TOWARDS EFFECTIVE CONFLICT PREVENTION:
A COMPARISON OF DIFFERENT INSTRUMENTS

Raimo Väyrynen

The inability of the international community to prevent and resolve violent conflicts in several regions of the world has given rise to strong political concern and, at the same time, stimulated new interest in their early warning and prevention. Both the United Nations (UN), governments, and non-governmental organizations have stressed the importance of acting early on and developing effective means to prevent the outbreak and the escalation of violent conflicts. This trend derives its momentum from the presumption that violence is easier to prevent and resolve at an early phase when issues are still specific and hence more amenable to transformation, the number of parties to the conflict is limited, thus reducing its complexity, and early measures are more cost-effective than later efforts1.

On the other hand, the recent emphasis on conflict prevention has been criticized for being both irrelevant and unrealistic. According to critics, "preventive diplomacy requires prescience, prescription, and mobilization" which cannot be generated in complex post-Cold War emergencies. Prevention is also said to omit the creative role of conflicts in society, to be illusory in assuming that preventive actions are inexpensive and risk-free, overlook problems of implementation and, in so far as it aims to eliminate the root causes of conflicts, suffers from economic and ecological determinism (Stedman, 1995).

With few exceptions, neither the advocates nor the critics of preventive action have systematically studied its tools and consequences in real conflicts. Thus, both active and passive approaches to preventive action tend to base their views on general and often stereotypical ideas of whether third-party actions, in general, are able to ward off the outbreak and escalation of violent conflicts. Instead of inconclusive debates on the advantages and drawbacks of these approaches, we need more solid knowledge on whether preventive diplomacy can be effective and, if so, under what circumstances. In other words, who should use what kinds of instruments of prevention against whom, and when?

In principle, such a systematic knowledge can be acquired in two different ways; either by systematic comparative studies of preventive diplomacy or practical political experience.2 Solid scholarly studies of preventive action in inter- and intra-state crises are few and far between. Almost the only exceptions are rare efforts to develop conceptual foundations for different types of preventive strategies and test them by case studies. Such studies have concerned, for instance, actors and specific preventive tools used by them to forestall the outbreak and escalation of violence in individual conflicts. These studies have started to enhance our knowledge about the limits of, and opportunities for, preventive action (which still remains, however, very rudimentary) (Chayes et al, 1995; Lund, 1996).

Political experience in preventive diplomacy provides a richer, but largely under-utilized, source of information. Alexander L. George has pointed out that theory and generic knowledge are usually more helpful in the diagnosis of specific problems than in providing prescriptions for action. Theory and knowledge may, however, contribute to policy prescriptions when couched in terms of conditional generalizations which identify: "(1) the conditions that favor successful use of each particular instrument and strategy, and (2) other conditions that make success very unlikely" (George, 1993: 17-18).

The utilization of practical knowledge is made difficult by the fact that few countries have formulated explicit policies to prevent deadly conflict. Neither there is much agreement on which means and strategies of prevention are appropriate and effective in each situation. In making choices between different instruments, governments face questions such as whether or not conflict prevention should involve only non-coercive means of mediation and persuasion, or if coercive tools should also be included in its toolbox? In the parlance of the UN, preventive diplomacy covers only operations undertaken under Chapter 6 of the Charter to settle disputes in a peaceful manner, while coercive measures are a part of enforcement policy under Chapter 7.

In addition, a better understanding should be gained on the priorities and relationships between the instruments of prevention used by the parties to the conflict, on the one hand, and by the third parties, whether governments or international organizations, on the other. In most cases the bottom-up prevention of conflict by local parties is not an option, either because of the intractability of the root causes of the conflict or because the crisis has escalated beyond the reach of local remedies.

Therefore, conflict prevention usually becomes the responsibility of third parties, at which point they have to make a primary choice between non-involvement and involvement. In the case of involvement, the main options are neutral mediation or more forceful intervention to stop the spread and escalation of violence. Mediation aims to improve the relations between the parties to a conflict, while forceful intervention tries to reorient their behavior by redefining the payoffs of the actors in conflict. Mediation relies mostly on good services and intervention on power to reshape the conflict situation (Princen, 1992).

A Conceptual Framework

This paper makes a distinction between three different phases of preventive action and three strategies of prevention. The phases are (a) conflict prevention, i.e. preventing disputes from arising between parties; (b) escalation prevention, i.e. preventing both the vertical and horizontal escalation of hostilities to more destructive forms of warfare and to involve additional actors; and (c) post-conflict prevention, i.e. preventing the re-emergence of disputes by reintegrating and reconstructing the society.3 In distinguishing these three phases of preventive action, I deviate from the standard solution that only efforts to forestall the outbreak of violence, and thus keeping the conflict latent, are regarded as prevention.

Conflict prevention can have different targets. The most comprehensive approach would target the root causes of violence. According to this view "a comprehensive preventive strategy must first focus on the underlying political, social, economic, and environmental causes of conflict" (Commission on Global Governance, 1995). Rather than addressing these "remote" causes of deadly conflicts, this paper is more interested in the more "proximate" causes, especially the manipulation of the incentives and payoffs of the conflict parties by means of different instruments of prevention. They provide more tangible instruments of preventive action than, for instance, the promotion of collective security and balance of power on the one hand or of development and democracy on the other.4

Provided that coercive means are considered legitimate in preventive action, it can rely on three different types of instruments, i.e. political, economic, and military ones. Political tools are derived mostly from diplomatic practice; fact-finding, monitoring, mediation, influence, promises, and threats. Economic tools rely on the manipulation of material costs and benefits by cutting off economic ties, promising their re-establishment, or providing outright rewards. Finally, military instruments boil down to the use of force or its threats in the case the parties continue to deteriorate the confrontation. It has been even suggested that defense policy in general should be defined in preventive terms: "it is the opportunity to pursue what I call preventive defense--that is, actions we can take to prevent the conditions of conflict and create the conditions of peace."5

In a more analytical vein, one can say that the parties to the conflict, or the third parties, can try to ward off the outbreak or escalation of violence by five different preventive strategies. They are reassurance, inducement, deterrence, compellence, or pre-emption.6 In reassurance parties try to convince the target about the futility of using force either by verbal appeals or commitments to specific future actions. In inducement parties initiate positive actions to stimulate constructive responses and to show that they are serious in their efforts to prevent violence. In an extreme case inducements may escalate into bribes by which "an actor transfers a resource to a target in exchange for a modification in the target's behavior" (Rothgeb, 1993: 110-117).

In a deterrence relationship the preventing actor communicates to the target that the violation of a norm would result, either because of denial or punishment, in major costs. While deterrence is a status quo policy, compellence requires an initiative by which the behavior of the target is altered. Compellence is a riskier behavior which is, after the failure of deterrence in the first place, continued "until the other acts rather than if he acts" (Schelling, 1980; Rothgeb, 1993: 139-140). Finally, in pre-emption another party is so convinced about the readiness of the target to use force that it decides to act in a premeditated manner to limit the extent of destruction. The Gulf War in 1990-91 is an example of conflict in which reassurance and inducement were never tried and deterrence failed. Therefore, compellence as a form of international enforcement was left as the only option (Rakisits, 1992; Gross Stein, 1992).

These five analytical strategies can be classified in several different ways. Conventionally, they can be divided into positive and negative tools on the one hand and active and passive means of influence on the other. Reassurance and inducement are active positive strategies, while compellence and pre-emption are active negative strategies. Deterrence, in turn, is a passive negative strategy which tries to prevent change by threatening to inflict negative costs. Passive positive strategies, such as the attractiveness of the economic wealth of a major power to bandwagon its policies, exist, but their role in preventive action seems to be limited.

Dominant means in active positive strategies tend to be either political or economic, since military instruments can seldom be used constructively (except for arms aid in specific cases). On the other hand, active negative strategies rely either on economic or military coercion which is expected to change the behavior of the target by the costs directly inflicted upon it. Passive negative approaches, relying on various forms of deterrence, tend to be military, but they may also contain economic and political elements. This preliminary evaluation suggests that political means of influence are the most versatile ones. They can be used in several types of preventive strategies and they address most effectively the causes of violence and its escalation. Economic and in particular military means have a more limited validity, even though they may have to be used when political and diplomatic tools fail to achieve desired results (Leatherman et al, 1996). On the other hand, the fact that political means, sometimes utilizing symbols and values, can seldom be used in isolation from economic and military instruments indicates a need to look at the interfaces of different strategies of influence.

Ultimately, the means of preventive action boil down to positive strategies (i.e. promises, persuasion, and rewards) and negative strategies (i.e. threats, coercion, and punishments). "Coercive diplomacy" is an example of a negative strategy, which aims to produce positive results by targeting the relevant actor with a specific demand, a time limit for compliance with it, and a credible threat of punishment. Thus, coercive diplomacy relies on the use of military or economic force that is selective, discriminating, demonstrative, and clearly informed by political objectives. In the end, coercive diplomacy is a strong form of politics rather than a form of warfare.

To avoid mistaking coercive diplomacy for a large-scale war and thus the risk of escalation, it should contain pauses and controls that give the target time and opportunities to understand the preventive motives behind the tools used. Preventive military actions can be limited in scope by targeting them, for example, at terrorist capabilities and facilities producing weapons of mass destruction.7 A further way to clarify the real purpose of coercive diplomacy is to communicate clearly those salient limits which the target should honor if it wants to bring the preventive coercive action to an end (Smoke, 1977: 241-242). Also economic sanctions appear to be more effective if applied in an instrumental and moderate way to achieve limited goals rather than used harshly and comprehensively to publicly humiliate the target (Eland, 1995).

In spite of the efforts to limit the means and clarify the messages of coercive diplomacy, it contains a risk of escalation, which can damage its original constructive purpose. Moreover, it is difficult to gain international acceptance for unilateral pre-emptive strikes unless there is strong proof available that the target is a genuine threat to international peace and security (Haas, 1994: 24-25). Recently, it has been increasingly stressed that positive incentives are probably a more effective mode of influence than coercive punishments, although they also may be more expensive because their costs cannot be substituted.

This is admitted also by Alexander L. George, the founding father of the concept: "whether coercive diplomacy will work in a particular case may depend on whether it relies solely on negative sanctions or combines threats with positive incentives and assurances". This amounts to concluding that coercive strategy may not be effective in preventing the outbreak of conflicts, though it may still be relevant in controlling the intra-crisis escalation of hostilities. Even then, a combination of carrots and sticks may achieve outcomes not obtainable solely by punishments or their threats. However, the terms of the preferred settlement of conflict must be clearly spelled out. This includes specific terms for ending the conflict, and provisions for their verification and enforcement to ensure that both parties hold to their commitments (George, 1991: 11).

Yugoslavia as a Test Case

The policy of the international community towards Yugoslavia in 1991-92 offers a test case on how preventive diplomacy has operated, and failed, in the post-Cold War context. In the preventive effort in Yugoslavia a plethora of international organizations was involved, including the Conference on Security and Cooperation in Europe (CSCE), the European Community (EC), the Western European Union (WEU), the North Atlantic Treaty Organization (NATO) and the UN. I have explored elsewhere the institutional causes of their failure focusing on such factors as the volatility of international order, inadequate coordination and division of labor between the organizations involved, and over-hasty imposition of the model of sovereign states on the federal, multinational reality of Yugoslavia (Vayrynen, 1996b).

This paper focuses, in its turn, on the means used in the preventive action in Yugoslavia rather than on the institutions involved. The basic question in this inquiry is the following; does the choice between different strategies and instruments of prevention have an impact on the outcome and, if so, what instruments either alone or in combination with others are the most effective ones? To be able to answer this query, I will briefly explore the political, economic and military means, as well as broader preventive strategies, used in the Yugoslav crisis (focusing primarily on 1991-92).

1. Political Instruments

The international community was slow to react to the burgeoning crisis in Yugoslavia. It was clear by 1989 - and certainly by the time multiparty elections were held in 1990 - that the Yugoslav republics were drifting apart and were ready to use force to achieve their political objectives. For Slovenia and Croatia, this objective was a loose con-federal system or full independence. This was resisted by Serbia and Montenegro, which sought a strong federal system capable of assuring Belgrade's dominance in southeastern Europe.

International action began in May 1991 when the EC sent a mission to Yugoslavia, headed by Jacques Delors and Jacques Santer, aiming to encourage a peaceful settlement. These and subsequent moves by the EC may be judged in different ways. On the one hand, it can be concluded that the Community acted quickly in the spring and summer of 1991 in launching its mediation activities (Lukic & Allen, 1996: 264-265). On the other hand, especially from a longer-term perspective, it can be judged that the EC acted too late, given that in March 1991 Serbia had already made a decision to use force.8 The latter assessment is consistent with a more general conclusion that EC initiatives in Yugoslavia were largely reactive and that anticipatory and proactive measures were neglected (Lucarelli, 1995: 9).

There are several reasons for the passivity of the international community. During the Cold War, any major crisis in the Balkans would have raised concerns in Washington and Moscow that their opponent would seek unilateral gains in the region. However, this situation had changed by the late 1980s. The Bush Administration defined the stability and coherence of the Soviet Union as its foremost priority and, in spite of the worsening situation in Yugoslavia, American and Soviet leaders paid practically no attention to the crisis in their mutual consultations. This represented a major change in comparison to the war and intervention scenarios of preceding decades.9

In principle, the Soviet Union supported efforts by international institutions to resolve the Yugoslav crisis. In reality, however, it remained wary of any external involvement in the "internal affairs" of Yugoslavia. The main reason for this was the parallel drawn by many observers between the breakup of the Federal Republic of Yugoslavia and the multicultural future of the Soviet Union itself. Moscow was especially concerned about the impact that the independence of Slovenia and Croatia could have on the quest of the Baltic states for independence. Above all, it did not want to establish a precedent permitting the West to overlook the principle of sovereignty and become politically involved in the internal problems of the Soviet Union. Moscow was also concerned about the possibility of external military involvement in Yugoslavia on the grounds that this might open the way for a more expansive Western military role in eastern Europe in general (Lukic & Lynch, 1996: 334-337; Wallander & Prokop, 1993: 97-99). For its part, following the Moscow coup of August 1991, the West avoided all moves in Yugoslavia that might have complicated the situation in Russia and the Soviet Union (Burg, 1995: 242-243).

In Western Europe, the EC was absorbed in drafting and (after December 1991) ratifying the Maastricht Treaty. It is ironic that while the development of a Common Foreign and Security Policy (CFSP) for the Community was a high priority, it failed to act decisively in the Yugoslav crisis. Instead, policy towards Yugoslavia became a bargaining chip between Germany, on the one hand, and Britain and France, on the other. Although the British and French governments were generally less critical of Serbia than were the Germans, they nonetheless acquiesced with Germany's determination to recognize Croatia and Slovenia in December 1991, largely in response to German concessions during the Maastricht negotiations on such issues as European Monetary Union (EMU) (Baun, 1995-1996: 621).

In addition to these agenda linkages, EC member states considered it important to retain a semblance of unity in their external actions. Unity of purpose and action was a key motivation for finalizing the foreign policy provisions of the Maastricht Treaty in December 1991. A failure to cooperate in recognizing Slovenia and Croatia would have boded ill for future cooperation in the external relations of the EC. Yet the Yugoslav experience also indicated that the CFSP would be unable to ensure collective action and would, in fact, prove little more than an incoherent combination of national foreign policies (Zucconi, 1996: 258-260).

In the multilateral context, the CSCE moved ahead in Yugoslavia in June 1991 by activating its emergency mechanism, sending monitoring and fact-finding missions to the region, and demanding immediate and complete cessation of hostilities in Slovenia and Croatia.10 However, these measures had practically no impact on the dynamics of the conflict. Even so, Hans-Dietrich Genscher (who chaired the CSCE Council of Ministers in 1991) viewed the CSCE as helpful in so far as it provided a forum for involving non-EC states - especially Russia - in crisis prevention in Yugoslavia, and gave external powers a degree of legitimate access to the republican leaders in Yugoslavia (Genscher, 1995). Genscher's judgement, however, may be viewed more from the vantage point of Germany's own interests than as a judgement of the collective effectiveness of CSCE crisis prevention in Yugoslavia.

These conclusions about a lack of effectiveness are also applicable to the actions of the EC. After the Yugoslav National Army (JNA) opted to use force to prevent Slovenian and Croatian independence, the EC brokered the Brioni Agreement on July 7, 1991, under the leadership of Hans van den Broek,. The Agreement sought to stop the fighting, provide a three-month breathing space in which the independence declarations of Slovenia and Croatia would not be implemented, and enable negotiations to begin between the republics. The Community also sent CSCE-sponsored missions to Slovenia and Croatia to monitor the implementation of the ceasefire.

The Brioni Agreement was ostensibly a triumph for the mediation efforts of the EC. In reality, however, it was facilitated by the collusion of interests between Slovenia and Serbia and sounded the death knell of the Yugoslav federation. The withdrawal of federal troops from Slovenia ensured the republic's independence, while Serbia gained additional military resources for later use in Croatia and Bosnia. The Brioni Agreement has therefore been assessed harshly: "it left every important item of contention unresolved...it simply put everything on hold" (Silber & Little, 1996: 166). It has been also noted that the Agreement pulled the rug from under the military and political factions in Serbia which were defending the federal order, thereby playing in the hands of radical nationalists in Croatia, Serbia, and Slovenia (Woodward, 1995a: 168-170).

A negotiation strategy based on the Brioni Agreement was pursued by the Hague Peace Conference, inaugurated on September 7, 1991. Chaired by Lord Carrington, the Conference was the main means by which the EC sought to broker a diplomatic solution to the crisis. The Hague Conference has since been castigated as a naive and futile exercise in multilateral peacemaking. However, this conclusion is too harsh given the extent to which Carrington genuinely strove to devise a constitutional settlement acceptable to all the Yugoslav republics.

Carrington sought a constitutional settlement which would have provided variable degrees of sovereignty to the individual republics. Such an arrangement would have enabled Slovenia to retain its de facto independence, tied Croatia into the confederation through various intergovernmental agreements, left Bosnia and Macedonia as constituent but semi-detached parts and Serbia and Montenegro as a core region. In fact, an "asymmetrical federation" of this sort had actually been proposed by Bosnia and Macedonia in the summer of 1991, before the Hague Conference had even begun. It now failed again as a basis for agreement by resistance from Croatia and - especially - Serbia (Silber & Little, 1996: 241-243).

Carrington was, nonetheless, able to produce a framework agreement entitled "Arrangements for a General Settlement". This aimed to establish a con-federal association of sovereign states based on existing borders, "unless otherwise agreed", protect minority rights, and promote economic co-operation and limited disarmament. Carrington's compromise was turned down, however, by Miloševi, since its implementation would have undermined his efforts to establish a Greater Serbia comprising Serbia, Montenegro and the Serb-inhabited parts of Bosnia and Croatia. An important reason for the Serb veto was that the provisions of the agreement on minority rights presented a double threat to Miloševi; they protected Serbs in Bosnia and Croatia without giving them the right to join Serbia, and conferred minority rights on the Albanians in Kosovo (Silber & Little, 1996: 164-165).

It can be argued that the Hague Conference never came close to a true settlement of the Yugoslav crisis. The process was dead when Miloševi chose, in October 1991 at the latest, to pursue a Greater Serbia by force of arms at the expense of a negotiated confederation. However, Lord Carrington, supported by the Dutch, came closer to an agreement in the Yugoslav crisis-- and thus the prevention of its escalation--than any other diplomatic effort at the time. The mediation process could only have succeeded in October and November 1991 with much stronger pressure on the Yugoslav parties from the EC, the United States (which was standing in the sidelines), and the UN Security Council.11

The EC was divided in its support for Carrington. As early as August 1991, for example, France had begun to collaborate with Austria in the UN Security Council in search of alternative paths of political influence in the crisis. True, UN Security Council Resolution 713 gave support to EC efforts in the Hague Conference in September 1991, but in a manner which can be construed as undermining Carrington's mission. It has been also argued that the West gave inadequate political and economic support to political forces in Serbia and Croatia which were opposed to the excessive nationalist policies of Miloševi and Tudjman (Woodward, 1995a: 147-181).

The expiration of the three-month moratorium on October 7, 1991 provided by the Brioni Agreement and the failure of the Hague Conference contributed to an escalation of violence in Croatia -notably Serb offensives against Dubrovnik and Vukovar. The German government intensified its campaign to recognize Croatia and Slovenia, buoyed in part by German domestic opinion (including a vocal Croatian lobby) which demanded support for Croatia in the face of the Serbian aggression. This recognition policy was premised on the break-up of Yugoslavia, which was later acknowledged formally in December 1991 by the Arbitration Commission of the Hague Conference chaired by Robert Badinter.

It has been argued that the German policy of recognition reflected its reluctance to participate in a military campaign in the former Yugoslavia. This observation is justified by the fact that Germany declared in favor of recognition only after Belgrade had used military force against Slovenia and Croatia. Thereafter, only two possibilities remained; either a collective military rollback of Serbian operations or the internationalization of the war through diplomatic recognition of the key actors (Lukic & Lynch, 1996: 270-273). Recognition was expected to create a political deterrent preventing Serbia from continuing its expansionist policy in Croatia and Bosnia (Maull, 1995-1996: 102-105, 121-123).

In opting for this approach, the EC, led by Germany, overestimated its own capabilities and underestimated Miloševi's tactical skills and commitment to his own political course. Most importantly, recognition did not, and in fact could not, contain any provisions on what would happen if Croatia and Serbia did not mend their ways. It was a one-move policy without any operational contingency plan on what might be done next.

The Community was badly divided on the recognition issue. In fact, it has been suggested that it was the very weakness of Western multilateral institutions and norms at the time that prompted Germany to defect from the US-EC consensus on establishing a confederal Yugoslavia. According to this interpretation, domestic pressures to recognize Slovenia and Croatia pushed the German government into a unilateral policy because it could not be sure that the other EC countries would go along with it (Crawford, 1996). However, Germany made a strong effort to ensure cooperation from other EC members and thus its defection from cooperation was partial rather than complete.

To avoid further divisions, the Community set up a timetable and a procedure on December 16, 1991, paving the way to recognizing all qualifying republics of Yugoslavia within a month. The key criteria for recognition were the inviolability of republican borders, a commitment to work for a comprehensive political settlement, and respect for human and minority rights (Woodward, 1995a: 183-189; Andersson, 1995: 343-346). Despite much evidence to the contrary, Genscher has since argued that the German-inspired strategy of preventive recognition worked because Serbia ceased its military aggression in Croatia (Genscher, 1995: 963-964). This argument is, of course, consistent with the interpretation that Germany's primary motivation was to prevent its own involvement in collective military operations in the Balkans.

Genscher omits the fact that Serbia had already achieved its main military objectives in Croatia--gaining control of one-third of Croatian territory--by the recognition date. The weakness of Genscher's argument can also be seen in the developing situation in Bosnia in early 1992, when recognition prevented neither Croatia nor Serbia from using military force to promote their political and territorial goals. In fact, it may even have encouraged their subsequent efforts to carve Bosnia up.

In addition, the recognition policy derailed any multilateral solution to the Yugoslav crisis. As Lord Carrington later described in his message to the meeting of EC Foreign Ministers on December 15-16, 1991: "I said very strongly that I felt that the timing of this was wrong. I pointed out that early recognition would torpedo the [Hague] conference. There was no way in which the conference would continue after that."12 It is worth noting that the recognition of Croatia was also contrary to the advice of the Badinter Commission. The basic lesson is that neither Croatia nor Serbia had any incentive to continue the multilateral process after Croatian independence had been recognized.

2. Economic Instruments

In addition to these political measures, the EC also tried to shape the situation in Yugoslavia by a combination of economic incentives and punishments. Economic instruments were potentially important, since Yugoslavia traded heavily with Western Europe and was dependent on it as a source of public and private money. As the crisis escalated in April and May 1991, the EC offered additional loans, credits, and an association agreement with Yugoslavia as an incentive for the federation to remain united. These carrots were not integrated, however, with any comprehensive political strategy (Maull, 1995-1996: 100-101; Lukic & Lynch, 1996: 262-263).

The EC policy of offering economic incentives to Yugoslavia did not last for long, however. This was partly due to the lack of US willingness to support economic rewards. Indeed, Washington suspended its aid to, and financial backing of, Yugoslavia in May 1991. This policy was not based so much on political considerations as on a reluctance on economic grounds to continue lending to debt-ridden Yugoslavia. In other words, the withholding of funds to Yugoslavia did not serve any clearly-defined political purpose. Instead, it exacerbated political tensions in the country fuelled by the economic crisis and pushed people away from liberals into the arms of the nationalists (Woodward, 1995).

As a part of a new economic strategy, the EC threatened in June 1991 to cut off economic aid to the parties if they did not agree on a ceasefire in Croatia and Slovenia. On July 5, 1991 the Community suspended all financial aid and banned arms exports to Yugoslavia to underpin its mediation effort at Brioni. The United States supported European efforts in September 1991 to stop the flow of arms to the region by pushing the UN to declare an arms embargo on all parties to the Yugoslav conflict (Security Council Resolution 713). This move froze the military superiority of Serbia, which controlled JNA arsenals of weapons. Subsequent American efforts to lift the arms embargo from 1992 onwards were opposed by Russia, which wanted Serbia to continue to benefit from it, and the UK and France which argued that lifting would jeopardize the safety of their ground troops deployed on peacekeeping operations in the Former Yugoslavia (Lukic & Lynch, 1996: 295-300).

EC sanctions were targeted more carefully on Serbia following the final Serb veto of the Carrington peace plan in the Hague Conference in early November 1991. The EC ended its economic cooperation agreements with Yugoslavia and announced that it would seek a UN ban on oil deliveries to Serbia. The latter move was dropped by the EC, however, in order to secure Serbian consent to deploy UN peacekeepers in the area (Goodby, 1995: 165-168). This incident suggests that the priority given to the deployment of neutral peacekeeping forces, and their protection for humanitarian operations, undermined efforts to influence Serbia's behavior by more coercive measures - in this case economic ones.

A more careful analysis of the timing and targeting of economic punishments suggest that the EC had a modicum of a coherent preventive strategy in the Yugoslav crisis. This is illustrated by Italian Foreign Minister Gianni de Michelis' offer to arrange an economic development package between the EC and Montenegro, with the intention of driving a wedge between Montenegro and Serbia during a crucial phase of the Hague Conference. However, although the initiative achieved some short-lived success, it was unable to prevent Montenegro from bandwagoning Serbia's policy in the longer term (Silber & Little, 1996:194-196).

In sum, economic instruments of prevention had only a very limited impact on the course of the Yugoslav crisis. Their use did not steer Serbia away from its expansionist policy, persuade Croatia to respect the rights of its minorities, or force the parties to agree on the Carrington plan. It has even been suggested that economic sanctions compounded the division between those forces which had decided to stay within Yugoslavia and those which wished to defect from it. Serbia did not expect help from the EC in any case, while Croatia had already received substantial promises of assistance from Germany. Different attitudes towards the use of economic instruments among the EC member states may have also deepened the political divisions which already existed between Germany and the others (Woodward, 1995a:175-176; Woodward, 1995b:144-145).

3. Military Instruments

Until mid-1991, the international community demonstrated a strong preference to use diplomacy, rather than military deterrence or compellence, to prevent the outbreak and escalation of the Yugoslav crisis. It also believed in the power of the European public opinion to sway the Balkan leaders. The Brioni Agreement prompted Genscher to declare that its diplomatic success made the use of military force to promote political goals inadmissible (Genscher, 1995: 940-941). This belief in the merits of mediation, together with a reluctance to use systematically more tangible forms of influence, ensured that "the EC negotiated with few instruments of coercion" (Goodby, 1995: 163-164).

The shift from economic rewards to punishments during the summer of 1991 reflected a growing recognition that stronger instruments of prevention were needed. By August , the Community had realized that even economic means were not adequate to cope with the unstable military situation in Yugoslavia and that some form of military means might be needed. This gave rise to proposals to set up a European "interposition force". Supported by France, in particular, which was seeking a more prominent role for the WEU in European security in general, and the Netherlands, the force was intended to separate the fighting parties and keep the peace. The creation of an interposition force would have added an element of military enforcement to the preventive tool box of the EC and encouraged closer cooperation between the EC and WEU (Lucarelli, 1995: 19-20).

Plans for an interposition force stalled, however, for several reasons. London (and Moscow) resisted politically; the command and logistical capabilities of the WEU were inadequate for such a task; and Serbia refused consent for a deployment. The establishment of a force was effectively ruled out by the EC foreign ministers on September 19, 1991, although they nonetheless asked the WEU to draw up a contingency plan for the potential use of military force. In November 1991, WEU planners recommended the use of naval force, but British opposition ensured that this plan was also rejected (Zucconi, 1995:168-171).

European unwillingness to send troops to Yugoslavia now made UN involvement essential. Serbian resistance to peacekeeping forces was overcome by Cyrus Vance in November 1991, when Miloševi realized that these forces could help him consolidate Serb territorial gains in Croatia. On December 12, 1991 the UN announced that it would send more than 10,000 troops to Yugoslavia. The UN effort to send peacekeepers, spurred by Western permanent members of the Security Council, was intended in part to inhibit Germany's unilateral drive to recognize Croatia. The British and French argued that premature recognition would harm the reconciliation process on which a successful peacekeeping mission depended (Goodby, 1995: 168-171). This effort failed, however, to persuade Germany to change its mind.

Despite limited successes, the UN Protection Force (UNPROFOR) failed in a number of respects. While it occasionally helped to stabilize the local situation and to assure the delivery of the humanitarian aid, it was unable to contribute to a long-term solution. One reason for this was that its underlying political arrangements did not address border and minority rights issues which, especially in Croatia, amounted to a political and military time bomb. Following major Croatian rearmament, the bomb eventually exploded in 1995 when Croatian forces re-established control over Western Slavonia and Krajina. UNPROFOR's mandate was also unrealistic in that it contained ambitious political goals without giving the forces adequate instruments and political support to pursue them.13

It has been asserted that the war in Croatia would not have spread into Bosnia in early 1992 if Slovenia and Croatia had not been recognized so soon. According to this reasoning, peacekeeping could have frozen the conflict in Croatia, and would not have pushed the Muslim-led Bosnian government towards a declaration of independence in the way that the premature recognition of Croatia made politically necessary. Without such a declaration, the Bosnian Serbs could have retained their ties with Belgrade. With the independence of Croatia and Bosnia, however, war in Bosnia could have been avoided by large-scale UN intervention - effectively converting Bosnia into a protectorate (Glenny, 1995: 61-62). The main weakness of this argument is that, without international recognition of Croatia, it would have been far more difficult to secure all-party agreement on the deployment of UN troops in the region.

Conclusion: Lessons of Preventive Action

The failure of preventive diplomacy in Yugoslavia can be viewed on two different levels. On the one hand, whatever measures third parties might have taken, republican leaders in Yugoslavia were deeply reluctant to compromise their political designs. Silber and Little have characterized the negotiations between them in the winter and spring 1991 as "conversations of the deaf" (Silber & Little, 1996: 147-153). As opportunities for local solutions to the crisis withered away, third-party intervention became increasingly necessary. As we have seen, there was no lack of effort. But there was a failure to produce tangible results. While third parties were not the original culprits in the Yugoslav crisis, their mismanaged policies in 1991-92 may well have contributed to the escalation in the conflict.

It is quite clear that deterrence, whether political or military, failed in Yugoslavia. The republican leaders were not overly worried about negative responses from either outside governments or international organizations. A main reason for the failure of deterrence was that the Croatian and Serbian governments were aware that they were unlikely to incur determined Western punishment for their aggressive role in the conflict. When political deterrence was tried by the pre-emptive recognition of the Slovenian and Croatian independence, the policy was so misdirected that it actually encouraged Croatia and Serbia to escalate their use of force in Bosnia.

Considering the effectiveness of various instruments of prevention, it is fair to say that in multilateral political mediation the EC made serious efforts to avert the escalation of crisis. It has been suggested that the Community's failure was, in part at least, due to Miloševic's perception that it was not neutral in the conflict.14 It is doubtful whether a more positive attitude by the Community towards Serbia would have helped much, however, short of its acquiescing with Miloševi's master plan of a Greater Serbia.

Active positive strategies were attempted early in the crisis with promises of economic aid and closer cooperation with the EC. But these failed to reorient the course of developments. Active negative strategies therefore remained as the only real alternative. The Yugoslav experience in 1991 provides counter-evidence to the now-popular idea that economic incentives are a more potent means to shape outcome of political processes than negative sanctions. At a minimum, the reasons for the failure of incentives must be specified in order to better understand how they may work more effectively in future.

To be more effective, the diplomatic preventive strategy should have been backed up by more coercive measures. It is unlikely that economic sanctions, even if more consistently and effectively implemented, would have made enough of a difference to the outcome. Thus, the international community should have been more willing to use military force early on, especially to protect Bosnia. However, major Western powers - especially the United States - were unwilling to provide ground forces and thus become politically involved. It is also conceivable that Russia would have vetoed a Security Council Resolution allowing Western-led military enforcement in Yugoslavia.

Michael Lund is obviously right in observing that "multifaceted action", i.e. the employment of several diverse instruments, provides a basis for an effective strategy of conflict prevention (Lund, 1996: 85-86). However, it is not certain that the use of more diverse and well-coordinated preventive instruments would really have enhanced their impact on the conflict dynamics. In fact, it is likely that in such intractable crises like Yugoslavia, only more coercive forms of diplomacy would have made a real difference. This is especially the case if violence has already broken out and the task is to prevent its vertical or horizontal escalation.

In addition to their domestic reluctance to become involved in the crisis, disagreements between the major powers repeatedly undermined more coordinated and coercive conflict prevention. In the EC, there were major disagreements between Germany, on the one hand, and Britain and France supported by the United States, on the other. Moreover, in contrast to the Gulf War, the vital interests of the West were not at stake and the mountainous terrain of the Balkans presented Western governments with more difficult military challenges than had been the case in the desert sands of Iraq and Kuwait (Joffe, 1992-1993: 32-36).

In sum, the established conclusion that mediators must closely coordinate the means and sequences of their moves, and thereby better control and resolve a conflict, is borne out in the Yugoslav case. However, even though EC policy in Yugoslavia has been correctly judged as "far more coordinated than in many other such situations", the divergent interests of its member states and the complex nature of the contemporary international system tended to undermine consistent and effective intervention by the EC (Webb et al, 1995: 177-178, 183-184; Crawford, 1996).

Western disagreements resulted in dual policies of prevention; Germany suppported Croatia, while the rest were unwilling to seriously challenge Serbia. This was, perhaps, the worst of all possible combinations. Political and economic support from Germany, and ultimately from the EC, led Croatia to believe that it could carry out reckless policies in Bosnia with impunity. Equally, the punishment of Serbia by EC and UN economic sanctions alone, without any real risk of military enforcement, made it possible for Miloševi to support - within the limits of Serbia's scarce resources - the military aggression of the Bosnian Serbs.

The ultimate lesson of Yugoslavia for an effective preventive strategy is quite simple. It requires good coordination of both the instruments of prevention and the policies of major powers which govern them. Prevention easily fails without such coordination, especially in complex and intractable conflicts where nobody can control the indirect and unintended consequences of decentralized political actions. As Robert Jervis has aptly pointed out: "Outcomes do not correspond to intentions because effects are often indirect, in two senses of the word. First, outcomes are often produced through a chain of actions and reactions. Second, the result of trying to move directly toward a goal may be movement in the opposite direction" (Jervis, 1993: 31).

In such situations, a determined and skilful political leader can play third parties against each other and strike advantageous deals with other parties to the conflict under the cover of an international intervention. This is especially possible if the intervention fails to define the basic parameters of the conflict and develop a long-term view of how these might subsequently be redefined. If third parties believe in a major one-off solution and lack credible contingency planning, preventive action is unlikely to succeed. The readiness to use limited, but effective, military force in the interest of collective security is a key element of such a plan.

A major problem in the Yugoslav crisis was that both political mediation and military intervention by the third parties took the republican and provincial borders for granted. There was a corresponding failure to consider other more creative and flexible alternatives until it was too late. The sovereignty model became a straitjacket which exacerbated, rather than ameliorated, the conflict. Ultimately, any successful strategy for preventing violent conflict requires foresight and the capacity for collective action.

Notes

* The author wishes to thank Dr. Kevin R. Warns for his editorial assistance.

1. See Boutros-Ghali, (1992) and Evans, (1994). Recent UN preventive activities are reported in detail in Boutros-Ghali (1995). Evans' blueprint for preventive diplomacy is supported by several case studies in Clements & Ward, (1994).

2. With its practical political experience, the Carnegie Commission on Preventing Deadly Conflict is developing an ambitious agenda involving a comprehensive understanding of the role of democracy, development, religions and other social forces, as well as different instruments such as negative and positive sanctions, in preventing violence. At the same time, The Center for Preventive Action at the Council on Foreign Relations, International Crisis Group, headed by former U.S. Senator George Mitchell, and London-based International Alert rely more on fact-finding missions and other forms of third-party intervention to crisis areas.

3. These conceptual issues are discussed in greater detail in my chapter (1996a). Doom and Vlassenroot (1995: 18-21) distinguish between pre-escalation stage of the conflict, open conflict and outcomes.

4. Democracy is explored from the perspective of conflict prevention by Diamong (1995).

5. Perry (1996) - the emphasis is in the original.

6. In the Cold War context these methods of preventive influence have been discussed in some detail by Lebow & Gross Stein (1994: 291-323).

7. The concept and forms of coercive diplomacy and its use in selected cases during the Cold War is further developed in George et al (1971).

8. See Burg (1995: 238). The year 1989 is regarded as the turning point by Goodby (1995: 160-62).

9. This conclusion is supported by Beschloss & Talbott (1993). This story leaves the impression that Bush and Gorbachev barely mentioned Yugoslavia in their discussions.

10. On the CSCE actions, see Remacle (1994: 213-28).

11. This interpretation has been offered by Glenny (1995: 56-65).

12. Quoted in Silber & Little (1996: 199-200).

13. The best analysis of the UN failure in Yugoslavia is still Higgins (1993). See also Eknes (1995).

14. Woodward (1995a: 178-79). The neutrality of the EC as a mediator is considered in greater detail by Wiberg (1995).

References

Andersson, Stephanie. 1995. "EU, NATO and the CSCE Responses to the Yugoslav Crisis: Testing Europe's New Security Architecture". European Security, Vol. 4, No. 2, pp. 328-353.

Baun, Michael N. 1995-96. "The Maastricht Treaty as High Politics: Germany, France, and the European Integration". Political Science Quarterly, Vol. 110, No. 4, pp. 605-627.

Beschloss, Michael R. & Talbott, Strobe. 1993. At the Highest Levels. The Inside Story of the End of the Cold War. Boston: Little, Brown & Co.

Boutros Boutros-Ghali. 1992. An Agenda for Peace. Preventive Diplomacy, Peacemaking and Peacekeeping. New York: The United Nations.

Boutros Boutros-Ghali. 1995. Confronting New Challenges. Annual Report on the Work of the Organization 1995. New York: The United Nations, Part IV.

Burg, Steven L. 1995. "The International Community and the Yugoslav Crisis", in Esman, Milton J. & Telhami, Shibley, eds. International Organizations and Ethnic Conflict. Ithaca, N.Y.: Cornell University Press, pp. 235-271.

Chayes, Abram & Chayes, Antonia Handler, eds. 1995. Preventing Conflict in the Post Communist World. Mobilizing International and Regional Organizations. Washington, D.C.:The Brookings Institution.

Clements, Kevin & Ward, Robin, eds. 1994. Building International Community. Cooperating for Peace Case Studies. St. Leonards, Australia: Allen & Unwin.

Commission on Global Governance. 1995, Our Global Neighborhood. The Report of the Commission on Global Governance. Oxford: Oxford University Press.

Crawford, Beverly. 1996. "Explaining Defection from International Cooperation. Germany's Unilateral Recognition of Croatia". World Politics, Vol. 48, No. 4, pp. 482-521.

Diamond, Larry. 1995. Promoting Democracy in the 1990s. Actors and Instruments, Issues and Imperatives. New York: Carnegie Corporation of New York.

Doom, Ruddy & Vlassenroot, Koen. 1995. Early Warning and Conflict Prevention. Minerva's Wisdom? Brussels: Vlaamse Interuniversitaire Raad.

Eknes, Åke. 1995. "The United Nations' Predicament in the Former Yugoslavia", in Weiss, Thomas G., eds. The United Nations and Civil Wars. Boulder, CO: Lynne Rienner, pp. 109-26.

Eland, Iva. 1995. "Economic Sanctions as Tools of Foreign Policy", in Cortright, David & Lopez, George A., eds. Economic Sanctions. Panacea or Peacebuilding in a Post-Cold War World? Boulder: Westview, pp. 29-42.

Evans, Gareth. 1994. Cooperating for Peace. The Global Agenda for the 1990s and Beyond. St. Leonards, Australia: Allen & Unwin.

Genscher, Hans-Dietrich. 1995. Erinnerungen. Berlin: Siedler Verlag.

George, Alexander L., Hall, David K., & Simon, William E. 1971. The Limits of Coercive Diplomacy: Laos, Cuba, Vietnam. Boston: Little, Brown & Co.

George, Alexander L. 1991. Forceful Persuasion. Coercive Diplomacy as an Alternative to War. Washington, D.C.: The United States Institute of Peace Press.

George, Alexander L. 1993. Bridging the Gap. Theory and Practice in Foreign Policy. Washington, D.C. The United States Institute of Peace Press.

Glenny, Misha. 1995. "Yugoslavia: The Great Fall". New York Review of Books, March 23, pp. 56-65.

Goodby, James E. 1995. "Conflict in Europe: The Case of Yugoslavia", in Goodby, James E., eds. Regional Conflicts. The Challenge to US-Russian Co-operation. Oxford: Oxford University Press/SIPRI, pp. 157-187.

Gross Stein, Janice. 1992. "Deterrence and Compellence in the Gulf 1990-91. A Failed or Impossible Task". International Security, Vol. 17, No. 2, pp. 147-79.

Haas, Richard N. 1994. "Military Force. A User's Guide". Foreign Policy, No. 96: 21-37.

Higgins, Rosalyn. 1993. "The New United Nations and Former Yugoslavia". International Affairs, Vol. 69, No. 3, pp. 165-83.

Jervis, Robert. 1993. "Systems and Interaction Effects", in Snyder, Jack & Jervis, Robert, eds. Coping with Complexity in the International System. Boulder: Westview Press, pp. 25-46.

Joffe, Josef. 1992-93. "The New Europe: Yesterday's Ghosts". Foreign Affairs, Vol. 72, No.1, pp. 29-43.

Lebow, Richard and Gross Stein, Janice. 1994. We All Lost the Cold War. Princeton, N.J.: Princeton University Press.

Lucarelli, Sonia. 1995. "The European Response to the Yugoslav Crisis. Story of a Two-Level Constraint". EUI Working Paper 37/95. Florence: The European University Institute.

Lukic, Reneo & Lynch, Allen. 1996. Europe from Balkans to the Urals. The Disintegration of Yugoslavia and the Soviet Union. London: SIPRI/Oxford University Press.

Lund, Michael S. 1996. Preventing Violent Conflicts. A Strategy for Preventive Action. Washington, D.C.: United States Institute of Peace Press.

Maull, Hans W. 1995-96. "Germany in the Yugoslav Crisis". Survival, Vol. 37, No. 4, pp. 99 130.

Perry, William J. 1996. "Preventive Defense". Trilateral Note, May 31.

Princen, Thomas. 1992, Intermediaries in International Conflict. Princeton, N.J.: Princeton University Press.

Rakisits, Claude. 1992. "The Gulf Crisis: Failure of Preventive Diplomacy", in Clements, Kevin & Ward, Robin, eds. Building International Community. Cooperating for Peace Case Studies. St. Leonards, Australia: Allen & Unwin, pp. 63-72.

Remacle, Eric. 1994. "Conflict Prevention and Resolution in the Former Yugoslavia: The Action and Inaction of the CSCE", in Calleja, James, Wiberg, Håkan & Busutti, Salvino, eds. The Search for Peace in the Mediterranean Region. Malta: Mireva Publications, pp. 213- 28.

Rothgeb, John M. 1993. Defining Power. Influence and Force in the Contemporary International System. New York: St. Martin's Press.

Schelling, Thomas C. 1980. The Strategy of Conflict. Cambridge, MA: Harvard University Press.

Silber, Laura & Little, Allan. 1996. Yugoslavia. Death of a Nation. New York: TV Books. Smoke, Richard. 1977. War. Controlling Escalation. Cambridge, MA: Harvard University Press.

Stedman, Stephen John. 1995. "Alchemy for a New World Order: Overselling 'Preventive Diplomacy." Foreign Affairs, Vol. 74, No. 3, pp. 14-20.

Väyrynen, Raimo. 1996a. "Keeping Closed the Gates of War: Maximizing the Peace Potential of Preventive Action", in Leatherman, Janie, DeMars, William, Gaffney, Patrick & Väyrynen, Raimo, Preventive and Inventive Action in Intrastate Crises. Joan B. Kroc Institute for International Peace Studies. University of Notre Dame (manuscript).

Väyrynen, Raimo. 1996b. "The Failure of Preventive Action in Yugoslavia". International Peacekeeping, Vol. 3, No. 4 (forthcoming).

Wallander, Celeste A. & Prokop, Jane E. 1993. "Soviet Security Strategies toward Europe: After the Wall, With Their Backs Up Against It", in Keohane, Robert O., Nye, Joseph S. & Hoffmann, Stanley, eds. After the Cold War. International Institutions and State Strategies in Europe, 1989-1991. Cambridge, MA: Harvard University Press, pp. 63-103.

Webb, Keith, Koutrakou, Vassiliki & Waters, Mike. 1995. "The Yugoslav Conflict, European Mediation, and the Contingency Model: A Critical Perspective", in Bercovitch, Jacob, eds. Resolving International Conflicts. The Theory and Practice of Mediation. Boulder: Lynne Rienner, pp. 171-189.

Wiberg, Håkan. 1995. "Making Peace in the Former Yugoslavia. Problems and Lessons", in Calleja, Wiberg & Busutti op. cit. 1995, pp. 229-53.

Woodward, Susan L. 1995a. The Balkan Tragedy. Chaos and Dissolution after the Cold War. Washington, D.C.: The Brookings Institution.

Woodward, Susan L. 1995b. "The Use of Sanctions in Former Yugoslavia: Misunderstanding Political Realities", in Cortright & Lopez op.cit. 1995, pp. 171-176.

Zucconi, Mario. 1996. "The European Union in the Former Yugoslavia", in Chayes, Abram & Chayes, Antonia Handler, eds. Preventing Conflict in the Post-Communist World. Mobilizing International and Regional Organizations. Washington, D.C.: The Brookings Institution, pp. 237-278.