Sand in the Engine: The Travails of an Irrigated Rice Scheme in Bwanje Valley, Malawi

Abstract

The establishment of the Bwanje Valley Irrigation Scheme (BVIS) in Malawi is a striking example of informed amnesia in development assistance. Despite the lessons learned earlier concerning a process approach to participatory irrigation development in Africa, in the case of BVIS outside interveners designed an irrigation system and parachuted it into Bwanje Valley as a black-boxed technology. Using a sociotechnical approach, this article analyses the travails of this irrigation scheme, showing that the conventional irrigation factory mindset is ill-suited for creating durable water networks. Achieving tangible improvements in rural livelihoods is better served by the interactive prototyping of water networks in situ, ensuring that new irrigation schemes are embedded in existing landscapes and complementary to existing livelihood strategies rather than supplanting them.

Acknowledgements

The authors gratefully acknowledge the research funding availed to Alex Bolding by the NWO (Dutch Research Council) under the ‘Shifts in Governance’ programme. During the fieldwork in Malawi, valuable assistance was provided by Rob Wilkens, seconded by DANIDA to the Department of Irrigation of Malawi, and Kees van Straaten, supervisor of this research at Wageningen University. We dedicate this article to Kees, whose untimely death on 22 December 2002 has been a great loss to us all. In him we lost a close friend and an inspiring colleague. This article is a tribute to his commitment to irrigation and development in Malawi, which started in 1995 with teaching irrigation and soil and water conservation at the Bunda College of Agriculture. While in Malawi, he also founded the Maravi Irrigation Company (MIC), a not-for-profit company that focused on designing and constructing irrigation systems in close interaction with farmers. It is our hope that this article will contribute to furthering his ideals.

Notes

1. This choice reflected an international trend. Across the globe construction of new irrigation schemes was curtailed, whilst emphasis was put on rehabilitation and cost recovery programmes in existing schemes (see Jones, 1995).

2. The plan included a proposal for the construction of BVIS comprising some 1,730 hectares.

3. It is not clear from Kishindo's study if women were also newcomers to Limphasa.

4. Amongst the rice irrigation schemes thus constructed or rehabilitated were Mwea (7,700 ha) and Tana delta (1,800 ha) in Kenya, Lower Moshi (2,300 ha) in Tanzania and Lower Anambra (3,850 ha) in Nigeria (Nippon Koei, 2005). Some of the larger irrigation projects entailed tied-aid grants where Nippon Koei was involved in all project stages.

5. Much of this material was available to the researchers: the original design drawings, copies of all feasibility and design studies, most of the weekly progress reports by the consultant and contractor, copies of the correspondence between the main actors, as well as access to the confidential files of the Department of Irrigation.

6. Maliro (1993: 18) compared yields attained in government rice schemes with those attained in self-help schemes in Ngabu ADD. Rice yields of 2.5 ton per hectare in two self-help schemes were found, 4.5 ton per hectare in government schemes for IET rice and 2.5 to 4 ton per hectare for blue bonnet grown in government schemes.

7. In one case, beneficiaries damaged newly constructed canal segments by digging out mice holes. Another case concerned a dispute over compensation for a destroyed crop. The beneficiaries blocked all construction work for two days before the matter was settled amicably.