Reviewed by
Olivier Barreteau
Cemagref, Montpellier
This book is a plea for diversity and is focussed on human behaviour, collective action and the commons. Diversity here means diversity of methods, models and disciplines. Even if it is not a book on social simulation strictly speaking, it devotes various chapters to certain methods used by social simulation scientists (e.g., experiments and agent based modelling) and deals with collective action, which is a typical issue in social simulation.Going even further than social simulation, this book provides an informative and comprehensive survey on a large bench of methods for CPRs management. Practitioners and scholars in that field are the target of the book as they can benefit from deepening specific methods which are used in their own experience. Furthermore, the authors' purpose is to provide an outlook on alternative methods so as to convince the reader to enrich his/her work through collaboration with others. Poteete and her colleagues delve into various research methods, e.g., ethnographic, case-study comparison, meta-analysis, experiment, agent-based modelling, with detail and rich references. However, it is notably clear that the aim here is not to allow readers to learn how to master all these methods. The idea is to prepare people to work with others who handle different methods.
Obviously, the authors are conscious of certain advantages and limits of each method. However, the particular conditions for methodological cross-fertilization are still poorly explained. Fortunately, a few examples of case studies where such hybridisation has been already done and a few cross-references throughout the chapters testify that the association of methods suggested by the authors is possible and relevant. The compatibility between assumptions and conditions to use each method does not raise major concern, but readers coming from specific backgrounds should be ascertain about this problem.
For JASSS readers, who are more interested in social simulation, this book is challenging and highly rewarding. Indeed, it gives enough room to social simulation methods, reports on empirically grounded ABM, which is also a key-point for social simulation as testified by any ESSA-European Social Simulation Association conference, and raises issues which are crucial for the JASSS community, as testified by the overall interest on the simulation of social processes of science.
The cutting edge knowledge on collective action introduced in this book is particularly challenging. The authors propose a revised model of collectivel action that brilliantly emphasizes the role of context, intended as broad conditions as well as micro-situations. This calls for conceptual models capable of encompassing the diversity of possible contexts, as in case studies, and working out their consequences on agent decision patterns. Following the description of methods in section 2, meta-analysis might be path to feed empirically grounded models without making them too much specific. One can imagine back and forth processes which associate meta-analysis, agent based modelling and simulation. Agent-based simulation would provide a frame to design categories for meta-analysis, whose outcomes would allow for simulation model revision which would then possibly suggest revised categories. The authors also point out the need for further investigation on the dynamics of socio-ecological systems. This fits a general purpose of agent based social simulation. However, the authors address too much attention on a diagnostic ontological framework which is static and may pose problems to study socio-ecological dynamics. This framework presupposes ex ante knowledge of the dynamics, whereas system complexity and the co-evolution of micro-situation and collective action processes would require situated action process analysis which is difficult ex ante.
This said, this book is something that social simulation practitioners should definitely read. The new model proposed to understand collective action can be tested in relation with other social science traditions already explored in social simulation. Situated action theories, anthropology of techniques and other pragmatic approaches to tackle human behavioural patterns in group and environment social interaction might add relevant dimensions to the "together" of the "working together", eventually leading to better tuned models.
In conclusion, Poteete and colleagues end up with a rather optimistic perspective. They consider collective work for crossing methods and disciplines in the field of commons management as a collective action problem by itself. Among the conditions listed to facilitate collective action, they spot trust among colleagues and stronger institutional rewards for people being part of interdisciplinary or multi-method work. With these conditions they consider that scholars don't need to harness several disciplines or methods, but to be solid on one and capable of communicating efficiently with others. Experimental approaches and social simulation could contribute by exploring diverse settings of coordination among scholars and simulation of knowledge flows in practice.
Return to Contents
of this issue