This paper sets out a novel computational method of testing the uses to which generic membership can help us understand large-scale movements in the history of ideas. It does so by taking a well-known test case, the so called ‘Adam Smith question’, as an easily identifiable (and well researched) problem in generic consistency. In brief the problem is this: Smith proposes one version of human nature based on sympathy in his Theory of Moral Sentiments (TMS) and another, completely orthogonal to it, based on self-interest in his Wealth of Nations (WN). This incoherence (if one assumes that Smith worked hard at creating a unified theory of human nature, which in itself is contestable) is said to be one of genre, the difference between moral philosophy and political economy.