Analysis of effort dynamics in the Zambian inshore fisheries of Lake Kariba - Eyolf Jul-Larsen

1. INTRODUCTION

Fishing effort remains one of the crucial concepts in fisheries management and research. According to conventional management thinking, fishing effort is considered the main factor influencing stock dynamics and stock productivity, and a lot of research is invested to investigate this relation. It therefore appears as somewhat of a paradox that - at least in the case of African fisheries - very little research seems concerned with the question of what causes changes in effort. But, this lacuna in research often does not prevent experts, policy makers as well as some researchers often from having strong convictions regarding what factors influence fishing effort in what direction. They are strongly influenced by common property theory where the maximization of individual profit at the expense of a collective resource, is assumed to drive effort to increase, but also by the famous G. Hardin who was inspired by the old Malthusian concern related to the danger of demographic growth. In the view of many fisheries managers, maximization of profit and population growth tend to make the continuous growth of fishing effort a “natural” law. But fishing effort is an aggregated manifestation of millions of individual and collective micro-decisions taken by the producers every day. As such they are influenced by the totality of possibilities and constraints that the fishermen and their families are facing of which general population growth and levels of income only represent a fraction.

However, some influential social scientists have invested considerable effort in challenging this view. By questioning the assumption of free access to fish, an assumption on which common property theory rely, they argue that a series of mechanisms at the local level often limit people’s access to the resource. In some cases it is argued that such mechanisms -often labeled community based management systems - are strong enough to prevent increase in effort beyond sustainable levels and that they may be considered to constitute types of local governance that distributes benefits amongst the local population. Similar lines of thinking are also the basic elements for community-based or co-management strategies in fisheries. These management strategies have gained a lot of support the last decade. It is characteristic though, that research which challenges the common property view of fisheries often has been more concerned by ‘proving’ that common property theory is wrong, by showing that free access does not exist, rather than by analysing effort development in a historical perspective through a reivew of fisheries statistics and other available data. With regard to southern African freshwaters, despite the considerable unreliability connected to catch and effort data, it is difficult to avoid concluding that, overall, fishing effort has increased considerably over the last 50 years (Jul-Larsen et al., 2003).

Ottar Brox (1990) proposes an alternative perspective to the study of fishing effort which although not focused on African fisheries, is of relevance for them. One of his objectives is to show that there are different types of changes in fishing effort and that the effects upon the resources, as well as upon social development, may be very different from that predicted by common property theory. By introducing an analytical distinction between population-driven and investment driven changes in effort,^[95] he argues that, until the last World War, fishing effort in the cod fishery in northern Norway was mainly characterized by fluctuations/growth in number of fishermen (population-driven) and that the relatively simple technology in use at that time made such fluctuations unproblematic and that indeed the fishery functioned as a commons but without causing a tragedy. The crisis in the cod-fishery first emerged after 1945 when introduction of increasingly capital-intensive technology (investment-driven change) became the characteristic variable in the effort development. In this way Brox wishes to emphasize that commons in fisheries does not automatically lead to tragedies and that they may serve useful functions, for the fishermen, their dependents and for the society at large.

Inspired by Brox’s analysis, and his analytical distinction between population- and investment-driven change in effort, this study investigates in some detail the type, the causes and the effects of effort development in the Zambian inshore fishery on Lake Kariba. The author draws mainly on data sources generated by earlier social science research, on government produced fisheries statistics, on data collected by the author during two shorter visits in 1995 and 1997, on data collected by the author in May-June 1998 and on a survey of 426 fishermen undertaken in September-October the same year by an assistant.

By examining available catch and effort data, the study first seeks to establish what type of effort development has taken place. Lake Kariba is found to be a typical case of population-driven changes. They seem to have dominated effort development since the fishery started in the early 1960s until the end of the 1990s. Despite a strong and stable demographic growth,^[96] the development is, over the years, characterized by substantial increases and decreases in number of fishermen. These fluctuations are found largely to rely on two different variables. Like in the study of Mweru fisheries by Gordon (2003), macro-economic conditions and wage labour opportunities in Zambia seem to effect the recruitment of fishermen to the fishery. It seems that when general economic problems increase the number of new fishermen grows and, when there are good job opportunities elsewhere the number of fishermen tends to fall. The other major variable influencing the number of fishermen are local access regulating mechanisms. That is, when the pressure on local resources increases, individuals who are not already part of the fishermen community are prevented from fishing on the lake. These mechanisms are particularly effective if inviduals are not from the area. In the case of Lake Kariba it is interesting to notice that the access regulating mechanisms have been brought into force not because of increased pressure on inshore fish resources, but because arable land and territories for tourist development are in high demand.

The results of this study clearly demonstrate the validity of Brox’s argument: provided that effort development is population-driven, commons in fisheries may fulfil important functions as a buffer and a safety-valve for many individuals in times of national economic crisis. Local, access regulating mechanisms on the other hand, which are important elements in various co-management strategies, tend to reduce this function and may, like in the case of Kariba, lead to under exploitation of important natural resources. It remains to be noticed that one important issue is not being dealt with in this paper, namely the question why investment-driven changes do not seem to have taken place on Lake Kariba in all these years. Answers to this question may be sought in another of the contributions in this publication (see Overå, 2003).

2. EFFORT DEVELOPMENT ON LAKE KARIBA ACCORDING TO AVAILABLE STATISTICS

Since its start in the late 1950s the inshore fishery on Lake Kariba has always remained a simple gillnet fishery, mainly undertaken by individual producers from small non motorized dug-out canoes. Fishing is done throughout the year and there is a constant and important mobility along the lakeshore and from the shore to a great number of unpopulated islands from where the fishermen - at least until 1994 - freely seek what they consider to be some of the best fishing grounds. Formally, access to fisheries is free for any Zambian citizen. The view among policy makers and fishery experts has been the conventional one that effort is continuously increasing causing reduced biological production and therefore to further increases of effort. We find this view being conveyed in most of the available policy documents and expert reports (Chipungu and Moinuddin, 1994; Walter, 1988) and even in some research works (Scholz, Mudenda and Möller, 1997). Musando (1996) is more cautious, but here too the image of a steadily increasing fishing effort is reiterated. There are many understandable reasons for this interpretation. First it should be recalled that many were produced in, or immediately after, the 1980s when effort arguably grew very quickly. In addition, the 1980s was the period where relatively good data on effort and catches were produced. However, this picture of steady increases in effort is hardly representative of what happened before and after the 1980s.

Data on effort and catches before 1980 are weak, unsystematic and scattered and it is not possible to establish the exact changes in effort with any reasonable degree of certainty. They derive from a multitude of different sources reflecting different methods of collection and seldom deal with more sophisticated variables than numbers of fishermen, boats and/or nets. Figures of total yearly catches are found in the annual reports from the Fisheries Department but it is unclear on what basis the figures were reached. It is only since the beginning of the 1980s that there has existed a fairly coherent system of data collection. The system is based partly on extrapolation of selected catch and effort data (slightly modified in 1992) and partly on frame surveys. These consisted of census of fishermen, nets and boats carried out in the Zambian part of the lake and were undertaken on average every three years. The two methods are tuned against each other. In this paper data from the 1980s are based on a combination of the catch and effort data and the frame surveys. In some cases data on fishermen and nets have been tuned against other sources considered to be more reliable (e.g. Walter, 1988). For the 1990s, catch data are estimated on basis of the catch and effort collection system, while data on numbers of fishermen and nets are based on frame surveys undertaken in 1990, 1993, 1995 and 1999. All are official data from the Zambian Department of Fisheries. Data on fishermen and nets in years when frame surveys have not been undertaken are simple interpolations made by the author.

Figure 1 shows the development in catches and numbers of fishermen and nets from 1982 to 1998. The data shows that effort must have grown considerably throughout the 1980s while there seem to have been a stabilization and later a reduction in effort after 1990.

FIGURE 1. Development in catches, fishermen and nets 1982-1998

A rough estimate of changes in catch per unit effort (CPUE) expressed as catch per fisherman and year and as catch per net and year for the same period is presented in Figure 2. The similar shape of the two curves indicates that the number of nets per fishermen is relatively stable. It varies between six and ten nets. In both cases CPUE seems to have grown in the early 1980s. From a perspective of conventional fisheries sciences it seems strange that catch rates are growing in a period when effort also increases. After 1984-85 catch rates are falling until they reach a minimum in 1992. This is of course the expected development. Since 1992 catch rates have slowly increased.

Variables related to the frequency of fishing (number of fishing nights) are not taken into account, but there are few reasons to believe that frequency of fishing varies very much (see e.g. Scholz, Mudenda and Möller, 1997:259). Furthermore, no significant technological changes have taken place except for a tendency towards a reduction in the average mesh size. The author therefore considers the picture as fairly representative for how effort has changed during this period, and the trends also conform to what fishermen and other local producers in the area report.

FIGURE 2. Development in CPUE expressed as caches/fishermen and catches/net 1982-1998

These preliminary results from the last 18 years are interesting. It would of course be desirable to have an idea of changes in effort for a longer period, but the data does not permit the development of such a view. However, if we go back to the earlier descriptions of the fishery, we discover that, technologically, it seems to have been practised without much change from the time the lake was created. So if we assume that the number of nets per fishermen, just like in later years, has not changed, this means that the number of fishermen in itself gives a fairly good picture of how effort has changed over the years.^[97] Although all the older data must be considered unreliable, those on numbers of fishermen are, for purely methodological and practical reasons, relatively more reliable than the others: it is easier to count fishermen than nets and catches. The development in numbers of fishermen from the start of the fishery is presented in Figure 3.

If Figure 3 is seen as a picture of the development of effort it clearly indicates that we are faced with a history characterized by a highly fluctuating fishing effort. We therefore propose to investigate, in some detail, major sociological factors that may explain the fluctuations in the number of fishermen presented in figure 3. For the years preceding the middle of the 1960s, the analysis mainly reflects works of other social scientists (Colson, 1962, 1971; Scudder, 1960, 1965, 1972). From then on the analysis is that of the author. Although aggregated quantitative data do not exist for the period 1971-1979, qualitative data exist which indicate in which direction effort developed and we shall integrate these into the analysis. If the analysis of the period after 1980 is more detailed it only reflects the availability of better and more reliable data.

FIGURE 3. Development in number of fishermen 1959-1999

3. INTERPRETATION OF THE FLUCTUATING NUMBER OF FISHERMEN

3.1 1958-1980: continuous ups and downs

The construction of the Kariba dam and the consequent creation of a lake covering some 5 350 square kilometres of land in the Zambezi valley on the border between what was then Northern and Southern Rhodesia entailed one of the biggest resettlement schemes ever undertaken on the African continent. In Northern Rhodesia alone it is estimated that as many as 35 000 people were directly affected. Most of them were of Gwembe Tonga origin^[98] and mainly practising an extensive type of agro-pastoral exploitation. The relocation process was complicated in that the non-inundated areas apt for this type of livelihood were already densely populated. There was therefore a strong need to establish alternative income generating activities and the colonial government made considerable efforts in facilitating the development of a commercial fishery in the new lake.

According to the then prevailing principles of indirect rule, land rights were formally allocated through the native authorities which in this case was the Gwembe Tonga Native Authority. It comprised the Mwemba, Sinazongwe, Chipepo and Simamba chieftaincies. Through negotiations with the colonial authorities the native authorities managed to acquire the authority to regulate the fishery on the lake. This meant that until 1963 when the system was changed, the Gwembe Tonga people exercised formal and exclusive rights to the fishery. These rights seem to have been respected in practice.^[99] The British wished to establish a commercial fishery and employed large resources for that purpose. This effort included a series of projects such as stocking of fish, the cutting of trees in large parts of the inundated areas, establishment of facilities for training members of the Tonga community in skills needed to become fishermen, establishment of fishing camps and marketing structures, as well as providing credits to fishermen for buying nets, boats and other material (Scudder 1965). The substantial government support and the high biological productivity that characterized the period immediately following the creation of the lake (Kolding, Musando and Songore, 2003) led to the realization of very interesting profits which attracted many young Tonga into the fishery. In the years following the start of the fishery, sometimes in 1958 or 1959, the number of fishermen grew rapidly to attain around 2 500 in 1963.

But, according to Colson (1971) commercial fishing was never taken up as an integrated part of the existing agro-pastoral economy of the Tonga households. On the contrary, fishing became an activity largely dominated by young men who sought a quick increase of their individual incomes. In this manner, fishing became an alternative to the long-established practice of labour migration for the young Tonga. The difference was mainly that an increased number of young people were given the opportunity through fishing. But, in a society where authority and power to a large extent were based on seniority, large incomes for younger people created a series of social problems between the fishermen, the households they belong to and their families. As is often reported to be the case in rural communities in Africa, the prosperous young men, difficult to control, were regarded as a challenge to the established leadership of elders to an extent that the most successful fishermen risked severe problems of non co-operation in their home communities.

This, together with concerns about fishing not being easy to combine with other economic activities and requiring a life apart from the villages for long periods of the year, led a great majority of the fishermen to reinvest profits into other productive sectors, such as agricultural cash cropping (also made possible by the creation of the lake) or trade. The removal of tsetse flies, which also were part of the government’s undertaking when building the dam, opened up for important investments in cattle.

By September 1962 the lake had reached its maximum level. But already by 1964 a 50 per cent decrease in total annual catches was reported, clearly indicating a strong reduction in biological productivity compared to the inundation period. Retaining the previous profit margins could only be done by intensifying the fishing effort, either through higher investments or through more intensive exploitation of the lake which meant also increased mobility on the lake and along its shores. Since profits already were channelled into other economic projects, increased mobility would amplify the need to coordinate with other activities and lead to stronger tensions in the local communities. Most of the newly established fishermen chose therefore to leave fisheries as quickly as they had entered the profession. In the three years after 1962 their number is reported to have fallen from approximately 2 500 to 500.

This pattern of leaving a newly created occupation and investing accumulated profits in more established occupations is a well known and quite common strategy (see e.g. Henriksen; 1974, Jul-Larsen 1981 concerning the attempt to establish a fisheries among the Turkana in Northern Kenya). The works of Colson (1971) indicate that the young men either resumed to labour migration, started cash cropping (cotton) or took on other types of salaried employment. A few remained in fishing, but it is symptomatic that these individuals disengaged increasingly from Tonga social life. At the start of political independence for Northern Rhodesia, as the Republic of Zambia, we must conclude that the objective of the colonial power to establish individuals of the Tonga as fishermen largely had failed.

As planned, the concessions given to the Gwembe Tonga Native Authority of restricting access to fisheries to the Valley Tonga were abolished by the colonial powers in 1963. The reason for this may be found mainly by studying the political preparations for national independence in 1964. The policy of indirect rule and territorial rights according to tribal identity and origin was to be substituted by a concentration of power at the level of the central government and the new constitution clearly stated the supreme rights of the government to all natural resources in the country. Therefore, the institutions creating native authority, which had been established by the colonial powers in the 1930s, such as the one for Gwembe Tonga, were formally dismantled. This does not mean that traditional political authority^[100] disappeared. The system certainly took new forms and new strategies were developed among its main actors in order to assure continued access to vital resources. Interesting accounts are found about how traditional authority and tribal identity have been strategically utilized in order to improve access to fishing grounds in southern African freshwaters (e.g. Chirwa, 1995; Aarnink, 1999). However, the disappearance of most of the Tonga from fisheries did not create much incentives for chiefs and village headmen to insist on the exclusive rights to the lake for the Tonga. On the contrary it could prove useful (and profitable) to allow new actors onto the scene.

Colson (1971) dates the entry of foreigners (i.e. people from outside the Gwembe district) to 1963 and the abolition of the native rights. However, Malasha (2002) indicates that their presence may be traced back to before the start of the fisheries. Already in the middle of the 1950s, the government had introduced a certain number of people from the north to help to clear trees from the areas to be inundated. Later, people with the same origins were recruited to help in the training of the new Tonga fishermen. Most of them belonged to different groups of Bemba speaking people who already had long experience from lake fisheries in Bangweulu and Mweru Luapula. A certain number of these individuals seem to have chosen to assimilate with the Tonga and ally with Tonga political leaders. The fishing experience of the Benga speakers and their local knowledge probably caused the Tonga leaders to treat them as allies and use them to recruit and build a new ‘tribe’ of fishermen on Lake Kariba. Unfortunately, not much is known about the relationships between the first foreigners and the local chiefs and headmen after 1963. It is not know to what extent the work of the foreigners depended on or was approved by the chiefs. In any case, the fact that the rights of the Tonga to the lake were abolished at the same time as the Tonga left the fisheries, opened up for the foreigners to invite other newcomers, mainly from their own area of origin, to come and join them. Initially the increase in number is limited. Figures do not exist for the early 1970s, but according to interviews with fishermen operating at that time, and who were still present in the area at the time of the study, the influx of foreigners was limited. The modest growth is somewhat surprising when we know that both the new government as well as the international community continued to support the development of fisheries in the lake^[101], and when we compare it with what happened some 10-15 years later. Although few data on the ethnic or regional composition of the fishermen exist after the mid 1960s, it is in the first half of the 1970s that the presence of foreigners becomes ‘common knowledge’ with regard to the Kariba fisheries.

Whether Valley Tonga or newcomers from outside, the fishermen were forced to reorganize their activities. The mode of work used during the first years was not longer appropriate. Not only fish had become scarcer but also fishing gear and other vital commodities became less easily available. In addition access to arable land was limited. The only way fishermen could compensate for these developments was to increase mobility and to constantly seek new resources as well as gear and fish traders. Contrary to the initial ideas, which envisage to group fishermen and traders in a selected and limited number of fishing camps, the new modes of operation meant the formation of a mobile, dispersed and individualized fishing population. Attractive areas were found around the more than 100 islands in the Zambian part of the lake. The shore-lines of these islands account for more than 20 percent of the total shorelines in the Zambian part of the lake (Pearce, pers. com.). One week a fisherman could spend in a village with fellow fishermen, while in the following weeks he would reside alone on one of the islands, before going back to the village to take his wife and their children with him to some inaccessible area on the mainland far from other people. Evidently, work units operating according to such principles have to be small and “one-man” enterprises were quite common, even though most fishermen were reported to be married and be accompanied by both wife and children. A consequence of this shift in exploitation patterns was a brake down of much of the infrastructure which initially had been put in place by the authorities such as houses and marketing facilities in the centralized camps. It meant a complete reshuffling of marketing structures which introduced serious constraints for increases in production.^[102] To some extent it was compensated by a shift in processing methods from sales of fresh fish towards more drying. But, little is known about how the commercialization of fish developed in this period.

By 1974 the Zimbabwean independence struggle had entailed a series of attacks by the Rhodesian army into bordering Zambian areas and the Zambian shores of Lake Kariba were particularly affected. The increased insecurity forced the Zambian government to officially close the fisheries in 1974. However, the closure did not lead to a complete halt of the fishery but rather to another restructuring. Information from fishermen operating during this period indicates that few of the foreigners left, neither the lake area nor the fishing. Instead, they were forced to choose fishing grounds and whereabouts according to the degree of security they provided, rather than according to how much fish they thought could be caught. At the same time fish traders are reported to have stopped coming to the lakeshore. The fishermen had therefore few opportunities but to seek as much protection as the Tonga villages could provide and seek to survive through barter and some cultivation of food crops for their own subsistence. Life stories from people having experienced this period tell us that a quite common strategy at the time for foreigners was to assimilate to Tonga customs and ways of life. No doubt fishing effort must have come down to a minimum in this period even if as many as 900-1 000 fishermen are reported to have operated in the Zambian part of Lake Kariba at the time of independence in 1980. It seems fair to assume that the number of fishermen may already have been of that amplitude when the fishery was formally closed six years earlier, since the recruitment of newcomers can not have been very high in that period.

3.2 The 1980s: the big increase in the number of new fishermen

At Zimbabwe’s independence the improvement in the security situation meant that fishing could be taken up as before. In some ways this is also what happened. But while the period from 1963 to 1974 had been characterized by a modest increase of fishermen, the 1980s saw a noticeable acceleration, particularly from 1983 and onwards. The accepted view has been that the foreigners continued to constitute the bulk of the newcomers, but already in 1985, Beck (1985) documented that the increase was constituted as much by Tonga as by Bemba people and the work of Walter (1988) confirms this. According to Walter’s findings, the Gwembe Tonga constituted between 30 and 40 percent of the fishermen in 1988. Furthermore, the data also show that among the foreign fishermen, there was by that time a great variety of origins and ethnic identities represented and Bemba speaking people only constituted somewhere around 35 percent of the total population of fishermen. These findings in many ways contradict the very resilient image among government staff that the increase has to be seen as a result of overfishing in the Bemba dominated waters in the north (Luapula, Mweru, Bangweulu) and a subsequent turn of these people towards Kafue and Lake Kariba.

The two questions that need to be answered are: why was the acceleration of newcomers in the 1980s so much stronger compared to the period before 1974? And, how to explain the changes in composition of the fishing population? Very little had happened in relation to the inshore fishery which can explain such dramatic changes and the main answers have therefore to be sought elsewhere. In the survey undertaken for this study in 1998, including 426 fishermen located in the Mwemba and Sinazongwe chieftaincies in the two eastern fishing zones of the lake,^[103] the foreign fishermen were asked about their immediate occupation before they started fishing on Lake Kariba. Seventy-three percent of those who started fishing in the 1980s reported to have come directly from some kind of wage labour, most often in the Copperbelt or in Lusaka and that the reason for shifting to fisheries was connected to the loss of their employment. The crisis in the Copperbelt and in the Zambian economy is generally seen to have started after the oil crisis of 1973. However, at that time the security situation at the Lake had already begun to deteriorate and a year or so later the fishery was formally closed. It was a risky venture to move to Kariba in that period. However, when it reopened people soon sought the opportunity represented by the underexploited lake.

After 1980 it is hence the general crisis in the Zambian economy and the subsequent reductions in wage employment which seem to constitute the main reason for the sharp increase of newcomers. This also explains the changes in the composition of the population. The crisis struck all ethnic groups and all regions of the country equally. The valley Tonga as well as Bemba and other groups were all involved in labour migration to the Copperbelt or sought wage labour in the cities and major towns. All had equally good reasons to seek the opportunities of fishing and the financial capital requested was as we already have indicated very moderate. The form and organization of the fishery remained very much the same as the type that developed after 1963; a fishery that was very dispersed and individualized with a high degree of mobility. The islands again became a preferred base to operate from. The normal procedure for entering the fishery would be to join an already established fisherman for six months to a year to learn enough about the local conditions (both environmental and social) to operate on your own. According to numerous interviews with chiefs and headmen, little was done by the local and the traditional authorities to prevent the newcomers in establishing themselves as long as they did not request, or claim land.^[104] If little was done it also had to do with the fact that the establishment of new individual settlements or fishing camps was extremely difficult to control. As we shall see this situation was soon going to change.

There was also an additional reason why Lake Kariba emerged as a particularly promising venture for many jobless people in the 1980s. In 1983 the Zambian authorities opened the fishery for Kapenta, a small pelagic offshore species which in 1967 had been introduced into the lake from Lake Tanganyika.^[105] Kapenta fishing requires mechanized rigs, (electric) light attraction, winches and big dip nets. The Kapenta fishery is undertaken at night by paid workers recruited locally, but not necessarily from the area. The fish is then taken either to one of the islands close to where the rigs are fishing or to the shore to be dried and commercialized. The fishery needs investments and technological knowledge far beyond what was within reach of the inshore fishermen. From the beginning all the operators were white settlers/entrepreneurs from the plateau who either established themselves, or a managing representative, at the lake. Most of them were also involved in a range of other economic activities.

For people who recently had lost their jobs and wages, the Kapenta fishery represented an alternative, even if they had to travel to Kariba to seek employment in the fishery. While waiting they sought to survive as well as they could and many of them found a means in the inshore fishery. Some who were not employed on the rigs or who simply changed their minds became part of the increase of inshore fishermen. It is unclear to what extent crew members on the rigs later sought to establish themselves as inshore fishermen, In our conversations with people it was referred to as a common strategy, but in the 1998 survey there is not a single fisherman who reports to have come from the Kapenta industry.

Kapenta operators soon discovered how difficult it could be to control employed labour and the catches landed by them. According to the operators, fresh and processed Kapenta soon became subject to a substantial illegal sale, taking place between the crews on the rigs and the traders. The inshore fishermen, occupying the islands and disposing of canoes were considered as essential allies in the conduct of this trade, but it goes without saying that in the absence of data their role is difficult to establish with certainty. The delicacy of the relationship between Kapenta operators and inshore fishermen may explain why no inshore fisherman in the 1998 survey wanted to admit any connection to the Kapenta business. Nevertheless, it is probable that ‘opportunities’ in the trade of stolen Kapenta contributed to the increased recruitment of inshore fishermen in the 1980s.

3.3 The 1990s: reductions among the ‘foreigners’

In 1989 the number of fishermen had attained around 2 500 individuals or about the same number as had been registered in 1962. From then on the number of fishermen seems to stabilize before it starts falling sometime in 1994 or 1995. With reference to the causes for the preceding growth, just being discussed, both the stabilization and the fall are surprising as the national economic situation and the prospects for jobs and wages in fact did not improve. Even if the introduction of multipartism in 1991 and the subsequent liberalization of the economy could tempt analysts to think that the macro-economic conditions would have improved, a recently published World Bank report (Rakner, Walle and Mulaisho, 1999) concludes that, on the contrary, living conditions and employment opportunities continued to decrease in the 1990s. The causes of the fluctuations in fishing effort - as indicated by the number of active fishermen - must be sought elsewhere. One possible reason for the stabilization of their number could have been the low water levels which started to appear around 1990 (Kolding, Musando and Songore, 2003). But this factor can not explain the reductions after 1994 as employment then started to fall at about the same time as the climatic conditions once again began to improve. We therefore propose to focus in some details on the competition for resources taking place locally.

As already mentioned, the local chiefs and headmen did not, or were not able to, intervene effectively with regard to the establishment of all the newcomers in the 1980s. As long as the newcomers’ activities only affected fisheries it was not really a problem. The competition for the fish resources were modest, even if the number of Tonga participating in the fishery started to grew again in the 1980s. However, the increased presence of foreigners inevitably led to competition and conflicts regarding other resources. In order to understand this situation a certain number of additional factors need to be taken into account.

Lake Kariba is characterized by considerable variations in climatic conditions which directly influence the biological productivity of the lake. In particular the hydrological levels of the lake (Kolding, Musando and Songore, 2003) are important. Furthermore, partly due to the considerable shifts in fishing effort over the years, and partly as a result of the dispersed location patterns of this fishing, the marketing systems on the lake have remained weak. In various ways both factors augment the insecurity of anyone who tries to adapt to circumstances by specializing in fishing and forces fishermen, irrespective of their origins, to seek access to alternative resources to sustain themselves and their families. The most obvious alternative resource is land for agriculture or for animal husbandry. If access to the lake has been relatively unproblematic to attain for people from outside, access to land has not. In fact, the persistent struggle by the Tonga to avoid outsiders being given access to land has been a permanent characteristic in the Tonga/outsider relations since the creation of the lake.

At the entry of the 1990s the Kariba inshore fishery found itself in a somewhat paradoxical, but not necessarily uncommon situation; it had become seriously constrained by social conflicts. These had built up progressively during the 1980s and were created as fishermen increasingly demanded access to agricultural land, while the local population would use drastic means to prevent them from getting it. A good example coming to the attention of the author occurred in 1997 when Tonga villagers, with the consent of their headman, fiercely opposed the establishment of a cemetery for foreign fishermen nearby their camp. The Tonga claimed - not without reason - that the existence of the cemetery would improve foreigners’ rights to claim agricultural land in the area.

The reasons why competition for land leads to such serious conflicts is complex, but should first of all be sought in the unclear rules for and underlying values about the principles guiding the allocation of land existing in the area. From research on local access regulating mechanisms in sub-Saharan Africa (for Zambia, see e.g. Berry, 1993 and Moore and Vaughan, 1994), it is known that there are normally not one set of (more or less consistent) regulating principles. Rather there tends to exist a whole range of principles, often based in contradictory logics. This leads to a situation of uncertainty and ambiguity as to how land and other vital resources are being allocated and regulated. Hence, different group of actors can, with some kind of legitimacy, claim rights to land (or parts thereof) by referring to all sorts of personal connections to it.^[106] On the one hand such a situation offers actors to seek access through a range of different strategies; on the other hand no-one really knows which principles are effective when and where. This unclear situation is also prevalent in the communities on the shores of Lake Kariba. The regulating principles increasingly being used by the foreigners to get access to land included marriage into Tonga lineages, alliances of various kinds with chiefs and headmen and investments. Some of the individuals employing these strategies have succeeded, some have failed and given up, but the majority remain in a state of uncertainty. Some Tonga may say they are allowed to cultivate, or keep animals in certain spots, while others contest any right to land whatsoever for foreigners. Inevitably, this has led to a considerable increase in conflicts, not only between outsiders and Tonga, but also among the foreigners and, may be most important, among the Tonga population itself.

To understand the stabilization in the number of fishermen the increased level of conflicts over land represents an important factor. It does so in two different ways. On the one hand is the stress it imposed on the outsiders causing some of them to decide to leave the area. On the other, and probably as important is that those in the fishing camps invested with the traditional authority realized that the growth in newcomers had to be stopped, and that - overall - the number of fishermen had to be reduced. However, the chiefs did not try to expel people already operating, but when possible limited recruitment of newcomers. According to themselves, they tried new means to reduce the recruitment of outsiders mainly by influencing the outsiders already in place, not to recruit apprentices. The author’s survey also indicates that the stabilization after 1989 mainly was due to reduced number of newcomers. While the average number of foreign newcomers in a total population of 313 foreign fishermen was 16.7 newcomers per year in the 1980s, this dropped to 10.7 newcomers per year in the 1990s.

The chiefs and the headmen had an additional reason to actively try to reduce the number of fishermen. Their location in a border area made smuggling and poaching a flourishing business. By occupying the islands and being among the few with access to boats, the fishermen were extremely difficult to control and the problems they faced in getting access to land only increased their incentives to use smuggling to diversify their risks. Although, not directly a responsibility of the traditional leaders, increased level of smuggling fuelled the general conflict level in the valley even among the Tonga population itself. According to Chief Sinazongwe, the high level of conflict over land and other activities forced the chiefs to take action. Since the late 1980s control over the recruitment of fishermen was constantly discussed among the chiefs and headmen. But one thing is to realize the problems, another is to find adequate means to solve them. Even though increased conscience and some control measures may explain a part of the stabilization of fishermen, neither the traditional nor the local authorities (councils, police, customs or fisheries department) were strong enough, to influence the number of fishermen to any significant degree. The local authorities certainly functioned as an ally to the traditional authority at a rhetorical level, but they lacked both means and incentives to exercise the control required by the chiefs.

What was going to make the big difference was the entry of the Kapenta operators in the overall competition for resources in the lake. Their non-involvement in the inshore fishery had made the Kapenta operators keep in the background during their first years of their operations. However, the problems of theft from the rigs soon made them realize that it was also in their interest to control and limit the number of inshore fishermen. Besides, the economic liberalization which followed the political changes of 1991, created prospects of tourism which interested the Kapenta operators as potential investors. The biggest islands became potentially important for establishment of lodges and wildlife sanctuaries. In May 1995 the author were informed by the Local Council in Sinazongwe that five major islands already had been leased to tourist/Kapenta operators on 50 years lease contracts. Also from this viewpoint the limitation and control of inshore fishermen became crucial for some of the Kapenta operators. The operators could do little alone; as whites and wealthy they were and are the ‘real’ strangers in the valley. They depended on the Tonga chiefs in order to be able to operate successfully. Chiefs had to approve the land allocations necessary to establish their production facilities and to support them in order for the operators to remain on good terms with the local population. The Kapenta operator/chief relationship was mainly established on individual basis and was often delicate and on occasions even hostile. Nevertheless it soon became quite tight and characterized by an understanding of some sort of mutual dependency. This dependency derived from the simple fact that, unlike all other local actors, the operators disposed of considerable financial resources and modern equipment which was crucial in exercising control of inshore fishermen. So, the chiefs had the legitimacy and the Kapenta operators had the means.

Therefore the alliance between the chiefs and the Kapenta operators, supported by the local authorities, is likely to be the main reason for the reduction in the number of inshore fishermen. Already in the early 1990 it is reported that Kapenta operators, in various ways, practically supported the chiefs and village headmen in their attempts to improve control over the fishermen. They provided the necessary means (speed-boats, petrol, etc.) for control trips to the islands and they supported the organization of local meetings. Already from the beginning of the 1990s, one can trace effects of this alliance, but paradoxically, it was an initiative coming from the government in 1994 which should render their alliance particularly effective.

General concerns about falling catch rates, and about what many believed to be overfishing, led the Zambian Department of Fisheries to launch a new co-management plan for the inshore fisheries early in 1994. In addition to employing traditional measures related to mesh sizes and prohibition of certain types of fishing techniques, the new plan would make it possible to allocate exclusive rights to fishermen in defined areas, by relocating and concentrating them into a more limited number of permanent fishing camps (Chipungu and Moinuddin, 1994; Malasha, 2002). It was argued that exclusive rights would eliminate the problems of free riders and irresponsible behaviour. The shores of the islands would still be open for fishing, but it would be prohibited to stay there at night. Such a prohibition de facto meant that the islands would become inaccessible for inshore fishing, given their distance from the shore. The new permanent camps were to be governed and managed by a committee elected among the camp population. In turn camp-committees should elect representatives to one of four “zonal” committees where, except for fishermen, there would also be representatives from local government, traditional authority, Kapenta operators and the traders’ community. The plan specified that the funding needed to run this co-management system should be assured by the retention of a part of the fishing licenses collected by the Department of Fisheries and by trade levies on fish collected by the Local Councils.

It appears that this plan contained so many ambiguities and unresolved issues that it became impossible to reach the objectives specified in it. Also, the fact that fishermen strongly resented parts of the plan made its implementation an illusion. In particular, the question of closing the islands to the fishermen was resented, even if all sorts of promises were made regarding improved living conditions in the new permanent camps. Nevertheless, most of the relocation of fishermen had been implemented less than two years after the plan had been presented. According to the Department of Fisheries’ frame survey of 1993 and 1995, the number of fishing camps had been reduced from 278 in 1993 to 67 in late 1995. As a result the conditions imposed on individuals wanting to operate as specialized inshore fishermen became so constraining that a considerable number of foreign fishermen chose to leave the Kariba fisheries altogether. According to the same frame surveys, the total number of fishermen fell from 2 238 in 1993 to 1 355 in 1995.

Unlike what many observers of Lake Kariba fisheries tend to believe it was neither the Department of Fisheries nor the management plan as such that caused the considerable reduction in number of fishermen in 1994-95. The reason why the relocation of fishermen took place so quickly was the fact that the alliance between Kapenta operators and chiefs made the initiative a reality. Long before the Department of Fisheries had started to consider specific actions, the Kapenta operators through their Kapenta Fishermen’s Association (KFA) and the chiefs (sometimes with the consent of the Local Council) started implementing the plan by pushing for a relocation of fishermen from the islands (Jul-Larsen, 1995). Chiefs, accompanied by with impressive courts, travelled around in the area by car and to the islands by speed-boat, explaining to the population how good things would become in the new permanent camps and threatening those who resisted moving. As always in such cases far more was promised than could be kept and particularly serious were many promises given regarding access to land. These campaigns and meetings were to a large extent funded and organized by the KFA^[107]. The drama was completed when in early 1995 the chiefs and the local Council managed to convince a troop of the military police force, attending a training camp in Sinazongwe, to undertake some ‘reality training’ by landing soldiers on the islands and remove inshore fishermen by force (Pearce, 1995). In such a situation it is understandable that the foreign fishermen, despite a lack of good opportunities elsewhere, preferred to abandon the area.

Even after the forced relocation, KFA continued to support the establishment of the camps, both practically and financially.^[108] But as time went by little was done on the part of chiefs or by local and central government in providing the infrastructure and the social services which had been foreseen. The chiefs met strong resistance from many of their village headmen concerning the promises made on land allocations, local government proved to be reluctant to allocate the promised part of the trade levies of fish to the management committees, and government had no additional funds to provide. The Fisheries Department never established the exclusive fishing zones which, from the point of view of fishermen, were one of the most important aspects of the whole plan. Despite the interest by some KFA board members, it became impossible for them to continue their support under such conditions and they slowly withdrew from the whole process. A common experience recorded from earlier relocation processes of rural populations in Africa, is that people start returning to their old places and continue to live and work as they did before. Here is how Sara Berry describes relocation of farmers in northern Zambia: “Moreover during the first decade of independence, the government pursued a policy of village regrouping which bore a close resemblance to colonial efforts to stabilize rural settlements. At first, villagers showed some enthusiasm for ‘village regrouping’, but voluntary relocations declined as promised services and amenities failed to appear. People who were already well situated in terms of access to jobs and markets or who were loathe to give up established gardens for an uncertain future never participated at all” (Berry, 1993).

Also in Lake Kariba the same tendencies soon became evident. Some of the individual Kapenta operators who had leased islands tried to prevent the return of inshore fishermen by use of force, but they failed. It is probably correct to say that, already in 1997, the fishermen made as much use of the islands as they had done before the relocation started. However, it seems as if the extremely mobile and individualized fishing that was common until 1994 were given up and that fishermen, at least to some degree, had accepted to use the permanent camps as a residential base (Overå, 2003). It should also be mentioned that one of the results has been a clear switch in the identity of fishermen in the camps. Tonga fishermen are now a much larger proportion of the total number of fishermen than before and for the first time since 1963, they were able to establish their influence and exercize some power in the fishing communities.

One important question remains: why has the number remained so low also after the dramatic reductions in 1994-95? A growth in numbers could have been expected as a result of the collapse in the alliance between the chiefs and the Kapenta operators. It was this alliance which provided the means and the legitimacy to enforce the relocation process and as per 1999, there were no institutions, alone or in cooperation, which could effectively control the recruitment and the mobility of inshore fishermen. From this perspective it is surprising that, according to Department of Fisheries’ frame surveys 1995, 1997 and 1999 no growth has taken place. On the contrary, there has been a continued but slight reduction; from 1 355 fishermen in 1995 to 1 263 in 1997, to reach only 1 170 fishermen in 1999. The author is uncertain of how to interpret this latest development and can not provide any well developed explanation. First of all the quality of the frame survey data needs to be checked. It may be that the number of fishermen recorded after 1995 is not directly comparable with numbers of fishermen reported before that year, since some informants claim that the frame surveys of 1995 and after only have included the “official” permanent camps. So, qualified interpretations must await study of the frame survey data.

4. CONCLUSIONS

Brox’s (1990) analytical distinction between population-driven and investment-driven changes in fishing effort relates to whether the changes are the result of changes in the number of fishermen or in individual accumulation of gear and technological changes. The reason for making this distinction is that it permits the analyst to study separately the underlying causes of change. These causes are not the same for population-driven and investment-driven changes. Therefore, often the consequences for biological sustainability, for social welfare and for equal rights to natural resources also will differ. This review of effort development in Kariba shows first of all that, although changes in fishing effort and changes in the number of fishermen is not one and the same thing, the number of fishermen probably constitutes the most important variable in determining effort development. Effort development in the inshore fishery of Lake Kariba is arguably much more population-driven than investment-driven. Furthermore, it demonstrates that population-driven changes are not simple reflections of demographic trends. The changes in Kariba are characterized by a considerable growth in, as well as by reductions in, the number of fishermen, even in a situation where the general demographic growth is noticeable and fairly stable. In Lake Kariba the main factors influencing the changes in number of fishermen can be summarized as follows.

First, the overall changes in what may be termed the macro-economic (to some extent also macro-political) conditions at the national and regional levels. The overall economic situation in the country and thereby the opportunity for jobs seem to have significantly influenced some of the most dramatic changes in numbers of fishermen. It is the war for independence in Zimbabwe that explains fishing effort at a minimum in the late 1970s, even after the general economic crisis had emerged in Zambia. And it is the same crisis that drives thousands of recently jobless people to try fishery on Lake Kariba. But this variable can not be seen in complete isolation; it must also be compared to changes in the opportunities created by the specific fishery in question. In 1963 when there are relatively good prospects for finding jobs in the Copperbelt and in cities inside or outside Zambia, the reduction in biological productivity in the lake is part of the context that explains the rapid decrease of fishermen at that stage.

Secondly, the analysis shows that changes in the macro-economic conditions together with the situation in the fishery do not always act alone. While the general economic crisis in Zambia to a large extent explains the growth in the number of fishermen in the 1980s, the prevalence of the crisis through the 1990s did not lead to continued growth during that period, but to stabilization and a subsequent decrease. In this context the analysis shows that local access regulating mechanisms also may be crucial. The larger the pressure on natural resources in Kariba, the more important local access regulating mechanisms become, even if these resources are not those of the inshore fishery. It is the pressure on arable land and on the territories of the islands (even the thefts of Kapenta may be seen as increased competition for resources) which explains why certain groups of actors such as chiefs, headmen and Kapenta operators see an interest in pushing and even implementing policies aiming to exclude inshore fishermen from the same resources. Although many social scientists are right when they insist on the problematic assumptions of free access regimes in fisheries, this analysis shows that assumptions that local access regulations are effective also can be quite problematic. The study shows that local access regulation is not particularly effective in maintaining biological productivity or in assuring the equal distribution of resources at the local level. Local access regulating mechanisms seem to have developed more in response to peoples struggle for control of and access to the resources, than as a manifestation of collective concerns for the sustainability of the resources or for the welfare of the communities.

The implications of these findings clearly demonstrate the utility of Brox’s analytical distinction. The role of macro-economic conditions in determining people’s wish to join fisheries as well as the great mobility of people implicates that this sort of fishery serves as a buffer and a safety valve to ease the consequences of fluctuations in the macro-economy. Such a situation makes it problematic to define “who are the fishermen” because those who are not fishing today may easily become fishermen tomorrow. And later, both tomorrow’s fishermen and those already in business today may decide to leave if conditions elsewhere improve. But, of course, the very high geographical and social mobility of Kariba inshore fishermen is not necessarily a bad thing from the point of view of society. They move over long distances and between countries and they easily switch occupations according to what they can obtain and what they think may generate the most interesting profits. From 1963 to 1994, it can be said that Lake Kariba functioned as a commons to the benefit of all those who fished there and without creating any fishery tragedy.

On the contrary, the establishment of strong access regulating mechanisms at the local level, which also was the intention of the proposed co-management regime, has limited the access to the commons to an extent that the lake is left considerably underexploited. By saying so the author does not underestimate the problems caused by the high number of foreigners in 1989-1990: the Tonga naturally have to protect what land had been left for them after the Zambian society took its share in the 1950s, the Kapenta operators are entitled to protect what was and is their property, tourism may prove beneficial to more than the investors and the chiefs in lake shore communities have a legitimate right and duty to minimize the levels of conflicts in their communities. But, the important question is: to what extent all these legitimate concerns can be adequately dealt with without removing what has proven to be an important safety valve for a great number of Zambians? Co-management in fisheries certainly has many faces.

5. REFERENCES

Aarnink, B.H.M. (1999), Socio-political struggles over fish, in Towards negotiated co-management of natural resources in Africa, Venema, B. Breemer, H.v.d. (eds.), LIT Verlag, Münster, 275-304.

Beck, R. (1985), Gwembe district rural development project, Pähl.

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), Common property theory; epistemological status and analytical utility. Human Organisation, 49 (3): 227-235.

Chipungu, P. & Moinuddin H. (1994), Management of the Lake Kariba fisheries, Zambia-Zimbabwe SADC fisheries project, Project report no.32, Kariba.

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

Colson, E. (1960), Social organization of the Gwembe Tonga. Human problems of Kariba Vol. I., Manchester University Press, Manchester.

Colson, E. (1962), Trade and wealth among the Tonga, in Markets in Africa, Bohannan, P. and Dalton, G. (eds.), Northwestern University Press, Chicago.

Colson, E. (1971), The social consequences of resettlement; the impact of Kariba resettlement upon the Gwembe Tonga, Manchester University Press, Manchester.

Gordon, D.M. (2003), Technological change and economies of scale in the history of Mweru-Luapula’s fishery (Zambia and Democratic Republic of Congo), 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, FAO Fisheries Technical Paper 426/2, FAO, Rome.

Henriksen, G. (1974), Problems of development in Turkana, Dept. of social anthropology, University of Bergen, Occasional paper no.11, Bergen.

Jul-Larsen, E. (1981), The lake is our shamba; certain socio-economic aspects regarding the Lake Turkana Fisheries Project, CMI Working paper A244, Bergen.

Jul-Larsen, E. (1995), The politics of resource management; an assessment of the sociological aspects in connection to developing resource management regimes on Lake Kariba, Zambia-Zimbabwe SADC fisheries project, Project Report no. 40, Kariba.

Jul-Larsen, E., Kolding, J. Overå, R., Raakjær Nielsen J. & Zwieten P.A.M. van (2003), Management, co-management or no-management? Major dilemmas in southern African freshwater fisheries, Synthesis Report, FAO Fisheries Technical Paper 426/1, FAO, Rome.

Kolding, J., Musando, B., & Songore, N. (2003), Inshore fisheries and fish population changes in Lake Kariba, 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, FAO Fisheries Technical Paper 426/2. Rome, FAO.

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, Harare.

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. and Zwieten, P.A.M. van (eds.), FAO Fisheries Technical Paper 426/2, FAO, Rome.

Moore, H. & Vaughan, M. (1994), Cutting down trees: gender, nutrition and agricultural change in the Northern province of Zambia, 1890-1990, James Currey, London.

Musando, B. (1996), Inshore fish population changes in the Zambian waters of Lake Kariba, M. Phil. thesis, University of Bergen, Bergen.

Overå, R. (2003), Market systems and investment “bottlenecks”, in Management, co-management or no-management? Major dilemmas in southern African freshwater fisheries, Jul-Larsen, E., Kolding, J., Overå, R., Raakjær Nielsen, J. and P.A.M. van Zwieten (eds.), FAO Fisheries Technical Paper 426/2, FAO, Rome.

Pearce, M. (1995), The removal and return of fishers in the Sinazongwe islands, in proceedings of the inshore working group workshop on Lake Kariba, Zambia-Zimbabwe SADC fisheries project, Project report no. 45, Kariba.

Rakner, L., Walle, N. van de & Mulaisho, D. (1999), The politics of economic reform implementation in Zambia, World Bank, Washington.

Scholz, U., Mudenda G. & Möller H. (1997), Some aspects of the development of the small-scale fishery on the Zambian side of Lake Kariba 1961-1990 and implications for fisheries management, in African inland fisheries, aquaculture and the environment, Remane K. (ed.), Fishing news books, Oxford, 255-271.

Scudder, T. (1960), Fishermen of the Zambezi. An appraisal of fishing practice and potential of the Valley Tonga, Human Problems in British Central Africa, vol. 27.

Scudder, T. (1965), The Kariba case: Man-made lakes and resource development in Africa, Bulletin of the atomic scientists, December 6-11 1965.

Scudder, T. (1972), Ecological bottlenecks and the development of the Kariba Basin, in The Careless Technology: Ecology and International Development, Farver and Mitton (eds.), New York Natural History Press, New York.

Walter, G. (1988), Lake Kariba, socio-economic baseline study, Ministry of agriculture and water development, Lusaka.