The concept of ‘tacit knowledge’ as the means by which individuals interpret the ‘rules’ of social interaction occupies a central role in all the major contemporary theories of action and social structure. The major reference point for social theorists is Wittgenstein's celebrated discussion of rule‐following in the Philosophical Investigations. Focusing on Giddens' incorporation of tacit knowledge and rules into his ‘theory of structuration’, I argue that Wittgenstein's later work is steadfastly set against the ‘latent cognitivism’ inherent in the idea of tacit knowledge and tacit rules. I also argue that the idea of tacit knowledge and tacit rules is either incoherent or explanatorily vacuous.Scholars of the emotions maintain that all anger requires an object of blame. In order to be angry, many writers argue, one must believe than an actor has done serious damage to something that one values. Yet an individual may be angered without blaming another. This kind of emotion, called situational anger, does not entail a corresponding object of blame. Situational anger can be a useful force in public life, enabling citizens to draw attention to the seriousness of social or political problems, without necessarily vilifying political officials.In the first half of this paper I show how H. L. A. Hart's theory of rules can resolve, or at least clarify, a central methodological problem in legal anthropology that was first posed in Llewellyn and Egebel's The Cheyenñe Way In the second half I explore and develop Hart's theory (a) of rules, and apply it to problems of agency and behaviourism in legal anthropology, and (b) of legal development, and apply it to the problem of rule‐scepticism in legal anthropology as it is posed in Roberts and Comaroffs Rules and Processes and elsewhere.As human beings, we share many historically developed, language‐game interwoven, public forms of life. Due to the joint, dialogically responsive nature of all social life within such forms, we cannot as individuals just act as we please; our forms of life exert a normative influence on what we can say and do. They act as a backdrop against which all our claims to knowledge are judged as acceptable or not. As a result, it is not easy to articulate their inadequacies in a clear and forceful manner. However, within most of our forms of life, we have a first‐person right to express how our individual circumstances seem to us. And by the use of special forms of poetic, gestural talk—talk that can originate new language‐games—we can offer to make our own ‘inner lives’ public. In this paper, I want to claim that this is just what Wittgenstein is attempting to do in his later philosophy: by use of the self‐same methods that anyone might use to express aspects of their own world picture, he is offering us his attempts to make the background ‘landscape’ of our lives more visible to us. These methods are explored below.Proponents of the view that social structures are ontologically distinct from the people in whose actions they are immanent have assumed that structures can stand in causal relations to individual practices. Were causality to be no more than Humean concomitance correlations between structure and practices would be unproblematic. But two prominent advocates of the ontological account of structures, Bhaskar and Giddens, have also espoused a powers theory of causality. According to that theory causation is brought about by the activity of particulars, in the social psychological case, individuals of some sort. Consistence would demand that structure be those individuals. But neither Giddens nor Bhaskar wish to reify structure to the extent that would fit it for a role as a powerful particular. If only human beings can be powerful particulars in these contexts, the only way that structures can be real must be as properties of conversational (symbolic) interactions. Human action is social just in so far as people direct themselves to engage well in joint activities with others.