In approaching the social conflicts that have wracked the infant Caucasian republics over the past fifteen years, it is useful to bear a broader comparative context in mind. State collapse and civil war have been the scourges of the period since 1945. Some 122 civil wars have wracked 73 countries, accounting for about 17 million deaths. Cross-national time-series datasets have helped us identify some broad patterns distinguishing states susceptible to civil war from those that have been relatively immune. Four points are especially noteworthy here. First, there is a statistically and substantively significant relationship between newly independent states and the onset of civil war. Twelve per cent of the new states admitted to the United Nations after 1945 fell into civil war within two years of independence. In this regard, the polities that emerged in Eastern Europe after 1989 proved even more susceptible than those created out of two earlier periods of imperial breakdown—in South Asia and the Middle East in the 1940s, and Africa in the 1960s. Of the Soviet and Yugoslav successor states, 30 per cent descended into hostilities in the decade after independence.

Second, there is an association between mountainous terrain—where rebels can hide from state militias—and the advent of civil war. Indeed, the Caucasian cases that are the focus of the book under review have ideal topographies to sustain guerrilla bands. Third, while the average length of all civil wars ongoing in 1999 is fifteen years, the average length of post-communist civil wars is just under four years. The latter are, to be sure, ugly, but they are shorter (and less bloody) than the average. Finally, all the post-communist wars were relatively quickly demobilized through an armistice—save Chechnya.

The quantitative data that revealed these patterns have their limits. Those analysing them can only speculate about the mechanisms that translate a propensity to civil war into actuality. Furthermore, these quantitative studies capture only a small part of the variance—they can explain the vulnerability of the post-Soviet Caucasus, but they cannot explain why Chechnya and Abkhazia suffered but Kabardino–Balkaria and Adjaria were spared. Georgi Derluguian’s ethnography of the Soviet collapse in the Caucasus provides tools that help turn correlations into compelling narratives. Derluguian traces the fate of three distinct social classes that had emerged after Stalin’s death in order to explain the events that proved so heavily destructive of wealth, life and security after 1991. His analysis reveals that it was the configuration of these classes in each of the republics which determined whether or not the post-Soviet polity found itself on the brink of chaos.

First, there is the nomenklatura, the high officials of the Soviet state. In the era of rapid development and constant purges, movement up the official status ladder was rapid, and offered educated youths from peasant backgrounds unprecedented opportunities for social advance. But post-Stalin, and especially under Brezhnev, growth slowed and young aspiring talents were blocked from entering the elite. Furthermore, the end of purges in the Brezhnev era allowed officials to use their positions for venal ends at no cost. Provincial Soviet life involved families buying state or party appointments, in order to then distribute bribe-friendly posts to relatives. All this was quite comfortable for the nomenklatura until the state began to unravel. They then had to make a historic choice: they could steal what they could of state assets and run; they could seek support from the newly reconstituted centre in Moscow to help them regain power; or they could transmogrify into nationalist elites and seek to lead independent states. How they chose had momentous consequences.

Second, there are the national intellectuals, a sub-class of the industrial proletariat. It may seem odd for Western sociologists to see intellectuals and professionals such as lawyers, doctors, teachers and journalists classified as proletarians. But this categorization is appropriate for analysis of Soviet society: as Derluguian points out, ‘like workers they were bound to state employment for the duration of their working lives, subordinated to a formal structure of bureaucratic management, and dependent on their wages’. To be sure, Derluguian acknowledges, there were celebrity artists and world-famous scientists in the major cities who escaped proletarian indignities. But in the provincial capitals, intellectuals received wages and lived in apartment complexes no different from those of skilled industrial workers.