As heterotopias ‘desiccate speech’ and ‘stop words in their tracks’, the attempt to clarify their spaces is probably futile. Yet, as Soja’s chapter shows, Foucault’s concept keeps being called upon, used and discussed. In fact, it seems that interest in heterotopias is still on the rise. In this chapter I’d like to propose that this renewed interest is due both to social changes that have occurred since Foucault’s article and to changes in the character of heterotopias themselves. In particular, three areas of discussion help to identify the reasons for both changes. First of all, there are the studies that propose a shift from the mass society of Fordism to the flexible socioeconomic organization of post-Fordism. This change has helped to introduce ‘difference’ as a new term in the equation normality/deviance on which

4 5 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2

characterization of space, I suggest that heterotopias are ‘spaces of representation’ and vanish when the social relations that produced them end. Finally, social differentiation has also enlarged the debate on the public sphere. The 1960s’ view of a universal and a-spatial public sphere is increasingly challenged by the notion of a multiplicity of public spheres that are public only for the social groups that produced them. Heterotopias are part of this group-specific publicness and make evident that public spheres always imply public spaces.