Britain's field trials of the environmental impact of GM crops - dubbed unsexily by Whitehall Farm-Scale Evaluations (FSEs) - have proved a watershed. Back in 1999, when the trials were first mooted, the plant breeders and biotech industry (represented by Scimac - the Supply Chain Initiative on Modified Agricultural Crops) thought the trials would be a useful means of buying time until popular feeling against GM technology died down a bit. If the trials were designed within sufficiently narrow parameters, their "success" could be virtually guaranteed. It has not turned out like that.
Throughout the four-year trials the government repeatedly asserted that if GM crops were found to cause harm to the environment, it would not license those crops. With oilseed rape and beet it has been convincingly demonstrated that GM varieties cause significantly greater harm to the environment than conventionally farmed crops. That settles the matter. No GM oilseed rape or beet in Britain.
With maize the opposite appeared to be the case. But this was due to three factors that make such a conclusion mischievous and misleading. First, the chemical herbicide used in the trials for non-GM maize was atrazine, an extremely toxic chemical already banned in the Netherlands and Germany and now being banned throughout the rest of the EU. So the claim that GM maize is better for the environment than the conventional alternative is invalid, because it is based on a comparison involving a chemical no longer in use.
Second, Bayer Crop Science, the biotech company that oversaw the trials, advised the farmers to carry out only one spraying of the GM crops with their herbicide (glufosinate ammonium). Subsequently, weeds were able to grow again and there was more food for the invertebrates and other wildlife in the GM-trial fields. But farmers in real life would almost certainly spray their fields two or even three times, because that would increase yields. The environmental impacts would very likely be that much worse.
And third, glufosinate ammonium (brand name Liberty) was used on its own for the GM trials - even though Scimac knew that it has been found to be an ineffective weedkiller when it is applied in isolation. This again distorted the environmental impact of the GM maize trials.
For both GM and non-GM maize, therefore, the trials were artificial and manipulated to bring about the desired outcome. If the maize trials had reflected genuine market conditions, they would almost certainly have produced the opposite results.
So, the GM field trials were performed in accordance with the biotech industry's own very narrow parameters and still science gives GM a decisive thumbs-down. But even were that not so, it would be quite wrong to assume that the FSEs constitute an adequate test of the impact of GM crops on the environment. They are, in fact, simply a test of herbicide management; that is, of the differential impact on wildlife in the fields (beetles, butterflies, worms, bees, etc) of using one herbicide for GM plants and another for non-GM plants. They do not test safety or risks to human health. They do not test the effects of GM on soil residues and bacteria, transgene flows or bird populations. Most crucially of all, they do not measure what would happen to the environment if farmers tried to maximise their yields (which is undoubtedly what they would do in real life), rather than trying to minimise environmental impacts for the sake of a trial. Nor do they bring organic crops into the equation; a comparison was made only with conventionally grown crops. Organically farmed crops would have shown much more favourable environmental results.
So a lot more work is needed before anyone can claim to know the full effects of GM even on the environment alone. That still leaves other crucial areas still uninvestigated. The most important is undoubtedly the effects of GM on human health.
It is scandalous that no systematic testing of the impact on human health of eating GM foods has ever been carried out. There are several reasons why this is now urgently needed.
First, GM genes are inserted randomly, out of the sequence that nature has evolved over hundreds of millions of years, and without knowing how the network of genes into which they are injected functions. Unpredicted and undesired effects are likely to be triggered as a result. Second, the vectors used to insert the GM material are viruses or bacteria, which often move (this is called horizontal gene transfer) into other organisms such as gut bacteria. This could compromise antibiotic resistance. Third, because GM organisms are novel products allergic reactions to them are common. Fourth, the herbicides used on GM are potentially hazardous; glufosinate ammonium, for example, is a neurotoxin and a teratogen (ie, it harms embryos). And fifth, the government-funded Medical Research Council and other medical institutions have acknowledged that we need to know more about the effects of GM on metabolism, organ development, immune and endocrine systems and gut flora.
In the face of all this risk and uncertainty, it is a scandal that the regulatory authorities, first in the US since 1992 and subsequently elsewhere (including the UK), have allowed the biotech companies to get away with the pseudo-concept of "substantial equivalence" as a way of avoiding proper testing. When the companies produce a new GM product, they examine whether it is broadly similar to its non-GM counterpart in terms of toxins, allergens (the substances that produce allergies), and nutrients. If there is "substantial equivalence", the new GM product is deemed to be safe.
This is unacceptable on two grounds. One is that some independent testing has found that GM products are not broadly similar in these respects. And much more importantly, food safety cannot be determined merely from an analogue; it requires independent and systematic testing of the GM product itself.
But, say the biotech apologists, GM food has been eaten in the US for seven years now with no ill effects. So what is the problem? The problem is that no epidemiological studies have been undertaken in the US to support this claim, and since damage to organ systems is not readily detectable and may be cumulative over time, the apparent absence of evidence is not evidence of the absence of harm. More worrying, though, is that the official US Centres for Disease Control believe food-derived illnesses to have doubled in the last seven years.
At least two further issues would need to be settled before there could be any question of introducing GM into the UK. One is co-existence: how can organic and conventional crops be protected from cross-contamination by GM? This may be an impossible problem to solve, because while the vast majority of GM pollen falls within the standard separation distances of 100 to 200 metres, some 1 per cent to 2 per cent appears to travel further, particularly on a windy day. Even that small proportion involves millions of GM pollen grains. And if cross-contamination does occur, as it inevitably would, how would an organic farmer be compensated for partial or total loss of business following loss of organic accreditation? It would require a statutory liability provision, but none is on the statute book and none is planned.
The second issue is labelling. Most people want to continue eating GM-free food, but the EU labelling laws soon to be put in place are based on a 0.9 per cent threshold. Food labelling will not have to indicate any GM content below this level. You wouldn"t know whether food was GM-free or if it had a GM content of almost 1 per cent.
But, incontestable though they are, all these further objections to GM are in one sense academic. The government itself set a criterion for rejecting GM crops; namely, whether they damage the environment. It has now been demonstrated by the government's own tests that they do. We must press the government to carry through what it unequivocally promised. As someone once said, there is no alternative.