"Organizational Routine and Intraorganizational Ecology"(in Japanese)

The purpose of this paper is to contribute to a more complete understanding of organizational learning during 1980s through 1990s. According to Huber (1991), an entity learns if the range of its potential behavior is changed through its processing of information, and then organizational learning is characterized in terms of four attributes: existence, breadth, elaborateness, and thoroughness. Related to organizational learning, Huber articulates four constructs: (a) knowledge acquisition, (b) information distribution, (c) information interpretation, and (d) organizational memory, but organizational memory is much in need of empirical study although the basic processes of (a), (b), and (c) depend on organizational memory. After the review of Huber, Cohen (1991) discusses the contact points of organization theory and contemporary cognitive psychology, and his key research idea is the distinction between declarative memory and procedural memory. Cohen & Bacdayan (1994) report an experiment in which paired subjects developed interlocked patterns and the evidence that the patterns have the chief characteristics of organizational routines and that their components are stored as distributed procedural memories. Moreover, Simon (1991) predicts and Cook & Yanow (1993) show through the case of three small workshops that a persistence of pattern survives a replacement of the individuals who enact the pattern. By piecing together new facts and ideas, we view an organizational routine as a system of interlocking, reciprocally-triggered sequences of skilled actions stored in a form of procedural memory. And the persistence of pattern of this system survives a replacement of its elementary individual memory. These propositions fall into line with the routine-based organizational learning theory characterized by both the emphasis on routines and the emphasis on ecology of learning (Levitt & March, 1998). Miner (1991) examines the survival of one type of routine through empirical study. But the ecological structure of organizational learning complicates the systematic modeling of learning process, and analytic work by using computer simulation was more common during the 1980s. This paper describes and criticizes the computer simulation model of mutual learning developed by March (1991). His simulation suffers fatal defects in the formulation of the model. Furthermore, he misinterprets non-equilibrium lock-in as equilibrium and he draws a wrong and opposite conclusion. We formulate an Excel version of March's simulation model in order to trace the causes of lock-in phenomena and equilibrium in the intraorganizational ecology of the persistence of organizational routine is a necessary condition to obtain a good organizational performance. Other findings confirm our conclusions.