Military Aircraft Sales Visualized (1990)

I am working on a new paper that uses SIPRI data to look at more fine-grained transactions than the simple TIV value that I used for my dissertation research.  The project that I am working on is using these data to test some hypotheses that are related to realist behavior by states.  I am in the midst of coding the data.  The coding itself has been a good project to get me down into the details of arms transactions by states.  I am looking at transactions by examining the trade registers for all years from 1990-2013.

I have completed my first year of data coding.  Overall there were 215 distinct military aircraft sales in which aircraft were delivered in 1990.  This involved 26 suppliers, the largest two were the USSR and the United States.  Even more states were the recipients of these aircraft sales.  This included the sale of all types of aircraft ranging from small helicopters to large cargo planes.  The most interesting sales are those involving fighter jets and jets that have other strike capabilities.

Since I can't help it, I have visualized the sale of aircraft between states as a network.  The graph probably needs some work.  I plan on cutting it down into chunks, aircraft by type, at a later time.  For now, though, I thought it would be a fun graph to post.

Data on arms sales come from the SIPRI data.

Just an update on R coding

I'm teaching an introduction to political analysis course at UNLV this spring.  I have gone out on a limb and am teaching the students R instead of STATA or SPSS.  I am more familiar with STATA, but do my network analysis work in R.  I am trying to become more proficient.  I had a problem with a project I am working on - trying to count the number of rivalry ties that a state has in any given year. Normally I would do this in STATA, but I decided to give it a whirl in R.  After two hours and numerous "googlings" I have produced the code to give me the count for each year of my data.  The code looks like this:

for(i in 1816:2010){
  fp <- file.path("network", paste("net", i, ".txt", sep = ""))
  outcount <- file.path("count", paste("count", i, ".csv", sep=""))
 
  file <- read.table(fp)
  #sum row - this is the number of out ties by a state
  count.row <-rowSums(file, na.rm=FALSE, dims=1)
  count.col <- colSums(file, na.rm=FALSE, dims=1)
  comb.ties <- cbind(count.row, count.col)
  total.ties <- rowSums(comb.ties, na.rm=FALSE, dims=1)
  write.csv(total.ties, file= outcount, row.names = TRUE)  
}

Created by Pretty R at inside-R.org

The data come from text files with a network matrix (sociomatrix) for the states involved in the international system.  I use these to examine network characteristics in other contexts.  Here, I have just used them as matrices and have used the summing features to get both the outgoing ties, states that identify others as rivals, and in ties, those that identify the state as rivals.  These are combined and summed for a total.  The total is then exported to a CSV file that can be used in other contexts.  It's blunt force, and perhaps inelegant, but it's mine!

The Enemy of my Friend - India Pakistan and Russia

This article popped up in my news feed this morning.  It is about Russia's anger that India is purchasing fighters from France. The gist of the argument is that Russia has been India's "true strategic partner" since the 1960s.  The evidence for this is that Russia has voluntarily abstained from selling weapons to India's primary enemy, Pakistan.

Russia now feels betrayed that India is not rewarding that loyalty. India is still the largest customer of Russian arms.  However, if India continues to seek alternatives, Russia could find a lucrative source of income drying up or at least becoming more difficult to assure.

Here is the money quote from the head of Rosoboroneksport:

“We do not sell military equipment to Pakistan, only because of our loyalty to India. But all other nations, with whom India does business, sell their weapons and systems to Pakistan. We are India’s true strategic partners.”

I had the data for arms networks laying around (for my dissertation) so I quickly put together graphs for both India and Pakistan.  These are the ego networks for the two states, which means that these are not the full world arms network, only those states involved in sales to either India or Pakistan.

Neither of these two graphs shows much difference in the composition of states selling weapons.  In the Pakistan network, shown at the bottom, Russia is located at the edge of the graph and is not selling directly to Pakistan.  In all other respects, though the networks are very similar.

I guess the question is now to what extent Russia has deliberately not sold to Pakistan.  If Russia has made a decision not to sell that is one thing.  If Pakistan has not tried to buy because it was being supplied by the US and its allies and wanted to maintain that source for its weapons, then this is a different conversation. The idea that Russia is engaging in a deliberate version of Heidegger's rule (The enemy of my friend if my enemy) is what is most fascinating to me about this entire episode.