ABSTRACT
Introduction In the last chapter we analysed the logic of self-transformation. We argued that self-transformation is based on a paradoxical decision, which can only be decided by chance. On the basis of this argument we will next ask the question: how is it possible that new self-descriptions, which are compatible with or viable within the organisation’s reproduction, come about by chance. Or, to highlight the underlying paradox: how can one explain the probability of such an improbability. This question can be answered using Luhmann’s general theory of evolution:1,2 we will argue that a decision on a new self-description does not automatically lead to the self-description being used as decision premise by the organisation. Instead, decisions on self-descriptions first pass through a selection mechanism before they are implemented. In the first section we will present Luhmann’s general theory of evolution, which is abstracted from its biological roots. This will serve as the foundation for developing an evolutionary model of self-transformation in the three further sections. Each section will deal with one of the three evolutionary mechanisms of variation, selection, and retention. 1. Luhmann’s General Theory of Evolution The theory of evolution was originally developed for explaining biological phenomena. As in the case of the theory of autopoiesis, here too we are dealing with a theory that originated in a different discipline and refers to a clearly different domain.3 Consequently, if we wish to make use of it in our specific 1 For Luhmann, the basis of the concept of evolution is the unfolding of ‘the paradox of the probability of the improbable’. See Luhmann (1997), pp. 413f.; Luhmann (1995a), pp. 433f. On this point, see also Campbell (1969), p. 73. 2 In organisation studies the concept of evolution (in its various interpretations) is frequently drawn upon. Most influential in this respect are the works by Nelson and Winter (1982) and Hannan and Freeman (1989). On the history of evolutionary thinking in the social sciences – economics in particular – see Hodgson (1993).
