84 Ethics and Systems Thinking
system managers give a better than usual supply to a part of their system where
they know investigations are taking place.
Such a supply usually has costs elsewhere. Digby Bevan (personal communica-
tion) visiting a warabandi pilot project in Andhra Pradesh in 1980 found, through
his initial questioning, that the farmers were happy and everything appeared to be
running smoothly; but further investigations revealed that the pilot scheme was
getting preferential treatment and receiving more water than needed while the
supply in the larger area of which the scheme was part was completely erratic. P. S.
Rao (personal communication) visiting a pilot project for rotational water supplies
on part of the Periyar Vaigai Project in Tamil Nadu went further down the minor
and found farmers who complained that they were getting less water than before
as a result of the pilot project. Anthony Bottrall (1983, p106) found that farmers
in a village immediately downstream of a pilot (showpiece) watercourse were
receiving very little water because extra supplies were being diverted to the pilot
watercourse in order to demonstrate its ‘success’. Similarly low concentrations of
wheat, mustard and cotton on the Rajasthan Canal control, and its higher concen-
trations of gram, probably owed something to the adverse effects of the privileged
water supply of the Naurangdeshar Distributary.
So, wherever research or a pilot project is undertaken and ‘success’ reported,
whether it is warabandi, on-farm development, canal lining, farmers’ organization,
rehabilitation, the management of minors or distributaries or something else, costs
to other parts of the system must be assessed. The question is whether water to the
special area means supplies to others, which are smaller, are less timely and less
predictable. Islands of salvation often draw resources to themselves and deprive
others. Many successes are really failures. Nor are they replicable unless water sup-
plies on the main system can be managed to provide a similar supply elsewhere.
Once again, analysis points to the priority of main system management.
Third, the normal process of learning from ‘islands of salvation’ can mislead.
The biases of rural development tourism (Chambers, 1983, pp10–25) apply.
Short visits by officials and researchers give rise to insights of uniform superficial-
ity. The same farmers are met by a succession of visitors and they tell them the
same things, while others are not met. What the farmers say can be slanted, some-
times false. A social anthropologist in South India once sat through a whole morn-
ing during which officials put farmers through their lines, rehearsing them in how
to answer questions expected from a senior officer who was coming the next day.
Then visitors are taken to the best places, and do not have time, even if they wish,
to probe multiple causation, to question biases in what they see or are told, or to
perceive costs to other places which offset the benefits they are shown. With any
policy – CADA, OFD, 8 hectare chaks, warabandi – rural development tourism
and the well-prepared visit impede learning and distort perceptions. Visitors
depart to their conferences and committees where error reinforces error as they
repeat to each other their common misperceptions. As with the Mohini Coopera-
tive, a belief can gain currency that an approach is replicable and should be
included in policy and plans. Busy politicians and policy makers need good ideas