Julian Assange’s conception of the late-imperial state in topological terms, or in those of graph theory, seems to stem from his politics as well as his professional background. But it’s also a natural tool for anyone with his practical concerns.
We’re dealing here with a star network in which each peripheral ‘leaf’ or vertex (embassy or consulate) shares an edge (i.e. transmits and receives information) only with the central node (the State Department).
This could allow us a new way to map Washington’s hub-and-spokes system of alliances and influence. Start with the number of cables exchanged per month between the State Department and its various embassies.
Take the proportion of this total correspondence accounted for by each individual point of origin (i.e. the source country or diplomatic mission). The time average of this figure tells us, in graph-theoretic terms, the relative weight of that edge – the strength of the pairwise link between vertex and State Department.
Without even reading the content of the cables themselves, this gives us useful information. Unfortunately, we don’t have all the documents sent in a given period, or know how the sample was drawn. But it seems that, for shorter time scales (say, year-by-year), we nonetheless have some idea of the actual variation and a rough guide to the proportions involved.
It would also be an interesting exercise to develop a vector-autoregression model relating the number of cables sent each month from embassies and consulates for which we’d presume some interdependence: e.g. Dili, Canberra, Honiara, Jakarta, Port Moresby, perhaps Kabul and Baghdad.
For a single diplomatic mission, we could test if the time evolution of data points, and the value observed in any given month, is merely dependant on its own lags or also on those of the other relevant embassies (or if it’s all just noise). How much information does the number of cables sent from embassy x at time t provide about the number sent from embassy y at time t + 1?
To know this wouldn’t be merely of academic interest, but would help us understand the role of regional ‘deputy sheriffs’ like Australia.
Unfortunately this can’t be implemented for the first hypothetical case that comes to my mind.
In the Cablegate leak, there are no (non-top-secret) documents out of the Canberra embassy from early 2006. Unsurprisingly, cables from the Dili embassy spiked just before the 2006 removal as Prime Minister of Mari Alkatiri (and again in early 2008, around the time of the curious assassination of the rebel leader, Alfredo Reinado).
The fingerprints of Australian military, intelligence and diplomatic personnel were all over these events, after Greg Sheridan fretted in The Australian:
[If] Alkatiri remains Prime Minister of East Timor, this is a shocking indictment of Australian impotence. If you cannot translate the leverage of 1,300 troops, 50 police, hundreds of support personnel, buckets of aid and a critical international rescue mission into enough influence to get rid of a disastrous Marxist Prime Minister, then you are just not very skilled in the arts of influence, tutelage, sponsorship and, ultimately, promoting the national interest.
What I am arguing is that, as part of a wider program of assistance involving lots of Australian personnel operating in South Pacific government agencies, deployments of Australian soldiers should be semi-permanently stationed in East Timor, Solomon Islands and, if necessary, other regional basket cases.