There is an intensifying geopolitical contest to connect the world by envisioning, financing, and constructing intercity infrastructure. In contrast to the audible rattling of American sabers, the European Union has launched a more subtle infrastructure development initiative. In both instances the objective is to compete with China, particularly Chinese state capital (Lee 2017), which is poised to establish a global monopoly over the megainfrastructure market. The result has been a planetary explosion of megainfrastructure projects, which has been described as an infrastructure “turn” (Dodson 2017), “push” (Ougaard 2017), and “scramble” (Kanai and Schindler 2019). Anthropologists have been instrumental in grounding the proliferation of infrastructure megaprojects in everyday life, and illuminating the dense webs of sociality and materiality that animate human settlements. Indeed, this scholarship uses material infrastructure such as pipes (Björkman 2015), electricity meters (von Schnitzler 2013), and roads (Harvey and Knox 2015) as “theory-making tools.” However, there is a scalar mismatch between the global infrastructure scramble and scholarship that interrogates urban infrastructure within cities. I argue that for infrastructure to work as a theory-making tool, analysis must account for the dynamic global trends shaping infrastructure development.

The infrastructure turn seems more like an infrastructure re-turn when contextualized historically. The United States and the Soviet Union competed for the loyalty of client states during the Cold War by providing assistance for large-scale infrastructure projects. Among regional planners, there was a consensus around the planning strategies that would serve to integrate territory into seamless national economies. This consensus collapsed with the neoliberal turn in the early 1980s, and a condition of many structural adjustment loans was the privatization of infrastructure. In contrast to the postwar era, planning in the neoliberal period was spatially blind and geared toward “getting the prices right.” The establishment of market-oriented territory was limited to clearly demarcated zones with lax environmental and labor standards, which had scant impact in most countries (Frick, Rodríguez-Pose, and Wong 2019). The results of decades of neoliberal governance disappointed even its most ardent supporters, and while proponents of neoliberalism have maintained their objectives—that is, attract foreign investment, upgrade manufacturing capability, and “move up” global value chains—they have begrudgingly come to accept an expanded role of the state. Spatial planning is now used to create market-oriented territory that can be plugged into global value chains. Tellingly, the World Bank has come to embrace spatial planning as an antidote to both market and state failure:

In recent years, a number of countries have experimented with various strategies to correct market and governance failures within and across industries. One approach is to work with spatial strategies such as growth poles, growth corridors, and special economic zones.

The production of market-oriented territory necessitates the construction of intercity infrastructure. Thus, rather than citywide infrastructure systems, the emergent regime of infrastructure-led development prioritizes intercity connectivity.

Designed transnational territories extend from a resource frontier to ports and export zones via industrial hubs. They are underpinned by logistics infrastructure that integrates settlements into a network geared toward the extraction, production and export of commodities. Examples include the Lamu Port-South Sudan-Ethiopia Transport Corridor (LAPSSET), Greater Mekong Subregion, the China-Pakistan Economic Corridor, and South America’s COSIPLAN. This logic also animates India’s corridor development strategy and China’s multifaceted Belt and Road Initiative. Such initiatives territorialize what Vinay Gidwani (2008, 74) refers to as a development machine: within a demarcated space, they rework “heterogeneous associations of humans and nonhumans, which, in combination, produce life functionings that would be impossible in the absence of connection.” Infrastructure serves to constitute these territorialized development machines, while it is simultaneously the conduit through which heterogeneous associations are staged.

If we situate our understanding of infrastructure at the scale of the transnational territory rather than the city, we are confronted by a set of issues surrounding ownership. The privatization of existing citywide infrastructure systems in the 1980s and 90s was straightforward in comparison to the construction of transnational intercity infrastructure. While China has pioneered a model of finance whereby bilateral loans are repaid with the profits generated by resource extraction, project finance is preferred by private capital in the Organization for Economic Cooperation and Development. In this model, syndicates of banks loan capital to a project sponsor (see Hainz and Kleimeier 2012), whose raison d’etre is to manage the infrastructure project at a safe distance from bureaucracies that may be staffed by opportunistic officials (Dorobantu and Müllner 2017). Their lack of collateral—not to mention the complexity of transnational projects—makes lending large sums a risky endeavor, and this is why the United States recently created an International Development Finance Corporation that can issue bilateral loans.

The transnational nature of contemporary development machines poses challenges to national governments, because sovereignty is widely distributed among a range of actors at multiple scales. National governments typically establish an authority to oversee the project, but in practice it remains unclear whose rules and norms prevail when conflict erupts. Should a local community that seeks to affirm land-use rights or prevent ancestral land from being appropriated confront local or national government officials, the project managers, multinational corporations, or the national development organization overseeing the project? Who has the authority to adjudicate conflicts over ownership of land and resources in a transnational development machine? While the establishment of transnational territories disrupts accepted notions of sovereignty, it is unlikely to foster identities that supersede other allegiances. People in Laos and Vietnam may prioritize national, local, religious, tribal or other identities, but they will not think of themselves first and foremost as from the Greater Mekong Subregion. Indeed, a myriad of protests have plagued the LAPSSET project as groups across the region have mobilized to lay claim to the spoils and disrupt the project.

Transnational territories undeniably rearrange associations in multiple ways, spatially distributing sovereign authority in the process. Meanwhile, populations who do not share the dreams of planners mobilize to contest dispossession and lay claim to infrastructure or the social surplus it generates. Infrastructure can indeed act as a theory-making tool, but it must be viewed through a bifocal lens that brings into view the multiscalar nature of the politics of infrastructure-led development.