How Cities Learn. Astrid Wood

offered innovation to each city. However, in exchanging local solutions, Toronto and Mexico City strengthened their relationship, thereby demonstrating that policy exchange is instrumental in forming relationships between localities. In McCann’s (2011b) study of Bing Thom, a Vancouver architect-consultant hired by Fort Worth, Texas to “Vancouverize” the city, he underscores the ability of a policy mobilizer to influence development across localities despite considerable differences between the places. These policy mobilities studies provide a foundation for my research of the manner in which cultural, economic and political relations between and within cities were used by South African policymakers to advance (or subvert) particular actions and decisions regarding BRT.

      It is also important to note the imbalance of power between cities of the global north and south (Massey 2011; Robinson 2011), as well as across cities of the global south (Bunnell and Das 2010 for urban policy transfer from Kuala Lumpur to Hyderabad, and Hains 2011 for transfer between Dubai and Delhi). Massey (2011), for instance, outlines a trade agreement between London and Caracas whereby London provided technical advice on a range of urban issues from transportation planning to waste disposal in exchange for a reduced price on oil, which London then used to fuel the city’s buses, funding a 50 percent reduction in fares for the poorest people of the city. The relationship between the cities became the subject of contestation in 2008 when the newly elected Mayor Boris Johnson cancelled the agreement. This account provides evidence of the “politics of place” as the “outcome of the contested negotiation of physical proximity”, which is also “explicitly relational” beyond the confines of the city (2011: 4). Such a contested terrain is also found by Robinson in her analysis of city development strategies (2011). She looks at the circulation of city strategies as a technique for governance, introduced in Johannesburg and then disseminated by the World Bank and Cities Alliance. “What might appear to be an instance of the local application of global policy discourse in the Johannesburg case”, Robinson writes, “was a strongly locally determined policy process, shaped by quite specific political dynamics” (2011: 34). Such claims offer a counterhegemonic view that policy practice can always be traced to foreign or western origins, a point critical to my study, which recognizes the multiplicity of sites of origin that are included or disregarded in the circulation of best practice.

How Cities Learn considers how policy exchange and adoption connects places. Mobilities theorists have provided a variety of metaphors for interpreting the way in which a policy moves through a system – the train moves along the tracks (Latour 1993, 2005), a car at a petrol station (Normack 2006) or water in a creek (Tsing 2000), all of which presuppose the relationship between the object and the conditions of its mobility. Tsing (2000: 337) reasons that “a focus on circulations shows us the movement of people, things, ideas, or institutions, but what it does not show is how this movement depends on defining tracks and grounds or scales and units of agency… If we imagined creeks, perhaps the model would be different; we might notice the channel as well as the water moving”. These entities – the train, car and water – circulate through the system while maintaining their connection to the spatial elements – the track, petrol station and creek bed. These movements underscore the “networked nature of interconnectedness” (Lester 2006: 135). Even though the relationship between the movable objects and place is provisional, ephemeral and transient – the train moves to a different part of the track, the car to another petrol station and the water through the creek – it remains linked through the system. This focus on intergovernmental mobility considers the way in which the (metaphorical) tracks, petrol station and creek bed promote circulations by creating a conducive context for their continuity. Neither the terrain nor its connections are “spatially fixed geographical containers for social process” (Hannam et al. 2006: 5), but these metaphors can help us understand the inter-urban and intra-urban relationships that facilitate the exchange of BRT across localities.

      Chapter 5 examines the relationships between importing and exporting localities. It investigates the influence of municipal politics that enable and/or constrain decisions to adopt – that is the way in which those cities learning concurrently are connected and disconnected by these multifaceted processes of knowledge accumulation. Such debates deepen and widen the space through which policy flows, by proposing that local municipal relationships, both competitive and cooperative, shape the circulation process. Moreover, it exposes policy mobilities as more than a course through which energetic policy mobilizers introduce proven solutions to unsuspecting policymakers. Rather, policy ideas move through the terrain of local politics, which house prevailing international and domestic cooperative and competitive relations.

      Tracing through Temporalities

      How Cities Learn considers the role of temporality and historicity in policy mobilities. It illustrates that learning is a progressive conversation in which slowly, incrementally policymakers warm to an otherwise foreign notion and through these continuous exchanges, ideas, practices and programs are relocated and localized. This research will take a broader perspective to show that policy flows through “waves of innovation” (McCann and Ward 2010: 175) which only seem to be arriving more frequently. There are peaks – periods of rapid diffusion facilitated by either need or opportunity, or both – and valleys – periods when circulation is minimal. In South Africa, the learning about BRT was longwinded and drawn out, incremental and at times delayed. How Cities Learn illustrates that regardless of the speed of circulation, policy implementation remains cumbersome because policy is always political, meaning that it takes time to localize policy.

Arguments supporting a crisis-driven approach towards policymaking through the hurried acquisition of international policy models, are frequent in the policy mobilities literature (Brenner et al. 2010; Brenner and Theodore 2002; Clarke 2009; Peck 2002, 2003, 2011a; Peck and Theodore 2001; Theodore and Peck 1999, 2001, 2011; Ward 2011). The term “fast policy” (Jessop and Peck 1998; Peck and Theodore 2015) has been used to characterize the rapid introduction of off-the-shelf prefabricated best practice policies. Much of the fast policy discourse rests on the prevalence of knowledge sharing between the UK and the US in the 1980s and 1990s. This scholarship proposes that explosive tactics create momentum to drive through systematic transformation, suggesting that a more incremental approach could not sustain implementation procedures (Peck and Theodore 2010a, 2010b). In so doing, it takes for granted these processes as hectic and hurried arguing that ideas and innovations are appropriated because of their prevailing success elsewhere, which make them easily executed within local policymaking cycles.

      Departing from the prevailing logic in the fast policy literature, in Chapter 6 I demonstrate that the learning process is in reality often lengthy, incremental and at times delayed. The dissemination of Singapore’s electronic road pricing system for taxing vehicular movement through the city provides an illustration of the multiple temporalities of policy learning. Whereas in London, a version of the pricing system was appropriated successfully, in New York, the Mayor’s proposal was defeated by the state legislature (Chua 2011). These arguments illustrate a difference between the speed at which a policy moves and the velocity at which it is implemented. Unlike Kingdon’s “policy window” (1995) which assumes a random confluence of people, choices, problems and solutions that come together at a particular juncture to enable learning (and if that window closes, the opportunity is missed), my findings suggest that ideas are acted upon through multiple temporalities. Policy adoption is an inherently political process and thus the speed at which a policy is adopted is distinct from its measures of long-term efficacy. The achievements of a new transportation system, be it related to its financial viability or impact on the city, may take two decades to become apparent and thus divorced from their assumed likelihood of success at the time of adoption. It seems then that there are a number of different speeds and temporalities through which policy flows.

      Chapter 6 considers the multiple temporalities through which circulated policies emerge and remerge before adoption, reasoning that without these multiple attempts, policy adoption is unlikely to occur. The arguments that follow bridge the lacuna between historical and policy mobilities studies, illustrating the gradual, protracted and idiomatic manner through which transnational best practice proceeds, in contrast to the spontaneous and hasty process documented in scholarly literature on