I met Likeness last month in Malawi. She is a peasant farmer with three kids and no prospects. The rains came late to her maize field near the Zambian border and then they stopped early. The result, just like in 2002, is misery: her crop failed, what she harvested has nearly gone, she has no work, and there's no money to buy food, fertiliser or seeds for next year. She is the face of world hunger, along with nearly one billion others caught up in the vortex of unprecedented food-price inflation and extreme poverty.
So could GM maize, or industrial farming and giant agribusiness - the "unmentionables" that Prince Charles railed against this month - make any difference? One man who may know is the head of Monsanto, Hugh Grant, who was in Malawi just before me at a conference on the future of world agriculture. He recalled how, after the 2002 famine, Monsanto sent Malawi hundreds of tonnes of hybrid (not GM) seeds. "Yields increased by 50 per cent to above 32 bushels per acre," he said. "Better seeds and fertiliser make an enormous difference."
Correct. As any farmer knows, you don't need GM crops to grow more food. Rather, you need good seeds and soils, better manures, crop rotation and irrigation. Education, markets, places to store the food where the rats can't get at it, all help farmers earn money. GM promises increases of 10-20 per cent in some crops. Good farming can more than double yields.
Another man who knows whether or not GM will help Likeness is Professor Robert Watson, chief scientist at the UK's Department for Environment, Food and Rural Affairs, and formerly Bill Clinton's scientific adviser. Watson recently chaired the International Assessment of Agricultural Knowledge, Science and Technology for Development (IAASTD), an immense and rigorous three-year, government-level scientific study of world agriculture, backed by the UN and the World Bank and independently peer-reviewed twice.
The 400 authors looked at the evidence and concluded that business as usual - industrial agriculture and trade rules tilted towards large corporations - can barely feed people today and won't be able to in future. The problem is that the present financial and trading systems work at the expense of the deteriorating environment and the growing numbers of poor.
But what about GM crops? The IAASTD authors kept the door open on the technology but said that it was not the solution for the world's poor. Instead, they called for more respect for the knowledge of local communities. This enraged the participating US-dominated agrochemicals and biotechnology industries, which walked out, claiming the whole exercise was unbalanced. The US, alone of all major countries, has refused to endorse the study.
GM acreage is growing worldwide, but it may never provide for the poor. In the 20-odd years since the first crop was sown, billions of dollars have been spent researching, developing and marketing the technology. But it is stuck on a very few commercial crops, and is still at single-gene transfer level. What's more, it is suited to monocultural farming, and the questions of ownership and safety just won't go away.
Over the years, there have been genuine safety concerns over individual GM foods but early fears have been allayed by US and EU government insistence that these are some of the world's most regulated foods. Activists still argue that there have been few major human health studies of an inherently unpredictable technology.
Back in 1994, the industry was promising crops that resist cold weather, drought, pests and disease, as well as plants that reduced the need for fertilisers. The world is still waiting. Last month, Hugh Grant said he now expected drought-resistant crops to be ready in the US "within six years"; it seems the science is more complicated than was thought. That hasn't stopped the industry enjoying an expansionist phase as agribusiness takes advantage of the food crisis, but anyone trying to assess the success or failure of GM can find themselves in a snake pit of claim and counterclaim.
Companies regularly overstate the potential gains of GM by under-reporting average yields in conventional production; activists seize on individual crop failures to propose that the whole technology is corrupt. Meanwhile, academics are partial to the big bucks that industry offers for GM research and development, and governments fear to upset their legions of small farmers.
One side paints a picture of the world's poor being denied a technology that could hugely improve lives; the other side claims industrial agriculture's heavy gun is aimed directly at it. Both are probably wrong.
Monsanto espies huge profits in places such as Malawi, where the whole country depends on maize. It's not legal to sell GM there but even if it were, the chances of Likeness and the small farmers like her, 90 per cent of the population, benefiting from it are utterly remote. Malawi is a land of conservative, uneducated and vulnerable farmers. They could not possibly afford the seeds or the herbicide, let alone take the risk. It would be criminal to ask them to.
Hugh Grant probably isn't losing much sleep about Malawi. The company is making record profits out of selling a high proportion of its GM seeds and herbicides to American and other farmers for the growing of biofuels. Of course, the company cannot be blamed if there's less food on the world market, or that maize prices have more than doubled. Instead, Monsanto's share price has risen to dizzy heights and the company has just raised prices for its seeds and herbicides by more than a third.
Now even farmers in the US are complaining about GM.
John Vidal is the environment editor of the Guardian