Tuesday, May 25, 2010

Reading :: Prince of Networks

Prince of Networks: Bruno Latour and Metaphysics
By Graham Harman

I came upon Prince of Networks through a reference from Tom Haskins' blog; he credited it as liberating him from some double binds in understanding actor-network theory (ANT). Many thanks to Tom for pointing out this book, which is quite good.

Prince of Networks is a philosopher's understanding of Latour's body of work. Specifically, Harman is trying to introduce Latour as "a key figure in metaphysics" (p.5). And he thinks it's about time:
The central problem of metaphysics is the interplay of objects and relations, and Latour sheds more light on both problems than perhaps any other contemporary thinker. It is odd that no one so far has seemed to realize this - bad luck for Latour, but once-in-a-lifetime good fortune for me, as the author of the first book on his metaphysics. (p.159)
Harman divides this once-in-a-lifetime book into two parts. Part I examines Latour's key works - particularly Irreductions, Science in Action, We Have Never Been Modern, and Pandora's Hope, after which he names these chapters - in terms of metaphysics, comparing these key works to parallel work in metaphysics. In Part II, he examines Latour's take on objects and relations more critically, drawing Latour's thought into a more critical dialogue with metaphysics.

Unlike Harman and (evidently) Latour, I don't have a grounding in metaphysics, so the second part was of quite limited value to me. But the first part was useful because I was able to see how Harman explains Latour's thought to himself and fellow metaphysicists, and thus was able to see how ANT's concepts circulated into other disciplinary dialogues.

Harman starts with Irreductions, a relatively short treatise that is typically appended to Latour's The Pasteurization of France. Here, Latour describes a 1972 epiphany on the road from Dijon to Gray, in which he realizes that "nothing can be reduced to anything else, nothing can be deduced from anything else, everything must be allied to everything else" (p.13, quoting Latour p.163). This assertion became a "new principle of philosophy" for Latour (p.13), a principle that led to four premises: "the world is made up of actors or actants"; "no object is inherently reducible or irreducible to any other"; "the means of linking one thing to another is translation"; "actants gain in strength only through their alliances" (pp.14-15). Latour also "rejects the rift between an inner substance and its trivial exterior" (p.17). And the concept of alliance, Harman says, doesn't mean that Latour is Machiavellian: he doesn't reduce truth to human power games, he accepts nonhumans as actors and measures reality in terms of how connected an actor is (p.19). Importantly, "any attempt to see actants as the reducible puppets of deeper structures is doomed to fail" in this understanding (p.21).

Harman moves next to Science in Action, which introduces the concepts of black boxes and action at a distance (p.33). Latour, Harman asserts, is not a philosopher of science so much as "a metaphysician working in a philosopher-of-science idiom" (p.36). In this idiom, Latour chooses to open certain "black boxes" - that is, to examine specific objects as performances rather than substances (p.44).

In the next book Harman examines, We Have Never Been Modern, Latour tackles modernism and critiques how it turns quotidian, fragile actants into abstract nominalizations: "capitalism, imperialism, science, technology, domination" (p.62, quoting Latour pp.125-126). Harman becomes very enthusiastic about such critiques, and even proposes that "Although in my view (but not Latour's) Heidegger is the greatest philosopher of the past century, there are certainly moments when he deserves to have his tyres slashed" (p.68).

Moving along, Harman next discusses Pandora's Hope. And here he begins to critique Latour more directly, particularly in Latour's handling of Plato's Gorgias. In Latour's reading, although Socrates and Callicles are nominally at odds, they have a common enemy: the crowds (p.86). Harman, however, thinks that although this critique has some merit, Socrates is really the open-minded and questioning fellow that he portrays himself as being, and in this sense Socrates is "the ally of Latourean metaphysics" (p.91). (Although I'm no metaphysician, I don't buy this argument.)

The second part of the book was, frankly, less interesting to me because it deals with a lengthy comparison and critique of Latour with various metaphysicians. I agree with Harman that "Latour ... cannot explain change" (p.130) - more precisely, ANT has no strong account of development (see my book Network). But I disagree with Harman's understanding of Latorean relationships, which he seems to interpret as componentization:
But Latour's philosophy does not even allow for the existence of microparticles due to the infinite regress implied by the principle that black boxes can always be opened. Hence, it is all the more important for Latour that he allow genuine non-relational reality to emerge at each level of the world. And though he never passes the buck of reality downward to an artificial stopping point in the purported final kingdom of quarks, he does pass it upward to the outward effects an actant has on its neighbors. But the buck must never be passed in either direction. The reality of an object belongs to that object - not to its tiny internal constituents, and also not to the larger collectives in which it is immersed. (p.163)
If I'm reading Harman right, he's taking the metaphor of the black box too literally here, as many tend to do. If you imagine yourself opening a literal box, you'll find only things that fit inside that box - smaller things. If you dismantle a machine, you get the components of that machine. But Latour's concept of the black box isn't that of components but the associations from which the black box is composed. If I opened my phone, I'd find components; but if I "opened" it as a Latourean black box, I'd find Google and Taiwan. When Latour does talk of tiny components, it's in the context of comparing them: for instance, in "Drawing Things Together," he emphasizes that whether we examine microparticles or galaxies, we tend to view representations that are physically the same size, about the size of a piece of typing paper. In Latour's conception, the world doesn't have "levels"; the world is flat.

In any case, despite my misgivings over a couple of points in Part II, I thought this book was a keeper. It's certainly worth reading if you're interested in Latour or in metaphysics.


bhawk said...

Been working on an RSA paper the past few days that looks at _Prince of Networks_. Nice to see your review pop up. One of my key issues is the spatial emphasis of ANT and lack of temporal projection. Like your take on the black box too. Might be useful.

Clay Spinuzzi said...

Right, I saw this book on the reading list for one of your classes and I thought: Hawk's way ahead of me. Looking forward to seeing that RSA paper once you're finished!

Clay Spinuzzi said...

Right, I saw this book on the reading list for one of your classes and I thought: Hawk's way ahead of me. Looking forward to seeing that RSA paper once you're finished!

Tom Haskins said...

Once again I'm delighted by your look at something we've both read. I'm also no metaphysician (or rhetorician), but I benefit from observing how conceptual frameworks get applied into new areas.

I've been thinking all the fuss about the opposition between Socrates and Callicles made by Harman, as well as Latour, reveals one of the biggest challenges of enacting ANT. It's easy to formulate a sociology, it's a huge step to allow everyone to have their own sociology. Likewise, it's almost a "no brainer" to make an issue of the accumulation and abuses of power, but it's a huge leap to see power granted by those "not in power" and "those in power" being not really having power over others to accumulate and abuse. Ant invites us to act on a highly developed sense of other - which is highly objectionable to selfish, power-tripping egos.

Your book: "Network" - gave me the question of "how to enact ANT as if it's an ontology?" That led me to place so much emphasis in my grasp of ANT on "taking interest in others' interests" and "observing assemblages of interests". That helps me not get ensnared in issues like ANT's denial/refutation of inner qualities, micro levels or developmental timelines. When everything is interests and interactions among those interests, so much else becomes conceptual baggage.

One takeaway I've gotten from this write-up of yours adds another way to enact ANT: by opening black boxes that contain associations, not components.

Thanks for all of this Clay!

Clay Spinuzzi said...

It's two-way - I've been learning a lot from reading your thoughts on The Shield of Achilles myself!

Related: I just finished reading a book called Between Hierarchies and Markets that tries to compare different conceptualizations of networks, including social network analysis, ANT, and Castells' "network society" work. Not sure what I think of it yet, but the author is trying to sharpen up the concept of networks to make it more analytically useful. Will review it soon.

Tom Haskins said...

I'm looking forward to reflecting on your thoughts on your reading of Between Hierarchies and Markets. I've added that title to my Google books library in case your insights provoke my interest in reading portions of it.

Alan Rudy said...

My sense of black boxes is that the term is better understood in action - black boxing or networkings that have been made sufficiently stable so as to be treated as black boxed. Its not so much that black boxes have components or relations, its that stabilized associations engender black boxedness and foster black boxing - acting as if networks of associations are things.

As I've noted before, one of the things that drives me insane about recent Latour is that black boxing and black boxedness have largely disappeared - except in so far as he argues that stable associations can be studied by means of a sociology of the social, while emergent associations can only be studied via ANT... a differentiation which appears to undermine the whole reason for unpacking stable and reified black boxed phenomena in the first place.

The point about Latour having a sense of space but not one of time is spot on because it is exactly the historical continuity of the stability of particular black boxings - actions grounded in the assumption that capitalism, the state, science, technology, nature, society are things - have been maintained (even if, as Latour oh-so-rightly shows, the networkings assumed to be stable are produced differently - and therefore are NOT the same "things" even though they are treated as if they are.)

What Latour never addresses is the capacity of some associations to project stability despite change... the essence of power, politics and history.