This has all been done without external support and finance, and demonstrates what’s possible when the motivation is right. According to the report the growing of a core crop of millet has resulted in the accumulation of some 20,000 tonnes of stored grain, across 40,000 households. This provides a source of resilience against future shocks, improving the sustainability of livelihoods in the district.
It all sounds too good to be true. Unfortunately I missed the launch of the report in London, and I have not been able to visit the areas in Gutu (clearly the effort is focused outside the new resettlement areas, as the farmers in our sample in Gutu are sticking solidly to maize), but the data is impressive, and the testimony passionate.
But there are questions about the indigenous grain strategy being advocated. I speak from experience, as in the 1980s, together with an NGO ENDA Zimbabwe, I was involved in a project that promoted small grains – finger and pearl millet, and sorghum – in Zvishavane district. The project supplied seeds, and supported the processing of the grains with the provision of ‘dehullers’. While it did make some in-roads, by and large the project failed. The dehullers are now archaeological relics and most farmers in the area plant maize.
Why was this? There are a number of complex intersecting reasons. First, growing millet is hard work. Finger millet is a difficult crop and pearl millet is subject to massive bird damage, from flocks of Quelea who descend in large numbers on any field. This is a big turn-off, as bird scaring is labour consuming and troublesome. Older farmers used to tell us that the problem is worse because millet fileds are now few. Being a first mover growing millet is brave. Second, millets take a lot of processing. The hard outer layer has to be removed to get the flour – hence the dehullers. Without these, it’s tough pounding, and much more difficult to prepare than women.
In discussions around the ENDA project, women always used to object. They didn’t want the hassle of going to the fields early and staying late – they had other caring work to do too – to scare the birds, and pounding for hours to get a few kilogrammes of millet flour was not worth the effort in their view. Finger millet in particular was not liked by women, as it encouraged beer drinking. While men would get quite motivated about millets in the discussions, it was women who often dominated the planting decisions, and it was striking that there was always much less millet planted than was discussed. In intrahousehold decision-making, women’s agency can be quite powerful. Third, is taste. Finger millet is good for beer, but many find the ‘sadza’ porridge of pearl or finger millet less a delicacy as is suggested in the new Harare ‘African’ restaurants. With the colour and consistency of concrete, pearl millet sadza is not my favourite food either! Several generations of people accustomed to easting white maize means that sadza from millet is difficult to sell (although it’s quite nice with soured milk I must admit!).
So there are reasons why adoption of millet is constrained. But the advantages of secure storage, as documented in Gutu, are potentially substantial. Millet stores well – for years. Unlike maize that needs to be consumed within a year, you can keep a granary full of millet over a full drought cycle. In the past, rainfall was patterned by cycles of a few years, with droughts coming more or less predictably. Having millet stores for the times when rain was less was essential for food security, and the store could be replenished when the rains returned. It was a perfect system for local level resilience. But with the move to maize, and the advent of food aid and relief programmes, these cycles have been disrupted. Climate change too has had an impact, as droughts are much more unpredictable these days, even if average rainfall has not shifted much.
In the areas I have worked in Masvingo and Midlands provinces, a key moment in this transition in the food crop mix and local food security system, was the devastating drought of 1991-92. This had a catastrophic impact on many fronts, and many were reliant on food aid through imports. Perhaps the most dramatic impact for the long-term was the disappearance of local varieties and land races, particularly of small grains. People had to plant their last seed stores, and when they didn’t grow, that meant the local extinction of a huge array of genetic variety, and with it the knowledge of what grows well where. A number of research and NGO projects – most notably the Community Technology Development Trust, whose head Andrew Mushita is a veteran of the ENDA experience – have tried to document and revive this genetic biodiversity, but with it lost from the farming and livelihood system, it is difficult to reincorporate.
Around that time, as part of a wider project on risk and farming systems, we did some modelling of risk responses under different conditions. Like all models it was only an interpretation of reality, but the approach used tried to simulate the type of stochastic variability seen in an increasingly volatile climate. The results were surprising. Despite the greater vulnerability to low rainfall episodes, a maize dominated strategy came out better than one focused on small grains in the model. This was because of the costs of production, and the value of maize. As long as this value (in the form of grain or cash) could be carried over to the following year, opting for risky maize made sense especially for the poor.
Farmers didn’t need a model to show them this – and especially women, for the reasons described – but it highlighted how complex decision-making under conditions of high variability is. As the model showed, mixed strategies made the most sense, with a smaller amount of millet as part of a mix. As the maize economy came under stress in the 2000s with the failure of markets, and government support through the Grain Marketing Board, new incentives to secure food locally emerged. In this period, for the first time in decades, the political and economic support for maize had disappeared. And without state support and the absence of a cash market because of hyperinflation, the maize reliance strategy became much riskier, and a local production system became preferable.
I suspect it’s a combination of these factors have pushed farmers in Gutu to take up millet again at the peak of the economic crisis. These were very different conditions to those in the late 1980s, when the earlier millet focused strategies foundered. Context matters a lot, and it is a combination of factors – markets, taste preferences, labour requirements and the wider political economy of crop support – that combine to make one technology more or less favourable. Maybe the experiences from Gutu suggest that the age of millets are returning, and we will have to get used to a different type of sadza.Agriculture