Production figures for 1985 show that ten years ago aquaculture occupied a meaningful place in the traditions of just five countries in Africa and the Middle East: Egypt, Iran, Israel, Nigeria and Iraq (FAO, 1996b).

This poor showing can be related to the fact that, until recent times, Sub-Saharan Africa was little interested in mastering water control and irrigation; historically, unlike in North Africa and the Middle East, hydraulic traditions there have remained limited and generally rather elementary (Gourou, in ORSTOM, 1979). Nevertheless, one does find examples of traditional brushparks, drain-in ponds and shellfish culture. The best known type of brushparks are the Beninese acadja and the Madagascan vovomora (Welcomme, 1972; Kapetzky, 1981; cited in ICLARM and GTZ, 1991). The best-known type of drain-in pond is the Beninese whedo, a very old tradition noted by Bourgoignie (1972). As for shellfish culture, the Sudan seems to be the country whose traditions date back farthest (Balarin, 1988b).

In North Africa and the Middle East, traditional water management offers a favourable foundation for the development of aquaculture: is it not the case that the region’s top aquaculture country is Egypt, which has a traditional involvement in hydraulics and aquaculture (Lacroix, 1993)? By the mid-1980s, the Nile Valley howash, an apparently very ancient type of drain-in pond, had spread to cover an area of over 30 000 hectares (ha) (Sadek, 1984, cited in ICLARM and GTZ, 1991). This foundation helps to explain the ex nihilo aquacultural success of Turkey, Israel, Syria and Saudi Arabia, and the first steps in aquaculture taken in Tunisia and Morocco.

Production trends over the past ten years show that a few emerging aquacultural countries are enjoying success today. This seems to be because of their deep-rooted history in aquaculture. Consider Egypt, for example, whose centuries of experience in aquaculture (FAO, 1996c) have allowed it to adapt to cultivating new species (Mugil cephalus, Dicentrarchus labrax, Sparus aurata), or Iran, whose albeit less ancient traditions have contributed to a remarkable Cyprinidae boom (Abdolhay, 1996). Similarly, on a somewhat smaller scale, there is the tripling of Cyprinidae production in Iraq (IOFC, 1996a).

Low aquaculture output in Africa and the Middle East raises questions for developers and researchers alike: Africa accounts for only 0.4% of total world production in volume terms and 0.5% in terms of value, with the corresponding percentages for the Middle East being 0.5% and 0.7% respectively (FAO, 1996b). Based on these figures, aquaculture in these regions can be ascribed to a situation of underdevelopment. At present, only six countries are producing in excess of 10 000 t: Egypt, Iran, Turkey, Nigeria, Israel, and Iraq (FAO, 1996b).

However, looking at recent aquaculture production trends within the overall context of African economies can only make for a more subtle, more hopeful appreciation. Indeed, although Sub-Saharan Africa has undergone ten years of continuous socio-economic decline (1983-93), as confirmed by falling per capita gross national product (GNP), aquaculture has grown considerably with production having almost tripled in volume over the same period (Lemercier et al.,1995).

The success of aquaculture production seems due to four clear factors: (i) the fruits of centuries of experience in aquaculture (Egypt, Iran, Iraq); (ii) good use of international aid (Nigeria, Syria, Madagascar); (iii) domestic efforts on the back of research-development and extension (Saudi Arabia, Syria, South Africa); (iv) and the strength of west Mediterranean markets (Turkey, Egypt and, to a lesser extent, Tunisia and Morocco).

Over the 1989-94 period, the following three types of success were seen in North Africa and the Middle East:

the inland aquaculture boom: Iran (from 17 000 to 32 000 t, due to the Hypophtalmichtys molitrix), Nigeria (8 000 to 14 000t, due to Clarias and Chrysichtys), Iraq (4 500 to 13 000t, due to the Osteichtyes), Turkey (2 700 to 7 300t due to Salmo spp), Syria (1 500 to 5 000t, due to Cyprinidae and Oreochromis spp), Saudi Arabia (0 to 3 500t, due to Oreochromis spp), Madagascar (300 to 3 000t due to Cyprinidae);

the marine fish farming boom: Dicentrrachus spp and Sparus aurata (Turkey, from 0 to 8 000t; Egypt, from 0 to 2 500t), Mugil cephalus (Egypt, from 1 500 to 8 000t), molluscs (South Africa, from 200 to 3 000t); and

the shrimp farming boom: Madagascar (0 to 400t), Kenya (0 to 100t), Saudi Arabia (0 to 80t).

The classification of aquaculture production systems varies according to the criteria selected; Shang (1981), for example, identifies ten criteria and underlines the many problems involved in any classification. From the viewpoint of aquaculture literature on Africa and the Middle East, the least flawed solution seemed to us to be a combination of two criteria: the end purpose of culture and the level of management intensity. Consequently, three main types of production systems emerged: (i) the subsistence system, which identifies with extensive culture where neither feed nor fertilizers are added; (ii) the small-scale commercial system, which is related to semi-intensive culture where some additional feed and fertilizers are provided; and (iii) intensive culture, which relies mainly on externally-sourced inputs.

According to Coche et al. (1994), the subsistence system can be identified with the extensive system and represents the system most widely in use in Sub-Saharan Africa. It operates mainly in rural areas with the purpose of meeting nutritional needs. One aspect of the extensive system, integrated aquaculture, has developed to but a minor degree despite many hopes.

There are three main types of extensive system: dam stocking, rice-fish culture, and pond culture. Fishstocking in dams has been a secondary consideration, yet widely promoted in Ghana, Malawi, Tanzania, Zambia and Zimbabwe, with annual yields of between 0.05 and 0.3t/ha. Despite an extension campaign conducted in 23 countries in Africa, rice-fish culture has seen relative success in Madagascar alone; while it produced an additional natural supply of fry in Egypt, average yields remained low, in the region of 200 to 400 kilograms per hectare per year (kg/ha/yr). Pond culture is the most widely used extensive system having been identified in 37 countries; average yields would seem to vary between 100 and 500 kg/ha/yr (ICLARM and GTZ, 1991).

The primary objective of the small-scale commercial system is the marketing of quantities of surplus produce, with a share of the output going to feed the smallholder’s family. The small-scale commercial system is generally run on a semi-intensive level of management with average yields in the order of 1t/ha/yr, yet rising to as much as 5t/h/yr. This system essentially involves pond culture which is a production system that tends to maximize the available by-products from crop or livestock farming, taking inorganic fertilizers, organic compost or manure for use in fish farming. As a rule, farmers tend to regard this type of aquaculture as a secondary activity behind their core agriculture or livestock-rearing work; the best farms - mostly situated in peri-urban areas - find themselves with the dilemma of having to choose between these three activities competing for the same feed and fertilizers (ICLARM and GTZ, 1991).

Virtually the entire output of the commericial/industrial production system is bound for domestic markets or for export, rather than for subsistence. As a rule, this system involves high capital outlay and is geared towards maximizing returns on investment. Such an objective is generally achieved by means of an intensive level of management, involving large-scale production and a reliance upon inputs from off-farm sources.

There are two main types of intensive system: (i) the pond, tank or raceway system; and (ii) the cage or pen. The former system involves dense stocking of fish and shrimp, high frequency of feeding (formulated feed), aeration, and feed removal. The best results would seem to be those obtained from the shrimp farming of Kenya and Madagascar, where output reached

2 500 t in 1996, with average yields of 3t/ha/yr (Kasprzyk et al., 1996). Cage and pen culture is generally practised south of the Sahara in brackishwater or freshwater as noted in experiences, past and present, in the Côte d’Ivoire, Kenya, Nigeria, Tanzania, Zimbabwe and the Niger. But the greatest success of all results has been that of Dicentrarchus spp and Sparus aurata in Turkey, which managed to capitalize on the comparative advantage of sheltered inlets as opposed to the raceway marine farming of the selfsame species in Tunisia and Morocco.

Annual fish consumption in North Africa and the Middle East is estimated at 5 kg and 8 kg per person/year respectively, which is below the world average (Ben Yami, 1996a). Aquaculture development in these regions is being restricted by the populations’ overriding appetite for meat; meat-eating habits help to explain why there has been so little development in the cooling facilities necessary for the domestic marketing of aquaculturally-produced fish.

It is obviously quite a different story in places that have deep-rooted aquatic traditions: for example, along coastlines (in the Yemen, Oman, Tunisia and Morocco), on the Nile and Shott El Arab deltas or along big rivers like the Nile, the Tigris and Euphrates. Needless to say, a population’s taste for fish is plainly going to work to the advantage of aquaculture development. Conversely, the absence of aquatic traditions has long acted as a brake on aquaculture development, requiring the sort of extension campaigns seen in Turkey, for example. Meanwhile, the case for developing aquaculture is encouraged by stagnating or even dwindling catches, as witnessed when the stocks of Black Sea pelagics depleted and caused a slump in Turkish fish consumption (Ersan, 1977a).

Generally speaking, fish occupies a more important place in the diets of Sub-Saharan African populations. The series of FAO-initiated National Reviews for Aquaculture Development in Africa (authored by Balarin or Deceuninck) has shown that fish accounts for more than half of popular animal protein intake and as much as 70% in Sierra Leone (Balarin, 1984e). These real fish protein needs are not wholly supplied by the national production systems of each country studied, which explains their quasi-systematic tendency to resort to importation.

These patterns of consumption, as highlighted by consistent growth in domestic demand, have been confirmed by more recent studies (Westlund, 1995; Lemercier et al., 1995). Lemercier et al. stress that despite a major growth in both marine and inland domestic output over the past few years, the balance of trade for fish products is now showing a deficit in terms of quantity even if it remains positive in value; it appears increasingly difficult to meet local demand. They estimate the average supply of fish products for the whole of Sub-Saharan Africa to have been at 8.4 kg per head in 1990, well short of the developing world’s overall average (13.9 kg) and far below that of the industrialized countries (23 kg). It is equally important to note that per capita supply is, on average, three times higher in western than in eastern parts, and that there are major differences within individual countries; both those where per capita supply is higher than 20kg (Angola, Congo, Gabon, Gambia, Ghana, Senegal, Sierra Leone) and those where it is lower than 2kg (Ethiopia, Burkina Faso, the Niger, Rwanda, Sudan).

The Sub-Saharan fisheries sector would seem to be at something of a turning-point. Internal demand is growing rapidly, production seems to have leveled off, exports are stagnating and imports are diminishing on account of the macro-economic difficulties suffered by the majority of countries (Lemercier et al., 1995).

Over the 1989-94 period, Sub-Saharan Africa received very large sums of international aid for aquaculture - $US 129 million. However, the tendency now is towards cuts in both the funding (from 30 to 9 million annually) and number of projects (from 24 to 9 annually). Of international aid, aquacultural production received the great majority (80% of the total), followed by institutional support (14%) and training which received less than $ US5 million.

Of the 28 countries benefiting from national aquaculture projects, the biggest beneficiaries were Guinea, Madagascar, the Niger and Nigeria, each of whom received $US 10 million. Over the whole period, more than 50% of the international aid came from the development banks, particularly the African Development Bank (AfDB) with bilateral aid only accounting for 20% of the total.

As far as North Africa and the Middle East were concerned, international aid for aquaculture was more than ten times less than the sums going into Sub-Saharan Africa. Indeed, for the 1989-93 period, international aid amounted to only $US 9 million - peaking in 1991 ($US 6.5 million) and 1993 ($US 1.8 million) - for a total of 11 projects, i.e., an average of just over two projects a year.

The allocation of aid in North Africa and the Middle East was different to Sub-Saharan Africa, with (in descending order) institutional support (45%), aquaculture production-processing-marketing (30%), and training and research (25%).

Just seven countries in the region enjoyed the benefits of national projects: Syria, Tunisia, Algeria, Egypt, Morocco, Lebanon and Libya. Over the whole period, more than 50% of the funding came from the United Nations Development Programme (UNDP), with a quarter coming from the development banks (particularly the Islamic Development Bank) with the rest from bilateral or multilateral donors.

In Africa and the Middle East, structural adjustment and international trade agreements have affected, on the one hand, the investment and operating conditions of aquaculture firms and, on the other, the demand for and the ways of marketing aquaculture produce.

Regarding investment and operating conditions, one of the effects of structural adjustment has been fewer tax breaks for aquatic production and marketing chains, for example through loss of fuel subsidies. In return, export incentives introduced into the new national investment codes have translated into tax relief or subsidies for aquaculture infrastructure and installation servicing (Tunisia). These export incentives help to explain the boom in Mediterranean Basin seabass and seabream aquaculture, especially in the region around Turkey and, to a lesser extent, Egypt, Tunisia and Morocco.

In Sub-Saharan Africa, the implementation of structural adjustment has resulted in the devaluation of the CFA franc and other national currencies (the cedi and naira, in particular), leading in turn to an increase in the prices of imported aquaculture inputs and marine products and, hence, the emergence of local technologies and domestic inputs. Overall, the structural adjustment policies have had a negative influence on demand for (especially aquacultural) aquatic products: the drop in real per capita annual income in Sub-Saharan Africa (-2.2% for the 1980-93 period) and the Arab nations (-2.1% over the same period) has brought down per capita demand for such products. This is a radically different situation from Southeast Asia, where a 6.9% rise in real annual income per inhabitant between 1980 and 1993 helped offset rising aquaculture input prices (UNDP, 1996).

Application of the GATT Uruguay Round agreements has progressively lessened the advantages provided by the Lomé Convention or the accords settled between the Maghreb and European Union (EU). Indeed, the scheduled phasing-out of protectionism (tariffs and quotas) means a gradual erosion of the measures protecting the African-Caribbean-Pacific (ACP) states. This is already affecting the competitiveness of African and Middle Eastern aquaculture operators, who now have to come to terms with tough competition on the European markets.

As the international trade agreements bring down the trade barriers and costs inherent to international trade, new restrictions on market access take their place. In the field of aquaculture, the exporters of Africa and the Middle East are confronted with new health and safety standards and environmental regulations. Regarding health and safety standards relating to production and marketing, the countries in question are increasingly obliged to accommodate EU directives in their national laws in such a way as to earn the sanitary approval of the importing countries. As for environmental regulations, and with allegations of environmental dumping, the countries in question are having to clarify their legislation by fixing the conditions for aquaculture project impact study requirements, the granting or refusal of aquaculture licenses plus the establishment of standards on pollutants, regulations governing trade in live aquaculture products, and the import of non-native aquatic species.

Notwithstanding this, Lemercier et al. (1995) suggest that the Uruguay Round agreements could have a positive impact upon the development of aquaculture in Africa: i.e., a price rise for fish in keeping with the pace of rising food prices in general, plus a more favourable positioning of African products on African markets linked to growing demand and a more competitive opening up of the African markets.

If sector-based policies, such as those concerning aquaculture, reflect the philosophy and objectives of development policies in general, then three separate periods can be identified: (i) colonial and post-colonial times; (ii) the period of infatuation and failure (the 1970s and 80s); and (iii) a new departure (the 1990s).

Apart from the rare countries, such as Egypt, that have always had a meaningful traditional involvement in the field, aquaculture was virtually non-existent in Africa and the Middle East during colonial times, even if colonizers did carry out a few small-scale localized experiments in Algeria, for example, or in Morocco, South Africa, Togo or Rwanda. Typically, experiments involved a transfer of European intensive aquaculture models. These high intensity capitalist models incurred high production costs, counterbalanced by the strong purchasing power of settlers and a desire for eating fresh fish; these are still in place in South Africa (René et al., 1994). The first initiatives in carp and then trout farming appeared in South Africa in the second half of the 19thth century for the purposes of recreational fishing; it was only at the end of the 1940s that commercial carp and trout culture became established (Hecht et al., 1990).

Meanwhile, development of fish culture became an imperative in the wake of the Second World War and during the 1950s as a means to tackle food supply shortages. Lazard et al. (1991) remind us of a few landmarks: the setting up of the Kipopo Pisciculture Research Station in the Belgian Congo (1948), the first Symposium on Freshwater Hydrology and Pisciculture in Africa in Entebbe (1952), and the setting up of the Bouaké Pisciculture Station in the Côte d’Ivoire (1956). By way of example, during the 1950-60 period, with the first tilapia trials carried out at the Kipopo station by the Institut National pour l’Etude Agronomique du Congo Belge having proved positive, the colonial authorities launched a vigorous extension drive which was followed by rapid development (Lazard et al., 1991). By the end of the colonial period, there were approximately 300 000 ponds in production in Sub-Saharan Africa as a whole (Satia, 1989).

The post-Independence slump in Sub-Saharan aquaculture came about for several reasons. The first was the departure of the colonists, e.g., from Algeria and the Belgian Congo, as they had been an important part in commercial piscicultural production and had possessed strong purchasing power. The second reason was the reduction in financial as well as technical aid for what was conventionally termed ‘peasant fish farming’. Furthermore, aid for extension efforts dropped at a time when local developers lacked sufficient technical knowledge. Other reasons for the decline have been suggested: flaws in the construction and choice of pond sites; no command of rearing techniques; lack of available agricultural by-products; no integration of fish farming with other activities; and assimilation of fish farming into a subsistence activity with no concern for profit-making (Lazard, 1992). Inadequate command of the grow-out cycle goes a long way toward explaining this decline: indeed, it was only in the 1970s that the monosex-with-predator technique was introduced for rearing Oreochromis niloticus (Oswald, 1996c).

During the 1970s, there was a resurgence of aquaculture projects because of their popular appeal amongst bilateral and multilateral cooperator. However, the resurgence was not preceded by any systematic evaluation of previous failures (FAO, 1987). Compared to the 1950s and 1960s, African pisciculture benefited from the improvement during the 1970s of rearing techniques: this was the decade when priorities shifted from finding biotechnical solutions to the search for economic equilibrium (Doumenge, 1986). In Sub-Saharan Africa, rearing techniques were improved for the following: the mastering of tilapia proliferation; the perfecting of reproduction in captivity and the rearing of other native species (siluriforms); the development of aquaculture structures other than ponds (pens and floating cages); the development of agricultural and agro-food industry by-products for fish-feeding; and the development of mineral and, above all, organic fertilizers to stimulate natural reproduction in ponds (Lazard, 1992). These acquired advantages fueled the popularity of fish farming in the course of the 1970s and 1980s, when two basic categories of fish-farming development can be seen as having been conducted in West Africa: industrial fish farming and peasant fish farming (Oswald, 1996c).

Industrial fish farming has often appeared the best way to rapidly overcome protein shortages. However, every major effort in the field has ended in failure: the Blohorn shrimp farm in the Côte d’Ivoire (Anon., 1993), the SIAL lagoon-rearing of Chrysichtys in the Côte d’Ivoire, and the Banfora farm on the Comoe river in Burkina Faso. This type of development did not result in any meaningful redistribution of income, it proved fragile and was overly dependent upon by-products, and was crippled by irreversibly high production costs.

Peasant fish farming was designed to be implemented by farmers, which meant very little initial funding as they would be digging their own ponds. A good number of development projects raised interest in this sort of fish farming amongst African peasants. By way of an example, more than 2 000 Côte d’Ivoire peasants decided to invest sums in excess of a million CFA francs. However, an a posteriori assessment showed that tilapia stocks were badly managed, predators were not used, the suggested densities were far too high (meaning that it was impossible to produce saleable-sized fish), ponds were poorly equipped and the state-run institutions took on an overbearingly important role (Oswald, 1996c). René (1994) suggests a further explanation: as the intensive fertilization technique originating from Asia and certain East European countries required skilled labour, it could only develop well in Egypt, where intensive aquaculture was already part of the country’s traditions, or in certain very specific areas close to an attractive market such as Kinshasa (Democratic Republic of Congo).

Several explanations for these failures have been suggested. The bilateral and multilateral cooperators - leading actors in the development of aquaculture in Africa - would appear to have given the various States far too big a role in the production processes. These cooperators also miscalculated the markets, did too little to strengthen institutional support and, finally, opted for an approach meant for production rather than the production and marketing chain.

Directly or indirectly, the public sector has long been deeply involved in piscicultural production via public or parapublic companies. As a rule, the State tended to opt for a more production-oriented approach (CEA/FAO, 1994). African and Middle Eastern methods of state intervention had two effects. First was a focus on biotechnical aspects to the detriment of institutional aspects, training and extension. Second was the reliance on technical cooperation which promoted foreign technology while relegating transfers of technology to a level of secondary importance. The choice of a production-oriented approach via state-run institutions, rather than a production-and-marketing-chain approach accounting for upstream and downstream sections of the production chain, did not help circulate these new technologies or weave a genuine economic fabric revolving around aquaculture activities.

Lemercier et al. (1995) stress the mistaken evaluation of the aquaculture products markets vis-à-vis other aquatic products or protein sources. Doumenge (1986) reminds us that domestic markets that are oversaturated with fish supplies or are too poor to absorb top-of-the-range products are not suitable for the expansion of aquaculture. In the Côte d’Ivoire, Weigel (1989) shows on the one hand that the amalgam of different fishery and aquaculture products defies specific studies that have highlighted the segmentation of aquatic markets and, on the other, that misconceptions in the domain of fish pricing and the production and marketing chains led many projects to produce incorrect estimates as to the marketing costs and means of distribution of potential aquaculture production.

Harrison et al. (1994) emphasize the misunderstanding of the institutional context, particularly the fact that training and extension remained a low priority for development projects of the 1970s because the needs had been ill-identified. These authors stress another recurring error: the fact that aquaculture was not examined within the context of the rural economy as a whole and, in particular, that women were left out of aquaculture projects. All this resulted in an incorrect evaluation of the constraints on mobilizing the factors of production, meaning that nobody foresaw the potential disputes over the use of land and water resources.

Weak output, growing deficits, lack of durability and persistent appeals to financial backers forced those backers to ponder the reasons for the impenetrability of the state institutions’ management of aquaculture activities. This resulted in a series of new approaches inspired by new trends in the social sciences: the systemic approach; the participatory approach; the promotion of an extensive system; and the overhauling of the market. In the field of aquaculture, these fresh notions were conceived by developers rather than researchers, which is why they were followed up with aquaculture planning and development proposals.

Aware of the fact that fish farming in Sub-Saharan Africa was often more or less integrated (as a secondary activity) into agriculture, the aquaculture development project managers found that the systemic approach conceptualized in the 1970s by rural economists could provide a satisfactory framework for examining that integration. As a result, developers working on the development of rural societies became interested in a production system approach revolving around a means of combining the factors of production. Broadly speaking, it involved giving emphasis to socio-economic organization and the methods of using natural resources rather than limiting oneself to biological and technological aspects: socio-economic and ecological conditions are what determine the choice of technological solutions and not the other way round (World Bank, 1991b). But the importance of bio-ecological factors in the field of aquaculture led researchers and developers to broaden the notion of a production system: the systemic approach applied to African aquaculture has been conceptualized in two ways depending on the origins (English or French-speaking) of their designers, although content analysis shows similarities in their respective notions of a resource system and agrarian system.

The ICLARM and GTZ (1991) document contains the following definition of resource system:

The term ‘agrarian system’ expresses the interaction between a bio-ecological system - represented by the natural environment - and a socio-cultural system via practices stemming mainly from acquired technical advantages; it aims at understanding the production processes, their changes and variations. It includes a range of interconnected notions ranging from basic technical operations to the production system passing via the crop or livestock management sequence and cropping or livestock farming systems. This notion of agrarian system is the one adopted by Oswald (1996), Dural (1994, 1995), Touré (1995).

The systemic approach has led to aquaculture project and sector planning proposals. As far as the projects are concerned, the choice of an integrated farming system linking fish farming with livestock-rearing according to Asian references (Pullin et al., 1995) has led to fish farming being regarded as just one of a number of project activities. Such has been the approach of projects of the International Centre for Living Aquatic Resources Management (ICLARM) begun in Ghana (Ofori and Prein, 1996) and Malawi (ICLARM & GTZ, 1991). The term currently in use is ‘integrated resources management’.

As for aquaculture sector planning, emphasis has been laid upon the notion of intersectoral management. Confirming the benefits of a systemic approach, the Study of International Fisheries Research Needs (SIFR) launched by the World Bank, UNDP, Commission of the European Communities (CEC) and FAO underscores the low level of institutional organization in the majority of integrated projects managed, in virtually every single case, by aquaculturalists under the institutional banner of the fisheries sector. The SIFR report (World Bank, 1991b) stresses that only an intersectoral approach can take natural resources management into account; however, such an approach would require a considerable interdisciplinary research effort (biology/ecology and socio-economics).

Many projects failed with the implementation of structural adjustment policies. Its resultant weakening of public services has gone together with the restoration of local institutions. Inspired by institutionalism, the participatory approach has become one of the new philosophies of aquaculture development. Two main constituents can be highlighted.

The first constituent is the reduction in public sector involvement which was supposedly to have been counterbalanced by the greater autonomy and responsibility of private operators or grassroots’ organizations (producers’ associations, non-governmental organizations/NGOs, private firms). In the late 1980s and through the 1990s, it made itself felt on the ground via a transfer of the ownership and management of the means of aquaculture production into the hands of private operators or grassroots’ organizations, as the State distanced itself from its former role as producer. Privatization took over from another option fashionable in the 1970s, i.e., the various guises of the cooperative movement (cooperatives, pre-cooperative or cooperative groupings).

The second constituent is the transfer of decision-making as to the appropriate production technologies and agendas to the producers themselves. The roles of the development institutions have become confined to offering technical assistance, leaving fish farmers free to develop their own technologies and agendas. Various types of approach have been systematized; Harrison (in Martinez-Espinoza, 1996) notes in particular the farmer participatory research (FPR) and the participatory action research (PAR), the rapid rural appraisal (RRA) and participatory rural appraisal (PRA). Instead of profitability or successful production targets, the participatory approach prioritizes training and, ultimately, the generalization of selected technologies by the fish farmers themselves. In Sub-Saharan Africa, the best-known projects involving participatory approaches have been the ICLARM project in Domasi (Malawi) and the Aquaulture for Local Community Development (ALCOM) project in Southern Africa.

To agree that aquaculture development projects should be integrated into existing socio-economic and technological organizations has meant taking the specific nature of African production systems into account, particularly the predominance of extensive systems in Sub-Saharan Africa (René et al, 1994). Africanist authors, such as Pelissier (Pelissier, in ORSTOM, 1979), stress the fact that the only productivity of any interest to the peasant is that of his labour and not of the land. The authors add that peasant logic gives priority to maximum income per work day, per unit of time worked, and even implies that focusing on high yields does not pay because it requires too much labour. Maximum labour productivity is assured by spatial expansion.

Other authors also point to the viability of these extensive systems and peasant rationality, since low yields from extensive farming are accompanied by satisfactory labour productivity. In other words, the Sub-Saharan farmer maximizes what he derives from his most scarce resource which, in Sub-Saharan Africa, is generally labour, not land (Couty, in Charmes et al., 1985). Couty underlines the fact that at the same time the extensive production systems enable the reduction of agricultural risks through spreading the use of scarce factors and the maintenance, or even growth, of labour productivity. Such is not the case with intensive systems where risk-reduction only comes at the price of some loss in the productivity of labour.

Applying these reflections to fish farming, Lazard et al. (1991) show that aquaculture has to take account of the fact that this relatively intensive new activity represents a technological innovation being brought into situations traditionally accustomed to extensive farming or livestock-rearing systems. Particular attention must therefore be given to integrating the technological innovations of fish farming into existing agricultural farming systems. Intensification must be gradual. This is the philosophy developed by three projects in particular.

The ALCOM project chose to promote a highly labour-intensive form of fish farming, mildly capitalist and simple in its technology. This extensive option would seem to have been quite successful, particularly in Luapula Province (Zambia), because the late 1980s saw a proliferation of ponds dug by the farmers themselves, irrespective of the low yields linked to low-density stocking and inputs (ALCOM, 1992). Constant attention was focused on the utilization of available resources and technical acquisition; the bulk of the effort then went into the monitoring of stocking checks, the suitable use of fertilizers and feed, water management and controlled harvesting.

The Fish Farming Development Project carried out in the mid-west of the Côte d’Ivoire made a similar choice by devising a system of reference comprised of techniques tailored to a context lacking in inputs: ponds dug by the farmers themselves, very low density, and the possibility of enlarging areas farmed through dams for pond drainage so as to allow the farmer’s labour to make satisfactory earnings from yields as low as 1 t/ha. Above all, technical choices were made in such a way as to allow the head farmer complete command over the factors of production such as the production of fry, the utilization of agricultural by-products, and direct marketing. This form of fish farming seems to have made quite a successful contribution towards the development of previously unexploited lowlands (Oswald, 1996d).

The successes of acadja (artificial reef) farming in Lake Nokoué (Benin), as described by Bourgoignie (1972) and Welcomme (1972) - an aquaculture technique that may be said to be extensive - has appealed to developers. Proposals have been made to enhance the system with acadja pens (Hem, 1997; Hem et al., 1995), and to extend the technique with funding from the Côte d’Ivoire, Canada and France, to the lagoon areas around the Gulf of Guinea, e.g., the Côte d’Ivoire lagoons or Lake Togo (Togo). However, a detailed analysis of the potential costs and profits of an acadja pen in Lake Togo has shown a number of problems that account for the limited subsequent expansion of acadja culture: the monopolization of space and property rights, deforestation and, finally, profitability due to the cost of branches of acadja in regions less wooded than Benin’s Oueme Delta (Weigel et al., 1989).

Generally speaking, the 1990s have seen a lifting of the confusion sustained by certain developers between two rather incompatible objectives: on the one hand, that of contributing to food security or the self-sufficiency of the respective farmers and, on the other, that of creating meaningful gains through the marketing of output on the domestic or export market. In both Africa and the Middle East, this confusion has exposed a weak understanding of the markets, a feature that ought to be regarded as stemming from a virtually exclusive focus on biotechnical rather than socio-economic aspects. The far too superficial understanding of the markets explains a few of the resounding failures in Sub-Saharan as well as North Africa: a far too rudimentary knowledge of the west Mediterranean Dicentrachus labrax and Sparus aurata markets made it impossible to produce reliable forecasts, thus giving a number of North African fish farming projects problems when they were unable to foresee the subsequent rise in the supply and fall in prices.

Nowadays, the confirmed success of the mariculture exports of Turkey or Egypt, or shrimp farming of Madagascar or Kenya is the direct result of the fact that the more recent aquaculture development projects take better account of the marketing dimension and that voluntarist State action has translated into development research, tax relief, and good terms for foreign investments. The stakes are high, for these are species with great market value. The examples of Tunisia, Madagascar, and Guinea serve to illustrate this situation.

In Tunisia, the voluntarist action of the State has translated into a funding mechanism for the aquaculture sector. The general provisions of the Investment Incentives Code stipulate that exporting companies shall be exempted from a certain number of taxes and duties as soon as their exports amount to more than 70% of total output. Exporting companies may be eligible for a variety of different levels of relief: total exemption from customs duties and Value Added Tax (VAT) on equipment plus other necessary producer goods, or abatements for income tax and company tax on net profits, etc. Financial incentives include subsidies for investment and research, with the State paying 50% of employers’ national insurance contributions for certain categories of employee (Ministère de l’Agriculture, PNUD-FAO, 1995).

In Madagascar, the promotion of Paeneus monodon shrimp farming has been underpinned in a number of different ways, notably through tax relief, because shrimp farming companies are now considered to be operating in a free trade zone and are exempt from taxes and duties and through active research development following construction of the Centre for the Development of Shrimp Culture, funded by the Japanese cooperation effort (Kasprzyk et al., 1996).

In Guinea, tax relief granted to aquaculture companies falls within the general system of exemptions and abatements which differentiates between various types of companies: small-scale units, industrial units or food factories. Relief, for example, can apply to company tax, flat rate tax, tax on turnover and apprenticeship, or import and export duties. The State is hoping that this will attract foreign and domestic investment for the development of shrimp farming (Anon, 1996b).

René et al. (1994) emphasize the shortcomings regarding the micro-economic data on aquaculture at the farm level. Micro-economic analysis is very often rudimentary in the preparation of projects as much as in their assessment. In particular, the majority of projects present summary accounts that either hide or undercalculate such issues as amortization, labour costs, land- and capital-related opportunity costs, fertilizer costs, etc. Taking three examples (Rwanda, the Niger and the former Zaire), the authors recall a number of basic truths such as the obvious link between the overall success of aquaculture development and the profitability of aquaculture activity at the farm level. The absence of the latter would explain the short-lived nature of projects once technical and financial assistance has come to an end.

Information is equally limited when it comes to analyzing the methods for mobilizing the factors of production. Lazard et al. (1991) emphasize the fact that pisciculture is a secondary activity in rural areas of Africa, and that it enters into competition with the dominant prevailing farming, livestock or fishing activities. It has to compete with these other activities for the land, inputs, aquatic space, labour and capital. More particularly, special attention needs to be focused on land ownership and (land and water) property rights, the local capacities and means of financing, and the possibilities and means of mobilizing labour.

Up until the early 1990s, no proper studies had been made on the relationship between aquaculture and household economics. Analysis of project documents shows that targeted populations were not studied in-depth and analysis of assessment reports confirms that not a single project was followed up with a detailed socio-economic study appraising the effects of having introduced aquaculture into the respective village economies. Fish farming is still in its infancy in Africa and the Middle East and, at this stage, monographic-type studies would be richly rewarding. In particular, the fishfarming systems operating without (or with minimal) monitoring deserve in-depth, case-by-case investigation; one especially interesting question would be to find out how hourly rates of pay in fish farming compares with farming or other activities (Lazard et al., 1991).

If one regards the degree of operational autonomy of the suggested fishfarming system as one of the key criteria contributing most to the development of fish farming, four principles may be put forward: developing fish farms should only count on local inputs; fishfarming activities should not constitute a burden on the household heavy enough to disturb its other agricultural work; workers should be paid according to a time scale that corresponds with the other agricultural activities of the household as a whole; and a large share of the households in the community concerned should directly or indirectly benefit from this new activity (Hem et al., 1995).

Four main references have been found on the socio-economic theme. They are, in chronological order, Balarin and Deceuninck’s national reviews of the 1980s, Lazard et al. (1991), Harrison et al. (1994) and, above all, the studies carried out in the 1990s within the framework of the ALCOM Programme.

A number of authors agree that the absence of in-depth studies on the social impacts of aquaculture is deplorable (Balarin, Deceunink). The integration of aquaculture activity into the existing social and economic organization should be judged as having had a positive social impact: Touré (1995), for example, refers to the example of fish farming in the mid-west of the Côte d’Ivoire.

As far as gender issues are concerned, information is virtually non-existent. Deceunink reports the existence of sexual divisions of labour in the Central African Republic and the Congo: for example, the men build the ponds and the women essentially take care of the maintenance, the feeding, and emptying and marketing of the fish (Deceunink, 1988b; Deceunink, 1985). It would be interesting to carry out a study on the working hours and share-out of aquaculture-generated earnings according to gender.

It is important to assess what is available: apart from the problem of finding skilled labour, it may even be difficult finding any labour at all. If labour is scarce, the amount of it supplied to aquaculture projects could also be calculated as a loss of labour for other activities; efforts must be made to render it compatible with seasonal patterns and to include traditional social labour management (Lazard et al., 1991). Christensen (1995) emphasizes two problems: labour input requirements need to be carefully researched and methods elucidated to reduce them; and novel methods to minimize the risk linked to the human capital investment and to the financing of pond construction need to be developed. Balarin and Deceunink look at the specific problem of supervising aquaculture extensionists: at the level of providing aquaculturists with adequate follow-up and monitoring the problem arises of the availability of competent agents. This is a problem shared, to differing degrees, by several countries: in Gabon, the shortage of top-level fishery managers; in Togo, the lack of extensionists; and in the Congo, the Central African Republic and Tanzania, the irregular follow-up and monitoring of fish farming.

Many authors above all raise the question of the amount and financing of credit. Aquaculture project managers and planners may find that it is in their interests to encourage recourse to informal credit, insofar as it comes under a traditional socio-economic system which will guarantee the reimbursement of borrowed sums irrespective of the practice of applying apparently very high interest rates (Weigel and Hem, 1984). The feeble development of institutional credit is confirmed by Satia (1988); in Sub-Saharan Africa due to the slow pace of procedures and the fact that credits are out of touch with the aquaculture cycle and badly managed, project managers there have had to take charge of the credit side of their projects and propose appropriate terms, e.g., for the promotion of pond culture in the mid-west of the Côte d’Ivoire (Oswald et al., 1988) or cage culture in the Niger (Patou et al., 1997). For satisfaction vis-à-vis the amount and terms of financing, one needs a minimum of knowledge about the informal and institutional financing systems and this is often not the case.

Generally speaking, aquaculture project documents or planners not only simplify the problem of aquaculture property rights but also impose exact copies of foreign legal notions on the societies concerned. This has occurred in French-speaking Africa where the State legally holds ownership over aquatic species. Yet, ignoring the traditional laws, the importance of which has been demonstrated (Bourgoignie, 1972; Weigel, 1985), can lead to disputes that will destroy attempts to develop aquaculture (Oswald et al., 1988).

Molnar et al. (1996) stress that a knowledge of property rights is important if one wishes to understand fish-agricultural farmers’ strategies: although not very profitable (if at all) fish farming can be undertaken as a means of appropriating marginal lands or as the best way of turning them to good use.

Before embarking upon an aquaculture development project, it would seem advisable to investigate the control over and usage of the areas of land or water in question. For the owner or user of these spaces, the decision as to whether to allot them or not is taken in line with the related opportunity costs. Prior negotiations with potential aquaculturalists as well as traditional authorities should reduce the risk of conflict.

The absence of economists in the field of African and Middle Eastern aquaculture is felt most strongly in the field of macro-economics. Broadly speaking, two features seem to be missing: a production and marketing chain approach, and an accounting for the direct and indirect effects of aquaculture development upon the local economy.

Apart from a few rare exceptions, there has until recently been a deplorable absence of real aquaculture production and marketing chains. This is in contrast to Southeast Asia where such chains have been built up through voluntarist public sector action, chiefly in three areas: (i) research-development; (ii) massive funding in the form of credit for chain operators; and (iii) industrial organization through the development of segments upstream (industrial hatcheries or feeds) or downstream (packaging and marketing).

Until recently, a focus on purely biotechnical aspects nullified the direct and indirect macro-economic effects of the development of national aquaculture schemes, as shown by the various national master plans, only a few of which have ever come to fruition.

With regard to Madagascar, one will notice the keen attention Kasprzyk et al. (1993) give to the need to introduce the missing segments of the production and marketing chain in, for example, matters concerned with research-development or aquaculture foodstuffs. With respect to Guinea, one will notice on the one hand the think-tanks on the articulation between the commercial/industrial and small-scale commercial production systems and, on the other, the analysis of the conditions for the feasibility of a feed factory during the drafting of the Guinean Master Plan for shrimp farming (Anon, 1996 b). On a broader scale, the Fisheries and Fish Farming Master Plan includes a Guinean aquatic production and marketing chain approach in its fish farming development plan (Marquet, 1996).

Another example that includes a very detailed account of the macro-economic effects of aquaculture development is provided by Tunisia’s Aquaculture Master Plan, produced under the auspices of UNDP and FAO. Indeed, it assessed the direct and indirect effects of the investments and running of various aquaculture projects proposed by the Master Plan. This methodology has been applied to each type of aquaculture: marine and inland fish farming, clam farming, mussel farming, and oyster farming. Significant aquaculture existed prior to the implementation of the Master Plan, and a comparison has been made of the situation before and after its implementation. The direct effects include the distribution of the intermediate consumption, linked to the value added of the distribution itself vis-à-vis imports and customs, and the breakdown of this value added between companies, households, public offices and foreign countries. The indirect effects include distribution of the intermediate consumption, linked to the value added upstream and downstream of the production itself, vis-à-vis imports and customs, and the distribution of this value added between companies, households, public offices and foreign countries. Finally, the effects on the balance of payments have been assessed, including the costs and earnings in foreign currencies relating to the investment and running of projects (Ministère de l’Agriculture PNUD-FAO, 1995).

Therefore, the positive experience of a few rare aquaculture master plans, as regards macro-economic information, leads one to recommend such an exercise before projects are even identified. The applications conducted by UNDP or FAO in the field of aquaculture offer a satisfactory methodological framework for taking macro-economic aspects into account.

While environmental concerns are relatively new to aquaculture in both Africa and the Middle East, environmental economics is newer still. The few authors in the field of aquaculture draw a clear distinction between the small-scale (subsistence or even commercial) extensive system and the intensive (commercial/industrial) system.

Generally speaking, small-scale aquaculture is environmentally friendly and has the potential for a low-input sustainable production system (ALCOM, 1994a,b). By way of an example, small-scale aquaculture as practiced in southern Africa has little or no deleterious impact on the biological and physical environment. However, in the long run, the aquaculture and fisheries in small water bodies can affect the conservation and management of watersheds and soils: this is one of the biggest environmental problems in Lesotho, for example. Thus, aquaculture and the management of small water bodies must be linked with watershed management, agroforestry, and soil conversion practices.

A few economic considerations appear in the work of Ayinla et al. (1994), who analyze the relationship between the increased earnings from the livestock-cum-fish farming system and the pollution and health hazards in Nigeria.

In a real piece of environmental economics work, Berg et al. (1996) attempt to build a model of the relations between the economic system and the natural aquatic system on the basis of the example of tilapia cage and pond farming in Lake Kariba (Zambia). Although intensive cage farming in the short-term perspective seems to be the only economically feasible method, semi-intensive pond farming or other less capital-intensive farming methods may be competitive if ecological and social aspects are also taken into account. The authors remind us that what is important is to avoid an aquaculture development based on a short period of prosperous growth followed by collapse with severe ecological, social and economic problems, as has been the case in shrimp farming in some areas of Southeast Asia.

The environmental impact of intensive commercial/industrial systems such as farming shrimp or seabass and seabream is entirely another matter for consideration.

As far as shrimp farming is concerned, Africa contains untouched lands that investors have managed to take advantage of, especially in Madagascar and Kenya. If current production has not yet created major environmental problems of a scale seen in some of the countries of Southeast Asia, the experiences of the latter have led planners to take environmental impact into account and they now recommend impact assessment studies prior to any aquacultural investment, as in Madagascar (Kasprzyk et al., 1996). It is with this preoccupation in mind that Fottland et al. (1996) look into ecological impact assessment via the example of the establishment of shrimp farms in the Rufiji Delta (Tanzania). The authors recommend clarification of a whole series of questions before a project can be approved: in particular, management of the natural resources in question by the local population in such a way as to adapt the project to the respective communities’ management capabilities and, equally, the cost of the externalities (pollution or diseases spread to humans) relating to the intensification of shrimp culture.

Despite interest in the subject, no in-depth studies in environmental economics have been identified in North Africa and the Middle East. Yet the challenge is nonetheless great: to reconcile aquaculture activities with tourism. Ersan (1997a, 1997b) points out the concerns of local authorities regarding pollution linked to the aquaculture of seabass and seabream in the Turkish waters of the Aegean. The Aquaculture Master Plan in Tunisia recommends clarification of the environmental impact of aquaculture installations vis-à-vis the terms of impact study requirements, the granting and refusal of permission to develop, and the list of pollutants subject to regulation (Ministère de l’Agriculture PNUD-FAO, 1995).

There are also shortages of data on marketing in general. While project documents may contain a few indicators, such as per capita consumption, the presentation of market estimates and marketing (or pricing methods) is generally rudimentary. This rudimentary presentation has frequently led to fish from aquaculture placed in the same category as other types of fish - particularly types tending to satisfy popular demand - hence masking its source. By way of an example, in Sub-Saharan Africa, fish from aquaculture cannot be compared to fish which satisfies popular demand but should be likened to the freshwater fish fetching high prices in urban or peri-urban areas. Therefore, while the strong demand for low-price pelagics in West Africa has often been put forward as the grounds to justify aquaculture projects designed to resolve supply shortfalls, the sales price for fish from aquaculture can in fact be prohibitive in the eyes of buyers with low purchasing power.

A first step toward remedying this shortage of data can be to simultaneously estimate potential markets in order to find out if there is a creditworthy demand, and to characterize the marketing channels in order to identify a sales strategy. Market estimation could be carried out on the basis of existing budget-consumption surveys: these often provide satisfactory indicators as to the state and evolution of demand. As for characterizing the marketing channels, this is done via the registering and identification of the main operators, the registering of the main dealing and processing places and, if possible, monographs on traders at the different stages of the channels (wholesalers, retailers).

A second step can be to understand the price structure via the determination of the marketing costs and margins. As each channel has its own corresponding costs and margins - depending on the relative costliness of the raw material, the frequency and means of transportation, the length of the marketing segment under consideration, and the sales method and processing means - generalizations are often inadequate. Particular attention should be devoted to the traditional channels known both for their performance insofar as they enable products to be widely distributed far from the marketing centres, and for their high consumer prices as shown by the high gross margins. Project managers have to consider the fact that re-organizing these networks with an aim to lowering costs, and therefore the retail price, for consumers would have the effect of reducing their role in redistribution.

Knowledge about trade networks and price structure is a must for determining the explanatory factors behind pricing. These factors are of both a qualitative and quantitative order, namely: the product’s physical characteristics; the supply of fish and substitutability; seasonal fluctuations; and marketing strategies. The latter deserve special attention in that monopolistic or oligopolistic situations (agreements on the part of operators in a situation of rent seeking) can have a decisive effect on marketing methods and price levels. Knowledge about pricing methods provides aquaculture project designers or planners with a helpful decision-making tool vis-à-vis marketing matters.

Many publications have reviewed the needs and nature of integrated/holistic economic information and the requirements for research on socio-economics. Since the Thematic of Aquaculture (FAO, 1987), there have been two major research interventions, by ALCOM (1994 a, b) and ICLARM and GTZ (1991), as well as numerous synthetic studies at the regional or national level that have collected and analyzed the information relating to small-scale aquaculture and research needs on socio-economics.

The purpose of the ALCOM Regional Programme has been to assist the Southern African Development Community (SDAC) member countries to investigate aspects of aquaculture development, and to test and demonstrate methods and approaches that are socially and economically viable, as well as technically feasible. This has been carried out through small pilot projects since 1986. ALCOM has produced about 60 reports and studies, as well as a regular newsletter; among these publications, some papers deal particularly with methods to collect information for planning and project design (Seki et al., 1993; Larsson, 1993; Wijkström et al., 1992; Wijkström, 1991; Wijkström et al., 1989; Wijkström et al., 1988; Anon, 1987c).

Since 1986, ICLARM has implemented a series of projects aimed at tackling the problem of how to develop, through research partnerships with national scientists and farmers, sustainable integrated agriculture-aquaculture systems on resource-poor African smallholdings. The work was centered in Ghana and Malawi. Two publications came out of workshops convened under these major projects: Prein et al. (1996), ICLARM and GTZ (1991).

The SIFR carried out between 1989 and 1991, was supported by a group of 18 multilateral and bilateral donors. The objectives of the Study were to: identify the specific constraints to fisheries and aquaculture management and development posed by the lack of information or the inaccessibility of existing knowledge; to determine high-priority research needs; to examine the capacity of developing countries to undertake research; and to propose a strategy and an action plan for improving donor support. Two publications present the results of the Study; Tropical aquaculture development: research needs and Fisheries and aquaculture research capabilities and needs in Africa (World Bank 1991, a, b). There are three major scientific concerns that sociological research, in cooperation with other disciplines, should address: (i) the understanding of the function played by aquaculture and other production activities in the allocation of production factors depending on social goals; (ii) a better understanding of production systems and their dynamics which would enable identification of ‘entry points’ for technological and organizational improvements (and not simply transfers) of farming systems; and (iii) the assessment of social impacts of external initiatives by using appropriate qualitative and quantitative tools developed by social sciences.

Due to criticism of a weak recipient-country involvement in the determination of priorities in SIFR as reported by Harrison in Martinez-Espinoza (1996), FAO (eight years after the initial studies assigned to Balarin and Deceunink), in collaboration with the Economic Commission for Africa (ECA) and the European Commission (EC), launched a regional aquaculture study in 1992 in Sub-Saharan Africa, with the objective of defining an indicative action plan for research on the basis of a series of detailed national reviews on aquaculture development and research needs. Two documents were published as a result of this regional aquaculture study; Aquaculture development and research in Sub-Saharan Africa. Synthesis of national reviews and Indicative Action Plan for Research (Coche et al., 1994) and Aquaculture development and research in Sub-Saharan Africa. National reviews (Coche (ed.), 1994). The authors highlight the lack of correspondence between identified development constraints and research priorities.

Between 1991 and 1993, the Overseas Development Administration (ODA) of the United Kingdom supported research which aimed to analyze the reasons for the supposed failure of small-scale rural aquaculture in Sub-Saharan Africa. Research was carried out in Zambia, Malawi and Kenya. Two texts refer to this research: Fish farming in Africa: what’s the catch? (Harrison et al., 1994) and Aquaculture in Africa: socio-economic dimensions (Harrison, 1994). Future research should be able to address the following issues: monitoring and assessing the content and nature of extensionists’ training from different perspectives (relevance, acceptability); assessing the fit of fish farming to other aspects of farming practices where extension for aquaculture is taking place; assessing changing attitudes as well as perceptions and priorities of the stakeholders (including the institutional ones); and monitoring the functioning of extension and the scope of farmer’s training courses.

In 1990 and 1991, the French Cooperation and Development Ministry assigned Lazard, Lecomte, Stomal and Weigel to a study of French-speaking Africa. The study, Fish farming in Sub-Saharan Africa (Lazard et al., 1991), set out to review the situation and develop guidelines for future work. The authors mainly stress the need to clarify the underlying objectives of the projects, to correctly study the market needs, and to identify the target producers; they also underlined the importance of a suitable institutional framework and a redefinition by donors of their tools and methods for the field of information and coordination. Elsewhere, Lazard highlights the past confusion between research, research-development, experimental development, and pilot project. This confusion actually showed that many projects described as ‘development’ projects had no biotechnically viable and cost-effective rearing system to propose when they were being put in place. Research was therefore often carried out as the projects came into operation (accompanying research), thus raising the crucial question as to the relationship between researchers and developers, with each being forced to leave his specialized field (Lazard and Koffi, 1996).

Taking the example of South Africa, Hecht et al. (1990, 1992) stress the importance of research support. In 1982, the Foundation for Research and Development of the Council for Scientific and Industrial Research appointed an Aquaculture Working Group, in order to assess the potential of aquaculture in South Africa, identify the research needs of the industry, prioritize the needs and formulate and coordinate an aquaculture research policy. The task of the Working Group was completed and in 1984 the guidelines for a National Aquaculture Programme were published. This provided a welcome boost to the aquaculture research momentum, stimulated growth of the industry and provided for the training of human resources. More than 70% of the present aquaculture operations came into operation soon after 1984. As has been stated by Hecht et al. (1990, p.15), "whether this was solely a consequence of the introduction of the National Aquaculture Program [sic] is obviously open to debate, but the development of technology and the dissemination of information and data certainly prompted several entrepreneurs to invest in aquaculture."

The very rapid growth of Mediterranean aquaculture explains why, in 1994, the MEDRAP II Project (FAO, UNDP) launched the Socio-Economic and Legal Aspects of Aquaculture in the Mediterranean (SELAM) network, whose aim is to establish a database for dealing with statutory and socio-economic matters, and to organize workshops, seminars and training courses. The work has chiefly concentrated on aquaculture production costs and the financial analysis of projects, banking and insurance for aquaculture, as well as aquaculture product marketing and the implementation and management of aquaculture enterprises. The countries of North Africa and the Middle East with a Mediterranean coastline therefore have at their disposal a network to focus on socio-economic and statutory aspects.