INTERNATIONAL JOURNAL OF URBAN AND REGIONAL RESEARCH DOI:10.1111/1468-2427.12798 1 © 2019 Urban research PUblications limited The authors would like to acknowledge the Swedish research agency FORMAS for funding (grant number 2016- 01273) and the Research Council of the University of Antwerp (grant number 37056). — BEYOND CIRCULAR THINKING: Geographies of Transit-Oriented Development Mattias QviströM, Nik Luka aNd Greet de BLock Abstract Transit-oriented development (TOD) plays a significant role within contemporary planning policies for ‘smart growth’ and sustainable development, particularly in Europe and North America. As a well-rehearsed practice, this planning model is due for critical assessment and improvement in terms of its ability to incorporate dynamic and heterogeneous socio-spatial processes as matters of concern. Analyses of the conditions for ‘making TOD work’ in the scholarly and professional literatures tend to focus on the ‘node’ and ‘place’ qualities. While elaborations on node analysis (primarily based on accessibility measurements) abound within empirical research, discussions of place-specific assets are limited in scope and often spatially bounded by the circle defined by a 10-minute walk. This essay examines the use of this generic ‘circle’ model, and how it normatively frames how place is understood in TOD studies. We argue that the circle enhances a Euclidean understanding of the site, which favours a static and homogeneous spatial analysis of accessibility and density rather than (other) place qualities relating to dynamic socio- spatial processes. Finally, we argue that relational geography can facilitate an analysis of place qualities beyond the circle––one in which both the continuities and shifting settings of the wider context are meaningfully taken into consideration. Introduction Transit-oriented development (TOD) has come to great prominence in planning, urban design and transport debates since its declarative origins in the work of Calthorpe (1989; 1993) and Newman and Kenworthy (1995), later buttressed by the empirical work of Bertolini et al. (1996; 1998; 2012) and Cervero et al. (1998; 2004; 2007), among many others. 1 As a model, it calls for a (re)unification of transport and land-use planning, with public transport (PT) acting as the backbone for sustainable development, expressed in compact urban form at PT nodes (i.e. stations). TOD is explicitly conceived to restrain sprawl, accommodating sustainable commuting as well as attracting environmentally- and economically-balanced growth (Calthorpe, 1993; Schwanen et al., 2004; Krueger and Gibbs, 2008; Renne, 2009; Papa and Bertolini, 2015). As such, TOD is closely related to the ‘Smart Growth’ and ‘New Urbanist’ movements (Burchell et al., 2000; Gibbs et al., 2013), which propose projects with density and mixed-use as Leitmotifs and local public-private growth coalitions as boosters (Goetz, 2013; Charmes and Keil, 2015; McFarlane, 2016). TOD strategies ostensibly redirect car-dependent diffuse patterns of growth into dense, fine-grained mixes of activities and land uses near PT nodes. The TOD site as mobility hub is thereby imaginatively shifted from infrastructural thoroughfare to urban public space. TOD acts in principle as the linchpin for decidedly urban projects––accessible, well-connected, and spatially-integrated ‘pedestrian pockets’ encouraging active transport (biking and walking). Supported by models in transport geography, the general policy assumption is that improved PT increases the accessibility profle of a location, 1 We say ‘declarative’ to acknowledge that de facto TOD models were widespread in patterns of rail-based urbanization such as ‘streetcar suburbs’ from the 1870s through to the 1940s (Warner, 1978; Luka, 2006; Mees, 2010; De Block and Polasky, 2011; De Block, 2014; Papa and Bertolini, 2015; Qviström and Bengtsson, 2015).