Earlier this week, I asked plant pathologist and molecular biologist Doug Gurian-Sherman to explain some of the science behind genetically engineered crops and their potential — or lack thereof — to feed a more populous, climate-changing world. Today, President Barack Obama is unveiling a $15 billion plan for food security in Africa. It's a plan that appears to contain a number of good elements, but it also promotes genetic engineering as a significant tool to help poor farmers produce more food and lift themselves out of poverty. [If you haven't already, take a minute to read and possibly sign the petition being circulated by CREDO Action — the activist arm of Working Assets — titled "Tell your senators Monsanto can't feed the world.]
In the last post, Gurian-Sherman covered why relying on GE technology is risky scientifically. So what should we be asking Congress to fund? Here, Gurian-Sherman outlines which existing alternatives might work better, from a biological standpoint.
The Ethicurean: Are there studies that compare the performance of a system using genetically engineered seeds and the inputs they're designed to work with, like Roundup herbicide, against the performance of other production methods?
Doug Gurian-Sherman: The UN Environment Program and the UN Development Program jointly sponsored a report [PDF] in 2008 that looked at 114 different studies in Africa using organic or near-organic methods and found that on average, crop yields increased by 116% — or more than doubled — after farmers started using these methods. Although these studies did not directly compare genetic engineering's performance, the yield increases from organic are much higher than has usually been reported for GE food crops in Africa.Can you explain some of the science behind why organic or near-organic methods seem to work so well in developing countries?
Sure. One production method that's been effective in dealing with several pests is called "push-pull." It's based on agroecological principles.
Imagine that you're growing a crop — maize, say — but the maize gets attacked by stem borers. What the pioneers of the push-pull method found was that plants produce all kinds of signals that insects use to find the plants. These signals are often volatile chemical substances produced by the plant. Stem borers are attracted to maize because of these signals. But there's a leguminous cover crop called desmodium that pumps nitrogen into soil and also sends a signal that repels stem borers. So if you grow it in rows between corn plants, it enriches the soil and repels stem borers. That's the "push" part of the equation — pushing the pests away.
"Pull" refers to pulling the pests to an area away from the crops you're trying to grow. So in the maize example, around the perimeter you'd plant an attractant for stem borers — for example, napier grass. In addition to sending signals that attract stem borers, some napier grass produces a sticky sap. When the stem borers lay eggs on the grass, the larvae get stuck in the sap and die. Napier grass is also really good fodder for animals. It can be harvested and used to feed cattle.
Researchers also found that desmodium fights one of the biggest weed problems in Africa: a parasitic plant called striga. Striga weeds attach themselves to corn or sorghum and drain the crops of energy. Desmodium causes striga seeds to germinate prematurely. So this is a system that does a whole bunch of good things at once.
When farmers begin using the push-pull method, their maize yields typically more than double. Compare that to genetically engineered Bt maize, which also targets stem borers: you might get anywhere from 20-40% yield increases with Bt, but that's less than half of what you get from push-pull, plus farmers won't see most of the other benefits. With push-pull you're also building soil fertility: you don't have to use a lot of synthetic nitrogen fertilizer because the leguminous cover crop does it naturally, and you have fodder for livestock as well as a corn crop. In general, with this system I'd predict that pest problems will be less than with Bt. No system is perfect, and there are limitations and problems that can arise with the push-pull system, but it shows a lot of promise.
So we need to put in context at what farmers get from the most successful GE technologies — those few that have been commercialized after 20 years. They simply don't compare well to knowledge-based systems that are biologically sophisticated.
Can you explain what you mean by "knowledge-based systems"?
Knowledge-based systems are based on an understanding of the biology and ecology of cropping systems, and how we can optimize them. Instead of looking at one piece in isolation, such as the harm caused by one insect or disease, the impact of the cropping system on the agricultural environment, including various pests, are examined together. They limit expensive "inputs" — the things such as synthetic fertilizers, pesticides, and seeds that farmers must buy each year. To the extent that knowledge-based methods like push-pull work and work well, they have real advantages over methods where farmers need capital and have to go into debt, if they can get access to credit at all, in order to produce food. And if crops fail, as they can anywhere in the world, indebted farmers are in a worse situation.
Knowledge-based agroecological systems have the advantage of being inexpensive. They usually use inputs like leguminous cover crops to add nitrogen to the soil, prevent erosion, and boost organic matter, and they rotate crops from year to year to keep pest levels down. That's a system based on understanding the biological interactions: you use knowledge rather than having to buy things.
It's a systems approach rather than a piecemeal one, as genetic engineering is — and even conventional breeding may be. Many of these knowledge-based systems do a number of things at once: they empower women (women do much of the farming in developing countries); they bring income into the family; they build the soil. When you build soil, the organic matter in the soil retains water better, so studies have shown that under drought conditions, crops grown in well-managed soil often have yields substantially higher, sometimes even double that of conventionally grown crops.
So agroecological farming systems have shown to be successful in drought conditions. What about the insect and weed problems that genetically engineered crops are supposed to be so good at combatting?
Well as I explained above in the push-pull example, knowledge-based farming methods do address pest problems, and usually not just a single pest problem like GE crops do. Also, remember that pests are more likely to become harmful when you grow the same thing year after year. If a disease organism or insect gets into the system, it can over-winter in the crop residues. For example, let's look at one of the worst insect problems in U.S. corn, for which most insecticide is used, and that the second type of Bt was developed to control: several species of beetle larvae called corn rootworms. If farmers were to grow several different crops in rotation, rootworm wouldn't be a big problem because the eggs, laid in corn, hatch in another crop where the larvae can't survive.
Weed seeds and pathogen spores also will build up over time in single crop monocultures. Growing the same crop over and over again is not biologically sound. Short rotations like the prominent corn-soy rotation in the Midwestern U.S. aren't much better.
If instead farmers use good organic methods, or nonorganic methods guided by good biology, they reduce weeds, insect pests, and diseases. Most pests are not adapted to live among or feed off of a lot of different kinds of crops, especially if the crops are not closely related. A pest that may feast on beans may not utilize a grain at all. When you rotate crops, it breaks the cycle of pest buildup. Weeds, pathogens and insects will be reduced. It's not perfect, but you can reduce problem organisms and encourage beneficial organisms. And you're able to do that, to achieve wide-ranging effects, because the system you're using is biologically based.
The approach of genetic engineering is to attack one issue — one disease — when inevitably there are a dozen different major problems for any of these crops. Farmers in some parts of Africa may see major insect pests, diseases, drought stress, soil problems, and different weeds all at the same time. GE technology deals with one problem at a time. And it's expensive.
I know that your work will be informing congressional debates over Obama's new development program. What should policymakers consider when they're thinking about how to spend that $15 billion? Are there specific questions they should be asking about any technology earmarked to boost development and food security in Africa?
There are a lot of different approaches to agricultural technology. Some of the issues that are important to consider are: Who does the technology favor — is it scale-neutral? Can it help poor farmers? Even if it can, can it disadvantage them compared to big farmers? How expensive is the technology, how easy is it to implement? What do the farmers themselves want? And a big question, of course, is how well technology works. So broadly, we should be asking, what does it accomplish and how expensive is it to implement?
There are limitations to all approaches, but with limited resources it is especially important to use the money wisely — and that means not only asking if something can work nominally, but how it stacks up against other methods and technologies, and how well it works in practice.
I'd like to add that often some of the best policies are completely unrelated to technology or even agricultural methods, but have to do with land reform that gives the landless and disenfranchised access to decent farmland that they were often dispossessed of in the past.
What's your sense of the Obama Administration's position on genetic engineering to date?
The overall direction is hard to know, but so far we have not seen evidence of any critical evaluation of the technology — the kind of comparative analysis that I am suggesting here. That kind of analysis may still find some role for genetic engineering, but as the UN's recent IAASTD report [summary, PDF] concluded, that role should be a limited one for a host of reasons.
We have also not yet seen evidence that this administration will regulate genetic engineering any more effectively than did the Bush Administration. For example, a major revision of USDA regulation that is currently in the works would further weaken already weak oversight of environmental impacts of genetically engineered crops. We [the Union of Concerned Scientists] and others have asked that the proposed rule be scrapped and that the USDA to start over, and there is still time for that to happen.
The decision on that rule will tell us a lot; if the Obama Administration continues a carte blanche policy for the biotech industry, it is not moving in the right direction. The actions of the State Department will also be telling. Will our government emphasize what has the best chance of doing the most good — such as improving infrastructure, agroecological methods, and access to land — or policies that favor a technology like genetic engineering that has accomplished little but is favored by multinational agribusinesses? If the administration follows a course of favoring genetic engineering, shifting the administration's stance will take a lot of outside pressure.