ANNEX A (AEBC/02/17)

SUMMARY OF RESPONSES TO THE LIABILITY GROUP’S SCENARIOS CONSULTATION

1. This paper summarises in some detail the main themes that have emerged from responses to the liability group’s scenarios consultation of 30 September 2002. A list of respondents is at Annex A1, with the consultation paper and covering letter at Annex A2. There has been a substantial response, from a wide range of stakeholders: in industry, agricultural organisations, farmers and NGOs. In addition, people giving evidence to the group on 5 November referred to the scenarios. That evidence is recorded in the minutes of the meeting. (Members of the liability sub-group have seen full copies of all responses by stakeholders.)

2. The order in which responses have been summarised in this paper is: firstly responses about generic questions in the consultation paper; secondly responses about the scenarios paper and its purpose, and about general matters related to liability; and thirdly comments about each of the nine hypothetical scenarios set out by the AEBC group to test the robustness of current legislation.

3. Some of the strongest messages are:

· that the scenarios described are not unique to GM crops, though responses differed substantially in implications drawn for possible commercialisation, and for requirements for liability and other provisions; and

· the need for co-existence and the importance of avoiding conflict between farmers;

There was not much specific analysis of whether a liability regime would be able to provide the answers, and what the relationship of such a regime should be to other instruments such as regulation. This is an area on which the subgroup will need to come to conclusions. (There was an interesting comment about the influence of perception in economic choices.)

Generic questions in the consultation paper.

Are there issues that may arise from the commercial growing of GM crops which are not captured by the scenarios?

4. Following conflicting reports from the USA and Canada of farmers who deliberately or accidentally infringe patent rights, would the AEBC consider the scenario of accidental presence of a patent-protected gene in a farmer’s crop and whether the farmer would be held liable for royalty payment under UK law. (NFU)

5. Several further hypothetical economic impact scenarios are offered, arising from: (1) lack of identifiable harm – health concerns; (2) lack of identifiable harm – insufficient confidence in working practices; (3) lack of identifiable harm – adventitious presence; (4) lack of harm – losses arising from precautionary measures; this is based on the actual decision not to grow maize for medicinal purposes. (A Turner & Sons)

6. The AEBC’s thoughts would be useful on liability relating to non-food crops for industrial use – whether derived by GM or not – building on the current example of high erucic acid oilseed rape (HEAR). (NFU)

7. NIAB offered to discuss additional scenarios.

Are the issues raised by the scenarios unique to GM crops, or might they equally flow from other approaches to crop development, such as through changes in agricultural practice or conventional plant breeding?

8. A number of respondents said emphatically that the issues were not unique (while others did not specifically address this, but made separate comments as in paragraph 17 onwards):

9. Liability issues associated with GM crop technology are no different from those arising from any other agricultural products. Once regulated and approved, GM crop technology introduces no new considerations on liability which are not or could not be addressed through existing legal provisions. (SCIMAC)

10. Any of the proposed scenarios could occur through introduction of crop plant varieties, or altered agricultural practices developed through more ‘traditional’ and unregulated technologies. (Simon Barber EuropaBio) None of the scenarios are unique to GM technology. Liability issues associated with GM crops technology are identical to those in the rest of the plant breeding industry; indeed there are parallels with all industries. Products take many years to research and develop and are costly to take to market. The regulatory regime for agrochemicals and GM seeds is extremely rigorous, with continuous detailed scrutiny before, during, and after registration. It is in the company’s interest to develop high quality reliable safe and beneficial innovative products. (Bayer CropScience Ltd).

11. Many of the issues raised are not unique to GM crops. (NFU, planning inspectorate, NIAB) Issues raised by monoculture are not confined to GM crops; issues relating to groundwater pollution through use of herbicides are not confined to GM crops, though might be exacerbated with GM crops; loss in value of crops can occur with conventional crops eg organic crops becoming contaminated with fertilisers or other chemicals; decisions made by supermarkets can be affected by a host of factors. (Planning inspectorate)

12. Where environmental harm is being considered, issues are not confined to GM crops, but applicable to many other novel techniques, though most changes are considered reversible, unlike for GM. End consumers are likely to tolerate a low level of cross-pollination between conventional and organic wheat much more readily than similar cross-pollination levels between GM and organic wheat. (A Turner & Sons)

Whether unique or not in that sense, how far are current liability rules adequate to impose the environmental costs of the harm on the person who has caused it?

13. Many respondents did not distinguish specifically between economic and environmental liability:

14. Present measures appear totally inadequate to begin to deal with any liability issue beyond that covered by Directive 2001/18. Even establishing evidence of harm would be very subjective and fiercely contested. Even establishing the facts in high profile incidents eg contamination of landrace Mexican corn varieties has met fierce resistance and denials - agreement on harm, let alone remedial action seems impossible within an acceptable time span. The practical implications of a farmer taking action against a large corporation, even with a well-supported case would be inconceivable; the resources needed to mount and sustain an action are far beyond the means of most farmers or even groups of farmers. (A Turner & Sons)

15. A liability regime should not distinguish between different production methods or causes of damage to the environment unless an increased risk of damage can be shown. Similarly, additional legislation is not needed where increased risk has not been demonstrated. (NFU)

16. There is no reason why GM crops should be singled out and a special liability regime created for them. Rules on how to deal with economic damage resulting from actions within a relationship between neighbours is by definition for civil law. There are well-developed, traditional and balanced rules in Member States for this type of dispute. (Simon Barber EuropaBio)

If unique in that sense, need liability rules be changed to reflect that uniqueness, and if so, how? Are there issues about aspects of a regime, such as burden of proof, threshold of liability (eg fault based or strict), foreseeability, costs or otherwise on which you would wish to comment?

17. Many respondents covered this area indirectly. They described where they felt liability should fall, and the type of regime needed. Their assessments follow:

18. There seems no reason why absolute liability should not apply to damage caused by GM crops – it does for fresh water. Introduction of GM crops is already causing significant additional general costs, and these should be borne by the companies profiting from GM technology. There should not be commercial GM growing until UK law has been changed to ensure all farmers, food manufacturers, processors and food retailers who wish to supply non-GM products can take legal action to recover damages for any losses they suffer through no fault of their own from GM contamination. European law prohibits use of GM ingredients in organic food and farming. It would be perverse for any legal regime not to recognise the market realities – the public expects organic food to be GM free to the level of commercially available, verifiable and repeatable testing, which is currently 0.1% (Soil Association)

19. Using the pharmaceutical industry as a model, manufacturers of GM products (seed, seedlings, clones/cuttings etc) would be liable in law, rather than – in that industry - the prescribers, if they use due care. The significant difference is that crops have widespread effect on the environment, while medicines do not, at any rate directly. If Government decided commercial GM planting will have significant economic and ecological benefits, there may be a case for Government to take liability. The public purse could enjoy any rewards. There are serious questions about ownership and control of GM technology, and potential dangers of allowing intellectual property rights to remain in the hands of commercial companies with their own financial interests. (Church of England Board for Social Responsibility)

20. As GM crops increase, potential for litigation is great. Disputes will not all fall into the area of ‘tort’ law, as the scenarios show. Special legislation will be needed to protect all players adequately. There must be no ‘grey’ area between responsibility for release of GMOs and the legal regime. Many situations apart from those in the scenarios could affect individual specialist growers, organic growers, SSSIs etc. Phrases like ’might reasonably be expected’ must be avoided – legislation on scientific issues needs scientific language. (National Council of Women of Great Britain: NCW)

21. In the long-term, organic and GM are mutually incompatible, as recognised by the EU Joint research Council report. If there were a major health problem from GM crops, who would be liable, the multinational producing them, the government that licensing them or the farmers growing them? Like disputes over cigarettes and lung cancer, it might turn on who knew what, when. Two examples of early concerns might be relevant to legal cases if problems emerge later, though no hard evidence now: loss of fertility in sows in Iowa fed large amounts of Bt corn, and use of the PAT gene, which may give a mild dose of weed killer in the gut. (Johnsons farm)

22. If commercialisation proceeds, all associated costs, risks and liabilities should be for the companies promoting GM crops; certainly not for farmers seeking to meet consumer demand for non-GM foodstuffs. (Farm) If, much later, commercialisation is considered, the law would need to be changed. Promoting companies and individual users, knowingly and willingly using the technology, should be liable for damage. Adequate insurance cover should be mandatory. Unforeseeability is one of the technology’s defining characteristics: it is reasonably foreseeable that the damage will occur in unexpected ways. Since common law cannot cover unforeseeable events, an underlying principle of common law is seriously flawed. (Carmarthen Gene Concern)

23. A principle should be established that a farmer who becomes a victim of circumstance is not responsible for the consequences – it is by definition not in his control. If a farmer loses income as a result of GM contamination, the seed company or company that owns the patent should ultimately be responsible for meeting claims for loss of income wherever they occur. It should be possible in each case to identify the particular modification and therefore the company involved. If multiple modifications occurred in one plant, liability should be shared equally between the companies owning the patents. (South Elmham Hall Farms Ltd)

24. The public are wholeheartedly against any GM crop being grown or sold, or seeds patented. Before damage is done to our environment, food or human health, a clear decision on liability is needed. This should lie with the large biotech companies who are going to make a profit from GM crops, not with the public. (R Neary)

25. The majority of farmers want the right to produce GM free or organic produce and need the law to protect their right. The public wants choice. The public and farmers will judge and condemn those they hold responsible for any contamination of the food supply or damage to the environment – the government, the biotech industry or both. Three tenets of ‘natural justice’ apply: the precautionary principle, the polluter pays principle and the burden of proof. Where there is any doubt on safety of a product or organism, the precautionary principle means it should be denied a commercial licence. Public rejection of GM foods is so strong that many would consider GM products a pollutant. The tax payer/consumer will not accept liability for pollution from GMOs, and farmers cannot afford to take on liability. Natural justice does not allow burden of proof to be with the victim. How can a farmer prove which crop contaminated his field; how can a consumer prove which link in the food chain contaminated their food; how can the government prove whose crop cross-pollinated with a wild relative? For all these reasons the patent holder of the introduced organism/product must be liable. (P Lundgren)

26. Before any decisions on commercialisation, there needs to be a clear and unequivocal means to address all of the potential liability scenarios. The greater part of the cost burden, such as those incurred through liability, segregation and regulation, should be borne by the biotech companies, processors, distributors and farmers choosing to grow the crops. (A Turner & Sons)

27. Genetic engineering must be dealt with independently, because the technology is entirely new and bears no resemblance to traditional breeding methods. Regulation of GMOs is still inadequate, even if more tightly regulated than other areas. Regulations on procedures such as disposal of genetically engineered waste are also entirely insufficient. Society has not asked for the technology and clearly does not to want it. Supposed benefits have yet to be proven. The opposite seems true – for example from ‘Seeds of Doubt’ and failure of Indian Bt cotton crops. Before commercialisation is even considered, consumers should be shown conclusive evidence GM will provide significant benefits; and adequate testing should be carried out, with regulation of labelling and scientific practices being tightened. It is far too early to be considering commercialisation. (Carmarthen Gene Concern)

28. There would be a need for protection for farmers who were using due diligence. Potential litigious situations could be minimised by practical codes of practice, to minimise gene flow through cross-pollination and maintain segregation along the production chain. (NFU)

Proposals from GeneWatch UK

29. The scenarios illustrate the difficult legal problems that may be faced in determining causation with multiple actors and complex cause-effect chains. Unforeseen, long-term and diffuse impacts are of primary concern to the public. As GM is new and controversial, demands for liability rules are greater than for existing practices. Clear liability for environmental and economic harm is a pre-condition for acceptance of commercialisation. Likelihood of benefits from the current range of GMOs is strongly contested and social ‘need’ is important. If the AEBC recommends that a strict liability regime including unforeseen impacts is not appropriate, it will have to justify this against a regulatory system which does not consider need, but takes benefits as given. This would not command public confidence.

30. The range of GMOs considered is very narrow. GM viruses and other micro-organisms, fish, insects, other animals and crops producing pharmaceuticals are all in active development. The Commission must actively consider these – or point out the need for further analysis. Alongside financial recompense, requiring institutions and individuals to accept responsibility for their actions could be an important social dimension of addressing so-called ‘organised irresponsibility’, which has been identified by Ulrich Beck and others as a contemporary problem in dealing with technological risk. Lack of willingness to accept responsibility has contributed to erosion of public confidence in institutions’ ability to manage risk fairly and robustly.

31. Co-existence must be resolved, by setting standards for all sectors, and putting in place mechanisms to ensure they can be met. Rules must be statutory and update the current voluntary guidelines and code of conduct. A new law could then be introduced to address economic loss. The only workable and fair option is for the company supplying GM seed (or holding the Part C consent) to be liable, when the source of harm is unclear. Agreement on co-existence should prevent undue conflict between farmers. In the absence of co-existence, a legal liability system will not function fairly, and courts will have to make decisions about public policy.

32. A six part system to resolve the dilemmas identified in the consultation document could be:

a. An expansion of the scope of the proposed Environmental Liability Directive to include all habitats and species harmed as a result of release of GMOs into the environment.

b. Strict liability for environmental harm which includes unforeseen impacts (where a state of the art defence is not acceptable) unless a regulatory process is introduced whereby social ‘need’ (ie a Fourth Hurdle) is established for the use of a GMO and agreement established that the liability for any environmental harm is to be taken by society.

c. A fund established by the biotechnology industry and independently administered, for remediation or mitigation of environmental harm arising from use of GMOs where causal relationships are difficult to establish or multiple actions have been involved.

d. Environmental liability provision to be a prerequisite for commercialisation. All other practices which may lead to environmental harm should be treated similarly in the future.

e. To clarify the situation in relation to economic liability, new laws should be established. However, agreed rules and practices on co-existence between GM and non-GM/organic farming must be agreed in advance. These economic liability laws should protect all farmers equally and where there are unpredicted impacts from use of a GMO or other organism, the producer not the farmer should be held liable if the farmer has observed the agreed rules.

f. Legally binding co-existence agreements and economic liability laws must be in place before commercialisation takes place.

To what extent is it likely that the EU’s development of a directive for environmental liability more generally will introduce a special regime for GM crops?

33. Environmental liability law is being developed at EU level, which is the appropriate level at present. (Simon Barber)

How far are current liability rules able to secure redress for the grower of a non-GM crop in the event of GM contamination of that crop? Should they provide for such redress? If so, how?

34. A number of responses covered this indirectly, and the Soil Association directly:

35. Current statute and common law do not provide adequate remedies in most situations where organic farmers are likely to suffer economic loss as a result of GM contamination. As the law stands, commercial growing of GM crops in the UK would introduce a new, unwelcome and unjust principle – the polluted pays. There is mounting evidence that commercial growing of GM crops will inevitably lead to contamination problems. The Soil Association’s recent report ‘Seeds of Doubt’ disclosed widespread problems of contamination of the non-GM food chain in North America from the supply of seed to farmers all the way to contamination in food on retailers’ shelves. It will be relatively simple to show GM contamination and economic loss has occurred, but the main legal difficulty will be proving the source of GM contamination. The further along the food chain the greater the difficulty, because of the many possible sources. The problem follows from the diffuse, ever-changing pattern of cropping and will arise whatever decisions are made on co-existence. The AEBC needs to make recommendations on this evidential point. Farmers will be most at risk if GM crops are grown commercially. Even now, GM contamination has potential to cause severe economic loss to organic producers, as shown by a recent case where imported organic soya has become contaminated after its arrival in the UK. The contamination was found during routine inspection of an animal feed mill. An early result is decertification of chickens fed on the contaminated feed – they cannot be sold as organic. – and the farmer will suffer significant economic loss. (Soil Association)

36. Unless liability for financial losses to organic farmers, following contamination, deliberate or otherwise, is clearly put on the biotechnology companies responsible for introduction of the GM crop, (or other source of GM contamination), there is a high risk of many organic farmers going out of business. If organic farmers lost certification, they would lose the price premium for their products, and consumers’ trust. Many of the scenarios illustrate the concerns small organic producers have about possible future GM contamination. Individual organic farmers with limited time and financial resources could not take on the biotechnology companies in the courts. Certifying bodies might not have the resources either. (D Williams)

37. A change in law specifically for GM crops would set a dangerous precedent and could ring the death knell for much of scientific advancement in more than agriculture in the UK. Other forms of agriculture, like organic, could also be under pressure, if reluctance to use controls for diseases like Bunt or Loose-smut in cereals or yellow virus in sugar beet threatened conventional farm crops. Adverse publicity is having a bad effect on the scientific community; there is already evidence that biotechnology research is declining. There is no scientific evidence that GM crops behave any differently to conventional crops. We have managed gene flow for decades. While biotechnology has much to offer in plant breeding, plant diseases will not be a thing of the past. (W Brigham)

38. US experience suggests that companies supplying GM seeds would not accept liability. An example is contamination of non-GM seed with Monsanto oilseed rape, where the farmer ploughed in the crop as a precautionary measure - neither Advanta or Monsanto were held legally liable. (Farm)

39. An example of precautionary measures to prevent potential loss: a planned crop of maize was not planted - intended for the extraction of medicines from the ‘silks’ from the cobs, because of the proximity of a GM crop trial site. Though cross-pollination was extremely unlikely, people’s perception of purity of the farm’s products might have been affected. There was a consequential economic loss from a related agreement over sale of oats. (A Turner and Sons)

Are insurance or other indemnity mechanisms likely to be able to underpin a liability regime for GM crops, and if not, how might they be so developed?

40. Insurance companies have said they could cover damage resulting from GMOs on condition the liability system is workable. The key will be to ensure ‘workability’. (Simon Barber)

41. Liability must have limits, to ensure companies would be willing to make insurance available. (NFU)

42. Farmers may not be able to find insurance cover: NFU Mutual does not cover liability for the Farm Scale Evaluations. (Farm)

Are there other instruments, of law or policy, which might better achieve the objectives of an extended liability regime?

43. Attempting to regulate any commercial activity to cover all possible misadventures is a fool’s errand. This is particularly true for activity involving novel applications of science, as seen in developments of in vitro fertilization and embryology. The presence of a governing authority such as the Human Fertilization and Embryology Authority has been essential, in order to respond to developments as they occur. (Church of England Board for Social Responsibility)

44. Each business should be responsible for ensuring to the best of its ability that its product meets the market standard. The supply chain as a whole has a responsibility to ensure standards are deliverable. It is particularly important to establish criteria for what constitutes ‘damage’ legally, and to review criteria in the light of developing knowledge. There is a particular need to agree de minimis thresholds for accidental presence of GM material in non-GM products. (NFU).

45. A Government interpretation of the complexities of the current legislation for a range of scenarios would help provide clarity. (NFU)

46. Conventional as well as organic farmers will be affected by GM crop growing. Perceptions, as much as actual cases of contamination, can have a strong impact on commercial businesses. Many likely cases of conflicting interests may be overlooked by restricting consideration of potential commercial losses to those where actual harm can be proved, (A Turner and Sons)

Perceptions of the scenarios paper

47. There were a number of robust comments about the scenarios paper from industry and science bodies. Following from these, the draft report will set in context and explain fully the purpose of the scenarios, to test the robustness of current liability provisions, to raise and test hypothetical issues.

48. The consultation document gives an extremely unrealistic, misleading portrayal of what might happen if GM crops were grown commercially in the UK. It may prompt unwarranted or disproportionate concerns about GM crops, and does not convey how far fetched or implausible many of the events described are in real terms. Overall the scenarios show disregard for regulatory controls, for the realities of practical agriculture in the UK, and predisposition towards presumption of harm, which gives cause for serious concern. (SCIMAC)

49. The scenarios are hypothetical and extreme; however they reflect some claims about possible risks, outside currently accepted research. Actions might be brought under the tort of negligence. Consideration is needed on how claims about GM in common circulation might affect the ‘reasonable foreseeability’ test. Potential actions are likely to affect the possibility and terms of insurance cover. Phantom risks’ are taken very seriously by the insurance industry. It seems preferable to describe the scenarios emphatically in terms of a claimant’s argument, rather than as objective. (Sense about Science)

50. The scenarios do not reflect the reality of farming, the environment, the crop science industry or the regulatory framework. Even taking a worst-case view of GM crops, the scenarios appear outlandish and extremely unlikely. The tone of the consultation paper is very disappointing, as it appears to perpetuate many myths associated with agricultural biotechnology. (Bayer CropScience Ltd) The scenarios make little reference to approvals, and assume the problems have not been picked up during the regulatory process: in risk assessment or public consultation. (NFU)

51. The scenarios are perplexing. Hopefully the intention is to set laudable goals for protection of the environment, not to find reason for introducing disproportionate and unscientific environmental liability legislation specifically for biotechnology products. Public and private plant variety developers, biologists and breeders have been encouraged for several decades to undertake and use basic and applied GM research. Government at national and EU levels have provided public monies, and have worked since 1980 through the OECD on the ‘biosafety’ of GMOs. They have developed and implemented regulations founded on these safety assessment principles, for both experimental field release and commercial release. In this encouraging policy environment, industry has invested financial and personnel resources. (Simon Barber EuropaBio)

General comments

Co-existence - or litigation

52. If commercialisation proceeds, there will be expensive, socially divisive legal wrangles between neighbouring farmers. The Soil Association’s ‘Seeds of Doubt’ affirms that GM crops have not brought increased profitability and have led to legal actions: farmer against GM companies, GM companies against farmers, wider food chain businesses against GM companies. Examples include the Aventis Starlink contamination, the Saskatchewan Organic directorate legal action following failed attempts for a Canadian moratorium on GM wheat, and the ongoing story of Canadian farmer Percy Schmeiser with Monsanto. (Farm)

53. Legal liability seems unclear. Other forms of agriculture can certainly be detrimental to the environment and to human health. GM planting has two-fold risk to farmers: some North American experience suggests high cost seed may not deliver long term cost reductions; and there is potential for group action against GM farmers where damage to biodiversity or organic planting is asserted. Such actions may trigger similar legal actions between farmers more widely. (C of E Board for Social Responsibility)

54. Growing GM crops has potential environmental benefits, to reduce use of insecticides and herbicides. There is large scale growing and consumption of GM crops in the USA, but Europeans seem opposed to these biotechnological advances and more concerned about the potential risks. Politicians, farmers and scientists should review the role of intensive methods of food production in the foreseeable future, in the light of food surpluses for many products within Europe, and in the light of increasing understanding of intensive farming and environmental damage. GM crops are part of an intensive production system and would form an integral part of the review. (Wellington College Belfast students)

55. How does AEBC see farm saved GM seed in future, particularly for labelling and traceability? (UFU)

56. Three reports are recommended: EU Environment Agency report, ‘Genetically modified organisms: The significance of gene flow through pollen transfer’ (Environmental issue report 28, 2002); Joint Research Council of the EU report: ‘Co-existence of GM crops with conventional and organic crops’, and the ACRE report: ‘Gene Flow from Genetically Modified Crops’. Commercial risk assessment is missing from the debate in the UK and needs to be addressed. (Farm)

57. If planning permission were sought for development involving GM crops eg for storage or processing, which might be equally well suited to conventional crops, there could be different levels of objection. (Planning inspectorate)

Comments on individual scenarios

Scenario 1: Monoculture

58. This scenario is not unique to GM crops – monoculture can occur through conventional technology. (English Nature, NFU, SCIMAC, Wellington College Belfast students,UFU) Examples of non-GM crops on the market are imidazolinone herbicide tolerant Clearfield crops. (Bayer CropScience) Problems of monoculture are well-documented. (NCW) Current UK intensive agricultural practices are thought to be associated with declining biodiversity. (NIAB) Managing biodiversity is an integral part of good farming practice, and should be encouraged for all crops. None of the people involved could or should be made liable. (UFU)

59. A good example for the scenario would be species loss because of lower habitat/cropping diversity, which has already happened all over the UK. The particular example given is unrealistic, as lower soil fertility would probably lead to more arable plant diversity and therefore more insect diversity. (English Nature)

60. The GMO release regulations now include indirect impacts on biodiversity of changes in management practice that could result from the adoption of a new product, before commercial release can be granted. (SCIMAC, English Nature)

61. The regulatory provisions could lead to conditions being placed on the consent, possibly a limit to the area that could be grown or practices to enhance biodiversity. Farmers found breaking consent conditions could be prosecuted. Post market monitoring could identify negative impacts on biodiversity from the crop - if necessary consent could be withdrawn. (English Nature)

62. Biodiversity is only protected in public law at present, but schemes are increasing with financial incentives to promote biodiversity. In future, biodiversity might be important as a financial asset. (NFU)

63. It will be important in future to ensure novel agricultural biotechnologies like GM are components of integrated farming systems which encourage diversity in cropping management and land use, which will in turn enhance biodiversity. EU and UK Government strategy should develop support systems to encourage this. (NIAB) Such monoculture would be extremely unlikely. (Bayer CropScience) Rotation is the farmer’s single most important tool in managing soil fertility, weed and insect pests. (SCIMAC) The scenario demonstrates lack of understanding of UK agriculture. If soil fertility declined, the crop would also suffer and the farmer would probably take steps to improve fertility. (NFU, Bayer CropScience, UFU) District wide monoculture does not exist. Farms have a patchwork of different crops, with areas of rough ground, trees etc maintaining a bank of insects and weeds. (Rothamsted showed wheat can be grown in the same field for 100 years.) (J&S Grant)

64. An unrealistic example, and entirely independent of the technology. Herbicide tolerance and disease resistance have been introduced by non-GM technologies. GM crops could - in contrast to the scenario - increase both soil fertility and biodiversity eg through reduced or no tillage. Market forces and agronomic problems tend to discourage large-scale monoculture. Damage to biodiversity could require compensation under the EU draft directive on environmental liability. The decision on what baselines determine damage will be crucial in developing a workable environmental liability regime. (Simon Barber EuropaBio)

65. The new Release Directive has monitoring requirements. There would be no redress in current law or the proposed environmental liability directive unless certain sites or species were affected. The scope of the environmental liability directive must be extended to cover all land and species - without this it will have little or no impact on the UK’s environment. Where it is not possible to establish cause effect links, but it is clear that a product or practices have contributed to an adverse effect, another mechanism is needed. The fairest would be an independently administered biotechnology industry fund. (GeneWatch UK)

Scenario 2: Direct ecological effects

66. This is not specific to GM. (English Nature, NFU, SCIMAC) Invasive and inadequately regulated non-GM plants include vines and rhododendrons introduced to the UK. (Bayer CropScience) Considerable research is underway using genetic markers and genome mapping to identify conventional breeding methods for drought or salt tolerance. (SCIMAC)

67. Potential for gene transfer to wild relatives is a requirement before environmental release of a GM variety. (SCIMAC, NFU) Aspects like ‘competitive advantage’ are a key part of the regulatory process; if the risk were considered too high, the crop would not be allowed in the first place. (Bayer CropScience) If an incorrect risk assessment is accepted, then responsibility lies jointly with the Secretary of State who allowed the release and the company making the release. (NIAB)

68. Transgenic techniques may be able to produce more radical changes than conventional or marker assisted breeding. To get commercial consent, the company marketing the GMO would have had to demonstrate to the satisfaction of the statutory agencies that ecological risks from the crop were acceptably low. So the liability question would turn on whether the company deliberately produced inadequate data, whether the Competent Authority(s) had been negligent, or whether the impact could have been predicted. The product would have to be withdrawn from the market immediately. The United States Department of Agriculture have recently had a similar scenario involving GM forage and amenity grasses. (English Nature)

69. It would be useful to obtain a decision on whether GM crops are ‘reasonable’ or ‘special’ use of land. For a variety approved for environmental release, there was a strong case to consider it a reasonable use. (NFU)

70. For such gene transfer it should be possible to identify the genes and companies involved. The biotechnology company holding the release consent should be liable. Native species can be harmed in several ways: disturbing an ecosystem through changes to the behaviour of the organism involved; by interfering with wildlife conservation policy intended to maintain genetic distinctness of populations; and as an offence to a person’s senses, knowing that foreign genetic material is in a wild plant. All types of harm should be recognised. The European environmental liability directive should include all native plants. (GeneWatch UK)

71. Effects on the lines suggested are unlikely. Conventional varieties do not cross with wild relatives, so GM crops are unlikely to do so. If there were crossing, it would be discovered early on and appropriate steps taken to avoid a problem. (J&S Grant)

72. The risk of transfer of such characteristics to wild relatives through cross breeding is a potentially serious environmental threat. The result could be ‘superweeds’, which would be very difficult to eradicate. (Wellington College Belfast students)

73. The risk of gene transfer of this type is unique to GM crops – although gene transfer occurs in nature, it is the modified gene that creates the risk in this scenario. The risk of gene transfer gives cause for concern about long term effects of GM use. Farmers using good practice cannot be held liable for gene transfer from a crop. If GM manufacturers are unable or unwilling to guarantee gene transfer will not occur, or insure against it, a form of licensing should be introduced to exempt growers from unforeseen problems resulting from use or consumption of the crop. The Government, with other Member States, would make the ultimate decision to allow the crop and should be liable for any future damage to the environment. (UFU)

74. Not specifically a GM issue. Flax could be modified to be salt tolerant through chemical induced mutation of protoplasts, with subsequent transfer to a related linum species that colonises a different ecological niche and displace other species. Sea beet already shows considerable salt tolerance, so is not a good example. A novel trait would require great care in environmental safety assessment, as it will be modified to occupy an ecological niche outside its normal one. Assessing for potential out-crossing and creating nuisance plants would be important in the regulatory process. (Simon Barber)

Scenario 3: Indirect effects on ground water

75. This captures many of the issues AEBC seeks to address. (English Nature)

76. The risk assessment would have to be revisited. Consent might have to be withdrawn, because the environmentally better herbicides became ineffective through introduction of the GM crop. The Competent Authority(s) that approved the crop might be held liable for not giving appropriate weight to evidence this could result. (English Nature)

77. The scenario could be expanded to include impacts on biodiversity of farmers changing herbicide regimes as a result of gene flow or gene stacking - spraying field margins etc to remove GM volunteers. The farmers might be unable to stay in agri-environmental schemes, with a loss in potential income, raising a liability issue. (English Nature)

78. This is not unique to GM crops: it relates to liability for consequences of herbicide use. Both a herbicide and GM crop variety would have to be assessed before given regulatory approval for commercial use, and the farmer would have to abide by statutory conditions for each herbicide, specifically designed to protect surface and groundwater. Failure to comply results in liability for prosecution. (NFU, SCIMAC, J&S Grant, UFU)

79. In addition to problems of multiple defendants, the damage would be too remote for law. (NFU)

80. The scenario is unrealistic. A wide range of herbicides are not soil acting or residual and can be used without increased likelihood of water pollution. Also, there is no evidence that herbicide tolerance changes species fitness – brassica rapa is not likely to become more invasive. (NIAB)

81. An outcome of the new release Directive should be better monitoring, which may help identify cause and effect. The seed and chemical companies should be held responsible, not individual farmers – companies should have ensured proper use of their products. The problem is not as complex as the scenario may indicate. There is also a question about whether regulators should be liable in this scenario: groundwater contamination is foreseeable and proper rules should have been put in place to prevent it. (GeneWatch UK)

82. The scenario is unrealistic, as weeds are already resistant to many conventional herbicides; it ignores the regulatory system for pesticides and existence of bodies to reduce the impact of problems. Clearfield herbicide tolerant non-GM crops have similar persistence profiles to GM varieties. (Bayer CropScience)

83. Not only a GM issue. Oilseed rape made tolerant to herbicides through non-GM technologies are already commercially available, and exactly the same scenario can be posed. Simple phenoxy herbicides that are not persistent should easily control glyphosate tolerant wild turnip in fields and on railway tracks etc. The outcome might come within traditional damage. Questions would then be who committed a fault; is there a causal link? (Simon Barber)

Scenario 4: Loss in value of crops

84. This scenario is not unique to GM or organic crops. Both farmers are responsible for meeting their required standard of purity. (SCIMAC, NFU, NIAB) The examples differ in legal status: production of certified seed is governed by a statutory regime; EU regulations make provision for adventitious presence of GM. Any loss of organic premium is a commercial issue between grower and customer or accreditation body if they require different standards. (SCIMAC)

85. In practice, neighbours would discuss cropping plans. (SCIMAC, NFU, NIAB, Bayer CropScience) They implement separation distances if needed. (NFU) The organic farmer might have to opt for crops which are not open pollinating or non-cross compatible crops. (NIAB) The SCIMAC guidelines have been developed in response to the need for co-existence and using experience with certified seed. (SCIMAC) Farmers have discussions for example with maize growing, where cross-pollination of sweetcorn by forage maize has a detrimental effect on the quality of sweetcorn. Another scenario of economic loss might be a GM crop of wheat being severely infected by bunt from an organic wheat crop whose seed has not been appropriately tested or treated. (Bayer CropScience)

86. It would be useful to have clarification of whether certified seed production or organic production is a sensitive use of land. Neither claim would be likely to succeed – with a failure on both parts to implement appropriate management procedures. (NFU)

87. The scenario underlines how important it is to have rules to which all parties agree, to prevent contamination either way. If rules are not followed, the relevant farmer should be liable for harm to the other. If the rules are followed, and the crop causing the contamination behaves unexpectedly, the seed company should be responsible for the economic harm arising. If contamination comes unexpectedly from further away, the seed company should be liable. New economic liability laws are needed, as currently the law would not assist either farmer. (GeneWatch UK)

88. In time, organic farmers will start to use GM technology, as they have more to gain than conventional farmers, once disease and pest resistance are built in. Under current laws, if the seed grown is not up to standard, the grower takes the loss. (J&S Grant)

89. The danger of contamination of organic crops would not be restricted to GM crops, and could be caused by any non-organic crop. New controls and compensation arrangements will be needed to protect farmers seeking to grow genetically pure crops. (Wellington College Belfast students)

90. There should be scope for co-existence. The established scheme for certified seed production sets minimum distances for each variety, which are accepted as realistic and achievable, allowing co-existence of growers with different criteria. In the same way organic crops have thresholds for adventitious non-organic presence, there should be a realistic tolerance for GM contamination. Maintaining zero tolerance for GM is unrealistic. As it is beyond the scope of detection, it should be categorised as ‘sensitive’ and outside the protection of the law. (UFU)

91. If contamination of the organic crop happened unknown to its farmer – mini whirlwinds may sound unlikely, but happen frequently during very hot harvest weather – was then sold to a seed company, and sold on before being found contaminated, the seed company and farmers supplied by the seed company would want to claim from the farmer who grew the seed crop. Farmers growing a crop from the contaminated seed could theoretically be liable under patent law to pay a fee to the company which patented the genetic modification. This has happened in North America, but is very wrong – the company should be responsible, rather than a farmer growing a contaminated crop unknowingly. (South Elmham Hall Farms Ltd)