I happened to listen to an interview with anthropologist Brian Larkin on the Cultures of Energy Podcast who has been studying infrastructure for some time now. The podcast focuses on Larkin’s interest in studying media and energy infrastructures, particularly as it relates to his field work in Nigeria–but it is also pretty free ranging. Larkin’s The Politics and Poetics of Infrastructure is mentioned a few times by the hosts as a significant piece of work, so I thought I would follow up on it, since, despite its huge impact factor in Google Scholar, I had failed to register Larkin’s work before. Below are some rough notes on reading through it.

One point by Larkin that I think is quite important that I haven’t seen talked about too much before is that it’s important not to take infrastructure as some fixed unitary thing. The very act of naming an infrastructure, and drawing boundaries around it is itself an interpretive act.

Given the ever-proliferating networks that can be mobilized to understand infrastructures, we are reminded that discussing an infrastructure is a categorical act. It is a moment of tearing into those heterogeneous networks to define which aspect of which network is to be discussed and which parts will be ignored. It recognizes that infrastructures operate on differing levels simultaneously, generating multiple forms of address, and that any particular set of intellectual questions
will have to select which of these levels to examine. Infrastructures are not, in any positivist sense, simply “out there.” The act of defining an infrastructure is a categorizing moment. Taken thoughtfully, it comprises a cultural analytic that highlights the epistemological and political commitments involved in selecting what one sees as infrastructural (and thus causal) and what one leaves out.

Larkin (2013), p. 330

Larkin also suggests some useful ways of thinking about the poetics of infrastructure, which emphasize the interpretive and aesthetic qualities of infrastructure:

[Infrastructures] emerge out of and store within them forms of desire and fantasy and can take on fetish-like aspects that sometimes can be wholly autonomous from their technical function. Focusing on the issue of form, or the poetics of infrastructure, allows us to understand how the political can be constituted through different means. It points to the sense of desire and possibility, what Benjamin (1999) would term the collective fantasy of society.

This idea of infrastructure as collective fantasy reminded me directly of notions of archiving it all, where simply by virtue of the web’s architectural support for seemingly infinite and instantaneous copies, the idea of the complete archive somehow seems more within reach. I’m specifically thinking of the efforts of the state to build massive data centers that attempt to archive everything uttered on the Internet. This collective fantasy is in fact more like a nightmare, driven by a fear of terrorism … a drive to collect everything so that future events can be forecasted like the weather, or so that past events can be prefectly reconstructed. But it is the infrastructure of the Internet itself that makes these surveillance poetics possible – or is it the other way around? Science and Technology Studies teaches us that these fantasies of surveillance and the Internet/Web are in fact [co-producing] each other.

I’ve used Internet and web interchangeably here which is kind of sloppy. Being precise about what the Internet is, and what the web is, and what work these concepts do is exactly what Larkin is driving at. The specific technologies that make up the internet and the web are different, and shifting all the time … but they are part of a system that extends outwards from and envelops the technological artifacts themselves. The decentering of the technical artifacts and looking at this system is what studying infrastructure is about.

Larkin positions the work of Thomas Hughes, who studied electrification (Thomas Parke Hughes, 1993) in the United States and large technical systems (Thomas P. Hughes et al., 1987), as important for Bijker, Bowker, Star, Edwards, Jackson and Yates who are already familiar to me for their work on infrastructure. I do remember running across Hughes in Russell (2014), so it’s probably worth my time to read up on him. In studying electrification Hughes studies the genaology of technology, and how particular technological innovations develop in specific circumstances, but in becoming infrastructures they are translated into new environments and situations. How this translation work happens is what Actor Network Theory (Latour & Porter, 1996 ; Latour, 2005) provides as both a method and theory for tracing these translations between humand and non-human actors.

It’s interesting that Larkin contests the idea that infrastructures are invisible until they breakdown, which he attributes to Star & Ruhleder (1996).

Thus many studies that begin by stating how infrastructures are invisible until they break down are fundamentally inaccurate. Infrastructures are metapragmatic objects, signs of them- selves deployed in particular circulatory regimes to establish sets of effects. It is commonplace, seemingly obligatory, for almost any study of infrastructure to repeat Star’s (1999) assertion that infrastructures are “by definition invisible,” taken for granted, and that they only “become visible on breakdown” (p. 380; see also Collier 2011, Elyachar 2010, Graham & Marvin 2001, Larkin 2008). But this assertion is a partial truth and, as a way of describing infrastructure as a whole, flatly untenable. Invisibility is certainly one aspect of infrastructure, but it is only one and at the extreme edge of a range of visibilities that move from unseen to grand spectacles and everything in between. (p. 336)

This is a pretty strong statement since the use of breakdowns as an analytic technique has a much longer lineage going back to Dewey and Heidegger (Koschmann, Kuutti, & Hickman, 1998). Perhaps Star’s focus on information infrastructures in particular is what lends a larger degree of invisibility: it really is difficult to think of the Internet as material infrastructure, even though it is. Perhaps the source of the misunderstanding here is that Star (with Dewey and Heidegger) claims that examining breakdown is a useful way of understanding infrastructure, not, as Larkin seems to be suggesting, that all infrastructure is invisible. I might need to reread Star a bit more to see if she really did make a strong claim like that – it seems uncharacteristic of her work.

I think part of what Larkin is doing here is side stepping a sociological understanding of infrastructure so that he can introduce the idea of an aesthetics of infrastructure that fits the anthropological mindset a bit better. The ways that infrastructure are part of our sensory experience as humans, and what these experiences signify is a very anthropological question. He goes on to cite work that has been done that examines the way materials like iron, and incandescent lighting have changed our experiences of living in very visible ways. In some ways this reminds me of recent work by Seaver (2017) on the cultural meaning of algorithms, and the use of ethnographic techniques to study them.

But one thing that’s interesting here is the way that the experience of infrastructure is situated in time relative to the infrastructure. Someone who lives through an infrastructural shift like the movement from flame to incandescent lighting will see the two infrastructures in relief in a way that it is hard to imagine someone seeing if they grew up within an already electrified society. It would take historical and imaginative insight to light a candle, and reconstruct what life was like before electrification. However to Larkin’s point this movement from flame to incandescent lighting is not really a breakdown. But it is a moment in which infrastructures can be seen more clearly because they are in direct relation to each other. This outsider view seems to be precisely what anthropology is about, which is another point that Seaver makes.

Overall this idea that technology and society coproduce each other, while not named specifically by Larkin, is what I’m left reading his article thinking about. Especially since he drops a reference to (Deleuze:1995?) to make the point that “types of machines can be matched to types of societies”.


Benjamin, W. (1999). The arcades project. (H. Eiland & K. McLaughlin, Trans.). Cambridge: Harvard University Press.
Hughes, Thomas P. et al. (1987). The social construction of technological systems: New directions in the sociology and history of technology. In (pp. 51–82). Cambridge, MA: MIT Press.
Hughes, Thomas Parke. (1993). Networks of power: electrification in Western society, 1880-1930. Johns Hopkins University Press.
Koschmann, T., Kuutti, K., & Hickman, L. (1998). The concept of breakdown in Heidegger, Leont’ev, and Dewey and its implications for education. Mind, Culture, and Activity, 5(1), 25–41.
Larkin, B. (2013). The politics and poetics of infrastructure. Annual Review of Anthropology, 327–343.
Latour, B. (2005). Reassembling the social: An introduction to actor-network-theory. Oxford University Press.
Latour, B., & Porter, C. (1996). Aramis, or, The love of technology (Vol. 1996). Harvard University Press Cambridge, MA.
Russell, A. L. (2014). Open standards and the digital age. Cambridge University Press.
Seaver, N. (2017). Algorithms as culture: Some tactics for the ethnography of algorithmic systems. Big Data & Society, 4(2). https://doi.org/10.1177/2053951717738104
Star, S. L., & Ruhleder, K. (1996). Steps toward an ecology of infrastructure: Design and access for large information spaces. Information Systems Research, 7(1), 111–134.