economy in the future; there is no obvious crop or other product which would make villagers dependent on a differentiated economy. Coffee appears to be a moot issue. And it is unlikely that Shunga's hoe-dependent maize farmers will be able to enter world maize markets and effectively compete with agriculture in Iowa. Emigration with resultant remittance income perhaps remains the one distant hope for the integration of Shunga into world markets. This though has not worked in the past, and there is no indication that it will work in the future.

There are a few other loose points which need to be made about economic life in Shunga:

--There is nothing unusually romantic about the subsistence economy in Shunga. It can result in conflict over resources as do any market-based systems.

--The use of Shunga as a labor-reserve may have very well retarded the development of Shunga's economy. No effective means of dealing with this issue from a macro-economic level since it began in the 1930s have emerged.

--The rule of law--in both contractual and criminal spheres--remains weak. In the criminal sphere, robbery, poisoning, house burning, necklacing, or in other words terror, are the main means for social control, rather than a functioning court system. In the contractual (civil) sphere, consensus remains the most important means for enforcing property rights. Such methods, are not predictable (i.e. irrational in the Weberian sense), relying as they do on surreptitious action. This is true even if in their own way, these systems "function." However, it is doubtful whether in a place like Shunga the property rights assumed by structural adjustment can be protected without some form of rationalized legal protection from the state. Unfortunately, structural adjustment as a group of policies does not address such issues.

For a number of pages now, I have proceeded with the heuristic exercise of searching for what, if any macro-economic policies have effected production in Shunga village. The conclusion is in effect, a rhetorical question: "Was it any of it really all that big of a deal??" This is different than asking, as most do, did it succeed or not in a particular national, regional, or rural economy. Put in logical terms I am considering not only the "alternative hypothesis" that structural adjustment led to changes (either positive or negative), but also the "null hypothesis" that nothing changed much at all. Does this add up to a net gain or decline in overall "standard of living?

Most importantly, there has been no significant change because the primary economic activity subsistence agriculture, persists. In 1985 before structural adjustment, villagers spent most of their time growing maize and beans for the subsistence economy. The same was true in 1995. In fact, the same was true in 1940, 1950, 1960, and 1970 as well. In other words, despite the adoption of various flavors of macro-economic reform, little seems to have changed in the most basic economic activities. Indeed, the only significant production for world markets occurred in response to coercion (in the 1950s), and in an unplanned fashion as the consequence of demands created by the current refugee crisis. The former gains were of course short-lived, while the latter--a sort of accidental capitalism via refugee welfare regime--have yet to prove themselves one way or the other.

Thus, despite the fact that Shunga is the very type of village which the structural adjustment programs were meant to benefit, I found no evidence that the reforms have been a shortcut to progress. On the other hand, the disasters predicted by the skeptics have also not occurred. Although, (as in the past) precise survey statistics are lacking, it is doubtful that illiteracy, mortality, or other indicators of health have risen or fallen much as a consequence of the austerity measures (8). Rather the slow changes begun with the first villagization programs in the 1930s continue, albeit without change in basic economic activity.

Up to here, this paper has been purposely atheoretical, and has not address broader questions about structural adjustment as a general macro-economic policy, the social dynamic of agrarian change, the uncaptured peasantry, neo-colonialism, the nature of property rights, general social processes, household dynamics, gender relations or the other issues which dominate academic discussions (see for example Berry 1993, World Bank 1994, Campbell and Stein 1992, Rigby 1992, Leys 1996, Cornia and Helleiner 1994). Such studies are interesting when careful inference is made to the types of places which they are based. Does Shunga meet this criteria? My answer is no it does not. Exactly why it does not can be found by examining the assumptions of some of the better studies.

How otherwise well-done studies miss villages like Shunga is in large part due to analysts’ insistence that pre-existing models of capitalism, class relations, etc. based on data from urbanized Africa, rural Asia, Latin America, and 19th century Europe be uncritically applied to places like Shunga. This is not an original point, and was well-developed in Koponen’s (1988:38-45) study of late pre-colonial Tanzania. I share his concern that there has been an artificial dichotomy developed between the "development discourse," and that of the Marxists. As Koponen notes, such models are poor tools for understanding what has happened in late pre-colonial Tanzania and are "inadequate concepts stemming from inadequate theories" which do not permit the flexibility necessary to identify new discoveries (Koponen 1988:40). I would extend his generalization to modern Shunga. in particular, and assessments of structural adjustment in similar areas.

Both the designers of Structural Adjustment policies, and its critics assume that government policy plays a central role in development. The assumption is that policy is central (see World Bank 1994 and Leys 1996). The capacity of a government policies to effect successfully productivity within its geographical boundaries is assumed; i.e. the ability of the nation-state to effectively implement policies is believed to be constant throughout a national territory. This is implied in the presentation of all "per capita" data which is uses total national population in the denominator. Indeed, it is only after this point that there analyses diverge: advocates of structural adjustment assume programs lead to a broadly-based development, while critics point to the volatility of exploitative class relations which emerge in such situations. To make the latter argument effectively, though, it is necessary to assert--without assessment of places like Shunga--that all of Africa has been proletarianized (see Leys 1996:132).

Advocates of the World Bank view take a similar view: "Better Policies Pay Off" is the explicit conclusion of the recent report Adjustment in Africa (World Bank 1994:4-7). If better policies do pay off, Shunga has missed out. Even assuming that there has been say, a 2% growth in production in Shunga (which would perhaps represent a macro-economic success) the low level of market production would mean only a trivial increase in cash per capita per family. For example, assuming generously a current income of $50 per capita in Shunga means a family of eight would net an extra $8.00 per year from this 2% growth rate. But this calculation still misses the central point about Shunga. As in the past, Shunga’s farmers feed themselves with the hoe. Nobody is getting rich from this, but then Shunga is not becoming a latter day version of Marx’ industrial proletariat either (9).

Analyses of 1990s Africa are of course no longer limited to a socialist/capitalist dichotomy. Nevertheless, a consistent feature has been the neglect of areas where the hoe remains the dominant means of production. This means that again, material progress is used as the dependent variable in analyses. A good example is Sara Berry’s well-respected book No Condition is Permanent: The Social Dynamics of Agrarian Change in Sub-Saharan Africa. Her network-based theory of agrarian change is based on a survey of African economies which have a history of plantation agriculture (Cocoa in Ghana and Nigeria), proletarianization via mining (Zambia), or a settler economy (Kenya highlands). Theory based on such examples cannot be generalized to places like Shunga.

There is though no indication that the state has had a more than passing strength in Shunga. Admittedly, at times the state was "strong:" both the colonial state and socialist state successfully implemented resettlement policies effecting everyday social life. Likewise, if what Iliffe reported about the massacre of 548 Waha by the German colonial powers is even partly true, it can also be brutal. But exercising such passing military strength is qualitatively different than maintaining the apparatus necessary to operate effectively an economy, be it market-based or socialist. Poorly developed state and market forces have meant that macro-economic policies, have had a very indirect impact at best. Ultimately, when the lorries returned from the resettlement programs (or more likely broke down), life returned to hoe-based swidden agriculture. Little progress was made in the long-term development production for national or international markets. Or in other words "development."

Theoretically, of course, this means that I have backed into a corner. If not in terms of market economics or class relations, what is happening in Shunga? But., as with Koponen’s (1988:360) broader treatment of pre-colonial Tanzania, there is no neat summing up. I would though like to point at a few factors which have been underestimated in analyses of villages which, like Shunga, continue to be embedded in subsistence production. Three in particular come to mind.

First, is the issue of what role "rationalized norm enforcement," or in other words the enforceable contract, plays in places like Shunga. A recurrent theme was the role that terror (which is arbitrary not rational) plays in economic decision-making in Shunga. This is poorly understood in Africa. Nevertheless, this same subject was emphasized when Max Weber brought up this subject in his discussion of the union between "fortress and marketplace" in his discussion of modern capitalism.

A second issue is the question of how a persistent frontier influences both the ability of the African state to extend itself into the marketplace. John Iliffe (1995) has recently made a case for the frontier being a constant when evaluating African life. Goran Hyden (1980) also contends that such a condition means that a peasantry will remain "uncaptured," a concept which is in turn borrowed from the classical sociology of Marx, Weber, and Durkheim (see Waters 1992b for a review). Domar (1970) has also discussed the negative relationship between the presence of "free" frontier land and what is in the effect the proletarianization of labor. William McNeill (1978) and Alfred Crosby (1986, 1994) also discuss how the presence of an empty frontier effects population growth rates. McNeill in particular raises issues of how settled rural areas like Shunga are necessary for the population for both the frontier and industrializing cities. The latter two references are a far stretch from answering the question of how the frontier effect the union of market and fortress, but given the heuristic approach I advocate, could well be relevant.

The third issue is that of proletarianization of African peasantry. Modern analysts, embedded in worlds where labor is commoditized and monetized easily forget that the separation of labor and its product (i.e. differentiation of labor) is a process completed in the West less than 200 years ago. Thus, an issue central to the minds of early social philosophers (10) has faded from the policy-making analyses of economists at the World Bank and the universities. For that matter, I have yet to see an assessment from the Marxists about why Marx’ (1978, 608) own ideas about the alienation or the peasant farmer cannot be applied to places like Shunga.

The sum of this theoretical discussion is that, as for development in Shunga, there are still no obvious generalizations about agrarian change in Africa. Theorists claiming so need to take pause, and consider the long history of efforts to develop places like Shunga. There are no shortcuts to what are ultimately complex questions.

Nevertheless, while I do not propose any new theory about agrarian change in western Tanzania, or anywhere else on the basis of Shunga, I do hope that this paper can perhaps form a brick in a theory of agrarian change. Future models proposed by political economists should take account of the nature of incremental change that has occurred in villages like Shunga. Indeed, I believe that the macro-economic models, whatever their focus, ignore Shunga at their own peril.

I worked in the area of Shunga from 1984-7, and in 1987 was married in the village. I spent a total of about seven weeks in Shunga in 1994-6. During this time, Heidi Kaletsch, a nurse who has worked in Shunga for 12 years was particularly helpful in describing daily life there. Magadalena Lanz, who has worked in Shunga since 1974 was also very helpful and insightful.

Much of the background information comes from car trips in the area around Shunga Village with men from Shunga and other villages. Such trips often provide splendid opportunities to discuss local history and sociology. Particularly interesting accounts have been offered to me by Mzee Abdallah, Mzee Chamwino, Mr. Mamboshella K., Mr. Mathayo Balikutsa, Mchungaji Julius, Mwenyekiti Patrick, and others. The idea for this essay came as a result of a similar ride in 1995 in Ngara District with Beth Whittaker, a graduate student from the University of North Carolina. During this ride I found that I could not articulate well why I thought existing explanations of social change in Tanzanian villages were inadequate. She later very kindly sent me a copy of Sara Berry's book which has stimulated my thoughts further. This paper then is the result of that conversation. I hope that this article explains to her better my views, even though she may still not agree completely.

A number of the Missionaries of Africa (White Fathers) living in the Kigoma area for many years have also influenced my thinking a great deal. Among them are Hank Van de Paverd, and the late Francis Tryers. Father Tryers lived in Kasulu form 1943 until his death a few years ago. He was an enthusiastic advocate of Ujamaa villagization policy, and offered eclectic and enthusiastic explanations of why places like Shunga changed for the better while larger macro-economic policies failed.

Like Marguerite Jellicoe (1978:x), my methods were "various." In Jellicoe's term, this meant that material "was all collected through participant observation and informal interview." This is a dignified way, I suppose, of saying that one's curiosities were satisfied by asking knowledgable friends questions about social life, mainly in a Land Rover, over banana beer, or coffee.

There are a few scholarly written sources about the Kasulu area. The most thorough English-language ethnography of the area continues to be Scherer (1960). Chubwa's (1986) Swahili language history and ethnography is more up-to-date and covers a wide-range of subjects. Other ethnographic writing includes, Mbwiliza and Gwassa (1976), Lovett (1994), McHenry (1973), and Waters (1992). References about Kasulu (Buha) are also found in old issues of Tanganyika Notes and Records, Man, and other sources written during the colonial era. Articles by J.P. Moffett and Captain C.H.B. Grant are among my favorites. Such articles do have a colonialist bias, but also have a richness of period data not easily found elsewhere. Good published bibliographies are found in Chubwa (1986), and Waters (1992a).

Bates, Robert. 1981. Markets and States in Tropical Africa Los Angeles:University of California Press.

Bates, Robert. 1983. Essays on the Political Economy of Rural Africa, Cambridge: Cambridge University Press.

Berry, Sara. 1993. No Condition is Permanent: the Social Dynamics of Agrarian Change in Sub-Saharan Africa, Madison: University of Wisconsin Press.

Bradshaw, York W., Paul J. Kaiser, and Stephen N. Ndegwa. 1995. Rethinking Theoretical and Methodological Approaches to the Study of African Development, African Studies Review 38(2):39-66.

Brain, James. 1973. The Tutsi and the Ha: a study in Integration Journal of Asian and African Studies 8:39.

Campbell, Horace, and Howard Stein, eds. 1992. Tanzania and the IMF: The dynamics of liberalization. Boulder: Westview Press.

Chubwa, P. 1986. Waha, Historia na Maendeleo, Tabora: TMP Book Department.

Cornia, Giovanni Andrea and Gerald K. Helleiner, eds. 1994. From Adjustment to Development in Africa: conflict, controversy, convergence and consensus? New York: St. Martin’s Press.

Crosby, Alfred.1986. Ecological Imperialism: the Biological Expansion of Europe 900-1900 Cambridge: Cambridge University Press.

Crosby, Alfred. 1994. Germs Seeds and Animals, Studies in Ecological History, Armonk, New York: ME Sharpe.

Domar, Evsey. 1970. The Causes of Slavery or Serfdom, A Hypothesis. The Journal of Economic History, 30:18-32.

Goebel, Alison and Marc Epprecht. 1995. Women and Employment in Sub-Saharan Africa Testing the World Bank and WID Models with a Lesotho Case Study, African Studies Review 38(1):1-22.

Gwassa, G C K, and J F Mbwiliza. 1976. Social Production, Symbolism, and Ritual in Buha: 1750-1900. Tanganyika Notes and Records 79-80(13-21).

Hyden, Goran. 1980. Beyond Ujamaa in Tanzania Berkeley: University of California Press.

Iliffe, John. 1979. A Modern History of Tanganyika Cambridge: Cambridge University Press.

Iliffe, John. 1971[1987]. Agricultural Change in Modern Tanganyika, Historical Association of Tanzania Paper No. 10. Nairobi: East African Publishing House.

Iliffe, John. 1995. Africans: the History of a Continent. Cambridge: Cambridge University Press.

Jellicoe, Marguerite. 1978. The Long Path: Social Change in Tanzania, Nairobi:East African Publishing House.

Koponen, Juhani. 1988. People and Production in Late Pre-colonial Tanzania: history and structures, Finnish Society for Development Studies.

Leys, Colin. 1996. The Rise and Fall of Development Theory, Bloomington: Indiana University Press.

Lovett, Margo. 1994. On Popwer and Powewrlessness: Marriage and Political Metaphor in Colonial Western Tanzania. International Journal of Historical Studies. 27(2):273-301.

Marx, Karl. 1978. The Marx Engels Reader, 2nd ed., ed. by Robert C. Tucker. New York: WW Norton.

Mauss, Marcel. 1967[1925]. The Gift Forms and Functions of Exchange in Archaic Societies. Cunnison., trans., New York: Norton.

McHenry, Dean, Jr. 1973. The Utility of Compulsion in the Implementation of Agricultural Policies: A Case Study from Tanzania, Canadian Journal of African Studies 305-316.

McHenry, Dean Jr. 1994. Limited Choices, The Political Struggle for Socialism in Tanzania. Boulder: Lynne Riener.

McNeill, William. 1978. "Migration" in Human Migration, Patterns and Policies, ed. by W H McNeill and Ruth Adams, Bloomington: Indiana University Press.

Ndegwa, Stephen N. 1992. The Struggle for Relevance in African Studies: and African Student's Perspective. Issue 20(2):42-45.

Rigby, Peter. 1992. Cattle, Capitalism, and Class: Ilparakuyo Maasai Transformations Philadelphia: Temple University Press.

Scherer, J.A. 1960. The Ha of Tanganyika, Anthropos 54:841-904.

Seppala, Pekka. 1996. Entrepreneurship and Diversification: the Decline or Professionalism in Tanzania, a paper presented in the 39th annual meeting of African Studies Association, San Francisco, November 1996.

Smith, Adam. 1937. The Wealth of Nations, New York: The Modern Library.

van den Walle, Nicholas. 1994. Adjustment Alternatives and Alternatives to Adjustment African Studies Review 37(3):103-117.

Vansina, Jan. 1991. Paths in the Rainforest Madison: University of Wisconsin Press.

Waters, Tony. 1992a. Lifeworld and System: Of Grain Mills and Water Development in Rural Tanzania, African Studies Review 35(2):35-54.

Waters, Tony. 1992b. A Cultural Analysis of the Economy of Affection and the Uncaptured Peasantry in Tanzania, Journal of Modern African Studies 30(1):163-75.

Waters, Tony. 1995. Africanists and the Challenge of the Mission Verandah, Issue 13(2)

World Bank. 1991?. Tanzania, AIDS Assessment and Planning Study, A World Bank Country Study. Washington: World Bank.

World Bank. 1994. Adjustment in Africa: reforms, results, and the road ahead. World Bank: Oxford University Press.

Summary Statistics Available for Shunga Village, Kasulu District, and Kigoma Region. Statistics for Shunga were obtained from the Hospital which conducts immunization clinics. Only since 1992 has coverage of Shunga itself been verifiable. Regional and District statistics were from the mimeographed document "Hali ya lishe Wilaya ya Kasulu," (undated), which was issued by the District Government. This data is based on 1992 and earlier censuses.

Total Population, 1992 4,458

Under 5 Population

1995 923

1994 865

1993 n/a

1992 873

1967 Population 207,595

1978 255,649

1988 320,518

Growth rate, 1978-88 2.8% per year

Average Family Size

1967 4.4

1978 5.6

1987 6

Total families, 1988 52,000

Height/Weight g.t. 80%, age 0-5

1991 75%

1992 73%

1993 69%

Total Number School Students Male Female

1987 8280 4251 4029

1988 7945 4117 3828

1989 7796 4018 3778

1990 9877 5002 4875

1991 6381 3272 3109

1992 6521 3249 3272

1993 6366 3063 3203

1988 Population 854,817

Infant Mortality (under 1)/1,000

1985 115

1975 163

Child Mortality (under 5)/1,000

1985 192

1975 269

Literacy, 1978

Male 50.1% literate

Female 27.5% literate

Literacy, 1992 14.8% illiterate

% No Education, 1992

Male 23.7%

Female 38.9%

Per Capita Income

Kigoma Region Mainland Tanzania Kigoma, % Mainland TZ

1980 1358 Shillings 2072 66

1982 1680 2729 62

1984 3039 3811 80

1986 5254 6715 78

1988 10,318 12,076 85