A central problem for this part of the book concerns social norms, their ontological character, our knowledge of them and its methodological implications. Chapter 5 likened them initially to laws of nature, external to us and forming a causal order. Under the heading of 'functional explanation', social structures were conceived as systems in static or dynamic, perhaps shifting, equilibrium, with individual actions explained as responses to the functional demands of the system. Durkheim's Rules and its example of crime were invoked in illustration. But, like Durkheim himself elsewhere, we were unsure how seriously to take the analogy between societies and organisms. So we fell back on a broader holism, which relied only on the autonomy of social facts. Individualism then entered a protest on behalf of neo-Classical economics by introducing Rational Choice theory and Game Theory. But, with a rational agent defined in Chapter 6 as a given bundle of ordered preferences, a stock of accurate information and an efficient computer, the protest invited a swift riposte. Holism retorted that preferences are dictated by the social structure or system, with calculations of expected utility merely a mechanism by which its demands are transmitted. Social norms can thus still be exhibited as the core of a social structure or, in so far as talk of 'systems' remains legitimate, as mechanisms in the dynamics of a social system.