Words and Ideas: Commitment, Continuity and Irreversibility 153to articulate this policy, the IBRFP project began to lose its support and, consequently,
its reality. It bore new policy labels of exclusion: ‘enclave project’, ‘niche project’, ‘repli-
cable model’, ‘parallel [to the state] structure’, ‘sectoral, downstream, micro-managed
project’. IBRFP had suddenly become the flared trousers^20 of the DFID wardrobe.In abandoning projects, lenders and donors have succumbed to a tragic pathology.
Notwithstanding the common view of colonialism, I found, in the late colonial
settlement projects that I studied in Africa, a strong commitment of local-level
administrators to ‘their’ projects and to the settlers. Administrators on the ground
were face to face with the realities. To varying degrees, subsequent political repre-
sentation also provided commitment and support to local projects. It is only now,
with lenders, donors and policy makers interacting and influencing one another
more and more, insulated in their capital city cocoons, that projects can more eas-
ily be abandoned. RIPS in Tanzania, the project in Brazil, and IBRFP in India all
provoke sad reflection on the costs of abandoning projects: staff demoralized, peo-
ple disillusioned, government discredited, ‘money down the drain’, benefits to the
poor forgone, and opportunities lost for ground-truthing, learning, innovation
and capacity-building. In our brave new 21st century of aid, for many poor peo-
ple, projects have not proved irreversible enough.
Commitment and continuity with policy
This view gains new relevance with the dominant development aid policies of the
new century. There are signs of a new consistency in two domains: in rhetoric,
extolling partnership, country ownership and policies that are pro-poor; and in
targets, with the international and now MDGs (Millennium Development Goals)
set for achievement by 2015. Sector programmes may, perhaps, not demand as
much long-term commitment as projects. But whether for projects or for sector
support, inconstancy is a feature of much aid. In this, agencies differ. United States
Agency for International Development (USAID), for example, stands out for its
short-term swings of policy and vocabulary, and its relative unreliability.^21 Overall,
the shift from projects to sector programmes and policy influence may have meant
a move for aid agencies from the grounded, bounded and stable to the more nebu-
lous, permeable and inconstant.
With projects, failures were harder to hide. There were reasons to hang in there
and try to make them work. With sector support, failures by lenders, donors and
governments may, perhaps, tend to be less embarrassingly conspicuous and respon-
sibility less attributable. More actors are involved. The scale is wider. The impacts
are further away. So responsibility and accountability are more diffuse. Both polit-
ical risk and institutional commitment have diminished. In consequence, it may
be easier to exit and to deny responsibility. The new dangers implied may not be
well recognized by lenders, donors and recipients. The dangers include, as Albert
Hirschman might note were he to revisit the aid scene, that less long-term com-
mitment and less continuity mean less creativity and less learning.