An attractive feature of the new project is that it is designed to be implemented through NGOs, a first-time major government-NGO cooperation bid in primary education. It is a recognition, also for the first time, by the Ministry of Primary and Mass Education (MOPME) that a different "non-formal" approach is needed to bring quality primary education service to the large number of children whom formal schools so far have left behind. And the project's funding comes from a 50 million dollars grant, rather than the usual loan, from the World Bank, complemented by six million dollars from SDC. A debatable component is the provision for stipend to children in 60 percent of the targeted upazilas as an "experiment," although NGOs have demonstrated that good service and not cash is the way to attract and keep children in school.

Non-governmental organisations in Bangladesh, especially BRAC, have pioneered a successful model of non-formal primary education for very poor and disadvantaged children who do not enroll in primary school or drop out early. Currently, one million such children are enrolled in some 32,000 non-formal schools, run directly by BRAC or through support to local small NGOs and community organisations. Children of age 8 to 14 years are taught the equivalent of five-year primary education in four years. Attendance in classes and completion rate of the course are much higher, and independent assessment has shown a better performance of children, in these schools than in regular formal schools. Most of the children go on to join formal schools at the secondary level.

BRAC and the other NGOs engaged in non-formal primary education do not argue that their schools are the model for primary schools in the country. They serve a small proportion of the disadvantaged children with a very special effort in management, supervision, and dedication of staff that only NGOs with strong leadership and professionalism can offer. The public system of primary education must improve and change to serve all children. Meanwhile, however, a relatively small number of NGOs, out of some 28,000 registered with one or another government agency, has developed and demonstrated a winning formula.

MOPME, the ministry responsible for primary education, instead of recognising and encouraging the efforts of NGOs in serving the children of the poor, has refused even to count the non-formal schools and children in them in primary education statistics. The specious argument offered by MOPME officials is that the constitution requires uniform primary education for all children and, therefore, non-formal schools cannot be regarded as "schools." No matter that there is hardly a uniform system of primary education in the country and what exists is seriously discriminatory to the poor. MOPME itself reports the existence of at least 11 types of primary institutions, other than the NGO-run non-formal ones. MOPME's concept of a "uniform system," an appropriate national goal, is more related to administrative arrangements for running a school, than the learning content and objectives and the quality of what children learn -- even if the delivery mechanisms are different. It is hoped that the ROSC project signals a change in the mindset of MOPME.

There is also a sorry record of government-NGO collaboration in non-formal education in Bangladesh. In literacy projects managed by the now defunct Directorate of Non-Formal Education (DNFE), which came under the purview of MOPME, there were widespread allegations of gross corruption and mismanagement in the projects. Implementation of the literacy projects, a project for basic education for urban working children (with some similarity to ROSC), and post-literacy and continuing education projects was carried out by contracting out the job to NGOs. The process of NGO selection and formation of Centre Management Committees, with excellent criteria and procedure spelled out on paper, and management, supervision, and monitoring of projects failed miserably. In the face of outcry in the media and independent researchers' assess-ment, DNFE was closed down by the Prime Minister's Office last Novem-ber. Earlier, the Total Literacy Move-ment, the flagship programme of DNFE, was discontinued midstream, in July 2003, due to the same type of complaints.

NGOs can and should be genuine partners, not just contractors for tasks assigned by officials who have little relevant professional know-ledge, in education programmes, including the new ROSC initiative. It can work only if the pitfalls of the present government-NGO coope-ration -- corruption, mismanagement, and control in the hands of officials without experience and knowledge of the subject -- can be avoided.

In this regard, the Campaign for Popular Education (CAMPE), a well-regarded forum of education NGOs in the country, has welcomed ROSC and has offered to help. CAMPE's membership includes most voluntary development organisations in the country active in basic education. CAMPE published this week a new edition of the Directory of Education NGOs, a compilation of basic data about more than 700 large and small organisations. A tremendous advantage for ROSC lies in the successful experience of NGOs in their work with out-of-school children.

CAMPE has offered to help identify the NGOs which have demonstrated their capabilities and which can serve as providers of technical support to the small local NGOs. The latter are expected to carry out the work of the learning centres. CAMPE has expressed its willingness to coordinate and support communication and social mobilisation necessary in the communities, help in identifying local service providers for the learning centres, and help assess their capabilities and performance. It is necessary for the government and the donors to bring NGOs (through CAMPE as a recognised forum of education NGOs) into a full partnership relationship from the very beginning of the project. It is, however, reported that the project design has been already developed and approved by the government and the donors and that an advertising agency has been recruited to carry out communication and mobilisation work in the targeted rural communities.

Much more than the 390 crore taka is at stake in ROSC. A genuine partnership relationship with committed and capable NGOs is essential for the success of the project and to avoid the well-known hazards of government-NGO coo-peration. The principal partners, including NGOs, need to work together from the beginning in designing, planning, and implementing the project -- sharing both responsibilities and risks. For this to happen, a change of attitude in the government and the aid providers is needed.

Dr. Manzoor Ahmed is Director of Institute of Education and Development, BRAC University.