ABSTRACT
In the literature on organisational learning a distinction is made between first and second order learning (Hedberg et al. 1976; Fiol and Lyles 1985) or, equivalently, between ‘single loop and double loop’ learning (Argyris and Schön 1978). The first is learning to do existing things better (more efficiently) and the second is learning to do new things. This is linked with the notion of ‘parametric’ change (Langlois and Robertson 1995) as opposed to ‘architectural’ change (Henderson and Clark 1990). Also related to this, March (1991) and Holland (1975) distinguished between ‘exploitation’ and ‘exploration’. In order to survive in the short term, firms need to exploit their present resources (or competencies or abilities) efficiently, and to survive in the long term firms need to develop novel competencies, to anticipate or create future market conditions. This combination of exploitation and exploration is arguably the main challenge for management, but it is a paradoxical task. To a greater or lesser extent, depending on the type of product, market, technologies and types of knowledge and competencies involved, exploitation requires fixed standards, routinisation and tight co-ordination, while exploration requires a loosening of structural ties and conditions. In this chapter I aim to contribute to the solution of this paradox. For that I make use of a recent book (Nooteboom 2000).
