Recent methodological contributions by John Davis (2006; 2007) have claimed that economics is becoming more pluralistic. In place of the monistic project of neoclassical economics a number of new research programmes have appeared: game theory, experimental economics, complexity economics and neuro-economics to name but a few. Tony Lawson (1997; 2003; 2007) and other proponents of critical realism in economics have challenged this optimistic picture. According to them, mainstream economics remains a monolithic enterprise dominated by deductivist modelling and an atomistic ontology. The division between the mainstream and the hetorodoxy, manifested in divergent methods, is the result of a fundamental incompatibility between the ontological presuppositions of heterodox and mainstream theories.

Recent developments in game theory provide a ready-made example for comparing these opposing methodological conjectures. This is not only because game theory is one of the areas cited by Davis as indicative of the new economic pluralism, it is also because game theorists have become interested in phenomena that have, until recently, exclusively of concern to heterodox economists. Convention is one such phenomenon, and it has received increased attention in game theory, notably through the work of Peyton Young (1993) and Robert Sugden (1986) who build on the early contribution of the philosopher David Lewis (1969).

For some time, discussions of convention has been tied to the analysis of John Maynard Keynes’s economic and philosophical writings. More specifically, convention has been studied by ‘radical Keynesian’ economists’, building on the Treatise on Probability (1921), Chapter 12 of the General Theory (1936), and Keynes’s Quarterly Journal of Economics article (1937). Perhaps unsurprisingly, the two literatures have very little overlap and mainstream game-theorists make sparse references to Keynes if any. Apart from the lack of dialogue between different theoretical accounts, our discussion is hampered by the absence of an agreed or ‘standard’ definition of convention in economics. We claim, nevertheless, that at least four general features of convention are widely accepted by economists. These are:

1. Conventions are arbitrary
2. Conventions involve regularities in behaviour
3. Conventions involve co-ordination between agents
4. Conventions are responses to uncertainty

There is little dispute about the significance of features 1-3. Uncertainty, on the other hand, has been interpreted in radically different ways and has been the locus of fierce debate between the heterodoxy and the mainstream in economics since the early part of the 20th century (Knight, 1921).

We contend that the controversy surrounding uncertainty is unsurprising as the Keynesian conception of uncertainty is the key to explicating the divide between the heterodox and mainstream theories of convention. Developments in the study of uncertainty within post-Keynesian economics have made explicit Keynes’s conception of ‘true uncertainty’ in terms that distance post-Keynesians from the mainstream view of uncertainty as risk. In a parallel development, EC economists have studied how this uncertainty transforms social practices. Their institutional focus has challenged the mainstream view of conventions as solutions, adopted in order to ‘reduce’ uncertainty and stabilise individual expectations.

We conclude by reflecting on what these contrasting approaches to convention reveal about the state of pluralism in economics and the distinctions between heterodox and mainstream approaches.