5. The architecture of the peace
The Truce and Peace Monitoring Groups
For a number of years, the BRA had been working towards internationalising the conflict through various UN agencies. So when an international monitoring force was agreed to in 1997, the BRA/BIG preference was for a UN force. Having always resisted the BRA’s internationalising strategy, the PNG Government also resisted UN peacekeeping. A Truce Monitoring Group (TMG) was established in November 1997, led by Brigadier Roger Mortlock of New Zealand. Half the 250 monitors were New Zealand military, with most of the rest being Australians (civil servants for the front-line roles, where there was distrust of Australian military involvement, though Australian military filled other roles) and military/police from Fiji and Vanuatu. Various informants said there were immediate positive effects from their arrival. Fighters no longer felt they could kill others with impunity. In the newly pacified space, lines of communication began to open up. While the sheer fact of arrival increased confidence that commitments for peace would be kept, the projection of the slogan ‘No ken stoppem peace’ (‘The peace cannot be stopped’) by the peace monitors was well crafted to build optimism in the invincibility of the peace.
Both the TMG and subsequent Peace Monitoring Group (PMG) were unarmed. This was driven by BIG/BRA anxiety that Ona and his remaining allies might misinterpret the intentions of an armed intervention—for example, as an Australian invasion to retake the mine. As Anthony Regan (2008) concludes: ‘By basing peacekeeping authority on a moral element instead of physical force, it enhances peace by contributing to a non-violent culture.’
In May 1998, the TMG was replaced with a new mission with a similar peacebuilding mandate: the Peace Monitoring Group (PMG). After a softening of initial resistance to Australian leadership, the PMG was led by Australian commanders. There had been successful confidence building through the Australian presence in the TMG. Much of the suspicion about Australian involvement centred on Australian commercial interest in reopening the mine. We were told stories of Australian helicopters ‘prospecting for gold’ and of Australians being tested with offers to buy gold. The 300-strong PMG was five-sixths Australian (Londey 2004:222) with Australian naval support offshore. Beyond monitoring the truce, the PMG was to educate citizens about the peace process, provide a framework for the full restoration of government services, oversee the collection and destruction of weapons, foster reconciliation, promote development and help establish a Bougainville police force. PMG commander Osborn saw part of their job as bringing conflicting parties together in dialogue and facilitating negotiations by providing a mixture of ‘ideas, information, communication and transport’ (Londey 2004:223). The PMG did not provide development aid but helped with medical assistance and transport for locals in response to immediate needs they confronted in the field, while trying to be careful not to distribute favours such as offering rides in ways that seemed to favour one faction over another. The peace operation ‘saved many lives, particularly those of mothers and newborn babies’. We interviewed peacekeepers who had babies named after them whom they had helped save by getting their mother to a doctor (sometimes to the concern of their wives!).
Between the deployment of the Truce and Peace Monitoring Groups, the Lincoln peace conference was held in January 1998, bedding down a ‘permanent and irrevocable’ cease-fire.
In 1997, when the TMG arrived, there could have been 800 PNGDF troops in Bougainville and 150 police mobile fighters armed with automatic weapons within PNGDF patrols (Londey 2004:219). The BRA fighting strength could have been more than 2000 on one account (Londey 2004:219). Anthony Regan’s comment on our text here is that the number of core fighters could have been only a couple of hundred, with many more being ‘home guards’. The BRA had perhaps 500 automatic weapons and 2000 or 3000 reconditioned World War II or homemade guns (Regan 2001). The Resistance had perhaps 1500 fighters, according to our interviews, though some Resistance leaders claimed 5000 men willing to fight, with a variable number of weapons ‘on loan’ from the PNGDF. Again, Anthony Regan makes the point that the overwhelming majority of the Resistance forces were little more than home guards.
Figure 5.1 BRA fighters posing with a US World War II bomber turret gun they had restored
Photo: Brother Bryan Leak
While the TMG was led by New Zealand, it was supported mainly by Australian resources. The Australian Defence Force was an advocate of Australian leadership, but all the diplomatic advice, including from Australia, was that the second-best after UN leadership for BRA acceptance and legitimacy with disarmament would be New Zealand leadership of the TMG. A New Zealand-led regional mission could also be put in place more quickly than a UN-led mission. The UN Security Council lent its support to the operation, but did not open a six-member UN Political Office in Bougainville (UNPOB), with some local employees, until August 1998, after the TMG had handed over to the PMG. Nevertheless, this office of the United Nation’s Department of Political Affairs (as opposed to the Department of Peacekeeping Operations) was a central part of the peace process, playing a vital role in bringing the leading parties to the conflict together to discuss the substance and process for the peace.
Ninety per cent of the BRA fighters had followed Kabui and their military commanders, rather than Francis Ona, into the truce. Ona saw the strength of will of the people of Bougainville for peace, which was evident within his own village, and began to adopt the position that he supported peace, but not the particular peace that his former deputies had negotiated. He declared a ‘no-go zone’ in the mountains centred on his village of Guava and encompassing the mine site. Both the TMG and the PMG respected the no-go zone, hoping that Ona’s hold-out force, which became known as the Me’ekamui Defence Force, would eventually fully join the peace and surrender their weapons. Ona almost certainly believed that a peace could not work without him. Instead, he gradually became more marginal and ultimately the Autonomous Bougainville Government and its leadership became more legitimate in the eyes of the people of Bougainville. Bougainvilleans had looked to Ona to get out in front in pushing for peace; his star faded when he failed to hear these hopes.
Outside the no-go zone, the TMG and PMG formed into small patrols that visited most of the villages in Bougainville, explaining the peace process, handing out literature and listening to complaints and concerns. They operated out of five regional centres in addition to the central bases (Regan 2001, 2010:65). The patrols were multinational and each included a woman. The female member of the patrol was important for building rapport with the women of the village. This assisted with maintaining the momentum of women’s leadership in peacebuilding. Many of the villages could be reached only on foot. Village meetings often ran for hours. In time, the emphasis shifted from visiting as many villages as possible to spending more time at villages strategic for the peace—for example, where there were a lot of weapons to be handed in or where there were potential spoilers. Peacekeepers learnt the social respect acquired by laying on a feast. They became good at building a relationship over food, with music and through sport (Breen 2001b:48). Even a volleyball team from Francis Ona’s village joined in the volleyball competition organised by the PMG; an Australian military surgeon bought them jerseys. Monitors who could play a musical instrument or sing were especially important assets. This leads to some lessons learned from Bougainville: make musical talents a selection criterion for peacekeepers, give singing lessons to interested peacekeepers, and prepare peacekeepers before deployment with some training in how to lay on a feast that indigenes will really appreciate. Australian and New Zealand military commanders saw the peace potential of music and organised several successful tours of Bougainville by military bands.
The Pacific Islander contingents increased more than the diversity of the musicality of the international peacekeepers, though they certainly did that as well:
I was with a patrol commander, a medic, a civilian monitor and one or two ni-Vanuatu and Fijian peace monitors. They were so much better than we were at picking up the vibe of a village. For example when we would fly in these people could tell us fairly quickly if there had been something that had happened since the last visit. They had a much quicker rapport with the locals than we did. (Australian peacekeeper interviewed by Reddy 2006:239)
A member of parliament for the Panguna area diagnosed Bougainville’s problem as one of distrusting everyone: Papua New Guinea, Australia, but most importantly one another. The TMG/PMG, and particularly the peace newsletter they distributed during their village visits, started to rebuild trust, in his view, by providing factual information that countered some of the wild rumours that were common.
The PMG gradually wound down and finally departed in June 2003. Both the TMG and the PMG were overwhelmingly acclaimed for the sensitive and helpful job they did, though the acclaim for the TMG and for Brigadier Mortlock was even more emphatic than that for the PMG. We consistently failed to draw any criticism during our Bougainville interviews from questions about the biggest mistakes made by peacekeepers or suggestions for how they could have done a better job, though a large number of respondents to the latter question said they should have stayed longer or stayed until all weapons were handed in and until Francis Ona formally joined the peace. There were the occasional allegations of Australians looking for gold from their helicopter and in rivers and another about an Australian soldier smuggling out orchids. This near-universally positive assessment of the monitors was in great contrast with the reaction to peacekeepers we report in the next two Peacebuilding Compared volumes on the Solomon Islands and Timor-Leste and in UN peace operations in other regions. As a top Australian intelligence official said: ‘Unlike so many UN operations, there was none of that frequenting of brothels. Monitors were highly disciplined and that was a key to success. New Zealand set a good tone; it worked well because they set good rules early on. [There were] strict rules on fraternisation.’
From July to December 2003, the Bougainville Transitional Team of just 15 remaining from the PMG assisted with the withdrawal of the peacekeepers and the completion of weapons disposal, which still had a long way to go and in 2002 suffered reversals from a number of de-containments of contained weapons in response to perceived or real local threats (UNIFEM 2004). A successor to UNPOB, the UN Observer Mission of Bougainville (UNOMB), continued the UN work in 2004 through to June 2005. All these missions came and went with no casualties from violence. They did leave, however, with perhaps a thousand or more usable weapons not contained.
The post-conflict peace process
Regan (2008) identified three phases to the post-conflict peace process: a process phase that did not attend to outcomes; the negotiated political settlement phase; and the implementation of the settlement phase. The first stage lasted from 1997 to June 1999. It involved establishing the peace process and its institutional architecture, including the Truce and Peace Monitoring Groups and a Peace Process Consultative Committee, a road map for steps towards a negotiated process signed at Lincoln, a ‘reconciliation government’ ‘to unify the Bougainvillean factions in advance of negotiating a political settlement’ (Regan 2007), establishing the UN political office in Bougainville (UNPOB) with a mandate to monitor monitors and the peace process more broadly. The Peace Process Consultative Committee, which included no women, was chaired mostly by the director of the UN observer mission. He became in time a more important mediator than the leadership of the Truce and Peace Monitoring Groups of a peace negotiated principally among local leaders. As trust grew, ad hoc meetings of two or more of the parties were also an increasingly important method for settling specific issues (Regan 2005a).
The second phase ran from June 1999 to the signing of the Bougainville Peace Agreement on 30 August 2001. It guaranteed: a referendum on independence for Bougainville, deferred for 10–15 years; a high level of autonomy for Bougainville; and demilitarisation by disposal of the weapons of local combatants and withdrawal of all PNGDF and police mobile forces. The third stage continued from 2001 to the present. It involved real disposal of weapons in a process that continued to 2005 but that left many weapons circulating in Bougainville, especially in the hands of Ona’s Me’ekamui Defence Force and Thomas Tari (the commander who captured the weapons at Kangu Beach; Box 4.1). Demand for new weapons from the militarisation of Tonu Village, where Noah Musingku sought refuge, and the realisation that guns could be an export industry for Bougainville criminals have seen in recent years an upsurge in digging up and reconditioning World War II weapons around the old American base in Torokina. In 2009, the US Government finally undertook to clean up the buried weapons it left behind on Bougainville 65 years ago in the interests of the security of the South Pacific (Regan 2010). The third stage of the peace process also involved withdrawal of the PNGDF in 2001, constitutional laws to implement the agreement that were jointly crafted and passed by the PNG Parliament in 2002, a participatory process for drafting a constitution for an Autonomous Bougainville Government (ABG) and an election for the ABG in 2005 that saw Joseph Kabui become President.
Anthony Regan (2010) characterises the peace process as ‘political negotiations across two divides’. The first divide separated the Resistance and the BRA. It was in fact a suite of divides along a continuum of positions between strong support for immediate secession and for immediate full integration with Papua New Guinea. The second divide arose after a consensus position among the Bougainville factions was reached. It was between the newly united Bougainville parties and a PNG state that feared even a deferred referendum on independence as a divisive and dangerous precedent.
The first suite of divides was bridged in mid 1999 as Bougainville factions prepared for political negotiations with Papua New Guinea by a process of advisers to all factions jointly developing nine options for the future. These ran from immediate independence to full integration into Papua New Guinea.
Next the advisers assigned tentative ratings to each option—high, medium or low—based on assessments of how well each option could contribute to dealing with the needs of post-conflict Bougainville reflected in 20 criteria (Regan 2002c). These criteria took account of Bougainville’s weak economic base and low level of administrative capacity, and so on. This process gave the highest rating to the option of a constitutionally guaranteed referendum on independence for Bougainville from Papua New Guinea combined with a high level of autonomy for Bougainville during the period of deferral. Next, the advisers’ analysis was discussed at length by the leadership, who agreed that the highest-rating option provided the best basis for a compromise solution. Only this option balanced the interests of opposing factions by keeping the question of secession alive, but leaving an ultimate decision on the choice between secession and integration to a later democratic process conducted when weapons had been disposed of, reconciliation achieved and the economy restored (Regan 2005a:26–7).
This consensus Bougainvillean negotiating position set the agenda for negotiations with Papua New Guinea that began in June 1999. The PNG negotiating team resisted it fiercely, seeing it as a threat to the unity of the central state that would presage disintegration were it emulated by other provinces. Australian Foreign Minister, Alexander Downer, broke through this deadlock by suggesting the compromise of, first, deferral of a referendum for 10–15 years, and second, substituting the Bougainvillean proposal of a binding referendum with a non-binding referendum subject to the sovereign approval of the PNG Parliament (Wolfers 2002:3). Why would the former BRA leaders agree to a referendum that the PNG Parliament could ignore? Backstage, Downer and the other influential international players persuaded them that the international community could and would pressure Papua New Guinea to honour a vote for independence at that time. Because so much of the PNG budget at that time was funded by Australia, it seemed plausible that this pressure would work. Downer argued that even without that clout over Indonesia, international pressure caused Indonesia to honour the 1999 East Timor independence referendum outcome. This compromise on the Bougainville compromise finally crossed Regan’s two divides. Minister for Bougainville Affairs, Moi Avei—as Peter Barter had done—built a high degree of trust with the Bougainvillean and PNG players at this critical moment (Regan 2002b:118).
The other key part of the agreement was that the Autonomous Government of Bougainville would have full authority to administer its own affairs, to legislate across the entire gamut of governance, including those areas most embroiled in the conflict: land, minerals and other natural resources and the environment. Key exceptions reserved for the central government were most foreign affairs and defence powers, central banking, international trade, customs and quarantine, international shipping and civil aviation, industrial relations and posts and telecommunications. Bougainville would have its own police, courts and taxation power.
The next priority was drafting a Constitution of the Autonomous Region of Bougainville. The draft was completed between 2002 and 2004 by a 24-member Bougainville Constitutional Commission that was broadly representative. The constitution was discussed, amended and ultimately adopted in November 2004 by a broadly based Bougainville Constituent Assembly.
Credible commitment architecture
As we have documented, Papua New Guinea and Bougainville had a history of deep distrust. Yet Papua New Guinea did have a comparatively good record of credibly committed constitutionalism. So the Bougainville negotiators derived assurance from entrenching the commitment to a referendum in the PNG Constitution. Double entrenchment was accomplished by requiring consent of the Bougainville legislature for any constitutional amendment relating to Bougainville (Wolfers 2006a). Even more creative was the embedding of implementation assurance in an architecture of ‘sequencing of, and linkages between, agreed-upon steps for implementation of key arrangements’ (Regan 2008). Each side committed to steps where it yielded up some power, provided the other gave it some power first. One party’s stepwise commitment was a precondition for the step-by-step commitments of the other.
The main linkages are between the provisions, on the one hand, for disposal of weapons by Bougainvillean factions and, on the other hand, the withdrawal of PNG forces from Bougainville and, more importantly, the constitutionalising and implementation of the agreed arrangements. The linkages involved completion of steps agreed to by one party being specified as a ‘condition precedent’ to be completed before the other party was required to take a separate step. In particular, the obligation on Bougainville’s ex-combatant groups to move weapons to secure storage arose only on Papua New Guinea both making the constitutional amendments implementing the agreement and beginning the withdrawal of its forces from Bougainville. In order to maintain pressure on Bougainvillean groups to dispose of their weapons, after the constitutional amendments implementing the agreement were passed by Parliament, they did not come into operation until the UN mission verified completion of stage two of weapons disposal (secure containment). Lack of substantial compliance with the agreed-on weapons disposal process could have resulted in the UN mission delaying elections for the ABG (any party to the agreement could call on the United Nations to verify and certify substantial compliance with weapons disposal and whether the level of security for the weapons was conducive to the holding of elections) (Regan 2008).
There is contextual genius in this architecture that gives parties who have taken pride in tricking one another incentives to honour their commitments. The sequencing has proceeded not always punctiliously and punctually, but pretty well. Particularly slow has been the transfer of specific governance competencies to the Bougainville administration (for a discussion of notice and other procedures for transfer, see Wolfers 2008). The Bougainville civil servants so far remain much more under the authority of their civil service bosses in Port Moresby than Bougainville ministers and the ABG President. This is partly a matter of Bougainville politicians needing time to develop experience in directing civil servants. Bougainville does not yet have its own court system; its judicial branch is part of the PNG judiciary. Capacity bottlenecks are the main reason why so few executive and judicial competencies have shifted to Bougainville so far. There is just so much to be done with building the full range of state capacities from scratch in such a small polity. In all areas, state capacity in Bougainville remains stunted. International assistance for this work has been less than it should have been—strikingly less than we will encounter in our next case study, the Solomon Islands.
The positive side of this is that Bougainvilleans are taking their time to construct their own autonomy in their own way. For Anthony Regan (2005a:44), this is part of the virtue of ‘extended timetables with which managers of international interventions are usually not comfortable’. In the Regan view, improved state capacity that comes slowly in response to local pressure for better service for citizens is more likely to be sustained and more genuinely responsive to needs as well. There remains a worry that the sequencing of commitments of the peace might work well up to the last step. To be blunt—and hopefully unfair—we fear there was good reason for Bougainville to be distrustful of the PNG state. If Bougainville votes for independence in a few years, will the PNG Parliament vote to honour their choice? Our interviews suggest many in the current PNG Parliament would not. There is much persuasive diplomatic work to be done by the many—nationally and internationally—who believe independence is not in the interests of economic and human development for Bougainvilleans. They need to get on with that conversation with the people of Bougainville just as they need to engage the present and potential future leaders of Papua New Guinea with the need—in the interests of such a hard-won peace—for national and international resolve to honour whatever the people of Bougainville decide is best for them.
The peace that was won in Bougainville was remarkable. Many experienced hardheads who were close to the action did not believe it could hold. The architecture of sequenced, linked commitments described above deserves some of the credit for it holding so far. If it does unravel in a final fateful feat of betrayal of Bougainville, we should not throw out the lesson that has already been learned about the contribution contextually attuned linked sequencing of commitments can make to peace. This is not to say that the particular sequencing that has secured 12 years of peace in Bougainville will work elsewhere. In most contexts, it would be wildly imprudent to delay weapons destruction for as long as it was delayed in Bougainville—just as having unarmed peacekeepers would often be a naive step in other circumstances. In Aceh (Braithwaite et al. 2010:Ch. 6), we saw that peacemaking failed when it opted for a truce deferring a political settlement and weapons destruction until confidence building proceeded; yet peacemaking succeeded in Aceh when a ‘nothing is agreed until everything is agreed’ political settlement preceded the cease-fire. What we laud in this book is the contextually attuned architecture of linkages that in this particular case saw the truce precede the political settlement. This is not to laud its specific sequence as generally applicable. We know that failures of credible commitment recurrently explain the outbreak of war (Collier et al. 2003). We know that peacemakers can learn to ‘think in time’ (Neustadt and May 1986) by asking contextual questions.
The Bougainville peace process was unusual in that the international peace monitors departed long before weapons disposal was completed. The PMG also made no real attempt to disarm the Me’ekamui Defence Force though elements of the force broke away and joined in, disposing of their weapons (Wolfers 2006b:7), as the peace process progressed. Francis Ona seemed to have no intention of an armed assault to take over Bougainville. He knew that the people of Bougainville would not support that—nor would many of his former fighters. When we asked a group of Resistance commanders what they would have done if Ona had sought to use his guns to re-conquer Bougainville, one said, ‘Our guns were in boxes, but we had keys.’ When we quibbled, pointing out that the United Nations held the second key, he said, ‘We had bolt-cutters too.’
Figure 5.2 BRA fighters, 1994, with homemade weapons and a World War II Japanese canon reconditioned at the Panguna mine workshops
Photo: Ben Bohane
Ona was a hold-out, as he saw it, because it was only a matter of time before the people of Bougainville would come to his view that the truce was a trick. Papua New Guinea would betray Bougainville and then the people would come back to him. There were some in the BRA who went with the peace thinking that if Ona was right after all, it would perhaps be as well that Ona was holding out with his guns in the no-go zone. Sam Kauona (2001:88) went close to saying this even in 2001: ‘We would rather he join us in the peace process, but in a way Francis provides the check and balance by staying out. He serves as a warning; as a reminder of the alternative if the peace process fails.’ It is still quite possible that Papua New Guinea will not honour its commitment to defer to the people of Bougainville if they vote for independence a few years from now. In 2005, Ona died without being vindicated. The Me’ekamui Defence Force, however, lives on. At different times since Ona’s death, different individuals—including Noah Musingku and Chris Uma—have claimed to have inherited Ona’s control of the force and its weapons. Some local informants said in 2007 that after Musingku had broken away to form with Fijian-trained fighters a different node of opposition to the ABG, there remained three factions of the old Me’ekamui Defence Force. One was the faction led by Uma, another was a group loyal to Ona and the third was a group who had been suspended from the Me’ekamui Defence Force by Ona for criticising his collaborations with certain white people. In early 2010, some unity has formed among many of the Me’ekamui factions, particularly in the no-go zone, joining together in the new peace processes led by President Tanis, but fragmentation remains or has increased around Noah Musingku and possibly Chris Uma.
The reality of Me’ekamui hold-outs has created an unstable situation in southern Bougainville in recent years. The considerable residual access to automatic weapons in the south has played into a variety of historical grievances between different groups and families, different armed factions from the war and into opportunities to exploit greed through roadblocks, extortion from the government and other criminal activities. A new push on weapons disposal is needed and President Tanis has been seeking to secure agreement to this. His successor, John Momis, promised during his 2010 election campaign that he would continue this push. Getting the agreement of the US Government to clean up its World War II weapons has been an encouraging start. A good example of the problem is the conflict between the Damien Koike and Thomas Tari groups, which have each been responsible for killings of members of the other group, with Koike attempting to kill Tari and Tari attempting to kill Koike. It was an inter-family, inter-warlord struggle for control in which each used the continued access to the formidable arsenals of the other to justify their own refusal to destroy weapons.
In 2001, when progress was stalling on the weapons disposal process, Australia provided A$5 million for small income-generating projects for ex-combatant groups in communities where weapons disposal was proceeding. Combatants were also promised amnesties and pardons for conflict-related offences as part of the 2001 peace deal. The three stages of the agreed weapons disposal process were completed only in May 2005 (Regan 2005a). This involved collecting weapons into locked boxes that were regularly audited by the PMG (Spark and Bailey 2005:601). At first, ex-combatant commanders of the units that surrendered the weapons kept the keys. In the second stage, locally contained weapons were consolidated into more centrally located double-locked containers. The UN observer mission held the second key. After UN verification of completion of the second stage, the third stage was final disposal/destruction of the weapons.
Seven and a half years between truce and disposal of (perhaps most) weapons was a long and risky wait. In fact, it was a double-edged risk. One side of that risk was what Thomas Tari did: refusing to dispose of the weapons at the end of the agreed containment stage, and later breaking open the containers and creating a certain amount of havoc with them as a post-conflict criminal entrepreneur. Another was the larger risk of restarting the war, which did not happen. But the biggest risk of that happening was posed by the Me’ekamui Defence Force, which was not a party to the peace or to weapons containment. The BRA and Resistance could only credibly promise to protect the unarmed international peace monitors because in the circumstances of an attack on them by the Me’ekamui Defence Force they could open the containers.
Anthony Regan’s (2010) most recent update—headed for press at the time of writing—suggests that the weapons retained by the Me’ekamui Defence Force (MDF) could have been much greater than previously believed:
It was estimated by senior BRA figures at the time that the MDF held about 400 to 500 weapons, though more recent estimates by former BRA and MDF leaders who, since 2009, have been seeking to facilitate development of a weapons disposal process involving the MDF elements in the former ‘no-go-zone’ in the vicinity of the Panguna mine…suggest that there could be a much greater number of weapons in the hands of MDF elements—perhaps as many as 2,500 (including a substantial, though as yet unknown, number of WWII weapons as well as significant amounts of ammunition from the former US/Australian World War II base at Torokina, most obtained since 2005). (Regan 2010)
Credible commitment and confidence in the peace
Delayed, graduated and partial weapons disposal linked to constitutionalisation of autonomy and a referendum provided a unique panoply of assurances. It would begin and progress only after the political settlement and constitutional changes began and ended (and after the PNGDF had completely departed). And it remained in place until the year of Francis Ona’s death—seemingly from natural causes. Stages of disarmament did not proceed until stages of politico-legal transition and stages of marginalisation of Francis Ona were behind the wary factions. By May 2005, 1900 weapons had been destroyed and UNOMB certified that stage three of the disposal process was complete and the community was now living with a sense of security. Some police think it could be accurate that two-thirds of all weapons that had been circulating have been contained, but the figure is considerably less than that for factory-made weapons. Thomas Tari’s arsenal (Box 6.1), which had been part of the peace process, and the Me’ekamui Defence Force arsenal, which was beyond the reach of the peace process, were significant on their own. Many weapons were also sold to be used in fighting in the New Guinea Highlands and in the conflict in the Solomon Islands (Alpers 2005:65, 69). Philip Alpers (2005:41) reports that two of the F-1 machine guns that Australia shipped to Papua New Guinea for use in the Bougainville war were seized by Southern Highlands Police in Wogia. National Intelligence Organisation officers in 2007 reported they still regularly received reports of the movement of guns into and out of Bougainville. It was far from a perfect weapons destruction process. Weapons destruction after war never works perfectly. In Bougainville, it worked credibly enough to consolidate peace but not well enough to give Bougainvilleans security from former BRA commanders who became minor post-conflict criminal entrepreneurs such as Thomas Tari, Damien Koike and Chris Uma, and indeed from Noah Musingku.
Yet it might have been better if the PMG had stayed at least until Thomas Tari, the Me’ekamui Defence Force and Damien Koike’s group were fully disarmed and all roadblocks that secured no-go zones with guns were dismantled. That might have been done with a patient winding down of the size of the commitment of personnel to quite a minimal level, but with a willingness to escalate—and indeed escalate to armed peacekeeping—had diplomacy at the roadblocks failed through the reality or the promise of armed violence. Volker Boege and Edward Wolfers both questioned in their comments on this conclusion whether it ever had been or would be necessary to mobilise for armed enforcement. Wolfers said: ‘The progress made in encouraging former hold-outs to engage in the peace process, including efforts by former President Joseph Kabui shortly before he died, and by his successor, James Tanis, would seem to belie the arguments for a delayed peace process or possible international armed intervention.’ Further progress in broadening the genuine engagement of various hold-out armed factions, with courageous support from church, chiefly and other traditional leaders, including many women who walked into armed camps to talk peace, in the six months after Wolfers’ comments would seem to vindicate his analysis. This steady expansion of the peace certainly does caution against any hasty resort to peace enforcement or police enforcement by making the point that patient local diplomacy has slowly born fruit.
While the top-down architecture of the peace we have described in this chapter has been critical to the peace that has been secured so far, an even more important ingredient is the bottom-up reconciliation and reintegration discussed in the next chapter. Yet a central conclusion of this book will be that this is a peace with a mutually enabling relationship between a top-down credible commitment architecture and bottom-up reconciliation.
 Sic. It should have been ‘stopim’.
 One Vanuatu peacekeeper we interviewed was concerned about his wife’s reaction to a baby being named after him and the dispatch of gifts to baby and mother in Bougainville. He was relieved when his brother was subsequently posted to Bougainville and able to renew the family connection, reporting back to his brother’s wife on the reality of the happy family circumstances of the birth of the baby. Another was greatly relieved to receive a photo of the baby named after him to show his wife how Bougainvillean was the skin colour of the baby!
 Commenting on these numbers in 2009, Anthony Regan said ‘[t]hose figures would need to be reconsidered in light of the weapons actually contained from 2001–2005—which were about 2000. The assumption by then was that this might be two-thirds of the total weapons, with perhaps 500 or so still in the hands of the Me’ekamui Defence Force, something similar retained by other elements—BRA/BRF, etc. But we also know there has been some leakage out, some new weapons in, and perhaps 400 or more World War II weapons dug up and brought into circulation.’
 Sport, according to some interviews, was especially helpful to the peace when combatants from different sides played on the same team, as opposed to team competitions that mapped onto divides from the conflict.
 This does not have to be inconsistent with moments when pushing through to a result for a negotiation deadline through a kind of exhaustion has no place. Brigadier Osborn (2001:52) described the negotiations the night before the scheduled cease-fire signing ceremony of 30 April 1998: ‘The signatories and the TMG had only agreed on its contents at 3.50 am that morning on HMAS Tobruk.’ Dr Bob Breen, who was on the Tobruk, said that the women leaders played a vital role that night and morning as go-betweens and in pushing the men to the point where they would sign. When we asked Breen to check if we had got this right, he contacted two other senior Australians who were on the Tobruk, one of whom ‘was adamant that the women shamed the men into signing the implementing agreement’, with the other ‘supporting this view’. The women had also played important goading and go-between roles at Burnham and Lincoln before the male leaders were ready to join hands. And they had often done that before in the jungles where battles were being fought.
 The Panguna Communiqué, signed at a large reconciliation meeting in August 2007 (Boege 2008:11; Wolfers 2008:189), was the culmination of a process of discussion by President Kabui and leaders of his government with Me’ekamui leaders in the preceding months. It greatly expanded the path for the restoration of government services to the no-go zone and the dismantling of roadblocks. Included as a minor part of this process was a two-day peacemaking conference organised by the Peacebuilding Compared Project, the Buka Open Campus of the University of Papua New Guinea and the ABG, attended by the President and by James Tanis throughout and also by Me’ekamui leaders on the occasion of the second anniversary of the ABG in June 2007. Willie Aga signed this communiqué on behalf of the Me’ekamui Defence Force. A video of this entire conference can be found on the Peacebuilding Compared web site.