Performance of Low External Input Technology in Agricultural Development 69made further adjustments, such as combining structures or altering the width or
management of vegetative strips.
It is more difficult to assess changes in farmers capacities for experimentation
(Sumberg and Okali, 1997). The strongest evidence is from Honduras, where the
project put particular emphasis on developing experimental skills. About one-fifth
of the participants reported at least one experiment carried out subsequent to the
project. In addition, a higher proportion of farmers who utilized LEIT reported
trying alternative pest control techniques in maize (although they used no less
chemical pesticides than other farmers). In the other two cases there was much less
evidence of farmer experimentation. Projects that emphasize simple field experimen-
tation can stimulate farmers to continue with their own investigations, but the case
study examples do not support the view that LEIT projects, on their own, can set off
a burst of local innovation. The comparison between FFS and local agricultural
research committees presented by Braun et al (2000) emphasizes that the develop-
ment of experimental capacities often requires a more long-term approach.
Because many LEIT projects feature group methods there is an expectation
that successful experiences will lead to further organizational growth and innova-
tion (e.g. Pretty and Ward, 2001). However, there was very little evidence that
participation in an LEIT project was sufficient to maintain the original groups
formed for the projects or to lead to further organizational development. In Hon-
duras, farmers who participated in the project belonged to more organizations, but
the project did not lead to any further group initiatives. On the other hand, the
satisfaction some project participants experienced in acquiring useful technology
may have encouraged their subsequent participation in other organizations. The
density of rural organization in Kenya is the highest of the three cases, and although
some of the catchment project participants went on to participate in other activi-
ties it is difficult to attribute this to the project itself. In only one site did part of a
catchment committee remain together for further activities. In Sri Lanka, the FFS
functioned for a season and then disbanded; the only exception was one case in
which another extension activity (in seed production) drew on the members of a
previous FFS.
LEIT projects are established in communities where social networks are already
in place. If the projects are competently managed, they may contribute to further
group-led initiatives. But the idea that a technology generation project has a high
probability of leading to further, spontaneous group action is erroneous. Commu-
nity organizations develop for various reasons, but permanent, autonomous capac-
ity requires particular incentives. The high density of donor projects in certain
areas can contribute to an image of organizational growth, as a group from one
project is inherited by a successor project. But significant organizational develop-
ment requires adequate incentives and local commitment, and often depends cru-
cially on outside resources as well. A community in Honduras that served as an
interesting contrast to the two principal project areas of the study provides a good
example. The community demonstrated a very high uptake of LEIT and exhibited
excellent local organization. But pre-project activity (including interventions from