In Growing Artificial Societies: Social Science from the Bottom Up, Brookings scholars Joshua M. Epstein and Robert Axtell have created a form of artificial life, existing only in a computer model, in which a variety of social and economic conditions can be manipulated and studied systematically.
Drawing on recent advances in computer modeling techniques, and a growing need to project real-world conditions far into the future, the project could fundamentally change the way social science studies are conducted in fields encompassing economics, national security, and demographics, among many others.
Viewed on a computer screen swarming with cyber-coded colonies, each vying for a share of a limited resource, the system allows researchers to watch entire civilizations as they grow, prosper, compete and eventually triumph or fail.
"We view artificial societies as laboratories, where we attempt to grow' certain social structures in the computer or in silico the aim being to uncover fundamental local or micro mechanisms that are sufficient to generate the macroscopic social structures and collective behaviors of interest," the authors write.
Life and Death on the Sugarscape
The artificial world of Epstein and Axtell is known as Sugarscape, a two-dimensional landscape containing some regions rich in a renewable resource called sugar and other relatively impoverished regions. Agents are born into this world with a metabolism requiring sugar, and with other genetic and culturally-transmitted attributes.
Movement of the agents is governed by a simple rule: "Look around as far as your vision permits, find the spot with the most sugar, go there and eat the sugar." Every time an agent moves, it "burns" sugar equal to its metabolic rate. Agents die if and when they burn up all their sugar.
When thousands of agents, each distinctive it its genetic and cultural makeup, begin to interact, "interesting things begin to happen," the authors observe. "The ecological principle of carrying capacity that a given environment can support only some finite population quickly becomes evident," they note. Skewed wealth distributions emerge; mass migrations to resource rich areas take place; social networks form; cultural groups coalesce, and these "tribes" interact in cooperation and conflict. In more elaborate versions of the Sugarscape, another commodity "spice" is introduced.
Trade: Far From Equilibrium
Now, the operating rule becomes, "Look around for another agent who has the commodity you need. Bargain with that agent until you reach a mutually agreeable price. Trade at that price."
The crude economies that result from the introduction of trade bring new possibilities and social problems to the Sugarscape. "While societies that engage in trade are able to support larger populations, trade may further skew the distribution of wealth," the authors note. "Overall, the economic markets that emerge from the interaction of Sugarscape agents do not generally reach the equilibrium of textbook economics; and these far from equilibrium economies raise fundamental questions for policy."
So-called "bottom-up" computer models like Sugarscape in which elaborate social structures emerge from the collective interaction of individuals following a few very simple rules have become practical only recently, with advances in computer hardware. Axtell and Epstein are confident that researchers will be able to tailor this new modeling approach to address a wide variety of social questions.
A CD-ROM version of Growing Artificial Societies is available. On it are 40 digital movies of the Sugarscape and related artificial societies, lasting nearly 1 hour in total. Some of the Sugarscape movies are available for downloading via the World Wide Web (http://www.brook.edu/), along with additional information and resources concerning this project.
Initial financial support for the project was provided by the John D. And Catherine T. MacArthur Foundation. Additional funding came from The Howard Gilman Foundation and The Pew Charitable Trusts.