related to social upheavals, internal conflicts or international wars also exist. They are, in fact, the basis of intelligence and diplomatic activities since centuries.

However, most of the early warnings issued in the last decades have not been taken seriously, or responded to, by the international community. In many instances, they were totally neglected. More sadly, some of them have been purposely not listen at. Reports on the developing famine in Ethiopia were available and circulated months and months before the situation turned into the most well-known famine of the mid 80's (the WFP/FAO donor conference on Ethiopia had been held without any response until the BBC footages showed the developing horrors at Christmas 1994). When early warnings are related to ethnic or social crisis, they have either not been noticed (Rwanda, in the early nineties a few years before the genocide actually took place) or been followed by inappropriate measures (Kosovo, which was raising a lot of concerns from the "educated observers" since mid-80s was not included in the Dayton agreements). The logical consequence of an early warning, which is a rapid response, has in most instances not materialised.

However, relatively simple cost effective measures could have been taken in terms of setting aside local seeds, and increasing local, national or regional areas under seed production. More important, strategies for in situ protective measures on the bio-diversity could have been implemented in a certain number of cases. In fact, most of these measures should become full fledge parts of any agricultural strategy aimed at promoting sustainable development.

Available information systems relating to disaster situations are plentiful. Yet they are not used properly, i.e. they are rarely listen to. Most of the information required is available, but often political priorities prevent a proper use of them. There is, nevertheless, room for improvements, at least in the way pre-crisis information can be collected and used before the disaster strikes.

The problem with preparedness measures is that they are an investment which will only pay off if and when the disaster actually strikes. This explains why neither local authorities nor Donors are very keen to finance them. Unfortunately, when disaster strikes, it is often too late. Then costs run high, appropriateness of the response is often limited and suffering reachs peaks that could have been avoided.

There is, however, a lot of confusion behind the term "preparedness". It is often understood only from a logistical angle. Pre-stocking of goods (and in relation to this paper - seeds) and transportation means is of course important. However, preparedness is more a state of mind than a physical process. With this overall acceptation of the concept, preparedness has to do with a few key strategic factors which include, but are not limited to :

In many man-made disasters, there will be no Ministry of Agriculture to deal with, no technical decision makers available for discussion and agricultural recovery. Farmers could be the only source of information. The thoroughness, but also the limitation of this information (site specific, little global view, limited knowledge on either the world seed system) the absence or excess of knowledge of the humanitarian business can lead to very biased assessement. By chance, there is no region on Earth which has not been studied, especially disaster prone areas. How can this information be collected, compiled and made available before the crisis, so that when it occurs, useful information is immediately accessible? Information systems connected to the INTERNET are one of the most recent assets of the international community. More and more, people's groups, NGO, local institutions can get "on the net" to look for required pieces of knowledge they might need. Structures like FAO/GIEWS or specialised centres can insure that this information is available in a "ready to use" form. In addition, trusted and tested solutions often do exist. What counts is the availability of the information about them. Local seed banks have been set up in Western Africa and in India. Trial and error programmes on that subject have already been in place in various areas of the world. There is no need to repeat the same errors, but avoidance of doing so depends largely on whether or not the actors are aware of these lessons, and make use of these lessons learned.

The establishment of seed stocks or lists of useful appropriate seed varieties and seed supply sources requires competent, dedicated and convinced staff. The human resource factor has often been seen as one limiting very much the efficiency of programmes. Training on matters related to seed issues with hands-off sessions for staff already working and specialised modules included into agricultural schools and universities curriculum can easily be organised.

Developing inventories which include a listing of local varieties with their characteristics and site specific adaptability will be crucial. One has to remember that in situ varietial conservation is often more efficient and cost effective that the establishment of costly national or international ex situ conservation systems: national seed banks require not only a sophisticated infrastructure of cold rooms, refrigerators and packaging units, but also agricultural stations for regular regeneration of the seed stocks. These national systems are, however, still necessary, since they could permit national inventories and co-ordination of overall biodiversity conservation.

International systems have also shown their usefulness in certain contexts. Most of the rice varieties of Cambodia were stored at IRRI. During the Khmer Rouge era (1975-78), some important varieties where almost lost (especially floating rice brands). In the mid 80's, IRRI managed to bring back some of these cultivars which then could be re-multiplied. In Angola, the national collection of sorghum, maize and beans was located in the premises of the Huambo National University of Agriculture. The military occupation of this town during the civil war and the destruction of the compound of the university resulted in the destruction of the collection. If some samples of the Angola varieties would have been provided to the regional phytogenetic resources conservation programme, the losses would not have been so detrimental. Now, not only the collection has to be started from scratch, but it is not sure if all of the varieties will ever be recovered. The decades of famines and population movements have taken a high toll on the national phytogenetic heritage.

Establishment of village based seed banks, with the necessary training and social work in order to sensitise the population of the importance for them to keep these banks in good conditions and to use and resupply the stocks is needed, and requires appropriate approaches.

These preparedness measures are especially important when a disaster has been followed by large scale displacements of rural populations. Indeed, when the refugees or internally displaced people go back to their areas of origin, appropriate types of seeds are often not available. Since most of these events (e.i. return and repatriation) are often predictable (this has been the case in most instances, apart of the recent massive return of Rwandan refugees from former Zaire), appropriate establishment of stocks of local seeds could have taken place. In Cambodi, for instance, the repatriation preparation process took more than a year. Alas, only very few people mentioned beforehand the importance of pre-stoking of Khmer traditional rice varieties. When the refugees came back, they only received "modern varieties" often very vulnerable to flood or drought. It did not help the economic re-integration of these returnees nor the establishment of an acceptable level of food security. This is the kind of failure which underlines how weak the international community’s preparedness response has been in recent times.

With a bit of proactive thinking, many steps can be achieved long before the emergency.

During the post conflict or post disaster period, access to traditional varieties will be crucial to re-establish farmer seed stocks and restore risk minimisation mechanisms based on socially-owned knowledge. Pre-crisis establishment of national seed banks and/or bio-diversity genebanks will be crucial for the post-disaster recovery. Regional co-operation is needed. The connection of national strategies and perspectives with regional ones will help to distribute the risk and share costs. Experience in Angola has shown that this can be done in a cost effective and efficient manner, provided that certain means are available and matched with a real will of local actors.

We also know that, in post-disaster periods, the absence of law and order can facilitate all kind of pillage. Seed companies can jump to collect samples of seeds they will use in their own research work without even mentioning the origin of the genetic material. Will the "protective components" enshrined in the Convention for the Protection of Biodiversity shall be actively defended in post disasters contexts? After all, countries back from the brink should as well be respected and their genetic resources protected.

Among the many factors contributing to the success of humanitarian disaster response operation in general and of sustainable agricultural recovery through seed distribution, one of the most important is the quality of the diagnosis.

This diagnosis depends, among other factors, primarily on:

• A proper understanding of the nature of the disaster, of its characteristics;
• A thorough analysis of the system that prevailed before the disaster took place, including the production processes (in terms of timing, requirements, relations with other sectors, etc.);
• An in-depth assessment of the way the disaster has affected life in the rural areas, including the agricultural cycles; and
• An identification and appraisal of the adaptive mechanisms and survival strategies put in place by the affected populations.

AUBERTIN, C. 1998. Les enjeux de la biodivisersité. Poche Economica, Paris, 112 p.

BESSET, J.P. 1992. René Dumont: une vie saisie par l’écologie, Collection"au Vif", Stock, Paris, 375 p.

CARBONNIER, G. 1998. Conflict, post war rebuilding and the economy: a critical review of the literature. The war torn societies project. UNRISD, Geneva, 76 p.

FAO. 1996. Conservation and sustainable use of phytogenetic resources for food and agriculture. FAO documents prepared for the World Food Summit, Rome, Italy.

FAO. 1997. Contributions for the development of seed security strategies in Disaster-prone Regions. Proceedings of the Florence International Workshop on seed security for food security, 231 p.

FAO. 1998. La production des semences de qualité déclarée au Rwanda; Kigali/Rome, 83 p.

GARY, J.M. 1995. Ethnobotany: a people and plants’ conservation manual. WWF/UNESCO, Chapman & Hall, 268 p.

GRALL, J. and LEVY, B.R. 1985. La Guerre des semences: quelles moissons, quelles sociétés? Ed Fayard, Paris, 410 p.

GRUNEWALD, F. 1985. Le riz et le sucre: manuel méthodologique pour la recherche développement en agriculture dans le sud-Thaïlande. GRET, Paris, 26 p.

GRUNEWALD, F. 1986. Alarme précoce, aide d'urgence et développement, quelques idées pour une stratégie. Bangkok, Asian Disaster Prepardness Center, 1986, 57 p.

GRUNEWALD, F. 1989. "Stratégies paysannes à la rescousse de l'autosuffisance". In: Affaires Cambodgiennes. Paris, L'Harmattan, ; pp.149-174

GRUNEWALD, F. 1993. "Pêche, forêt et agriculture dans un Cambodge à peine sorti de la guerre civile". In: Revue Tiers Monde, XXXIV, N134.

GRUNEWALD, F. 1993. When the rain will return: emergency assistance, rehabilitation and development. Text presented at the Conference Colloque "Développement, il y a urgence, Lyon", November 1993, 13 p.

GRUNEWALD, F. 1995. "Action before, during and after the crisis: the experience of ICRC in retrospect". In: International Review of the Red Cross; N° 306, May-June 1995; pp. 263- 281

GRUNEWALD, F. 1995. "Protection of biodiversity in times of war and during post conflict periods". Report of the first International Conference, Huambo (Angola) July 1995, ICRC, 25 p.

GRUNEWALD, F. 1996. "The other cost of war: protection biodiversity in times of conflict". In: Plant Talk : plant conservation worldwide, N° 822, Novembre-Décembre 1996; pp. 633-654

GRUNEWALD, F. 1996. "For or against food aid". In: International Review of the red Cross; N° 312, October 1996; p. 15

GRUNEWALD, F. 1997. Au delà de la survie: concepts et pratiques de la réhabilitation agricole au CICR. "Working papers of the Relief Division", Genève, 27 p.

MACRAE, J. and BRADBURY, M. 1998. Aid in the twilight zone: a critical analysis of humanitarian-developement aid linkages in situations of chronic instability. UNICEF, New York, NY. 75 p.

MAZOYER, M. et al. 1998. Histoire des agricultures du monde: du néolitique à la crise contemporaire. Seuil, Paris, 531 p.

ODI. 1997. Fournir des semences pendant et après les situations d’urgence. Programme de semence et de biodiversité. Revue de l’Etat des connaissances N°4, London, 151 p

ORSTOM. 1989. Le risque en Agriculture. Collection: A travers champs, ORSTOM, Paris, 619 p.

SPERLING, L. 1996. Rwanda: Household survey on status of potato crops. Seeds of Hope, ISARD, Kigali, 27 p.

SPERLING, L. 1996. Rwanda: Nation Household survey on status of beans, sorghum and Cassava crops, 1995: the impact of war and varietal erosion, Seeds of Hope, ISARD, Kigali, 60 p.

SPERLING, L. 1996. Executive summary and reflection of the Seeds of Hope socio-economic analysis in Rwanda: the impact of war on agricultural production. Seeds of Hope, ISARD, Kigali, Rwanda. 24 p.

USAID/DESFIL. 1996. Seeds for disaster mitigation and recovery in the Greater Horn of Africa. Chemonics International, Inc. Washington, D.C. 137 p.

Notes