For the last couple of years I have increasingly become hostile to structural models of the social. The reason for this is not that I do not think that the concept of structure doesn’t often present highly useful and illuminating models of social relations. Like any formalization, a model allows us to discern patterns and relations that are not immediately evident from the standpoint of the buzzing confusion of the empirical world, and to discern possibilities of combination that would otherwise be impossible to conceive. Problems with the concept of structure arise, however, from two sources: 1) as employed in Continental social and political theory, the concept of structure all too often renders too much invisible, leading us to ignore the work of engineering or public works through which social networks are organized, maintained, and reproduced. As a result, we end up asking the wrong sorts of strategic questions in our political theorizations. When we look at a mathematical category, for example– a diagram that is not remarkably different from a structure –we are dazzled by the elegance and the simplicity of the model, the manner in which it brings a sort of clarity to the world, forgetting that the elements related by these arrows require painstaking processes through which they are brought together.

2) When we ontologize structures, treating models as realities rather than descriptions of realities, we are led to ask all sorts of misguided questions. Suddenly it is the map that does all our explanatory work, rather than the map functioning as the tracing of a set of fuzzy patterned interactions that themselves require explanation. This is always the problem when we appeal to “social factors” to explain some phenomena. We treat the very thing to be explained as what does the explaining, making appeals not unlike appeals to Zeus to explain lightening. But worse yet, when we ignore how patterns, relations, and ongoing forms of organizations are formed, our social explanations lead to a sort of theoretical pessimism as a structure or a network of power takes on the appearance of an iron law of necessity from which escape is impossible. The crystalline beauty of a diagram or a structure takes on the appearance of a pre-established harmony. However, what we should aim for is not a pre-established harmony, but, as Latour puts it in Irreductions, a post-established harmony. The real mystery is not how it is possible to produce change, but rather how certain networks manage to hold together at all.

read on!

As I reread Graham’s Prince of Networks, this time back to front, these issues are very much in the forefront of my thought. In particular, Graham manages to illustrate just why the concept of translation is so important in conceptualizing social networks. As Harman writes in a characteristically gorgeous passage,

…the means of linking one thing with another is translation. When Stalin and Zhukov order the encircling movement of Stalingrad, this is not a pure dictate trumpeted through space and transparently obeyed by the participation of actors. Instead, a massive work of mediation occurs. Staff officers draw up detailed plans with large-scale maps that are then translated into individual platoon orders at the local level; officers then relay the orders, each making use of his own rhetorical style and personal rapport with the soldiers; finally, each individual soldier has to move his arms and legs independently to give final translation to the orders from above. Surprising obstacles arise, and some orders need to be improvised– the enemy melts away at unexpected points but puts up stubborn resistance in equally startling places. (my emphasis) Move from war to logic, we find that even logical deductions do not move at the speed of light. Deductions too are transformed one step at a time through different layers of concepts, adjusting themselves to local conditions at each step, deciding at each step where the force of the deduction lies and where possible variations can be addressed or ignored. No layer of the world is a transparent intermediary, since each is a medium: or in Latour’s preferred term, a mediator. A mediator is not some sycophantic eunuch fanning its masters with palm-leaves, but always does new work of its own to shape the translation of forces from one point of reality to the next. Here as elsewhere, Latour’s guiding maxim is to grant dignity even to the least grain of reality. Nothing is mere rubble to be used up or trampled by mightier actors. Nothing is a mere intermediary. Mediators speak, and other mediators resist. (15)

The problem with structural models and discursive models more generally– not to mention with models that think change is produced through “ideological critique” –is they treat these structures, concepts, and signifiers as if they are transparently disseminated throughout the social field faster than the speed of light. What they forget is the labor of translation and mediation, or the way in which differences must be passed from one actor or object to the next in a field that is never a smooth space. If there is something deeply refreshing and liberating in the concept of translation, then this is because it brings back into view the processes, the interactions, and the resistances that organization must navigate in producing itself. Suddenly the bits of papers upon which messages are sent, the presence or absence of electricity, whether or not there is a storm, the sinews and muscles of the soldiers, supply lines, satellites, etc, become relevant to understanding the dependencies of an ongoing social organization and how it is formed. In other words, our attention is drawn to problems of engineering. As a result, strategic possibilities are multiplied profoundly, for it becomes possible to discover all sorts of weak links. In addition to this, it becomes clear that organization is not the result of one centralized agency, for all these actors– both human and non-human –form alliances through which a particular patterned organization forms and sustains itself in time. It becomes difficult to determine whether we are acting or being acted.

It might be objected that such a complex set of interactions leads to a pretty grim prognosis where the possibilities of political change are concerned, as where before we might restrict ourselves to a few key phenomena such as certain ideological signifying constellations or an event through which a situation might be leveraged, we now discern a non-linear network of translations and mediators without center or origin. However, this is not the case. Latour’s thesis is that “[t]he more connected an actant is, the more real; the less connected, the less real” (PN, 19). In other words, drawing on the hackneyed example of the game of Go, the power of a piece is not an intrinsic feature of the piece, but is rather a function of its relations to other pieces on the board. The power of the Go piece can fluctuate depending on the relations that come into being.

The key issue of engagement thus becomes one of forming alliances.

Since actants are utterly concrete, they do not have an inner kernel or essence encrusted with trivial additional properties. Actants are always completely deployed in their relation with the world, and the more they are cut off from these relations, the less real they become. Pasteur initially stands alone in his fight with Liebig over the cause of fermintation, or with Puchet over spontaneous generation. Gradually, Pasteur amasses a formidable army of allies. But notice that not all of these allies are human. Despite the word ‘Prince’ in my title, latour is no Machiavellian reducing truth to human power games. Instead, Pateur’s motley allies include mighty politicians who grant him funding, pieces of glassy or metallic equipment, and even bacilli themselves. Actors become more real by making larger portions of the cosmos vibrate in harmony with their goals, or by taking detours in their goals to capitalize on the force of nearby actants. (19)

While I do not share Latour’s thesis that objects are nothing but their effects on other actors or objects, his account of the degree of reality belonging to an entity and its relations to other entities is powerful. As alliances are formed and relations grow, new objects come to be produced, rebounding on the existing network relations and forcing further organization transforming translations to take place. It is in this growth of relations in the formation of alliances that real change is to be located. Perhaps the reason that the necessity of getting down in the dirt with mediators and translations is so ubiquitous in so much political theory as that in deploying our differences we come to see that these differences are not deployed transparently, but rather that the difference that issue from us enjoy adventures that we cannot master or control.