This state of the art paper reviews literature published up to the end of 1994 on economic, security and environmental regionalization in southern Africa. In the field of economic regionalization a distinction is made between three main integration models, trade, integration, neofunctional integration and development integration. Furthermore, the concept of new regionalism, which is in focus for the UNU/WIDER project The New Regionalism. Implications for Global Development and International Security' is used. Existing main regional organizations are categorized in accordance with these regionalization concepts and the conclusions from the literature analysing their records as regionalization instruments are summarized.The study offers a number of conclusions as to the relevance for southern Africa of the various regionalization models. Both theory and experience show that the trade integration model is badly equipped to be the basis for regionalization in Africa, as the production structure of most African countries contradicts the assumptions on which the model is based. With South Africa finally included, a regional complementary pattern emerges, which may change the southern African situation in this respect. Still the problem of strong regional imbalance has to be handled. It should also be noted that weak financial institutions and lack of foreign exchange in a number of countries in the region will continue to constrain the intra-regional trade. The impact of the present international doctrine of openness towards the rest of the world is also discussed.The neofunctional integration model is compatible both with the existing integration pattern in the region and a reluctance among the governments in the region to become part of supra-national institutions. Strong players in the economic market have an obvious interest in this type of regionalization which can take place without much political negotiation.The development integration model is especially interesting as a potential instrument for change in a region like southern Africa, where history has created a strong regional imbalance due to the South African dominance. One of the main difficulties in the implementation of the model is often lack of necessary political commitment leading to political decisions at an early stage of the process. It remains to be proved that governments in southern Africa, including South Africa, have this commitment.The concept of new regionalism as conceived in the UNU/WIDER project is closer to development integration than to the other integration models. It is, however, a still broader concept, in which the perspective of 'development from within and below' s more emphasized and including also security and environmental perspectives.The security concept is used in it broadest definition, far from the classical realist one. In southern Africa the benefits of handling potentially destabilizing issues both to increase mutual confidence in the military sector and to contain factors such as competition for water resources, uncontrolled migration flows and organized crossborder criminality from developing into local or even inter-state armed conflicts.When it comes to regional environment co-operation southern Africa follows the rest of Sub-Saharan Africa in that most environmental activities are discussed, planned and sometimes implemented at the national level. The main exception is water management, where inter-state cooperation regarding development of river basins occur, and a more comprehensive regional perspective is needed.The development in southern Africa since 1990 suggests that the most dynamic regionalization forces are to be found among actors (private and public) in the economic field. They act quicker than intergovernmental forces, who will negotiate over long periods before they possibly come to conclusions as to how regional integration and cooperation shall take place. The capacity of state structures to implement regional integration, including creating regional structures, is weak. The resulting regionalization pattern will therefore depend less on inter-governmental action than on nongovernmental forces with their own benefit as main aim. To this shall be added that many governments' political will to integrate is much weaker than their pro-regional rhetoric.The regionalization process in southern Africa will not follow one single model. However, one factor has to be especially focused when choosing analytical instruments, namely the asymmetrical structure of the region. This means that features from the development integration and new regionalism models have to be included. It also means that governments in the region have to attend to regional issues and both the political will and the state capacity have to improve in order to use the dynamics of the market forces in a way that avoids contraproductive developments.