Analysing disrupted health sectors: a toolkit:
Previous page |
Learning from previous recovery processes
Despite the efforts of researchers and practitioners over the last two decades, learning from previous conflict and post-conflict experiences has not thriven. Responses to new crises range from the mechanical replication of previous approaches, to the starting anew every time, in this way discovering again some well-known lessons. The difficulty of comparing different situations and of drawing correct, appropriate lessons from them is not exclusive of conflict-affected health sectors.
Investigating the field of health sector reform, McPake and Mills (2000) argue that the propensity for either none or too much transfer of approaches reflects two symmetric fallacies: “the search for a single best model” on the one hand, against “the belief that nothing can be learned from other contexts” on the other hand. They suggest a way out of this unhelpful dichotomy, by using a conceptual framework, which recognises three groups of valid conclusions: a) generalizable to most or all situations; b) specific to a given context; and c) valid for a sub-group of situations, considered sufficiently similar. Clarifying the nature of the conclusions should prevent decision-makers from transferring a b-type conclusion to another crisis, or from transferring a c-type conclusion to a crisis of different character. The snag of this sensible approach is that it demands to decisions-makers a robust knowledge of both previous crises and of new ones, a knowledge that very few involved players hold.
Objective constraints limit the access to the required knowledge. A first group of problems lies with the nature of the lessons learned. Learning from crises is difficult, even for insiders, who are exposed to fragments of evidence, rarely assembled into a coherent picture. Thus, comprehensive lessons are rarely learned, whereas incomplete ones may be wrong or misleading. Further, the shortage of solid, widely accepted information makes most lessons arguable (particularly the controversial ones). Parties displeased by a certain conclusion find always room for demolishing it. Politics and vested interests inform and sometimes distort knowledge, hence limiting its transferability.
A second group of constraints relates to the way knowledge is managed and exchanged. Available knowledge is dispersed across agencies, research centres and NGOs. Much knowledge, held by people, remains unwritten. And written knowledge tends to get lost in the crisis environment. Language barriers, frontlines, mistrust, displacements, short assignments, all hinder the exchange of information among participants.
A third group of difficulties is linked to actors. The insiders of a health sector entering a crisis are usually not conversant with the issues, nor with the international debate related to health care provision in conflict and post-conflict settings. Conversely, outsiders, failing to understand the specific features of a new crisis, bring with them the lessons learned elsewhere, which may not apply to the new context. Researchers are in a better position to facilitate the transfer of knowledge, but they convey it according to academic etiquette, hardly the most appropriate communication tool, given the target audience. Additionally, they may publish their findings with years of delay, when decisions have been already made, and tend to remain in country only during short time spans.
A fourth group of obstacles relates to the way the aid enterprise is structured. Donor agencies have been singled out as poor learners (Berg, 2000). They shape the policy debate, often undermining learning from previous crises, by imposing corporate policies. These lack sometimes a strong empirical basis, may be inappropriate to the local context (Strong, 2003), and are usually presented stripped down of assumptions and caveats to policy-makers. Also, by fragmenting the health field and flooding it with their own priorities, donors raise the level of noise to such an extent that learning, both from inside experience and from outside knowledge, becomes impossible. Short programming cycles and assignments compound the picture and distort learning: what looked promising at a given point in time might emerge as a flop later on, when some unanticipated side effects emerge. In this way, wrong lessons may be drawn or right lessons be neglected.
The host of barriers to learning from experience is discouraging. Some of them, of a structural nature, cannot be fully overcome, but just controlled to lessen their impact. Other constraints might be addressed only by a radical change of the way the aid system operates. Hence, the most likely outcome of future crises is the familiar constellation of wrong lessons applied, right lessons ignored, insensitive decisions made and ineffective action pursued.
‘Best practice’ is unlikely to be found embodied in a specific ‘best’ policy or model. Rather, it emerges from a judicious balance of context-sensitive exploration, rational appraisal of alternatives, restrained generalisation of specific experiences. The same approach, to be applied across different crises, should lead to the drawing of different conclusions and to generating different policies. The lesson lies in the methods adopted to pursue an adequate understanding of the picture, rather than in the choices triggered by such an understanding.
For an excellent discussion of learning in the humanitarian field, see Van Brabant (1997). Annex 12 offers to the consideration of interested parties some condensed reviews of documented recovery processes. An insightful, detailed discussion of these issues may be found in Bower (2002). By looking at the variety of situations and responses, the actors of future recovery processes should be able to recognise similarities as well as differences, and make decisions accordingly. For a discussion of the country context and its relationships with the health sector, see Module 3.
Analysing disrupted health sectors: a toolkit:
| Next page