Reviewed by
Matthias Meyer
Hamburg University of Technology
This book discusses epistemological and methodological issues concerning “relational methodologies” (strictly speaking, it would be better to speak of “relational methods”, as I do in the following. Methodology is already the discourse and a reflection on the methods). According to the editors, relational methods share the common basic assumption that it is more appropriate to conceptualise social and economic reality from a relational perspective by recognising that “facts and objects – and its dynamics – occur under at network form” (p. ix). The editors of this book subsume under the term “relational methods” network analysis, NK simulation modelling, neural network models and agent-based simulation modelling. The book consists of 14 chapters contributed by the editors and additional (co-)authors. The first section, which has four chapters, discusses theoretical aspects, while the second section provides in the remaining 10 chapters examples of applications of these methods.The foundation of the book’s design is the frequent reciprocal disregard of contributors to abstract epistemological debates as well as those who address the same problems, but in a more applied way by focusing on operational problems. The book is designed to bring together these two different attitudes. It is not a handbook in the traditional sense and it explicitly states that their intended readers are not beginners. Rather than an audience wishing to understand the basic concepts, methods and tools in this field, the “right reader is one who is already familiar with at least one of these methodologies and wishes to know a critical view underlying its appropriate applications and limitations” (p. xii). It is a reader who wants “to understand the deep implications of the relational nature of the methodology with whom she is familiar, and who wishes to approach the other methodologies with whom its shares the methodological trait” (p. xii).
One finds in the first section a presentation of the four key network-based methods. The opening four chapters should not be seen as a simple introduction, but more an attempt to review these methods and their usage as well as to develop them further. For example, the chapter by Lucio Biggiero on network analysis develops the interesting thesis that the delayed use of this method in economics compared with psychology and sociology can be explained by the incompatibility between the conceptual framework of the general competitive equilibrium and the relational view. Given one of the main tenets of the book, namely the strong interdependence between methods and the conceptual as well as epistemological level of science, this leads to an interesting corollary: network analysis gaining momentum in economics has deep implications for the conceptual frameworks used in this field and its epistemology. Another example is the contribution of Marco Valente on agent-based simulation modelling. He positions the use of simulation in economics and the social sciences as a tool with which to specify and explore explanatory mechanisms. This requires that simulations overcome their black box character and provide more transparency about the intermediate steps that translate input into output values. While the author discusses this mainly with respect to technical requirements (e.g. software), it also has interesting implications for the statistical analysis of simulation models.
These are just two examples of the types of stimulating explorations the reader can expect from this book. In addition, it provides in the second section a selection of chapters on different topics, although the applications emphasise phenomena such as industry clusters, R&D collaborations and inter-organisational knowledge flows alongside some interesting contributions to organisational theory and team performance.
In sum, this book provides both an overview and a deeper understanding of these four relational methods. In this respect, I would like to commend the editors for defending a position of methodological pluralism and their related effort in developing nine criteria for deciding which method is most appropriate for the problem at hand. The book is well written and I would highly recommend it to JASSS readers interested in the kinds of methodological and epistemological considerations outlined in this short review.
Return to Contents
of this issue