The five case studies that follow serve several purposes. In the Amazon cases, the first example illustrates the nature of an ongoing traditional fishery, while the next two describe two rapidly developing commercial fisheries which combine both traditional fishermen and out-siders. Both Zambezi examples relate to longterm studies of two separate Central African fisheries, one on the Middle Zambezi and one on the Kafue. Both illustrate the speed and magnitude of the changes that have transformed traditional fisheries around the world. Although previous management strategies utilized within such fisheries may be applicable to present conditions, at the very least they will have to be adapted to those conditions. While each case history is different, the Amazon and Zambezi transformations illustrate trends that are worldwide.
The process of commercialization and intensification of inland fisheries that has occurred throughout the developing world in recent decades is well illustrated by several case studies of fisheries located in the Amazon basin of Brazil and Eastern Peru.
The initial “primarily subsistence” stage is typified by the case of the Cocamilla Indian fishing folk who live along the Huallaga River in Northeastern Peru. A traditional Cocamilla community of 350 persons studied by Stocks (1983) in the mid-1970s exploits several areas of the Huallaga floodplain including the open channel of the river, drying ponds left in shallow depressions when the river recedes during the dry season, and small tributary streams. Their most important fishery, however, is a “varzea” lake, a term used in this case to refer to a “living” lake that is permanently connected by natural canals to the main river channel. Though the bulk of the fish caught by the Cocamilla is for local consumption, approximately 15 percent is marketed via traders who seasonally visit the area.
The Cocamilla employ a wide range of techniques in exploiting the fish resources of the varzea lake. Using small one or two man canoes, they use a simple hook and line method along the shoreline during daylight hours. They also utilize several varieties of nylon gill nets as well as cast nets thrown from the front of the boat. Though less common than in the past, spear fishing is also practiced on the lake. Finally, each December the Cocamilla also organize a communal fish harvest using a local vegetable poison. After a small portion of the lake is closed off by nets, the poison is placed in the water. The stunned and dying fish rise to the surface where they are speared or pulled from the nets that encircle the area. (See Hickling, 1961:108–237, and Welcomme, 1979:182–199, for a more general description of traditional riverine fishing methods.)
The intensity with which the Cocamilla fish varies considerably over the year. During periods of low water when the fish are most easily caught because they are concentrated in a relatively restricted area, fishing intensity is high. During periods of high water fish are more difficult to catch and as a result the Cocamilla spend less time fishing and increase the amount of effort expended in hunting forest animals that offer an alternate source of protein. The timing and intensity of fishing is also conditioned by the agricultural activities of the Cocamilla.
Despite the lack of sophisticated gear and motorized boats characteristic of large-scale commercial fisheries elsewhere in the Amazon, the productivity and efficiency of the Cocamilla fishermen are quite high. Based on detailed measurements over an entire year, Stocks found that the yield of fish from all sources in the area ranged from 0.6 kilogram of fish per hour of labor in April, at the peak of the flood period, to 5.6 kilograms per hour in November during the low water season. The overall average for a sample of three fishermen over the year was 2.12 kilograms of fish harvested per hour of labor. Assuming an eight hour day, this results in a respectable yield of about 17 kilograms of fish per man day. In the varzea lake, the main source of fish exploited by the community's fishing folk, the Cocamilla harvested roughly 17 metric tons per square kilometer of water.
Since the Cocamilla fish primarily for subsistence, Stocks is not able to provide an estimate of the cash income achieved by the community's fishing folk that could be used as a measure of the relative success of their fishery. Stocks does discuss, however, the nutritional status of the community that can serve as an alternate indicator of the well-being of the population. Based on measurements of food consumption taken three times over the year, Stocks found that the Cocamilla diet was adequate in terms of both calories and protein consumed, though there was some seasonal variation in the availability of protein. Over the year, the caloric intake was consistently around 2000 kcal. per person per day. The daily protein intake, gained primarily from fishing with a supplement from hunting, ranged from a high of about 75 grams per person per day in the low water season when fish were most easily harvested to 50 grams of protein in the high water season. These figures show that, though there may be a seasonal decline in protein intake, the diet achieved by the Cocamilla, derived from a combination of fishing, hunting, and agriculture, is sufficient to support the population in a good state of health.
Though the Cocamilla men do the bulk of the fishing, women still contribute significantly to the production of fish. Women participate in the communal harvests using fish poison each December and also often assist in cast net fishing by paddling the canoes while a man tosses a net from the front of the boat. Though Stocks does not specifically examine differences in level of living among the Cocamilla, his discussion implies that there is little socioeconomic differentiation in the community. This is also seems to be the case among other subsistence-oriented fishing groups in the Amazon such as the Shipibo (Bergman, 1974; Behrens, 1981) and the Siona-Secoya described by Vickers, (19).
Two recent studies by Goulding (1981) and Smith (1981) of larger scale fisheries in the Amazon basin of Brazil serve to illustrate the worldwide process of rapid commercialization that has taken place in recent years. Smith discusses the fishing industry surrounding the town of Itacoatiara on the Amazon River and Goulding examines the fishery around the town of Porto Velho on the River Madiera, a major tributary of the Amazon. Like the Cocamilla area of Eastern Peru studied by Stocks, these two fisheries, especially the more remote Porto Velho fishery, were primarily subsistence-oriented until the late 1950s or early 1960s. At that time, the region's population began to increase rapidly with the construction of the first roads in the region that culminated with the completion of the Trans-Amazon highway in the early 1970s. In 1950, for example, the population of Porto Velho was only about 12,000 persons. By 1970, after the town had been connected by highway with the city of Cuiaba in the province of Mato Grosso, the population had quadrupled to some 50,000 people. By 1975, the number of inhabitants had doubled again to 100,000.
During the same period, more sophisticated types of fishing gear, most importantly nylon gill and seine nets, were introduced into the area. In addition, with the growing local and export market for fish products, ice plants were opened to help in the preservation of fish for long distance shipment. During the same period, large-capacity motorized fishing vessels were also introduced. After less than fifteen years, seine nets now account for 70–75 percent of the total catch in the Itacoatiara and Porto Velho fisheries. Though traditional fishing techniques, similar to those described for the Cocamilla, are still in use (primarily for subsistence), their importance is declining rapidly. Both communities are now the base for fleets of motorized fishing vessels that often travel considerable distances in order to reach productive fisheries. A much larger fishing fleet from the major port town of Manaus, upstream from Itacoatiara on the Amazon, grew from 140 in 1970 to 800 vessels in 19 (Smith, 1981:122).
As in most floodplain fisheries, there is a pronounced seasonality in fishing activities along both the Amazon and Madiera Rivers. In both fisheries, the low water period accounts for the largest part of the annual catch, though some of the largest harvests may occur during upstream fish migrations that take place at periods of rising or high water. During times of the year when the returns to fishing are low, most Itacoatiara fishermen turn to agriculture or wage labor to supplement their income. Many own or rent small plots of land on which they cultivate both subsistence crops and cash crops such as jute. Others may work as wage laborers for large landowners or for one of several industries in the area such as in lumber, jute, or rubber processing plants.
The highly commercialized fishing that began in the Itacoatiara and Porto Velho areas in the late 1960s resulted in an initial jump in the productivity of the fisheries (stage three). After several years of exploitation, however, yields in the vicinity of the two ports began to drop significantly (stage four). At Porto Velho, for example, Goulding's data show an increase in the productivity of the fishery from only 450 tons of fish annually in 1968 to 1900 tons in 1974, a year when there was a sharp increase in fishing effort and also unusually low water levels. Since 1974, however, there has been a steady decline in the annual production of the Porto Velho fishery with a yield of only about 900 tons in both 1977 and 1978.
At the same time that total yields have been declining, the catch per unit effort (c.p.u.e.) has also fallen. In 1974, Porto Velho fishermen enjoyed a return of about 54 kilograms of fish per man day. By 1977, the c.p.u.e. had dropped to only 26 kilograms of fish per day. Though there was a slight rise during 1978–79, the efficiency of the Porto Velho fishery is still only about half the 1974 level. At Itacoatiara, Smith reports a similar c.p.u.e. of about 23 kilograms per day for the late 1970s. These averages, of course, mask significant seasonal variation in efficiency and also do not distinguish between different techniques of fishing such as gill and seine nets (for details see Goulding, 1981:60–68).
Based on ten years of data for the Porto Velho fishery, Goulding states that “given the present economy of the Amazon fisheries, somewhere between 20 and 30 kilograms per man day fished is the minimum level at which the commercial fisheries can function economically.” At a rate of return below this he would expect the number of fishermen to decrease or the price of fish to increase. It is interesting to note that despite the sophisticated gear employed in the Itacoatiara and Porto Velho fisheries, their level of efficiency, around 25 kilograms of fish per man day, is not much higher than the returns of about 17 kilograms per man day achieved by the Cocamilla using their much less sophisticated techniques.
In summary, though there is apparently no evidence that heavily exploited fish species are in danger of extinction over the region as a whole, local depletion of commercially important species has become a problem and the productivity of fisheries in the Amazon has dropped significantly after an initial boom during the late 1960s and early 1970s.
The high cost of equipment in the commercial Amazon fisheries is beyond the reach of the majority of fishermen. At Itacoatiara, for instance, Smith reports that only a few of the wealthier fishermen are able to afford the very expensive lampara seine nets that cost $500 each. These same men also own the larger motorized boats capable of undertaking the long distance fishing trips that are the most productive means of exploiting the fish resources of the region. Less well-off fishermen are able to fish using small motorless boats and less expensive gear, such as gill nets or indigenous techniques, but are only able to reach the nearby fisheries that are no longer very productive. More often a group of fishermen will share the cost of renting a large boat or work as crew members for an owner of a large fishing vessel who takes 50 percent of the catch, dividing the remainder among the crew.
Though the owners of large boats can make a considerable profit, most small-scale fishermen appear to make a marginal living in Itacoatiara. Smith says that locally fishermen are considered to be rather low on the socioeconomic scale, slightly above the lowest category, migrant laborers, and about even with peasant farmers. Most fishermen earn an income above the minimum wage of $2.00 per day, but during the slack season when fish are difficult to catch they are forced to turn to other occupations such as subsistence farming or wage labor to support themselves.
Thus, we see that though fishing is still a viable economic activity in the commercialized fisheries of the Amazon basin, many of the small-scale fishermen, appear to achieve a barely adequate level of living. What is more, there is a growing gap between the majority of small-scale fishermen, many of whom are no longer independent self-employed workers, and the few well-to-do fishermen and entrepreneurs who control the major equipment and large fishing vessels used in the fisheries.
The Middle Zambezi fishery that commences at the bottom of the gorges below Victoria Falls in Zambia-Zimbabwe and extends over 300 kilometers downstream to the junction of the Kafue with the Zambezi illustrates the rapidity with which a traditional fishery can become commercialized. The data that follows is drawn from Scudder's longterm study of the development of the Zambian side of the fishery where the Gwembe Tonga form the large majority of the population.
In the early 1950s, there were no commercial fishermen among the Gwembe Tonga nor was anyone else commercially fishing the river. Gillnets or motorized boats had yet to be introduced, and no fish were exported from the Middle Zambezi Valley. The local people were nonetheless knowledgeable fishing folk who were familiar with all the larger species in the Middle Zambezi Valley which they fished with a range of traditional techniques adapted to specific water conditions and the behavior of the species sought (Scudder, 1960). Using different techniques except when they collaborated in poisoning isolated pools during the dry season, both men and women were skilled fisherfolk during the appropriate seasons, with fish providing an important source of protein.
As a result of the construction of the Kariba Dam in the latter half of the 1950s, and the filling of Lake Kariba between 1959 and 1963. the men commercialized their fishing in less than five years. During this period, expansion of the Lake Kariba fishery was a model of fisheries development and management. On the planning side, the new fishery was restricted to the local population for the first five years so that at least some of the indigenous population would have the opportunity to become professional fishermen before the lake was opened (in 1964) to all comers. A closed system coinciding with the rainy season (when agricultural activities peaked and general like Labeo move up tributaries to spawn) was also introduced from the start. As a further contribution to the development of a local limited access fishery, the government stationed both mobile and localized extension staff in the area, provided a revolving fund for loans to acquire gill nets (and accessories) and boats; imported improved boats to be purchased with those loans; imported outsiders and trained locals to build those boats; laid out the major fish camps in which markets were built; and established a major fisheries training and research center which subsequently became a national facility.
The Gwembe Tonga responded positively to these incentives. Eight months after the dam was sealed in December, 1958, 407 fishermen were counted in Zambian waters using 93 boats (of which 87 were dugout canoes) and 748 gill nets. Between 1960 and 1964 the proportion of improved boats increased significantly and the first outboard engines were purchased. Repayment rates on loans were excellent, with some fishermen paying off their debts in half the required time period. By the end of 1962 over 2000 Gwembe Tonga, using over 5000 gill nets, were fishing on the Zambian side. Recorded landings of fish exceeded 3000 tons, rising to over 4000 in 1963.
The local District Council also contributed in a major way to the rapid commercialization of the fishery (from which it also profited by taxing fish exports) by making and improving feeder roads to at least some of the fish camps and by establishing a lake transport system for fishermen and traders based on 26 footer inboard diesels. Though problem ridden, this service proved invaluable, not only allowing fishermen in remote camps to transport their dried product to the market centers but also to concentrate more on their fishing. During the boom years hundreds of small and medium scale private traders flocked to the lakeside, the fish trade being strictly a private sector operation which employed a significant number of outsiders (subsequently some local people became fishermen/traders, with a very small minority becoming full-time traders).
Catching one to three tons of fish per year, the majority of the fishermen made substantial savings some of which they invested in cattle and the education of their children. As noted previously the fishery was also the most important mechanism for drawing local women into a market economy for they came to the fish camps by the thousands to brew beer and sell farm produce.
As the initial period of high productivity that characterizes new reservoirs came to an end in the mid 1960s (stage three), the fishery went into a decline as total landings dropped to about 1000 tons (stage four). This decline was exacerbated by opening up the fishery at that time to outsiders and by increasing substantially the number of outboard driven mechanized units through a new loan program. As catch per unit effort declined, the better fishermen who had mechanized their operations were saddled with debts that they could not pay off while most nonmechanized canoe fishermen reverted to a subsistence mode of production. The majority of local fishermen, however, left the fishery, turning their attention back to agriculture or to other activities. Though the boom was over, investments made with savings during the good years nonetheless continued to play a major role in the agricultural and commercial development of the area during the 1970s (though the decline of the gillnet fishery continued throughout that period because of the war for Zimbabwean independence, an upturn began in the early 1980s).
While the Kariba case illustrates the speed with which a traditional fishery can be transformed into a highly commercialized artisanal one in which savings are reinvested in the development of the encompassing area, the rapid development of the Kafue Flats fishery illustrates how a highly productive fishery can be overwhelmed by its very success.
Covering 4,340 square miles at full flood (Gay as quoted by Welcome, 1979), the Kafue Flats fishery, in terms of both its biophysical and socioeconomic components, is one of the best studied in the world. On the socioeconomic side we have relied in particular on LaMuniere's unpublished Ph.D. thesis (1969) and on a series of more recent unpublished papers by Hayward (1981–1983).
The history of the Kafue fishery can be divided into three time periods of which the first lasts until approximately 1917, the second from 1917 to the mid-1950s and the third from the mid-1950s to the present. These coincide roughly with Goulding's three stages of commercialization, with our fourth stage occurring during the past few years. During the first time period, the surrounding peoples fished for consumption only. the large majority making their living as herder/farmers (the Ila) and as farmer/herders (the Plateau Tonga and Sala). According to Lehman, the Flats themselves were named after the Twa who “are reported to claim ownership of the fishing rights of the plain,” every group having a “specific limited area which it defended against intruders” (1977:41–42). Their exclusive access rights were rejected, however, by the colonial administration which declared the fishery open to all comers.
Even before colonial times, however, the Twa were fighting an unsuccessful battle against encroaching Lozi fishermen from Barotseland to the southwest. During the latter part of the nineteenth century the Lozi, under their Paramount chief, had extended their influence into the Kafue Basin, with an annual tribute first levied against the Ila and the Tonga in 1882. According to LaMuniere “the Paramount claimed the right to hunt, fish, to own land and to keep representatives in Ila and Tonga land” (p. 41), with official parties of Lozi fishermen sent to exploit the Flats at least by 1917. Thus began the incipient commercialization of the fishery, with the number of independent Lozi fishermen (that is, fishermen working for themselves rather than the Paramount) increasing over the years. By “1935–40, a large body of Lozi fishermen could be found on the flats every June at the beginning of the fishing season” (pp. 42–43). Though there were no reliable estimates of their numbers, by the time of the 1959 census(when their dominance over the fishery had already begun to decline), they totaled 3000 to 4000 fishermen.
LaMuniere attributes the rapid commercialization of the fishery during the third stage to three factors. The first was the introduction of relatively cheap, nylon gill nets in the mid-1950s, which allowed anyone who could save one month's salary as an unskilled worker to immediately enter the fishery as a small-scale owner (of one net)/operator (prior to that date Lozi fishermen had made their nets largely from old automobile tires, an arduous and time consuming process, to say the least). The second factor was the increasing demand for fish on the Copperbelt and in the rapidly growing national capital of Lusaka which was less than 50 kilometers from the eastern end of the fishery by an all-weather macadamized road. Increased demand stimulated the fish trade, with LaMuniere ranking “the expanding activities of fish traders” (p. 49) as a third factor exceeding in importance the introduction of gillnets.
During the second half of the 1950s a multiethnic population of fishermen drawn from throughout Zambia as well as from Malawi and Tanzania arrived in large numbers. In 1961 LaMuniere found that 70 percent of those in his sample has arrived during the past ten years. “The influx of fishermen from the Copperbelt was further increased in 1958 by the fact that a copper recession had set in and that thousands of men, in mining and allied industries, were finding themselves temporarily out of work.”
As the numbers of fishermen and nylon nets (both gill and seine) rose during the 1950s, estimated production increased from 1700 tons in 1954 to over 11,000 tons in 1958 at which time total fishermen exceeded 5000. During the next two years landings dropped significantly as a result of reduced flooding and heavy fishing pressure during periods of low water with yields rising again following a return to more normal flooding. Between 1961 and the construction of the Kafue Dam in 1972 they ranged from 2894 tons to 7992 with a mode of 5500 to 6500 tons. Though the dam, plus the construction of the Itezhitezhi Dam in the late 1970s, radically altered the flood regime of the flats, thereafter yields continued to average about 6,000 tons. While landings initially fell after dam construction because of the permanent inundation of many seining beaches, the fishermen “adapted to indefinite periods of high water by seining from boats anchored in flooded coves, by grass islands, or in open shallow lagoons” (Hayward, 1981 ). As a result landings rose again, with annual variations now correlated with the extent of the reservoir drawdown.
The commercialization of the fishery also provided significant employment for traders. Though no accurate counts have ever been made of traders, Hayward counted an average of 56 traders serving three harbors in the eastern sector of the fishery. Small-scale traders predominated, some of whom sold fresh fish daily in the adjacent Kafue industrial estate. As with fishermen during both the late 1950s and since 1974, many traders formerly had been unemployed, with 75 percent of those interviewed by Hayward stating that “they began the trade because they had no job” (1981 :13).
Though socioeconomically the fishery has weathered past reductions in annual landings, Hayward's analysis shows that currently the fishery is in a bad state. Since the mid-1970s Zambia's prolonged economic downturn has seriously. reduced employment opportunities, with many unemployed and underemployed men turning to fishing and the fish trade. Before the onset of the recent drought (which has devastated large areas of Southern Africa), approximately 6000 fishermen were harvesting an equal tonnage of fish. The majority, however, were catching less than a ton per year, which is probably a minimal figure for maintaining an adequate living standard in terms of poverty datum line considerations. As hard times have continued, upper income civil servants and other urban residents have also entered the fishery as a means of stretching (and in some cases doubling) their incomes. As absentee owners of boats and seining equipment, they recruit their labor force from among the increasingly marginalized unemployed who have become a rural proletariat. In Hayward's words, “economic pressure within the society as a whole is thus increasingly pushing the higher economic strata into informal activities and reproducing overall socio-economic stratification in the informal sector as well” (1981:7).
Though the flooding of the eastern portion of the flats (which is the most heavily fished and densely settled because of relative ease of access and its proximity to the major markets) favors the use of gillnets over seining equipment, theft of gillnets has become such a serious problem that their use has been drastically curtailed. At the same time the cost of a seine net has doubled in price since 1968 while the price of fiber glass boats has increased by a factor of six, with the cost of a single net and boat (without outboard) exceeding $2000.
With such costs restricting purchase of boats and seining equipment to the wealthy, upwardly mobile small-scale fishermen are increasingly turning to illegal techniques to increase their yields. The two most important are kutumpula and use of small meshes (below the legal minimum of three inches), which often are used in conjunction. Kutumpula fishermen use gill nets which they set in a semicircle prior to beating the surface of the water with wooden “paddles” in order to drive fish into the nets. Initially yields increased, while theft of nets decreased since now they were under constant observation while in the water. By 1981, however, the population structure of the fish was being altered in Hayward's opinion, with the proportion of fish over three pounds dropping on the Lusaka market. Hayward was also concerned that kutumpula fishing with small mesh sizes was also interfering with spawning. The drought no doubt has worsened the situation in the past two years as more fishermen chased less fish in shallower waters using illegal techniques which were just too efficient in their ability to capture existing stocks (Hayward, 1982:24–25).
The cases studied illustrate a number of important characteristics Except in the most isolated areas, they document the rapidity with which traditional fisheries can become commercialized. This commercialization can involve primarily local traditional fishermen, outsiders or a combination of the two. It can generate significant savings which may catalyze a process of development within the area of the fishery (Kariba) or be primarily exported from the area (Kafue).
The Amazon and Zambian cases also illustrate the extent to which riverine fisheries can absorb significant numbers of unemployed and underemployed people as both fishermen and traders, as well as provide investment opportunities to private sector entrepreneurs and salaried professionals. But as operations intensify, either within the Amazon or the Zambezi river systems, selective fish stocks, catch per unit effort and living standards can all be adversely affected, with many fishermen becoming a marginalized population of rural poor.
It will not be easy to reverse this situation which we believe is becoming increasingly common in a number of countries. In such extreme cases as the Kafue Flats fishery, not only has the influx of outsiders “broken older informal controls” but “apparently nothing can be done on the local level to curb excessive fishing practices” (Hayward, 1981a and 1982, respectively). On the other hand, to date government management strategies have also proved to be completely ineffective. If a solution is possible, the evidence assessed during this survey suggests that it will require a cooperative effort between strong organizations of local fishermen (and perhaps traders) and supportive outside agencies (both governmental and nongovernmental). Since few successful examples of such cooperation exist, solutions will also require a willingness to experiment with new ideas and new institutions, some of which are explored in subsequent sections of this report.
