WHAT KINDS OF PREVENTION ARE EFFECTIVE? GETTING AHEAD OF THE CURVE

Much extant research looks at failure: coun­tries that faced potential violent conflict, and where no preventive effort was tried or opportunities were missed (e.g., Zartman, 2005).

Basic ingredients

The first wave of this research looked mainly at preventive diplomacy (direct prevention), and thus relatively late stages of confrontation (e.g., Miall, 1992; Manuera, 1994; Lund, 1996). It suggests convergence around ele­ments that appear to be associated with effective avoidance of violence:³⁸

1. Act at an early stage (Miall, 1992:198)., that is before a triggering event (Wallensteen, 1998: 15), “early, early, early" (Jentleson, 2000: 337).

2. Be swift and decisive, not equivocal and vacillating (Wallensteen; Jentleson, 2000: 343; Hamburg, 2002:146; Harff, 2006: 6).

3. Use talented, influential international diplomats who command local respect (Jentleson, 2000: 336; Miall, 1992:193).

4. Convince the parties that the third parties are committed to a peaceful and fair solution, and oppose the use of force by any side (Jentleson, 2000: 341).

5. Use a combination of responses, such as carrots and sticks, implemented more or less coherently (Hamburg, 2002: 146-47; Wallensteen, 1998: 15; Jentleson, 2000: 336; Leatherman, 1999: 182-94; Zartman, 2005:14; Byman, 2002:217).

6. Provide support and reinforcement to moder­ate leaders and coalitions that display non­violent and cooperative behavior Zartman, 2000: 310.

7. Build local networks that address the various drivers of the conflict, but avoid obvious favoritism and imbalances (Wallensteen: 15; Jentleson, 2000: 336; Hamburg, 2002: 147; Leatherman, 1999: 199).

8. If necessary to deter actors from using violence, use credible threat of the use of force or other penalties such as targeted sanctions (Jentleson, 2000; Zartman, 2005: 202).

9. Neutralize potential external supporters of one side or the other, such as neighboring countries with kin groups to those in a conflict (Miall, 2000; Hamburg, 2002:147).

10. Work through legitimate local institutions to build them up (Wallensteen: 15).

11. Involve regional organizations or regional powers, but don't necessarily act entirely through them (Wallensteen: 15) Jentleson: 339; Miall: 198).

12. Involve major powers that can provide credible guarantees, but use UN or other multi-lateral channels to ensure legitimacy (Jentleson: 337; Wallensteen: 15; Hamburg, 2002: 147; Leatherman et al. 1999: 216; Zartman, 2005: 13).

The studies also find that certain local and regional conditions significantly enhance the chances of success (e.g., Miall):³⁹

1. Domestic leaders who are relatively secure and feel a self-interest in stability, and thus are open to third parties facilitating or mediating emerging disputes.

2. Majorfactionsthatshowsome mutual ability to manage societal disputes and carry out public policies that benefit all communities.

3. Accommodative policies and procedures such as voting systems and opportunities for political participation that blunt the impact of grievances felt by one side or the other.

4. Relations between major political groups that have been peaceful in the recent past.

5. One side is not much more powerful than another.

6. Weak group solidarity or political mobilization within one of the protagonists, such that they cannot mobilize beyond a certain level.

7. The country is small and relatively dependent on the international community economically, politically, and militarily.

Towarda theory of prevention: timing and sequencing

While very useful, these findings do not reveal the utility of particular instruments at different stages. It is widely accepted that different interventions are needed at different moments (e.g., Lund, 1996: 191; Rothchild CAII, 1996: 44). As indicated, it is also believed that several kinds of instruments are needed. But such a multi-pronged strategy cannot mean everyone doing everything in every stage and place. More is not necessarily better. Consequently, the leading current research question being urged for the field is which mixes of differing instruments are most effective in which stages of conflict and contexts, other things being equal (e.g., Miall, 1992; Nicolaides, 1996; Harff, 2005).⁴⁰ Case­studies and ‘large n’ quantitative studies have begun to mine recent experience (e.g., Rubin, 1998, 2004; Nicolaides; Rowsbotham and Miall; Leatherman et al.,)⁴¹ to get at this issue. Differing levels of analysis, typologies, and cases have impeded the task of cumu­lating and verifying findings, and many are partly deductive rather than empirical (e.g., Lund, 1997; Leatherman; Kriesberg, 2003: Rothchild, 2003: 45). Nevertheless, gather­ing up what extant findings and grounded reasoning suggest so far can provide useful heuristic guidelines for policymakers about which combinations of instruments to apply to the early stages of conflict.⁴²

To explore the available evidence, we examine below what research suggests are most useful of the basic types of prevention at each of three distinguishable early phases of conflicts. These phases lie in the realm of unstable peace between a peaceful equilib­rium where conflicts are managed predictably, on the one hand, and tensions are beginning to escalate into confrontation, significant violence or organized armed conflict, on the other (cf. e.g., Mitchell, 1981; 2006; Lund, 1996; Lund, 1997; Kriesberg, 2003; Ramsbotham and Miall, 2005).⁴³ To frame the following discussion, we pose here a familiar assumption that “soft” measures must be followed by “hard” ones, the more a conflict escalates - e.g., diplomacy must precede the use of force.

Latent conflicts

These arise when exogenous or endogenous changes are generating underlying but unac­knowledged strains among societal groups but they have yet to mobilize to express their interests.⁴⁴

A priori instruments: structural and direct

As described earlier, one prominent a priori instrument involves global and regional organizations promulgating standards or reg­ulations backed by incentives in order to encourage present or prospective member states to respect human rights, adopt demo­cratic procedures, settle disputes peacefully with their own minorities and neighboring states, or submit to restrictions on terms of trade (e.g., Lund, OECD-DAC, 1998; Jentleson, 2000: 338; Hamburg, 2002: 147; Cortright, 269-72).⁴⁵ The evident effective­ness of this instrument in reducing potential causes of conflict seems to derive from the conditional incentives offered to leaders who have already subscribed to particular norms, at least nominally, and are already in power before particular conflicts ensue, thus avoiding the difficulties of intervening where parties have already violated the norms and become entrenched in opposed positions on specific disputes. When agreeing to them, a regime's future stakes are not immediately apparent, compliance can be voluntary, there is time to adjust a country's policies, and individual actors cannot argue they are being singled out.

Ad hoc structural instruments

Vigorous structural measures can help specific governments to alleviate underlying socio­economic sources of conflicts or institu­tional and policy deficits that keep countries from addressing those problems meaning­fully and peacefully. When in the 1980s, international lending institutions began to pressure developing countries to privatize para-statals, reduce public spending, remove price subsidies, stabilize monetary systems, and liberalize trade regulations (Muscat, 2002: 196), the rationale was not solely economic productivity and growth, but polit­ical stability, an implicit theory of peace. In fact, considerable large ‘n’ research suggests that economic liberalization such as free trade policies are highly correlated with lower levels of poverty, and that development correlates with lower levels of conflict (e.g., Hegre et al., 2002; Goldstone et al., 2003). Failing to enact reforms, on the other hand, is likely to deepen poverty and inequities that increase the chances for upheaval.

However, critics argue that structural adjustment measures can increase political instability and thus risk of conflict, especially in the poorest countries by reducing income and increasing competition among prospec­tive losers and gainers during de-statalization. In this view, globalization increases vulner­ability to complex humanitarian emergen­cies by liberalizing trade, increasing capital mobility, raising debt, lowering commodity export prices, and reducing foreign direct investment (e.g., Rapley, 2009).

This debate revolves in part around differ­ing time frames. To derive the ingredients of peace from ahistorical econometric methods that pinpoint the highest correlations among indicators in large numbers of countries ex post facto is not to understand how these correlations came into being over time and the ways that the variables actually behaved and interacted within particular countries.⁴⁷ Though austerity measures may provoke violent protests in the short run, the evidence of political instability is mixed and context­specific (Muscat, 1995). Such adjustment policies may not create fundamental threats to regimes (Bienen, 1986). In fact, early policies toward natural resources, trade access, diver­sification, corruption, price shocks, and ethnic quotas can boost growth (Collier, 125-40).⁴⁸ Whether such policies mitigate or worsen conflict also depends on how these interna­tional and domestic policies are designed, introduced, and implemented.⁴⁹ Social safety- net programs can be used to compensate groups that are especially hard-hit by short­term effects of economic austerity.⁵⁰ In any case, normal policies of international lending institutions applied automatically without tailoring them to each country context may be especially destabilizing in the poorest and least capable states.⁵¹ In short, economic reform may have better chances of success at this stage, than when politics are more polarized, but they need to be conflict­sensitive and accompanied by compensatory measures.

As against such conditional aid,⁵² donors also provide outright aid such as in health and education to alleviate social needs and thus encourage economic activity. Such support programs are believed to have stabilizing effects because they can create new markets and increase social interaction (Cortright, 1997; Collier, 134). A drawback is that such assistance is implemented through divisible projects and programs, so benefit allocations may reflect the differential access of a society’s ethnic groups, causing ‘horizon­tal inequities’ (Stewart, ), especially where prebendal or patronage mechanisms distribute resources and life chances as is common in Africa. When the competitive pressures of democratization arise, ruling parties have especially strong incentives to use social and economic programs to win and reward supporters. Thus, conflict-blind aid intended to alleviate poverty may actually privilege certain and identity groups and intensify inter­group rivalries (Graham, 1994).⁵³ Donors often find that even well-intentioned support may visibly affect the relative position of politically significant groups in a society and thus exacerbate the sources of conflict (Collier, 138). Where there are politicized ethnic divisions, aid programs may contribute more to conflict than do macro-economic reforms because they are more or less ‘lumpy.’⁵⁴ Implementing programs through multi-group and locally-run mechanisms may help to avoid obvious partiality and bridge such cleavages (e.g., Anderson).⁵⁵

Both economic reform and outright aid are less likely to provoke conflict if developing societies have institutions that manage the social strains and inequalities that global­ization can cause (Rodrik, 1997). As many donors concluded that structural adjustment could not work unless bolstered by effective governance (Stokke, 1995: 26), the latter became another entry point for structural con­flict prevention. International agencies now widely subscribe to the view that democracy­building is an effective way to achieve domes­tic stability.⁵⁶ Again, the evidence arises from strong cross-sectional statistical associations in a large number of countries between democracy and peace between and within nations (e.g., Russett, 1993). At the stage of latent conflict, such support for building institutions that can regulate emerging social conflicts is promising (Nicolaides: 53). Some countries like Indonesia though ethnically fragmented have taken genuine steps toward popular democracy and maintained relative stability.

However, views that any steps toward more democracy are gains for conflict reduction (e.g., Diamond, 1996:40-8) do not recognize that democratization also risks destabiliza­tion. Studies of actual dynamics of change in particular countries find that the risk of conflict often rises during periods when authoritarian systems are shifting to more pluralistic structures (e.g., Mansfield and Snyder, 1995a,b).⁵⁷ Alternatively, transition­ing polities may remain ‘partial’ or ‘illiberal’ democracies (Ottaway, 2003; Zakaria, 1997) in which the regime’s hold on power is not challenged, political and civil rights are abridged, and representation occurs through informal power-sharing within cliques. If such autocratic or oligarchic regimes (‘anocracies’) continue to resist meaningful democratic reform, they could simply stagnate econom­ically as well as politically, inviting state breakdown and violent conflict.

At the same time, it is unclear whether such regimes necessarily lead to stagnation and violent conflict or can evolve grad­ually toward more openness and stability. Informal power-sharing among less than fully accountable political leaders, though falling short of formal democracy in a Western sense, does not lead inevitably to conflict.⁵⁸ In fact, intra-elite co-optative bargains, though less than ideal by Western standards, may be a pre-requisite for political stability and thus eventual development (e.g., Rothchild, 2004, Byman). So once again, the likelihood of conflict may be determined more by whether governments make accommodative adjustments, such as allowing for some political activity (cf. Cramer and Weeks, 2002: 41f).⁵⁹ Positive discrimination programs to increase access of minorities to government jobs and services can co-opt group resentments (Rothchild, 2004: 47). These diverging scenarios make the current national politics in authoritarian countries such as Saudi Arabia, Uzbekistan, and Egypt, and more pluralistic but weak systems like Kyrgyzstan and Azerbaijan, crucial focuses for early warning and conflict prevention.⁶⁰

Appropriately, especially since 9/11, ana­lysts have looked increasingly to ‘supply-side’ programs that support institutions of the state to make governments more effective from the inside. Several analysts argue that before rep­resentative democracies can function effec­tively, basic institutions of the state need to operate effectively.⁶¹ Fragile and failed states need to have effective ministries, local authorities, and judiciaries delivering health, education, roads, sanitation, and justice.⁶² State strengthening includes professionalizing a country’s security forces, both to restrain them from abusing its citizens and enable them to provide security.⁶³ National laws also need to provide guarantees such as property rights (Kapstein, 2004), enforce policies governing the economy, establish regulatory agencies such as for banking and trade, and respect civil and political rights and criminal laws through courts,⁶⁴ including protections for minorities and other limits on arbitrary power: ‘constitutional liberalism' (Zakaria, 2003).⁶⁵

On a broader plane, much research has weighed the utility for preventing ethnic conflicts of constitutional engineering that allocates political authority through differing options: unitary systems versus federalism, autonomy or partition; presidential versus parliamentary systems, and proportional ver­sus plural electoral rules (see e.g., Horowitz; Wimmer, 2004). Federalism is often presented as a possible means of conflict resolution or prevention, for devolving policymaking can shield minorities and be more responsive to regional or local interests. But decen­tralization has both calmed and divided societies (Siegle and O'Mahony, 2007). Proportional representation and winner-take- all voting helped in South Africa and not in Northern Ireland. Again, how such differing arrangements affect the risk of conflict in a given country depends on other particular factors, such as the political relationship between contending identity groups and the politics of change.

In sum, all such economic, political and constitutional structural changes envision ultimate states of affairs that, if attained, would undoubtedly reduce conflicts signifi­cantly. But the challenge is getting to these endpoints without destructive conflict. In the short run, reforms such as structural adjustment and majoritarian elections are not always feasible, and can be counter­productive if applied too quickly or with insufficient attention to a country's balance of power, political economy, and potential for backlash and deeper polarization. Liber­alizations that fragment power have to be balanced by stabilization that consolidates it (cf. Paris, 2001), such as state and societal institutions with authority to reconcile competing interests and force compromises. Many ideal liberal-internationalist solutions set aside the difficulties and pitfalls of getting reforms adopted and do not calculate the risk of destabilization in view of the capacities of differing societies for peaceful change.⁶⁶ Merely prescribing ultimate ide­als is as useful as a doctor advising an obese patient with heart trouble to ‘lose weight.'⁶⁷

Ad hoc direct instruments

Structural policies do not necessarily engage the specific stakeholders in emerging national conflicts, although they require consent or at least toleration by host governments where they are applied. Critical to their adoption and implementation are the processes and channels through which governing elites make decisions about them, steps that affect the prospects for social conflict. This reality thus calls for direct forms of preventive engagement even at this stage of latent conflict.⁶⁸ But despite the frequent obeisance expressed to the idea of engendering ‘local ownership,' structural programs often treat the leaders in a country not as active agents of change but automatons who respond to incentives and disincentives in some Pavlovian stimulus-response international experiment.

Obviously, direct prevention is premature if no conscious sense of a serious prospective harm or opportunity is present (Berkovitch; Nicolaides 1996: 52). Where societies see no serious problem that needs fixing, it is hard for third-party would-be preventors to explain why they are needed. Pointing to a conflict of interests might actually destabilize the situation (Kemp: 50ff).⁶⁹ Or, if no aggrieved parties have stepped forward, it is unclear whom one can talk to. But once underlying problems are beginning to surface as con­tentious issues, direct engagement fostered by trusted third parties is best carried out within existing institutions and ruling processes, thus giving standing regimes the chance to respond in ways that do not immediately threaten their status while allowing them to address emerging problems. State elites acting early on to deal with structural conditions can be effective prevention (Rothchild, 2003: 46). Whether or not governments have accepted inter-national standards through agreements they have signed, they may take umbrage at criticism and dismiss outside pressure. But fact-finding missions from institutions such as the UN can overcome resistance, especially

if complemented with direct support that addresses the deficiencies (Rothchild, 2003: 46-7). As it is better to foster compliance than rely only on ex poste condemnations of deviations (Nicolaides, 1996: 54), multi­lateral organizations have also moved from simply promulgating and pressing standards on a government to hands-on assistance. The Office of the UN High Commissioner on Human Rights, for example, has shifted from simply monitoring human rights to help­ing governments comply, through creating national institutions that build human rights capacity. A related approach is the Lome consultations the EU holds with governments in Africa, the Caribbean, and the Pacific for incrementally establishing democratic institu­tions, thus allowing for flexibility regarding which countries are expected to meet which benchmarks by when.

Manifest limited conflict

The stakes of conflict increase when wider forces of change elicit awareness of conflict­ing interests and energize affected groups, issues come into the open, and potentially diverging positions are decided upon and voiced (Miall, 2007). Accepted forms of protests may be underway as well as irregular acts, including violence. The aim is both to prevent confrontations that escalate, hard­ening of positions and polarization, rising fears, and mutual defensive measures that create security dilemmas and to find bases for cooperation. For some, this is the most strategic moment for prevention, as the tasks of earlier and more basic structural prevention are seen as too demanding and complex (Ottaway and Mair, 2004). Some rawness of sores of discontent may be needed to expect positive change to occur (cf. Stedman, 1995). Structural measures continue to be useful - but now, less for alleviating the underlying sources of the conflict than as ‘purchase’ (Rothchild, 2003), to ‘sweeten’ an agreement, that purveyors of direct prevention can use tactically.

Direct measures thus become more essential. Opposed groups often have little inclination to initiate mutual engagement, at least until they fail to achieve their objectives unilaterally through first trying coercive or violent means. Still, some may seek outside help at this early stage more often than may be realized (Nicolaides, 1996: 49), as when the Barre regime was under challenge by various clans. Direct methods through which third parties can intervene peacefully include the classic array of official and non-official interactive methods. All these are intended to get parties in closer contact and communication for more accurate information about mutual interests and needs, dispel ignorance and fear, and expose them to more options, possibly leading to agreements (e.g., Rothchild, 2003: 46; Zartman and Rasmussen, 1997).⁷⁰

One direct approach uses non-binding interactions such as various types of conflict transformation workshops that precede, fol­low or operate under or alongside official ‘track one’ diplomacy or political processes (Fisher, 2005; cf. Ropers, 2005). Rather than take up substantive issues to seek settlements through adversarial, judgmental approaches, these gentler methods or ‘soft mediation’ (Nicolaides, 1996: 51) create a non-threatening milieu to simply facilitate inter-party communication, thus expecting to elicit more committed participation and pave the way to locally decided and ‘owned’ accommodations (e.g., Zartman and Rasmussen). One study found that extensive mutual communication rather than hard bar­gaining has been more effective (Bercovitch, 1998: 243). In 2003, for example, UNDP and Guyanese leaders agreed to a whole series of governmental and civil society dialogues that resulted in the country’s first ever non-violent elections in 2006. Success may depend greatly on whether they are spearheaded by prominent outsiders who command respect (Lund and Myers, 2007). Yet, even if a small society and government is immersed in workshops, if improved relationships are not translated into legal and policy changes that institutionalize and uphold agreed rules even on stormy days, the usual political styles can return (Lund and Myers, 2006). It is difficult to instill new habits unless they are embedded in locally run institutions (Nicolaides, 51). A point is reached when the question is whether a body politic adopts such habits on its own without third-party therapy. Such non-formal methods are not intended as alternatives to tougher approaches, but complementary (Fisher, in Zartman and Rasmussen, 1997: 241).

A innovative hybrid of a priori, ad hoc, structural and direct engagement that lies between non-formal facilitation and formal mediation is the work of the OSCE High Com­missioner on National Minorities (HCNM), an office mandated to become proactively engaged in ethnic disputes arising in the 1990s. The first able incumbent and his successors have made innumerable visits to Eastern Europe and newly independent states to meet with leaders and minority groups. They facilitate dialogues, recommend policy remedies to chief executives and parliaments, and show how OSCE norms may apply, including drafting model legislation. Only very rarely have they publicly pressured the parties, but crucial to the success that many analysts judge this innovation has often had in reducing divisive tensions and eliciting accommodation is the eventual reward for good behavior of economic aid and member­ship in the EU, NATO, and other Western bodies (e.g., Hopmann, Mychajlyszyn).

Still, leaders in conflict-vulnerable soci­eties and weak states are often disinclined to compromise and/or they affirm positions and agreements they cannot enforce (Nicolaides, 1996: 52). If their recalcitrance breaks off communication or thwarts opportunities for joint problem-solving, third parties may need to get more directive by engaging parties in ‘muscular mediation' or formal negotiations with teeth (e.g., Jakobsen, 1996: 24), such as proferred aid or ‘coercive diplomacy,' such as threats to cut off aid (Rothchild, 2002: 48f), impose economic sanctions, or use force (e.g., George, 1994: 199).

Military measures can also be used for direct prevention, but not yet in the form of a threat or actual use of force. The usual foreign policy debate over ‘force versus diplomacy' tends to pertain to high levels of confrontation. But before that stage, the overlooked but promising instrument of preventive deployment (Nicolaides, 44f) can act as a deterrent by inter-positioning forces even before any hostile actions have occurred. The only clear example has been UNPREDEP, the UN force that posted 1100 troops along Macedonia's border with Albania and Serbia from 1992 until 1999. Though its firepower could not withstand a Yugoslav army attack, UNPREDEP created a tripwire that would likely trigger more forceful responses. Its removal in 1999 was followed two years later by an insurgency that originated in border areas UNPREDEP had once patrolled (Lund, 2005). As significant, it had a calming effect on domestic inter­ethnic relations (Lund, 1997).⁷¹ Similarly, peace zones secured militarily can contain actual or potential conflict by cordoning off specified areas, with or without the consent of a government (Nicolaides 45), such as in Northern Iraq under Hussein.⁷²

Escalating violent conflicts

Positions are hardening, relationships break­ing off, parties disengaging. Irregular expres­sions of grievances grow into wider violence, foretelling possible organized conflict. Major hostilities look imminent. The aim is to avoid an irrevocable spiral.

To pre-empt increasing intransigence, invoking and enforcing a priori norms might still be effective. Less than totally punitive measures can activate those in the country who support peaceful resolution. But using coercive diplomacy in the absence of a clear pattern of overt violence or gross violations of norms may be seen as unfair and illegitimate (Nicolaides, 1996: 44) because it presumes actions would occur for which the evidence is equivocal. Another mistaken reflex is to try to address the supposed ‘root' causes of a conflict such as ethnic or religious differences, economic disparities, or lack of democracy, as if they mainly now drive the violence. But such ad hoc structural measures are less and less useful as well as feasible, when it is the violence that drives violence. What is most urgent is to halt the spiral through potent political and military direct prevention.⁷³

The tougher tools of formal diplomacy, though difficult, may arrive at short-term settlements to buy time such as ceasefires (Nicolaides, 1996: 52; Rothchild, 2002: 54; Heldt, 8). These are more likely to be effective to the extent a strong mediator or team is skillful in instilling the parties with an urgent sense of the costs that can come from further bloodshed (Rothchild, 2002:55). They also work better if accompanied by potential rewards that buy off the parties and help them fulfill an agreement, including the offer of development aid (Cortright, 1998, Rothchild, 2002), and/or punishments that pressure them to agree. Where there is asymmetry in power between the parties, measures to strengthen the power of the weaker party may budge the stronger.

Where the parties remain obdurate, coer­cive diplomacy such as sanctions or threat of force may be needed to reverse undesired actions or compel desired actions. Threats of the use of force were used when, for example, Presidents Bush and Clinton issued several warnings to President Milosevic not to support any armed activity in Kosovo as he had in Bosnia. Such threats are more likely to be effective if issued before possible escalations of hostile actions occurs, or if they follow immediately upon initial manifestations of violence (Nicolaides, 44-5), not ex post facto. Threatening to expel a state from an international organization is less effective once significant investment in a violent course has occurred. The more that the conflicting parties inflict physical harm on each other, they cannot just back down the ladder they climbed up, for mutual hurt and increasing fear remain (Mitchell, 2005; cf. Rothchild, 2002: 51). By the same token, indictment by a war crimes tribunal is not likely to prevent the perpetrator continuing to fight, and can be counter-productive, once they are named and being hunted down, as they have no incentive to refrain from fighting, unless some provision allows amnesty. If sanctions are actually used, they must be comprehensive to be effective (Jentleson, 2000: 337). But such coercive diplomacy is less applicable when the threat is a breakdown of a state since the source of the problem is hard to target (Nicolaides, 42). Similarly, non-targeted sanctions have been widely criticized as having considerable neg­ative side-effects for the general population while benefitting well-positioned elites.

One of the few joined debates in this scattered literature pertains to this stage: when are conflicts ‘ripe for prevention?' Some ana­lysts believe it more propitious to act before the outbreak of any significant violence. Violence ‘crosses a Rubicon' from which it is very difficult to return (Jentleson, 2000), creating huge challenges for intervenors (cf. Edmead, 1971 cited in Berkovitch, 1996: 251). Others believe that some initial fighting that gets nowhere, a ‘soft stalemate,' is needed before parties will no longer be tempted to try violence to see if it gets them gains (Berkovitch, 1996: 251). Thwarted violence or blocked confrontation are thus needed to soften parties up to compromise.⁷⁴

Third-party willingness to use force can also influence the calculations of actors regarding their use of force. Much discourse in conflict prevention assumes military force to be antithetical to peace. Some NGOs that first stepped up to undertake conflict resolution responsibilities in threatened countries tend to oppose any form of force ideologically, or to downplay the role of any coercion in favor of non-coercive methods and policies such as diplomacy and, lately, development assistance. But some analysts suggest that sticks as well as carrots need to be exerted more or less simultaneously - with flexibility shown regarding what quotients of each are applied in specific situations (Jentleson, 2000; Byman, 2002: 219). ‘...while coercion rarely is sufficient for prevention, it often is necessary' (Jentleson, 2000: 5). Deterrence through the threat of using force may often be a pre-requisite for effective negotiations and, by implication, structural initiatives. Threats of force can encourage allies within a country to spring up. Still, threat of force must be made clear and credible by clearly conveying a concrete demand and the cer­tainty that non-compliance will be punished (Jakobsen, 1996: 3), such as through pos­sessing capabilities and having domestic and international backing that can be sustained. They also need to be targeted precisely at specific actors who might otherwise escalate their actions, be potentially more costly to the parties than their persevering, identify the proscribed behaviors, and be accompanied by realistic alternative solutions (Nicolaides, 42-4).⁷⁵ The chances increase if the balance of power favors the threat sponsor and the value to the targeted actor of ignoring the threat is greater than the costs of compliance (Jakobsen, 1996: 3-5).⁷⁶

Alternatively, if the threat of force is not backed up with credible force when there is non-compliance, they run the risk of encour­aging aggression by calling the bluff of the international actors (Nicolaides, 45).⁷⁷ Alack of follow-through or half-hearted measures can embolden their target (Nicolaides, 1996: 42-3) if that party comes to believe that the threat is empty. Empty threats toward Bosnian Serbs had adverse effects when the latter did not follow through in protecting safe areas such as in Sbrenica (Jakobsen, 1996: 24).⁷⁸

Actual use of force may be needed to limit emerging violence such as being visited upon a threatened minority group (Nicolaides, 42). Several argue that timely introduction of a relatively small force in Rwanda in May of 1994 would have stopped Hutu extremists from continuing to carry out their plans to kill thousands of Tutsi and Hutu moderates (Feil, 1998; Feil (1998) cited in Jentleson, 2000:16; cf. Melander, 10f.). Butthishasbeen questioned (Kuperman, 2000). The tactical question is what amount is sufficient to restrain or reverse the undesired behavior.⁷⁹

If violence does cease, security guarantees are in place and diplomatic processes are in play, neither freezing of the violence nor diplomatic agreement is sufficient by itself to move the actors to tackle the abiding political and socio-economic problems that occasion a conflict. For these, assistance is also needed for programs in institution building and development, now that they can operate in an environment that is basically stable and not constantly threatened by violence.

All in all, this quick review supports the notion that differing kinds of interventions are needed at particular settings and stages of conflict, and in certain combinations of hard, soft, and other kinds of measures. However, they complicate the simple sequencing that is often presumed: that the greater the hostilities in a conflict, the more that coercive measures are needed.

If one looks at the whole early period, the research does support a general picture in which increasingly coercive measures are needed to the degree a conflict escalates. However, the emerging literature qualifies that simple formula and adds altogether new elements to the equation. Before societal strains become salient, a priori regimes whose specific implications are unforeseeable but hold out attractive incentives can socialize leaders into international expectations. If enforced and resourced, these standards can foster structural and institutional changes that make more likely the peaceful management of transitional stresses from economic reform and democratization. But such liberalization needs to be accompanied by compensatory measures. Democratization needs incremental steps for effecting peaceful transition such as power-sharing arrangements, accompanied by conditional material aid for implementing changes. As political and policy disputes over such changes inevitably arise, sym­pathetic international envoys or missions with significant authority can usefully enter the picture, much earlier than usual, to midwife their resolution - playing ‘good cop' by persuading incumbent leaders to inaugurate changes before they lose control. During such potentially unstable periods - contrary to the assumed sequence whereby military power is a last resort following the exhaustion of diplomatic efforts - security assurance may be essential for undergirding the ensuing domestic political negotiations. Where regimes choose to resist openings and move to repress them violently, firm coercive sanctions and credible threat of military force can deter them, and actual use of effective deadly force can halt their extremes. In short, the conventional scenario (derived perhaps from a Cold War crisis paradigm in which sovereignty is supreme and engagement comes late in the form of diplomacy or military action) does not sufficiently factor in structural measures, hands-on institutional support and positive incentives, and deterrent military measures. Regrettably, however, as useful as all these research findings may be as guidelines to action, they are not followed because decisionmakers do not have such lessons at their fingertips.⁸⁰