Customer Reviews

Growing Artificial Societies: Social Science from the Bottom Up (Complex Adaptive Systems)

5 star:		(2)
4 star:		(2)
3 star:		(1)
2 star:		(3)
1 star:		(0)

 

 

This book is not a "how-to" book. They do not provide all of the code for thier sugarscape model. Yes, they provide some snap-shots of code for the reader, but those are instructive as to how to organize one's own code for your own ideas and models. If you want the entire code go to Swarm or RePast web pages and look for it in objective C or Java.

I was introduced to this book in a graduate archaeology course. Now, 3 years later I've returned to it for my dissertation. What this book does it explain how simple rules and ideas can create rather complex outcomes. What are the affects of having agents vision be only 5 cells compared to infinite sight? Can simple biological questions such as resolution of vision have a profound affect on our social structure? There are a bunch of, respectively, simple questions that this book address or introduce to explain the power of this method for the social sciences.

If one is looking for a "How To Book" you should go to Ascape, RePast, Swarm, or any of the other agent based modeling software research groups. What this book does is provide the reader with the conceptual issues and the foundation for what this method can do, that's it.

This book was part of a graduate research class I was in. We built thier simulation from the ground up, but found many errors and simulation artifacts with in the book. Though the simulation was a very good one, they left or ignored key details, and the book only discusses the conceptual model. Building the model from the information in the book can be an exercise in futility. They do not give much detail, and what they do give, they hide within footnotes and seperate critical information with pages of analysis. The alanysis unfortunately doesn't talk about model deficiencies and other simulation artifacts the modelers introduced. In the end, an excellent simulation, regardless of how they put it together, and the errors their model injected into it.

The authors do an impressive job of demonstrating how agent based simulations can be applied to social systems. In the past, modeling of this sort was limited to traditional analysis techniques such as applied differential equations. While some are critical of this work because they point out the number of assumptions inherent in this model, they also neglect to consider the greater degree of assumptions and over-simplifications implicit in pure mathematical models (eg, linearity, continuous functions, etc.) An advantage of agent based modeling is that one can consider all sorts of rules which do not lend themselves to purely mathematical models. Consider queuing theory as an example. While there exist basic mathematical models for queue analysis, once a certain threshold of complexity is reached, these models fail, and one must look to computer simulation as the alternative. While their results are speculative, the authors have successfully demonstrated emergence of complex behavior from simple rules. One such example is an unexpected diagonal migration path emerging from an orthogonal movement rule.
In the future, this type of social modeling will be the accepted norm and practitioners will look back at this work as a foundational reference.

Granted, this is not a cookbook for creating the simulations described. However, it gives a good picture of the power of agent simulations, and shows the basics of behavior modeling. In this respect, it is an excellent text. I would suggest it for an advanced undergrad course, rather than graduate level.

The idea of the "generative" approach to social science, as described in this book, is that we attempt to understand the workings of a society not through "discursive" models (as in qualitative social science) or through game theoretical models with extremely simplified assumptions and homogeneous agents (as in more "quantitative" approaches), but by creating a simplified version of the society and letting it "run." So we use software (or pen and paper; Schelling's pioneering models of segregation were studied without the benefit of a computer) to generate an environment and simplified agents that interact with one another, and then we see what happens. The advantages of this approach is that one can add complexity to a model in a controlled manner and study the resulting dynamics (rather than merely static equilibria); so, for example, one might start with agents who move and eat, then see what happens if you add the ability to trade, and so on (besides, it's fun to play around with such models). The disadvantages, however, come from the very flexibility of the approach, which allows for highly complex models: it is sometimes hard to tell whether or not a particularly interesting result is simply an artifact of the simulation or the consequence of some particular simplifying assumption (though it should be noted that the same is often true of traditional models, where striking results, like the pareto-optimality of free markets, are sometimes basically artifacts of the simplifying assumptions made for the sake of creating "tractable" models).

Epstein and Axtell report in this book the results of a pioneering 1994 simulation: "Sugarland". They start with a very simple environment, the "sugarscape", consisting of a two-dimensional surface (technically, a torus) with an unevenly distributed resource ("sugar") and add some very simple agents who can move around the sugarscape according to simple rules and "eat" the sugar. As they add more complex rules, they basically "generate" a number of features of societies - like migration patterns, cultural transmissions, combat, trade, credit relationships, disease transmission, etc. - and study how some simple rule changes affect these patterns. The book does not describe in detail how to do this (they give some information that might help in replicating the simulation, but no code); instead, one can read it as a kind of advertisement for the generative approach to social science. Epstein and Axtell excitedly argue that many phenomena can be understood as "emergent" effects of the interaction of agents following simple rules, and hence that the best way of understanding them is by looking at which simple rules are able to generate them. In general, this idea seems reasonable; it is unlikely that we will ever fully understand polities and economies through models composed of homogeneous, fully rational agents. We will need to study complexities that become tractable only through simulation.

But though Epstein and Axtell's results are often suggestive, it is not always clear that they have properly looked at the ways in which minor changes in the parameters of the simulation might disrupt the patterns they find, or sufficiently thought about how to interpret the results of their simulations. As a result, the book is somewhat disappointing: if you are already convinced of the utility of the generative approach, you may not learn much here, and if you are not, then you may think that the authors have not really addressed the important objections to the generative approach. A good example is in the chapter of the book on combat and cultural transmission. Here the assumptions made about how to model cultural transmission and combat between agents seem rather arbitrary (rather than based, for example, on research about human or animal combat), and though they explore a number of alternative specifications, the results seem only lightly grounded theoretically, more an artifact of the simulation than an illuminating model of actual cultural transmission or combat. (By contrast, the chapter on trade and credit is probably the most solid in the book, since they ground the results in economic theory and systematically explore some of the space of alternative rule specifications and parameter values for their agents. Nevertheless, their results there do not go beyond what most sophisticated economists already knew, and do not add much knowledge about the properties of dynamic adjustment in markets - we get no bubbles or other interesting phenomena, for example, and we would need to introduce something like production functions to make the model better reflect actual economies).

Of course, the book is 16 years old right now, and new advances have been made in agent-based simulation and "generative" social science. Besides, as I mentioned above, the book is perhaps best read as an advertisement for the virtues of the "artificial societies" approach rather than as a contribution to the study of actual societies; it opens your appetite rather than satisfies it (I wanted to do some programming as I was reading it, to try to see if I could replicate their results). But it is probably fair to say that the book does not quite succeed at showing that the artificial societies approach is actually worth taking seriously, unless you are already convinced of its merits.

This book is an opportunity missed. The subject is interesting (and contrary to the views of another reviewer, I think there is valuable research being done here).

The model seems to be well thought out, although its very limited scope (a 50 by 50 playing field) makes me almost sure the results can have little meaning. I was continuously troubled by the fact that they described their world as a torus (wrap-around like a doughnut) but none of the illustrations supported this. I didn't buy the version with the CD-ROM, but frankly, I'm glad I saved my money.

Moreover, at almost every paragraph, I felt the authors had contrived the result they desired.

For a much more stimulating read, try "Turtles, Termites, and Traffic Jams : Explorations in Massively Parallel Microworlds" by Mitchel Resnick,

This short book is written by the pioneers of the agent-based modeling technique. It describes their pioneering model, starting with the most basic iteration. Each chapter describes the process of adding a layer of complexity to the model, and the results therein. It contains no "errors" or "artifacts." It is a "how to book," for those with at least a passing interest in social science and the ability to download shareware from the web (i.e. NetLogo). Any person with an interest in this field must be able to discuss this book, as well as the author's Anasazi model, among others.

The following is from the Sept 1997 issue of "Doctor Dobbs Journal", also available at the Electronic Review of Computer Books (www.ercb.com/ddj/1997/ddj.9709.html): Cellular automata can indeed generate complex behavior; the problem is, how do you determine what, if anything, that behavior means? A pendulum is billions of simple entities (atoms) interacting through simple rules (electromagnetic forces and gravity); does that mean that the swinging motion of a pendulum tells us something profound about the economic cycle of capitalist economies? By changing the parameters in the authors' "Sugarscape" worldlet, you can get its little agents to migrate, to trade, and so on. But what the authors don't report is how many combinations of parameters they tried that didn't produce behavior that could be given an intriguing label...in short, all the things you would need to know to judge for yourself how significant their results really are. ..."Growing Artificial Societies" is an example of "cargo-cult science." Its authors enact the rituals of science without seeming to understand the reasons for those rituals