Entropy vs. Organizational Learning and Dynamic Capabilities: The Thermodynamic Analogy

The paper offers a theoretical framework bringing together the matters of organizational knowledge, dynamic capabilities, organizational learning, knowledge creation and innovation by drawing parallels with thermodynamics and informatics and making use of such concepts as entropy, chaos and disorder, equilibrium and uncertainty. The idea of entropy originated in the XIX century as a means of explaining the basic principles governing the operation of - at the time - steam engines, but gradually developing into one of the most fundamental concepts in modern Physics. Being effectively a measure of disorder in a system (and thus, in its original sense, explaining the system's inefficiency through the dissipation of 'useful' energy), it made its way into quite a few other fields, often quite remote from thermodynamics, wherever the notions of uncertainty and/or chaos and disorder could be made use of. In informatics and cybernetics, for example, it is used to describe how the act of acquiring information reduces uncertainty (e. g., if one checks out the weather forecast, their own predictions of the next day's weather are likely to become somewhat more accurate in probability terms). In economics and sociology it was in use, sometimes avoiding the direct application of the term, from mid-XX century, appearing in the works of scholars such as Shannon and Weaver, Pareto, Capecchi, Mller, McFarland, Carvat and Kucera as well as others, leading to the formulation of the unified Social Entropy Theory in 1990 ((Bailey 1990)). Somewhat surprisingly, no examples of its use can be found in the area of organizational knowledge and capabilities; it would appear that the leap somehow has not yet been made. In this paper we attempt to bridge the gap by highlighting the analogy between an organization and a black box process with inputs, internal process and outputs, not entirely dissimilar from Carnot's engine. It, in turn, lets us draw parallels and make a number of propositions concerning the relationships between capabilities, learning, knowledge and other related matters. The resulting framework allows further elaboration in two directions, both towards the development of more advanced mathematical apparatus and its operationalization with high applied potential. Although relying to a degree on some basic knowledge of scientific concepts and making fairly limited use of mathematical notation, the paper is aimed at the general audience and, hopefully, will be of interest to scholars and practitioners alike.