Conclusion

Violence was certainly pervasive in the life of steppe nomads, in the sense that the whole society, rather than just a class of warriors, was exposed to violence to defend oneself or protect one’s community.

It is exceedingly difficult to find unequivocal evidence of the reasons for a transition from low to high levels of violence, and to determine whether a political, economic, climatic or other cause may be behind the social and political crisis that ignited an acceleration and expansion of militarisation and warfare. There is certainly not one single cause, but equally certainly the starting point of the crisis has to be found within the nomadic world.

The basic argument to explain the transition from a multi-polar system of aristocratic domains - within which a further system of class divisions and clan orders operated - to a centralised system of power based on a single monarch, a ruling clan, and rigid political and military hierarchies, is to resolve the social crisis and put an end to exhausting cycles of violence. This ‘supra-tribal' condition (as it has been defined) shows considerable differences from case to case, according to internal customs and traditions of the people themselves, and in some cases the ‘supra-tribal' chieftain, as noted above in the case of the Kitan, was elected for a limited term only, indicating the reluctance of the tribal aristocracy to relinquish their autonomy and privileges indefinitely to any single person.

One may ask why, once a new political order had been created, war continued, often beyond the ecological and territorial boundaries of the steppe state itself. These campaigns are what the reputation of the nomads for savagery is pinned on. The answer may well reside in the excessive financial requirements of the new state, which the local economy of the steppe, based on very limited if any surplus production, could hardly satisfy.^[50] Lavishly wealthy aristocracies, large standing armies and a plethora of service agencies that furnished court and emperor with all that they required con­tributed to a large increase in the state budget, which tended to balloon generation by generation as the imperial clan grew and its requisitions became more burdensome. In order to sustain this political machine, the state had to increase its revenues, and one way of understanding the political changes occurring within Inner Asian empires is to focus on their ability to extract revenues from their neighbours. Violence or the threat of it was one (but not the only) resource that nomads had at their disposal to obtain additional revenues. Violence was, however, often resorted to in order to exact tribute, impose advantageous trade conditions and conquer sedentary populations that could be taxed.

The violence applied to all these processes varied from time to time and from situation to situation but it cannot be regarded as indiscriminate.

The standard definition of the roving and stationary bandit theory is that ‘uncoordinated competitive theft by “roving bandits” destroys the incentive to invest and produce, leaving little for either the population or the bandits. Both can be better off if a bandit sets himself up as a dictator - a “stationary bandit” who monopolizes and rationalizes theft in the form of taxes'.^[51] Based on this definition, one may question whether the supra-tribal chief is a variant of the ‘stationary bandit' who imposes his dictatorial rule over the nomadic population. As we have seen, the existence of a supra-tribal leader requires a measure of consent from the rest of the tribal aristocracy. However, there is no question that the supra-tribal chieftain has the right to tax the nomadic population in the form of requisition of people and animals for corvee labour and army service. But the surplus that could be extracted was minimal, given the low productivity of the grassland. What, then, happens when the amount of taxes that can be extracted from the population falls far below the state's requirements?

This is where Olson's theory becomes interesting to the extent that it may stimulate a further reflection about the projection of Inner Asian military power outside the nomads' own economic zone, and they become either roving bandits, engaging in raids against the sedentary population to loot and extract ‘protection' payments, or stationary bandits, engaging the fiscal exploitation of conquered areas. Violence was required in both cases, but the latter also required a transformation of the state itself through the adoption of institutions that allowed for a rational and systematic fiscal system.^[52] Such transformations are the building blocks of the history of steppe nomads, and the use of violence, as well as attempts to limit it, are an integral part of that evolutionary process.