John Ishiyama, a professor at North Texas University, has an interesting new paper at Communist and Post-Communist Studies using network analysis on the transformation of the North Korean elite from 1997-2012. The paper is framed in terms of the ongoing debate about whether there are identifiable factions in the regime or not, and particularly differences between moderates and hardliners.
Ishiyama uses on-the-spot guidance data for the period, classifying a limited group of top individuals—30 to be exact—into the categories of Conservative, Moderate or Open, meaning unknown. The classification draws on earlier work by Lim and Gause, supplemented by Michael Madden’s analysis at North Korea Leadership Watch. Data was inputted into the network program UCInet to generate graphic representations of the nature of the network—defined as frequency of joint appearances—during four periods: January 1997 to July 2002, the early “military first” period; July 2002 until July 2009, covering the onset of the second nuclear crisis; what I have called the “high succession” period through Kim Jong Il’s death, and then the first ten months of 2012.
The graphics show a higher concentration of military hardliners during the first period, followed by a subtle shift toward moderates in the 2002-2009 period, including Jang Song Taek, Kim Ki Nam, Marshal Ri Myong Su and General Hyon Chol Hae, each of which are profiled. The number of visits is less during the high transition period, but Ishiyama argues the network does not change.
Finally, in the early Kim Jong Un period Ishiyama finds an initial focus on military visits followed by a shift toward more civilian sites, including his new emphasis on leisure sites. Ishiyama notes the decline in Jang Song Taek’s visits, the marginalization of Kim Ki Nam, Ri Myong Su, and Hyon Chol Hae and the rise of a new group with uncertain views including Ri Jae Il, Colonel-General Hwang Pyong So, and Vice Marshal Choe Ryong Hae (who has recently been purged, as we predicted he would be).
Ishiyama has to confront the puzzle of the fact that the change in the nature of the network during the 2002-2009 period is not accompanied by a continuation of the moderate reforms launched in 2002; rather, we see what Marc Noland and I have called “reform in reverse.” In addition, this period coincides with the onset of the second nuclear crisis and the on-again off-again process of the Six Party Talks, yielding progress briefly in 2005 and again in 2007-8 but ultimately collapsing.
In sum, there does not appear to be any link between the network and policy. One theoretically plausible reason advanced by Ishiyama is that the regime’s moderately more reformist inclinations are stifled by the souring external environment. Yet even if we concede Washington’s bungled policymaking during the first Bush administration, there is the question of the almost immediate reversal of the 2002 reforms and the fact that the conflict started as a result of a fundamental breach in the Agreed Framework in the form of North Korean dabbling in enrichment.
The alternative explanation is that the conservative-moderate distinction may not make much sense in such a system; the coding of officials along this dimension is not well-explained and is ultimately not well known. It is plausible to assume that there is some heterogeneity in individual-level preferences. But that is quite different from the claim that these are politically salient: that groups vie over or influence policymaking in any enduring way. Rather the leader chooses those he prefers, but also dispenses with them at will as well. In such circumstances, the regime cannot be characterized as “factionalized” because the factions are not enduring features of the regime’s political economy but subject to central control.
The succession also poses puzzles. Under Kim Jong Un there has clearly been an influx of a cadre of younger elites in to the circle surrounding the new leader. But the most salient fact about them is not their factional identity but that they have replaced both previous prominent moderates and conservatives. What we are seeing is not a shift in factional composition defined along a conservative-moderate spectrum but rather a shift of generations and loyalties.
We are still inclined to see the regime as personalist with shifting coalitions related to power-sharing and coup-proofing aims rather than ideological divisions; we have made these points in our posts on Jang Song Thaek. But it is refreshing that Ishiyama and others are using these network techniques and we will report on similar work we have been doing with Luke Herman and Jaesung Ryu along similar lines (forthcoming in Asian Survey).