1
TERRITORY

Self-determination conflicts are above all about territory. Even if they do not involve an outright demand for independence, but simply for internal self-rule, the demands made and the rights they invoke are either directly or indirectly linked to territory. Questions such as 'Whose homeland is it?' and 'Who has the right to govern this territory?' fuel these conflicts and are used to mobilize popular support. In the case of Crimea, Sasse (2002, p. 3) points out that the territory 'had multiple ethnic claims to it, all of which were plausibly historically grounded'. Even conflicts that are not initially about territory, but rather about the nature of the regime, can take on a territorial dimension. Rebels may abandon the idea of reforming the joint state and may indeed see the potential, for their community and for themselves, in having a separate state. The Sudanese People's Liberation Movement (SPLM) for decades fought for the creation of a reformed 'New Sudan' and only later came to demand an independent South Sudan (see e.g. Rolandsen, 2011).

The importance of a territory does not depend simply on its material value. Territory is symbolically important as the home of the imagined community. As Toft puts it, 'no matter how barren, no territory is worthless if it is a homeland' (2003, p. 1) and even barren territories can become the object of intense, violent struggles. For the central government, one of the main concerns is to maintain the (de jure) territorial integrity of the state and ensure that no territory is beyond their control. Territory goes to the heart of statehood; it matters both for legitimacy and identity, and for security and political power. If a group controls its own territory it is much more likely to feel secure, especially following a bloody intra-state war, and control of territory comes with power; this provides guarantees for the group as a whole and significant privileges for its leaders. It is therefore unsurprising that territory is one of the most contentious issues in peace talks, and the territorial solution agreed on significantly impacts on other parts of the peace agreement.

The International Community has generally promoted solutions that maintain the existing state, and although there have been some movements towards negotiated secessions in particularly protracted conflicts, the preferred option remains for self-determination to be realized through various forms of autonomy arrangements. The merits of autonomy are however contested. Proponents argue that autonomy offers a viable compromise between demands for separate statehood and for a unitary state (Ghai, 2008, p. 245). Critics however see it as an unstable solution which empowers separatist forces and risks causing the break-up of the state (see e.g. Roeder, 2009).

This chapter analyses post-Cold War trends in territorial solutions and identifies common problems encountered in the twenty cases. The analysis focuses on how these territorial solutions interact with the external and internal context, including dynamics and competition within the communities. It focuses particularly on the position of the separatist movement, which is often under-analysed in the existing literature. It finds that autonomy is at the core of almost all the peace agreements, either as a permanent or an interim solution. Somewhat surprisingly, these agreements typically do not include power-sharing provisions and the dominant model can be described as 'simple autonomy'. The overall track record is better than argued by the critics of autonomy, but the specific form of autonomy is crucial and the chapter points to two under-analysed factors that have led to serious tensions and even the collapse of settlements: the capacity of the autonomous territories and the ambiguity of autonomy provisions, especially in relation to the delineation of internal borders and the extent of autonomy. When it comes to the survival of autonomy arrangements, the devil is in the detail. The chapter finds only limited support for Bell's (2008) argument that we are witnessing the emergence of a 'new law of hybrid self-determination'. Although we do find some attempts to blur sovereignty, including trans-border dimensions and postponed independence referenda, the fudging of sovereignty is extremely difficult to sell to the parties and to implement. It will be interpreted in terms of centralization or secession. This is compounded by the prevailing dichotomous view of sovereignty: you are either sovereign or you are not; sovereignty cannot be divided, it cannot be shared. 'Simple autonomy' does not require the state to be fundamentally redefined and the fudging of sovereignty is limited, but it does address many of the demands made by separatist leaders. This helps explain its popularity, but also points to its limitations.

The Popularity of 'Ethnic Autonomy'

Wolff observes that 'territorial self-governance has been included in a very significant number of actual and proposed settlements' (2009, p. 28). Roeder agrees that 'ethnofederalism', in which the constituent units are homelands for ethnic minorities, is now embraced with enthusiasm as a solution to intra-state conflicts (2009, p. 204). He points to calls for a federation of autonomous ethnic regions in Iraq; the Boden plan for Abkhazia, which suggested making Abkhazia 'a sovereign entity . within the State of Georgia'; and the 'Dual federation' once suggested for Nagorno Karabakh, which would have linked the disputed entity to both its de jure parent state, Azerbaijan, and its kin-state, Armenia (ibid., pp. 203-4). More recently, a federal solution has been mooted for Syria (Reuters, 2016) and extended autonomy for Eastern Ukraine is also envisaged in the vague, and as yet unimplemented, Minsk II Accord (Sasse, 2016).

Why is territorial self-governance, in its various forms, such a popular response to separatist conflicts? As Wolff points out, the rationale is that it allows 'the different segments of diverse societies to realize their aspirations for self-determination while simultaneously preserving the overall social and territorial integrity of existing states' (2009, p. 28). Ghai (2008, p. 245) similarly argues that autonomy constitutes a mid-point of competing claims; that of separate statehood and a unitary state. It therefore presents a possible compromise in a situation where compromises are hard to come by and where zero-sum rhetoric dominates. For the separatist movement it offers self-rule and a degree of protection against the central government. For the central government it ensures the survival of the state within its existing border. This is particularly important in states with more than one potential separatist region. Autonomy may reduce the level of control exercised by the leaders of the central government, but it does not significantly undermine their power and privileges.

Autonomy, moreover, covers a variety of solutions. At one extreme, we find cultural autonomy which affords minorities self-rule over education, media and cultural affairs. Cultural autonomy does not have to be territorial and it does not come with wider legislative, executive and judicial powers. At the other extreme, we find solutions that maintain only on paper the sovereignty of the central government over the disputed territory. For all intents and purposes the autonomous region acts as an independent state with its own parliament, government, courts and security forces. In between these extremes we find various degrees of territorial self-rule; the extent of exclusive powers varies and the degree to which the arrangement is guaranteed for the future also varies. In some cases self-rule is combined with shared rule, for example in a federal system, whereas in other cases the ties to the centre are minimal. It is therefore not a blueprint as such, and as a solution it is clearly adaptable to the conflict situation (Hannum, 2004, p. 275).1

Its proponents may concede that autonomy, especially in the form of ethnic federalism, has a 'terrible track record' (Snyder, 2000, p. 327), but they counter that the major federal failures, such as the Soviet Union and Yugoslavia, were largely 'sham or pseudo-federations' which therefore provide limited lessons for today's mediators (McGarry & O'Leary, 2009, p. 9). The very idea of solving intra-state conflicts by giving autonomy to separatist groups has however been strongly criticized as a misguided and dangerous strategy. Roeder (2009, p. 206) argues that ethnic federalism or autonomy is 'particularly imprudent after civil wars' and Lake and Rothchild (2005, p. 112) find that 'warring factions have never realized full political decentralization along territorial lines as part of a civil war settlement' between 1945 and 1999. Ethnically-defined autonomy is criticized for reifying divisions; instead of addressing the divides that caused the conflict, it actually exacerbates them. As Erk and Anderson (2009, p. 192) point out, the paradox of group recognition is that it 'perpetuates the very divisions that it aims to manage'. But not only that, it also creates proto-states which reduces the cost of secession, thereby making it a realistic option (ibid.). Roeder (2009, pp. 212-13) similarly argues that the 'balance of coercive and defensive capabilities is likely to shift to the advantage of the secessionists' and they can therefore force...