Of what 'green infrastructure' do we dream?Infrastructure lies at the heart of many disputes within environmental policy and planning. Take for example recent proposals for massive federal investments in infrastructure in the United States. These investments promise climate resilience, secure jobs, and large improvements in environmental quality occurring alongside a new era of public lands policy. Despite these lofty goals, fierce debate rages over the relationship between infrastructure spending, social wellbeing, and economic performance. Within these debates, some have argued for a deeper inclusion of 'green infrastructure' (GI), as a key component of sustainability transitions. And yet, debate continues over what GI is and how it should be implemented.Existing scholarship makes clear that GI means different things to different individuals and institutions. Deeply contested, GI manifests via varying conceptual orientations: through a water/stormwater management approach, an urban ecology/ecosystem services approach, a landscape conservation/greenspace planning approach, and integrative approaches across these overlapping orientations (Matsler et al., 2021;Szulczewska et al., 2017). Even within each of these conceptual orientations, significant differences exist in the scope and interpretation of what constitutes GI (Wright, 2011). These are not simply epistemological issues related to disciplinary silos; rather they stem from incompatible ontologies over what type of work should be expected from ecosystems, and its close corollary, the nature of the relationship between humans and landscapes (Gabriel, 2014;Gandy, 2002;Wachsmuth, 2012). Like many issues facing sustainability policy, these ontological concerns do work beyond the abstract, revealing the radically different and often alienated experiences of humans with each other and their resident ecosystems (Grabowski et al., 2019).Ontological instability adds considerable confusion to the implementation and practice of GI. The GI concept was originally popularized as a way to connect various ecological habitat typessuch as street trees, forest patches, parks, restored streams, and floodplainsin a cohesive planning framework to provide multiple benefits (Benedict & McMahon, 2001). In practice however, GI is more often implemented to address specific and singular purposes: for example, GI is built as an engineering solution for stormwater (Fletcher et al., 2015) or flood (Thorne et al., 2018) management, or entangled with political solutions to 'undesirable' land uses as in the case of clearing slums (Millington, 2018) and redevelopment of vacant properties (Schilling & Logan, 2008). In each of these cases, significant conflict exists over what counts as 'green infrastructure' to meet the socially defined goals driving urban and infrastructure redevelopment.In addition to different conceptualizations of what biophysical elements are considered 'part of' GI, GI is also socially and politically mutable. This instability presents itself as a contradiction of meani...
