Probabilistic political economy in geographical contextIn July 2008 a number of economic geographers presented papers at an interdisciplinary conference organized to celebrate twenty-five years since the publication of Farjoun and Machover's seminal book, Laws of Chaos: A Probabilistic Approach to Political Economy (1983). Within the political economic community this work represented a break from the hitherto deterministic models of`classical' political economy that had dominated discourse over the theory of value as the determinants of prices, outputs, and profits in capitalist economies (Goodwin and Punzo, 1987). Specifically, Farjoun and Machover's innovation was to introduce probabilistic reasoning and statistical mechanics to the field of political economy. Accordingly, the conference concentrated on four main themes: (1)``Laws of chaos'', reflecting on the impact of the book, (2)``Theory and methods'', exploring the concept of statistical equilibrium in political economy, (3)``Models and empirical reality'', connecting models in probabilistic political economy to empirical reality, and (4)``Disequilibrium and out-of-equilibrium dynamics'', examining the properties and empirical plausibility of probabilistic models of capitalism. More recently, these themes have received a good deal of attention within the broader political economy community, with both theoretical innovations and empirical testing being conducted from the perspective of a probabilistic reasoning. For example, political economists have begun to employ notions drawn from`econophysics' to explore the nature and existence of statistical equilibrium as an alternative to general equilibrium models, money, income and wealth, and the theory of value (Cockshott et al, 2009). Economic geographers have a longstanding and, nowadays, sometimes forgotten tradition of employing probabilistic mathematics, statistical mechanics, and information theory to understand the random space economy (Curry, 1998). Early work on spatial interaction, spatial structure, and their integration used potential models and statistical mechanics to account for spatial patterns (Bennett et al, 1984). Subsequently, these models were extended to incorporate dynamic adjustments between pattern, flows, and process through the use of systems models of catastrophe and bifurcation. More recently, spatial modelers have begun to employ notions drawn from evolutionary dynamics, complexity, and agent-based models to understand the evolving economic landscape (Wilson, 2006).Whilst geographical political economists have tended to eschew mathematical and statistical models, there are some researchers ödubbed`regional political economists'ö who have focused on modeling economic dynamics in terms of evolving interdependencies between heterogeneous sectors and regions, where aggregate prices, outputs, and profits are conceptualized as the mean of the distribution of these variables. Within this tradition, Farjourn and Machover's work had an early impact through the work of Sheppard and Barnes (1...