INTRODUCTION

Genetically modified organisms (GMOs) are a fact of modern agriculture, and are here to stay. GMOs are also a fact of public preoccupation and opinion, which politicians must takeinto account. FAO recognizes the great potential and the complications of these new technologies. We need to move carefully, with a full understanding of all the factors involved.In particular, we need to assess GMOs is terms of their impact on food security, poverty,biosafety, and the sustainability of agriculture. Will GMOs increase the amount of food in theworld, and make more food accessible to the hungry? Clearly, GMOs should be seen not inisolation as technical achievements. Hence, I will discuss not the specifics of GMOtechnology, but the context in which they are developed and deployed, and about how public opinion and government policy on GMOs are formed.
The public in many countries distrusts GMOs. They are often seen in the context of globalization and of privatization and even as “antidemocratic” or “meddling with evolution”. There are as yet few perceived advantages for the public, because GMO applications to date have concentrated on reducing costs for producers without direct consumer benefits. In particular, it has been a tactical error of the industry to concentrate on pesticide-resistance as one of the earliest applications, as this has stimulated environmental concerns. The public often confuses the industry with the science. And consumers worry about risk, not about scientific freedom.Scientists in both the private and public sectors clearly see genetic modification as a major new set of tools. They are also participants and spectators in a major shift of research from the public to the private sector, which will undoubtedly influence the future direction ofresearch and research investment. As shareholders in the GMO debate, scientists must recognize that there is also a substantial public distrust of science. Obviously, the industry looks at GMOs as opportunities for corporate profit yet, at the same time, recognises that public acceptance may be a stumbling block. In turn, governments often lack coherent policies in relation to GMOs, and have not yet developed and implemented adequate regulatory instruments and infrastructures. As a result, in most countries, there is no 2 consensus on how biotechnology and GMOs in particular can focus on the key challenges of the food and agricultural sector. Governments need to be more proactive in addressing these questions and, in this, the role of scientists in public service will be crucial.

Definition of Genetically Modified Crops (GMOs)

Before I go more in depth about the pros and cons of genetically modified crops, let me first give a definition of what genetically modified crops are.
Genetically modified crops (often abbreviated as GMOs) are simply crops, whose genetical material has been modified. There are two ways to do this: 1. Traditional selection and breeding (much like breeding animals), 2. Modern, scientific modification of the crops.
In this article, my strengths and weaknesses of genetically modified crops list will deal with the second, the scientific effects and applications.

Process of Genetically Modifying Crops

As the scope of this article is not to describe the detailed process of how GMOs are modified, I will just very briefly describe it.
First and foremost the genetic material of the two or more crops whose genetic property or properties will be mixed has to be fully mapped. The phrase "genetic mapping" means to have a full and exhaustive recorded knowledge of the genes, and the sequence of genes of the genetically mapped organism(s).
When each of the genes (and their functions) of the particular crops have been identified, they are then separated in a science lab. These genes are then cloned and injected into the sequence of genes embryonic form (sometimes to stem cells) of the recipient crop. Finally the seed of the modified crop is planted and grown in greenhouses through traditional methods.
Risks to Natural Biodiversity

There are many genetic transformations in crops, such as altered starch, oil, and fat content, which will probably have little or no adverse impact on biodiversity. Most of the present generation of GM crops carries transformations for the insertion of genes for herbicide tolerance and insect resistance into existing crop varieties. My comments will therefore focus on the genetically modified herbicide tolerant (GMHT) and genetically modified insect resistant (GMIR) crops which are closest to commercial use in Europe, but are being used commercially now over some 40 million hectares worldwide.

Gene Flow and Transfer of Traits to Other Species

Recent research confirms that genes introduced into some genetically improved crops will spread into related native species (Chevre and others 1997). Gene transfer is almost inevitable from crops that have interfertile relatives in adjacent natural ecosystems, but not from crops such as the maize and cereals grown in Europe, whose closest relatives are on the other side of the oceans. Should we worry about this? After all, genes have been moving for many years from “conventionally” bred crops to wild relatives; for example, in the UK hybrids occasionally occur between oilseed rape (Brassica napus) and native species like wild turnip (B. rapa) (Raybould and Gray 1993; Department of Environment, Transport and the Regions 1999). The difference of course is that genes inserted into GM crops are often derived from other phyla, giving traits that have not been present in wild plant populations, and if introduced accidentally, may change the fitness and population dynamics of hybrids between native plants and crops, eventually backcrossing into the native species and becoming established. So the issue is not so much the rate of gene flow (on which there has been copious research), rather the impact that this might have on agriculture and biodiversity (on which there has been almost no research). Conventional plant breeding, using mutagenesis and embryo rescue techniques, also produces lots of completely new genes in crops, about which we know very little. Interestingly, these are often the very crops being used by organic farmers and being sold as “natural foods”! Most geneticists would argue that most “foreign” genes introduced into crop/native hybrids would in fact decrease their fitness in the wild, leading to rapid selection of these genes out of the population. This is particularly true of genes designed to prevent germination of saved seed, like the so-called terminator gene - if this were to “escape” it would commit instant suicide and certainly not spread into the natural world as has been suggested by some anti-GM campaigners. There is no difference to the farmer between buying seed with terminator technology and buying hybrid seed, because neither can be saved and grown next year. There is a serious issue about whether farmers in the developing countries should become locked into a cycle of dependence on patented seed, but the genetics of this technology is not a direct environmental threat(see Pinstrup-Andersen and Cohen, This volume). Transfer of certain genes, such as resistance to insects, fungi and viruses could increase fitness (ability to reproduce) of any resulting hybrids, possibly forming aggressive weeds or plants that swamp wild populations. Weeds having tolerance to a range of herbicides could also emerge; these would be difficult to control in agriculture, or in natural ecosystems like grasslands. Farmers may eventually need mixtures of herbicides to control them, causing yet more damage to biodiversity.
There is already evidence from North America that this “multiple tolerance” and resistance to herbicides is beginning to emerge(see Cook, This volume). If nontarget plants acquired insect resistance from GM crops, they could damage food chains dependent on insects feeding on previously nontoxic wild

Genetically Improved Crops and Agricultural Intensification

The prospect of gene transfer causes concern for crops that have wild relatives in the same ecosystem, and occupies reams of headline comment in the press. Perhaps of greater importance is the fact that management of some genetically improved crops would be very different from conventional intensive agriculture or organic farming.
In the United States, genetically modifed herbicide tolerant (GMHT) crops are grown under a regime of broad-spectrum herbicides applied during the growing season. Farmers report almost total weed elimination from GMHT crops, which include cotton, soybean, maize, beet, and oilseed rape.They also report substantial reduction in herbicide use (see Pinstrup-Andersen and Cohen, This volume). Recent research in the United Kingdom confirms that weed control in GM beets and other GMHT crops is likely to become much more efficient (Read and Bush 1998). These results are hardly surprising since this is the main purpose behind the technology. This GMHT system will soon be available, at least experimentally, for virtually all mainstream agricultural crops, including vegetables. Broad spectrum herbicides used on commercial scale GMHT crops during the growing season may be far more damaging to farmland ecosystems than the selective herbicides they might replace. Using these herbicides in the growing season may also increase the impact of spray drift onto marginal habitats such as ancient hedgerows (field margins common in Europe) and watercourses. It is not only the volume of herbicides that is the issue but their efficiency and impact on wildlife. When insect resistance and herbicide tolerance are combined in the same crop variety, there may be few insects capable of feeding on the crops and few invertebrates and birds would be able to exploit the weed-free fields. In Europe we already have massive declines in farmland birds, with several previously common species now close toextinction. The problem with assessing the environmental impact of these changes in management is that the regulatory system and the public has very little scientific data on which to assess the real risks, and potential benefits, from adopting GMHT crop systems. Formal risk assessments submitted by the biotechnology companies as part of the regulatory process deal with this issue inadequately. In the United Kingdom, the Department of Environment, Transport and the Regions and the Ministry of Agriculture, Food and Fisheries have realized this, changed the regulatory system, and commendably have started some field-scale experiments to try to answer some of these important questions. The development of new crops with improved tolerance to abiotic factors (such as drought, salinity, frost) and the potential advent of ‘pharmed’ crops producing vaccines and GM biomass systems, may also change crop management, perhaps increasing demand for arable land in the long term, and putting further pressure on natural biodiversity on marginal land. Agricultural Intensification and Declining Wildlife If we want to make predictions about how intensification enabled by GM crops could affect biodiversity, we can turn to evidence of declines in farmland plants, insects, and birds resulting from agricultural intensification in Europe over the past 30 years. Factors responsible include abandoning traditional crop rotations, increased pesticide efficiency and drift, use of artificial fertilizer, drainage, and intensification of soil cultivation(McLaughlin and Mineau 1995). There is overwhelming evidence demonstrating that the use of more effective pesticides (including herbicides) over the past 20 years has been a major factor causing serious declines in farmland birds, arable wild plants, and insects. Pesticides not only have direct toxic effects on wildlife but they also enable modern crop management changes to take deliver early warning of dangers in crops or the place. Winter-sown crops, for example, rely heavily on effective fungicides. Thirty years ago winter sowing was unknown in the United Kingdom and winter stubbles were widespread, providing an essential food source for wintering flocks of birds. There are many examples of declines in farmland wildlife in the UK and these are typical of intensively managed farmland throughout Europe. It is important to remember that although these declines in biodiversity have been severe in many intensively managed areas, there are still viable populations of many farmland-dependent species throughout Europe. Some of these, however, are only just surviving the impact of intensive agriculture. Twenty-five of the 200 species of British “arable plants” are now “Nationally Scarce” and a further 24 are “of conservation concern” and included in the 1983 IUCN Red Data Book (RDB) (McLaughlin and Mineau 1995; Wilson 1994). Not only have many arable plants become threatened but there has also been a marked shift towards a less diverse, grass-dominated flora (Kleijn and
Snoeijing 1997). More effective herbicides are responsible; similar trends have been observed elsewhere in Europe (Eggers 1984; Andreasen, Stryhn, and Streibig 1996; Wilson 1992, 1994). Changes in herbicide practice have also been a major factor in reducing the distribution of insects such as the common blue butterfly (Aspinall 1988), the larvae of which feed on broad-leaved weeds.
Over half of British farmland birds are now in serious decline and 13 are red-listed (Siriwardena and others 1998). The 78 percent drop in grey partridge (Perdix perdix) numbers observed in the United Kingdom between 1972 and 1996, has been directly attributed to increased herbicide and pesticide efficiency. Skylark (Alauda arvensis) populations have declined by 75 percent over this period mainly due to increased pesticide efficiency (Campbell et al. 1997). Recent research implicates agricultural intensification in the decline of other songbirds(Ewald and Aebischer 1999). Besides the aesthetic and scientific reasons for conserving biodiversity within and around agricultural crops, there is another important utilitarian reason for wanting to do so. This is the need to maintain the food chain links between native species and crop systems. This link is vital if we are to preserve the function of biodiversity to chemicals used to manage them. Without these links, we are unlikely to be able to detect any dangers arising from the new agriculture by monitoring wildlife; toxicity of DDT and aldrinbased organochlorine pesticides (Sheail 1985) and showed up the potentially lethal effects of PCBs before toxic levels built up in humans. This is not just an issue for the industrial countries. It is a natural alarm system which is probably the mostthe first organism in the food chain will increasingly be Homo sapiens. This “natural early warning system” has served agriculture and the public very well over the past 50 years. It detected the cost-effective way of monitoring environmental safety in developing countries. We abandon this biological system at our peril.

Virus Resistant Crops
Some viruses infect people or animals and other viruses infect plants. Plant viruses reduce the productivity of annual crops and can kill fruit trees.
Some plant viruses are spread by insects. Plants can be protected from those viruses by using insecticides or other pest management methods. There is essentially nothing else that a farmer can do to protect his crop from virus damage, except to grow a different crop. But genetic engineering a plant to protect it from a particular kind of virus is quite easy. A gene from the virus which encodes a protein in the virus' outer coat is copied into the plant's DNA. The plant then makes the coat protein, which is harmless, but which stimulates the plant's natural defenses. Virus resistance traits have been introduced into many crops, including squashes, tomatoes, potatoes, tobacco and, perhaps most dramatically papaya.

The Potato Famine
In 1840s Ireland, the potato crop was devastated by a late blight fungus (Phytophthora infestans) and Irish people starved en masse. That fungus could reappear at any time in any place and wipe out a potato crop. Some varieties of potato have previously had some resistance to late blight fungus, but now a fungal strain has appeared in Russia that destroys those previously resistant varieties. This year (2001) a similar fungus appeared in potato fields in Prince Edward Island and 630 million pounds of potatoes, the island's principal crop, had to be destroyed. But very recently, scientists were able to transfer a gene from alfalfa to a potato plant and the resulting potato plant is able to resist the fungus and thrive.
Potatoes also rot. A principal cause of potato rot is the bacterium Erwina carotovora, which has been called the flesh eating bacteria of the plant kingdom. Now a gene that confers resistance to E. carotovora has been coupled to a control gene that turns on when a plant has been wounded, and this construct has been transferred to experimental potatoes. As the researchers hoped, the modified potatoes, when punctured by a toothpick and exposed to E. carotovora, had almost twenty times less rot than unmodified potatoes.

Slow Ripening Fruits
There are many fruits which ripen after picking. After they reach optimum ripeness, they begin to deteriorate. This is necessary for the life cycle of the plant, which relies on the sweet and pulpy parts to nourish the seeds. A ripe fruit literally digests itself.
When this process is rapid, it effectively means that the fruit cannot be enjoyed out of season, or far from its growing area. For example, there is a popular Malaysian papaya variety which is unavailable outside Southeast Asia because it ripens so rapidly that it cannot be shipped very far. But it is quite easy to genetically engineer a fruit so that it does not ripen so rapidly. It doesn't even require a gene from another organism. Instead, a gene involved in the ripening process is copied with the message in reverse order. So now that plant has two genes with mirror image structure.
The way an organism uses the information in a gene to make a protein involves copying the gene (DNA) onto a messenger molecule, known as messenger RNA. The modified plant copies both the original gene and the mirror image gene to produce both types of messenger RNA. But since these messenger RNAs are exact complements of one another, they can wrap about one another just like the two strands of DNA, effectively blocking both messages. This means that the plant makes very little of the enzyme that causes ripening. This genetic engineering trick is called ``antisense technology''.
The Malaysian papaya was transformed in this way and therefore a slow ripening variety will soon be available.
The very first genetic engineered plant to be commercially developed as a whole food was a slow ripening tomato, called FlavR Savr. It was developed by Calgene, Inc. Because it could remain on store shelves for a long time, it could be left on the tomato plant until optimally ripe, and therefore the FlavR Savr tomatoes sold for a premium compared to other tomatoes. Although consumers initially liked Calgene's tomatoes, they didn't ship well and the variety was eventually dropped.
Controlled Ripening
A coffee bush ripens a few coffee beans each day for many months. The best quality beans must be picked just after ripening, so picking coffee beans is very labor intensive. It would obviously be preferable if the beans would all get ripe at the same time.
Genetic engineering will make this possible. There is a coffee gene which turns on to initiate the last stage of ripening. Scientists modified a control gene so that the ripening gene does not turn on until the plant is sprayed with a triggering substance (patented and sold by the company that developed the coffee variety). Therefore all the beans on a bush reach the same not quite ripe stage and stop to wait for the triggering signal. The farmer decides when to spray the bush so it can be picked completely clean a few days later.
This can substantially improve the life of the small farmer. He can take a short vacation without losing part of his livelihood. He can work fewer hours per day, or he can pick all his crop in a few days and increase his income by working at another job. A large scale farmer would need fewer workers to pick the same quantity of coffee beans, and could afford to pay them a higher wage.
The control of when a crop is harvested would be valuable for other crops besides coffee. For example, the quality of grapes declines rapidly after they reach their optimum sugar content. Grape farmers now have to mobilize every available hand to harvest all their crop in a very short time. Their lives would be simpler if they could spread the harvest effort over a few weeks instead of a few days.
Large scale crops are harvested with special equipment. A farmer would not need to own a combine if he could rent it for the few days it was needed. But that wouldn't work if his neighbor needed to rent it for those same few days. If neighboring farmers could control when their crops become ready for harvest, they could share scarce and expensive equipment.
Saving the Banana
Wild bananas have seeds. They reproduce sexually, like beans and oak trees. But they aren't easy to eat. Bananas grown on plantations have no seeds. They are reproduced by taking cuttings from older banana plants.
Cultivated bananas are seedless because they have three of each type of chromosome instead of the normal two of each type. Such plants are called ``triploid''. They are always sterile. Genetic triploid freaks arise from time to time in nature, but modern breeders can also use chemicals or electric shocks to create triploid mutant cells.
Bananas have been cultivated for many thousands of years and there are about three hundred different banana varieties. Each variety was developed by crossbreeding wild bananas. Whenever a promising variety was been produced, the breeder caused it to be triploid, hence seedless. That plant was cloned by propagating cuttings, and it became the parent of its variety. All bananas plants of the same variety are genetically identical, like identical twins.
Of the three hundred varieties, only one single variety completely dominates international trade. It is called Cavendish. It is possible that you have never seen a banana other than a Cavendish banana.
Certain kinds of fungus can infect and kill banana plants. In many parts of the world, Cavendish banana plants are being attacked by a fungus called black Sigatoka. Since wild bananas can reproduce sexually, they are not all identical and some wild bananas can resist black Sigatoka. If the black Sigatoka fungus is present in a region, the resistant banana types become prevalent. But the Cavendish banana plants have no resistance. They are all identical so they all die.
To grow bananas commercially, growers must spray their plants with fungicides. Year after year, the black Sigatoka fungus has been evolving resistance to these fungicides, so growers have to spray more and more fungicide each year. Approximately one third of the cost of raising a banana is the cost of spraying it with fungicides, and it gets more and more costly each year.
A form of black Sigatoka banana disease, now spreading around the world, can tolerate all known fungicides. Soon it will attack bananas in Central America and the Carribean islands, the heartland of banana culture. The Cavendish banana will become virtually extinct. Agronomists estimate that this will happen within ten years.
This is not a fairy tale. It has happened before. Forty years ago, the most popular variety of banana was one called Gros Michel. But Gros Michel was susceptible to a fungus called ``race 1 Panama disease''. Now it is gone. Cavendish bananas, which are not susceptible to race 1 Panama disease, replaced them. (A related fungus, race 4 Panama disease, can infect Cavendish banana plants. At present it is only found in Malaysia and some nearby countries.) In turn, some other variety of banana could replace the Cavendish. It would look different and taste different but it would still be a banana.
There is only one practical way to save the Cavendish banana. It must be given some combination of genes from wild bananas which are not susceptible to the fungus. But this can't be done by any natural technique.
Cross-breeding can create other banana varieties because wild bananas reproduce sexually. But Cavendish bananas reproduce only by cloning. Conventional breeding would have to rely on rare mutated Cavendish banana plants which can produce seeds and which can therefore be crossbred, in theory. Even the mutated plants produce only a tiny number of viable seeds, as few as two or three seeds in a hundred pounds of bananas. No banana breeding experts think that they can breed fungus resistance into a Cavendish banana variety in only ten years.
Biotechnology can rescue the Cavendish banana. Scientists in Belgium have used genetic engineering to transfer some fungal resistance genes from wild bananas. The transformed Cavendish plants are not susceptible to the fungus and they can then be reproduced into nursery stock by the usual method of taking cuttings.
The Eggplant in Winter
The edible part of an eggplant is formed from the ovary of its flower. In this way, it is like the edible flesh of an apple, a pepper or grape. When we eat these fruits, we discard the seeds. But the plants only transform their ovaries into fruits when they start to produce seeds, although in the case of an eggplant, its seeds are so tiny that we ignore them. Eggplants will only set seeds in warm weather, so to grow them in the winter in an unheated greenhouse, the grower must use a chemical to trick the plant into beginning fruit development without setting seed. Such fruits do not grow very large or very fast under these conditions. So eggplants are expensive in the winter.
But now scientists in Italy have transferred two genes into a variety of eggplant, which not only allows the plant to set fruit in cool greenhouse conditions without chemicals, but also increases productivity of the same plant in either hot or cold weather.
The eggplant variety that the Italian scientists created is seedless. One of the two transferred genes is a switching gene which is turned on only in the ovary part of a flower. That gene turns on the other transferred gene, which makes a protein involved in synthesizing a growth hormone. The growth hormone makes the ovary grow into the fruit, just as it would have done in a traditional eggplant making seeds. Neither gene requires either seed setting or warm weather.
Where does one get seeds to produce large numbers of seedless eggplants? The transformed plants produce pollen, so they can be crossed with traditional eggplant varieties and the hybrid produced by that crossing has the seedless and self-starting property.
The scientists report productivity increases of 37% for the new variety, and they believe that the seedless type would be more marketable.
Most policy and regulatory activities regarding GMOs are national. International agreements have a role to play in harmonizing national regulatory systems, for example on transboundary questions, and when trade standards are involved. International agreements can also help governments, particularly the smaller governments, to rationalize their national capacities and investments through common action. In relation to GM crops, the major instruments concern risk-assessment and management in two areas: food and health; and the environment. GMOs in food and agriculture have been around long enough for us to have some background against which to assess possible risks. There has been as yet no proven effect on human life. Moreover, a recent study of GMO releases into the environment in Britain showed that they had not survived in nature, as, in fact, they only would if they had an evolutionary advantage, which remains an important possibility. But let me say that state-of-the-art knowledge may not always be adequate to assess the risk of GMOs to health or the environment. Of course, lack of evidence of adverse effects is not the same as knowledge that genetic modification is safe.

In FAO we deal with environmental and food risk-assessment questions under a more general concept of “biosecurity”, namely the application of sanitary and phytosanitary measures (SPS) and a methodology of risk analysis for food and agriculture, including fisheries and forestry. They include:
♣ the protection of animals or plants from pests, diseases, or disease-causing organisms;
♣ the protection of human or animal life from risks arising from additives, contaminants, residues toxins or disease-causing organisms in foods, beverages or feedstuffs;
♣ the protection of human life or health from risks arising from diseases carried by animals, plants or plant products, or from pests; and
♣ the prevention or limitation of other damage from pests.

Moreover, FAO has specific standard-setting responsibilities that relate directly to GMOs under the Technical Barriers to Trade (TBT) and the Sanitary and Phytosanitary (SPS) Agreements of the World Trade Organization. There are several important international mechanisms in the context of GMOs. Firstly, the Cartagena Protocol to the Convention on Biological Diversity, which was adopted in January 2000. It will come into force after fifty ratifications and is the main international agreement in relation to GMOs (“LMOs”, or “Living Modified Organisms”, in the language of the protocol). It covers the transboundary movement, transit, handling and use of allLMOs (except pharmaceuticals) that may have adverse effects on the conservation and sustainable use of biological diversity, taking also into account risks to human health. It allows for standard-setting in relation to the handling, transport, packaging and identification of LMOs. Secondly, food safety aspects of Genetically Modified Organisms are, at international level, dealt with by the FAO/WHO Codex Alimentarius, which covers all aspects of food safety. The Codex is currently working on standards for risk assessment for labelling, and for several other food safety aspects of GMOs. The Codex standards are recognized by the SPS and
TLC Agreements.
Thirdly, the International Plant Protection Convention (IPPC) has the objective of preventing the spread and introduction of plant and plant product pests, including weeds and other species that have indirect effects on both wild and cultivated plants, and to promote appropriate control measures. This also applies to risks associated with LMOs. The IPPC sets International Standards for Phytosanitary Measures (ISPMs), which are also recognized in the SPS agreement. It is also in the process of establishing practical cooperation with the CBD and its Biosafety protocol. An IPPC expert working group met in September 2001, in coordination with experts from the CBD, to develop a detailed standard specification for an ISPM that identifies the plant pest risks associated with LMOs, and ways of assessing these risks.

Access and benefit-sharing

In a world governed by Intellectual Property Rights and concentrated research investments, genetic resources are the raw material to which biotechnology is applied: in agriculture, these are not wild products of nature, but the fruit selection and development by farmers throughout the world since the Neolithic. In contrast to other areas of industry, this poses the immediate question of how to guarantee continued access to those who have been involved before: the farmers and the breeders. Governments are now completing the negotiation, within FAO’s Commission on Genetic Resources for Food and Agriculture, of a new, binding convention, the International Undertaking on Plant Genetic Resources, which we expect to be approved by our Conference in November. The Undertaking will create a Multilateral System of Facilitated Access and Benefit-sharing for the world’s key crops. This is a major step, because dealing bilaterally with such widespread resources would involve unacceptably high transaction costs, and could impede agricultural progress. Multilateral access provides multilateral benefit-sharing, which includes the sharing of the benefits arising from the commercialization of materials from the Multilateral System through a mandatory payment. The access of breeders to genetic material for further breeding– which becomes ever more difficult with GM crops under patents is a public good that needs to be protected. FAO is involved in several instances, such as in the World Intellectual Property Organisation, where concerns of food and agriculture and IPRs are being discussed.

GMOs and scientific responsibility

Science never took place in a vacuum, but now more than ever, it needs to stand up to public scrutiny, to engage with public opinion, and address technical and ethical issues in the application of science to the needs of the poor. The current debate over transgenics and the labelling of foods containing ingredients from GMOs has highlighted the need for transparency. There is growing awareness of the public’s ambivalence towards genetic engineering and its aims. The research sector should advertise the fact that it disposes of an important instrument for quality control. The scrupulous honesty of the peer review process, under the watchful eye of scientific institutions, including scientific academies, the great journals, is a guarantee for quality, if not always for relevance The food and agriculture sector could well take the example of the declaration last week by twelve of the world’s most prominent medical journals of a policy to prevent the corporate sponsors of research exercising control over the analysis, interpretation and reporting of research results. However, the general public is largely unaware of peer reviews and standard setting (which takes place through similar processes) and public trust remains low.

Conclusion

Genetic modification has increased production in some crops. But the evidence we have suggests that the technology has so far addressed too few challenges, in few crops of relevance to production systems in many developing countries. Even in developed countries, a lack of perceived benefits for consumers, and uncertainty about their safety, have limited their adoption. The scale of investment involved, and the attraction of advanced science, may distort research priorities and investment. Genetic modification is not a good in itself, but a tool integrated into a wider research agenda, where public and private science can balance each other. Most of the short term successes may be derived from marker assisted breeding and diagnostics rather than from transgenic crops per se. Steering research in the right direction and developing adequate and international agreements on safety and access is a difficult and responsible task. While we are more aware than ever of the need to manage international public goods responsibly, the political tools to do so are weak, and, in a globalised economy, the voices of small countries and poor producers and consumers often go unheard. I believe that scientists have moral responsibilities to speak for the weaker segments of society, because they sometimes best understand the likely results of not doing so. In the preparation of this text, special thanks go to Clive Stannard, and the contributions of Jim Dargie, Niek van der Graaff and Andrea Sonnino are gratefully acknowledged.

References

Andreasen, C., H. Stryhn, and J.C. Streibig. 1996. “Decline of the flora in Danish arable fields.” Journal of Applied Ecology 33, 619-26. Aspinall, R.J. 1988. “The changing geography of the common blue butterfly in Northumberland.” Entomol.Rec.100, 1-6.

Campbell, L.H. and others. 1997. A review of the indirect effects of pesticides on birds. JNCC Report 227. Chevre, A.-M., F. Eber, A. Baranger, and M. Renard. 1997. “Gene flow from transgenic crops.” Nature, 389 (30 October), 924. Department of Environment, Transport and the Regions. 1999 Investigation of Feral Oilseed Rape Populations: DETR Genetically Modified Organisms Research Report No 12, DETR, London

Eggers, T. 1984. “Some remarks on endangered weed species in Germany.” 7th Internat. Symp. On Weed Biology, Ecology and Systematics: 395-402. English Nature. 1997. Position Statement on Genetically Modified Organisms. English Nature, Peterborough, PE1 1UA, U.K.

Ewald, J. A., and N .J. Aebischer.1999. Pesticide use, avian food resources and bird densities in Sussex Joint Nature Conservation Committee Report, Peterborough, UK. Kleijn, D., and I.E. Snoeijing. 1997. “Field boundary vegetation and the effects of agrochemical drift: botanical change caused by low levels of herbicide and fertilizer.” Journal of Applied Ecology 34,1413-25.

McLaughlin, A., and P. Mineau. 1995. “The impact of agricultural practices on biodiversity.” Agriculture, Ecosystems and Environment 55, 201-12.

Raybould, A. F., and A. J. Gray, 1993. “Genetically modified crops and hybridisation with wild relatives: a UK perspective.” Journal of Applied Ecology 30, 199-219. Read, M.A., and M.N. Bush. 1998. “Control of weeds in genetically modified sugar beet with glufosinateammonium in the UK.” Aspects in Applied Biology 52, 401-6.

Sheail, J. 1985. Pesticides and Nature Conservation, Chapter 7, 103-122, Clarendon Press, Oxford U.K. Siriwardena, G. M., S. R. Baillie, S.T. Buckland, R. M. Fewster, J.H. Marchant, and J. D. Wilson. 1998.

“Trends in the abundance of farmland birds: a quantitative comparison of smoothed Common Birds
Census indices.” J. Appl. Ecol. 35, 24-43. Wilson, P.J. 1992. “Britain’s arable weeds.” British Wildlife 3(3), 149-61. Wilson, P.J. 1994. “Botanical diversity in arable field margins.” BCPC Monograph No. 58: Field Margins - Integrating Agriculture and Conservation: 53-8.