The “lords” of Malombe; an analysis of fishery development and changes in fishing effort on Lake Malombe, Malawi - M. Hara and E. Jul-Larsen

1. INTRODUCTION

For the last 15 years or so, in much of the grey literature as well as in the general narratives about the status of freshwater fisheries in Southern Africa, the fishery of Lake Malombe has been considered a crown example of a common tragedy. As has been shown elsewhere (Zwieten, Njaya and Weyl, 2003) the fishery has seen a rapid and considerable growth in fishing effort. Great changes occurred in the period from the 1950s to the 1980s with regard to types of gear used and species targeted. Lately the dominant Chambo (Oreochromis spp.) has almost disappeared and total catches have decreased dramatically. This development is very different from the situation on other African lakes. On lakes like Mweru or Chilwa total catches have remained fairly stable despite a considerable increase in fishing effort. These fisheries are analysed in other papers in this publication.

The objective of this paper is therefore to investigate in some detail the characteristics of the effort development on Lake Malombe and to provide a sociological interpretation of the changes. We will do so by distinguishing between population-driven and investment-driven changes in effort. Ottar Brox (1990)^[50] defines population-driven growth as increase in fishing effort resulting from increased human participation in fishing. This growth is often characteristic of the frontier phase of the development of national economies. In the context of Lake Malombe, it means increases in number of gear owners and crew members. Investment-driven growth is defined as growth in capitalization and in technological level. In our case this means increase in number of fishing gears per unit, improved technology and/or technological innovations of gear in use. We will demonstrate that until the 1970s, growth in effort was both population-driven and investment-driven, but that population-driven growth probably started to slow down sometimes in the 1970s and that it almost stopped around 1980. Since that time the number of production units has remained fairly stable until in the middle of the 1990s when it started to fall. During the 1970s and the 1980s the growth in effort was almost exclusively investment-driven, i.e. a growth due to accumulation of equipment and technological changes within the existing units. During the 1990s the reduced number of production units has been followed by reduced investments within these units so that effort reduction during this decade has been caused both by less people and less investments.

Our analysis of the sociological characteristics in the Malombe fishery shows that the dynamics behind effort development are quite complicated and only by taking account of the background of the owners and their relations with the rest of the population in general, and with the crew members they hire in particular, can the effort development on Lake Malombe be properly understood. The analysis demonstrates how labour migration options, owners’ possibilities to invest in alternative economic activities, as well as their ability to control their labourers and to manage their units, are all factors influencing the effort development just as much as do the volumes and the prices of catches.

As Brox (1990) demonstrates, investment-driven growth has generally different and often more serious and long lasting sociological effects than population-driven changes. The fishery on Lake Malombe (and on the southeast arm of Lake Malawi) represents one of the few cases known in southern African freshwater fisheries where investment-driven growth in effort -although moderate - has been more important than population-driven growth. It is therefore also the objective of this paper to explain how this development was possible and what seem to be the main constraints confronting further growth of this type. We will show that the investment-driven growth depended completely upon an influx of financial resources from outside. We also show that in the type of economy and production system that prevails around the lake, there are several mechanisms that reduce the incentives and the possibilities for investment-driven growth. Most of them seem to be associated with the particular type of patron-client relationship that exists between boat owners and crew members. Fisheries on Lake Malombe have for the last 25 years or so been dominated by less than 400 boat owners, of whom only 200 remain in the fishery. Within an estimated total population of more than 50 000 people the gear owners, in light of their economic wealth and social influence, are easily seen as some kind of lords. However, their lack of control over their own production units is an excellent illustration of the investment problems reigning in many African fisheries.

1.1 Method

Three sources of data and information have been used in this study: a field survey, relevant literature and archive material, and data from a parallel study on the biological development in Lake Malombe.

The approach used in the field survey was to build life histories of gear owners and crew members. For this, 42 gear owners (22 from the west bank and 20 from the east bank) were interviewed. 15 crew-members were interviewed (six from Malombe west and nine from Malombe east). In addition to interviews, a list, compiled in 1993 by the Fisheries Department, of all gear owners and details of their fishing gear was used to evaluate changes in gear numbers and gear ownership between 1993 and 1999. The Fisheries Department field staff (Messrs Kasuzweni and Thindwa on the west bank and Bezai on the east bank) helped in providing an update of the list. Where possible, the individuals on the list - or their relatives - were interviewed.

The second important source of information was literature concerning the historical political economy of agriculture and fishing in southern Malawi. Four authors - Chirwa (1995), Mandala (1990), McCraken (1987) and Vaughan (1982) - were particularly valuable. Furthermore, official government reports and other archival material relating to the period in question have been used extensively. Finally, the findings about trends and changes in fishing effort is also based on the parallel investigation into the biological explanation for the development of fishing effort within the framework of the same project (Zwieten, Njaya and Weyl, 2003).

1.2 Lake Malombe, its present fishery and the people

Lake Malombe is described as “an impoundment of the outflow of Lake Malawi via the Shire River, 12 kilometres south of Lake Malawi” (FAO, 1993:2). The lake is about 390 square kilometres in area and averages seven metres in depth with a maximum depth of 17 metres. It experiences the same climate regime as the southern part of Lake Malawi. This means that the influence of trade winds on productivity is paramount. Unlike in Lake Malawi though, rainfall run-off is believed to contribute significantly to the productivity of Lake Malombe. The hydrological parameters of the Shire River are nevertheless crucial for the conditions of the lake. In the early decades of this century the water levels of Lake Malawi were extraordinarily low and the outflow of the lake is reported to have stopped completely around 1915 (McCracken, 1987:417). During the next 20 years there was no outflow from Lake Malawi and the water levels in the Shire and Lake Malombe therefore varied dramatically^[51]. It was only after the Bomani floods of 1939 that the water levels became less unstable (Mandala, 1990:6).

On the west bank side, the lake is bordered by the Mpiri Piri hills, which lie within three to seven km from the lake while on the east side it is bordered by the Mangochi hills. On the Southeast banks is the Liwonde National Park. The fishing villages are thus confined within narrow strips of land along the lake on both sides making the population densities very high and the amount of farming land within the village areas very little. This is especially so on the west bank where the population density averages over 500/km² (Bell and Donda, 1993). On the west bank is the main road from Lake Malawi and Mangochi Township to the urban centres of Zomba, Blantyre and Lilongwe. The east bank is very difficult to access. This is especially so in the rainy season as the area then can be reached only through a dirt road that branches off from the Mangochi-Namwera road.

A census conducted by the Ministry of Agriculture (Liwonde, 1998) showed that 8 396 “farm families” existed in the villages on the west bank while another 2 657 families were to be found on the east bank. Using five persons per household, which is the national average household size, the population in the fishing villages can be estimated at 42 000 people for the west bank and 13 300 for the east bank. The annual rate of population growth in Mangochi district is reported to have been 1.7 percent per year in the period 1987-1998 (Malawi Government, 1999:19).

Our surveys showed that in June 1999, boat owners operated 173 production units - all types included - from the western shores of the lake while 22 units were operated from the eastern shores. Each production unit is owned by one person and the average number of crew members in the units is estimated at ten. Based on these figures, we see that only two per cent of the families on the west bank and less one percent of those on the east bank are owners of fishing units. On the west bank fishers represent no more than approximately four percent of the total population while they constitute less than two per cent on the east bank.

While the figures might underestimate the real numbers of people in the fishery, it is difficult not to view these figures as an illustration of the rather modest role the fishery is playing in the local economy although the total population figure includes children under the age of 15 years and adults over 65 years of age; groups that we must assume to be economically inactive. Furthermore, the figures only say something about the quantity of people involved and nothing about the financial importance of the sector. Nevertheless, they clearly indicate an important aspect of everyday life in Malombe which often tend to be overlooked: for most people fisheries is neither the only nor the most important economic activity.

2. DEVELOPMENT OF THE FISHERIES BEFORE 1950: POPULATION-DRIVEN GROWTH IN EFFORT AND DEVELOPMENTS OF MARKETS

The development of fisheries around Lake Malombe should not be seen in isolation from what has happened in the far more important fishing area constituted of the Southeast Arm of Lake Malawi located less than 15 km to the north. Many of the fishermen exploiting Lake Malombe regularly have been exploiting Lake Malawi as well. In these two areas people are generally of the same origin with the same language and culture. Most of them identify as Yao, a tribe originally from an area east of the Rovuma River in Mozambique. This tribe has dominated the local political structures in the area since the 1860s. The economic tradition of the Yao is one of trade, developed through their long contacts and relations with Arab traders on the East Coast of Africa. The contact with Arabs also greatly influenced the Yao to adopt Islam (Rangeley, 1970). Thus although Christians are found in the area, the great majority of the population are Moslems.

Furthermore, macro-structural factors in the Malombe and Malawi fisheries - both political and economic - have largely been the same. Most of the Malombe area belongs to the same administrative and native authorities as areas along Lake Malawi, with Mangochi town (formerly Fort Johnston) the administrative centre located equidistant from the outlet of Lake Malawi and the inlet of Lake Malombe. The broad lines in the development of Lake Malombe fisheries are therefore the same as those of the southern parts of Lake Malawi. This is also reflected in the literature in which fisheries development on the two lakes tends to be dealt with together.

McCracken (1987) reminds us that commercial fishing in the area is an old venture and not something that started emerging in the 1950s. One of the major events that stimulated fishing for sale on the southeast arm of Lake Malawi is said to have been the need to provide fish to South African soldiers going to East Africa during the First World War. On Lake Malombe it is reported that the flood in 1925 led people to turn to fishing. People’s gardens had been inundated and as a result, many peasant farmers switched to fishing or fish trading as a means of raising money to buy food and, for men, also to pay their hut tax. But due to the unstable water levels in the Shire we must expect that extensive commercial fishing in Lake Malombe must have been difficult to perform before the water levels of the river stabilized.

In the 1920s and 1930s European and Indian traders had started to transport fish by lorry from Lake Malawi for sale in Zomba and Blantyre. Partly, this was fish they bought from African fishermen and partly it was their own catches from commercial production units that emerged in Lake Malawi after 1930. Also, a great number of Africans started to engage in fish trade, transporting fish by bicycle (McCracken, 1987). The different groups of traders demonstrated an interesting mixture of competition and collaboration where African traders often represented elements in the bigger trade networks of the Asians and Europeans. Given that the trade route from Lake Malawi runs along the western bank of Lake Malombe, we must assume that an increasing part of its fish was integrated into the same system, even if the centre of the commercial activities was located further north.

People had historically been using fish traps, weirs and nets made from Bwazi or Chopwa (a fibre from Poolzolzia hypoleuca). The 1930s also saw an increase in use of twine from disused bicycle tyres (locally called Linye) for making nets. At the same time, the Indian and European (mainly Greek) production units on Lake Malawi introduced manufactured cotton (later nylon) nets. They also introduced new technologies such as beach seines, and from 1938, trawling. If African fishermen only modestly copied these techniques in the years before 1950, we assume that the main reason was related to a combination of costs and the relatively high efficiency of their traditional gear.

Throughout the 1920s and 1930s, the colonial government made no attempt to stimulate the growth of the fishing industry. This was partly attributable to its policy of keeping Africans as a labour reserve for the European settler estates in the Shire highlands (Vaughan, 1982) and also a deliberate policy that attempted to force or coerce people in rural areas to go into cotton farming at the expense of food crops or other economic activities such as fishing (Mandala, 1990; Vaughan, 1982). The first official reports concerning fishing date from around the late 1920s. Two concerns seem then to have been of importance for the authorities: on the one hand the state of the resource and, on the other, peace and order.

Most important were probably questions related to peace and order. The increasing entry of European and Asian investors into the fishery and the emerging conflicts among various groups of Africans led to increased tensions. Authorities saw it as being extremely important to minimize these conflicts and tensions. Interestingly, and for various reasons, the policies of the government did not favour the foreigners; on the contrary one may say that most of the regulations introduced during this period favoured the position of the African fishermen compared to their European and Asian competitors. While African fishermen were largely left to themselves, a number of restrictions concerning licenses and gear were put on the foreigners (Chirwa, 1995; McCracken, 1987; Hara, 2000). However, this mainly applied in the Southeast Arm since the government had banned non-African fishing on Lake Malombe^[52].

The other concern was related to the state of the resources. On this issue the government was ridden by rather paradoxical influences of thought. In the first half of the 1930s, the concerns seem mainly to reflect the well-documented tendency among colonial officers to view African methods of hunting and fishing as detrimental to the ecology and the reproduction of exploited species^[53].This led to the prohibition of most of the African fishing gears such as traps, weirs and poison. These regulations are reported not to have had much effect since local authorities did not enforce them. Towards the second half of the 1930s government attitudes changed and they came to see the non-African fishing methods as being the most challenging ones. This meant that fisheries on Lake Malombe were, to a large extent, left unregulated by officials except for the 1933 requirement to pay fees on canoes, through local chiefs, under the Native Authority Act (Chirwa, 1995:361-5).

The first attempts by the colonial government to stimulate the industry were in the early 1940s following two surveys by British led specialists, one of which had been led by Professor Platt in 1938 (McCraken, 1987). The report had pointed out the crucial importance of fish as the sole major source of animal protein in Nyasaland, since the country did not have good sources of meat. Platt’s Nutrition Unit experimented with intermediate fishing technologies for producing fishing equipment and processing fish. But the results of these experiments were largely disappointing. Apparently, the new techniques were less effective than what African fishermen were already using. In the 1940s, the industry was further depressed by the shortage of bicycle tyres due to the wartime restriction on the importation of rubber, which in turn meant a significant fall in the number of traders.

3. THE 1950S AND 1960S: INTRODUCTION OF MANUFACTURED GEAR AND NEW PATTERNS OF INVESTMENTS

The second attempt by the colonial government to stimulate the fishing industry was made in the 1950s and it was more successful. The 1952 Nyasaland Report noted that African fishing was not a purely subsistence activity, but rather an undertaking that had a commercial side to it. Most of what fishermen produced was either sold to local consumers on the beach for people’s own consumption or distributed among the members of the fishing crews for their personal needs. Some African fishermen on Lake Malawi who had been using inshore seine nets started increasingly to fish offshore using manufactured gillnets (Government of Nyasaland, 1952:53). It became more and more common for African fishermen to fish on commercial basis and the purchase and use of manufactured nylon nets among this group increased. Seine nets were also increasing in number and becoming longer as fishermen started to clear more and more beaches for their operations. In the 1950s it was still common among African fishermen to use nylon webbing for the bunt only and linye for the outer panels in the beach seines. Bigger nets and seines made it increasingly difficult to operate from dugout canoes and led to the emergence of bigger and more costly planked boats.

A combination of factors stimulated the development of the African fishing industry in the 1950s: the Government introduced a policy of the bulk sell of nylon nets and twine and plank boats to African fishermen; the establishment of the Horris Hickley netting company in 1959 (which became Blantyre Netting Company in 1979) made manufactured nets more and more common; and for the first time, Africans who had the potential were offered loans from the African Loans Board and the Native Development and Welfare Fund for the purchase of improved equipment (Government of Nyasaland, 1953:70-72).

Furthermore, the government introduced a policy that restricted non-African commercial fishermen from operating in Lake Malombe^[54]. The arguments against permitting commercial fishing by non-Africans in Lake Malombe were twofold. First, the government feared that commercial fishing would be biologically detrimental to the fishery in such a small lake as Malombe. But, and this is the second reason, it also feared that a non-African presence would cause conflicts between African and non-African fishermen. The Director for Game Fish & Tsetse Control, M. J. Borley under which the administration of fisheries fell, argued that “where the fishermen were heterogeneous and where the standard of the non-native fishery is so far above that of the native as to put it in a totally different class”, it seemed politically important to avoid even the risk of overfishing. In such a case, it appeared to him not improper to apply restrictions to the non-native effort as soon as effort and yield data indicated that they were very probably necessary, without waiting for incontrovertible proof that they were essential. Delaying imposition of restrictions and pending incontrovertible proof, entailed the risk of damage to the basic economy of the native fisherman through circumstances quite outside his control. Borley argued that if deterioration had resulted from delay in introduction of controls, the disadvantage of impaired race relations and political feeling would have been added to straightforward economic loss.

Finally, the creation of the Federation of Southern and Northern Rhodesia and Nyasaland in 1953 led to the lifting of the wartime ban on fish exports and allowed the large-scale purchase of fish for export mainly to Southern Rhodesia (Government of Nyasaland, 1962:81-82).

What we see emerging first in the southeast arm of Lake Malawi but also in Lake Malombe during the 1950s is a process that clearly differentiates between two groups; new African commercial fishermen on the one hand and, on the other, those who continued to fish mainly for their own consumption and for local barter. McCraken notes that in the 1950s, already, “perhaps the most remarkable feature of African fishing was the emergence of a small group of “big time” capitalist fishermen who employed labour on a regular basis and invested in imported nets and boats” (1987:427). Under this commercial orientation, group ownership of equipment was abandoned and instead nets and boats became private property of owners most of whom left the day-to-day control of the fishing to a son while they supervised the drying and selling of fish on shore. The 1962 Fish Ranger’s report noted that there were 30 active African commercial fishermen on Lake Malombe^[55].

However, this process is explained not only by changes in government policies. According to the same report, of the 30 commercial fishermen, only seven had obtained government loans of 200 to 300 pounds sterling. The rest had raised the money needed from elsewhere. With reference to the situation in the Southeast Arm, Fisheries Officer A.D. Sanson stated in 1956 that: “Since 1950, there have been a few Africans who have attempted to set up fisheries on a real business basis. I think I am right in saying that not one is originally a fisherman himself. They are all African business men, some from other parts of the country and some belong to this part of the country and have spent most of their lives in South Africa or Southern Rhodesia and have come back to set up business. Local fishermen are employed. There are no cooperative fishing concerns; every one of these businesses is being run by one man^[56].” The growth in labour migration through organized recruitment agencies such as the Witswatersrand Native Labour Association (WENELA) to South Africa, but also to other places, were hence one important factor. It strengthened the processes which led to a more commercial fishery and in turn caused more and more of those who returned to invest their savings into the industry.

Finally, fish trading among Africans continued to increase. Partly because of the comparatively low wages and poor conditions that prevailed on settler tea and tobacco estates in the Shire Highlands, people started to take up fish trading in increasing numbers as it was more lucrative. Also it gave them an independence which wage labour denied (Vaughan, 1982). In addition to this, the price paid to peasants for cotton did not compensate them fully for their labour. African traders had started moving their fish by lorry, often teaming up with other traders to hire a lorry for the transport of their individual consignments. By 1962, there were an estimated 400 fish buyers in the Lake Malombe area^[57]. This indicates that the investments required for entering fish trade must have been considerably lower than those required for commercial fishing. But despite the reported existence of 13 traders for every commercial fisherman, the reports of the Commission of Inquiry into the Fishing Industry clearly show that fishermen often experienced sale problems, particularly during the rainy season^[58]. Fishing in Lake Malombe was concentrated in the northern part of the lake and landings on the west bank due to good access to roads and markets in that area. The few fishermen that fished on the east bank transported their catch by boat to Mtundu and Nkupekupe on the Upper Shire and to the north-west part of the lake.

4. THE 1960s AND 1970s: GROWTH IN EFFORT; MORE PEOPLE, MORE GEAR AND TARGETING OF NEW SPECIES

Whereas in the 1950s it was still common to use Linye for the outer net panels of the seine nets, the African commercial fishermen increasingly dispensed with the use of local materials in the 1960s and started to use nylon nets exclusively, thereby improving the efficiency of their operations. Furthermore, the use of static gillnets was abandoned in the late 1950s as these nets frequently were either destroyed by crocodiles and or stolen if left fishing over night. Instead, people started fish driving, locally called chiombera^[59]. This was practised both during day and at night. Fish driving is reported to have been introduced in 1959^[60] and within a few years, became the most common technique in Lake Malombe. Normally a net of as much as 1 000 yards long was set in a circle and a powered boat steamed round inside the circumference, with crew members beating the water to drive the fish into the net. This technique resulted in enormous increases in catches, fuelling further the commercialization of the fishery. In addition, government introduced a policy of shooting crocodiles to lessen the problem of net destruction. According to the 1962 Fish Ranger’s report, all the 30 production units of the commercial fishermen were motorized (the common type of engine they used was the “Seagull” 102 long shaft).

Sensing that they could increase their catches further by using gillnets as active gears like commercial trawlers on Lake Malawi, most gillnet fishermen started to trawl their gillnets using two boats and outboard engines. Used in this way, it was necessary to increase the depth of nets. Thus gillnets used as a tolora (trawler) or kandwindwi, (named after the trawler operated by the MALDECO Fishing Company on Lake Malawi) increased from 24 or 40 to a depth of 80 meshes^[61].

With the commercially valuable Chambo in abundance and increased numbers of migrant returnees, the use of the seine nets became more and more prevalent in the 1960s and 1970s. At their height of operations, around 1980, seine nets had grown in size up to 1 000 metres long. While the 1962 Fish Ranger’s report indicate that there were five Chambo seine nets operating in Lake Malombe, the Director of Fisheries Frame Survey figures show that the number had increased to 27 by 1981 (Zwieten, Njaya and Weyl, 2003)^[62]. The increased use of the seine net also saw the invention of the Chalira, a small seine net with small mesh sizes that was operated behind the bigger seine net in order to catch the fish that usually escaped from the larger seine as it was pulled close to the beach.

The 1960s and 1970s also witnessed the emergence of a seine net targeting new species: the so-called “Kambuzi seine net”. Like the Chambo seine it is operated from the beach, but has smaller meshes and was initially about half the length of the Chambo seine. It was meant to target Kambuzi^[63] like so many of the other gears used in Lake Malombe, the Kambuzi seine net had been developed in the Southeast Arm long before it emerged in Malombe. It is interesting to note that fishing for Kambuzi began well before the stress on Chambo became noticeable. An important reason for this is that in the 1950s, the tea and tobacco estates in the Shire Highlands (Thyolo, Mulanje, Zomba, Chiradzulu and Blantyre) had instituted a tenant labour system (McCraken, 1987) and they required cheap food for their tenants. Kambuzi was seen as one of the ideal food items. During the 1960s, Kambuzi also came into increasing demand among the low-income groups in the growing urban areas and there also proved to exist an export market to Southern Rhodesia. In addition to the stress on Chambo, caused by increased fishing effort, it seems that the increase in demand for fish and the growing consumer acceptability of smaller species also fuelled the expansion of the Kambuzi fishery. At the beginning of the 1980s, the demand for Kambuzi must have been high since the Department of Fisheries figures of estimated catch show that there existed between 60 and 70 Kambuzi seines in Malombe in 1981 (Zwieten, Njaya and Weyl, 2003).

The 1960s and 1970s continued to be good years for labour migration to South Africa, and to some extent also to Rhodesia, resulting in an increasing number of successful returnees who were willing to invest in fisheries. Except for the returnees, little is known about the relationship between work migration and economic activities at home. However, more recent information -including life stories collected for the present study - indicates that specific networks between migrants and some of their relatives remaining at home, developed in this period. These networks served to transmit and invest remittances from the migrants before the migrants decided to return to their home area for good, thereby rendering work migration even more effective in economic terms. Fisheries Department Frame Survey figures show that by the early 1980s, the number of owners of fishing units had reached around 230 (Zwieten, Njaya and Weyl, 2003).

The speed of the technological changes during these two decades also indicates that another important social process must have taken place simultaneously. Towards the end of the 1970s, it had become exceedingly difficult to operate at a subsistence or small-scale basis. In the Chambo fisheries, various types of active catch methods such as chiombera and tolora, as well as Chambo and Kambuzi seining, requiring heavy investments in the form of costly nets, outboard engines and planked boats had completely taken over from the passive methods practiced earlier. With this noticeable increase in investment-driven growth of effort we must assume that the catch rates for the older passive gears must have diminished, rendering the low investment fisheries very difficult to sustain, to say the least. It is therefore reasonable to assume that the great majority of the 230 boat owners in 1981, constituted the type of commercial fishermen that A.D. Sanson described already in 1956 (see citation above).

In his study of the development of access rights to fishing in the southeast arm of Lake Malawi W. C. Chirwa concludes, “while no single group won the contest for access rights and control, the traditional leaders were the greatest losers as they failed to use their political position to block the success of other groups” (1995:377). In the context of Malombe, where the African commercial fishermen did not have to face competition from non-African fishermen, they were much more successful in asserting full access rights than were the African commercial fishermen on Lake Malawi. But, they were successful at the expense of those who were unable to follow suit in the race for new and more effective fishing gear. The new and extended fishing units required an extensive use of labour, but the people recruited to work as crews were often different from those who had just lost their access rights due to lack of investment capability. By the end of the 1970s the fishery in Malombe had to a large extent lost its role as a “commons” and since then the “Lords of Malombe” succeeded in establishing some sort of control over the fish resources in the lake.

However, the control of access exercised by the commercial boat owners has many aspects. The increased need for labour could not be fulfilled by recruiting labour from within the family as had been the convention until then. In the beginning it seems as if crew-members from outside the families were compensated through some sort of direct payment. McCracken (1987) shows that already in 1956 owners had considerable problems related to the control of their crews. Some owners then blamed crew members publicly for theft of catches and misuse of equipment. Towards the end of the 1970s, the pay structure changed to a share system based on 50 percent of the catch allocated for the investments and the other part for the labour. With minor modifications and variations between the different gear types, this system remains in use to the present.^[64] The new system resulted in increased earnings for the crew members. Since their earnings depended on how much they caught, this gave an incentive for crew members to work harder and fish longer hours. Day and night fishing using gillnets became common. To what extent this shift altered the power relations between gear owners and crewmembers or solved the reported control problems of the gear owners will be discussed below.

The commercial gear owners also invested in the marketing of fish. It has been reported from the late 1950s that many gear owners had bought or intended to buy their own trucks and pickups^[65]. Using these, they could move their own catch and also buy from their fellow fishermen, cutting out the middlemen. As fish sold retail at more than 100 percent mark-up, such forward integration further increased the earnings of commercial gear owners (Hara 1993). The growing urbanization in Zomba, Blantyre and Lilongwe in the 1970s and 1980s also acted to increase demand for both Chambo and Kambuzi. But contrary to what happened in the fisheries where the investments of some excluded the participation of others, investments in the trade system never excluded the participation of small-scale traders operating from bicycles and various types of public or private transport. Given that by using bicycles and pubic transport fish could be sold in a much dispersed and often inaccessible hinterland of small villages and regional towns but also in big urban centres, the trade system continued to remain decentralized and open to a wide range of people. In the early 1990s, an estimated 3 000 fish traders operated on the southeast arm of Lake Malawi, the Upper Shire River and Lake Malombe. Of these, approximately 50 percent concentrated their activities in Lake Malombe (FAO, 1993).

5. THE 1980s

5.1 Reduction of Chambo catches and technological changes

Fishing effort on Lake Malombe had increased dramatically during the 1970s: in 1981, on a water surface of 390 km², 230 gear owners hired 1 400 crew members fishing from 350-360 planked boats and operating more than 100 km of gillnets, 27 Chambo seines and between 60-70 Kambuzi seines (Zwieten, Njaya and Weyl, 2003: Figures 11 and 12). Most of the boats were motorized. Total catches in the early 1980s fluctuated between 7 000 and 12 000 tonnes of which 6 000 to 9 000 tonnes were Chambo (Zwieten, Njaya and Weyl, 2003: Figure 1).

The 1980s saw both a decline in nominal catches of Chambo as well as in its relative importance of the total catches. By the early 1990s, Chambo catches had been reduced to around 1 000 tonnes while the total catches remain close to what they were a decade earlier. During this period we also observe great changes in technology and catch methods. By 1990 the amount of gillnets had been reduced to 45 km and only six or seven Chambo seines are reported to remain in the fishery. On the other hand the number of Kambuzi seines tended to grow in the first part and later started to fall so that the number of seines in the early 1990s remained somewhat lower compared to what it had been early in the 1980s. Also the use of outboard engines became more and more rare and by 1994 there were no engines left in the Malombe fishery (Fisheries Department, 1995).

The big change in the fishery for Kambuzi came when the Nkacha net was put into common use in the first part of the 1980s. Already in 1976, a Lake Malombe fisherman, Mr Galimbe Paudala, invented this new gear and fishing method which targeted Kambuzi in the open waters of the lake. But most of the fishermen only seem to have adopted the method and start using the Nkacha after the Chambo catches began to fall in the 1980s^[66]. Until 1983 the number of Nkacha nets remained well below 50, but thereafter the number increased rapidly and in 1991 there were 180 such nets in Lake Malombe. Since the Nkacha was an open water seine, this meant that the fishers could fish all over the lake. Quickly, the Nkacha became the most important fishing gear as almost all gear owners switched from Chambo and Kambuzi seines to Nkacha nets. The number of gear owners and assistants did not grow much through the 1980s. Some additional gear owners seem to have been able to establish themselves, and in the early 1990s they are reported to number around 300, but the number of crew members remains stable (Zwieten, Njaya and Weyl, 2003: Figure 11).

The development of fishing effort in the 1980s represents something new in Lake Malombe. While effort from the early 1950s to the late 1970s had been growing as a result of increases in number of fishermen and in investments, the 1980s experience a noticeable reduction in the population-driven growth of effort. Besides, the new investments which continue to flow into the fisheries are no longer just added to already existing investments. It could have been expected that the Nkacha net would come as an addition to already established practices but instead they are used to substitute the previous investments in order to target new species in new environments. This change is a simple reflection of the disappearance of the Chambo, it does not make much sense to continue to chase a species which has become impossible to capture and this is the reason for giving up the Chambo gear^[67]. Whether the change from gillnets and Chambo or Kambuzi seines to Nkacha nets represents an investment-driven increase of fishing effort is hard to say since it is difficult to compare one fishing method to another^[68]. Nevertheless, there can be little doubt that although the investments and the technological development very much continued, the fishing effort tended to stabilize during the 1980s when it comes to investments.

5.2 Reduced recruitment of fishermen: the role of ‘lords’ and crew members

We must also ask to what extent the reduced recruitment of gear owners and crew members that took place in the 1980s reflected changes in the biology of the lake (reduced catches of Chambo as well as reductions in overall catch rates), and/or to what extent they were directly caused by social changes that took place, in particular the emergence of the new elite. The question is whether the stabilization in the number of operators was caused by economic incentives to leave the fisheries when signs of heavy fishing pressure emerges or whether it was the economic influence in general of the “lords” and their dominance in the fishery in particular that made them able to control - and even deny - individual’s access to the resources of the lake so effectively, quite irrespective of whether the individuals wanted to participate as co-owners or as crew members. To answer these questions we need to investigate thoroughly who came to constitute this elite, how it was reproduced and what opportunities and constraints the elite faced in its daily economic operations.

To become a gear owner, especially a seine or a Nkacha net owner, means to have made it to the top of the social profile in Malombe. Only a couple of hundred individuals, out of a total population of more than 50 000, find the means to invest in a fishing unit. In the 1950s we saw how the Fisheries Officer in Mangochi linked the emergence of the first African commercial fishermen on Lake Malawi to businessmen from other parts of the country and labour migration returnees who invested their savings in fishing material. On Lake Malombe the emergence of commercial fishermen took place some years later and shows a slightly different picture. The authors interviewed a small number older people who had established themselves as gear owners in Malombe in the 1960s. These interviews indicate that very few “foreigners” seem to have invested in the Lake Malombe fisheries. Primarily labour migrants originally from Mangochi district started to invest in Chambo seine nets, in boats and in outboard engines, even if some local businessmen and retired civil servants also emerged among them.

It is important to notice that those constituting the elite today, only very rarely are the same individuals - or representatives of the families - that formed or constituted this elite 30 or even 20 years ago. Among the 42 gear owners interviewed, only three or four of them had been owners in 1980. Twenty four of them report that neither their father nor any of their uncles had owned fishing gear. However, this does not mean that the remainder - 18 of the interviewed gear owners - are sons or nephews of commercial fishermen, nor does it mean that they inherited their fishing gear. Only three or four cases are known were a presently active boat owner is the heir of a commercial fisherman. What this tells us is that the elites are not a fixed and stable group, but rather an assembly of those individuals who at any time have been successful enough to achieve a certain financial status adequate enough for investment in fishing, which many of them may lose again or, if they do not, may be dissipated at the time of the death of the person. All the life-stories indicate a pattern which is common in rural Africa; economically wealthy people are almost without exception “self-made”. One of the main reason for this is that inheritance rules often are ambiguous and fluid and a source for serious conflicts (Berry, 1993). Matrilineal descent, like we find in Yao society, often complicates the principles of inheritance further. Common practice^[69] is often that a person inherits the estate of his/her uncle, that is the person’s mother’s brother. Even then, it is usual that the equipment will be sold and money shared among the deceased person’s family (sisters, father and mother) other than being left to a nephew to continue with fishing. In this way, the family avoids in-fighting for the estate. Apparently, even in cases where the deceased gear owner had expressed the wish that his children should take over his estate, it is common for his father or uncle to over-ride his wishes and take over the estate, unless the deceased had backed his wishes with a written will. Only 13 percent of the gear owners registered in 1993 and who had deceased before 1999 had left their equipment to their sons. In such cases, the sons were old and strong enough to assert their claims to their father’s estate or their father left the gear in their hands well before they died^[70]. The problems connected to the inheritance of gear have also led to some unintended consequences regarding ownership. In order to avoid the dissipation of assets in the inheritance process, ten women, either wives or sisters of the deceased, have managed to postpone the inheritance process and to establish some kind of control over the fishing gear. It is unclear, however, how and when the first female gear owners emerged. As will be shown below the female owners are under severe constraints and it is therefore improbable that they will constitute an important economic part within the elite.

Since inheritance is not an option, those who want to become boat owners have to find other means and these means prove to be fairly similar to what the first commercial fishermen used^[71]. In most cases it means finding a source for accumulation of capital outside the fisheries. In the 1980s the most common options remained migrant labour savings, credits or loans and fish trading. The contract system to the South African mines continued but ended towards 1985 and also the migration directed towards the ‘informal’ sector - in South Africa as well as in Zimbabwe - started to show signs of stress. It was still possible to find interesting sources of income even if it was much less attractive than it had been in the past decades. The reduced opportunities, which have continued into the 1990s, can be observed in general social changes in the Upper Shire area. More than a reduction in emigration it has resulted in an increase of migration failures. Avery high percentage does no longer achieve any measurable success in their quest to earn and save money for re-investment back home.

Fish trading seem also to have been a good alternative to labour migration in the 1980s. Sources of credit or loans could also be an option. In the 1970s and 1980s certain shop owners and the Blantyre Netting Company were sometimes willing to offer nets on loan to potential gear owners. Accumulation of savings as a crew member seems to be very difficult and only four of the 42 owners interviewed refer to this type of savings as their source of starting capital. Earnings are limited and receiving them in small daily instalments under the pressure of social and economic needs of an extended family system does not encourage savings which would require decades.

If it is difficult to become a gear owner, it is not necessarily easier to remain one once you have established a fishing unit. The management of the unit and its labour in particular has been found to constitute the main critical factors. Managing crews in commercial fishing units in Malawi has never been an easy task. The first entrepreneurs loudly complained to the Commission of Enquiry in Fort Johnston in 1956^[72] about how they in various ways were fooled by their crews who sometimes destroyed equipment or stole fish they had caught. This is in no way an uncommon feature in African fisheries. On the contrary, the direct control of labour is often reported to be critical^[73] particularly since norms and rules that regulate owner/crew relations are seldom shared. On Lake Malombe we found that, despite the seeming economic power of the ‘lords’, their ability to control the crews are often restricted. Some crews have over the years ascended their position to one that resembles that of shareholders. Sometimes they decide on the timing of work, the recruitment of new crew members or the sale and price of fish without consulting the owner. Despite the introduction of a share system in the 1970s some crews have later stopped contributing towards payment of operational and fixed costs for the unit, meaning that these are now solely borne by the gear owner.

The reasons behind the paradox of a numerically limited and economically strong elite with restricted power over their own employees are complicated and can only partially be dealt with in this work. The owners’ problem is that few sanctions exist that do not hurt them as much as it hurts crew members. Owners can withdraw their nets and boats from operations in order to force changes or they can dismiss their crew and recruit new ones. In both cases the owner risks serious losses as a result of thefts, malpractices or inexperienced fishermen. Besides, the relationship between owners and crew members is not a single-function one, but complex and strongly embedded in the wider socio-cultural framework of the Upper Shire region. Sometimes one finds that crews are able to mobilize general support among the population preventing the owner’s possibilities to employ new crews, or they organize and manipulate boycotts of specific owners who they claim are engaging in practices thought to be inconsistent with agreed norms and practices.

The owner/crew relationship proves even more complicated and in some cases the influence and power of the crews are found to serve the interests of their masters. One distinct feature of the 1993 survey of owners is the absence of non-local investors; that is of investors who do not claim to be from the Malombe area. Only a handful was identified, despite the fact that owners, crews and even individuals not participating in the fisheries, all claim that access to the lake is open to everyone. Thorough investigations showed that access regulation to a large extent was exercised by crews who put extra restrictions on the operations of ‘foreigners’. It was reported that the outside boat owners seldom were allowed to inspect the catches of their boats at landing and that all maintenance and repairs were under the control of the crews. This kind of owner could chose between appointing a local person to look after his interests or simply wait and receive his share from the crew after they had taken theirs. As a result, very few foreign gear owners are able to run a profitable fishing business. One may say that the barriers for “foreigners” entering fishery in Malombe mainly centre around their lack of inside knowledge about local rules and a certain general reluctance among local people to accept “foreigners” as boat or gear owners. This gives fishing crews possibilities to increase their influence. The same kind of cultural barriers were reported to apply towards the few women boat owners. They must contend with negative social and cultural perceptions to their participation in the fishery in what is a largely Moslem community.

This brief presentation of owner/crew relations demonstrates the complexity of factors influencing the effort dynamics on Lake Malombe. The reduction of Chambo catches is arguably part of the explanation for why the growth in number of owners stabilized during the 1980s. However, the reduced possibilities to accumulate wealth through labour migration were also important. The incapacity among the elite to convert economic power into direct social control indicates that their economic power were less important than what could be expected in explaining the reduced recruitment of owners. On the other hand, an increased power among the crews - mainly used to struggle against the owners - seem to have effectively prevented recruitment of new groups of owners, thereby indirectly supporting the interests of the same owners. As for the stabilization of crew members this is also affected by reduced catches and technological changes from Chambo and Kambuzi seines to Nkacha nets. However, the increased power of crew members is also a factor which strongly influences crew recruitment. As shown, the owner’s possibility to successfully manage his fishing unit is closely related to his control of the crew. But control of the crew is obviously related to the number of individuals to control. The owner has therefore a strong incentive in keeping the number of his employees as low as possible.

6. THE 1990s: STOP IN INVESTMENTS AND REDUCTION OF FISHING EFFORT

The 1980s saw dramatic changes in catch composition of the fishery on Lake Malombe, although the total biomass remained fairly stable in the lake. This picture changed again in the 1990s when total catches decreased from approximately 10 000 tonnes per year in the beginning of the decade to 3 000 tonnes per year in the period after 1995. The large share of the catch is Kambuzi, while Chambo has almost disappeared and now only constitute 200-300 tonnes per year. Parallel to this reduction we also see a clear reduction in fishing material such as boats and Nkacha nets. Chambo and Kambuzi seine nets have almost disappeared. Only the number of gillnets shows a modest growth in the 1990s (Zwieten, Njaya and Weyl, 2003: Figures 1, 11 and 12). The development in terms of gear owners and crew members is more unclear. The frame surveys of the Fisheries Department show an increase of owners in the 1990s. This increase was faster than what was observed during the 1980s; from approximately 300 owners in 1990-1991 to more than 400 owners in 1999 (Zwieten, Njaya and Weyl, 2003: Figure 11). These figures stands in clear contrast to the results of an update in 1999 of the 1993 gear owner register. This update shows that of the 332 gear owners that were recorded in 1993, 210 (more than 60 percent) had left the fishery by 1999. Meanwhile 73 new gear owners had joined the fishery, making it 195 gear owners operating on Lake Malombe in 1999 (see Table 1)^[74].

TABLE 1. Summary of the June 1999 updated information on the Lake Malombe 1993 gear owners’ register

It is difficult to explain how the big discrepancy between these two sources of data has emerged. Part of the explanation is probably that the 54 units, which according to table 1 have migrated and that for the great part now operate on the southeast arm of Lake Malawi, still are accounted for in the frame surveys. This accounts for about half of the discrepancy. Nevertheless, the accuracy of the update and a series of parallel indications supporting its results make us conclude that the number of gear owners operating on Lake Malombe has been considerably reduced since 1993 and that the number even may be lower than what it was at the start of the 1980s.

The frame surveys referred to by Zwieten, Njaya and Weyl (2003: Figure 11) state that the number of crew members has remained stable. The 1993 registry does not provide data on crew members, but given that the number of owners is believed to have fallen, one could assume that the number of crew members also has decreased. However, this is not necessarily the case. As the number of seine nets and Nkacha nets has declined, the owners have tried to compensate and to maximize effective fishing time by hiring an increased number of crews which are fishing consecutively within the same unit. In the late 1990s most Nkacha units had two and some even had three crews each and the number of crew members per owner is therefore probably higher in 1999 than what it was at the beginning of the decade.

There can be no doubt that the dramatic decline in catches is the main reason behind the decline in investments and fishing effort. While gear owners bought second and third units in the 1980s, it has become increasingly difficult to own and manage more than one net. Very few people in the 1993 registry have increased the gear they own. In most cases, those who had two or more Nkacha nets have either sold one or combined the two to make one. Most owners of Chambo seine nets who have either migrated to Lake Malawi or retired their nets. A good number of the Kambuzi seine nets were converted into Nkacha nets in the early 1990s or the units migrated to Lake Malawi. The few seine nets remaining on the lake are based at the outlet of the lake just outside the Liwonde National Park where they seem to benefit from the protection of the Chambo provided by the park.

However, the decline in catches does not explain the particularities of the reduced fishing effort in terms of the relative importance of changes in number of owners and crew members compared to the changes in volumes and quality of the fishing gear. In order to understand these particularities the constraints found to influence decisions and operations of the owners are important. Many of these constraints continued to prevail and were aggravated in the 1990s. All the sources for accumulation of external funds have become more difficult to access. Legal labour migration to destinations outside Malawi is no longer an option, even though various types of “informal” and very risky travels to the neighbouring countries are reported to continue to help potential newcomers. The update of the 1993 register shows that most of the 73 new entrants report that they obtained their capital from migrant labour savings in South Africa or Zimbabwe. Fish trade has lost much of its potential due to the fall in catches. It takes more and more time to gather a large enough consignment to take to the market so that traders make fewer trips. Since Kambuzi has less value than Chambo, trade in Kambuzi is unlikely to generate surplus capital. Whereas shop owners and the Blantyre Netting Company were willing to offer nets on loan to gear owners in the 1970s and 1980s, they have become more and more reluctant to provide such facilities in the 1990s. The ability of loanees to repay has declined due to declining profits and the sources for credit or loans have virtually disappeared and where they exist the interest rates are high and security difficult to provide.^[75] As demonstrated in Table 1 the result is that new entrants more and more rely on buying used equipment.

The control of the “lords” over their crews has continued to weaken, and the owners have had to pay a stiff price for the introduction of more than one crew into their units. Due to the sharing of the material a crew no longer accepts any particular responsibility for the care of the equipment. Abuse and the careless use of equipment are reported to have become common in many cases. With the decline in catches the mobility of crew members between fishing units has increased and the feeling that crew membership is ganyu^[76] has become increasingly entrenched. The lack of long term tenureship in a unit adds to the problem of responsibility and appropriate use of equipment.

The rules and norms regarding owner/crew relations have certainly not become clearer or more consistent over the last years and the management problems have increased, partly as a result of the increased number of employees. Until the 1990s most gear owners processed their catch before selling it to traders, thereby doubling their earnings. The crew members’ share was based on the price of the wet catch. They only got their share after the gear owner had sold the processed fish for the week in question. Crew members allege that most gear owners sold the catch to themselves at a very low price, thereby reducing their share. Some also did not give them their share as agreed at the end of each week or after the sell of a particular consignment. Due to such malpractices by gear owners, the crew members started pushing for changes to the system. If the gear owner wanted to process the catch, he had to compete with other potential buyers. Secondly, crew members decided that they would get their share immediately after the sell had been conducted. Thus gear owners who wanted to practice forward integration had to compete with other buyers (fish traders) for “their own” fish. Increasingly, traders bought the fish direct from the crews and processed it on their own other than buying processed fish from gear owners. For many gear owners, this meant an end to forward integration. Those who had bought pick-ups to market the processed fish in the cities sold their vehicles or deployed them for other businesses such as hiring them out to fish traders or for the transportation of people (matola). This brought about a fundamental change in power relations between gear owners and crew members. Whereas before the gear owners would take over the catch from the crew as soon as it was landed, nowadays he stands by as traders bid against each other for the fish until the highest bidder is found. The going price has to be agreed between him and the crew. The insistence by crew members that they get their share immediately after the sales have been conducted has also added to his short-term cash flow problems especially after catches declined in the 1990s.

For the owner the system presents a catch-22 situation; it can be difficult to save enough money to carry out major net repairs needed. At the same time, a net in poor state is not as efficient as one that has been properly maintained. The crews usually do not agree to forego their share of the money for a few days so that the gear owner can use accumulated funds for such expenses. The practice of many crews has become to abandon the net and to seek another owner if the net is not productive anymore. Thus if the gear owner cannot save enough money from his share of the catch or cannot borrow money for repairs elsewhere, the result will be that the net will continue to decline in quality and eventually, the crew members will abandon it if the owner does not suspend the operations before that happens. Inability to keep up with repairs leading to decline in productivity has been one of the major reasons for suspending fishing or complete wear out of nets, leading many owners to divest from the fishery. An added factor is that whereas the net owners in the 1970s and 1980s would subtract the operational costs such as money for fuel, twine for net repairs and breakfast for the crewmembers, the latter demanded that the sharing should be based on gross sells other than nets sells. Thus, the gear owner has been left with the sole responsibility for all the costs (capital and operational). As if this is not enough, the share proportions in the Nkacha have been changed to 45 percent for the gear owner and 55 percent for the crew members, since the diver (mtiwi) gets 10 percent of the gross sells before the sharing is done. The gear owner must also contend with increasing theft of fish or gear components by crew members as most Nkacha units started to deploy more than one crew to a unit.

One of the few possible responses from owners has been to revert to less capital-intensive gears than the Nkacha nets. Less expensive gear usually can be operated by smaller and fewer crews and the gear owner thereby is better positioned to exercise effective control over his employees. The modest increase in numbers of gillnets, longlines, handlines and traps in the fishery in the late 1990s (Fisheries Department, 1999) may have been caused by such considerations, but the problem remains that the catch rates for these types of gear remain very low.

As already mentioned, many of the management and control problems have been amplified by the lack of commonly shared rules, values and norms among owners and crew members. But it is probably also correct to say that, during the 1990s, the crew members have increased their control in terms of operational decisions and pricing of the catch and that many gear owners have lost much of whatever control they may have had earlier regarding these decisions. Those who in terms of economic wealth seemed to emerge as the lords and the elite of Malombe have at present far less influence regarding their own production units than what could be expected. This is an important factor explaining why so many owners, over the last years, have decided to leave the fishery.

However, in order to leave the sector people must also have alternatives. This seems never to have been a problem on Lake Malombe. The Yao are historically traders and business people (Rangeley, 1970) and this seems to define the attitude of the gear owners towards fishing. Fishing is not an occupation they are culturally or emotionally tied to. A fishing unit is considered an investment that must pay back in terms of profit; otherwise they will seek to sell it and invest their money in something considered more promising. Table 1 indicates that out of the 210 owners who left fisheries after 1993, 126 are still economically operational in the area. Some have moved into other business activities such as tobacco farming^[77], or run minibuses or pick-ups for transporting people. Others again have built houses in Mangochi Township or in some larger villages where they consider the demand for rooms to rent to be high.

Such switches in economic activities are not necessarily very difficult or risky. Contrary to what has often been assumed the commercial gear owners have never exclusively relied on the fisheries in Lake Malombe for their incomes. On the contrary the life stories of gear owners demonstrate that all of them are involved in a range of different economic activities and that diversification always has been considered the best strategy. In addition to commercial agriculture, people transport and renting houses it is common for gear owners to run shops, maize mills, beer halls, local video cinemas and/or rest houses and to invest in cattle and in dimba gardens (Mandala, 1990). The various businesses are supposed to support one another and secure stable access to cash. Forward integration of the fishing business into fish trading used to be popular strategy among the gear owners, but fish trading has declined as an option. Most of them say it is hardly profitable anymore as the fish has become too expensive, especially after the crew members started to intervene in the pricing of the fish.

The shift in focus from fishing to other economic activities is also to some extent a self perpetuating process. The more the attention of a boat owner is directed towards other businesses, the less time he has to follow up and control his fishing crews. Many gear owners or ex- gear owners emphasized this dilemma and insisted that there exist a limit for how much time that can be spent in other activities before the fishing business starts suffering. The analysis of owner/crew relations demonstrates that many boat owners crossed this limit in the 1990s. It seems as the wealthiest among them decided to give up the fishery while those who can not afford larger investments in other sectors continue to operate as gear owners.

7. CONCLUSION

What makes effort development on Lake Malombe particularly interesting is what took place after 1950. Before this time, fishing effort was mainly characterized by population-driven changes, and the development seems to have been fairly similar to what has been the case on many other southern African lakes. However, from that time the investment-driven changes grew progressively in importance and in the 1980s it came to dominate the effort development completely. Zwieten, Njaya and Weyl (2003) show the serious effects this growth has had on the environment and the biology of the lake. Although not a proof, the case of Lake Malombe is an indication that investment-driven growth of effort, under such conditions, is more of an environmental and biological challenge than population-driven growth of the type observed in so many of the other water bodies of the region.

The case also demonstrates how the investment-driven growth as such caused the gradual reduction in the population-driven growth on the lake. The increased efficiency of the Chambo seine nets reduced the catch rates of the gillnets and other more conventional gear to a level where the operations of the latter became almost impossible. Thereby, a new group of “lords” managed to appropriate the access to the lake at the expense of other and less wealthy segments of the population. This effect is well-documented in many parts of the world when investment-driven changes of effort start to dominate in the development of a fishery (Brox, 1990).

However, as interesting as the effects of investment driven changes are, the findings connected to factors constraining investment driven changes to occur. First, the analysis shows how intimately related the investments in the Malombe fisheries have been related to economic opportunities external to the fisheries. Not only was it labour migration to countries abroad that facilitated accumulation of money and thereby initiated the investment processes. This is already well documented in the literature (McCraken, 1987), but our analysis also shows how important the flow of externally generated wealth has been to keep the investment level going. At least on Malombe it seems as if neither the seine net fisheries for Chambo or Kambuzi, nor the Nkacha net fishery, ever became able to sustain the actual level of investments observed in the 1970s and 1980s. More money was probably floating into the fishery than out of it, and this explains the high turnover rate in terms of gear owners. The reasons are not to be found in pure economic considerations connected to the volumes and the prices of catches. On the contrary, evidence exists showing how the potential profit margins were good for a long period of time. The main problems are more to be found in the institutional arrangements which are found to dominate in the Malombe fishery in particular and in the Upper Shire society in general. Unclear rules and norms regulating the life of extended families and social security networks makes it very difficult for people to make the savings required to keep a relatively capital-intensive operation going. Even more important is that the same lack of clarity is found to dominate in the regulation of owner/crew relations. Symptomatically, there is strong distrust between owners and crews and this leads to an extended need to control each other. Beside the reduced catches in the 1990s, the problems connected to the owner/crew relations are probably the most important factor explaining why so many owners have left the fishery in recent years. The Malombe society is in no manner unique regarding these sociological characteristics and similar phenomena are increasingly being reported from many rural communities in sub-Saharan Africa. This is why this case study is particularly instructive when it comes to understand why investment-driven growth of effort so seldom seems to take place in the fisheries of southern Africa.

Finally, the Malombe fisheries also show how inter-connected African artisanal fishery is to all other kinds of economic activities, both rural and urban. It adds to demonstrating the weak empirical foundation behind some popular and widespread representations about African fishermen being marginalized people and African small-scale fisheries being a “last resort”. In Malombe we have shown it to be the opposite: the most influential fishermen are found in the richest segments of the community and their decisions always reflect their assessment of the earning potential of the fishery on the one hand against other economic activities on the other.

8. REFERENCES

Bell, R. & Donda, S. J. (1993), Community Participation Consultancy Report, vol. I, Fisheries Department, Lilongwe.

Berry, S. (1993), No condition is permanent: The social dynamics of agrarian change in sub-Saharan Africa, University of Wisconsin Press, Madison.

Brox, O. (1990), The common property theory: epistemological status and analytical utility, Human Organization, vol. 49 (3): 227-235.

Chirwa, W. C. (1995), Fishing rights, ecology and conservation along Southern Lake Malawi, 1920-1964, African Affairs, vol. 95: 351-377.

FAO (1993), Fisheries management in the South-East arm of Lake Malawi, the Upper Shire River and Lake Malombe, with particular reference to the fisheries on Chambo (Oreochromis spp.), CIFA Technical Paper 21, FAO, Rome.

Fisheries Department (1995), Frame survey results for Lake Malombe for the year 1994, Bulletin No. 18, Fisheries Department, Lilongwe.

Fisheries Department (1999), Frame survey results for Lake Malombe for the year 1998, Fisheries Department. Lilongwe.

Government of Malawi/UNDP (1996), Management for development programme: Mangochi Socio-economic Profile, Lilongwe.

Government of Nyasaland (1952), Colonial reports 1951, Her Majesty’s Stationery Office, London.

Government of Nyasaland (1953), Colonial reports 1952, Her Majesty’s Stationery Office, London.

Government of Nyasaland (1962), Colonial reports 1959, Her Majesty’s Stationery Office, London.

Hara M. M. (1993), Fish marketing and consumption in Malawi, in Marketing and Consumption of fish in Eastern and Southern Africa (selected country studies), Reynolds, E. (ed.), FAO Fisheries Technical Paper: 332. FAO, Rome.

Hara, M. M. (2000), Could co-management provide a solution to the problems of artisanal fisheries management on the Southeast Arm of Lake Malawi? (PhD thesis). University of the Western Cape, Cape Town.

Liwonde, ADD (1998), Census Report on Farm Families in the Nansenga and Maiwa EPAs for the purposes of distribution of the Starter Packs.

Malasha, I. (2002), Fisheries co-management: a comparative analysis of the Zambian and Zimbabwean inshore fisheries of Lake Kariba, D.Phil. thesis, Centre for Applied Social Sciences, University of Zimbabwe.

Malasha, I. (2003), The emergence of colonial and post-colonial fisheries regulations: the case of Zambia and Zimbabwe, in Management, co-management or no-management? Major dilemmas in southern African freshwater fisheries, Jul-Larsen, E., Kolding, J., Overå, R., Raakjær Nielsen, J. & Zwieten, P.A.M. van (eds.), FAO Fisheries Technical Paper 426/2, FAO, Rome.

Mandala, E. C. (1990), Work and control in a peasant economy: A history of the Lower Tchiri Valley in Malawi, 1859-1960, University of Wisconsin Press, Madison.

McCraken, J. (1987), Fishing and the Colonial Economy: The Case of Malawi, Journal of African History, 28: 413-429.

Peters, P. E. (1999), Agricultural commercialization, rural economy and household livelihoods, 1990-1997. Report provided under the ASAP Support Project, US Agency for International Development, (mimeo).

Rangeley, W. H. (1970), The a Yao, The Nyasaland Journal, vol. 16 (1): 7-27.

Vaughan, M. (1982), Food production and the family labour in Southern Malawi: the Shire Highlands and Upper Shire Valley in the early colonial period, Journal of African History, 23 (3): 351-364.

Zwieten P. A. M., Njaya F. & Weyl O. L. F. (2003), Effort development and the decline of the fisheries of Lake Malombe: does environmental variability matter? in Management, co-management or no-management? Major dilemmas in southern African freshwater fisheries, Jul-Larsen, E., Kolding, J., Overå, R., Raakjær Nielsen, J. & Zwieten, P. A. M. (eds.), FAO Fisheries Technical Paper 426/2, FAO, Rome.