This article, now out of print, was published by Pluto Press in 1992 as Chapter 5 of Humanity and Nature: Ecology, Science and Society
Ill fares the land
To hastening ills a prey
Where wealth accumulates and men decay.
Oliver Goldsmith1
Agriculture marked a major turning point not only in human evolution but also in the ecology of the planet. The impacts of agriculture did not occur all at once but have emerged over the last 10,000 years or so.
[1] New ecological formations—cultivated fields, orchards, gardens and pastures—were created. Of the approximately 14,000,000,000 ha of land area on our planet, some 11 per cent was in cropland in 1981-3. Another 24 per cent was listed as permanent meadows and pastures, and 31 per cent as forests. The rest of the land area is lumped into the category 'other,' including wetlands, deserts, tundra, etc. The figures are very rough, but the indicate that a major part of the earth's land area consists of habitats created by human activity.
These new ecological formations affect the climate of the earth. The heat balance of our planet depends on the absorption of the sun's energy and its various transformations. But part of that energy is reflected back into space by clouds or by the earth's surface. The fraction reflected back is the albedo, and stands today at about 0.14, that is, about 14 per cent of the sunlight that gets into the neighborhood of earth is reflected. But different surface materials have different albedos. The snow-covered Antarctic has an albedo of about 0.85, an oak forest about 0.15, grain crops or grassland bout 0.23, and bare ground has an albedo which depends on its color. Thus the change from forest to crop land will lower the albedo of the earth during the height of the growing season, but raise the albedo when the land is bare. Irrigation also lowers the albedo and therefore raises the temperature of the earth.
More local effects also occur. The burning of sugarcane before harvest put so much ash into the Cuban atmosphere that it provoked increased rainfall just when this was least desirable because it interfered with the transport of the cut cane to the mills. The Cuban meteorologist Fernando Boytel (no date) argued that Cuba has a partly continental climate because the shallow muddy waters along the north coast behaved like land with respect to the sunlight. Therefore erosion can contribute to the heating of the earth.
[2] Unlike many natural ecological communities, cultivated areas have sharp boundaries, forming islands of distinct habitat. Whether we regard micro plots, even down to one or a few individual plants, as separate habitats or as part of a larger 'mixed crops' patch depends on the purpose of the analysis. From the economic point of view, the peasant may see a whole hectare as a mixed garden, but a bug that lives only on guavas may perceive isolated islands of single trees in a sea of weeds.
The nature of such island mosaics varies across agricultural systems. Plantation economies have often produced formations of thousands of hectares of a principal crop with isolated pockets of other land use, strips of forest along rivers and gulleys or the steepest slopes. Often the large plantations have fringes of minifundia around them, peasant homesteads where the farm laborers produced for their own maintenance.
[3] There has been a long-term trend toward reduced diversity of crops. Subsistence agriculture requires a certain diversity of crops both to provide for qualitatively distinct uses – different nutrients, medicinals, fuel and raw materials for construction and crafts. The development of commerce has allowed for local and geographic specialization. Trade makes crops interconvertible so that a single product can become many. This causes a drastic reduction in local diversity. One or a few principal crops and their associated weeds dominate the farmland, while the original native vegetation is banished to the inaccessible and waste land, and kitchen gardens preserve some of the original crop diversity. Plants which are not quite cultivated, yet are usefully gathered, may survive and be protected.
On a world scale some 3,000 kinds of plants are used out of about a quarter of a million species described. Of these about 150 are grown commercially and some others are cultivated for home use without reaching the market. Finally, humanity depends for its food mostly on a few principal crops: the grasses wheat, rice, maize, sorghum and millet; cassava, yams and potatoes; the legumes chickpeas, pigeon peas, dry beans and soybeans; sugar cane and beets.
[4] Along with reduced local diversity came the extension of ranges of a few principal crops. First they were introduced into regions similar to their native habitats in other parts of the world. But then they expanded into areas of less suitable habitat where production was less reliable unless special technologies, in the first place irrigation, protected them from the climate. Plant breeding led to the adaptation of plants to new habitats. For instance, for crops such as soybeans that are very sensitive to the day length, different photoperiod races with light responses suitable to different latitudes have been developed.
Today the geographic distribution of crop plants depends not only on their physiological needs but also on available technology and the costs of these methods compared to the prices of the crops. Growing seasons can be extended by irrigation, maize can be harvested before maturity and finish drying indoors with the help of heaters, smudge pots can protect orange trees against frost, vegetables can be grown hydroponically on islands that lack soil or in greenhouses in the tundra. The area sown to a crop varies from year to year according to market calculation, with both total production and yield per hectare responding more to prices than to weather.
[5] New species arose. The plants themselves have been selected under historically new conditions for traits desired by farmers. In some cases these traits made continued survival without human care impossible. For example, wild grasses shatter. While this is adaptive for the plants it makes harvesting more difficult although not impossible. The stick and basket are the appropriate tools for gathering shattering seeds and are used for crops like wild rice. The closed, non-shattering heads such as the ear of corn make harvesting much easier and bring in new tools such as the sickle.
The new cultivated habitat is also the environment for weeds. These are plants which have adapted to the conditions of cultivation, benefit from land clearing and fertilizers and irrigation. In a few cases, such as rye, the weeds themselves have been domesticated in turn and have become crops, while crops such as indigo have escaped from cultivation to become weeds.
Finally, new animal life has formed around the croplands. Many herbivorous insects are specialists feeding on only one or a few host species. It seems to be the case (Strong, 1979) that the number of herbivore species which feed on a plant in a given region increases with the abundance of that plant and its physical structural complexity. Trees provide more different habitats than bushes which provide more habitats than herbs and grasses. Crops are also fed upon by mites, roundworms, birds and mammals and are invaded by fungi, bacteria and viruses. It is estimated that various kinds of herbivores consume some 20 per cent of the world's agricultural production.
The herbivores themselves are the food resource for large numbers of predators and parasitoids. These are predaceous insects, mites and roundworms and parasites of the herbivores (microorganisms). Intermediate between the predators and the parasites are the parasitoids, mostly wasps and flies which lay their eggs in their hosts' bodies.
Other members of the community are the predators' predators, the parasitoids' own parasitoids, decomposers which feed on the plant and animal residues and their predators. A whole community is organized around cultivated plants and their associated weeds which may number hundreds of species.
[6] The patter of human settlement has been dominated by agriculture and livestock. Cultivation can support many more people than can be maintained by gathering and hunting. As long as people had to be supported from local production, population concentrated in the most productive regions, often along rivers. But 'a region' is not defined by natural conditions alone. Is a single patch of meadow a regions, or does the region include adjacent forests and mountains? This depends on the socially organized movements of goods. Anthony Leeds (1980) studied the historical ecology of the hill country around Austin, Texas. In pre-conquest times it was an agricultural region supporting native population from local production. After conquest by European immigrants, it became a cattle region, no longer self-supporting but providing meat for the city and receiving other necessities by trade. Today it is a recreation area for Austin and no longer involved in significant food production at all.
At present the 5,000 million people in the world are fed from perhaps 1,400 million hectares of farm land. But the distribution is very uneven. Australian sheep ranches may have tens or hundreds of thousands of hectares while a Bengali farm might support a family from a single hectare and 90 per cent of the Salvadoran rural population is landless. Even in countries with high population densities the concentration of land holding allows some to have major investments in commercial agriculture while others can only supplement off the farm wages with a little gardening.
Over the centuries the productivity of world agriculture as a whole has been increasing slowly but unevenly. Japan and England have kept records for a long time. In Japan, rice yield in 800 CE was 1.2 tons per hectare. In 1600 it had risen to 1.5 t/ha, by 1820 it had reached 2 and now it stands at around 6. Wheat yields in England rose from 0.4 t/ha in 1225 CE to 0.8 t/ha in 1600, 2t/ha in 1900, and now stand at about 3.8. In recent times the improvement of agricultural yields has accelerated (for example, Braudel, 1981).
Machinery has allowed for large-scale extensions of cultivated areas through the pumping of water for irrigation and the tractor increased the area a single person could work. After the First World War, the nitrogen fixing industry developed to produce explosives became a chemical fertilizer industry. After the Second World War, chemical warfare was aimed at insects and microorganisms, and the expansion of pesticide industries led to further advances. Finally, plant breeding, more recently accelerated by 'biotechnological' methods, has increased the capacity of plants to use nitrogen and resulted in the 'green revolution' increases in productivity.
But the march of progress has been uneven. Indirect methods of estimation by anthropologists and archeologists have suggested yields of cassava of 20 t/ha in Mayan Central America in areas no longer cultivated. The Barind region of present day Bangladesh, one the heart of Golden Bengal and the breadbasket for the ancient Hindu capital at Gaur, is now a badly eroded semi-arid neglected area of that poor country. In some cases single crops have had to be abandoned due to pests and diseases, such as coffee retreating from Sri Lanka before the rust disease. The advance of agriculture is neither universal nor inevitable.
The Critique of Modern Agriculture
The successful long-term increase in agricultural yields has given rise to a widespread optimism. It was believed that the development pathway that had proven so effective – increased and more diverse technical inputs to agriculture from off the farm – could continue indefinitely and that any problems created along the way could be solved by the same methods that created them.
This optimism was organized around the developmentalist ideas about progress: progress, or modernization, takes place along a single path from backward to modern. However, in recent decades a growing uneasiness with this pathways has emerged, for the following reasons:
[1] Modern technical high-input agriculture is quite inefficient in terms of energy and resources. Its efficiency is limited to output per unit area, which is greatest in the Netherlands, Denmark and Japan, and to output per unit of labor, which is maximized in the U.S. The labor efficiency is usually exaggerated tenfold however: while one farmer may be involved in producing food for 40 consumers, food is really produced by one farmer plus five workers making the mechanical, chemical and energy inputs to the farm and another five handling, processing, storing, shipping, packing and selling the food. Therefore in the United States one food production worker on or off the farm produces food for about four consumers.
[2] Modern agriculture undermines its own productive base. The extensive use of irrigation uses up the reserves of the aquifers, making it necessary to dig ever deeper to mine water. Irrigation results in the salinization of the land. Heavy machinery or overgrazing of cattle compacts the soil. Annual cropping results in the loss due to erosion of some 5-7 tons of topsoil per hectare and per year in the United States, in the worst areas up to 50 tons per hectare, while natural processes can replace only 1-2 tons (Pimentel et al., 1987). In other parts of the world where sloping land is farmed or attempts are made to cultivate the rain forest, erosion losses are greater, and when this takes place on shallow soils the impact is felt sooner than on the deep soiled prairies of North America.
Modern large-scale monocultures have lower organic matter content that the soil previously held. This increases the need for irrigation and nitrogen fertilizer and reduces the diversity of the community living in the soil. One result is increased pest problems. Another is the depletion of the minor nutrients which are not included in the mass fertilizer application. This comes about because increased yield means greater quantities of material removed from the field and not restored. But in addition, the heavy doses of nitrogen, potassium and phosphorus interfere with the uptake of the minor elements such as boron, zinc, copper and manganese.
In affluent economies this declining productive capacity of the land can be hidden for a time by increased inputs of fertilizer, pesticides, irrigation water and mechanical energy. But the economic conditions under which this solution is viable become increasingly stringent and eventually the land has to be abandoned, first passing through stages of rain fed farming and pasture.
[3] Modern high-technology agriculture increases the vulnerability of production. Previously, diversity provided a buffer against the uncertainties of the environment. While some plants suffer from abnormally wet years, other prosper. Early frost might kill the corn but the oats will have been harvested. Pest outbreaks usually affect only one or a few crops, allowing the others to mature normally. A price decrease in one crop may be compensated for by a rise in another, or on-farm diversity can at least feed the family even if cash income is lost.
Diversity is lost when the single criterion of profitability excludes considerations of risk, food value, environmental impact, employment opportunities, taste and other factors. Not all crops are equally profitable. Since 'green revolution' plant breeding emphasized an increased capacity to respond to nitrogen fertilizer, grasses (grains) improved more that legumes, which have their own nitrogen fixing mechanisms. Therefore high-yielding wheat has tended to replace chickpeas in India. The absorption of Third World agriculture into the world market with its increased demand for meat has created situations in which it is more profitable to grow feed than food or divert farmland into pasture.
Large-scale monocultures are ideal environments for pests. Some of these pests have emerged as significant problems only with modern agriculture. Monoculture provides large expanses of available food where populations can spread continuously without barriers. Commerce often allows the pests to spread even faster.
Pesticides eliminate natural enemies, allowing minor herbivores to explode into major disasters. The brown plant hopper, so devastating for rice, seems to be a problem only after intensive pesticide use. Mites as pests of orchards, army worms (Spodopter species), fruitworms (Heliothis and others), whitefilies and other common pests are referred to as secondary pests because their emergence as problems follows pesticide use.
Modern plant breeding, especially around the green revolution, has emphasized yield above all other considerations. For instance, a dwarf variety of wheat puts more of its product into grain and less into stems and roots. But this means that weeds can outgrow the wheat and more active weed control is required. Reduced roots means more need for irrigation. In general, the new varieties are bred for performance along with a whole technical package. In its absence they perform poorly.
Fertilizer technology, especially large doses of inorganic nitrogen, makes plants more attractive to insects since nitrogen is often a limiting nutrient for them. And many plant diseases seem to be more severe with nitrogen fertilization. This seems to contradict the widely held belief that healthy, well-nourished plants are more disease resistant. However, abundant inorganic nitrogen is more favorable to the microorganisms in question than to the plants, which shows increased vegetative growth but more succulent tissues and some imbalance of other nutrients.
In market economies, modern agriculture increases economic vulnerability. The use of irrigation, fertilizer and pesticides provides at least temporary protection against the vagaries of nature. But since these are produced off the farm and must be paid for, variation in prices can have a greater impact on the food supply than variations in rainfall. The root zone of a wheat plant in Argentina now has as part of its environment the politics of the Middle East by way of the price of oil and therefore fertilizer.
When variation in production passes through the market its impact is amplified. For example, only about 20 per cent of the world rice production is traded internationally. Therefore a 5 per cent decrease in production results in a 50 per cent reduction in rice on the world market and corresponding to this, an enormous rise in prices. Finally, the inequality in a society usually means that a small fluctuation in production is allocated by the market in such a way as to affect the poor disproportionately. For instance, if there is a 5 per cent decrease in production of rice, this results in a price increase. The affluent will still eat as before, but if the decrease is absorbed by only the poorest 25 per cent of the population, then they eat not 5 per cent less but 20 per cent less rice.
Unpredictable economic fluctuations are sometimes exacerbated by deliberate refusals to sell to particular countries as a form of political pressure. Under these conditions monoculture and specialization are even more dangerous to a developing country.
[4] Modern agricultural technology harms the health of farm workers and their families and also the general public by way of pesticide contamination. In 1972, the World Health Organization estimated that there were some half a million cases of pesticide poisoning annually with a 1 per cent death rate. This is certainly an underestimate because the symptoms of pesticide poisoning are very variable and not always recognized.
[5] Modern high-technology agriculture pollutes the environment. Pesticides get into the air, the soil and the water. Fertilizer run-off gets into streams and lakes where it provides nutrients that permit bloom of algae. During this bloom, oxygen is produced which returns to the atmosphere. But when the algae die the organic matter sinks to deeper water and decays. The decay process uses up oxygen so that we get zones of oxygen deficiency which are lethal to fish and invertebrates. Fertilizers react with the soil conditions and sunlight to release nitrogen in the form of nitrogen dioxide (NO2) which is a photooxidant affecting health and acid deposition and a greenhouse gas.
[6] The quality of food is debased. The single-minded pursuit of yield as a determinant of profit has often resulted in the sacrifice of protein content or other aspects of nutritional value to bulk. Heavy applications of nitrogen unbalance the mineral composition. The mineral content of vegetables can vary by as much as an order of magnitude depending on the conditions under which they were grown. The requirements of storage and transport take precedence over food quality. The competitive marketing of food has led to cosmetic criteria as spurious indicator of quality which encourage unnecessary pesticide use since many insects have effects limited to the appearance of the outer surface of fruits.
[7] Modern technology alters social relations in the countryside to the detriment of the poorer farmers and laborers. Differential access to the technology through acce to credit or information contributes to the failure of many farms and the concentration of land holding and production inf fewer hands. In the United States the initial investments in land and farm machinery prevent many aspiring farmers from entry into agriculture. In the Third World it is often more profitable to replace peasant labor, which uses some land for subsistence crops, by wage laborers dependent solely on the income from working the large farms. The displaced peasants may become landless laborers or migrate to the cities. Jobs created by the new technologies are fewer than those eliminated except where year-round cultivation is made possible by irrigation and mechanization.
[8] The position of women is often undermined. Even in countries where women are responsible for most of the farming, new technologies are usually given to or fall into the hands of men. Modern post-harvest processing equipment displaces cottage level husking and milling which were usually women's occupations. And the greater unevenness of labor demand during the annual cycle works to the disadvantage of women with with children, who can run farms with divers crops making more level demands on their time.
[9] Agricultural science still operates on a very narrow intellectual base, emphasizing single inputs for each purpose and ignoring the broader implications of recommended technologies. Given the complexity of nature, many of the consequences of introducing innovation appear as 'unexpected' side effects. Research is dominated by the search for marketable input commodities rather than ecological knowledge to reduce the need for inputs.
This constrains the techniques to be applicable within the confines of a single farm. For example, if a pesticide is effective against the adult of an insect which does not damage the larva, it will not protect the crop. If, in addition, the insect adult is highly mobile so that next year's invasion does not come from the survivors of this year's reproduction on the on this farm but from the survivors in the whole region, it will not benefit an individual farmer to use this product and the market researchers will recommend against its development. But a region-wide integrated pest management program might find the product useful.
Integrated pest management (IPM) (Levins, 1986) has a dual significance. On the one hand it reflects an awareness that promiscuous pesticide use is counterproductive and is a small step toward ecologically rational methods. Pesticide use is reduced by careful evaluation of what pest levels cause sufficient economic harm to justify intervention, and ingenious monitoring procedures have been invented to carry this out. The timing of applications is planned in such a way as to minimize impact on beneficial insects. The release of laboratory-raised parasitoids, the careful timing of planting dates, the breeding of resistant varieties, the use of pheromone traps and repellants have all contributed to the reduction of pesticide use.
But the other side of IPM is that it is a rear guard action by industry offered instead of biological control as a program which acknowledges ecology but still attempts to preserve a major role for the agricultural chemicals industry. Therefore the strategy of IPM has been to invent a much-enlarged bag of tricks for farmers to buy. These include both commodities and services, as computerized decision-making consultants enter agriculture.
Since the selling of commodities as farm inputs is the most profitable way of turning agricultural research into profit, the dominant research strategy is one of increasing the diversity of off-farm inputs. An alternative to this strategy gives research priority to the design of almost self-operating systems that require less intervention.
[10] Specialized research, cried out in the industrial countries or the major urban centers of the Third World, is part of a technical and economic package that confronts indigenous peoples as cultural invasion, disrupting their ecology, knowledge, technology, beliefs, land use system, autonomy and demography. Thus modern agriculture must be seen as a seccessionalstage which creates the conditions for its own replacement. The question is whether it will lead to a wasteland or an ecologically and socially integrated pattern that protects nature, production and people.
Most of these criticisms are not familiar although usually not presented together. Nevertheless, the dominant pathway of development is still the high-technology package imbedded in a market economy.
Developmentalism
We argued in Chapter 2 that environmental destruction continues in the face of evidence of its destructiveness due to the various combinations of greed, poverty and ignorance that determine development policies in different parts of the world. In agriculture, greed takes the form of commoditization. Poverty acts to create an urgency to produce which is impatient with criticism of undesirable consequences and unable to take the risks of changing production systems or challenge the science and technology of countries that seem to be successful. Ignorance is structured into a complex system of beliefs, information and gaps in information, philosophy of science and supporting institutional and policy arrangements which discourage any radical departures. Central to this structure is the developmentalist outlook which supports the following propositions:
Modernization is the transition from labor-intensive to capital-intensive technology, mobilization of energy to move or transform huge quantities of matter in order to free people from the overwhelming physical burdens of farming.
This is indeed descriptive of some of the major changes that have taken place in agriculture in the last century. However, it is still a nineteenth century thermodynamic model in which success is measured by the quantitative movement and transformations. An alternative model, however, is more concerned with the flow of information rather than bulk, and with very small inputs producing big effects. The energy involved in a nerve impulse or the total bulk of a hormone or even the power consumption of a super computer are negligible compared to the effects they have.
Thus we have some precedent for suggesting that the evolution of agriculture can be from labor intensive to capital intensive to thought and knowledge intensive, and that intellectual effort can be directed less toward inventing new inputs for agriculture and more toward designing agroecosystems that are as self-operating as possible and require minimum input of capital and labor.
Modernization simplifies for efficiency, so that the evolution of agriculture must be from diversity to uniformity and specialization at the level of the enterprise or region.
This again is a roughly accurate description of what has happened. But this homogenization is more a consequence of the economic history and organization of the industry that of the nature of production. We have already indicated that monoculture is ecologically inefficient. The diversity of peasant production way partly a consequence of peasant technology, partly a deliberate choice to have a variety of foods or a hedge against disaster, and partly a consequence of a patchwork land tenure so that the countryside was a mosaic reflecting also the different needs and decisions of different farms.
We suggest that the evolution of agriculture can be from the partly random diversity of the tiny peasant landholdings through the uniformity of agribusiness plantations to an ecologically and socially rational planned diversity. An approach to such planning is outlined below.
Modernization requires a transition from small to large scale.
The argument here is the one of economies of scale. Large fields allow for larger equipment; large farms allow for more investment in equipment and even to allot resources for gathering technical and economic information or hiring expert technicians and managers; large enterprises can have access to credit and will be less vulnerable to the whims of the market.
In opposition to the obvious harmful consequences of at least some giant enterprises, ecology-minded groups have put forth the counter proposal that 'small is beautiful.' This is partly an argument for diversity and partly for scaling down of machinery so that is will be compatible with small farms which are viewed as socially desirable.
However, large planned units are not always unsuccessful, and the uniform rejection of large planned units is more an ideological commitment that a careful analysis; we return to this question below. On the other hand, the widespread failure of family farms in the United States is well known. Here we have situations of economic failure with at least relative technical success in the short run. There does not seem to be any convincing evidence at present that one or another organization of agriculture is clearly, universally and overwhelmingly superior. Rather, the experience of each depends on its particular history and context.
When we discuss the scale of farming, it is necessary to differentiate between the unit of production and the unit of planning and remuneration. The often extremely localized qualities of different pieces of land due to topography, soils, flows of moisture and past history of use argue for small or variable sized units. Interactions between patches of land with different vegetation types modify the microclimate and can be used to improve the conditions of production. These effects occur roughly on a scale of ten times the height of the plants. Residues of one kind of patch can be used as inputs on another. Proximity makes this easier to do. The movement of beneficial insects, birds and other organisms from one patch to another is an essential part of ecological pest management. This requires that distances between patches be compatible with the mobilities of the creatures in question and would usually argue for modest-sized patches. The desirability of farmers' getting to know each of their dairy cows individually so as to care for them better and detect disease sooner argues for a maximum herd size below 100 animals, while epidemiological dangers also urge limits on poultry units.
But while individual patches of crop or other activity should be small enough to allow for the advantages, the whole array of patches may be quite large. The management of water is best carried out on the scale of a whole watershed and would require resources such as labor from all subunits. The management of disease or pests must take into account the range of a unit population of pests, their mobilities. The maximization of benefit for the ensemble as a whole, no matter how benefit is defined, is different from the maximization of benefit from each patch separately.
But if some patches will be more productive than others, and may be devoted to less productive activities than would be possible in order to improve the whole, people deriving their income from single patches of vegetation will be unequally rewarded for efforts that may be equally hard. It would not be feasible to ask some farmers not to grow the most profitable crop in order to protect their neighbors from pests. Further, remuneration should in some way reflect the productivity of the whole set of patches.
Therefore the problem of scale requires combining the advantages the advantages of detailed local adaptation with larger scale coordination. How that coordination is to be achieved cannot be settled by ecological argument.
Modernization aims at an increasing control over nature, the freeing of farming from the unpredictable caprice of weather and pests
There is some truth in this assertion. The uncertainties of rainfall can be compensated for by irrigation, unexpected pest outbreaks by pesticides, unauthorized plant growth by weed killers, and so on. However it must be remembered that in some ways this merely shifts the focus of uncertainty from physical and biological factors to economic ones while contributing to the long-term vulnerability and productivity loss of the system. Further, the goal of increasing control has often led to the notion that any plant that wasn't sown is a competitor, any antenna that twitches threatens crop destruction. The vegetation between fruit trees is clipped to golf course height, the cows are kept in air conditioned boxes, nature is treated as an enemy. The uncertainty of nature is a real problem for farmers. It can be confronted in several ways: we can enhance diversity, so that declines in one crop are compensated for by increases in others. We can breed for tolerance of extreme conditions such as deep-water rice, salt-tolerant grains and vegetables, drought-tolerant crops, etc. There can be systems of redistribution so that losses in one area are made up for by increases in others. And there can be improved prediction within systems that minimize the vulnerability to the unusual.
Good agricultural practice does not require complete control. Not all weeds or herbivores are really harmful. An insect that eats lower shaded leaves of tomatoes can improve yield and ripening quality of the fruit. Some fungi improve the flavor of grapes. Many grains depend for production only on the topmost leaves (in maize, the leaves above the ear.) Beans can tolerate heavy visitations of whiteflies after they are older than about three weeks, provided the whiteflies do not carry viruses.
An alternative to the impractical goal of increasing control is one of design of agroecosystems such that we can get away with not controlling most of what happens and rely on the resilience, robustness and feedbacks in the system to ensure that our needs are met. A diverse even if not censused fauna and flora is a protection against pest outbreaks. A rich soil with good organic matter and nearby forests would buffer against fluctuations of rainfall. Strong technical interventions could be held back as last resort emergency measures.
The smaller the object of study, the more scientific it is, and modern science is increasingly specialized.
This is the reductionist bias which has dominated Euro-North American science since its inception, creating a hierarchy of the sciences and allowing 'modern' biology to be identified as molecular genetics. This also results in the typical errors of deriving programs for the large from tests of the small and places much of research out of reach of poor countries and non-specialists. When we recognize the legitimacy of science on all levels we also make possible mass participation of science and the mobilization of vast intellectual resources. It is particularly necessary to break down the barriers between social and biological sciences in agriculture to give scientists and appreciation of the history both of their problem and of thinking about it.
Modernization requires the replacement of superstition by science.
The term superstition implies that traditional or folk knowledge is necessarily wrong. Attitudes toward prescientific knowledge in medicine and agriculture have gone through many changes in recent times. The reality that all knowledge come from direct or indirect experience and reflection on that experience in the light of previous belief. Therefore each system of knowledge comes from a point of view and from a particular location in the world which carries with it interests and indifferences and both insights and blindnesses. Modern science is a particular episode in the history of knowledge in which the creation of knowledge has become part of the social division of labor, organized around a special training and socialization or recruits, the organization of separate institutions and supported by a self-conscious concern for rules of inquiry.
The history of agricultural development schemes is full of examples of new technologies being rejected for apparently irrational reasons that proved to be well founded later on. Peasant knowledge tends to be local, detailed and intimate where scientific knowledge acquires its theoretical insights at a certain distance from the particular. Both kinds of knowledge are needed for ecologically rational agriculture. Particularly, as we seek technologies that are gentler they are also more site specific: where the objective is to kill everything that creeps on the field it almost doesn't matter what they are or how they relate to each other, but if we want to nudge the ecosystem toward better production and productivity a detailed knowledge is critical. Things that work in one place will now work somewhere else, often for quite subtle reasons.
Therefore, a precondition for an ecological agriculture is the creative combination of the detailed, intimate but very local knowledge that people have of their own circumstances with the abstract, theoretical and general knowledge that only scientific distance, comparison and intellectual detours can produce. This exchange of insights requires conditions where the farmers can meet with the agricultural scientists as equals on terms of mutual respect.
The Design of Sustainable Agricultural Systems
At the time of writing there is a growing concern about the sustainability of development. The notion of sustainability is itself an object of conflict. For those who are satisfied with existing economic and power relations it is the maintenance of productivity, keeping society as it is. For some it is the removal of the most obvious abuses and a democratization of access to opportunity for production and consumption while leaving both production systems and consumption patterns intact. for others sustainability is economically productive, ecologically rational and socially just.
Sustainability can be posed on the level of the single farm, the region, the nation or globally, and on time scales from years to centuries. As we expand the temporal and spacial scales we also usually have to expand the substantive scope since constraints which operate in the small become objects of decision making in the large. New problems and new degrees of freedom are added. Questions of land tenure, the availability of water in aquifers and rivers, the climatic and hydrological impact of forests, the choice of crops for a region, the structure of markets and prices, population and employment trends, dietary preferences and the organization and priorities of research are all givens from the point of view of the single farm but must be examined, questioned and decided about on a larger scale.
It is necessary for us to look at several scales. The smallest scale and shortest term horizon confronts the individual peasants, farmers and extension workers as practical issues of survival. There are severe constraints at this level, exerted by local custom, the availability of markets, transport and storage or processing facilities, the priorities of creditors, and access to information and seed. Tenants may not be sufficiently secure in their tenure to consider land improvement. They may be constrained by the priorities of landlords for whom farm income is only one part of their income. If they are also money lenders and food merchants and pesticide vendors, then the diversification of crops, expansion of the farming season and the reduced dependence on purchased inputs goes against their interests, and they oppose farm scale programs for sustainability and ecological and social rationality. Expansion of farming opportunities for women may improve the economy of the household but weaken the dominance by men and cause further conflict. It is not at all obvious that 'improvement' appeals to everyone and will be accepted once its merits are demonstrated.
Therefore the single-farm measures discussed below may not always be practical. However they can be introduced where the social conditions permit and otherwise serve as models of the possible.
The individual farmer can in principle make decisions about land use within the farm, choose crops and techniques for water and soil management and pest control, especially for the less mobil pests.
The prevailing analysis of farmers' decisions is based on neoclassical economic models. They assume that the farm is a business with the goal of maximizing profits. A crop responds to an input such as water or fertilizer or labor, at first rapidly and then with diminishing return. The optimum level of input is that at which the cost of further input just balances the price received by increased production. A lot of sophisticated modeling has gone into describing the responses of crops to inputs of several different kinds simultaneously. Then development strategy consists of trying to manipulate the costs and prices in such a way as to get farmers to make desired decisions. In the research program of the 'green revolution,' plant breeding was aimed at changing the response curve of crops so that farmers would decide to consume more inputs and produce more crops.
A more integral approach to farm decisions would differ from the neoclassical model in several major ways. First, instead of a single criterion such as profit maximization, multiple criteria are used including long-term social goals. Second, the constants of the neoclassical model which depend on soil, moisture, microclimate and pests which determine the technical decisions in that model are here co-variables with yield. That is, any use of inputs and any yield level affects the conditions for the following year, so that yield and productivity evolve together. Third, the neoclassical model assumes that almost any inputs can be bought, so that yield and price are weighed against input cost, but in the integral model there are constraints on inputs set by ecological and social factors.
The diversity of crops makes land use patterns and crop interactions a major concern. Crops interact at different levels. Through sharing of a common constraint, they compete for resources. Through having different responses to weather, different pest problems and separately determined prices they jointly provide buffers against uncertainty. Their chemical composition makes them partly supplementary and partly interchangeable in the provision of nutrients. Their different microenvironmental requirements will usually suggest where to place them spatially. The residues of one productive activity such as manure and straw for compost or mulch may serve as inputs to another.
There are also direct agronomic interactions which offer advantages to diversity: first, plants affect the microclimate around them. Wind is interrupted by the growing plants. Leaves can intercept the mud splash that sometimes spreads fungal spores from the soil to the susceptible plants. Mutual shading can be beneficial or harmful. Second, combinations of plants with dense superficial root mats and deeper taproots can be used to manage the flow of moisture and the building of the soil. Third, combinations of crop improve pest control. Some species serve as trap crops for the pests of others. For example, corn can divert fruitworms from green peppers, and beans can perform this service for tomatoes while also fixing nitrogen. Others repel the pests that affect their companions. Garlic repels slugs invading lettuce beds, marigolds repel soil nematodes. Mixtures can confuse herbivores looking for their preferred food plant, provide nesting sites or nourishment for beneficial predators and parisitoids or fungi and attract pollinators.
These interactions occur over different spatial scales from close intercropping to alternating strips or mosaics of fields. The desirable scale depends on enhancing positive interactions while minimizing harmful effects. For instance harmful shading is a close up effect, but the movement of predators can take place over greater distances. These considerations and the problems of mechanization can determine the spatial separation that farmers eventually use.
Because of the complexity of these interactions among different agricultural activities, ecological issues enter more directly into economic calculation than is the case with simple profit maximization.
Usually there are a few major crops produced for market in any region. The existing markets, credit, services, skills and knowledge are all focused on these crops with little room for major departures. However a much greater number of potential crops are available for home consumption and small-scale local sales. Most peasant agriculture combines market and subsistence production. The subsistence crops provide a buffer against market uncertainty or disastrous weather, often absorb labor during the off season for the major crops, provide nutrition and medicine and raw material for crafts. Collective farms and state farms, even if specialized toward one or two commercial crops, generally also produce for the maintenance of their own members or workers. The subsistence fields on state farms are increasingly important in Cuba and provide the opportunities for agronomic innovation, crob interactions and learning how to manage diversity. The combination of economic and agronomic interactions listed above would result in decisions in which crops are not necessarily grown only where they do best considered separately. For instance, only a small area is required to meet the garlic need of a country. It can then be produced quite efficiently. But it may turn out to be preferable to grow all the country's garlic in protective bands around lettuce beds. The garlic production may be less efficient this way but the whole agricultural enterprise improved.
Some of the most interesting work on spatial patterns on the farm has been done by the Permaculture group in Australia. Their model is the independent yeoman farmer with sufficient land for diversity. In the permaculureee scheme, concentric rings around the residence contain activities with diminishing demand for constant attention. Thus the kitchen garden is close to the house, field crops further out, then orchards, pasture and woodland. The farm is also divided into quadrants with land use related to prevailing sunlight and wind direction, and modified further by detains of soil and topography. Since the growth of forest is a long-term project, permaculturists also consider the stages of tree growth, with temporary crops making use of the land between the young trees.
The rather specialized social assumptions of the permaculturist model make it less directly useful for the tiny peasant plots of much of the tropics or for large-scale production units that aim to feed cities. But their approach to serious consideration of spatial organization can be adapted to other systems.
Water is a flow-through on the scale of the individual farm. It enters by way of rain, run-off from uphill, subterranean water movement in the aquifer, and as irrigation water. It leaves as run-off, percolates down to the aquifer, evaporates from the surface or passes through the plants and leaves as transpiration.
As with any flow-through, the total input is externally determined (except for the purchase of irrigation water) and the output is equal to the input in the long run. Our object in water management is to regulate that flow so as to minimize the erosion caused by run-off and balance available moisture to the crop needs. In general plants grow best when the actually available water is equal to the potential evaporation for the soil and the transpiration through the plants, combined into the single term evapotranspiration. But the development of plants is never so integrated with rainfall as to maintain this balance throughout the growing season. Therefore various measures can be taken to approach that balance.
Water can be stored directly in ponds. Ponds can also provide for the production of fish, ducks and geese. Residues from the land can fertilize the ponds and fish waste and algae can fertilize the land. Ponds modulate the microclimate to a distance about equal to their width in the downwind direction. The immature stages of dragonflies and other predatory insects so that ponds contribute to pest control. Finally, the water is readily available for normal domestic use, irrigation and fire fighting. However, the construction of ponds can be costly, there is evaporation from the surface, and land is removed from terrestrial production.
Water may also be stored in the soils of forests and in cropland with good organic matter content. This is perhaps the optimal way of storing water since there is no major outlay of resources for construction and the measures which are required are also desirable for other reasons, for instance, preservation of soil organic matter.
Soil moisture also serves to moderate the fluctuations of temperature and can sometimes prevent frost damage. The large-scale drainage of areas of Florida may be responsible for the more frequent losses of orange crops to frost despite and unchanged average temperature.
Sometimes the problem is too much water. This can be dealt with by improved drainage and also by the use of plants with transpire more water back into the atmosphere. Eucalyptus trees are especially effective in this regard and were used in Italy to drain the Po valley marshes.
Water requirements must be measured by the crops themselves. Sometimes it makes more sense to adjust the plants to the water regime that to try to create a water regime for the plants. Whenever crops are grown outside their zone of adaptation, water management becomes a major production cost and the year to year variability of the yield is usually increased. Water requirements can be reduced by selecting crop species whose water requirements match the actual water regime.
Pest and disease management within a single farm is most effective for some of the less mobile pests and diseases where the results of one year's efforts affects the next year's outbreak. Minimal input methods include mixed cropping, trap crops, repellant crops, resistant varieties, and the design of ecosystems which maintain populations of the natural enemies of the harmful organisms.
The first line of defense would be the predators such as predatory mites, spiders, ants, ground beetles, earwigs and some bugs and nematodes. They tend to be generalists and can maintain themselves even when the pest is absent. Sometimes they need help in the sense of protecting nesting sites or providing food supplements. They serve a preventive role since they are there before the pest outbreak, but usually respond slowly to the outbreak if it occurs.
The next line of defense consists of specialized parasitoids, usually wasps and flies. Because of their specialization on only a few host species they will be rare or absent before an outbreak and lag behind the pest. But their dependence on the pest population also means that they respond rapidly to the outbreak and can become very abundant. The adult parsitoidsss usually require a source of nectar for their maintenance, so that nectar sources should be grown to attract them. Sometimes it is even desirable to maintain pest populations at low levels so that when the outbreak occurs the parasitoid will not lag too far behind. For instance the scale insects which attack citrus in Cuba are very seasonal, coming with the rains and the flush of new leaves. But they are less seasonal on guavas. The guavas can keep a population sufficient to maintain the wasps that prey on them.
Insect diseases can be both preventive at the endemic level and can increase rapidly when an outbreak occurs. The fungi Metarhiziu, Beauvaria and Hirsutella have been used for this purpose.
There are technical constraints affecting the use of these methods on the scale of a single farm. If very mobile pest can be killed only after the damage is done, it will not protect this year's crop. And if the outbreak next year comes from a wider area than a single farm, post-damage control will not help that farm very much. Further, populations of parasitoids, birds and bats (excellent predators) usually depend on larger areas than a single farm.
Some kind of coordination on a larger scale is necessary. But this is often quite difficult to achieve. The Hessian fly is a major pest of wheat in North America. One remedy is the growing of resistant varieties which yield satisfactorily during an outbreak but which are inferior to the susceptible wheats when the flies are absent. During the rising phase and peak of an outbreak it is to the advantage of all farmers to switch to the resistant varieties. Their joint efforts cause a decline in the Hessian fly population over a whole region. But once the outbreak is receding it is to the advantage of each farmer to return to susceptible varieties before the neighbors do so as to derive the benefits both of the higher yielding variety and the protection their neighbors' resistant variety provides for all of them.
Regional Agroecology
When we shift focus from the single farm to the region, new degrees of freedom become available but also new requirements for an effective system.
The availability of water, pretty much a given for an individual farm, can be managed regionally. Now the exploitation of the aquifers, the land use around rivers and the preservation of forests become objects of decision making.
The prices and costs which are givens for on-farm decisions are affected by what all the farmers do. Here the inequality among farms plays a role. If the prices received, the costs of marketing or of inputs, and the constraints on inputs vary from farm to farm, not only is there inequality of living conditions but also an increased total accumulation of profit despite reduced total production. Therefore the farmers and planners interested in total production would favor those measures which reduce the variation such as rural roads, market integration and eased access to credit. And the goals of production and equity are compatible. But the merchants who can benefit from the variation or investors interested in the total rate of accumulation would find these measures deleterious and oppose them.
On the other hand variation in natural conditions an dinn the responsiveness of the crops to inputs has the effect of increasing both total profit accumulation and production while increasing inequality. Therefore technical innovations which benefit mostly the better off farms will have a greater impact on production than measures which improve the yields of the poorer farms. This means that planners with interest in increasing total production will have common interests with the better off farms even if they have no direct personal ties to the wealthier landowners. They will tend to sympathize with strategies of 'building on the best' and to accept growing inequality as a 'cost of progress.' This makes research strategy a political issue imbedded in class conflicts of other kinds.
Regional land use policy will affect the food supply, equity, employment, resource use and population movement. Therefore planning on the regional level might proceed as follows:
[1] Classify land areas by potential uses as determined by topography, soil, access to water and climate. For instance the steepest slopes may be unsuitable for any direct production and conserved under natural vegetation for environmental protection, gentle slopes may be available for natural or managed forests or tree crops, and so on.
[2] List potential crops including forest, orchard and field crops and grazing. 'Potential' here refers to immediate potential and therefore to foods which are part of the diet or export market.
[3] For each crop, give its use value as food, industrial raw material, contribution to water management, soil improvement, effect on local climate, conservation and effects on other crops, and additional values such as recreational. For each use we can combine the contributions of different crops.
[4] List the economic value of each land use as providing foreign exchange, farm income and jobs and having requirements of water, labor, organic matter and so on. For instance the pasturing of beef cattle has a very low demand for labor, provides and export product, ruins soil and makes manure available for vegetable crops.
So far these are technical evaluations. Of course they can be refined further to take into account the impact per year and long-term trends. The next stage is a more political matter. The existing information has to be combined in order to contribute toward policy. For instance, it may be decided that all the food requirements of the region be met locally, that in order to discourage migration to the cities there must be an average of at least one full-time job equivalent per hectare, that average water needs cannot exceed one cubic meter per year, that a given amount of firewood be harvestable within 15 minutes walk of each household, that half the land area be wooded, that manure production suffice for 1,000 ha of vegetables and that pesticide use by kept below some tolerable threshold. The formulation of such goals is both a political and a technical process requiring both scientific expertise and broad democratic participation both to set reasonable goals and to make their implementation likely.
National Level Agriculture
Although national agricultural policy must include the same issues as we encountered regionally, these are subordinated to several other macro-scale issues: the land tenure question, the relation between production for consumption and export, land use, urban-rural balance and so on.
The first question is who owns the land, who works the land and how are they remunerated for their labor.
Some caution is required here. The notion of ownership is itself bound to particular societies and has many different meanings. Ownership really arose under capitalism. It includes the rights of use and abuse, sale or rent, and complete diversion from agriculture. Sometimes these rights are restricted by zoning laws, or some rights such as 'development rights' might be sold separately from the agricultural rights.
At present there are several forms of land holding in agriculture: the owned or leased family farm of Europe and North America is a modern capitalist enterprise producing for the market and making use of purchased inputs. It has a high ratio of capital to labor and adds only a small though essential part of the the vluee in food production. Economic decisions are based mostly on profit maximization.
The trend, especially in the United States, has been toward the decline in the number of family farms and an increase in their size. Family farms are faced with the double uncertainties of nature and the market, and have a precarious existence which discourages risk and subordinates long-term planning to short-term solvency. Since the objective of farming is to support the family, it is advantageous to keep working thourh as long a season as possible and on all the land. Therefore even crops which yield a lower profit may still be cultivated when other crops cannot be grown, or on land that won't support the principal crop. This encourages a certain amount of diversity. The major advantages of the family farm are the direct personal motivation of the farmers which encourages intense and disciplined labor, and the detailed intimate knowledge of the land and of individual animals, which allows for locally directed technical choices.
But even farms of several hundred hectares are usually too small to determine their own pest populations and water relations, which must be treated as givens.
The large corporate farm in the US is usually only one of a number of investments by its owners. It is often held only for a short time, used of tax purposes and then sold. Agricultural production does not allow for a more rapid turnover of capital that the reproductive cycle of plants or animals permits, and is beset with the dual uncertainties of nature and the market. Therefore it is not the investment of choice. Large corporations have tended to control the manufacture and sale of the inputs and the processing and marketing of the harvests while leaving to the farmer the risks and constraints of production in the field.However, is some areas such as California corporate agriculture has been profitable for fruit and vegetable growing.
The economic rationality of the corporation is somewhat different from that of the family farm. The objective is the maximization of the total profit from all investments taken together, not from each enterprise separately. Therefore if a farm yields a maximum profit which is below the average rate of return of the other enterprises of the corporation it is regarded as unprofitable. Since labor can be hired and fired, there is no need to produce when profits would be low. Further, since the corporation is not committed to any particular farm or even to farming as a whole, it can decide to use up the productivity of the land as quickly as possible and invest the earnings elsewhere. If the rate of reproduction of productivity (soil formation ndd leaching of salt) is slower that the discount rate of the economy then this kind of exploitation becomes 'rational.'
In Third World countries much of the agricultural land is in peasant households, large estates and corporate plantations. Peasants are farmers living at the periphery of modern capitalist societies and producing both for subsistence and for exchange with the surrounding markets. The may have title to their land or be tenants paying rent in money or kind, and may also engage in wage labor and craft production. The majority of the rural poor are landless and work for wages, sometimes in the form of housing and small plots as well as well as money. Peasant farming becomes increasingly incorporated into the wider economy so that, quoting Marx (The Eighteenth Brumaire), 'the small holding of the peasant is now only the pretext that allows the capitalist to draw profits, interest and rent from the soil while leaving it to the tiller of the soil himself to see how he can extract his wages.'
The degree of land concentration varies widely, and struggles over the land are a major factor in the political and social processes of these countries. Revolutionary movements in China, Vietnam, Kenya, the Philippines, Guatemala, and El Salvador have been motivated in part by a struggle for land reform. In El Salvador some 90 per cent of the rural population are landless. Indigenous communities in Andean and Amazonian regions of Peru, Bolivia and Ecuador have seen their own survival as dependent on retaining or recovering control of their land.
Collective agriculture occurs in a number of different forms: the Soviet and East European collective farms, the Israeli kibbutz, the Saskatchewan group farm, Cuban cooperatives, the Chinese commune. They differ in inputs consumed, internal organization, relation to the surrounding economy, and the extent to which participation is voluntary or coerced. This seems to be a major factor in determining the success or failure of the enterprise. Where collectivization was imposed coercively, production is often reduced by low motivation and resentment toward the administration experienced as coming from the outside. Even in voluntary collectives uneven motivation is often a problem. Where the collectivity is imbedded within a capitalist economy, as in the various experimental communes in the US founded for religious or political reasons, the Mexican ejido or the Israeli kibbutz, the pressures of the market act as coercions on the long-term development and lure the younger members of the collectivity away from farming. Some collectives such as many kibbutzim have resorted to hiring labor from off the farm, the farm then functioning as a collective employer.
The Chinese commune system was one of the largest, most integrated collective systems. The commune included many villages and often thousands of people, and assumed state functions below the provincial level in rural China. It organized diversified agricultural production, industrial enterprises processing the harvest and manufacture of inputs and tool, and engaged in watershed management, irrigation projects and reforestation. The results were mixed, and the system was abolished after the death of Mao Zedong. Subsequent criticisms of the communes have to be seen not as the self-critical admissions of error but as accusations by the winners of a political struggle against the losers, so that a full evaluation of the experience is not yet available.
Perhaps the best assessment comes from William Hinton. He claims that about one third of the communes were highly successful, a third were disastrous and the rest survived. The crucial difference between success and failure seems to be the extent to which democratic organization had real content or was perverted into an authoritarian hierarchy allocating resources and privileges in an atmosphere of corruption. At any rate, in the post-commune system of privatized farming, collective assets passing into private hands has produced some prosperity, while large-scale ecological projects have fallen into decay, class differentiation is reappearing, and a get rich quick mentality has shifted decision making away from sustainable models.
State farms are a common form of landholding in revolutionary societies. The state farm has the advantages of economies of scale, an on-farm scientific staff and better access to equipment and supplies. Its major failings have been a developmentalist, industrial model leading to high chemical and mechanical input, low-diversity, short-term economic urgency, and a management system which undercuts motivation and prevents adaptation of production to very local circumstances.
In Cuba, some 80 per cent of farm land is in state farms. A comparison of Cuban agricultural performance with that of neighboring countries is inconclusive. Yields are higher than in the other countries in some crops, lower in others. Rural health, educational and cultural resources are generally better than those of the other countries. This has helped Cuba slow down the rural-urban migration and prevented Havana from becoming a smaller scale Mexico City. Ecological rationality is spreading, with a policy of replacing pesticides with biological and natural control, and reforestation exceeds cutting in contrast with the regional trend.
Whatever the pattern of land tenure, the governments attempt to determine what happens on the land. This is done by positive planning, by price manipulation, subsidies and services, with different degrees of motivation and success.
The first priority has to be meeting peoples' consumption needs. This can be done both by national production and by exports exchanged for products that can be produced better elsewhere. Governments have wrestled with the balance between export and consumption agriculture. Exports can be exchanged not only for food imports but also for industrial inputs that are so important in development. Especially for tropical countries, it is possible to produce crops that the industrial temperate zones cannot produce, and therefore have almost guaranteed markets. However there is an element of uncertainty here: industrial countries look for substitutes for their imports, such as corn syrup to replace cane sugar, or synthetic fibers to replace cotton. A second problem with agricultural exports is the increasingly unfavorable terms of exchange for Third World agricultural products compared to industrial goods. A third issue is the vulnerability of the economy to international price fluctuations or discriminatory trade policies.
In terms of ecology, export production requires large volume, and this encourages specialized enterprises with monocultures and all the attendant dangers. Competition on the international market often imposes standards of appearance of products which lead to excessive use of pesticides. Yet export is necessary both to help finance the economy as a whole and i order to exchange for foods that cannot be produced locally. For instance, it does not make sense to grow wheat in the tropics. Yet bread had become an important part of the diets of the peoples of almost all countries, especially in the growing urban sector. Therefore wheat flour has to be imported, and this requires the export of tropical products.
It would seem that a mixed strategy is required which combines production for export and for home consumption.
Meeting the consumption needs of the people means satisfying first the existing diet and dietary aspirations. For instance, in most poor countries where meat was rarely seen, the consumption of meat is a basic consumer aspiration. This has led to the priority expansion of poultry and cattle production in Cuba. However, there are both ecological and health reasons to consider dietary changes. Meat production is one of the least efficient ways to turn solar energy into food. One kilo of beef requires 100 times more water for its production that 1 kilo of wheat, and consumes 16 kilos of feed grain. And Cubans already consume 79 grams of protein daily per capita, perhaps double the recommended requirements from a nutritional point of view. Already cardiovascular disease are a leading cause of death in that country, and increasing meat consumption must be a contributing cause.
Grazing is also destructive to the land, mainly through compaction. In the US, western rangelands have lost about one third of their carrying capacity for cattle as a result of prolonged grazing. Large-scale meat production unlinked to field crops is also a major nitrogen polluter of water supplies.
Therefore a long-term land use program should consider promoting changes in diet while satisfying present demand, but changing the eating habits of a population is not easy. Eating is a social activity as much as a nutritional one.
Within each region, land use provides employment as well as harvestable goods. Since different land uses have different demands for labor, the mix of activities must be in equilibrium with available and anticipated labor. Rangeland and conservation areas have the lowest demand for labor, basic grain crops are intermediate, and vegetable crops the highest. Additional requirements enter from ecology. For example, it may be that for every 1,000 ha of field crops we need 5 ha for forest for pest control purposes, or 10 ha of forest to provide the firewood to cook the food, or 100 ha to modulate the water supply.
Agriculture does not affect only agriculture. Any large-scale land use pattern also affects health and climate. The effect of farming on health enters through the use of chemicals, the role of soil in determining the nutritional qualities of food and as a detoxifier of the atmosphere and the effects of water management on mosquitoes and other vectors of disease. The climate effects come about through the altered albedo, wind erosion putting dust into the air and reducing the transparency of the atmosphere, the effect of moisture on the variability of temperature.
As we become more sophisticated in understanding the necessary relations among parts of a system, we find more requirements on that system and more different degrees of freedom to achieve them. Thus there is no final perfect scheme for agriculture but a strategy for designing the system which is open-ended.
