(i) NGO projects rarely address wider scale structural factors that underlie rural poverty;

(ii) NGOs have limited capacities for agricultural technology development and dissem-
 ination, and limited awareness of how to create effective demand-pull on government
 research services;

 (iii) the activities of different NGOs remain uncoordinated, and information exchange is
 poor especially among small NGOs where transaction costs are high.

These strengths and weaknesses of NGOs, and their implications for NGO-GO relations, are
discussed in more detail below, and illustrated by examples from Africa, Asia and Latin
America.


 Successes and Failures of NGO Technology Development

There are five main areas in which NGOs have been innovative and relatively successful.


Diagnostic and Technology Development Methods

Conventional public sector approaches to agricultural technology development have difficul-
ty in coping with the wide range of agro-ecological and socio-economic conditions charac-
teristic of the complex, diverse and risk-prone areas in which many of the rural poor live
(Chambers et al., 1989; Richards, 1985). In such areas, agricultural technology development
must not merely be on-farm and farmer-managed, but participatory in order to draw on local
knowledge and to meet farmers' needs, opportunities, constraints and aspirations. The
approaches introduced in GOs have frequently been expensive and time consuming, and often
not participatory (Biggs, 1989a). Some NGOs, on the other hand, have been innovative in
developing more parsimonious approaches.

For instance:

* In Kenya, the Diagnosis and Design methodology practised by ICRAF partly grew out
 of the development of methods by CARE and Mazingira in the early 1980s to elicit
 rapid farmer assessment of tree species (Buck, 1993).

* In Chile, NGOs were responsible for the elaboration of farming systems perspectives,
 and their subsequent teaching to other institutions (Sotomayor, 1991).

* In India, Myrada has been instrumental in developing participatory rapid appraisal
 methods and training for both other NGOs and government staff in their implemen-
 tation (Fernandez, 1993; Mascarenhas et al, 1991).


GATEKEEPER SERIES NO. SA43