infrastructural urbanism and fracture-critical networks – mammoth // building nothing out of something

infrastructural urbanism and fracture-critical networks


[Amos Coal Power Plant, from Mitch Epstein’s fantastic series American Power]

Adrian Lahoud has a lengthy post on infrastructure and urbanism at Post-Traumatic Urbanism; the post is well worth reading. A handful of somewhat scattered comments on it follow.

I strongly agree with the emphasis on “complex urban interdependencies”, in addition to “physical artefacts” of infrastructure. I think Lahoud’s making roughly the same point (that infrastructures cannot be considered as purely physical objects, but must be understood within their social, political, and economic contexts, as well as that those contexts are growing increasingly complex and difficult to process) that Varnelis (along with the other authors) does in The Infrastructural City, which is also well worth reading.

I also appreciate the effort made to link infrastructure and politics:

“If politics means making decisions that divide, then nothing divides quite like the kilometres of concrete and steel that make up a freeway or rail line. By understanding infrastructure as the ‘structuring of access’ we foreground the way it unevenly redistributes opportunity (and cost) in accordance with power. As such it forms a crucible for political activity.”

1I admire the attempt to expand the object of study, but worry that if we define everything as infrastructure, then we’ve defined infrastructure as nothing, as to include everything it must have no exclusionary properties. Lahoud appears to have had the same worry:

“This raises the question as to what isn’t infrastructure. The answer to this would be to say that the property of something being infrastructural or not, does not properly belong to the object itself, it emerges through the relation said object has with other objects. If this relationship is a dependent one, in which one object relies on the other for its functioning, then we might say that the second object plays the role of infrastructure. However if the relation between the objects is characterised by autonomy – that is to say independence – then we could not say that the object operates infrastructurally.”

My general thinking is that it’s not particularly helpful to redefine words in a way that makes the conversation inaccessible to someone who isn’t aware of your redefinitions — which is why I’d shy away from using infrastructure to generally mean something other than the common understanding (physical artifacts) — but I’m largely on-board with the notion that, particularly for design purposes, the most interesting and important properties of infrastructures are relational, derived from the active properties of the infrastructures, and therefore best described by the adjective and adverb (infrastructural, infrastructurally), not the noun (infrastructure). Funnily enough, I made this same distinct (between infrastructure and infrastructural) in my thesis.


[‘The Diversity Machine and Resilient Network’2, from Lahoud’s studio]

However, I’m not sure that I find the student projects fully convincing instantiations of Lahoud’s rhetoric. They still seem to be “big plans”, which, even on a generous day, I’m ambivalent about the value of. I’d like to think that learning to work infrastructurally (to use Lahoud’s language1) means working more flexibly (despite rhetoric which differs sharply from modernist rhetoric, the two designs presented appear to be close kin of modernist residential housing collectives and contemporary superblocks, as both take a large piece of land and develop urban fabric wholesale upon it), less directly (designing the infrastructure upon which the city grows, with an awareness of how the shaping of the infrastructure will affect the growth of the city, but not presuming to design the city itself) and accepting a degree of loss of control over the aesthetics of the resulting city fabric (which presents a host of drawing problems — how do you draw something which you are not presuming to design and still manage to communicate the importance of the work you’ve done in designing the scaffolding? — but still seems to me to be a humility worth developing).

2 Credited to the student team of Martin Abott, Georgia Herbert, Clare Johnston, Joshua Lynch, and Alexandra Wright

Having expressed that reservation, I do greatly appreciate the thrust of one of those student projects, ‘The Diversity Machine and Resilient Network’2, which argues that, though Beirut’s “urban fabric… lacks consolidation… optimization or efficiency”, this is not a weakness, but a strength: “it is precisely the ‘redundancy’ of the distributed social infrastructure and relative autonomy of the neighbourhoods that lends the city its resilience.” Though made more specifically in reference to urban form and less in reference to infrastructure, this point reminds me of two things.


[PXP Lease on Jefferson, from CLUI’s slideshow “Urban Crude”, at Places]

First, as faslanyc noted in the comments on a previous post, the impact of an infrastructure on the territory in which it resides should be evaluated not just by its scale, but also for its degree of distribution and connectivity. I don’t think there’s a simple distinction to be drawn between good (distributed) and bad (single entity, to use faslanyc‘s term), but Lahoud and his students do bring up one of the major advantages (resilience) that a distributed infrastructure has over a monolithic infrastructure. A second major advantage motivates Uchitelle (insofar as he is motivated by something more than infrastructural nostalgia) in the “Superproject Void” article I noted previously: distributed infrastructures connect diverse territories and, in doing so, produce economic benefits which cannot be replicated by monolithic and singular infrastructures, no matter how large.

Second, and directly related to that first advantage, I’m also reminded of the article Fracture Critical, which ran recently in Places and draws an interesting parallel between two ways of designing specific infrastructures, fracture-critical and fracture-resistant, and ways of designing larger systems. According to the article, fracture-critical infrastructures and systems (the term orginates within engineering) are defined by four primary characteristics: “lack of redundancy, which makes a structure susceptible to collapse should any individual component fail”, “interconnectedness”, “efficiency”, and “sensitivity to stress”. Though I’ve said that redundancy is one of the key characteristics of a distributed infrastructure, this definition of fracture-critical suggests that there may be an inherent tension between “redundancy” and “interconnectedness”. Given that the consequences of a networked super-project being fractured would be enormous, I suspect that there’s a place for being cautious about the design of such projects, even while recognizing their value.

2 Responses to “infrastructural urbanism and fracture-critical networks”

  1. [...] wanted to thank mammoth for some thought provoking feedback regarding the relationship between the studios ambitions and [...]

  2. [...] Lahoud has a thoughtful response to mammoth’s earlier post “infrastructural urbanism and fracture-critical networks” (itself a response to another post by Lahoud on a recent studio he led), discussing how to properly [...]