When does a frozen conflict thaw?

In recent years, we have observed an increasing destabilisation of the world order. Part of it is due to the thawing of certain territorial disputes that, after being “frozen” for a number of years, have recently flared up in major escalations. The Russian invasion of Ukraine, the war in Gaza, the 2025 India-Pakistan Crisis, or other events that received less media coverage, such as the 2023 Azerbaijani offensive in Nagorno-Karabakh or the end of the ceasefire in Western Sahara in 2020, are all examples of thawing frozen conflicts that have somehow contributed to the current international instability. Beyond those disputes that have recently escalated, we have other disputes that, despite remaining frozen, keep posing a looming threat to international peace, such as the Taiwan or the Kosovo issue. Most of these disputes are related to territories that either have declared independence or seek some degree of separation from their original states. Despite having achieved de facto independence, however, their sovereignity remain significantly disputed on the international level, with a significant portion of the international community (at times the whole international community) continuing to recognise the territory as legally part of their original state. The dispute usually starts with an all-out war between the central government and separatist groups, resulting in a frozen situation where a ceasefire consolidates the insurgency’s de facto control over a certain portion of the territory, but no peace deal is brokered that finally ends the conflict. The role that these disputes are playing in shifting the international order makes it ever more important to understand when they violently thaw after a period of freeze. That is what I explore in my recent article “From ice to fire: A quantitative analysis of violent thawings in non-international frozen conflicts.” Here I illustrate some of the findings from my study.

A frozen conflict can undertake three phases. There can be a stalemate where the active hostilities are completely halted, or they continue with low intensity in the field; it can escalate violently into new all-out military confrontations; or it can undergo processes of peaceful transformation. But when do the parties decide that a military solution is more convenient than maintaining the status quo or proceeding with a negotiated deal? First, we can highlight certain events that have a “disruptive” effect on the stalemate, shifting the cost-benefit evaluation from maintaining the status quo to resorting to a military solution. The rise to power of a new leadership with an uncompromising stance on the issue can destabilise the stalemate. Long-standing disputes that remain highly prominent in public discourse can serve as a powerful tool for rallying support for a new incumbent. After the 2003 Rose Revolution, for instance, Saakashvili came to power by vowing to restore Georgian territorial integrity. His aggressive rhetoric towards the separatist region of South Ossetia, which enjoyed de facto independence since 1992, contributed to the escalation in 2004, shortly after its first election, and in 2008, a few months after he had regained power.

Alliance shifts represent another set of disruptive events that can trigger a violent thawing. These occurrences not only impact the immediate balance of power but also signal broader geopolitical realignments that can encourage one party to attempt to extract better conditions through military means. That can happen in different ways. The fact that key allies diminish their support because they are reapproaching the other side reduces the potential costs of a re-attack while increasing the chances of a decisive victory. For instance, the growing alliance between Russia and Azerbaijan led the Armenian government, formerly backed by Moscow, to gradually talk about normalising relations with Baku, going as far as recognising the disputed Nagorno-Karabakh as part of Azerbaijan’s territory. Once the de facto authorities in the region were left without solid allies, Azerbaijan had the incentives for re-attacking, finally dissolving the separatist entity that had been de facto independent since 1994. Another possible scenario is that the party that is damaged by an international rapprochement decides to intervene militarily to boycott the process, as Hamas did in the 7th of October attack amid advanced talks between Israel and Saudi Arabia for normalising relations within the framework of the so-called Abraham Accords. Finally, one party may take advantage of the other side’s fragmentation to play one faction against the other. For example, during the Kurdish Civil War, the Iraqi government sided with the forces of the Kurdish Democratic Forces, former adversaries in the 1991 Iraqi uprisings, against the Patriotic Union of Kurdistan to foster its influence in the separatist region.

On the other side of the spectrum, we have stabilising factors that consolidate the stalemate, rendering a violent escalation less likely. A peacekeeping operation can decrease the probability of violent thawing in several ways, including imposing reputational costs on the party that defies its mandate, reducing distrust between the belligerents, preventing accidents, and facilitating the implementation of peace arrangements. In this respect, relatively stable frozen conflicts, such as that between Serbia and Kosovo, are usually monitored by broadly-mandated peacekeeping operations.

The presence of a powerful patron backing the de facto state is another factor that strengthens the stalemate. The frozen conflicts in the post-Soviet space offer a good example of it. Territories where Russia maintained a military presence, such as Transnistria, Abkhazia, and South Ossetia, have displayed relative stability. On the other hand, Nagorno-Karabakh, which enjoyed the military protection of a much weaker patron such as Armenia, experienced several violent thawings over the years.

Finally, autonomy arrangements can temporarily accommodate the separatist demands, promoting trust between the parties while reducing the incentives for new military confrontations. For instance, the partition of Bosnia-Herzegovina into two federal entities was one of the pillars of the Dayton Agreement, which froze the conflict in 1995 and maintained the dispute relatively stable over the years. On the other hand, the disbandment of the Southern Autonomous Region, established as part of the Addis Ababa Accords, contributed to the re-ignition of the conflict in South Sudan in 1983.

Disclaimer: Views expressed are of the author(s) and do not necessarily reflect the views of The Statecraft Institute.