Some weeks ago when I sketched out an approach to modeling literary history (in Underwood and Sellers 2015: Beyond narrative we have simulation) I hadn’t yet read Michael Gavin’s Agent-Based Modeling and Historical Simulation (Digital Humanities Quarterly, 8.4, 2014). Now I’m wondering what it would take to transform my sketch into an agent-based model.
Here’s Gavin’s basic characterization of agent-based modeling:
Agent-based modeling, sometimes called individual-based modeling, is a comparatively new method of computational analysis. Unlike equilibrium-based modeling, which uses differential equations to track relationships among statistically generated aggregate phenomena — like the effect of interest rates on GDP, for example — ABM simulates a field of interacting entities (agents) whose simple individual behaviors collectively cause larger emergent phenomena. [...] Unlike text mining, topic modeling, and social-network analysis, which apply quantitative analysis to already existing text corpora or databases, ABM creates a simulated environment and measures the interactions of individual agents within that environment. [...] the intellectual work of ABM centers on identifying the relationships among individual rules of behavior and the larger cultural trends they might cause.
In this way, agent-based modeling is closely associated with complex-systems theory, and models are designed to simulate adaptation and emergence. In the fields of ecology, economics, and political science, ABM has been used to show how the behaviors of individual entities — microbes, consumers, and voters — emerge into new collective wholes. John Miller and Scott Page describe complex systems: "The remarkable thing about social worlds is how quickly [individual] connections and change can lead to complexity. Social agents must predict and react to the actions and predictions of other agents. The various connections inherent in social systems exacerbate these actions as agents become closely coupled to one another. The result of such a system is that agent interactions become highly nonlinear, the system becomes difficult to decompose, and complexity ensues" [Miller 2007, 10]. At the center of complexity thus rests an underlying simplicity: the great heterogeneous mass of culture in which we live becomes reconfigured as an emergent effect of the smaller, describable choices individuals tend to make.
And that, of course, is what I was after. But rather than just setting the simulation in motion and watching for emergent effects, I want to see if a particular emergent effect – a long term directional shift in the properties of texts – can be generated from purely local interactions.
Once I’d read through Gavin’s paper, which runs through some models (and which I recommend), I remembered that Graham Sack had something about simulating the emergence of literary genres at DH2014 and that I’d excerpted his précis in a post from last year, DH2014: Computing the Literary Mind – Look at This! You can also view his video:
Since there’s a bit of text to read on the screen, you might want to view it over at Vimeo rather than in this narrow column.