Ulrich Witt (2005) recently emphasized that economic theories still have difficulties in understanding "why" and "how" organizations change. On the one hand Neoclassical approaches and New Institutional Economics based on the equilibrium notion, tend to focus on what organizations should do, but ignore the question of how they reach it. On the other hand, evolutionary economists tend to forget the intentional dimensions of change by focusing on fitness and adaptation. With the first perspective, change is considered as a goal without process; with the second, it is an infinite process that barely has a goal.
Evidence captured by empirical analysis may change this perspective. As Marta Feldman (2000; 2003) or Benedicte Reynaud (2005) present it, routines are not performed by blind agents, but "by people who think and feel and care" (Feldman 2000, 614). Even if we accept Nelson and Winter's (1982) analogy, and if we compare routines with genes, it seems necessary to integrate the fact that unlike biological processes, economic mutations do not appear randomly (Metcalfe 2005).I Obviously, they are often the product of a conscious intention. As Witt (2005) explains, changes in an organization are not only conducted by a mere adaptation to external conditions, but also rest on strong internal considerations where hierarchical decisions have an important role to play. This perspective may help to analyze the cognitive and intentional dimensions of change in organizations.
It is interesting to note that more than a century ago, Thorstein Veblen (1898) was aware of the importance of human intentions in evolution, emphasizing the "teleological" aspects of economic actions. For Veblen, "the change is always in the last resort a change in habits of thought" (1898, 391). Following this perspective, it seems necessary to restore the teleological aspects of human behavior in modern evolutionary analysis. In other words, to escape from a strict Darwinian analogy and to analyze human changes as goal-oriented processes.
The aim of this paper is to contribute to the above mentioned perspective by characterizing two different types of change processes that have features of both "equilibrium" and "evolution.". In doing so, we will follow Pierre Garrouste (2007), who considers that these two economic conceptions may be more complementary than they are substitutable. The first two sections study the dynamic equilibrium process and the teleological processes, respectively. The third section discusses the implication of this typology to the study of organizational learning, and concluding remarks are provided in the final section.
Learning and the Dynamic Equilibrium Process
Learning can be defined as a process of permanent changes in behavior that result from environmental interactions (Lazaric, Monnier and Paulre 1995; Garrouste 1999). Two kinds of learning are compatible with this definition. Behavioral learning is a learning process that does not modify the main features of the cognitive system. Learning to bike is an example of behavioral learning. Even if it implies a small cognitive dimension, it does not change the biker's main intellectual conceptions (its representations system). In contrast, cognitive learning is a learning process that modifies the agent representations and interacts with its specific visions or preferences.
Learning appears as a typical intermediate concept that can fill the gap between the equilibrium and evolutionary approaches. On one hand, it needs the evolutionary approach to clarify its process in terms of path dependency or irrevocability (Georgescu-Roegen 1971); on the other hand it at least needs the equilibrium logic to understand its final perspective, its rules or its goal. This feature is important in terms of distinguishing a learning process from a random change. Learning is an organized change. For this reason it is a process that follows rules, and that needs pedagogy. The final results of a learning process may be infinite, but they are not without...