5
Learning and Mislearning
Robert Chambers
How we learn and mislearn about canal irrigation systems has not itself, to my
knowledge, been a subject of research. Yet how our beliefs are formed and sus-
tained deserves a hard look. This chapter examines examples of research and inno-
vation. Two sets of accessible data are considered. The first is three studies made of
the left bank of the Mahi-Kadana Project in Gujarat. The second is action research
undertaken on the Mahanadi River Project and Hasdeo Bango Project in Madhya
Pradesh. In analysing these, and in subsequent discussion of pilot projects and ‘islands
of salvation’, my approach is critical. But it is easier to criticize than to do. It is not
hard to find holes in any pioneering research or action, especially when the arena is
as complex and varying as canal irrigation. The courage of those who run the risks of
research and action is to be applauded. The spirit of this chapter is positive, to learn
lessons and to contribute to the collective struggle, to get closer to the truth.
Mahi-Kadana: Seeing Parts and Missing Links
Unusually, the Mahi-Kadana Project in Gujarat has been the subject of three sepa-
rate and substantial multidisciplinary studies, each with a different purpose, orien-
tation and disciplinary composition. Each study reveals different aspects of the
project, and leads towards different practical conclusions.
The Mahi-Kadana Project itself developed in two stages. The first was the
construction of a weir across the river Mahi at Wanakbori in 1958. This diverted
run-of-the-river flows primarily for the kharif (monsoon) season, to supplement
rainfall. The second stage was the construction of the Kadana reservoir upstream,
in 1978. Storing and releasing water from this reservoir was designed to enable rabi
(winter) and summer crops to be grown. The three studies were concerned with
the larger Mahi Right Bank (MRB) and not the smaller Mahi Left Bank, and all
discussion and figures which follow refer only to the MRB. In 1980–1981, the
Cultivable Command Area of the MRB was reported to be 213,000ha, with an
Reprinted from Chambers R. Learning and mislearning. In Managing Canal Irrigation. Cambridge
University Press, Cambridge 1988 © Oxford and IBH Publishing Co. Ltd.