When you gaze from the top of the Temple of the Giant Jaguar, 96 metres above the floor of Tikal, you look down upon the crowns of giant rainforest trees, the branches of which crack and snap as howler and spider monkeys leap and chatter. Had you been dropped here from afar, you might have been forgiven for thinking you gazed upon a wilderness.

The Petén rainforest of northern Guatemala is one of the world's hotspots for biological diversity, containing 200 species of mammals and 500 of birds. You would be right to be awed, but wrong about the wilderness. During the Mayan Golden Age, between 250 and 900 AD, the capital city, Tikal, supported a population of between 10,000 and 40,000 people.

Since the mystifying collapse of the Mayan civilization, people have farmed with slash-and-burn methods. Fields are cleared in the forest, cropped for a year or two and then abandoned as families move on to new sites. Over time, as population has increased and as others have come to log the forests, so farmers have had to reduce their fallow periods and thus returned to former fields too soon for natural fertility to have been restored. Both agriculture and the forest come under pressure - yields remain low or fall and the forest steadily disappears.

All photos: Jules Pretty

But on the edge of the Petén forest farmers are using a magic bean to improve their soils and to save the rainforest. The velvet bean (Mucuna pruriens) spread to farmers when several Honduran and Guatemalan NGOs, in particular World Neighbors, Cosecha and Centro Maya, found that planting it between rows of maize substantially increased cereal yields. Mucuna can fix into the soil 150 kilograms of nitrogen per hectare annually - a free, natural fertilizer for farmers. For every hectare it also produces annually 50-100 tonnes of biomass, plant material that is allowed to fall on the soil as a green manure, suppressing weeds and helping to build the soil. In this bean lies the protection of the Petén rainforest. Build the health of the soil and farmers no longer want to burn trees to create new fields. It is, after all, difficult and dangerous work and farmers would love an alternative. An improvement to soil health changes the way that farmers think and act. They see the benefit of staying in the same place and of investing in the environment for the future.

Further west, along the border country of Guatemala and Mexico, lies the Usumacinta River. It is another area of extraordinary biodiversity. Inside the forest, this silent and eerie natural cathedral, the air is heavy with humidity and pierced with sunlight from the canopy far above. By some wonderful coincidence, this is the Cooperativa La Felicidad, or Happiness Co-operative. Here the whole community now grows mucuna in its fields and has begun a journey across a new frontier of thought - towards settled and sustainable agriculture.

I asked one farmer, Gabino Leiva, about the 'bean manure', as they call it. He said: 'The bean manure destroys the weeds, the beans simply kill them, and all the crops flourish much more. This is what we all need to do - manure our soil for increased production.'

It is technically easy. Improve the soils through low-cost, environmentally sound methods, and we can save the remaining rainforests and reclaim the land for the people who live there.

Photo: Jules Pretty

Fundamental principles
For most of our history the daily lives of humans have been played out close to the land. Since our divergence from apes, humans have been hunter-gatherers for 350,000 generations, then agriculturalists for 600, industrialized in some parts of the world for 8-10, and dependent on industrialized agriculture for just 2 generations.

The conventional wisdom is that, in order to increase food supply to feed the hungry of the world, we need to redouble efforts to modernize agriculture. After all, it appears to have been successful. But there are major doubts about the capacity of such systems to produce the food where the poor and hungry people live, let alone to continue high levels of production in industrialized countries without harmful effects. Farmers worldwide need low-cost and readily available technologies to increase food production.

This is where sustainable agriculture comes in. A more sustainable farming seeks to make the best use of nature's goods and services while not damaging the environment. It does this by integrating natural processes such as nutrient cycling, nitrogen fixation, soil regeneration and natural enemies of pests into food-production processes. It also minimizes the use of non-renewable inputs that damage the environment or harm the health of farmers and consumers. It makes use of the knowledge and skills of farmers, thereby improving their self-reliance, and it seeks to make productive use of people's collective capacities to work together to solve common management problems, such as those associated with pests, watersheds, irrigation, forests or agricultural credit.

Sustainable agriculture functions in many different ways within landscapes and economies - it produces food and other goods jointly for farm families and markets, but it also contributes to a range of public goods, such as clean water, biodiversity, carbon sequestration in soils, groundwater recharge and flood protection.

As sustainable agriculture also seeks to make the best use of nature, so technologies and practices must be locally adapted. This is most likely to emerge from new types of social relations based on trust. Human abilities in the form of leadership, ingenuity, management skills and capacity to innovate are vital inputs for agricultural sustainability. Agricultural systems with high levels of social and human assets are more able to innovate in the face of uncertainty.

Does sustainable agriculture work?
These are all fine ideas, but can they work in practice? At the University of Essex we recently completed the largest-ever survey of sustainable-agriculture improvements in developing countries. The aim was to audit progress and assess how such initiatives, if spread on a much larger scale, could feed a growing world population that is already food-insecure.

Photo: Jules Pretty

We looked at more than 200 projects in 52 countries. We calculate that almost nine million farmers are using sustainable-agriculture practices on about 29 million hectares, more than 98 per cent of which emerged in the 1990s. These methods are working particularly well for small farmers, as about half of those surveyed are in projects with an average of less than one hectare per farmer, and 90 per cent where farmers have less than two hectares.

We found improvements in food production are occurring through one or more of four different mechanisms.

The first involves intensifying a single component of the farm system, with little change to the rest of the farm. It might mean raising the productivity of home gardens with vegetables and/or tree crops, planting vegetables on rice bunds (embankments), the introduction of fish ponds or a dairy cow.

The second entails adding a new productive element to a farm system, such as fish or shrimps in paddy rice, or agroforestry (where trees are grown with crops as an important crop themselves), which boosts food production and/or income.

In the third, better use of nature increases total farm production. Examples of this might be water harvesting and irrigation scheduling or the reclamation of degraded land. These can lead to new dryland crops and increased supply of water for irrigated crops, allowing more intense cropping.

The fourth involves improving yields of staple crops by introducing new regenerative elements into farm systems, such as legumes or integrated pest management - or perhaps new locally appropriate crop varieties or animal breeds.

What is happening to food production? We found that sustainable agriculture has led to an average 93-per-cent increase in per-hectare food production. The relative yield increases are greater at lower yields, indicating greater benefits for poor farmers, and for those missed by the recent decades of modern agricultural development. The increases are quite remarkable, as most agriculturalists would be satisfied with any technology that could increase annual productivity by even one to two per cent.

It is worth emphasizing: these projects are seeing a close-to-doubling of productivity over several years - and this still underestimates the additional benefits of intensive food production in small patches of home gardens or fish ponds.

Photo: Jules Pretty

Changing whole systems
What we do not yet know is whether a transition to sustainable agriculture, delivering greater benefits at the scale occurring in these projects, will result in enough food to meet the current needs in developing countries - let alone future needs after continued population growth and adoption of more urban and meat-rich diets.

But what we are seeing is highly promising. There is also scope for additional confidence. The evidence indicates that productivity can grow over time: as people learn and share better techniques, there are stronger social structures supporting this kind of agriculture and the soil itself gradually gains in fertility.

The issue of asset accumulation is important. If agricultural systems are low in natural, social and human assets then a sudden switch to 'more sustainable' practices that rely on these very assets will not be immediately successful - or at least not as successful as they might be. But, given time, they do work.

In Cuba, for example, newly established urban organic gardens produced 4,000 tonnes of food in 1994. After just five years this had grown to more than 700,000 tonnes, partly because of an increase in numbers of gardens, but also because the productivity of those gardens has steadily risen over time.

Sustainable systems become more productive when human capacity increases, particularly in the form of farmers' capacity to innovate and adapt. Sustainable agriculture is not a concretely defined set of technologies, nor is it a simple model or package. It needs to be conceived of as a process for social learning. Lack of information on agro-ecology and the skills needed to manage complex farms is a major barrier.

Photo: Jules Pretty

We know much less about these resource-conserving technologies than we do about the use of external inputs in modernized systems. How farmers learn about technology alternatives is vital. If they are coerced then they may only adopt them for a limited period. But if the process is participatory and enhances farmers' ecological literacy, then the foundation for redesign and continuous innovation is laid.

As Roland Bunch and Gabino López put it, writing about central American agriculture: 'What needs to be made sustainable is the social process of collective innovation itself.'

Successes are still regrettably in the minority. But time is short and the challenge is simply enormous. We must now take seriously the opportunities offered by this revolution in agricultural and food systems. There is already promising evidence that it can work. We should ask: what would be achieved if we all realized another future was possible? The state of the world and its communities is at stake.

Sustainable agricultural and food systems can right many wrongs, but salvation will not come from these sources alone. Ultimately, if there is to be change centred on both individual transformations in thought and collective changes in action, then it is a question of politics and power.

When governments or policy-makers hesitate, those who believe in this vision will have to organize and show the strength that comes from collective will. Most people have something to gain by supporting a sustainable agricultural revolution. Some will feel they have too much to lose. Others will progress beyond the boundaries that limit our thinking about what is possible. There is nothing to suggest that only certain sorts of people are able to make such a transformation. Equally, there is nothing to say that everyone will.

This millennium offers all kinds of opportunities to make the future sustainable for life on earth. Putting the culture back into agri-culture, the wonder and magic back into nature, and connecting up food systems around agro-ecological, knowledge-based and community-oriented principles can, I believe, help in these wider transformations.