In 1697, the author Daniel Defoe, famous for his novel Robinson Crusoe, wrote a book titled Essays Upon Projects.The word 'projects', during Defoe's times, meant unscrupulous means of making money, usually by building structures, such as housing or public utilities. The 'age of projects' as Defoe (1702) called it, inaugurated an era preoccupied with the built form. In the 19th century, this built form consolidated as 'improvement', or governments seeking to uplift subjects by building public works projects. In the 20th century, governments, private finance and experts all came together to recreate ideas of improvement into that of development or a common condition of progress for human kind. A key facet allowing a variety of actors to 'improve' and 'develop' societies has been infrastructure. That is, built forms which facilitate and enable (and disable) social life, economic activity, commerce, movement and aid colonial and postcolonial state formation.We employ the term infrastructure in an analytical fashion, arguing that a variety of new histories, in this case of colonial South Asia, can be anchored around the concept of infrastructure. Specifically, we argue that using infrastructure as a heuristic category allows for a re-evaluation of the history of the political economy of modern South Asia, which has been studied through the prisms of space, capital and the social. Anthropological research has defined infrastructures as physical and material objects, such as roads or water pipes, through which people, capital, finance, non-humans and technologies move through. Brian Larkin shows how anthropologists have explored the political rationalities that underlie technological projects, the systems that infrastructures are implicated within and the affect and sentiment they inspire. Other interventions have conceptualized infrastructure as a social relation (Elyachar, 2010) and as a form of 'calculative reason' (Mitchell, 2001) which informs a vast field including governance, engineering, transport and communication (Carse, 2017). To quote Larkin (2013, 339) again there are a number of ways to 'analyse and conceive of infrastructures'. However as Nikhil Anand, Hannah Appel and Akhil Gupta (Anand et al., 2018, pp. 8-9) argue, most ethnographies study the neoliberal era, but there is little discussion of the 'public'