Due to the escalation of the Liberian civil war, and in the absence of a significant international response, the Economic Community of West African States (ECOWAS) set up a Standing Mediation Committee (SMC) in May 1990 to seek ways of bringing an end to the violence. Two months later, attempts by the Liberian Inter-Faith Mediation Committee (IFMC) to mediate between the warring factions began to lose impetus. In August, the SMC adopted IFMC proposals as the ECOWAS Peace Plan and assumed the lead role in peace efforts. Although failing to secure a ceasefire, it quickly assembled and deployed a military peacekeeping force, the ECOWAS Ceasefire Monitoring Group (ECOMOG). It also convened a national conference to elect a civilian-dominated Interim Government of National Unity (IGNU), and set about securing Monrovia for the installation of this government. In all these moves, it met stern resistance from the rebel National Patriotic Front of Liberia (NPFL) and from its backers within the francophone states of ECOWAS.

Since then, ECOWAS has orchestrated a series of diplomatic initiatives which have sought to establish the basis for peace in Liberia. With the benefit of hindsight, it is clear that these initiatives have repeatedly and tragically failed. They have not met the needs and interests of Liberia's warring factions and their leaders, much less those of the civilian population. The following survey looks at the peace agreements marking the course of ECOWAS diplomacy, summarising their important clauses, their strengths and their major weaknesses. It is divided into three sections. The first considers the accords signed before the July 1993 Cotonou Accord, the second looks at the Cotonou Accord itself, while the third analyses the post Cotonou agreements. This grouping reflects the pivotal significance of the Cotonou Accord in the shifting political context of the peace efforts.

There were seven major ECOWAS-sponsored agreements signed before the Cotonou Accord. These were the Bamako Ceasefire of November 1990, the Banjul Joint Statement of December 1990, the February 1991 Lome Agreement, and the Yamoussoukro I-IV Accords of June-October 1991. The Bamako, Banjul, and Lome Agreements may be co-considered as the first stage of ECOWAS diplomacy, underwritten by Sir Dawda Jawara, President of The Gambia and Chairman of the SMC.

At the Bamako meeting, the NPFL, the Armed Forces of Liberia (AFL) and the Independent National Patriotic Front of Liberia (INPFL), the three factions active at the time, committed themselves to the ECOWAS Peace Plan and to an immediate ceasefire. In Banjul, they pledged to organise a national conference which would reconstitute the IGNU with the inclusion of their own representatives within 60 days. The Lome Agreement built on the earlier accords and specified modalities for the implementation of the ceasefire. ECOMOG was given the prime role in disarming the three factions, who were to assemble in allotted camps. However, until the formation of the reconstituted interim government, it would carry out these duties in liaison with a Technical Committee comprising members of all the factions. ECOMOG was also to supervise the handing in of weapons, the registration of troops and the receiving of the sick, the wounded and the displaced.

The NPFL only signed up to these agreements under intense military and diplomatic pressure. Furthermore, in securing Monrovia in October 1990, ECOMOG had actively collaborated with AFL and INPFL forces to resist the NPFL advance. In the context of mutual suspicion resulting from these facts, hopes for disarmament seemed far-fetched and the accords soon floundered. They were effectively nullified when the NPFL realised it was being outmanoeuvred in the national conference of March 1991 and resorted to wrecking tactics. Refusing to recognise the authority of the conference, the NPFL leadership then set about consolidating its National Patriotic Reconstruction Assembly Government (NPRAG), an alternative administration based in Gbarnga. The IGNU was reconstituted in Monrovia without NPFL participation.

The pre-Cotonou diplomatic process was revived in June 1991 with the signing of the Yamoussoukro I Accord. This agreement secured an apparent rapprochement between the NPFL and IGNU and built the impetus for the three further accords signed by these parties. Orchestrated by the late President Houphouet Boigny of Cote d'Ivoire, the Yamoussoukro agreements were under-written by a newly formed ECOWAS `Committee of Five' which was effectively led by the Ivorians. They were further facilitated by the involvement of the Atlanta-based International Negotiations Network (INN) led by former US President Jimmy Carter.

The culmination of this new stage of diplomacy was the Yamoussoukro IV agreement signed in October 1991. While this accord reflected an appreciation of the conflict much improved on that of Lome, it still gave too central a role to ECOMOG. It was also over-optimistic about the timetable of encampment, disarmament and demobilisation, which it envisaged could be completed in 60 days.

This said, Yamoussoukro IV, like its predecessors, was compromised by a lack of factional commitment and by antagonism within ECOWAS. Although technically flawed to some degree, it failed more due to continued NPFL aggression and Nigerian fears of francophone-NPFL collaboration. These factors led elements within ECOMOG to provide covert support to the emerging United Liberation Movement of Liberia for Democracy (ULIMO) which cut its teeth opposing NPFL-backed insurrection in Sierra Leone. Especially after ULIMO-NPFL fighting had intensified in western Liberia, the NPFL refused to disarm to ECOMOG. The Yamoussoukro process was finally driven to its grave when the NPFL launched `Operation Octopus', their ill-fated attempt to capture Monrovia in October 1992.

The Cotonou Accord is, without doubt, the most comprehensive accord signed on Liberia, and all the subsequent agreements merely clarify or amend it. Its 19 articles cover ceasefire, disarmament, demobilisation, the structure of transitional government, election modalities, repatriation of refugees and a general amnesty. It was facilitated by ECOWAS in collaboration with the United Nations and the organisation of African Unity (OAU). It was signed by IGNU, ULIMO, and the NPFL, whose military set-backs had forced their return to the negotiating table.

The Cotonou agreement marks a major watershed in ECOWAS diplomacy for a number of reasons. Firstly, it represents the end of the era in which factions signed accords merely as half-hearted or duplicitous responses to external pressure. With the Cotonou Accord, the peace process began to focus more specifically on the relationships and interests of the factions themselves, and hence acquired a much more `Liberian' character. Secondly, it heralded a new stage in the peace process in which Liberian politics came to be dominated by the armed factions. For all intents and purposes, the civil state was ushered towards retirement at Cotonou as ECOWAS sought to institute a power-sharing regime which was genuinely inclusive of all armed interests. Thirdly, while previous agreements had made ECOMOG the sole executor of the peace process, the Cotonou Accord recast the role of the force in recognition of its questionable neutrality. It stipulated first that ECOMOG be expanded to include troops from outside West Africa, and second that it operate in close association with a United Nations Observer Mission (UNOMIL). Taken together, these developments represent an unequivocal effort on the part of ECOWAS to allay the fears of the factions and to accommodate their aspirations within the peace process.

By March 1994, the Cotonou Accord appeared to be showing a degree of promise. The Council of State, the legislative assembly and the supreme court of the LNTG were all installed, and symbolic encampment and demobilisation had commenced. However, the logistics of implementation had already begun to lay bare the deficiencies of the accord.

The peace formula adopted at Cotonou disarmament - resettlement - election did not have an impressive record, having failed woefully in Angola and Cambodia. Only in Zimbabwe-Rhodesia in 1980 had the approach been successful. This was due partly to the willingness and capacity of regional powers to secure compliance from their clients, and partly because forces receiving the incoming Patriotic Front combatants had all arrived at their designated bases according to schedule. These success criteria did not hold for Liberia. On the one hand, Burkina Faso, Cote d'Ivoire and Nigeria would or could not wield the requisite influence over the NPFL and ULIMO. On the other, delays in the deployment of UNOMIL and the new battalions of ECOMOG derailed the timetable for the implementation of the accord. Many felt that these delays exposed the timetable as wildly unrealistic. They certainly allowed plenty of time for the emergence of new factions, including the AFL-sponsored Liberia Peace Council (LPC), and for a renewal of hostilities between the NPFL and its various enemies.

The process of implementation also highlighted a number of other major flaws in the Cotonou settlement. The first of these was that it made no provision for resolving disputes arising out of appointments to the executive of the transitional government. Such disputes stalled the establishment of the LNTG, which had been the pre-requisite for disarmament, and continued to plague the efficacy of the administration for months after its installation. They were also a primary cause of destabilising splits in both ULIMO and the NPFL. Another major shortcoming was that the Cotonou Accord failed to clarify the position and the role of the Council of State. It was unclear on whether the members nominated by the various parties would represent their own factional interests or whether, upon assuming office, they would be independent decision makers serving the needs of all Liberians. This is significant as the leaders of the warring factions had strong vested interests, particularly economic ones, in maintaining instability. Finally, considering the sheer magnitude of the task of resettlement and reconstruction, the Cotonou Accord also placed unrealistic expectations on the deeply divided LNTG. By mid-1994, it was clear that the Cotonou Accord was inoperative.

Between the collapse of the Cotonou Accord and the outbreak of renewed fighting in April 1996, there were three new agreements signed on Liberia: The Akosombo Accord in September 1994, the Accra Clarification of December 1994, and the August 1995 Abuja Accord. All of these agreements were secured under the resolute auspices of incoming ECOWAS Chairman and Ghanaian President Jerry Rawlings, with degrees of support from other ECOWAS figures, notably in Cote d'Ivoire and Nigeria. The UN, the OAU, and a range of international NGOs also lent their support at various times. Although each set out to amend and strengthen the Cotonou Accord, the post-Cotonou agreements still failed to address the underlying issues and interests propelling the war.

The principal objective of the Akosombo Accord was to grant the LNTG and the armed factions who dominated it a more central role in the management of Liberian state affairs. To this end, it amended the Cotonou Accord in a number of ways. First, it placed the responsibility of overseeing the peace agreement not only on ECOMOG and UNOMIL, but also on the `security forces' of the LNTG. Second, it ceded effective control of the Council of State to the more powerful factions, removing the requirement for consensus decision-making and instituting the authority of a simple majority. Thirdly, it also permitted factions to review the status of their appointees at all levels of the administration at any time.

The Akosombo Accord ran into major problems over appointments and nominations to the LNTG. Strong resistance was registered by some ECOWAS members, notably the Nigerians, and by a range of Liberian political parties and civilian interest groups. A major concern was that the agreement disproportionately favoured the NPFL and ULIMO-K who by this time controlled most key executive portfolios and were jointly pressing for the aged Chief Tamba Tailor to be appointed Chairman of the Council of State. The increasingly central role assigned to armed factions within the LNTG also led to groups aligning and realigning themselves in a violent scramble for position. This in turn produced significant breakdowns of command within individual factions, notably ULIMO.

Compared with the three signatories of Akosombo, no less than eight groups signed the Accra Clarification. However, while this agreement brought new groups into the peace process, along with those who had been left out of earlier accords, it did little, if anything, to advance the search for a solution. After signing the Accra Clarification, the parties failed to reach an agreement on the composition and chairmanship of the Council of State, with the AFL and `coalition' forces particularly struggling to determine their joint nominee. Crucially, while factions and political leaders haggled, the implementation of other aspects of the agreements remained stalled. Perhaps most significantly, Accra attempted to co-opt the civilian Liberian National Congress without heeding its concerns on the militarisation of the peace process.

The Abuja Accord managed to bring the leaders of all warring factions into the transitional government, which was installed with due ceremony in September 1995. As such, it represents in many ways the fulfillment of the aspirations of the post-Cotonou peace process. It did not however bring peace to Liberia. Discontent continued to fester within groups and individuals who believed themselves sidelined in the settlement. Factions continued to guard their territorial and commercial resources jealously, with ongoing violence between NPFL and LPC and between various sub-groupings of ULIMO. The lack of security and international support continued to impede the full deployment of ECOMOG and UNOMIL. As elements of NPFL and ULIMO-K sought to consolidate and extend their power within the framework of the LNTG, tensions intensified. Increasing numbers of factional fighters entered Monrovia, ostensibly to ensure the security of their leaders. On April 6th 1996, police forces, backed up by NPFL and ULIMO-K fighters, moved to arrest ULIMO-J leader Roosevelt Johnson on charges of murder. This led to an eruption of violence between NPFL/ULIMO-K and ULIMO-J/AFL/LPC which was the worst seen in Monrovia since before the Cotonou Agreement.

A couple of general points can be made concerning the shortfalls of the Liberian accords. Firstly, all have failed to address the vested interests profiting from the anarchic status quo. This might have been achieved through the institution of effective mechanisms for establishing and monitoring embargoes on illicit trade. Secondly, in seeking to accommodate the demands of armed groups rather than shoring up civic and political institutions, the accords have all been vulnerable to the emergence of new and proxy factions. Such factions have typically not been signatories to existing agreements and as such have not been under obligation to respect them.

In addition to these general points, some specific failings can be highlighted concerning the two discernible phases of the peace process. In the pre-Cotonou phase, it is clear that all aspects of the process were severely undermined by the ambiguous nature of ECOWAS intervention. If the aim of early initiatives was to protect a civilian state from the illegitimate accession of the NPFL, then the rapid deployment of ECOMOG and the resolute defence of the IGNU is understandable. Less so are the attempts to pressure the rebels into a peace process whose aims they clearly held in contempt. On the other hand, if the idea was to minimise instability by seeking a neutral compromise with the NPFL, then the stern confrontations and partisan collaborations of ECOMOG make little sense. What is clear is that pre-Cotonou, ECOWAS itself was riven by uncertainty and by competing interests working at crossed purposes. Under these circumstances, the early accords were probably doomed to prolong Liberia's suffering. Enduring peace was certainly not served by the churning out of agreements to which the sponsors, never mind the protagonists, were clearly ambivalent.

If the early accords were void of much meaning, post-Cotonou initiatives were more misguided. A major flaw in all of them was that, while increasingly ceding power to the factions in the executive arm of the transitional government, none of them established mechanisms for resolving the inevitable disputes over the interpretation of provisions. Another destructive element in the post-Cotonou efforts was the way in which faction leaders were permitted to extend their political power without acceding to the military pre-conditions specified in the accords. In protecting their accumulated interests, all factions were hugely reluctant to proceed with disarmament, a phenomenon abetted by a lack of international censure of faction leaders and the inordinate delays and shortfalls in the deployment of military observers.

While a thorough, binding and well-resourced settlement is yet to be reached in Liberia, the resurgence of warfare in April 1996 seems to have created a watershed in the peace process. In order to facilitate elections and secure a more just and peaceful political dispensation, Liberians and their international partners must seize the moment. However, if they are to successfully transform organised violence in Liberia into constructive, unarmed political conflict, they require the vision and the courage to engage with the evolution of the peace process thus far and to learn from its manifold shortcomings.