Four ways to fail at defining “infrastructure”.
It’s encouraging to discover that I’m not alone in having difficulty establishing a working definition of infrastructure. I’ve found few accounts of infrastructure that can specify unambiguously what it is they’re even looked at. There’s no lack of definitions though (often several different and conflicting definitions in the same account): I’ve encountered four basic types:
Exemplary lists that indicate the kind of thing the writer means by infrastructure: “airports, harbours, roads, sewers, bridges, dikes, dams, power corridors, terminals, treatment plants” (Bélanger, 2010); “wires, ducts, tunnels, conduits, streets, highways, and technical networks that interlace and infuse cities” (Graham, 2000). This kind of definition makes it the reader’s business to sort out what the listed things have in common, so it’s kind of a non-definition definition. It does however, succeed in conveying something open-ended and heterogenous, like a Latour litany. As Levi Bryant writes, litanies undermine implicit essentialism:
“Through the creation of a litany of heterogeneous objects, the object theorist is forced to think that heterogeneity as such rather than implicitly (and often unconsciously) drawing on one prototypical object that functions as the representative of the nature of all objects.”
That these lists are so common in describing infrastructure suggest that these properties of open-endedness and heterogeneity aren’t just quirks of description, but something inherent to be grasped.
Documentary definitions from historical or institutional sources. Dictionary definitions are the classic version of this. Bélanger and Edwards both cite The American Heritage Dictionary, for example. A step further than this is to identify documents from policy or practice. Edwards offers a definition from the 1997 President’s Commission on Critical Infrastructure Protection:
“By infrastructure . . . we mean a network of independent, mostly privately-owned, man-made systems and processes that function collaboratively and synergistically to produce and distribute a continuous flow of essential goods and services”
These are useful for gaining an historical perspective on what infrastructure was understood to be at a particular time, although obviously there’s no guarantee that the dictionary editor’s definition is either particularly well-founded, nor an accurate description of what infrastructure was even then. Edwards’s documentary definition is a more interesting case, because as a policy document, it’s performative not just descriptive. That is, it’s a definition that makes a difference. Infrastructure was made or managed in a particular way because it was defined like this.
Functional definitions that specify what infrastructure is by what it does. For example, that American Heritage Dictionary definition says infrastructure is the “collective network of roads, bridges, rail lines and similar public works that are required for an industrial economy to function”. The defining feature is that this set of things allow “an industrial economy to function”. Common themes in this kind of definition are service and dependency. But what do ‘function’ and ‘service’ mean? To answer those questions means diving into more general writing about technology. Thomas P. Hughes offers in ‘Evolution of Large Technical Systems’ the idea of a “common system goal”: that what characterises these systems is that they’re teleological, oriented by some governing idea of purpose. There must be complications in this, though: Frischmann describes infrastructure as “shared means to many ends”. Are the ends of infrastructure singular or multiple?
Structural / formal definitions that say infrastructure is something that takes a particular shape or topology. Most commonly, infrastructures are described as networks, but occasionally they’ve been seen as landscapes (Graham mentions the idea; Bélanger explores it much more thoroughly: 2006; 2009; 2010), and Varnelis points to new infrastructures taking the form of an ecology (2009), or a field (2011). I’ve defined it this way myself, but now I’m not satisfied that a purely structural or formal definition is adequate; at least partly because of the range of forms that infrastructures take, but also because it overlooks how they work as if an infrastructure was little more than a particular type of pattern.
In practice these definition types are blended, and all the people I’ve referred to here employ several. None of them are really adequate on their own, because they each fail to capture something significant. A viable working definition (for whatever “working” means in my case, which is pretty slippery in itself!) is probably going to be a layered or composite thing.
Bélanger, P. (2006). Synthetic surfaces. In Waldheim, C., ed., The Landscape Urbanism Reader, pp. 239–265. New York, NY: Princeton Architectural Press.
Bélanger, P. (2009). Landscape as infrastructure. Landscape Journal, 28(1):79–95.
Bélanger, P. (2010). Redefining infrastructure. In Mostafavi, M., ed., Ecological Urbanism, pp. 332–349. Baden, Switzerland: Lars Müller.
Douglas, C. (2011). Off the grid. infrastructure and transformational space. Access, 30:45– 57.
Edwards, P. N. (2003). Infrastructure and modernity. force, time and social organization in the history of sociotechnical systems. In Misa, T. J., Brey, P., and Feenberg, A., eds., Modernity and Technology, pp. 185–226. Cambridge, MA: MIT Press.
Frischmann, B. M. (2012). Infrastructure. The social value of shared resources. Oxford, UK: Oxford University Press.
Graham, S. (2000). Introduction. cities and infrastructure networks. International Journal of Urban and Regional Research, 24(1):114–119.
Varnelis, K. (2009). The Infrastructural City. Networked Ecologies in Los Angeles. Barcelona: ACTAR.
Varnelis, K. (2011). Infrastructural fields. Quaderns, (261).