DATA AND METHODS

This study is based upon fieldwork involving peri-urban settlement in Kunduchi and Kibamba, two divisions within Kinondoni district, each approximately 18–20 kilometres north and west of Dar es Salaam's city centre. This research took place during two phases: an 18-month ethnographic study conducted from 1998–1999, and a survey project conducted during a 6-month period in 2010. The principal line of investigations during the 1998–1999 visit was to visit and observe contemporary urban farming by residents who had lived in Dar es Salaam for a decade or more but subsequently moved to the city's periphery to engage in farming. To accomplish this, I had the opportunity to stay with families in two separate neighbourhoods: Kilongawima, which became a separate urban ward of Dar es Salaam in 2000, and had a 2002 population of 4,174 (United Republic of Tanzania 2004), and a neighbourhood known locally as ‘Agape’, named after a nearby Lutheran Church missionary centre. This low-density neighbourhood lies on the outskirts of the urban ward of Mbezi Centre, which had a total population of approximately 10,000 inhabitants. Since Agape was not an administrative entity, its population was not reported separately in the 2002 census.

A great many landowners of Kilongawima and Agape whose lives I learned about in 1998 and 1999 were former civil servants who had recently retired or received severances as the Tanzanian government reduced its footprint and privatised parastatals following the 1990 Zanzibar Declaration. Because many of these former civil servants had sought control of land in order to, among other things, create urban gardens and engage in animal husbandry, each neighbourhood represented an ideal location for participant-observation. Though by 1999, each area had been considerably built up with substantial housing, walls and fences, outbuildings and access to telephone service, electricity and water, the neighbourhoods were largely unsurveyed at the time; residents relied on informally registering their holdings at the local ward chairman's office, allowing them to secure an ‘offer of a right of occupancy’ from the Ministry of Lands and Surveys. A great many of the conflicts I encountered during my research in the late 1990s involved border disputes between neighbours, since the ‘offers’ upon which they based their tenure were vague as to property lines.

Yet, while the area represented low-density but built up urban neighbourhoods, there were a great many other individuals who did not have formal control of the land on which they were living. In 2010, I began a survey project to gain an overview of contemporary peri-urban residents, interviewing 612 adults of various backgrounds in the adjacent wards of Mbezi, Kimara B and Temboni, and 211 individuals in Kilongawima. In this survey of 811 adults residing in the neighbourhoods, 269 respondents, or about 33%, stated that they engaged in at least some cultivation. A majority, 169 respondents, were women, and the remaining 100 were men. The 269 respondents were split almost evenly between those claiming to own land (135) and those who did not (134).

Since the implementation of the Land Act of 1999, nearly all of the residents with whom I familiarised myself during my 1998–99 ethnographic research had arranged to have their holdings surveyed and formally registered in line with the stipulations of the act. However, there was also an intermediate category of landholders whom I met in 2010, who had acquired access to land under policies and directives that governed tenure prior to the Land Act but had yet to survey and register their holdings. While most with unsurveyed plots self-identified as ‘owners of land’ in my 2010 survey, their holdings were not formally recognised by the Lands Commission, thus putting them in a rather precarious state. They sometimes had substantial holdings, buildings and other infrastructure. But without formal registry, they left themselves vulnerable to the potential of either government confiscation or private interests claiming the land for themselves.Footnote ⁴

My principal aim in the 2010 survey was to learn more about the lives of the many individuals who lived in these areas, but did not control land. It was also hoped that it would serve as a longitudinal study to document changes that had occurred since the implementation of the Land Act, among both landowners and those without legally recognised ownership status. Each area had undergone significant changes, not only including those that came about through the Land Act, but because of changes of infrastructure, transportation and the growth of new urban agglomerations, many of which had not yet sprung into existence in the 1990s.Footnote ⁵ Although most of the 72 questions of the survey simply asked for a ‘yes’ or ‘no’ answer, they provided a vehicle offering the opportunity for myself and research assistants to ask qualitative questions about the experience and perceptions of suburban cultivators as we encountered them in the course of the survey. It is these descriptive data of the 269 self-described individuals involved in cultivation in the areas that represents the main focus of this article.

VARIETIES OF SUBURBAN CULTIVATION

One of the challenges in examining the lives of cultivators involved developing a vocabulary to discuss the variety of ways in which individuals practiced cultivation in and around Dar es Salaam. The English word ‘farming’ denotes a range of activities, including cultivating and animal husbandry, presumably taking place in rural areas. Those whose principal occupation is farming are commonly thought to be engaging in food production on a noticeably larger scale than horticulturalists or gardeners. Taken a step further, their activities are aimed at producing a surplus for market; though the term ‘subsistence farmer’ is sometimes used to refer to those whose production is largely earmarked for household consumption.Footnote ⁶ The term ‘farmer’, like its twin ‘peasant’, also represents a kind of foil in opposition to the concept of urbanite or town dweller, who presumably eschews land management for the purpose of producing food. Hence, the term ‘urban farmer’ has always represented a problematic one in Western discourse, since it seems to confound the categorical distinction between rural and urban life.

While its Swahili equivalent, kilimo, often is used with the same general range of meanings as its English gloss, residents of Kilongawima and Agape made a clear distinction between kilimo and ufugaji, or raising animals – even though both are often conducted together in many households around Dar es Salaam.Footnote ⁷ Householders with urban holdings extensive enough to keep sheep, pigs, goats or milk cows often referred themselves as farmers, wakulima; though neighbours would call their activities ufugaji. Since nearly all land has been allocated to private individuals, there little room to allow animals to graze freely. Much of the grass growing wild in the region surrounding Dar es Salaam has tiny sharp nettles rendering it indigestible for domesticated animals. Householders who keep larger animals found it more practical to grow edible stands of grass themselves, to collect and transport edible grass where available, or to use household vegetable waste as feed, rather than to bring their animals to grass growing in vacant lots or open areas. For suburban residents, it is almost axiomatic that those who kept pigs, sheep or dairy cows must have secure tenure and substantial economic means, since it was essential to build enclosures and supply animals with proper feed (Mlozi Reference Mlozi1997: 22–3).

At the other extreme, there were those who kept small seasonal vegetable gardens adjacent to their places of residence, as well as those who planted limited amounts of maize around the neighbourhood to take advantage of the Masika heavy rains, who simply referred to their activity as gardening (kilimo cha bustanibustani). As with its English equivalent, produce obtained through gardening was largely consumed within the household, or offered to neighbours as a goodwill gesture or in exchange for small amounts of money. Such gardens were seldom, if ever, a commercially viable endeavour or substitute for purchasing staples at local markets. A few of these gardeners raised chickens for eggs or household consumption, but the number of residents doing so has significantly diminished in recent years due to the prevalence of diseases, as well as the risk of theft. Such gardeners generally did not consider their activities to be kilimo – in fact, a great many engaged in such activities were women who considered themselves mama wa nymbani, whose principal occupation was raising children, and who did not consider themselves involved in significant income-generating activities.

Intermediate between the agribusinesses that support urban animal husbandry, and the casual urban gardener, were a number of individuals who had plots of as much as a hectare of land or more under cultivation, often along local streams, in lowlands where water collects, or other amenable open spaces still found throughout the area. They grew vegetables with commercial viability, including tomatoes, mchicha (a spinach-like vegetable), sweet potato leaves, okra and papaya, in quantities that clearly extended beyond their household needs. A few continue a long-standing practice of planting cassava, which is highly resistant to drier conditions between the rains. Those involved referred to it as hand irrigation or ‘sprinkle’ farming (mwagiliaji), which requires the laborious process of filling buckets by hand and splashing the water over crops. Several explained that they carefully controlled the amount of water they provided to each row of vegetables in order that they could control their ripening to meet the demands of the local markets. Some also took advantage of seasonal springs that become available when the water table rises during the rains, or dig catchment pits enabling them to extend the time in which they cultivate for additional months beyond the rainy seasons.

Though such intermediate level cultivators are producing for market, many did not control the land on which their fields were located. A few obtained permission from TANESCO, Tanzania's power utility monopoly, to cultivate under power lines that crossed the Mbezi area. Others illicitly cultivated along the road reserves, a 14-metre clearing on either side of the Morogoro Road, clearly marked with concrete pylons and considered strictly off-limits to cultivators. Open space cultivation on utility land is commonplace throughout Dar es Salaam, and mostly supported by the Land Act, which grants that cultivators, with the permission of utilities controlling non-hazardous open spaces, be allowed to obtain a residential licence to utilise a plot (Tanzania 1999 V.23.(3) a–c). The only stipulation was that cultivators avoid tree crops or other plants that interfere with transmission lines or the utility's facilities.

The remainder of the land, representing most of the open cultivable land in Mbezi and Kilongawima, is currently under rights of occupancy of private individuals, meaning that cultivators have made arrangements with the owner to cultivate. It was surprising that many of these intermediate cultivators, with the full consent of the individual registered as the land holder, were cultivating and often living on the plot, not exactly as a renter, but as a ‘guard’ of the land who used and occupied it rent-free. Many of those on these plots therefore had clientistic relations with an officially recognised, but often absentee landlord.Footnote ⁸ Sometimes, these occupants were relatives of the landowner. But a great many others, representing the majority, had no familial or ethnic affiliation with the absentee owner.

An explanation for this widespread practice is that the Land Act, though somewhat more liberal than earlier ordinances in security of tenure, mortgaging and sale, only recognises use rights in land. Individuals of economic means may act as absentee landowner, perhaps holding multiple plots in various parts of Dar es Salaam, provided that the land remains developed. Once registered, the registrant was expected and required to put the land into productive use, by either cultivating, or developing ‘unexhausted improvements’ on the parcel. Though an individual who failed to develop land may have had some slight risk of government confiscation, there was also the possibility that others might occupy the land and challenge the current owner's right of occupancy. Additionally, many feared that a vacant lot might become an illicit dumping ground, in which case it might be declared hazardous or abandoned (Tanzania 1999 VI.51·1.a–e), thereby risking forfeiture.

Such relationships between landowner and cultivator often began when an individual was hired to clear or maintain land or care for animals. In such cases, the individual developed relationships with non-landholding non-relatives who had lived in the area for a substantial time, and were willing to let them cultivate and perhaps build a wattle-and-daub shed or house on vacant property to give it the appearance of development. Since many became familiar faces in the neighbourhoods in which they initially worked, they often used this opportunity as a stepping stone to find another neighbour willing to employ them for temporary or day labourer jobs. If and when the owner is ready to build on the land and request their removal, their familiarity with the neighbourhood helps them locate other vacant properties, controlled by absentee landlords, from whom they can arrange permission to occupy or cultivate.

The degree of implicit trust in such arrangements may seem somewhat counter-intuitive; the conventional wisdom is that Africans would only trust their own kinsmen or members of their own ethnic group to take on this degree of responsibility for protecting resources. But in some instances, these landless residents have lived in the area for many years, and perhaps proven to the satisfaction of neighbours that they can be sufficiently responsible to protect the interests of those with rights of occupancy. In Kilongawima, several long-established landless residents had been chosen by their neighbours to serve a political role as wajumbe, or ten-house cell leaders. Their role suggests that they have come to represent ‘indigenous’ inhabitants of the neighbourhood in relation to the landed but more itinerant neighbours who have only settled in the area in recent years.

Another stark example of such informal arrangements is a large tract straddling Mbezi-Temboni, referred to as ‘kwa Msuguli’. General Msuguli successfully led the 1979 invasion of Uganda that ousted dictator Idi Amin. As reward for his services to his country, Tanzania provided him with a substantial tract of land. His family, partially in deference to the principles of socialism prevalent during Msuguli's day, allows individuals to cultivate in open undeveloped spaces with the understanding that they are obligated to leave if and when the family requests. Some of these cultivators are renters living nearby in neighbouring developed areas. Others have no other home, and live in wattle and daub houses within the Msuguli tract. Thus such individuals, having never enjoyed rights to land in the area, nevertheless represented long-term neighbours to those with more permanent rights of occupancy in the area.

DILEMMAS: SUBURBAN FARMING WITHOUT FARMERS

While many individuals cultivated, involving themselves in kilimo to some extent or another, very few self-identified as farmers, wakulima. Early in the course of field testing the survey instrument, it became apparent that this represented a particular challenge for studying their situation. The survey posed a series of questions about occupation: were individuals salaried employees, retired, did they run a business, work in a trade or as a day labourer? Consistent with field experiences when I investigated urban cultivating in 1998 and 1999, some of those who controlled extensive holdings and also kept at least some animals in Kilongawima and Mbezi, volunteered that they were farmers. They often described their income-generating activity as ‘agribusiness’, even if selling crops, eggs, milk and meat typically represented only one among a number of different income-generating activities: such urban farmers worked in the city or were retired, ran additional small businesses, or exploited local resources in demand around the city.Footnote ⁹

Not surprisingly, the household gardeners neither considered themselves farmers, nor their principal occupation to be farming. However, what initially came as an unexpected outcome was that nearly all of the intermediate open-field cultivators, who clearly spent considerable time tending fields with commercial-scale production, insisted that they weren't farmers either. Many of these cultivators had other occupations such as running vigenge, or open market stalls along roadsides, performing trades such as tailoring or hairdressing, or casual labour such as feeding and caring for animals belonging to local landowners. Of course, those with an occupation or business outside cultivating generally considered themselves first and foremost to be business people, tradesmen or employees, but not farmers. A few portrayed their cultivation as an mradi or small-scale income-generating scheme ancillary to their main occupation.

But even the many who had no other income-generating activity other than cultivating bluntly told me that work they were doing in the fields was not their occupation at all.Footnote ¹⁰ Illustrative is a conversation with a senior individual who had been living in the Mbezi area for about 20 years. He was born and raised in Tanga District, where he had worked for many years as a day labourer at various tea and sisal plantations. About 20 years prior, he became sick, and received treatment at a hospital in Dar es Salaam, where one of his children lived. After his recovery, he remained in Dar es Salaam, never securing a permanent job. He had been cultivating crops such as tomatoes, corn, mchicha and other vegetables since that time, on land on which he had no right of occupancy. But in spite of having a long conversation about his life story in the middle of the nearly three hectare field he cultivated just off the roadside in Mbezi, he insisted that he had no occupation. So far as he was concerned, he had long left the world of work, and had cultivated simply to pass the time, yet could not rightly be considered a farmer.

I was fortunate that many of these individuals were willing to patiently explain to me, a somewhat naïve North American, why many felt that they were not in fact farmers. As I listened to their explanations, they struck similar notes, offering two broad but similar reasons as to how it is possible that they could spend considerable time engaged in kilimo without being wakulima. More senior members of the community, including the above-mentioned individual, more often than not offered this first perspective: In their view, farming was not a vocation, but something that they as Africans simply viewed as being part of their daily round of activities. In the rural villages in which most were born and raised, they or their senior relatives engaged in multiple income-generating activities, from day labour on local plantations, to small-scale trading or even local government employment. These potentially extra income-generating activities defined one's job or vocation. And while crop cultivation often resulted in production for local or regional markets, the farm or garden was an integral part of the maintenance and well-being of the household. Thus everyone was expected to participate in cultivation and animal husbandry to some extent or another. In contrast for those living in the industrialised West who feel that farming represents both a vocation and an occupational category, many Tanzanians did not necessary view farming as a distinct occupation.

The second explanation contrasted with the first, but offered a window into the dilemmas and anxieties of many suburban residents, younger ones especially, whether or not they had petty income-generating activities outside of farming or were unable to find other employment. Numerous informants – even those with extensive fields under cultivation – stated that they weren't farmers because ‘we do not have capital [Mtaji]’. Though the kinds of crops they were producing – tomatoes, mchicha, cucumbers, seemed to be directed to market, and yields seem to be sufficient to allow individuals to make some profit, and earn a modest living, they did not feel that it was truly farming unless they had sufficient inputs to produce something of scale. When citing examples of capital, some suggested having rights of occupancy in land would allow them to expand operations, but no one in my survey openly stated that lack of space was at issue. More often than not, they listed such amenities as electric water pumps, vehicles, fertiliser or structures as the kinds of capital that would enable them to scale up sufficiently to be committed farmers.

It seemed that there was an implicit comparison with their landed neighbours, many of whom engaged in far more intensive urban gardening and animal husbandry on relatively small plots. Several of those with extensively developed rights of occupancy did volunteer that they were primarily farmers. This was especially true of retirees, who arrived in the areas shortly after receiving their severances, which they used to invest in developing their holdings to engage in cultivating and animal husbandry. Most had arrived in the area at a time when the ideology of Ujamaa, or rural socialism was still in vogue, and the area still retained much of its rural character. However, by the early 2000s, each area's character had shifted dramatically, and those engaged in ‘agribusiness’ and who were able to support resource-intensive crops such as bananas and pineapples had the economic means to introduce inputs that were only available through the urban infrastructure, such as deep water wells, pesticides, fertilisers and electric pumps. Many of those without rights of occupancy, landless, and viewing these long-term residents from afar and likely unfamiliar with their origins, nevertheless were keenly aware that they had the capital to produce on a much larger scale than those without such amenities.

Tellingly, there exists another category of entrepreneur whose members also occasionally identified themselves as farmers. The shoulders of the main trunk roads leading out of Dar es Salaam, including those that pass alongside Mbezi and Kilongawima, are commonly alive with hundreds of ornamental plant seedlings, either sprouting from paint cans or in polypropylene plastic bags filled with top soil. These bagged plants represented the inventory of seedling vendors. While a few of the plants being sold by roadside vendors are vegetables or fruit tree seedlings, most are ornamentals with no agricultural use. Local landowners represent the vendors' customer base, purchasing them to decorate the areas surrounding their households. Vendors often work in groups, and most have no rights of occupancy in their roadside stands. However, like farmers in local fields, their principal responsibility is to make sure that seedlings are watered and guarded against theft or damage. They also periodically travel, some as far as Mozambique or Zambia, to obtain inventories of exotic ornamental seeds not native to Tanzania. Though their agribusiness involves no tilling of land, they sometimes identified themselves as farmers because of their capital investment in seed and plant inventory. But once again, their business model presumes a customer base of landowners with the means to afford to decorate their landscape with ornamental plantings.

PROSPECTS: SUBURBAN CULTIVATION AND THE SHIFTING LANDSCAPE

Although many of the now planned and surveyed neighbourhoods previously described were already in existence by 1998 and 1999, to a large extent, most of the areas in question were still underdeveloped during that time. Ordinarily, those with titles to the land, whether obtained through formal registry with the Ministry of Lands, or through other then-recognised but less formal kinds of tenure, invariably lived within easy reach of major trunk roads like the Morogoro or Bagamoyo roads. Or, they preferred to acquire land on the outskirts of hamlets found scattered throughout the Dar es Salaam region that traced their origins to Ujamaa villages that had been created and provisioned by the Tanzanian government in the 1970s. Being close to trunk roads or population centres assured easy access to transportation, as well as to inputs less likely available at a distance, such as piped water, electric lines, telephone service and veterinary services.

Another common sight in the 1990s were absentee farms and gardens belonging to weekend farmers, who lived and worked in Dar es Salaam, but raised vegetables for household need and occasional sale on the open downtown market. The national infrastructure – especially roads connecting Dar es Salaam to other regions of the country – had deteriorated significantly during the 1980s and 90s when the nation had suffered a number of economic setbacks. There therefore was a demand for agricultural products in Dar es Salaam, whether an urbanite grew the crops themselves, or brought them to a city market while still reasonably fresh. Many of the longer-term landless residents of the present-day neighbourhoods began as hired hands who could break soil for these commuter gardeners, or acted as guards of the fields while their owner was absent in the city. Being close to a concentrated population or transportation hub assured not only access to an urban garden on the outskirts of the city, but also a ready market for produce, eggs and meat from more developed residential peri-urban farms.

Being an early resident, as well as being at arm's reach from the city and population centres in an emerging but as-yet undeveloped neighbourhood, assured adequate space in which to engage in peri-urban agriculture and animal husbandry. Many of the earliest residential settlers in these neighbourhoods started by building one-room square sheds, not only as a hedge against accusations that they were not utilising the land, but also as a temporary dwelling as they attempted to establish a homestead. Often, they used their sheds as the first room of what eventually became a ‘Maputo’ or Mozambique-style house – a modular rectangular design in which rooms could be added on to the original shed in stages as funds became available to the owner. If successful at cultivation or animal husbandry, it was inevitable that a household would also include outbuildings to keep animals. And even by the late 1990s, fences and walls around a compound had become commonplace – to keep animals in, unwanted intruders out; but most importantly, to demarcate the boundaries of the proprietor's urban farm at a time when marked and surveyed plots were few and far between.

The remainder of this article focuses on the factors that have governed and likely will continue to govern the prospects of suburban cultivators. Although I have already mentioned that conditions on the ground favour the fortunes of those with a recognised right of occupancy over land in these areas, and conversely create a precarious situation for those without such rights of occupancy, there is nevertheless a shift from the input-intensive, residential farming and animal husbandry, to the open form of interstitial gardening practices by most non-land holding residents. Thus, while landholders enjoy a relatively more secure comfortable position, it appears less and less likely that farming will play a major role in the repertoire of activities assuring their livelihoods.

The reasons do not necessarily have to do with availability of space or security of tenure. Rather, it involves other historically significant changes in Dar es Salaam in particular and Tanzania as a whole: availability of water and competition of markets for produce. Many of the vanguard are giving up on suburban residential farming, towards utilising their land primarily as a residence, pursuing agriculture in other areas beyond Dar es Salaam if not giving up farming entirely. On the other hand, those precariously holding on may be better able (or have no choice but) to weather changes brought about by spiralling infrastructural costs of resources such as water, or shrinking profit margins due to competition within urban markets.

In the 1990s, many of these larger-scale farmers relied on the municipal water supply. Major water mains were located along the Bagamoyo and Morogoro Roads, major trunk roads that pass the areas where the earliest of residents settled. Initially, homeowners connected PVC hoses between the main and their houses or farmsteads. Since these plastic hoses frequently developed leaks, unpaved roads and trails that led to people's houses and businesses often were flooded with puddles. And because of the rudimentary manner in which the water supply was connected, they paid bills based upon estimates of water usage – something that one long-time resident described as ‘stealing from the community’, since most engaging in agriculture used far more water than their estimates suggested. This plentiful supply of water north and east of the city enabled more affluent householders to grow water-intensive crops such as bananas, okra or peppers, and made it possible to keep larger livestock such as milk cows and pigs. Often non-landholding residents of Mbezi and Kilongawima could gain permission to use taps belonging to those who controlled land – and frequently, they benefitted indirectly from the runoff created by leaky pipes and water that flowed off more substantial peri-urban farms. (see Drechsel & Dongus Reference Drechsel and Dongus2009: 73).

At the other extreme, there are still many residents without any access to a nearby municipal water source in more remote areas of Mbezi and Temboni. Like many areas of downtown Dar es Salaam unserviced by municipal water, householders depend on entrepreneurs, often young men on bicycles lugging large plastic jugs to sell door-to-door, for their day's supply of water for drinking and washing. Private water delivery has become such a big business throughout Dar es Salaam as a whole that, even DAWASA, Dar es Salaam's water utility, maintains a fleet of trucks to deliver water from its grounds to individual household customers or small-scale distributors beyond reach of their water mains. There have also been accusations of unscrupulous entrepreneurs drawing water from municipal water supplies, loading it onto trucks, and selling it at a mark-up to neighbours who lack access to piped water. In Mbezi Centre, one vendor has even been accused of depleting the local water supply by selling municipal water from DAWASA supplied pipes. Neighbours complain that they had been forced to buy water that should be flowing directly into their taps from the vendor who is capturing it closer to the Morogoro trunk road.Footnote ¹¹

A great many of the landless cultivators around Mbezi are concentrated around the Mbezi River. Kilongawima is in a similar situation, since many of the non-landholding urban farmers with any sizable fields cluster around a runoff stream created by wastewater from the Kotex plastics factory, located on the Africana Road, just south of the ward. Except for a few who have been able to acquire petrol powered or manual foot pumps, they engage in mwagiliaji, carrying buckets of water from the stream to nearby gardens daily, or create short irrigation ditches to feed fields. The very existence of streams and other out-of-season sources of water for urban farmers was only possible through the existence of medium- and high-density housing in the Mbezi and Kilongawima areas and the existence of an urbanised infrastructure. Elders who had lived in the Mbezi region for the past 40–50 years informed me that the Mbezi ‘river’ used to be a seasonal stream that only flowed during the months following the Masika rains. For most of the year, it had remained dry. Due to the presence of houses that rely on municipal water supplies, runoff has assured that the river flows, and water is available year round for the largely landless cultivators who work in the vicinity of the river.

Unfortunately, while a great deal of the runoff represents usable unprocessed water that has leaked from households and industries, some of this runoff comes from untreated wastewater. I myself observed on numerous occasions that there are still many houses whose toilet pipes simply flow out the house and spill into nearby hills and valleys. There was also the ubiquitous presence of latrines that were located in gullies that likely formed as runoff channels during the rains. Also, in the areas near the Morogoro Road, there is runoff from the road itself. It is also a common sight to see individuals washing their laundry in the river each morning, and laying clothing along its banks to dry. Most producers attempt to avoid these contaminated water sources, but there is always a question due to the proximity of their plots to these contaminated sources. Thus, landless suburban cultivators often face a situation similar to urban farmers in the Msimbazi river valley and denser areas of Dar es Salaam, in that householders are uncertain whether it is safe or not to eat locally grown produce, which may serve to dampen local demand for their surplus.

MARKETING PRODUCE WITHIN THE CITY

Often, the market for these crops fluctuates significantly based upon two factors. The first depends upon local demand for their produce, and the second revolves around competition with producers from other regions of Tanzania. In some instances, producers market directly to nearby customers in the local neighbourhoods in which they cultivate. A neighbour can always visit a field, and sometimes select produce that the farmer harvests and immediately hands over to the customer. A common sight on roadsides and residential settlements are vigenge, or makeshift stands selling produce and other goods, and a few of the more entrepreneurial cultivators retail their own produce to nearby residents, and even passing motorists. If an individual's produce is something of high demand to specialty customers, such as grass for cattle feed, pineapples or avocados that are popular among tourists staying at hotels along the Indian Ocean, producers can prearrange sales, or at the very least, maintain a steady relationship with a specific customer assuring that as these things become available, the customer will buy them.

A great many other cultivators depend on intermediaries to purchase their produce and deliver it to customers. Such intermediaries do not themselves usually participate in cultivation or animal husbandry, but go from field to field several times a week to buy ripe produce to sell elsewhere on the urban market. In the course of the survey, I ran into several of these intermediaries, who explained that they usually deliver a day's collection to the stands and along roadsides and in market areas such as Mbezi Centre, Kibanda cha Mkaa or Afrikana. Many local cultivators are unsure exactly where these produce buyers sell their product. Likewise, for landowners with dairy cows, they also depended on morning visits by milk collectors, who would often pour the 1 or 2 litres that each cow produced into plastic containers that were then delivered to wholesalers who had distribution networks throughout Dar es Salaam.Footnote ¹²

There was probably a period in the 1970s, 80s, and into the early 90s, when local produce of this sort played a significant role in feeding the city. This development helped account for the presence of fairly large and sophisticated urban agribusinesses that proliferated in peri-urban areas in the 1980s and 90s, often utilizing several hectares of land, and including not only garden crops supplying household needs, but also the keeping of livestock. Such urban farmers were distinguishable from the casual urban gardener because of their heavy reliance on the municipal water supply, electricity and other urban inputs. They also commonly employed farm hands to maintain operations. While occasionally these more sophisticated operations were found in densely settled wards closer to the city centre, over time, these intensive farmers tended to flee to lower-density settlements as the city grew. Such areas allowed ample space within which to garden and raise livestock, and greater ease of access to natural resources that served as inputs into their farms. Additionally, low-density settlements at a distance from the urban centre meant fewer complaints from neighbours or close attention by government officials.Footnote ¹³

Aside from the fact that difficulties in obtaining sufficient and reliable inputs of water, fertiliser and arable land impose limits on the scale of commercial cultivation and animal husbandry possible within the suburbs, there is another factor that has changed the face of suburban cultivation. With improvements in the roads and transportation over the past decade, small urban producers face stiff competition from other sources of produce. In surveying vendors at local produce stands and shops carrying vegetables, I learned that a majority do not buy their supplies from neighbouring suburban farms, but take local transport weekly to the wholesale markets Manzese or Kariakoo, closer to downtown, which in turn obtain their produce from more distant regions of Tanzania such as Tanga or Morogoro District. Considerably more produce is brought into the city than in past decades due to better connections with more distant regions.Footnote ¹⁴

This increase in the reliability of transportation has also affected the market for milk and other products of animal husbandry. For urban farmers with the economic means to cultivate and keep animals, many have opted to seek grants of occupancy in Bunju, Kibaha or Bagamoyo regions, just beyond the limits of city expansion. Alternatively, many have sought land as far away as Kisarawe, Morogoro or Tanga, where there is considerably more space to engage in either animal husbandry or agribusiness, and easier access to needed inputs to support them. Whereas once having pigs or a sizable herd of milk cows represented a meaningful income-generating source, allowing the opportunity to employ several hands to feed and milk the cows, now many of those former farm hands have found themselves unemployed. Many individuals who initially came to the area to work as farm hands were later turned away because an owner of cattle, pigs or goats scaled back their operations and could no longer afford to employ them. Some have taken up cultivation on whatever space is available, while others search for various kinds of day labour opportunities, frequently expressing uncertainty about their futures.

CONCLUDING THOUGHTS

The historical, infrastructural and legal circumstances though which these peripheral settlements were formed, as well as the clear distinction between relatively affluent residents who enjoy a security of tenure and others who have a tenuous and somewhat precarious situation in these areas, help define the suburbs of Dar es Salaam. Unlike either rural producers or committed urban farmers seeking niches in the central city upon which to cultivate, the endeavours of suburban cultivators may be best understood in terms of the relationship between their activities and the decentralised but now largely private interests of those controlling access to land and resources in the suburban neighbourhoods in which they are situated.

This distinction, in my view, warrants a reassessment of the nomenclature that often designates these peripheral settlements as either ‘urban’ or ‘peri-urban’. Although there is often considerable ambiguity and overlap in the distinction between suburban cultivators and, on the one hand, rural peasant, and on the other, urban farmer, the differences are centred upon their relationships to the configurations of tenure that are to be found there. Residential suburban cultivators, for the most part, either control surveyed land, or cultivate through formal or informal arrangements with private landholders. A great many of those cultivating in the suburbs do not even consider themselves farmers, reserving that term for the few with the capital to invest in infrastructure to cultivate and raise animals on a commercial scale.

The fact that many individuals engaging in peri-urban cultivation do not consider themselves farmers reflects the findings of a recent World Bank study of other cities in the developing world, suggesting that access to credit and capital increasingly defines farming in many developing countries (2013: i–iv). One of the infrastructural necessities is surety of tenure as defined by a secure sense of land ownership. In the case of Dar es Salaam, those engaging in cultivation, whether casual rain-fed gardening or somewhat more intensive farming and animal husbandry, depend upon the urban infrastructure into which cultivators have immersed themselves. In many cities throughout Africa, it almost ubiquitous that farming represents an adaptable strategy compatible with other concurrent urban occupations (2013: iii). This tends to be the case in Dar es Salaam as well to a certain extent; however, this generalisation applies more often to those with a land base upon which to build a repertoire of income-generating activities.

Because suburban farms are generally small, and cultivation requires open space, they are usually quite visible on the landscape to the casual observer. Suburban farms therefore have the qualities that lend themselves to serving as civic spaces, in much the same way that bars, houses of worship, verandas and markets play similar roles in a fully urban setting (Lewinson Reference Lewinson2007). Perhaps the only possible exception would be for those who could afford to place their gardens and livestock behind fences and walls within their households. For less fortunate suburbanites, including those who engage in open space farming, their civic engagement is quite often visible to the public in much the same way as urbanites' activities occur on street corners or in public parks. Hence, in the suburbs, the more affluent residents announce their presence on the landscape to the casual observer indirectly through their property and conspicuous consumption; the less fortunate make their social relations of production known more directly through their bodies and labour.