The spatiality of theorizing the urban experience, of thinking cities in a world of cities, maps well onto the core elements of a comparative imagination − thinking across different ‘cases’ to produce conceptualizations which contribute to wider understandings of the processes being analysed, and which might then, in turn, be considered in relation to other contexts or cases. However, the comparative methods which have conventionally underpinned urban studies have not been well placed to serve the project for a more global urbanism (Robinson 2011). Theorizing cities can benefit from a comparative imagination, but comparative methods need to be refitted to support a more global urban analytical project, including a substantial reconfiguring of the ontological foundations of comparison. For example, what might be considered a ‘case’ needs to be redefined to avoid the restricting and territorializing trap of only comparing (relatively similar) ‘cities’: we might rather compare, for example, specific elements or processes in cities, or the circulations and connections which shape cities, thus rendering urban experiences comparable across a much wider range of contexts, and building research strategies which are adequate to the complex spatiality of urban forms (Robinson 2011; Ward 2010). Comparators − the ‘third term’ which establishes the grounds of comparability of cases ( Jacobs 2012) − need to be selected so they are relevant to a diversity of urban contexts, rather than seeking relatively

similar cities for comparison as is conventional. In this way, they will be able to stretch theoretical concepts to the breaking point required for the reinvention of urban studies for global analysis, and not simply reinforce parochial and limited understandings (cf. Pierre 2005). More generally, the status of the case itself needs to be reimagined, in relation to both the wider empirical processes shaping particular outcomes, and the conceptualizations which are an important ambition of comparative strategies. This is essential to ground an adequate post-structuralist comparative method which moves beyond both quasi-scientific explanations and a view of the world in which wider structures are drawn on to explain complex specific outcomes.