INTRODUCTION

In chapters 1 and 2 one conclusion was that for a foundation of innovation policy we need to move beyond macro analyses of factor productivity and into the innovation system, with micro-level analyses of actions and interactions of the actors involved, in the context of institutions and institutional change. R&D does not directly yield innovation, but a pool of ideas and emergent technologies, products, and practices, which require testing, design, commercialization, marketing, and the development of production. This requires entrepreneurship and organisation, internally within firms and externally, within networks of firms. R&D may also serve to explore new developments and opportunities, build up absorptive capacity, knowledge on sources of knowledge and contacts for future collaboration. The innovation process is not linear, but entails feedback from testing, as well as trial and application to the generation of new knowledge. In this chapter, we attempt to contribute to further insight on the micro level of actions and interactions, from the perspective of learning by interaction, which is also central in the literature on innovation systems (Lundvall 1988).

As in the literature on innovation systems, an important source of inspiration here is an evolutionary perspective, which recognises the fundamental importance of uncertainty and unpredictability, in the emergent nature of innovation, and the crucial role of diversity as a source of novelty. This in itself has important policy implications, as it points to the limitations of rational planning and design of innovation. However, we also recognise that learning is inherently social, and hence we should derive insight also from theory of cognition, communication, language, and motivation, and these do not necessarily conform to details of Darwinian evolution as found in biology. In other words, while cultural evolution may derive important insights from neo-Darwinian theory, it is not necessarily isomorphic to it.

This chapter outlines and elaborates a ‘cyclical model’ of innovation, where the development of new knowledge and new practices (exploration) and the improvement of existing practices (exploitation) build upon each other. A key theme is how such a combination can be achieved, in spite of fundamental differences between exploitation and exploration, and between the conditions of markets and organisation that they require. Among other things, the analysis offers a deepening of our understanding of why collaboration furthers innovation and the positive role of differences in insight (‘cognitive distance’).