Learning to bridge the gap between science-based decision support and the practice of farming: Evolution in paradigms of model-based research and intervention from design to dialogue

Application of science in agriculture has been primarily to the ongoing improvement of material technologies. But there has also been an expectation that science, in a 'systems' mode, could, and should, contribute to improved planning and decision making by farmers. For over 30 years computerised models of farm economies or production systems have claimed the ability to identify superior alternatives for management action. Over this period, model competence has improved immensely, and farm computer ownership has grown to high levels, but this has not generated conspicuous or sustained enthusiasm among farmers or their advisers. This paper examines the experience of model-based interventions in farming practice in search of insights to both past failure and future possibilities for models with seemingly impressive capability to be relevant and significant to managers in the challenging task of achieving sustainable farming. The strategy is to 'stand back' far enough to see conceptual and historical 'connections' between research and farming from a vantage point where the difference between a systems view and a philosophical view becomes indistinct. An adaptation of Karl Popper's 'three worlds' model serves as a 'map' of the differences among types of knowledge and among three paradigms for scientific intervention in practice. This aids explanation of the 'gaps' between research and practice when: (1) researchers design 'best practice' for practitioners using theoretical models, (2) researchers provide practitioners with practice-guiding tools, and (3) researchers with theory and models collaborate with practitioners to research 'best practice' in the context of practice. The benefits and challenges of an approach for bridging 'the gap' which uses multiple rationalities and research paradigms are discussed. Additional keywords: systems, DSS, action research. AR 9 R.c R. L. c Br thwe o ar