In the halls of international institutions, a common refrain of diplomats is that “there are no military solutions.” The case of the Krajina should make these diplomats uncomfortable, for it clearly shows that sometimes there are military solutions. Additionally, the Krajina raises another, even more uncomfortable question: does ethnic cleansing work?

First, a word on what this question does not mean. Ethnic cleansing is a notoriously nebulous term, ranging from mutually agreed population swaps at one end of the spectrum to genocide at the other. Obviously, in addition to being a heinous tactic, large scale massacres like Srebrenica are highly destabilizing and can galvanize global opinion against the perpetrators, as the Bosnian Serbs discovered in 1995 when they were subjected to punishing NATO airstrikes. Long-term attempts to cleanse or subjugate minorities can “undermine the social cohesion of a country” and “is thoroughly counterproductive as a tool of state-building.” A far better strategy is to integrate minorities into the political and economic spheres of the country. But what if the ethnic security dilemma renders this impossible?

Ethnic cleansing in the former Yugoslavia “can be understood primarily as a policy designed to construct homogeneous polities over contested territories whose people’s allegiance could not be assured.” Klejda Mulaj argues that that “[f]or the most part, ethnic cleansing in this region reflects the official perception of minorities as a threat undermining the cohesion of the dominant nation.” But in the case of Croatia it was more than this. Croats were by far the dominant majority in Croatia from the beginning, and the fear came primarily from the fact that the Croatian Serbs were tied to larger Serb populations in Bosnia and Serbia, who together had formed a plurality in Yugoslavia. The Krajina Serbs were far more threatening because of their ties to Belgrade and the threat a “Greater Serbia” posed to the territorial integrity of Croatia. Operations Flash and Storm resolved this problem: by the end of the summer of 1995, Croatia’s borders and political nature were secured, and in subsequent years, Croatia could, and did, re-admit large numbers of its Serb population without them posing an existential threat to the nature of the state.

Ethnic cleansing is also harder to define when populations voluntarily or preemptively leave. By March of 1991, some 20,000 Serbs had already fled Croatia, most bound for Vojvodina. “From the very start, fear itself created large numbers of refugees.” Throughout the war, all sides would use a range of tactics to encourage widespread flight. Rape as a weapon of war was used to terrorize people into leaving. So were prisoner of war camps, where the prisoners’ families were told their loved ones would be released only if the families agreed to leave. Most Serbs in the Krajina fled before the Croatian army had the chance to throw them out.

Uncomfortable as it may be, clearly delineated territory certainly makes peace deals easier. This was true even in the most notorious cleansing episode of the war. During the Dayton negotiations, Western negotiators were “secretly relieved” that Srebrenica and Zepa had fallen because it made the subsequent division of Bosnia much easier to enforce.

“Neater maps” could now be drawn.

There is also strong evidence that federations with autonomous zones are sub-optimal at best, and can often be unworkable and undemocratic, particularly in lower-income countries. Most developing world federations “remain intact because of the authoritarian practices of their central leaderships, but this is a highly inefficient mode of governance.” Most federations created in the 20th century either broke apart or became partly or fully centralized. The former Yugoslav states are testament to this, with the most stable ones (Slovenia, Croatia) being the ones most firmly defined in terms of ethnic character and national identity, while the ones with significant ethno-geographic minority populations (Bosnia, Macedonia, Kosovo) are the least stable and the most dependent on international stewardship. Even today, Serb-controlled parts of Kosovo and Bosnia and Herzegovina would probably secede if allowed to vote on the matter. And even relatively peaceful freezing of conflicts can have deleterious effects on the lives and economic opportunities of inhabitants. Disputed “limbo worlds” like Nagorno-Karabakh, Somaliland, North Cyprus, and Abkhazia “have to try harder” to establish the legitimacy that the international community has not given them. They are often dependent on external support or indefinite peacekeeping missions; are at risk of becoming garrison states like the Krajina did; and face the near-constant risk of reconquest, or, failing that, of becoming “permanent second class state[s].” These are no conditions to build a stable, functional, liberal, and prosperous society at peace with its neighbors.

The region also has a history of large-scale population transfers that were followed by lengthy stability between former enemies with previously disputed territories. Greece and Turkey ended their war with an agreed compulsory population exchange involving two million people. At the cost of tremendous suffering to the displaced individuals, it achieved the goal of stability between the two states. During the Second World War, the Nazis not only perpetrated the Holocaust, but also deported large numbers of people, particularly in Poland, to “Germanify” territory they had seized. When the war was over, Eastern European countries retaliated by deporting nearly 12 million Germans back to Germany, over a sixth of whom died “from a combination of war, hunger, cold and disease.” The Czech Republic’s revenge against Hitler’s occupation of the Sudetenland was the “organized transfer” of almost two million Germans. The fact that no subsequent German leader would have again likely sought to use German minorities as an excuse for expansionism was now buttressed by the fact that there would be no Germans left to afford the opportunity. Of course, the causality of the stability that followed is disputed. Did the absence of Germans abroad end German revisionism and expansionism? Or was it the division of Germany between great powers and the onset of the Cold War? Or was it cultural changes within Germany itself after the fall of the Nazis?

The question of what would have happened if Serbs had freely been allowed to return in the immediate aftermath of Operation Flash, as Croats were, is a counterfactual and is thus unknowable, though surveys suggest that, through the former Yugoslavia, more people would have returned home had the right housing and economic conditions existed. However, minority returns can be destabilizing because they render “control of that territory once again uncertain, thus re-creating the same security dilemma that will help to escalate the conflict in the first place.” Zagreb’s initial policy avoided this outcome by ensuring undisputed Croat control of all of Croatia’s territory and the ethnic spoils therein. As with the founding of most European states, it was “the fait accompli of physical control through military force” that won the day.

In doing so, the HDZ created a climate conducive to more conciliatory returnee policies later on. The 2000s offered Croatian leaders a political landscape in which they could be more accommodating than in the 1990s. There were several reasons for the shift. First, with the war over and the regional political situation largely stabilized, the ethnic security dilemma posed by a breakaway Serb province was removed. Second, the subsequent Serb returnees were too few in number to pose a real threat, and they understood that they were “effectively finished as a united political force.” Third, the government’s plodding acceptance of more Serbs in power was done “mostly under international pressure and to meet other incentives, such as EU membership.” Fourth, electoral alliances of convenience (notably between the HDZ and SDSS) helped align political motivations to make returns easier. In short, the defining of national identity and borders through military force helped lead to political stability which allowed for greater returns and tolerance than likely could have happened otherwise.

Some have argued that the validity of an “ethnic security dilemma” was undermined by the fact that 100,000 Serbs ultimately did return, and that the state allowed them to. However, at less than a 20th of the population, this smaller Serb community posed no serious political threat to the established order, and even after returning faced consistent employment discrimination and underrepresentation in the civil service wherever proportional representation was not required by law. Also telling is that, despite a superior economic situation in Croatia than in Serbia in the early 2000s, two thirds of Serbs chose not to return.

The moral calculus of long term peace against short-term suffering is beyond the scope of this paper. But it seems clear that solidifying the ethno-nationalist identity of states through partition can, under the right circumstances, yield greater stability than other outcomes. Carter Johnson found that “partition is a uniformly effective tool in preventing a recurrence of war and low-level violence, but only if it includes the physical separation of ethnic groups.” Johnson also found, however, that incomplete partitions (such as India and Pakistan and Israel and Palestine, where territory or populations are still disputed) set the stage for long-term instability, and that “[b]ecause partition without the separation of ethnic groups does not increase the likelihood of securing peace, population transfers become necessary.” Whether these are through negotiated population exchanges, forced evictions, or voluntarily departures, the plight of the moved populations is usually tragic and harrowing. But as Andrew Bell-Fialkoff wrote in Foreign Affairs in the summer of 1993, whether people leave voluntarily or are compelled to go, fear and a desire for stability and peace “will accomplish the same end. ... With no sizable minorities left within any state and with the warring factions securely walled off behind “national” boundaries, the best that can be hoped for is that the motors of conflict will be disabled and the fatal cycles of violence that have marred Balkan history will finally have reached their end.”