History of Aquaculture

Chapter 1
The History of Aquaculture In Traditional Societies

Malcolm C. M. Beveridge, David C. Little
Institute of Aquaculture
University of Stirling
Stirling
Scotland
U.K. FK9 4LA

The Origins of Aquaculture
The origins of aquaculture are lost in history and little evidence remains to direct even a serious investigator of the subject. There are, after all, no aquaculture-specific artifacts to guide archaeologists. There is often little to distinguish abandoned ponds, even supposing we were able to find them, from dams of various types or from systems for producing inundated arable crops such as rice. Surface water would be stored to support communities or households in many cultures and the predominant use for domestic/agricultural purposes may disguise a secondary role of holding or growing fish. Where rainfall is seasonal the focus of communities is around surface water bodies that may have originated as little more than natural depressions but became modified by the dependence these communities had on them. The presence of fish bones and shells in refuse heaps at early human settlements or representations of fish on cooking pot shards indicate only that the occupants ate these foods, not how they obtained them. The tools used in aquaculture are common to farmers and fisher folk and remains of net-like materials or hooks tell us only about how fish met their fate, not whether they were caught from a river or from a fishpond. With few exceptions there is also no genetic record of domestication to draw upon, an important distinction between fish and livestock being that fish didn’t need to be domesticated in the way livestock did in order to rear them in captivity.
In this chapter, we examine what is known about the origins and development of aquaculture among traditional societies: who practiced it, and why, what distinguishes it from industrial and post-industrial aquaculture and its legacy. We briefly consider agriculture and current theories concerning its origins and development. We draw upon examples from traditional societies in four continents - Africa, the Americas (with the exception of Hawaii, which is considered in detail in the following chapter), Asia and Europe - and from various periods in history. Unfortunately, our review is not entirely balanced as there remain enormous gaps in knowledge.
We begin with a consideration of what aquaculture is and how it differs from hunting.
Aquaculture: some definitions
It is important at the outset to be able to distinguish aquaculture from fisheries and agriculture. It may be differentiated from fishing because, as in agriculture, some measure of care or cultivation is involved. Reay’s (1979) definition of aquaculture as "Man's attempt, through inputs of labour and energy, to improve the yield of useful aquatic organisms by deliberate manipulation of their rates of growth, mortality and reproduction" appeals from a biological perspective. However, it omits the other key component that distinguishes it from hunting: the concept of ownership or the extension of access and exploitation rights. The recently revised definition of aquaculture used by the UN Food and Agricultural Organization states that aquaculture is:
‘ .. the farming of aquatic organisms including crocodiles, amphibians, finfish, molluscs, crustaceans and plants, where farming refers to their rearing to their juvenile and/or adult phase under captive conditions. Aquaculture also encompasses individual, corporate or state ownership of the organism being reared and harvested ..’
(Rana 1998).

            In this definition, both husbandry and ownership are seen as intrinsic. However, many traditional forms of aquaculture are based on the exploitation of multipurpose water bodies in which the organisms themselves are ‘common property’ i.e. neither ‘owned’ by an individual nor by some corporate body or the state. For present purposes we assume the key criteria distinguishing farming from hunting are that:

• there is some form of intervention(s) to increase yields; and
• there is either ownership of stock or there are controls on access to and benefits accruing from the interventions (for parallels in agriculture see Bromley, 1992).

            Another key point in the FAO definition: end purpose is not at issue and fish owned and reared other than for food is regarded as the products of aquaculture.
Nevertheless, as is discussed further below, differentiating between hunting and farming in the aquatic environment remains fraught with difficulties, in large measure because comparatively little effort has been expended on documenting and analyzing the range of methods used in exploitation of aquatic environments.
A further set of definitions is necessary in order that we can compare aquaculture in traditional societies with contemporary practices from an ecological standpoint: these relate to resource use, or the differences between ‘intensive’, ‘semi-intensive’ and ‘extensive’ aquaculture. According to Coche (1982), in extensive aquaculture, the aquatic animals must rely solely on available natural food, such as plankton, detritus and seston. Semi-intensive aquaculture involves either fertilization to enhance the level of natural food in the systems and/or the use of supplementary feed. Such feeds are often low-protein (generally <20%), usually compounded from locally available plants or agricultural by-products, and complement the intake of natural food which is higher in protein (Hepher, 1988). In intensive aquaculture, animals are almost exclusively reliant on an external supply of high protein food (generally >20%), generally based on fishmeal and fish oil. These definitions broadly relate to use of environmental resources - so-called ‘goods’ - although they ignore other resources such as land and water and seed. The intensity of production methods also has implications for use of environmental ‘services’; the more external food that is supplied per tonne production, the greater the wastes and the greater the demands on the environment to disperse and assimilate these wastes. However, the terminology is insufficiently well defined to be any more than a general guide. The term ‘semi-intensive aquaculture’, in particular, covers a huge diversity of aquaculture practices and ranges from minimal inputs to fairly substantial inputs of feed.
The origins of agriculture
‘People did not invent agriculture and shout for joy. They drifted or were forced into it, protesting all the way.’
Tudge (1998) Neanderthals, Bandits and Farmers
Introduction
The long accepted view of how and why agriculture began and spread has recently undergone some revision (Harris, 1996: Diamond, 1997; Tudge, 1998). While the established view remains that it started some 10 - 12 000 years ago there is a growing recognition that different peoples adopted food production at different times. While some cultures, such as the Chinese, developed agriculture independently, others learned from neighbors or colonizers while a few, such as the Aboriginal Australians, appear never to have acquired agriculture at all (Flannery, 1994). The advantages of farming are readily apparent. Much plant and animal biomass is difficult or dangerous to harvest, labor intensive to prepare or of poor nutritional value. By farming, it is possible to select and grow crops and animals that give high nutritional returns per unit expenditure thereby increasing food supplies. Agriculture also generates food surpluses and food storage, prerequisites to the development of settled, politically centralized, socially stratified, economically complex and technologically innovative societies (Diamond, 1997). However, Tudge (1998) has recently argued that this view still under-plays the importance of farming throughout much of our two million year history, especially from the late Paleolithic - some 40,000 years ago - onwards. He believes that a variety of ‘proto’-farming activities, a term coined to describe an ad hoc collection of activities that coaxed more food out of the environment, such as crop protection and game management, was part of the repertoire of responses to times when demands for wild foods outstripped supplies. Tudge also contends that when food supplies improved through upturns in abundance of game or more clement weather or death or emigration of people, they returned to what they enjoyed best: hunting. His contention is that farming was hard work and to be avoided unless absolutely necessary.
Although he makes little reference to farming prior to 12,000 years ago, Harris (1996) too believes in the concept of ‘proto-agriculture’ and has elaborated an evolutionary classification of systems of plant and animal exploitation; a simplified, unified history of agriculture (Fig. 1.1a). There is much food for thought in these ideas and we believe that many of the concepts currently being considered by agricultural historians, especially with regard to terrestrial animal production, provide insights into aquaculture, how it evolved and was practiced among traditional societies.
Aquaculture and ‘Proto-aquaculture’
Aquaculture too began in different societies, both agriculture- and fishing-based, and followed a pattern of development in many respects similar to that of agriculture. There is good evidence in aquaculture for Tudge’s theory of people opting in and out of plant and animal cultivation according to their needs, although as we hope to show these needs were not always related to food. There is also evidence of ‘proto-aquaculture’, defined here as activities designed to extract more food from aquatic environments, such as:

• transplantation of fertilized eggs;
• entrapment of fish in areas where they could thrive and be harvested as required;
• environmental enhancements, such as development of spawning areas, enhancement of food, exclusion of competitors or predators, etc.;
• holding of fish and shellfish in systems - ponds, cages, pens - until they had increased in biomass or until their value had improved.

            Each activity might on its own be considered as no more than stock enhancement and thus within the definition of what might be considered as managed fisheries. However, we apply the critical concept of control of access and benefits to draw a line between where managed fisheries end and proto-aquaculture begins. If the effect of such interventions increased supplies sufficiently to satisfy needs and to result in an equitable distribution of benefits, then it may have evolved no further. We propose the relatively small degree of control over the life cycle of the animal and the low impact of the intervention on fish or shellfish production is used to distinguish proto-aquaculture from aquaculture. The definition of proto-aquaculture is compatible with those of Coche (1982) regarding extensive, semi-intensive and intensive aquaculture and is further characterized by low consumption of energy (see above).
The proposed definitions do not neatly distinguish fishing from fish farming, but perhaps this is only to be expected when dealing with something like the transition from hunting to farming. It also ignores some of the more contentious issues, such as ranching, a term used to describe the release of juveniles into the wild, only to be recaptured later as adults. However, it is useful in helping explain how aquaculture might have first developed. It is very much a working hypothesis and others with more detailed and accurate information, more insight and more time for reflection and debate will undoubtedly construct a better framework.
Proto-aquaculture and the origins and pattern of development of aquaculture
Why might people in traditional societies have begun farming fish and shellfish? The answer, according to the agricultural historians, would be because it was necessary (Boserup, 1965). It is clear that aquaculture began in various parts of the world and at various points along the aquatic food supply line, between water and plate. The farming of fish and shellfish is by definition an activity of settled societies, originating among both fishing and wetland farming cultures as well as at points of trade. While we may surmise that conditions similar to those favouring the development of agriculture would have usually been necessary, i.e. that foraging and hunting (fishing) were insufficient to satisfy demands for fish, provision of food was not always the most important driver for the development of aquaculture.  
Stewart (1994) and others believe the importance of fish in early hunter-gatherer societies to have been under-estimated. Rudimentary proto-aquaculture techniques would probably have evolved among such societies, although evidence is scant. Native North American peoples living on the Pacific seaboard are believed to have transplanted the eggs of spawning salmon in an attempt to improve fish survival and returns.
Many proto-aquaculture activities relied on some sort of holding facility. The simplest to construct would have been earth ponds. In some parts of the world these would have been little more than mud walls constructed to temporarily hold water and fish following the seasonal flooding of a river. Such systems are still in use in some parts of the world today. The ‘whedos’ or fish holes of Benin are one such example (see Welcomme, 1972, for details). The practice of communal construction of weirs on small rivers and streams in Asia to store water outside of the monsoons principally to ensure adequate irrigation for wet rice cultivation during lulls in the rains and allow early seed bed preparation is also common. Attempts to increase fish yields would have been a logical next step, by affording protection from predators and, perhaps, by feeding fish with household scraps or farm wastes (see Fig. 1.1b).
Among fishing-based societies, a number of scenarios in which proto-fish or -shellfish farming arose is readily envisaged: the short-term storage of catches until there were sufficient fish or shellfish to make the journey to market worthwhile; the transport of live fish to market; the holding of catches until prices improved. These strategies are still seen among fisher folk today: modified traps, netted off shallow areas of lakes, cages of the sort still seen in parts of Indonesia, traditional floating cages used in the Great Lake area of Cambodia (see Beveridge, 1996).
If the theories of the agricultural historians hold true for aquaculture, then we can also expect to see aquaculture wax and wane as the result of changes in supply relative to demand for fish and shellfish.
Africa

            The earliest evidence of fish culture of sorts purportedly comes from ancient Egypt where fish often had a sacred as well as prosaic role in society. They were strongly associated with the cyclical life-giving forces of the Nile and the New Kingdom Egyptian view of the world, tilapia in particular being strongly linked to the goddess Hathor and the concept of rebirth (Desroches-Noblecourt, 1954). In his account of tilapia in ancient Egypt, Chimits (1957) reproduces a 4,000 year old bas relief figure from the tomb of Thebaine showing what appears to be an artificial, drainable pond being fished by a nobleman (Fig. 1.2). Many New Kingdom tomb scenes also show tomb owners sitting on chairs, fishing tilapias from their ponds, their wives standing behind and assisting with the catches. Although rod and line fishing is believed to have been common among all classes in Egypt at that time, the fishing activities of the nobility were limited to that of their ponds. Their interest in fishing stemmed from religious rituals associated with death and rebirth and not with pleasure or sustenance (Desroches-Noblecourt, 1954; Brewer & Friedman, 1989).
This was aquaculture at its most simple, more proto-aquaculture than aquaculture, involving little in the way of inputs or husbandry or pond management. Tilapia would have been transferred from nearby rivers to the ponds where they readily would have bred. While some food may have been provided, it is unlikely to have been important since production of fish for food was not the objective. It is believed that the practice persisted into the New Kingdom until the importance of rebirth in the world view of Egyptians waned. Although Brewer & Friedman (1989) detail many peculiar beliefs and taboos among the priesthood associated with fish, early travelers to Egypt confirm that fish was of tremendous importance in the Egyptian diet. The Roman traveler Diodorus Scullus is quoted as saying that ‘.. the Nile contains every variety of fish and in numbers beyond belief: for it supplies the native not only with fish freshly caught but also yields an unfailing multitude for salting.’ Herodotus, too, who travelled here some 2500 years ago reported that ‘.. all Egyptians in the Nile Delta possess a net with which, during the day, they fish ..’. Given the fertility of the river, the abundance of fish and the skills of the fishermen it is not surprising that these proto-aquaculture activities developed no further and indeed waned with the change in religious significance of fish. Its revival would have to wait several millennia until the early years of the present century.
Asia
China and freshwater fish farming

            Although possibly pre-dated by events in Africa, Asia - and China in particular - is widely regarded as the cradle of aquaculture. The chronology is complicated, but is summarised in Table 1.1. Many factors are thought to have constrained livestock development and predisposed China to develop aquaculture earlier than elsewhere in Asia. The wet rice agro-environment evolved relatively late in China’s history, population pressure stimulating colonization of low-lying deltas. Such conditions would have both inhibited any development of mixed farming based on ruminant livestock and crops and supported fish culture and production of livestock such as pigs and poultry that thrived on rice by-products and water-based scavenging respectively. The process of agricultural evolution in southern China, from crop-dominated to mixed farming (Grigg, 1974; Little & Edwards, 1997) was, therefore, molded by the limits and potentials of a flood-prone environment. Any diversification from a rice monoculture required a process of ‘ditching and diking’ that produced deeper areas suitable for fish and higher dike areas for horticulture. Although areas of wet rice production were relatively sparsely populated until the Han dynasty (Grigg, 1974), the adoption of increasingly intensive, and irrigated, production in suitable areas prompted a rapid increase in population and demand for aquatic products that would also have also been an important factor in stimulating aquaculture. With the development of ditch-dike systems, other crops such as beans, green vegetables and tobacco could also be grown alongside (Bray, 1984). As today, fish would have been a common component of seasonally inundating rice paddies. Given the lack of animals and the prevalence of wetlands it is thus not surprising that fish was a prominent component of the diet in many areas.
The development of irrigation in China and elsewhere was driven by the fact that rice grows best when provided with water of the right quality and in the right quantity at the right time. While there is no neat chronology of evolution of irrigation systems in China, it is clear from the widespread existence of clay models of irrigated agriculture systems recovered from graves throughout southern China that by the Han Dynasty (2300 - 1700 BP) ponds were being widely employed for water storage (Bray, 1984; Li, 1994). To some this indicates the earliest that aquaculture might have been developed. In a single grave over 18 varieties of aquatic plants and animal, that are still used by the Chinese today, were found within an intact rice-field model. These included lotus flowers, seeds and leaves, water chestnuts, soft-shelled turtles (Trionyx sinensis) grass carp (Ctenopharyngodon idella) and Carassius auratus (Guo, 1985 in Li, 1992). These areas of southern China had high densities of people culturally dependent on aquatic foods. As population densities increased, demand for fish and other aquatic foods would have increased and the practice of holding and growing fish would become increasingly attractive compared to reliance on increasingly exploited and inconsistent wild stocks. In the floodplains of China and elsewhere in Asia soil is excavated to construct elevated, better-drained areas for establishing homesteads and raising crops. Although the resultant excavations may be referred to as fishponds by aquaculturists farmers refer to them simply as ponds, an indication of their multipurpose nature
Others, however, including the Fisheries Society of China, refer to the short treatise published by the statesman Fan Li some 2500 years ago (2500 B.P.). It describes common carp (Cyprinus carpio) farming in sufficient detail to provide incontrovertible evidence that fish culture had developed well beyond a proto-aquaculture activity and that aquaculture was well-established by this time (Li, 1994). The monograph details the design and layout of fishponds, carp breeding, and fry and fingerling rearing techniques. Fan Li’s account is of ‘semi-intensive’ monoculture of carp, although there remains some debate as to the species (see Balon, 1995). The integration of fish - presumably carp - culture with that of aquatic plants and vegetables is apparent from written records dating from 2200 - 2100 B.P. while written records of rice-fish culture date from the period 1975 - 1780 B.P. (Yang, 1994).
Despite the long history of freshwater fish farming in China, there are few documentary accounts and details are fragmentary at best (see Table 1.1). According to Li (1994) fish culture expanded from rice paddies and ponds to lakes - this implies the use of cages, pens and/or enclosures - during the Han Dynasty (206 BC - 220 AD). Provided herbivorous species were used, culture could have relied on semi-intensive or even extensive methods. If, however, omnivores were farmed, then there would have been a greater reliance on supplementary feed. There is strong evidence that small dams, constructed by farmers primarily for water storage purposes, were also used to produce lotuses, water chestnuts, fish and turtles (Bray, 1984). A milestone for aquaculture in China seems to have been reached at the beginning of the Tang Dynasty in 618 AD with the culture of combinations (polyculture) of carps (Ling, 1977; Li, 1994). However, while we might expect that this would have been promoted as a means of increasing yields from semi-intensively managed systems, historians claim polyculture to have been less due to an appreciation of ecology and the synergies of growing together species with complementary feeding habits than the fact that the word for common carp in Chinese - ‘Li’ - sounded the same as the emperor’s surname. As a result, the catching selling and eating of this species was banned for the next 300 years or so. Availability of suitable wild seed was was critical to China’s aquaculture development, and the same was true for Indian carp polyculture in the subcontinent and culture of Vietnamese silver carp (Hypophthalmichthys harmandi ), mud carp (Cirrhinus molitorella) and common carp (Cyprinus carpio) in the Red River Delta, Vietnam (Chevey & Lemasson, 1937). There is also evidence that there were traditional methods of producing seed that pre-date the now widespread use of hypophysation in hatcheries. Dry ‘bundhs’, seasonal ponds that fill quickly at the time of the first rains, were used to stimulate spawning of Indian major carps in West Bengal over a hundred years ago (Sharma & Rana, 1986).
Other key events in Chinese aquaculture include the gradual integration of fish pond with various crop and livestock production systems (see Table 1.1), leading to what is widely regarded as the most complex integrated aquaculture system of all, the fishpond-dyke-mulberry system of Zhujiang, southern China (Ruddle & Zhong, 1988). The fishpond-dyke-mulberry system was strongly output orientated to meet both local demand for a variety of products - live fish, fruit, etc. - and distant markets for products such as silk. Until recent decades the yields of Chinese carp polycultures are likely to have remained low (<2 t ha-1 y-1), however, and mainly dependent on feeding low nutrient, volunteer fodder to the grass carp. This situation, as indicated by the oft-quoted Chinese proverb ‘one grass carp feeds three other fish’, reflects the paucity of, and competition for, nutrient inputs on traditional farms in southern China. The macrophagous grass carp, whose manure enhanced the productivity of food organisms for other fish species in the pond, filled a ‘niche’ usually filled by ruminants, for such low value feed (Little & Edwards, 1997, 2000). Collectivization of agriculture in the 1950s saw great changes in the types of nutrient inputs, labour and markets available and a resultant increase in complexity of Chinese integrated aquaculture.
The influence of China on aquaculture in the region has undoubtedly been great, but poorly studied. Chinese emigrants may have taken common carp Cyprinus carpio and techniques of fish farming with them first to Korea and then to Japan (Drews, 1951; Ling, 1977), from where McLarney (1984) believes the practice gradually spread to other parts of Southeast Asia. The culture of the exotic common carp with indigenous anabantids from swamps such as giant gourami (Osphronemus gouramy), kissing gourami, (Helostoma temmincki) and snakeskin gourami, (Trichogaster pectoralis) and the riverine carp, silver barb (Barbodes gonionotus) and nilem (Osteochilus hasselti) in Indonesia is belived to stem from their introduction by Chinese immigrants over two centuries ago (Edwards et al. 1997). Until recently ethnic Chinese immigrants growing fish around urban centres imported Chinese carp in well-boats and, latterly, by air. Nevertheless, as we say above, the creation of excavated areas in floodplains, the presence of high densities of people used to exploiting aquatic foodstuffs are all the conditions necessary to initiate proto-aquaculture/aquaculture. There is evidence that away from the major floodplains wild stocks of aquatic animals are more prone to over-exploitation and that the stocking of juvenile and breeding wild fish was more likely to be a response to local fish shortages in these circumstances. The stocking and management of local common carp in ricefields in mountainous northern Lao and Vietnam (Chevey & Lemasson; 1937, Little & Pham; 1999) is an example. The accumulation of land and other resources by elites may also be a prerequisite to the development of aquaculture in agricultural societies. The construction of large ponds would require mobilization and organization of labour beyond the resources of poorer households. This perhaps explains the domination of large Hindu landowners in traditional aquaculture in Bengal (Lewis et. al., 1996). There are almost certainly rich histories of aquaculture development in India, Indonesia and elsewhere. Unfortunately, these are much more poorly documented than in China (see Table 1.3 for summary chronology for Southeast Asia).
Mariculture
Documentary evidence suggests that mariculture in Asia is less than 1000 years old. The culture of milkfish (Chanos chanos) in brackish water coastal ponds is believed to have developed in Java, Indonesia some time between 1200 and 1400 AD under the influence of the Hindu rulers (Schuster, 1952). Convicts were apparently sent to the coastal fringes to work in the salt marshes and to guard coastal fires. They were forbidden to practice agriculture or trade and their conditions were so harsh that they are believed to have survived largely through damming of creeks and trapping of fish and crustaceans. Schuster (1952) believes coastal pond fish farming evolved among these ‘untouchable’ fish-eating peoples from the seasonal flooding of salt ponds during the wet season.
Mariculture developed independently in Japan during the Tokugawa era (1603-1868) (Ling, 1977; McLarney, 1984). Some traditional forms of coastal mariculture, such as the use of barachois in Mauritius, are still practiced to a limited extent. Here, shallow lagoons, isolated from the sea by stone walls fitted with screens, are stocked with mullet (Mugil cephalus) or rabbit fish (Siganus sp.) fingerlings. Some oyster farming may also be practiced.
Europe
Introduction
Historical evidence suggests that mariculture in the Mediterranean pre-dates that in Asia by a millennium or so. The Etruscans practiced active management of coastal lagoons along the Adriatic and Tyrrhenian coasts in the 4th and 5th centuries BC. These proto-aquaculture activities evolved into what is generally known today as ‘vallicoltura’, involving the installation of permanent or semi-permanent embankments to enclose the lagoons. Sluices were incorporated within the structures in order to trap fish inside the shallow, productive lagoon environment where they could be harvested when desired. This form of mariculture is still practiced to a limited extent around the Northern Adriatic, Tyrrhenian Sea, Sicily and Sardinia (Kirk, 1987; Ardizzone et al., 1988). Shellfish farming, involving the and relaying of oysters, was well-established in the Adriatic 2100 to 2200 years before present (Balon, 1967b, in Balon 1995), but probably pre-dates this by many hundreds of years.
The Romans are documented as being among the first to build coastal aquaculture ponds, most likely before the end of the second century BC (2200 B.P.). In his Natural History (Vol IX) written some hundred years later, Pliny the elder notes that a certain Lucinus Murena, had excavated substantial areas of fish ponds (‘piscinae’) at Grotta Ferraia, close to the summer residence of Cicero. This idea was apparently adopted by other members of the nobility, partly for the purpose of holding live food fish, but also as a demonstration of wealth and status. Pliny the elder also recounts that:
‘.. at Baculo in Baiae Disrict, the pleader Hortensius had a fishpond containing a lamprey that he fell so deeply in love with that he is believed to have wept when it expired. At the same country house Drusus' wife Antonia adorned her favourite lamprey with earrings, and its reputation made some people extremely eager to visit Baculo’.

            Aquaculture perhaps, but scarcely for food! However, while these ‘saltae’ or ‘maritimae’ were built by the rich nobility primarily for spectacle, a practice that was widespread and persisted for several centuries, common people also built ponds (‘dulces’) for food production for income generation (Kirk, 1987; Zeepvat, 1988). Both freshwater and saltwater ponds were built, the former being stocked with a variety of coarse fish and salmonids, the latter with eel, mullets, turbot and sea bass. According to Zeepvat, classical Roman literature gives the impression that the keeping of fish in artificial ponds was commonplace throughout the Mediterranean provinces of the Empire.
Continental Europe
The common carp (Cyprinus carpio) was undoubtedly key to the development of aquaculture in continental Europe over much of the past two millennia. During the 1st and 2nd centuries A.D Roman attempts to culture local fishes in piscinae on a significant scale to increase supplies of fresh fish failed and as a result the common carp from the Danube were imported (Balon, 1995). Gradually the practice of common carp culture developed by the Romans spread westwards, although it did not reach England until the late 14th century (Currie, 1991).
The monasteries, the repository of knowledge throughout the Middle Ages, were undoubtedly important in the development of fish - especially carp - culture among the farming communities of central and Western Europe. There are several reasons for this. First, fish played an important symbolic role in early Christian motifs, fish representing Christian souls, and ponds of fish are common in mosaics decorating early Byzantine churches (Drewer, 1981). Second, a number of the Christian monastic orders, such as the Cistercians, were skilled agriculturists. The Cistercians also led in the development of fulling mills in which the processing of large quantities of wool took place. The requirements for waterwheels to drive the heavy timber and stone hammers that pounded the wool to clean and thicken it, in the mid-14th century led to the construction of dams that were usually stocked with fish (Binnie, 1987). According to Binnie (1987), the Cistercians and other religious Orders influenced the populace at large to adopt the practice of eating fish on Fridays and during Lent. The growing importance of fish and the preference for fresh rather than salted led to not only the royal and ecclesiastical establishments but also wealthy landowners to construct ‘stew’ ponds.
No single authoritative history of fish culture in Europe, however, has yet been produced, the principal reason being the lack of study. Hoffman (1996), however, has marshaled current knowledge and understanding to give an excellent preliminary synthesis of pond fish culture. Hoffman associates the development of pond fish culture with environmental degradation and population growth that led to growing demand for freshwater fish. By the reign of Charlemagne (768-814 AD) carp culture was sufficiently important for the emperor to have issued edicts to tenant farmers regarding the maintenance and management of ponds, protection from poaching, the regulation of fishing and the sale of fish (Balon, 1995). By Medieval times, fish had become an important staple in the diet (Hoffmann, 1996). Initially, fishponds seemed to have been built in Europe to ensure supplies of fish for the owner(s) and his/their dependents, especially during Lent when eating of meat was prohibited. Small fish were returned to the ponds to ensure continuity of production. Any surpluses tended to be used as gifts to cement social relations or for sale, although as Hoffmann (1995) states, medieval fishponds were not managed to cater for what were undoubtedly considerable markets for fish for stocking and for consumption. A number of the medieval power dams, such as the Cistercian-owned Lucelle dam in northwestern Switzerland and the Bruna dam, Sienna, Italy, were also concurrently used for fish culture Schnitter (1994). The latter was owned by the State that had a certain responsibility to ensure there was sufficient food for its populace and woe betide any that failed in this (Adams, 1984; Hoffmann 1996).
Economic growth during the 14th-16th centuries promoted widespread development of fishponds throughout much of continental Europe (Adams, 1984; Schnitter, 1994) although for reasons given below not in England. Carp farming was very extensive and often practiced in modified floodplains difficult to exploit profitably except for pasture. In Hungary, for example, ponds were often located in areas of poor, unproductive soils that were likely to flood. The ponds ensured a continuous supply of food, particularly during times of war, feeding both rich and poor. In the Czech Republic, in the late 14th century there were 75,000 ha of ponds although production was very low, yielding 2,250 tonnes of fish (Adamek & Kouril, 2000). Bohemia appears to have had the greatest concentration of ponds. During the 15th and 16th centuries, some 700 ponds, totaling an area of 1800 km2, were built (Votruba, 1987). Ponds also flourished in the district of La Dombes, between Lyon and Bourg, France, for a period of several hundred years from the early Middle Ages, eventually covering an area of some 180 km2 (Sahrhage & Lundbeck, 1992).
From the late 16th century onwards, pond fish farming in Germany and France fell into decline and did not begin to recover until the development of trout culture in the mid-19th century. The causes were various. In Bohemia, they have been ascribed to over-production coupled with the pauperisation of farmers that culminated in the Peasants’ Wars of 1524 (Sahrhage & Lundbeck, 1992), while in the Czech Republic frequent wars and the development of agriculture were blamed (Adamek & Kouril, 2000). In France, the secularization of the church by Napoleon in 1803 is accorded some of the blame (Sahrhage & Lundbeck, 1992). The main reasons, however, were economic. Just as it had once proved highly profitable to convert farmland into fishponds, so the fortunes of fish culture turned full circle: increased profitability of other crops, coupled with State incentives to convert ponds into farmland, and a contracting market finished carp farming in many areas. The decline in the consumption of fish, a commodity always more expensive than meat, had begun during the 17th century, first among the nobility and later among the middle classes.
England
As in continental Europe, the Romans must be credited with the introduction of fish culture to England. At the time Zeepvat (1988) wrote his review, some dozen ponds, ranging in size from 30 m2 to 1770 m2 had been uncovered at Roman villas in southern England and there are undoubtedly more to be discovered. The earliest of the ponds dates from 65 AD, the latest, at Bancroft, Buckinghamshire, from the first half of the fourth century AD. Many of the ponds were primarily ornamental in nature, while others were principally designed for bathing. However, many served a secondary purpose of production and storage of fish for the table and a few even are surmised to be primarily for food production: Zeepvat (1988) presents data from an excavation at Shakenoak, Oxfordshire, dating from the second century BC. Here a complex of ponds was uncovered in the 1970s, all of which seemed to have been purpose-built for fish production. One pond is surmised to have been a breeding pond, another a production pond, while a third is believed to have been used as a holding tank for fish awaiting shipment to market. The villas and their ponds were abandoned when the Romans left Britain in the 5th Century AD.
According to many historians, the greatest consumption of fish in Medieval England was by the religious communities such as the Cistercians, in whose houses ‘meat and lard were never eaten … save for those seriously ill and by the hired workmen. The diet was therefore confined to bread, vegetables and dishes mainly from fish, flour, eggs and cheese, with honey as an exception. Throughout Lent and Advent and on certain vigils, eggs and cheese were banned, as were all kinds of imported spices’ (Knowles 1950). However, recent research by Currie (1989, 1991), has cast doubts on the traditional view that the monastic orders were primarily responsible for the development and dissemination of pond fish culture in England. There is much evidence to show that the ascendant Norman aristocracy built the first large-scale fish pond developments, primarily in order to cement their new-found status as landowners. There are records of fishponds from several areas of Norman England, dating back to the end of the 11th century (Roberts, 1966; Stearne, 1970-71; McDonnell, 1981; Roberts, 1985; Binnie, 1987; Currie, 1991). In 1069 William the Conqueror, the French nobleman who had subjugated England only a few years before, ordered the River Foss at York to be dammed just above its confluence with the Ouse in order to protect his newly built castle, leading, incidentally, to the creation of a fishpond. Many more fishponds were built in England over the next two and a half centuries. Indeed, Roberts (1966) claims that ‘.. fishponds were normal appurtenances to manors over much of lowland England during the middle ages ..’ and that ‘.. few monastic establishments were without a supply of freshwater fish, frequently kept in ponds..’ while ‘ .. fish production [was] on some scale from royal ponds during the thirteenth century’. By the late Middle Ages, there were at least 135 pond systems in Oxfordshire alone (Chambers & Gray, 1988).
Other accounts highlight the role of royalty in pond fish production. Henry III (1216-72), for example, made numerous well-documented gifts of live fish to his nobles, promoting the development of fishponds on the estates of the wealthy (McDonnell, 1981; Binnie, 1987). The ponds owned by the Cistercians, Augustinians and Gilbertines and other monastic orders were either granted by wealthy secular patrons in association with gifts of land or were built somewhat later during the medieval period and on a smaller scale.
As mentioned above, fish were important in the early Christian church in both symbolic and practical terms. More than 100 fasting days per year were introduced, during which only cold-blooded animals such as fish and crustaceans could be eaten and even more elaborate rules regarding consumption of fish were followed by monastic orders (see Knowles, 1950). However, the idea that monastic fishponds provided much-needed fresh fish for the monastic orders during fasting days is rejected by Dyer (1988), Currie (1989) and others. They assert that cheap salted sea fish was eaten during periods of penance and cite contemporary records stating that fresh pond fish were to be kept for special occasions. The populace at large was rarely able to eat meat or fresh fish. Dyer (in Currie, 1991) states that fishponds were seen as bastions of privilege and that peasant discontent, such as peasants’ revolt of 1381, in which the ‘stews of Southwark’ were wrecked, were often targeted.
The royal fishponds fell into decline during the period 1300-1480 due to the decline of the number of royal houses, the costs of pond maintenance and the relatively poor economic returns and because of the rising numbers of retail fishmongers (Steane, 1988). With the dissolution of the monasteries in the 16th century, many monastic fishponds were abandoned, never to be productive again. In the latter part of the medieval period, a few wealthy freemen began to build ponds (Roberts, 1966; McDonnell, 1981; Currie, 1991). There is evidence that wealthy peasants owned ponds in the Forest of Arden, Warwickshire, for example, and that part of the produce was sold. Currie (1991) also cites evidence of fishmongers in mid -14th century London feeding and keeping fish in ponds for sale. However, the practice could never be described as widespread. In post-medieval times, the keeping of common carp in ponds became popular for a while among land-owning nobility, the fish being used for both domestic consumption and sale (Currie, 1991). However, Taverner (1600), in his discourse on fish culture, mentions that fishpond keeping was no longer as popular as it once had been. In leasing estates to tenants, fishponds were often neglected. There was a decline in eating freshwater fish such as carp, bream and perch during the 17th and 18th centuries. According to Chambers & Gray (1988), the 1869 edition of the redoubtable Mrs Beeton’s Book of Household Management states that freshwater fish are seldom purchased.
The Americas
In pre-Hispanic Mexico the chinampas that developed and flourished in lakes Xochimilco and Texcoco in the Valley of Mexico under the Aztecs, represent only one of several integrated wetland agriculture/aquaculture systems (Coe, 1964). The term ‘chinampa’ derives from the Nahuatl language and means ‘net of branches’, The chinampas were created by cutting and piling turfs onto the lake bottom around lake margins to form a patchwork of peninsulas and islands, each typically 100 m long by 5 - 10 m wide, interspersed by a grid-like pattern of canals. The key feature of the chinampas is that lake water could infiltrate the agricultural beds, maintaining a supply of water to the roots of the plants throughout the year. A wide range of crops was grown, including maize and tomatoes. Organically rich mud recovered from the canals would be piled onto the chinampas, thus maintaining fertility (Armillas, 1971).
Although there have been many changes and much has been lost, the systems still exist and remain ‘the biologically richest agro-ecosystem … in which most of the flora is managed and used’ (Jiménez-Osornio & Gomez-Pompa, 1991). Fish play no part in the present-day system, and although it is believed that they once did it is difficult to say how significant or to what extent fish production was deliberately enhanced. It is thought that fish were purposely trapped in the canals where they could be readily be harvested as required and that crop and vegetable by-products may have been used to enhance fish production. There were attempts to experiment with a more highly integrated form of chinampa-based aquaculture in Tabasco, Mexico, and in the Akanyaru marshlands of Rwanda in the 1980s, with mixed success (Micha & Chavez, 1997).
Fish farming methods
Introduction
Almost without exception, pre-20th century fish farming was of herbivorous/omnivorous species such as carps and roach, milkfish, mullets and tilapias. Culture of piscivorous pike, as was practiced in Medieval European fishponds, was done in polyculture systems where the pike undoubtedly fed on smaller stocked carps and other coarse fishes. The monoculture of carnivorous salmonids was initiated in the 19th century as a response to meeting the growing demand of recreational fishing and to environmental degradation on the north Pacific coast of the Americas and was limited to the production of juvenile stages for stocking. The culture of aquatic carnivores for the table had to wait until after the 2nd World War.
Aquaculture for religious or for social status reasons aside, proto-aquaculture and aquaculture were about increasing fish and shellfish production. The fundamental principles involved can be derived from the figure of Pitcher & Hart (1982) for the factors that determine exploitable stock biomass (Fig. 1.3). To increase the harvestable yields from a stock, it is necessary to increase stock biomass (recruitment and growth) while minimizing losses through disease or predation. Stocking, increases in food supply, either through fertilization or provision of additional food, concern for animal welfare and exclusion of predators may all be employed in fish farming, although profitability is usually the ultimate criterion. Here we examine what was known about aquaculture methods among practitioners in traditional societies.
Broodstock management and spawning
Early fish farming relied either on trapping or transplantation of wild fry (carp, milkfish, mullet, etc.) or on breeding and production of fish such as tilapia, roach, etc. within the production system. Accounts of catching and transportation of fry for stocking of fishponds are apparent from two written works dating from the early thirteenth century (Song Dynasty) in China (Li, 1994).
How well did early fish farmers understand the life cycle and principles of farming fish? It is clear from the he treatise on fish culture written by Fan Li, some 2500 B.P., that the principles of carp breeding were well-appreciated by that time in China (Li, 1994). McDonnell (1981) believes that in England the basic concepts behind holding fish in separate enclosures for store (‘stews’) and for breeding would have been familiar to Medieval Cistercian sheep farmers, implying that in many of the larger fishpond systems there were separate breeding/fry rearing ponds and production ponds. Hoffman (1995) also cites evidence of separate carp fry rearing ponds or ditches associated with the ducal fishpond complex at Laperrière-sur-Saône, Burgundy, in the mid-fourtneenth century. Most of the fish cultured in ponds in England at this time - bream (Abramis brama), dace (Leuciscus leuciscus), perch (Perca fluviatilis), roach (Rutilus rutilus), tench (Tinca tinca) and pike (Esox lucius) - (McDonnell, 1981) would have readily acclimatized to the muddy, stagnant pond environment and provided there were a few aquatic plants present, such fish would have happily bred. During harvest of Medieval European fishponds, smaller fish were usually retained for restocking purposes (see Hoffman, 1995, for example). However, there is also extensive documentary evidence showing considerable and well organized movements of fish for stocking purposes (Roberts 1966, McDonnell 1981). Over a twenty-year period, for example, parcels of fish, typically numbering 45 bream, 30 pike and larger numbers of smaller fish, were sent as gifts from the Royal Foss pond in York to stock the ponds of the Archbishop of York, the Fountains and Bylund Abbeys in Yorkshire and, further afield, ponds belonging to the Earls of Cornwall and Lincoln (McDonnell, 1981; Binnie, 1987).
The first carps kept in ponds in Roman and medieval continental Europe would have been wild Danube forms. Little is known about how these fish were propagated or whether any selection took place. Although the first documentary accounts of common carp breeding in ponds did not appear until the 13th century, Balon (1995) asserts that breeding in captivity had occurred as early as in Roman times in Europe, albeit that spawning would probably have been a rather hit-or-miss affair. By the mid-late 16th century articles and even scientific investigations of carp production were being published. It is apparent from around this time that domesticated strains that differed in colour, shape and scale pattern from wild forms, were being produced. Balon (1995) believes domestication of the goldfish (Carassius auratus) in China preceded that of the common carp Cyprinus carpio in Europe, probably some decades earlier in the 16th century. Balon cites Chen (1952) in asserting that domestication in China resulted from the Buddhist belief that freeing a captive animal was an act of self-purification, and that the freeing of a rare, naturally occurring mutant form, was a greater deed. ‘Ponds of mercy’ were established at temples and the rich began to culture fish in jade jars, leading to inbreeding of aberrant forms - trailing fins, bulging eyes and scale and pigment: a perverse form of selective breeding, in other words. The culture of koi carp in Japan some two hundred years later may or may not have arisen from this practice (see Balon, 1995, for review). Today, the release of captive aquatic animals, especially fish and turtles, is still an everyday practice in many parts of Southeast Asia as is the husbanding of wild stocks in and around temple grounds. Protection and feeding fish in rivers and canals proximal to Buddhist temples and within temple ponds to make merit is widespread in Thailand, Laos and Cambodia. Such refuges may be an important part of conservation of wild stocks is areas with heavy exploitation and extensive water control.
Pond construction and management
Much effort and expenditure often went into the construction and maintenance of ponds, as is particularly evident from European records. Medieval European fishponds in varied in size from a few hundred m2 to more than 80 ha and depths could be as much as 6 -8 m. Most fishponds were formed by the damming of a stream in a valley, the dam wall being constructed from earth and stone and strengthened with wood. Early ponds tended to exploit the contours of the land; ponds built from the 16th century onwards were more likely to be linear in conformation, thereby facilitating netting (Chambers & Gray, 1988). Often the ponds were multipurpose; however, purpose-built fishponds were also constructed in close association with a millpond or moat surrounding a manor house or large farmhouse. Few of the ponds built in England during the early medieval period could be drained without creating a breach in the wall and a simple overflow was usually all that was present to ensure that ponds did not burst their banks. Many of the ponds built in France, however, had deep drainage ditches fitted with wooden sluices that could be lifted, thereby allowing complete drainage of the ponds (Hoffmann, 1995).
Once built, ponds had to be maintained. Binnie (1987) asserts that silting was not a problem in medieval English fishponds and that pond catchments deliberately may not have been ploughed (i.e. used for crop culture) but instead were used for animal grazing. However, there are many contemporary written accounts of periodic de-silting. A detailed account at one such medieval pond is given by Roberts (1985). The Alresford fishpond, built around 1200, was an enormous structure (a ‘magnum vivarium’), covering some 80 ha. In the winter of 1252-53, after fish had first been transferred to a nearby stew pond, a breach was deliberately made in the 70 m long, 6 m high weir in order to allow the pond to drain. The following winter 122 labourers using 24 carts were employed to clean out the silt, the breach mended and the pond allowed to refill and then restocked with 1072 roach, 603 bream, 229 perch and 115 pike. The factors that most often precipitated pond draining were deteriorations in stock ‘through disease or fouling of ponds’ (Binnie, 1987). Although breaching of pond walls and de-silting was expensive, especially with such large ponds, the removal of silt was seen as essential in maintaining pond volume and the practice of allowing pond soils to dry out seems to have been long recognised as important in maintaining productivity (Binnie, 1987). It is known that during dry fallowing periods, mineralisation of sediment-bound nutrients occurs, resulting in renewed productivity once ponds are refilled. Such recycling of nutrients is further enhanced if grazing animals are allowed to within these temporal pastures. There was also a tradition of inter-cropping in some parts of Europe. Taverner (1600) advised crops should be grown and cattle herded in fishponds during alternate dry years. In Hoffmann's account of the management of fishponds at Laperrière-sur-Saône, Burgundy, during the middle of the fourteenth century (Hoffmann, 1995) draining and harvesting of ponds were often finished in time for planting of summer crops. During 1339-40 and 1350-51, extensive and costly repairs necessitated that the dry pond beds be farmed for a year, yielding crops of barley, oats and hemp. In central Europe the practice evolved of rotations of 3 years of crops being followed by three fish crops was widely practiced (Wunder, 1949, in Hickling, 1962).
Feeding practices
Technical literature related to feeding of fish is scant. Rumsey (1994) quotes mention by John Halver and Peter Milne and others of illustrations dating from more than three millennia of people feeding fish in impoundments. However, as still seen in some of the more extensive forms of aquaculture today, it is likely that the majority of the diet of fish grown in ponds and rice paddies would have been natural, principally plankton and benthic invertebrates and detritus. Natural productivity alone would have been able to sustain production of several hundred kilograms per hectare from monoculture. Production from polyculture, where synergistic feeding relationships are exploited, is likely to have been somewhat higher, although probably no more than 600-700 kg ha-1 y-1, depending on the nature of water body, proximity to human habitation and how nutrients drain into water bodies.
Manuring has been widely used for centuries in Asia. In England too, Chambers & Gray (1988) cite evidence of human effluent being deliberately channeled in monastery fishponds during the later medieval period. Feeding was certainly known to the Chinese and to the Romans. However, Hickling (1971) and Currie (1991) believe that it would have been beneath the upper classes that owned fishponds in 11th-13th century England to try to increase revenues through feeding. But this was certainly not the case with the aspiring lower orders who owned or leased fishponds for commerce and who did feed their fish. In many systems fish would also have been fed, most likely on wastes from the table and agricultural by-products and surpluses. Such practices would have been carried out on an experimental basis initially, but would have become accepted practice as yields were observed to increase. The integration of fish culture with agriculture in the pond-dyke system in southern China would have encouraged the use of such surpluses and wastes as fish food. Supplementary feeding would almost certainly have been necessary for fish reared in cages, there being insufficient plankton or seston available to sustain growth under most circumstances (see Beveridge, 1996, for review).
With the development of salmonid farming in the 19th century, the use of animal protein such as oysters and horseflesh in diets became necessary (Maitland, 1887). Rumsey (1994) quotes anecdotal mention of fish feeds in books by Garlick, Norris and Stone published in North America during the latter half of the 19th century, although he says little detail is provided. Fish diet development started in earnest in the 1920s (Ellis, 1994; Rumsey, 1994). In Germany Schaeperclaus (1933) published a monograph on pond fish culture in which he discusses the inclusion of marine fish, meat and offal from warm-blooded animals, fish and shrimp flours and waste meals from oil extraction of soya, sunflower, etc., in trout diets. He also recommends the inclusion of 25% of beech or poplar sawdust as binder!
Discussion
Despite the lack of serious archaeological or historical research there is sufficient evidence to conclude that the origins of aquaculture date back several - possibly as many as five - millennia. Early aquaculture - proto-aquaculture as we have termed it here - would probably have involved some simple form of intervention, such as transplantation of fertilized eggs or spawning fish or the enhancement of spawning grounds to increase production. Such actions would require that a claim to ownership or the control of access to and exploitation rights of stocks were established. A parallel development may have been the use of feed or shelter, probably first to attract fish in natural ecosystems and make them easier to exploit, before these aspects were incorporated into more formal culture systems. These forms of proto-aquaculture have parallels with the development of  animal husbandry (see Fig. 1.1 and  Table 1.2). The concept would have been readily understood, and perhaps informed from experience with livestock, and if inappropriate food were given it would have acted as a fertilizer, stimulating the productivity of the food web. As with agriculture, aquaculture would have to wait until the 20th century for the arrival of inorganic fertilizers, but scientific inquiry into their use, and widespread adoption by fish farmers lags their use in terrestrial systems by several decades. Evidence suggests that captive breeding of fish and the management of ponds were later developments, as indicated in the aquaculture development scenario illustrated in Fig. 1.1b. Without control over propagation, selection for improved growth and disease resistance, let alone shape, colour, flesh quality or domestication, such a significant component of agricultural husbandry, could not occur. In fact, the opposite may have happened. Although in certain contexts there would have been a large degree of passive selection for response to cultured environments, the practice of harvesting larger fish, leaving the smaller, slower-growing fish to serve as broodstock was, and still is, commonplace (see Silliman, 1975, for discussion). Removing the largest individual appears to be a common practice in the history of aquaculture the world over, perhaps illustrating residual close links to ‘hunting’ and the role of raising fish in building social capital. That the fish thrived and grew often owed more to chance than a profound understanding of satisfying the animals’ needs; that had to wait until the development of the biological sciences in the 19th century.
Proto-aquaculture activities have been recorded in many parts of the ancient world, including Egypt, China and Mexico. Why did aquaculture develop in some places rather than others and what governed the methods of culture and the pattern of development? Fish has almost always been universally important in the Asian diet. Although extensive China’s coastline was small relative to the inland land mass. With its wetland-based agriculture and, from the Han Dynasty onwards, its high densities of aquatic food-eating people it had strong incentives to develop inland aquaculture and it is here that such advanced aquaculture methods as controlled spawning of fish, polyculture and integrated aquaculture first developed. By contrast, until recently in Japan and other countries with extensive coastlines, marine fish and shellfish resources were sufficient to supply their populations’ needs. It should also be reiterated that religion, politics and power and social issues were sometimes more important in aquaculture development than increasing the supply of food or economic viability. In central and western Europe, for example, fish farming was primarily developed in the first instance by the monastic orders in order to ensure supplies of fish for days when eating of meat was forbidden. Subsequent pond aquaculture developments in continental Europe occurred in areas where soils were too poor to sustain agriculture (Hickling, 1962). In medieval England pond fish culture was also used by the arriviste Norman rulers to help consolidate political power.
There are many parallels between the development of agriculture and aquaculture, with people opting in and out as the need for fish/food changed. However, unlike agriculture which has been the most important way of obtaining food on land for at least 10 000 years, aquaculture has until recently contributed little in real terms to world fish or shellfish production. Instead of evolving towards cultivation, hunter‑gatherer methods of procuring food from the aquatic environment developed along a different path; by improvements in tracking methods and by increases in killing power. There are several reasons why the exploitation of food in terrestrial and aquatic environments did not develop in the same way. First, food in inland waters and in the sea has, until recently, been abundant. Changes in fishing grounds, increases in fishing pressure and development of fisheries technology were sufficient to meet demands and there was therefore little incentive to learn to farm. Moreover, the aquatic environment was hostile and something to be feared. It must have seemed impossible that a structure that could hold fish securely and withstand the forces of the tides and currents, waves and storms, could be built in the sea. There were other technical problems, too, to overcome. While the breeding and husbandry of animals and the harvesting and planting of seeds was readily achieved on land, it has proved difficult to breed many aquatic species and to hatch the eggs and successfully rear the offspring. These problems in part stemmed from the fact that people were dealing with organisms that were very different from themselves and with an environment about which they were largely ignorant. It was not until the rise of the biological sciences in the 19th century that the mysteries surrounding the physiology and reproduction of aquatic animals and the role the environment played in controlling these processes began to be understood. These constraints were greater for coastal than inland environments, which may explain the earlier development of the latter. The environments for fish culture to develop were to some extent a consequence of the overall development of intensified flood plain agriculture in Asia.
Like agriculture and fishing, aquaculture was shaped by changes in society - wars and colonialism, politics and power, the waxing and waning of religions with their associated cults and schisms, the rise of science and industrialization (Tables 1.2, 1.3). However, it is not solely concerned with the production of food but has also played a role in a number of religions and in consolidating social position. The production of fish for religious purposes in ancient Egypt, Christian Europe and Buddhist Japan has contributed much to the development of aquaculture as a whole, especially with regard to selective breeding. Certain ironies are also apparent. Unlike hunting whose image is associated with nobility fishing is often associated with poor people of low social standing (Sahrhage & Lundbeck 1992). The construction of fishponds and the keeping of fish as in the Roman Empire and medieval England have long been associated with social position and would have required significant resources unavailable to ordinary people. Where demands for fish for religious or social purposes declined or the economics of production changed, as in ancient Egypt and in 18th century France fish farming withered and died.
Despite its ancient roots, however, fish and shellfish farming should be regarded as very much a post-second world war phenomenon. Although locally important, in global terms farmed aquatic production prior to 1950 was insignificant. With the possible exception of China, few of today's important industries, particularly where mariculture is concerned, owe much to ancient traditions or technologies or have a history extending back more than 30 or 40 years. Even in China until recently farmed production was but a fraction of that from capture fisheries. Data assembled by Li (1994) and the FAO (Lu, 1992; Immink, 1997) show that the spectacular - approximately 50-fold - growth in cultured freshwater fish production has occurred since the founding of the People’s Republic in 1949 as a result of political, economic, technical and demographic changes. Some 82,000 man-made water bodies have been constructed for hydropower, flood control and irrigation purposes, increasing the inland water surface area by 2.05 million ha (Lu 1992). In the late 1950s the techniques for spawning Chinese carps were developed, freeing Chinese farmers from their reliance on wild-caught fry while the development of the mixed economy some 20 years later have given further incentives to increase productivity. Elsewhere in Asia aquaculture was rare. Until recent decades. Huet’s (1972) generalisation that ‘In the Far East where all farmers are fish farmers and vice versa ..’ was based on observations of a few clusters of activity around Bangkok and in West Java (Edwards et al., 1997). Edwards et al. estimate that probably far fewer than 10% of small-scale farms in the region culture fish and of those that do, most produce fewer fish than their resource base allows.
And what is the legacy of traditional aquaculture? Much present day mariculture, particularly the intensive salmon and shrimp farming industries that have grown so rapidly during the past two decades, has a basis in industry. It has been technology-led and is heavily dependent upon the environment not only for the supply of essential resources such as food, but also to disperse and assimilate waste food, faeces, excreta and chemical residues (Beveridge et al. 1997). In parts of China traditional aquaculture practices are being abandoned in favour of intensification. Land prices are too high, water is in short supply, cheap feeds are becoming ever more easily available and there is a strong demand for fish from the increasingly affluent and urbanized population. However, for large areas of rural Asia, and elsewhere in the developing world, resource-intensive aquaculture is not, and is unlikely to be, a viable option. Throughout Asia, the stigma associated with use of night soil in ponds has led to a decline in its use, compounded by increased mixing of human and industrial waste in urban sewage systems (P. Edwards, pers comm.). Trends to separate aquaculture from broader farming systems are largely commercially driven but are in conflict with longer term, and rational, practice. Just as industrial livestock production is now being challenged in Western Europe on public health, animal welfare and, most immediately, environmental grounds, we expect a similar fate for unsustainable aquaculture development.
Acknowledgments
Colleagues James Muir, David Penman and Krishen Rana kindly helped with information and references. Last, but not least, we thank Dr. Barry Costa-Pierce for his knowledge and enthusiasm on the subject of the ecology of aquaculture systems.
References
Adamek, Z. & Kouril, J. (2000) A long aquaculture tradition in the Czech Republic. Aquaculture Europe, 25 (1), 20-23.
Adams, N. (1984) Architecture for fish: the Sienese Dam on the Bruna River - structures and designs, 1468 - ca. 1530. Technology and Culture, 1984, 768-797.
Ardizzone, G. D., Cataudella, S. & Rossi, R. (1988) Management of Coastal Lagoon Fisheries and Aquaculture in Italy. FAO Fisheries Technical Paper, 293. FAO, Rome.
Armillas, P. (1971) Gardens on swamps. Science, 174, 653-661.
Balon, E. K. (1995) Origin and domestication of the wild carp Cyprinus carpio: from Roman gourmets to the swimming flowers. Aquaculture, 129, 3-49.
Beveridge, M. C. M. (1996) Cage Aquauclture. 2nd edn. Blackwell Science Ltd, Oxford.
Beveridge M. C. M., Phillips, M. J. & Macintosh, D. C. (1997) Aquaculture and the environment: the supply and demand for environmental goods and services by Asian aquaculture and the implications for sustainability. Aquaculture Research, 28, 101-111.
Binnie, G. M. (1987) Early Dam Builders in Britain. Thomas Telford, London.
Boserup, E. (1965) The Conditions of Agricultural Growth. Allen & Unwin, London.
Bray, F. (1984) Biology and Biological Technology. Part II: Agriculture. Science and Civilisation in China. Volume VI, (Ed. By J. Needham). Cambridge University Press, Cambridge.
Brewer, D. J. & Friedman, R. F. (1989). Fish and Fishing in Ancient Egypt. Aris & Phillips, Warminster, U.K.
Bromley, D. W. (1992) Making the Commons Work. Theory, Practice and Policy. ICS Press, San Francisco.
Chambers, R. A. & Gray, M. (1988). The excavation of fishponds. In: Medieval Fish, Fisheries and Fishponds in England, (Ed. by M. Aston), pp. 113-128. British Archaeological Reports, British Series, 182 (i).
Chevey, P. & Lemasson, J. (1937) Contribution à l’Étude des Poissons des Eaux Douces Tonkinoises. Institut Océanographic de I’Indochine. 33e Note. Gouvernment Général de L’Indochine, Hanoi.
Chimits, P. (1957) Tilapia in ancient Egypt. FAO Fisheries Bulletin, 10, 211-215.
Coche, A. G. (1982) Cage Culture of Tilapias. In Biology and Culture of Tilapias (Ed. by R. S .V. Pullin & R. H. Lowe-McConnell), pp. 205-246. ICLARM, Manila, Philippines.
Coe, M. D. (1964) The chinampas of Mexico. Scientific American, 211, 90-98.
Currie, C. K. (1989) The function of fishponds in the monastic economy. In The Archaeology of Rural Monasteries (Ed. by R. Gilchrist & H. Mytum), pp. 147-172. British Archaeological Reports, British Series, 203.
Currie, C. K. (1991) The early history of the carp and its economic significance in England. Agricultural History Reviews, 11, 97-107.
Desroches-Noblecourt, C. (1954) Poissons, tabous et transfromations du mort. Nouvelles considerations sur les pelerinages aux villes saintes. Kemi, 13, 33-42.
Diamond, J. (1997). Guns, Germs and Steel. A Short History of Everybody for the Last 13,000 years. Chatto & Windus, London.
Drewer, L. (1981) Fishermen and fish pond: from the sea of sin to the living waters. Art Bulletin, 63, 533-547.
Drews, R. (1951) The Cultivation of Food Fish in China and Japan: A Study Disclosing Contrasting National Patterns for Rearing Fish Consistent with the Differing Cultural Histories of China and Japan. PhD thesis, University of Michigan, Ann Arbor.
Dyer, C. (1988) The consumption of fish in medieval England. In Medieval Fish, Fisheries and Fishponds in England, (Ed. by M. Aston), pp. 27-38. British Archaeological Reports, British Series, 182 (i).
Edwards, P., Little, D. C. & Yakupitiyage, A. (1997) A comparison of traditional and modified inland artisanal aquaculture systems. Aquaculture Research, 28, 777-788.
Ellis, J. N. (1994) In search of history: early days and trout feeds. Aquaculture Magazine, July/August, 50-51, 54, 56-59.
Flannery, T. F. (1994) The Future Eaters. Reed Books, Australia.
Grigg, D. B. (1974) The Agricultural Systems of the World. Cambridge University Press, Cambridge.
Harris, D. R. (Ed.) (1996) The Origins and Spread of Agriculture and Pastoralism in Eurasia. Smithsonian Institution Press, Washington.
Hepher, B. (1988) Nutrition of Pond Fishes. Cambridge University Press, Cambridge.
Hickling, C. F. (1962) Fish Culture. Faber & Faber, London.
Hickling, C. F. (1971) Prior More’s fishponds. Medieval Archaeology, XV, 118-123.
Hoffman, R. C. (1995) 'Carpes pour le Duc ...:' the operation of fishponds at Laperrière-sur-Saône, Burgundy, 1332-1352. Archaeofauna, 4, 33-45.
Hoffman, R. C. (1996) Economic development and aquatic ecosystems in Medieval Europe. American Historical Reviews, 101, 630-669.
Huet, M. (1972) Textbook of Fish Culture. Fishing News Books, Farnham, Surrey.
Immink, A. J. (1997) World Aquaculture Production 1950-1983. Report for the FAO, Rome. Institute of Aquaculture, University of Stirling.
Jiménez-Osornio, J. J. & Gomez-Pompa, A. (1991) Human role in shaping of the flora in a wetland community, the chinampa. Landscape and Urban Ecology, 20, 47-51.
Kirk, R. (1987) A History of Marine fish Culture in Europe and North America. Fishing News Books, Oxford.
Knowles, D. (1950) The Monastic Order in England. Cambridge University Press, Cambridge.
Lewis, D. J., Wood, G. D. & Gregory, R. (1996) Trading the Silver Seed. Intermediate Technology Development Group, London.
Li, K. M. (1992) Rice –fish farming systems in China: Past, present and Future.. In Rice-Fish Research and Development in Asia, pp 17-26, (Ed. by C. R. dela Cruz, C. Lighfoot, B. A. Costa Pierce, V. R Carangal & M. P. Bimbao). ICLARM Conference Proceedings 24, ICLARM, Manila.
Li, S. F. (1994) Introduction: freshwater fish culture. In Freshwater Fish Culture in China: Principles and Practice, (Ed. by S. Li & J. Mathias), pp. 1-25. Elsevier, Amsterdam.
Ling, S. W. (1977) Aquaculture in Southeast Asia. University of Washington Press, Seattle.
Little, D. C. & Edwards, P. (1997) Contrasting strategies for inland fish and livestock production in Asia. In Recent Advances in Animal Nutrition in Australia, pp. 77-85. (Ed. by J. L. Corbett, M. Choct, J. V. Nolan & J. B. Rowe). University of New England, Armadale, Australia.
Little, D. C. & Edwards, P. (2000) Integrated Livestock - Fish Farming Systems: the Asian Experience and its Relevance for Other Regions. .FAO Fisheries Technical Paper. FAO, Rome. (in press).
Little, D. C. & Pham, T. A (1999) Common carp indigenous ricefield-based culture in Northern Vietnam. AARM Newsletter, 4 (3), 4-5.
Lu, X. (1992) Fishery Management Approaches in Small Reservoirs in China. FAO Fisheries Circular 854. FAO, Rome.
Maitland, J. R. G. (1887) The History of Howietoun. J. R. Guy, Stirling, U.K.
McDonnell, J. (1981) Inland fisheries in medieval Yorkshire 1066-1300. Borthwick Papers, 60. University of York.
McLarney, W. (1984) The Freshwater Aquaculture Book. Hartley & Marks, Point Roberts, Washington.
Micha, J.-C. and Chavez, M. (1997). Development of agro-piscicultual ecosystems in tropical marshlands. In: Integrated Fish Farming, (Ed. by J. A. Mathias, A. T. Charles & H. Baotong.), pp. 347-358.CRC Press, Boca Raton.
Pitcher, P. T. J. & Hart, P. J. B. (1982) Fisheries Ecology. Croom Helm, London.
Rana, K. J. (1998) Recent developments in aquaculture statistics. Fishery and aquaculture statistics in Asia. In Proceedings of the FAO/SEAFDEC Regional Workshop on Fishery Statistics. 19-21 August 1997. Vol. 11, pp.242-254. FAO, Rome.
Reay, P. J. (1979). Aquaculture. Arnold, London.
Roberts, B. K. (1966) Medieval fishponds. Amateur Historian, VII, 119-126.
Roberts, E. (1985) Alresford Pond, a medieval canal reservoir: a tradition assessed. Proceedings of the Hampshire. Field Club and Archaeological Society, 41, 127-138.
Ruddle, K. & Zhong, G. (1988) Integrated Agriculture-Aquaculture in South China. The Dike-Pond System of the Zhujiang Delta. Cambridge University Press, Cambridge.
Rumsey, G. L. (1994) History of early diet development in fish culture, 1000 BC to AD 1955. Progressive. Fish-Culturist, 56, 1-6.
Sahrhage, D. & Lundbeck, J. (1992) A History of Fishing. Springer -Verlag, Berlin.
Schaeperclaus, W. (1933) Lehbruch der Teichwirtschaft. Paul Parey, Berlin.
Schnitter, N. J. (1994) A History of Dams. A.A. Blakema, Rotterdam.
Schuster, W. H. (1952) Fish Culture in Brackish-Water Ponds of Java. Indo-Pacific Fisheries Council Special Publication No 1. FAO, Rome.
Sharma, K. P. & Rana, R. (1986). Seasonal ponds - a potential source of fish seed in Rajasthan, India. In The First Asian Fisheries Forum, (Ed. by J.L Maclean, L.B.Diizon & L.V.Hosillos), pp.87-88. Asian Fisheries Society, Manila, Philippines.
Silliman, R. P. (1975) Selective and unselective exploitation of experimental populations of T. mossambicus. Fisheries Bulletin, 73, 495-507.
Stearne, J. M. (1970-71) The medieval fishponds of Northamptonshire. Northants Past and Present, IV (5), 299-310.
Stewart, K. M. (1994) Early hominid utilization of fish resources and implications for seasonality and behaviour. Journal of Human Evolution, 27, 229-245.
Taverner, J. (1600) Certaine Experiments Concerning Fish and Fruite. William Ponsonby, London.
Tudge, C. (1998) Neanderthals, Bandits and Farmers. How Agriculture Really Began. Weidenfield & Nicolson, London..
Votruba, L. (1987) Historissche talsperren und wasserseicher in der CSSR. Historische Talsperren, Band 1, pp. 401-406. K. Witter, Stuttgart.
Welcomme, R. L. (1972) An evaluation of the acadja method of fishing as practiced in the coastal lagoons of Dahomey (West Africa). Journal of Fish Biology, 4, 39-55.
Yang, H. (1994) Integrated fish farming. In Freshwater Fish Culture in China: Principles and Practice, (Ed. By S. Li & J. Mathias), pp. 219-270. Elsevier, Amsterdam.
Zeepvat, R. J. (1988) Fishponds in Roman Britain. In Medieval Fish, Fisheries and Fishponds in England, (Ed. By M. Aston), pp. 17-26. British Archaeological Reports, British Series, 182 (i).