Norwegian Institute of International Affairs, Peacekeeping and Multinational Operations No 6, 1995.

 

ISSN 0804-6794 © Norsk Utenrikspolitisk Institutt


Executive Summary

1. The wrong lessons regarding the use of military force have been learnt from Somalia, the watershed experience of the second generation of UN operations.

2. The phases of international involvement in Somalia roughly correspond to the three eras of UN operations generally: diplomatic peacekeeping, the use of military force, and the exercise of political authority.

3. The self-perceived mandate of peace-enforcement had its origins in a debate about the limited use of force. It has been confused now with high-intensity enforcement against an international aggressor.

4. Somalis did not merely revert to clan identification after the collapse of the state in 1991. Conflict was dictated by the whims of political factions and warlords.

5. Colonialism destroyed the social fabric that had previously maintained order in a decentralized community. In the factional conflict, therefore, there was no check or balance to restrict individual power.

6. An apocalyptic cycle resulted in which anarchical conditions led to violence, the use of food as a weapon, which intensified violence and accelerated fragmentation.

7. With UNOSOM I, the international community responded slowly and alienated the local population. The UN entered the factional conflict as it sided with one faction and was challenged by another.

8. An early attempt to marginalize the factions failed. To succeed it needed to be more comprehensive and supported by the UN Secretary-General.

9. The UN applied the conventional, inter-state peacekeeping instrument to a challenging, internal conflict that required much more. A diplomatic approach cannot respond to intensely social and political conditions.

10. Worsening famine conditions led to the enforced delivery of food by UNITAF. A dispute erupted between the US and the UN Secretary-General regarding the scope of the operation. By its military design, it could only fulfil limited goals.

11. The entire process in Somalia came to be dominated by military imperatives. This would prove unsuitable for responding to long-term social and political ground requirements.

12. The logic of UNOSOM II was flawed as it prepared to “assist” a non-existent centre. The UN would have to exercise more authority than it was prepared to, and did so ineffectively as a result.

13. A wide range of political tasks were assumed, but conceived in a diplomatic and military manner.

14. There was failure to appreciate the political dimensions of the new mission.

15. This led inevitably to an atmosphere of confrontation with local factions.

16. A seamless transfer of authority from UNITAF to UNOSOM II created a vacuum. More tasks to be done with fewer resources gave the impression of a failing operation rather than the start of a dynamic new effort.

17. There were too few civilians in the operation to develop a political operational concept that could fulfil the mandated tasks. Instead, a military plan was relied on almost exclusively.

18. The loose relationship between national contingents in a US-led coalition was inherited by a UN Force Commander. The challenging environment and independence of contingents fragmented UNOSOM II.

19. Contingents disagreed on rules of engagement, causing incoherence in the purpose and functioning of the mission.

20. UNOSOM II needed to establish a political authority locally, but its domination by military imperatives led to a combat phase. Military failure was followed by too little reconstitution of authority being done too late to be sustainable.

Figure One: Administrative Map of Somalia (not included)


Preface

Peacekeepers never retreat, they only ever withdraw. Yet the March 1995 departure of UN forces from Somalia resembled more the former than the latter. In the early morning of 1 February, Pakistani troops abandoned the UN compound, formerly the site of the US embassy, which is now to be returned to the US after the UN spent $160 million on it in the last two years. Within moments, hundreds of looters and armed men loyal to the mission’s nemesis, General Mohammed Farah Aidid, swarmed into the walled enclosure and stripped it unceremoniously of every piece of wood and wire, every movable object and removable part of the five permanent buildings, including their light fixtures and chairs.

Tension rose at the airport, the final point of departure and the site of the landing of the initial 500 Pakistani troops in September-October 1992. Just as they had been veritably imprisoned by a hostile environment then, the last UN troops too were kept in a corner by aggressive militia, with only one way to move: out of Somalia. An international operation with US marines mounted to secure the UN withdrawal was one of the final acts of the international community in Somalia. The Financial Times dubbed it “Operation Abandon Hope” [1]. It leaves behind the supremacy of the very warlords it had hoped to tame, unchallenged by new institutions or alternatives to leadership for the Somali people. While mass starvation was finally ended with the arrival of US troops in December 1992, the prospect of continued anarchical conflict threatens to invoke the spectre of famine again. The impact on the country has not been positively sustaining, with the fate unknown of those Somalis who had remained loyal to the UN. What was perceived as the “easy option” proved impossible given the manner with which it was approached.

However, the understanding of the reasons for this is being lost or forgotten. As much printed material as there is available on Somalia, in UN reports, press coverage, and the cottage industry of secondary source publications in academic and practitioners’ journals, only a limited amount has reflected the internal strengths and pitfalls of US and UN operations in the field. Much of the UN’s documentary material never reached headquarters in New York. It was quite a closed operation, certainly by the critical stage of combat operations in summer 1993. The net result is that the wrong lessons have been learnt from what has turned out to be the watershed experience of the second generation of UN operations over the last six years.

United Nations secretariat officials, including the Secretary-General, permanent representatives at the Security Council, the United States diplomatic, military and political establishment, and politicians world-wide have opposing, simplistic and largely mistaken interpretations about what works and what does not in fostering peace. This report intends to dispel conclusions that are hardening rapidly as stereotypic misconceptions, and which are charting a skewed course both for UN operations and for the organization itself. It assesses the essential problems in Somalia, the deep-rooted crisis of anarchy and the apocalyptic conditions this spawned. In this context, it considers the blunt instruments used by the international community to respond to the crisis, such as inter-state, diplomatic peacekeeping between warring factions and intensive military force. It concludes that far from encouraging retreat from UN operations as a whole, the Somalia experience indicated, rather like a laboratory might in which different tools were tested against the same problem, what the missing link in UN operations is, namely a political capability in the field. It argues that the development of this should be pursued vigorously if so-called “warlords” are to be checked effectively elsewhere.

The Norwegian Institute of International Affairs (NUPI) dispatched a research mission to the operational area in February 1994, at the time Western nations, including the United States, were withdrawing from the field. The team was composed of: Åge Eknes, Director of NUPI’s programme on United Nations peacekeeping and multinational operations; Toralv Nordbø, Military Advisor at the Permanent Mission of Norway to the United Nations; and Jarat Chopra, Research Associate and Lecturer in international law at the Thomas J. Watson Jr. Institute for International Studies, Brown University, and Visiting Researcher at NUPI. Members of the team have also concluded a number of interviews with officials in New York, Washington, London and elsewhere involved with the missions to Somalia.


Glossary

AID (US) Agency for International Development
APC Armoured Personnel Carrier
CCPO (UNOSOM II) Chief Civilian Personnel Officer
AOR (UNOSOM II) Areas of Responsibility
AWSS (UNITAF/UNOSOM II) Authorized Weapons Storage
CMIO (UNOSOM II) Chief Military Information Officer
CMPO (UNOSOM II) Chief Military Personnel Officer
COO (UNOSOM II) Chief Operations Officer
COS Chief of Staff
DC (UNOSOM II) District Councils
DFC Deputy Force Commander
DPKO (UN) Department of Peace-Keeping Operations
FC Force Commander
FPM (UNOSOM II) Force Provost Marshal
HRS (UNITAF) Humanitarian Relief Sectors
ICRC International Committee of the Red Cross
JMCC (UNOSOM II) Joint Military-Civilian Committee
LAS League of Arab States
MSC (UN) Military Staff Committee
NATO North Atlantic Treaty Organization
NSC (US) National Security Council
NGO Non-Governmental Organization
OAU Organization of African Unity
ODA (US) Office of Disaster Assistance
OIC Organization of the Islamic Conference
OPCON (US/UNOSOM II) Operational Control
OPLAN 1 (UNOSOM II) First Operational Plan
OPLAN 2 (UNOSOM II) Second Operational Plan
ORCI (UN) Office for Research and the Collection of Information
QRF (US/UNOSOM II) Quick Reaction Force
RC (UNOSOM II) Regional Councils
ROE Rules of Engagement
SDM Somali Democratic Movement
SNA Somali National Alliance
SNF Somali National Front
SNL Somali National League
SNM Somali National Movement
SPM Somali Patriotic Movement
SRC Somali Revolutionary Council
SRSG Special Representative of the (UN) Secretary-General
SSDF Somali Salvation Democratic Front
SSNM Southern Somali National Movement
SYL Somali Youth League
TACON (US/UNOSOM II) Tactical Control
UAE United Arab Emirates
UN United Nations
UNITAF Unified Task Force
UNOSOM I First United Nations Operation in Somalia
UNOSOM II Second United Nations Operation in Somalia
USC United Somali Congress
WFP (UN) World Food Programme


I. Peace-Enforcement

The eminent historian of international relations, F.H. Hinsley, concluded in his classic survey of plans and proposals throughout the ages for securing and maintaining peace [2] that the greatest experiments have followed in the wake of the worst wars. In the wake of the Cold War, the international system has conducted a number of experiments in an effort to create an instrument that can secure peace and create the conditions for its maintenance. However, unlike the privilege of natural scientists, there is not an international laboratory, or a controlled environment in which to test propositions and theories, assess their consequences and record the lessons of success and failure. Instead of the scientific method, the method of trial and error has dictated progress in the international system.

United Nations (UN) operations established to address age-old scourges of humankind – war, famine, death – are costly ventures subject not to rational logic and field requirements, but the vicissitudes of inter-governmental politics and a bureaucratic sub-culture. Unremarkably, these blunt instruments have not developed an adequate means of scientifically or even honestly gathering the results of trials undergone in the field or of errors made in rapidly evolving social conditions, into which UN operations have increasingly deployed. Some wheels have been repeatedly reinvented while others remain to be conceived.

In this haphazard manner, the international response to the scourges afflicting Somalia was regarded by UN Secretary-General Boutros Boutros-Ghali, public commentators and practitioners in the field as an experiment in “peace-enforcement.” However, this was an ill-defined concept, interpreted literally and confused therefore with the much broader concept of “enforcement” under Article 42 of the UN Charter.

The use of armed force was considered the ultimate sanction of the United Nations system and in the event that non-military sanctions proved inadequate Article 42 provided for “action by air, sea, or land forces as may be necessary to maintain or restore international peace and security”. The drafters of the Charter outlined a scheme for the provision and direction of such forces in articles 43-50. UN member states, particularly the powerful permanent members of the Security Council – the United States, the Soviet Union, France, Britain and China – would agree to make available on call to the Security Council armed forces and facilities for the successful conduct of international military operations. These permanently available forces would be under the strategic direction of a Military Staff Committee (MSC) composed of the defence Chiefs-of-Staff of the Security Council permanent members. The MSC would be the vital link between the legal legitimacy and political authority of the Security Council and the effectiveness of operations in the field, by translating broad political decisions into specific military instructions.

In this system of “collective security” the term “collective” had a special meaning. It did not refer merely to a group of states cooperating as a coalition in a particular venture, but to the subordination of military assets to a central authority in advance of any crisis. This would establish a degree of objectivity, automaticity and consistency in responding to crises – the ideal of any system of law and order. The UN’s version of collective security was intended as an instrument to halt and reverse acts of aggression by a recalcitrant member of the international community. In the pursuit of establishing such a system, the ideal was pitted as an opposite to the prevailing anarchical system of balances of power. With conceptual attention focused on the two opposites, the span of activity within collective security at one end of the pole was not adequately distinguished.

Consequently, the tools available to any system of collective security were black and white: non-military action or the unlimited, and therefore uncontrollable, use of force, as in traditional warfare. However, international law did not regard the international aggressor as an “enemy” and although since the Nuremberg Tribunal individuals could be held criminally responsible for crimes against peace, an aggressor state is not considered criminal in the manner that a murderer is in national legal systems. The sovereignty of states renders the crime, not the perpetrator, the enemy. Enforcement actions, therefore, were restricted in how they halted and reversed aggression. The use of force was more restricted than it was under the laws of war limiting traditional warfare because a “state of war” is unlawful in the first place and once entered, the laws of war are intended to prevent at best excesses of force. But enforcement actions, derived from international authority, were to use only the amount of force proportional to the force used in the initial violation and not more than strictly required by operational necessity. To match these legal requirements, a specific military doctrine was necessary to distinguish enforcing international rules of law from subduing an opponent in pursuit of national victory. Such a concept was not available to the international community in 1945.

As a result of the conceptual history it inherited, the UN Charter outlined only black and white options for responding to aggression: a span of non-military activities and sanctions on the one hand and military action if these failed on the other. The only instructions on how to use force were the broad powers under Article 42 and the customary restrictions of proportionality. Since the Charter was designed to respond to the massive force used in inter-state aggression, responses to aggression, even if limited by proportionality, were also conceived on a grand scale and certainly had not anticipated the more limited and sophisticated needs of responding to intra-state threats to the peace. In any case, the genuinely “collective” use of force was not tested since the ideal scheme of the UN was stillborn as its drafters witnessed an increasing polarization between the United States and the Soviet Union. As its name suggested, for the United Nations to function it had to rely on the kind of cooperation that existed between the Allied nations of the Second World War. Instead, it fell victim to the political stalemate of the Cold War: the two superpowers could not agree on the numbers of forces that should be placed at the disposal of the UN and consequently the idea was left to atrophy.

Although the MSC was established, it has functioned only pro forma. There has not been an adequate international strategic directorate to translate international political decisions into collective military action. While such a capacity existed in national ministries and departments of defence, these did not have an international mandate and had to guarantee national interests, not develop objective UN strategies. Therefore, when time came to employ armed forces collectively, there would be no one in the UN system whose job it was to adapt military instruments from warfare to enforcement.

This test had to wait four decades and for the Cold War to run its course. There was a mistaken belief that Article 42 action could be carried out only by the instrument envisioned in Chapter VII of the Charter. But Article 42 does not state this and action could be undertaken by a force established under the implied powers of other Charter provisions. Nevertheless, Article 42 has never been relied on explicitly as a legal basis for UN action: armed force in Korea in the 1950s was based on “recommendations” to member states under Article 39 of the Charter; the Security Council authorization permitting Great Britain the use of force to interdict the passage of oil into Southern Rhodesia in 1966 was a kind of Article 41 1/2, or the use of force to underwrite economic sanctions; and the 1991 Persian Gulf War was based ultimately on Chapter VII as a whole and used the language of Article 42 without expressly referring to it.

In all of these cases, since there was no collective instrument to implement United Nations decisions, “enforcement” relied on the willingness of individual states to act independently or participate in loose-knit cooperative arrangements or coalitions. A single nation dominated each action and at no time was there a high-intensity enforcement operation functionally integrated as part of the UN system. Consequently, a commonly accepted approach to enforcement was not developed and in the rare cases of UN Security Council resolutions authorizing national action, member states conducted warfare as they knew how, against North Korea and Iraq.

In principle, an operation based on a Security Council resolution sub-contracting specific tasks to a state or group of states may still be regarded a collective action provided the mission is accountable and responsible to the direction of the UN. It is more important for the political direction of an operation to be collectively determined than for armed forces to be collectively commanded. In Korea, the US command was not as directly responsible to the UN as it should have been, but the use of force was limited in any case by the political restrictiveness of the Cold War. In the Gulf, the use of force was not limited in this manner, by a Cold War check or balance, nor was it by the rules of proportionality inherent in enforcement action, and with a carte blanche Security Council resolution arrived at by diplomatic pressure, the United States conducted unfettered military operations.

These were subsequently criticized [3]. Nations which voted in favour of Security Council Resolution 678 authorizing “all necessary means” had not expected the degree of force that would be used nor the interpretation of “all necessary” to mean “unlimited.” There were charges that the UN had been hijacked and the Security Council taken advantage of unlawfully. The US Pentagon on the other hand found the Gulf to be a model which fitted the established US war-fighting strategy embodied in the so-called “Weinberger principles,” which outlined the criteria for US engagement: the ability to use overwhelming force against a clearly identifiable enemy. Furthermore, the Pentagon felt that “coalition warfare” was “the way to go” for America’s future military engagements. Others felt that the Gulf War was an aberration and its scale was unlikely to be repeated soon, that there was a great deal of other activity that needed attention.

While this operational debate continued, the Gulf War turned out to be effectively a turning point. Control over the actions of nations sub-contracted by necessity was to be tightened. The distance between UN resolutions and operations in the field had grown too great, rendering UN decisions mere fig-leaves for ulterior motives of member states. So when the United States offered troops and leadership of another coalition in November 1992 for Somalia, a Security Council and Secretary-General without alternatives gratefully accepted but with reservations. They clarified and limited the mandate and tried to ensure its direct involvement in decisions made by qualifying “all necessary means” in Security Council Resolution 794 of 3 December 1992, and through reporting and joint consultative procedures. In this manner there was a convergence of international direction and sub-contracted great power assets into some kind of joint operation that crystallized with the need of the US to withdraw in the spring of 1993 and the attempt to unite UN contingents and US coalition forces in a single operation. The overall experience of Somalia would eventually lead by the summer of 1994 to a complete divergence of this process and a return to, more robustly than ever, great power military operations with UN fig-leaf resolutions authorizing such actions.

Certainly by early 1993 there was emerging in military circles the mistaken distinction drawn between “UN operations” and coalition operations with UN resolutions. In the former, it was argued, national forces assisted in the implementation of UN political decisions. In the latter military imperatives determined the nature of the response first and only then was a resolution to be sought for approval. This distinction was operationally true for historical reasons, because the UN had only commanded simple peacekeeping operations. But it could not be sustained in law: an operation, whether under the secretariat’s command and control or sub-contracted to a coalition, is either a UN operation or it is not. Operations mandated by UN resolutions should fly the blue flag and not a gray combination of national flags. The blue flag flew in Korea and not over coalition forces in Somalia, yet the former was less under UN direction than the latter.

The mandate of the Second United Nations Operation in Somalia (UNOSOM II) was not “enforcement” in the manner of reversing the acts of an international state aggressor. Instead, it was perceived to be “peace-enforcement,” which was something less, but it was not clear what. The term “peace-enforcement” had a very specific origin. It appeared in June 1992 for the first time in paragraph 44 of Boutros-Ghali’s An Agenda for Peace. This was a proposal more for a mechanism than a concept, namely “peace-enforcement units.” It distinguished these from the forces envisioned in Article 43 of the Charter, which were to respond to international aggression, and recognized that the black and white options of non-military sanctions and the massive use of force were inadequate. The UN, it said, had been called on to restore cease-fires that had broken down and this required something more than what the secretariat had at its disposal.

The specifics of paragraph 44 referred to the mechanism of the units. They would be available on call, undergo advance training, and be deployed under the authority of the Security Council and commanded by the Secretary-General. The deployment of units was to be considered provisional measures under Article 40 of the Charter. It did not elaborate further on the meaning of “peace-enforcement.”

The concept was not well-defined partly because it emerged from a debate that was just maturing in early 1992. Despite the dramatic revival of enforcement possibilities, by far the greater activity of the UN was in the expanding area of peacekeeping. The concept of peacekeeping had been developed by the UN in response to the restrictive political atmosphere of the Cold War. There had been need for an instrument that was something more than the pacific settlement of disputes outlined in Chapter VI of the Charter, but which did not encroach on the enforcement mechanisms of Chapter VII. This led to the narrowly constrained practice of inserting UN forces between belligerent armies that had agreed to halt hostilities and wanted a symbolic guarantor of their armistice. Peacekeepers were referees with limited physical means.

The experience of peacekeeping developed several basic assumptions:

  1. a Force had to operate with the full confidence and backing of the Security Council;
  2. a Force operated only with the full consent and cooperation of the parties in conflict;
  3. command and control of the Force would be vested in the Secretary-General and Force Commander, under the authority of the Security Council;
  4. the composition of the Force would represent a wide geographic spectrum, although conventionally excluding the permanent members of the Security Council, and contingents were supplied voluntarily by member states upon the Secretary-General’s request;
  5. force would be used only in self-defence, although self-defence included defence of the mandate as well as the peacekeeper; and
  6. the Force would operate with complete impartiality.

With the end of the Cold War, the United Nations entered its second generation as an institution. There had not been an operation deployed for a decade after the 1978 establishment of a mission in Lebanon. But then, diplomatic agreement was reached on a number of outstanding conflicts – in Afghanistan, Iran/Iraq, Namibia, Central America and elsewhere – and the UN secretariat was requested to guarantee each negotiated settlement. Although there was limited success, peace agreements increased in number and complexity, and outstripped the capacity of the UN to successfully underwrite the terms belligerents had agreed to. The peacekeeping formula continued to be applied to non-peacekeeping tasks and operations in the field began to dysfunction, particularly during their deployment phases, first in Namibia, then in Western Sahara, the former Yugoslavia, Cambodia and Angola.

In early 1992, some observers recognized that UN operations had developed beyond the conventional parameters of traditional peacekeeping. They called for a “third option” a second generation of operations that would occupy the “middle ground” between peacekeeping and large-scale enforcement. The principles of peacekeeping were necessarily challenged. Consent of the parties in conflict was not always forthcoming or sometimes withdrawn once given. In response UN operations would have to consider using force. This would require sufficient assets to do so and the inclusion in operations of armed forces from Security Council permanent members. In turn, the issue of command and control of assets would become more acute. Impartiality would be defined as the objectivity with which the mandate was executed rather than the degree of submission to the will of the parties in conflict. This would give the Security Council an even more crucial role in developing clear mandates, continually supporting the force in the field with the solidarity of its will and expanding mandates as changes in ground conditions required. These developments reflected that a new set of tasks had been assumed by the UN and required an operationally-driven set of categories that did not fit neatly into the existing Charter provisions of Chapter VI and VII.

“Peace-enforcement” had its roots in this middle ground, but its ill-definition in An Agenda for Peace led to its confusion with large-scale, conventionally conceived enforcement. This was symptomatic of conflicting approaches to the development of doctrine for UN operations. The Agenda was a politically drafted document and not operationally driven, so it referred to vague “peace” terms, in addition to peacekeeping and peace-enforcement, such as peace making and peace building, and not to specific tasks. Therefore, despite the Security Council check and balance on the US-led coalition, the Unified Task Force (UNITAF), and the UN’s control of UNOSOM II, the domination of the US throughout the process and the lack of a specific conceptual doctrine within the UN obscured the effective distinction between “enforcement” and “peace-enforcement.”

This set the stage for a flawed and uncontrollable experiment in Somalia of historic significance. It was not to be a genuine test of the “third option” or of anything that could be labelled “peace-enforcement.” Rather, in the absence of a “peace-enforcement” doctrine, operations shifted between the black and white options of no force or too much force. Yet the wrong lessons have been taken away about the middle ground on the pretext that the use of force has failed the test.

The United States, the United Nations machinery and the UN Secretary-General all had different perspectives on the intensity operations in Somalia were to have. The United States brought to the field not only its Gulf War experience, its coalition model and “new world order” peace-fighting strategy, but in fact Central Command was responsible for both missions and many of the same troops and officers were deployed from the Gulf to Somalia. United Nations officials in the field, despite having assumed a vastly different array of tasks, were still rooted in the traditional peacekeeping school and carried this baggage in an overwhelmingly hostile environment. The small number of civilian staff under such conditions led to a virtually paralytic scope of non-military activity.

The Secretary-General, somewhere between these two extremes conceptually, effectively pushed for the first UN enforcement operation, without any concept of operations for this other than his mere reference to “peace-enforcement.” He later admitted that An Agenda for Peace had only “touched on peace-enforcement” [4]. Yet he wanted the experiment to be as extensive as possible. But this combination of perspectives, none of which were ideal for the host of problems in the Somali environment, was sure to fracture and be overwhelmed by false conclusions.

United Nations officials were left feeling that the use of military force should be avoided in peace operations, since the degree of destruction in Somalia was not matched by achievement of many objectives. The United States concluded that failure was due to the fact that not enough force was used. And the Secretary-General returned to black and white options, the very conditions that had led to the need for a third option.

He stated in his “Supplement to An Agenda for Peace” that “Peace-keeping and the use of force…should be seen as alternative techniques and not as adjacent points on a continuum” [5]. He falsely concluded that the most successful operations in recent years respected the traditional peacekeeping principles of consent of the parties, impartiality and the non-use of force except in self-defence. In fact, the principal cause of failures in the last five years had been the application of the peacekeeping model to non-peacekeeping conditions. Even within the “Supplement” this reaffirmation of traditional peacekeeping stands in stark contrast to the kinds of “peacekeeping operations” described, [6] which are clearly beyond the kinds of conditions in which peacekeeping developed. That operations have been deployed on this basis unsuccessfully seems to have been forgotten.

Furthermore, the Secretary-General disregards the essentially political framework in which enforcement must be conducted. He states: “The logic of peace-keeping flows from political and military premises that are quite distinct from those of enforcement; and the dynamics of the latter are incompatible with the political process that peace-keeping is intended to facilitate.” [7] This is a confusion of political activity and diplomatic processes. It is as part of the latter that peacekeeping is traditionally deployed and participating in political processes requires something more if the UN’s involvement is going to succeed. The Secretary-General essentially divides the third option in two and relegates part of it back to expanded, but traditionally conceived peacekeeping, and the other to enforcement. He goes so far as to regard “peace-enforcement” as synonymous with high intensity enforcement of Article 42 of the Charter. In paragraph 23 he lists the “instruments for peace and security” amongst which he includes “peace-enforcement”; yet the section corresponding to this is entitled “Enforcement action” and describes the kind of massive force operations that characterized responses to inter-state aggression [8]. Anything other than peacekeeping or large-scale force is therefore obscured.

It is with large-scale force that the United States is most comfortable. In the wake of its experience in Somalia, the year-long inter-departmental debate in Washington about the criteria for engagement in UN operations crystallized in spring 1994, one month after the US withdrawal from Mogadishu. On 3 May, President Bill Clinton signed Presidential Decision Directive 25, the “U.S. Policy Guidance on Reforming Multilateral Peace Operations”. This effectively applied the Weinberger principles to peacekeeping. US participation depended on the potential for results in a period of limited duration. It was a “quick-fix” concept requiring clear conditions, a free hand to operate in the field and an exit strategy.

These were military imperatives dominating the formula that would sever political solutions from unrestricted military operations conducted in a veritable vacuum. It amounted to the idea that once the military were deployed the political role had concluded until the military job was done. This was the development of a catch-all sledge hammer which had proven to be dysfunctional again and again in the essentially social and political environment of internal conflicts. It was the transformation of military imperatives into a political strategy and foreign policy and inevitably a recipe for disaster.

This has since been exacerbated by partisan party politics and the introduction of legislation into the US Senate and House of Representatives in January 1995 as part of the Republicans’ “Contract with America”. This requires the President to certify to Congress that:

  1. the US participation in a UN operation is necessary to protect US national security;
  2. the commander of the unit retains the right to report independently to superior US military authorities, and can decline orders judged to be illegal, imprudent or beyond the mandate agreed upon by the US until superiors override this;
  3. US forces remain under US administrative command for discipline and evaluation; and
  4. the US retains its authority to withdraw.

It also imposes destructive financial restrictions on US-UN peacekeeping relations.

The skewed US and UN reactions to Somalia suggest the wrong lessons have been learnt. Somalia illustrates that the missing piece of the puzzle of UN operations is the need for a political capability. The severing of enforcement from a political process by the Secretary-General and military imperatives from political direction by the US will exacerbate the fog of peace missions. Somalia emphasized the need for precisely the opposite: the subordination of diplomatic, military, humanitarian and civilian tools to a political framework that could function in combination, not merely in coordination, with the local social and political process and the daily evolution of local conditions.

The phases of international involvement in Somalia roughly correspond to the three eras of UN operations generally. The First United Nations Operation in Somalia (UNOSOM I) that deployed gradually throughout 1992 was based on the traditional peacekeeping concept. This was the diplomatic use of troops. The military force employed during UNITAF in late 1992 and early 1993 reflected the attempt to enhance the military competence of the UN in its second generation. UNOSOM II after spring 1993, albeit shifting between these two models, needed a strong political capability to succeed. The first two are blunt instruments and have their place. By themselves, neither could have responded to the conditions in Somalia. Aspects of both could have been employed as part of a comprehensive strategy to respond to the social and political demands of Somali anarchy. The lessons from UNOSOM II need to be reassessed in this context if the Somali experience is to be instructional and not destructive. The “middle ground” debate about limited use of force needs to be less about military enforcement than developing a policing capability that can enhance the political competence of UN operations in a third era. This can ensure that diplomatic and military tools are balanced within an overall political framework in the unavoidable Somalias of the future.

II. Anarchy and the Apocalypse

1. Clans and Factions

Anarchy, it could be said, was a principal characteristic of Somali social organization before the colonial powers’ scramble for Africa reached the Horn. That is, if anarchy is understood to mean the absence of a central government. Anarchy is not a vacuum, however: The alternative to modern, sovereign government may not be disorder, but an unfamiliar order; and even when the absence of centralized authority does result in disorder, there will be proliferation, not stagnation, in political fragmentation social and political factions.

In Somalia, there is a fundamental difference between the chaotic disorder and uncertainty that followed the collapse of the Somali Republic in January 1991 and the lack of centralized authority in pre-colonial times. The conventional view holds that “Somalia has dissolved into its traditional segmentary divisions” and rather than being anarchical, Somali society has returned to its traditional institutions of clans and sub-clans [9]. This does not explain, however, why

At no time in the recorded history of Somalia has nearly one-third to one-half of the population died or been in danger of perishing due to famine caused by civil war. This calamity surpasses all previous ones and can be appropriately called “Dad Cunkii”, the era of cannibalism. [10]

The successors to the Somali state were not clans and sub-clans, which were reverted to by the population as social institutions, but political factions. These armed groups, which commanded some resources and managed to control limited territorial areas, may have been divided along clan or sub-clan lines initially, but their perpetual fragmentation led to alliances of convenience that crossed clan lines. Established and led by what has been referred to in the Somali context as “warlords” and their cronies who galvanized support through clan allegiance, factions became the repositories of power by the force of arms. However, the evolution of factional violence did not proceed according to clan loyalty, but to the vicissitudes of uncontrolled warlords. This was only a recent phenomenon in Somali history. But why was there no check or balance in society to prevent the emergence of factions as they had been stopped in the past?

The pre-colonial Somali community extended beyond the geographical boundaries of modern Somalia to Djibouti, the Ogaden region of Ethiopia and the Northern Frontier District in Kenya. In this general area there was perhaps the most homogeneous ethnic community in a single place on the African continent. Islam had been commonly adhered to since the eighth century and almost the entire population today is Sunni Muslim. Somalis spoke the same Cushtic language and shared common customs and traditions and a single culture. Of the population estimated in 1992 to be 6.5 million, 98.8% were ethnically Somali and only 1.2% Arab or Asian. This included the 250,000 Bantu and Swahili-speaking groups settled along and between the Shebelle and Juba river valleys in southern Somalia.

Despite this homogeneity, there was no collective Somali consciousness. The overwhelming majority of Somalis were nomadic. The ideal of nomadic independence developed amongst the Somali majority, who then despised settled agricultural cultivation, the formal cooperation this required, and the hierarchy, authority and governmental organization that resulted. In fact, the all-pervasive Koranic jurisdiction was distinguished from any kind of centralized civil law, which was popularly referred to as “Al-Jabr” – “the tyranny” [11]. A traditional Somali proverb stated: “Me and Somalia against the world, Me and my clan against Somalia, Me and my family against the clan, and Me against the family.” [12]

Somalis did not identify with other ethnic Somalis regardless of origin, but differentiated themselves according to their patrilineal genealogical origins. Extended family groups formed by finding a common heritage in a common ancestor. These clans further divided into sub-clans and eventually immediate families. Families constituted the basic building block of society and the household the basic unit of production. With such complete fragmentation, and given inequalities between clans, sub-clans or families, even within the equalitarian nature of Somali society, and given the appetites of the strong or the weak that invariably feed on this inequality, there is not a reason to suggest why Somalis should not have descended into perpetual and internecine violence of the kind that characterized 1991.

Except that in pre-colonial times there prevailed a kind of pan-Somali code of conduct called the Xeer. This was “a set of rules and norms…socially constructed to safeguard security and social justice within and among Somali communities,” to which Muslim values were added as Islam spread throughout the area [13]. Clans and families could not muster enough resources and were not sufficiently centralized to exploit inequalities, and in any case the communitarian nature of Somali society meant that there were not classes stratified according to resources. As a social contract, the Xeer regulated conflicts “in the absence of centralised coercive machinery” through the generally accepted ethic of “the absolute necessity of relying and living on one’s labour/livestock rather than exploiting others”. It was a self-regulating system in which one was prevented from dominating others.

The source of individual authority, then, was not the capacity to control others, but the capability to ensure others were not controlled. Most individuals were involved in productive activity, and the effectiveness of one’s household management in this regard was a prerequisite for leadership. Leaders would be elected by majority votes of informal councils or assemblies known as the Shir composed of any member of a particular lineage. Shirs also made most political and judicial decisions, limiting the exercise of leadership powers. Decisions were implemented by committees of elders appointed for a specific purpose by the elected leaders. In this manner, the ideal of decentralization was guaranteed and checks and balances against exploitation were safeguarded.

Under colonial administration, a new set of relations developed. The Somali nomadic space was divided into five parts. To protect the trading links of its colony at Aden from other powers such as France, in 1886 Britain declared northern Somalia a protectorate (British Somaliland). In 1888, France established its own colony at Djibouti (French Somaliland). The Ogaden was ceded to Ethiopia in 1897 and later the Northern Frontier District became part of British colonial Kenya. Italy claimed southern Somalia in 1905 and consolidated its control by 1927 (Italian Somalia). Despite Somali indignation, political, social, and above all psychological or perceptual fragmentation meant there could not be any immediate concerted response to the arrival of European powers.

Colonial administration necessitated centralization of authority if a European minority was to govern an indigenous majority. This was not just the imposition of the centralizing character of sovereignty and the nation-state system that had originated on the European continent over several centuries, and therefore operated in a context of checks and balances on power which would be lost when exported. Colonial centralization necessarily challenged the traditional source of authority in the pastoral political system. It introduced into daily life commercialism and therefore the domination over the pastoral producers of non-producers – the merchant or the colonial state-enterprise – which claimed control of pastoral surplus production. The communitarian social order without resource-based classes “was superseded by an economy in which the competition for access to commodities, the consumption of objects beyond one’s productive capabilities, and the accumulation of wealth in the urban centres were paramount” [14]. The establishment of a class of Somalis not engaged in production but with means of authority in turn generated in the colonial state system a Western-educated and acculturated élite capable of participating in political life.

In the cities, the legitimacy of authority no longer relied on an individual’s productive capability. Commercialism and the predominance of a new culture which sought to concentrate capital through exploitation displaced the Xeer ethic of mutual non-exploitation. By this was discarded the fundamental check and balance on power without there being available a replacement: an independent state-system with foreign origins could rely neither on European nor indigenous restrictions on power. Clan assemblies and committees of elders no longer made sense as a political system; they continued as social institutions in the face of state centralization.

The focus of Somali concerns centred, not on Somali exploitation of Somalis, but on European exploitation of Somalis. There was the natural desire of an indigenous people to expel foreign powers, but there was also the desire to attain the resources and exercise of authority of the colonial administrators. This was one consequence of the Xeer’s displacement and the absence of new rules of political behaviour distinguishing, if not separating, commercial and political enterprises. In fact, resisting colonial domination meant mustering the resources to do so, competing for control of produced surplus in the manner of the colonial master, and in so doing accelerating the replacement of Somalis’ own pastoral nomadic culture. Most significantly, developing the means to challenge colonial authority meant Somalis had available the tools to fight one another. This dissolved the Xeer glue that regulated conflicts and held Somali society together in some kind of anarchical order. It laid the roots of future chaos when the independent Somali state finally collapsed.

Resisting colonial administration displaced the primacy in Somali society of the clans; divided as they were it was easy for them to be ruled. Islam was a unifying force and leaders based their cause on the concept of jihad against the “infidels”. There developed a kind of proto-nationalism in which the concept of clan was extended to all Somalis. This pan-Somali identity was linked to territory in the geographic nationalism that emerged after the Second World War. The nation of the Somali community being tied to territory, the notion of settlement contradicted a nomadic past. It was another feature that at once was a centralizing tendency and served to fracture traditional institutions. Order now relied almost exclusively on the capacity of the state structure to take root and govern effectively.

This was not to be. In 1960, British Somaliland and Italian Somalia gained independence and united to become the new Somali Republic. This comprised only two of the five parts of the Somali ethnic community. Since this was now identified with a geographical area, the notion of a “Greater Somalia” developed, including the Somali communities in Djibouti, Ethiopia and Kenya as legitimate parts of the Republic. Foreign policy in independent Somalia aimed in part to “liberate” Somalis in adjoining states and the five-pointed star on Somalia’s national flag which symbolized the colonial division was a reminder that they could be united. In the absence of a Xeer to check or balance the exercise of government authority, the will to dominate extended beyond the borders of the Republic. Secessionist movements were supported in the Ogaden and the Northern Frontier District. Somalia fought violent wars with Ethiopia in 1961, 1964 and 1977-78. Hostility between the two continued throughout the 1980s. A war with Kenya lasted from 1963 to 1967.

Internally, unchecked exercise of political authority and imbalanced economic exploitation had its consequences. Despite the inheritance of democracy, the multi-party parliamentary state was supported by a weak civil society with fragmented institutions that had not been reconstituted to sustain representative government. The economy was stagnant: there was intense competition among élites for resources while there was not enough surplus to be reinvested to expand the productive base. Instead, resources were obtained through overseas loans. This “made the state the most lucrative source of funds. It was the competition among the elite for these resources that ultimately led to the degeneration of the major political parties and the demise of parliamentary governance” [15]. The number of political parties and candidates for parliamentary seats increased dramatically between 1964 and 1969, since the best access to state resources was as an elected official, and, once elected, officials had the resources to get re-elected.

These were the conditions that led to dictatorship. The ruling clique was drawn from the two organizations that had agitated for independence, the Somali Youth League (SYL) and the Somali National League (SNL), established in 1945 and 1947, respectively. It was able to consolidate its control over state resources. However, since even state resources were limited, and since the parliamentary system enabled a majority of members denied their share to defeat the Government, “it was imperative that important changes in the political process be brought about to save the system from consuming itself. In essence this meant shedding the “democratic,” if not the electoral, paraphernalia, and imposing a petit-bourgeois dictatorship on the Somali society that required the support of the army.” [16]

However, after the assassination in October 1969 of the President, Abdulrashid Al Sharmarke, who had been Somalia’s first Prime Minister, the army stepped into the confusion and established a military government, the Supreme Revolutionary Council (SRC). Under the chairmanship of the army’s senior officer, Major-General Muhammed Siyad Barre, it dismissed Parliament, banned political parties, canceled the 1960 Constitution, renamed the country the Somali Democratic Republic and adopted an ideology of “scientific socialism.” In 1976, pressured by his Soviet patrons, Barre established the Somali Revolutionary Socialist Party and proclaimed himself Secretary-General.

Under Barre’s dictatorship, the Somali state came to an end [17]. The SRC was initially welcomed by the public as a means of ending the scramble for resources in government, but it soon became apparent that Barre was the greatest looter of all. The regime lost credibility following the 1977-78 defeat in the Ogaden. The ideology of scientific socialism was discredited following the Soviet departure. A new competition for power amongst the elite and members of government was triggered, but this time armed force replaced electoral manipulation. Barre responded by rewarding those loyal to him with more and more state resources, and any opposition was persecuted and violently punished. Somali nationalism and the need for popular legitimacy were discarded. Instead, to replace those he purged from his administration and the military, Barre sought loyalists in his immediate family, his Marehan sub-clan and Darod clan. State institutions became instruments of Barre’s executive and were no longer Somali institutions that could survive a change of power: the remnants of the state either followed Barre out of Mogadishu or were picked apart after his downfall.

Since the face of Barre’s rule came to be associated with his clan, opposition to his power was also organized along clan lines. But this was not the re-emergence of the Xeer-bound clan led by the Shir, its chiefs and the elders. The fact of lineage had not been altered by the displacement of clans during the struggle for independence. So, in the 1964 and 1969 elections, candidates motivated by profit and not ideology were distinguished from one another by their clan affiliation; and in turn, voters were mobilized on this basis. But the candidates had not emerged from the leadership process of the clan; they were members of the surplus-controlling elites who now employed lineage connections for electoral gain. In this manner, the clan was becoming a basis for individually-led, power-motivated factions.

Barre’s own regime fed on itself and increasingly fragmented as independent centres of power emerged along sub-clan and family lines. Public law and order had long since collapsed; Barre’s cabinet functioned as a hierarchy stratified along clan lines and by degrees of loyalty; and finally, one of Barre’s main pillars of power, the military, disintegrated into clan militias, dividing and ruling itself in the manner it had done in the rest of the country for two decades. As the only defence against Barre’s regime and in the institutional collapse that followed, people turned to clan connections as a support mechanism. Armed factions managed to exploit this and create constituencies for their political and personal agendas.

Opposition factions emerged gradually, then proliferated until Barre’s collapse. Following Somalia’s defeat in the Ogaden in 1977-78, a group of officers mostly from the Majerteen sub-clan of the Darod, that had been denied access to central power, led a failed coup attempt in April 1978. They subsequently established the Somali Salvation Democratic Front (SSDF) in 1981, which was given sanctuary in Ethiopia and operated in central Somalia. In the north, populated by the Issaq clan, the Somali National Movement (SNM) was established in 1981 to challenge the supremacy of the south. The south had dominated political life since independence: the first three prime ministers were Darods, and while the fourth was an Issaq, he was overthrown by Barre, another Darod. They operated in the north with Ethiopian support and in 1988 eventually attacked Hargeysa, the regional capital, which was then razed to the ground by Barre in response. In 1989, the Hawiye of central Somalia, the largest clan, established the United Somali Congress (USC). Although it emerged late in the process, it managed eventually to overthrow Barre in 1991. By this time, there were more than 15 armed factions operating in Somalia, nominally representing parts of all six Somali clans.

The loose association between faction and clan was second to political expedience. Factions such as the USC and SNM were loosely based on clans, others on sub-clans, such as the Somali Patriotic Movement (SPM) established in 1989 to represent the Ogadeni sub-clan of the Darods. Sometimes clans spawned more than one faction other than along sub-clan lines: the Somali Democratic Movement (SDM) of the Rahanwin clan split into two sub-factions in early 1992 and supported opposing sides in the post-Barre civil war. Sometimes sub-clans split into splinter factions: the SPM of the Ogadeni’s split to support other opposing sides in the war. Furthermore, despite factional identification with clans, many factions were composed of individuals from a variety of clans and sub-clans: there were Hawiye guerrillas who had operated with the Issaq SNM in central Somalia, for instance.

Most significantly, alliances emerged between factions, sub-factions, clans and sub-clans and splinters of each. For instance, the USC split in 1991 between the Abgal sub-clan led by Ali Mahdi Mohamed and Aidid’s Habr-Gedir. Both allied themselves with other clans and factions, and in 1992 Aidid formally established the Somali National Alliance (SNA). It included Aidid’s Habr-Gedir USC, a sub-faction of the Rahanwin SDM, a splinter of the Ogadeni SPM, and the Southern Somali National Movement (SSNM), a southern coastal faction. While Mahdi was courting the SSDF and the pro-Barre Somali National Front (SNF), his own sub-clan split into three factions leaving him a narrow base of support.

By 1992 the four strongest warlords controlled major centres. Aidid and Ali Mahdi divided Mogadishu between them. Bardera was controlled by General Mohamed Said Hersi “Morgan,” a son-in-law of Siad Barre, his commander of the former Somali army and the leader of the SNF. Kismayu was controlled by Colonel Ahmed Omar Jess and his SPM. All four had been personally responsible for large-scale massacres. Mahdi and Aidid had fought each other brutally throughout 1991 and 1992 and continued to order killings throughout the UN’s presence in Somalia. Morgan had been responsible for razing the city of Hargeysa in Barre’s war against the SNM in the north. As late as 8-10 December 1992, Jess massacred some 100-200 civilians he felt were a threat before UNITAF forces could reach Kismayo [18].

The mob had replaced tyranny and it could be said, not that Somalis had returned to the anarchical order of the clan, but that factional warlords had unravelled the last shreds of the Somali social fabric. Without the Xeer’s check and balance, the collapsed state meant that transferred to the warlord was the appetite of sovereign power, unrestricted by political institutions or a social order. The individual and the faction were free to dominate all other Somalis. The opposition had not intended to liberate Somalia from Barre’s dictatorship, but to replace him with themselves. They could agree only on the removal of Barre and nothing else beyond. The factions were poised for mutual destruction.

Figure Two: Clan lineage of Somalia (not included)

2. An Apocalyptic Cycle

Somalia in 1992 resembled an apocalyptic image in which the four horsemen emerged in something of a vicious cycle. As a result of anarchy, war and famine and death plagued the country in an increasingly uncontrollable cycle. Anarchical conditions, and the inequalities inherent in this in the absence of the rule of law that transforms the position of the weak as against the strong, perpetuated civil war and caused in turn increasing fragmentation, which in its turn caused more competition and conflict. The destruction of the means of food production and the willful use of food as a weapon led to famine. The conditions of internecine violence that led to the famine also limited the access of international humanitarian assistance. This intensified the famine and exacerbated in turn the security conditions that led to it as fighting over power and fighting over food for survival became increasingly connected. Even the limited control over their members that factional warlords exerted at any given time could not prevent self-interested looting by individuals.

The roots of the cycle were the anarchical conditions and the proliferating, hydra-like warlords; the vast numbers of armaments and the famine were symptoms of this illness. Great theoretical tomes have been written on anarchy, but what does it look like? In the streets of Mogadishu it amounted to the biggest bully with the biggest stick exerting influence over a local area and its population, until being displaced by a bigger bully with a bigger stick or by several smaller bullies. Some individuals were organized enough to command factions as warlords. As it turned out, the warlord was best challenged by another warlord: local civilians were not strong enough, unless they too were willing to engage in a ruthless competition for power; and the international community did not have the finesse to displace him. These were the two attributes needed to counter the warlord: strength and finesse.

Responding to the warlords was the ultimate challenge of the international community and it should have been the basis of the overall framework in which a comprehensive approach disentangled the vicious cycle. Instead, there was a tendency to respond primarily to the symptoms of the problem, and to minimalize overwhelmingly the essentially political tasks or to exclude them in all but name. But it is also true that the symptoms were serious.

By 1991, there was no effective government organization in Somalia, centrally, regionally or locally. Central authority disappeared with the fall of Siad Barre in January 1991. Before then, local government authority was based on elected regional and district councils, but in which the central regime determined the candidature of officials. Each of the original 16 administrative regions of Somalia contained 3 to 6 districts, except the region around Mogadishu, which had 15 districts. There were a total of 84 districts. None were functioning in 1991.

The mechanisms of both oppression and law and order were either abolished in late 1990 or had collapsed by January 1991. The Western-style court system arranged in four tiers – with a Supreme Court, courts of appeal, regional courts, and district courts – ceased to function. Barre’s National Security Courts, under direct executive control, were abolished in October 1990. The well-respected Somali Police Force, the less accountable People’s Militia and Barre’s instrument of authority, the National Security Service, were all disbanded in January 1991.

Although the Somali National Army was disbanded in January 1991, as late as January 1992, at the height of the civil war, remnants of the army abandoned some 40,000 weapons. As a consequence of Cold War competition, the Somali military was at one time the best armed on the African continent, later being superseded by Ethiopia. The army ground forces were organized into 12 divisions of 4 tank brigades, 45 mechanized and infantry brigades, 4 commando brigades, 1 surface-to-air missile brigade, 3 field artillery brigades, 30 field battalions, and 1 air defence battalion. This was channeled into the clan militias that formed from the dissolution of the army. The Somali Air Force was composed of 3 fighter ground attack squadrons, 3 fighter squadrons, 1 counter-insurgency squadron, 1 transport squadron, and 1 helicopter squadron. These simply ceased to operate by 1992; the carcasses of the aircraft could be seen strewn about airfields throughout Somalia.

There were vast quantities of arms available. Both the Soviet Union and the United States had alternately provided aid to Somalia. There were also caches of arms that had been built up from open and black markets. There was a virulent cross-border trade, especially in ammunition. In addition to this, there was a militarily trained population: it had been a requirement under Barre of school and college graduates and civil servants.

Between November/December 1991 and March 1992, the war accounted for as many as 30,000 deaths and 27,000 wounded. In the same period, some 300,000 died of hunger or hunger-related causes. At the beginning of 1992, 3000 Somalis, mostly women, children and old men, were dying daily from starvation. By June 1992, 5000 people were dying each day, 1.5 million were on the brink of death and 4.5 million people were nearing starvation. With a UN-estimated population in 1991 of 7.7 million, these figures indicated all the people of Somalia were in a life-threatening position. By September 1992, 25% of all Somali children under the age of 5 had died.

The refugee flow to neighbouring countries grew to unmanageable proportions. The refugee camps in Kenya had a capacity of 130,000, yet by mid-1992 there were 300,000 individuals registered. They were arriving at the rate of 1000 per day, and by November 1992 this figure increased to 2800. It was expected that by 1993 there would be 500,000 refugees in Kenya. In mid 1992 there were 500,000 refugees in Ethiopia and by November over 1 million had crossed into Kenya and Ethiopia. There were also 65,000 refugees in Yemen, 15,000 in Djibouti and 100,000 in Europe in August 1992.

The internally displaced were no less numerous. While 100,000 were forced to flee from Mogadishu in December 1991, by November 1992 250,000 had moved into Mogadishu. Another 60,000 had gone to Kismayo and Baidoa. Some 250,000 were displaced in the northern region. This left small groups in the countryside isolated and vulnerable.

The collapse of the civil infrastructure was complete. There were no basic services such as electricity, communications, transportation, schools, health services, or available money to be used in the markets that always seem to survive such calamities. In March 1992, there were 500,000 people in Mogadishu without basic services. Of the estimated 70 hospitals in Somalia in 1988, only 15 remained partially operational and these were entirely dependent on external assistance. The hospitals had no water, electricity, drugs or any basic equipment. The collapse of the primary health care structure interrupted vaccination programmes and resulted in a dramatic increase in incidences of malaria, tuberculosis, tetanus, leprosy, parasitic and venereal infections, as well as skin and eye ailments.

Deteriorating sanitary conditions posed a major threat to public health. The streets were blocked with refuse. Rain threatened to exacerbate the problem. Potable water disappeared in the collapsed infrastructure. Large groups were at risk of epidemics and a number of people died.

The economy was devastated. Agriculture and livestock production were responsible for 2/3 of the country’s employment and 3/4 of its foreign exchange. By early 1992, 70% of the country’s livestock had been lost. In the absence of veterinary services and medicine, cattle were dying of disease by the thousands. War devastated farming areas, particularly cereal-producing locations, and some 500,000 farmers had fled to refugee camps in Ethiopia, Djibouti and Kenya. A 2-year old drought severely compounded the situation. Fishing communities in the north lacked nets, engines, spare parts, fuel and refrigeration equipment.

On 30 July 1992, James Kunder, director of the US Office of Disaster Assistance (ODA), described the situation in Somalia as “the single worst humanitarian crisis in the world”. On 24 July, the UN Secretary-General criticized the members of the Security Council for giving what he maintained was undue priority to the situation in Yugoslavia, which he described as a “war of the rich” [19].

III. UNOSOM I: The Diplomatic Instrument

1. Seeking Consent

The international community responded to this apocalypse slowly, and when it did, it did so only in a fragmented manner. The late response alienated Somalis and turned them against the UN even before its arrival. They felt the international community had abandoned them. This was a feeling that later turned into outright opposition to what was perceived to be a partisan effort.

Aidid had not trusted the UN from the beginning. One of Barre’s tactics of undermining opposition was to exile the disfavoured to diplomatic posts abroad. Aidid had been banished as Ambassador to India and he remembered that at the time there were cordial relations between Barre and the Egyptian Minister of Foreign Affairs, Boutros Boutros-Ghali. Aidid felt that both Boutros-Ghali and James O.C. Jonah, then UN Assistant Secretary-General from the Office for Research and the Collection of Information (ORCI), were supporters of Barre. Jonah had visited Mogadishu 3-5 January 1992 and failed to negotiate a cease fire. Aidid charged that a statement made following the visit, which he surmised had been written by Boutros-Ghali before, officially labeled him a warlord [20].

Ali Mahdi, on the other hand, commanding a force of 5000 in comparison to Aidid’s 10,000, [21] weaker as he was, yet having proclaimed himself head of state, consistently cooperated with the UN in order to strengthen his own position. This in turn made the UN more suspect in the opinion of other factions. From the beginning of 1992 an effective alliance began developing between the UN and Ali Mahdi, which persisted despite the size, strength or intentions of UN forces in Somalia. It has proved to be the case consistently that the UN, rather than supporting a standard or structure or concept of authority to facilitate a local transition from conditions of conflict to an order of civil society, it adopts itself the nature of authority locally. Whatever its mandate, whether to facilitate local belligerents or to defy them, it tends to manoeuvre in the direction of maximum consent or accepts the restrictions placed on it by the powerful party. In combination, this meant that unless the UN was willing to exercise independent political authority, it had to become much like another Somali faction. The Somalis joked that “all Somalis are generals and the UN is an admiral” – a reference to the fact that each of the factions was led by a general, while the Special Representative of the UN Secretary-General (SRSG) during UNOSOM II was US Admiral Jonathan T. Howe. Given the very nature of the environment, Howe found this to be an advantage initially [22].

Over the next two years, the UN never managed to transcend its alliance with Mahdi and its alienation of Aidid. The divergence only grew wider and wider and the UN effectively served to drive a wedge deeper between the two. The reason for this is significant and is an insufficiently recognized difference between operating in inter-state and internal conflicts. In international conflicts, the primary objective of peace efforts in previous decades, to fulfill the Charter objective of ensuring peace and security between states, was to achieve a cessation of hostilities. Provided violence stopped, the conflict could be contained and neighbouring or distant but interested states could be protected from its spread. The longer-term concern for the underlying causes of the conflict could be relegated to pacific means of dispute settlement, which did not pose immediate threats to the remainder of the international community. Treating the belligerents as exclusively whole entities, like other full members of the international community, meant the UN was not concerned with the body politic or social conditions prevailing in each, regardless of their relevance to the conflict. There may have been only some minimal concern for the gray space in the no man’s land occupied by UN forces. This was a diplomatic posture that firmly placed peace operations between belligerents.

To deploy between factions in internal conflicts, however, is to further fragment a collapsed state. Divisions are formalized, tying factions more firmly to defined territory delimited by cease fire lines, and factions are treated as static, whole and legitimate entities. These are further destabilizing artificialities in an environment in which factional fringes are ill-defined and their coherence perpetually evolving. Factions are rarely monolithic. The focus should not be on the area between factions, but above factions. An overall, comprehensive framework requires a political authority that can exercise executive powers independently of the factional malaise, yet functions jointly with factional and other local leaders as part of a transitional total social and political process. A UN effort that was not centralized in its own organization, that did not have independent means of exercising political authority and that had not integrated in its strategy genuine local needs had to shatter in the context of extreme fragmentation. That is exactly what happened.

Every “player” in Somalia had his own political agenda. Coordinating these would not be enough. There had to be a means of jointly making political decisions in the field on a regular basis. Each of the factions had its objectives, as did other kinds of leaders, such as the elders, intellectuals, and women’s groups, who sought to sidestep the warlords. The UN Secretary-General was pushing Somalia onto the international agenda and wanted as robust an operation as possible, pushing for more and more troops as a response to perpetually deteriorating security conditions in the absence of any other options. But since this manoeuvring was conducted by the secretariat and the initiative had not come from member states, there were not the resources to back the unrealistic expectations of a restrictively-conceived operation. The Security Council had in late February 1992 just approved the blueprint for Cambodia, the UN’s largest and most ambitious operation in its history to that date. Croatia had exploded and Bosnia was about to dominate the international agenda overwhelmingly. UNOSOM I became more of an idea than an operation. James Jonah was approaching the problem according to traditional peacekeeping principles while Mohamed Sahnoun, former Algerian Ambassador to the UN and the SRSG as of March/April 1992, tried to foster coordination between the different political kingdoms of the UN and a local political process. He attempted to address the source of the problem, the anarchy.

2. Attempted “Regionalization”

When Sahnoun began operating in April 1992, his “team pursued a strategy of putting the clan system to work for Somalia” [23]. Since clans had survived as social institutions, there were old authority patterns that could be resorted to, even if the elders had played more of an advisory than leadership role in pre-colonial society. Nevertheless, they could “help identify and promote new leadership that could become a credible alternative to the faction leaders”. By the middle of October 1992, factional leaders and community elders had assented to the notion of a conference on national reconciliation.

In addition to this, Sahnoun pursued a strategy of decentralization, or “regionalization”. He recognized the limitation of focusing too much on Mogadishu. The UN could, and eventually did, get bogged down in the Mahdi-Aidid struggle. Sahnoun proposed:

to divide Somalia into four zones to encourage decentralization of all operations. The four zones would include Bossaso, Berbera, Kismayu, and Mogadishu. This decentralization would have made both UNOSOM and relief agencies less dependent on the conditions prevailing in Mogadishu and promoted the new regional leadership so badly needed by Somalia.

Geographic “decentralization” may have referred to operating away from the bottleneck in Mogadishu, but it did not imply UN autonomy in the regions; operations would still have to be centrally orchestrated. This was a moot point: although the strategy was endorsed by Security Council Resolution 767 of 27 July 1992, only in mid-October were regional offices about to be opened, when it was too late.

This kind of grassroots political process was doomed by several problems. To be effective in such anarchical conditions, the UN had to provide a centre of gravity around which alternate leadership could grow and strengthen, even to a certain degree in cooperation with the factions themselves. However, in the absence of local physical infrastructure, any UN assistance to mitigate this or a UN political infrastructure, the process was too fragile and too easily disrupted by the warlords, particularly as security and famine conditions worsened throughout 1992.

Worse still, the UN bureaucratic structure and habit undermined rather than strengthened the process. UN agencies refused to decentralize. They would not leave Mogadishu, Nairobi or Djibouti, and refused to deploy officials in the four zones. “Delivery of emergency relief and rehabilitation operations continued to be concentrated in Mogadishu, giving Aidid and his allies a certain leverage. This (geographic) centralization also undermined UNOSOM’s efforts to organize grassroots structures in local communities within regions.”

Furthermore, the UN failed to take the advice of its Office of Legal Affairs to investigate the crash in July 1992 of a Russian plane with UN markings leased to the World Food Programme (WFP). It had apparently delivered currency and military equipment to the north of Mogadishu to troops loyal to Ali Mahdi. Another such illegal flight crashed in late October and it was unclear how many others there may have been in contravention of Security Council Resolution 733 of 23 January 1992, which had imposed a “general and complete embargo on all deliveries of weapons and military equipment to Somalia”. It was incredible to Sahnoun “that although the UN’s name and reputation were at stake, no serious investigation was undertaken and no legal action for redress was pursued”. The trust of the warring parties and the sympathy of Somalis towards the UN was eroded: “Suspicion of all UN personnel was spreading in Mogadishu…(The incident) rekindled the old perception of many Somalis that the UN and some countries were biased in favor of Ali Mahdi.” An essential feature of a successful political process, the support of the population, was being lost. Ali Mahdi’s view was rare in Mogadishu: “Let me confirm that the UN has done a lot of good things” [24].

Above all, Sahnoun lost the support of the Secretary-General. Sahnoun was increasingly frustrated by the behaviour of the UN specialized agencies, which he publicly criticized, and the management approach at UN headquarters in New York. He clashed with James Jonah. In October 1992, Sahnoun convened in the Seychelles a cross-clan conference of Somali intellectuals representing the four zones to discuss national reconciliation. During the meeting, Boutros-Ghali sent a message to Sahnoun challenging the reasons for his presence there and ordering him to refrain from criticism of the UN agencies. “It was more than one could tolerate from the UN bureaucracy that had inspired such criticism, ” he felt and resigned as SRSG, issuing a formal statement to the press on 29 October 1992.

3. A Symptomatic Approach

Throughout this inescapably flawed and fragile political process by far the greater focus at UN headquarters in New York was on the symptoms of insecurity and famine. While the International Committee of the Red Cross (ICRC) and NGOs such as Save the Children, Irish Concern, CARE, Medecins Sans Frontieres and others were operating in the Somali interior, some UN agencies complained of security conditions and did not even maintain offices in Somalia, preferring to remain in Nairobi or Djibouti. This invited the scorn of NGOs. The UN agencies were unable to meet their pledges. Between January and June 1992 some 50,000MT (metric tons) of food were needed each month, or a total of 300,000MT for the entire period. During this time, WFP delivered in total 18,857 of its pledged 68,388MT. The ICRC on the other hand, between February and June, brought 53,900MT of food into Somalia through 20 different entry points, by sea, air and across the Kenyan border. The food was distributed through community kitchens, providing two meals a day and feeding some 600,000 individuals, rather than distributing raw stuffs [25].

Part of the problem of this fragmented approach was the prevailing but mistaken belief that agencies, NGOs and the UN can be coordinated. Like the factions and warlords, each has its own political agenda, usually devised out of context in a headquarters far from the field. The field office behaves like a diplomatic mission, representing the headquarters policy inflexibly. Again, needed was a means accepted by each NGO and agency headquarters for collective decision-making in the field on an on-going basis. Instead, they could not agree on common policies. Sahnoun proposed that the agencies harmonize their payment rates for services and use of facilities to limit the impact of militia and looters on UN operations. He also suggested the selling of the food being delivered in order to undermine the market established by looters flogging their take. Yet the agencies resented the spectre of any authority other than their own and rejected such interference, thus exacerbating the famine conditions by mid-1992 and in turn the security conditions.

The overall flaw of UNOSOM I was to attempt to apply to the problem as a whole the wrong instrument, the peacekeeping model, and at the wrong time. In a diplomatic fashion, this treated internal factions as state territorial entities. It was inevitable that the bureaucratic habit of the UN, seeking to function within the umbra of familiar institutions, in the absence of state officials, gravitated towards the factions and warlords. Who else were the UN officials to talk to, they wondered? The warlords were at least identifiable to some degree, a critical requirement in a diplomatic culture. But UN officials behaved with the factions as they would with sovereign governments. Therefore a primary objective when the UN became apprised of the Somali situation was to achieve a cease fire. But the limitation of this could have been anticipated: when factional fighting subsided, it was replaced by general lawlessness, banditry and looting, which was tolerated by powerful factional leaders such as Aidid because they could not control it.

Nevertheless, a concerted effort led to a cease fire between Mahdi and Aidid on 3 March 1992. But it was a fragile agreement: Mahdi was perpetually cooperative and Aidid consistently intransigent. Mahdi accepted most anything the UN proposed, Aidid did not. Both men met Boutros-Ghali at the UN in January and the terms of a cease fire document were negotiated. The signing took place in Mogadishu in the presence of James Jonah. The document provided for the free entrance of humanitarian aid; a cease fire; and the deployment of 50 observers to monitor the cease fire.

On 17 March the Security Council passed Resolution 746 in response to the cease fire, but only called for a technical mission to determine how to proceed. All three aspects of the agreement were delayed: humanitarian assistance did not arrive in any great quantity; Aidid refused to permit the deployment of the observers unless they were unarmed and without uniforms as a matter of Somali sovereignty, although he eventually compromised on the issue of the uniforms; and just because there was a cease fire between Mahdi and Aidid, this did not mean that other factions would not be fighting. Aidid charged that Siad Barre attacked within 60 kilometers of Mogadishu in collusion with the UN.

Finally, on 24 April 1992, Security Resolution 751 authorized the deployment of the 50 observers and officially established UNOSOM. While a cease fire had been concluded only between Mahdi and Aidid in Mogadishu, Security Council resolutions referred to a cease fire throughout Somalia. Nevertheless, it directed the 50 observers to supervise the line of separation, the “Green Line”, dividing Mogadishu between Mahdi’s and Aidid’s forces. It was not a spacious no man’s land but the width of a single street. Nor was it like pacifying a battlefield or a front between two state armies; it merely divided two areas of disorder. In any case, the observers were not fully deployed until 23 July. They did not have the required logistical support, had a tendency to observe from their hotel rooms, and when they did venture out they were completely vulnerable – at one point two were ambushed and wounded. Given the limitations of a misconceived task, some proved more valuable in negotiating free passage of convoys from the port to Mahdi’s northern Mogadishu.

Resolution 751 also authorized the deployment of a security force to:

provide security for United Nations personnel, equipment and supplies at the port of Mogadishu and to escort deliveries of humanitarian supplies from there to distribution centres in Mogadishu and its immediate environs. They would also, as necessary, provide security for United Nations personnel, equipment and supplies at the airports in Mogadishu. [26]

The Secretary-General envisioned a strength of 500 infantry troops for the purpose. This was an entirely unrealistic expectation given the very conditions he described to justify the security force in the first place. The notion of distribution being the critical problem rather than the supply of food became the dominant position. By October 1992 estimates of international assistance being looted before their distribution ranged from between 15 to 40% and 10 to 80%. In his 24 August 1992 report, the Secretary-General proposed the deployment of four additional security units of 750 each for Bossaso, the southwest, Berbera and Kismayu. This total of 3500 was approved by Resolution 775 of 28 August. Another 719 personnel, including three logistical units, were authorized on 8 September, bringing the total UNOSOM I authorized strengths to 4219.

However, the increase in the number of security units was not coordinated with UNOSOM in the field or factional leaders. Sahnoun had finally achieved agreement between Aidid and Mahdi at the beginning of August 1992 on the deployment of the initial 500 troops, months after their authorization. Had they arrived within the month, they might have been able to deploy. Though, they would still be reliant on the good wishes of the factions, which were daily inconsistent. Instead, characteristically slow UN troop movements meant the security unit was not to arrive in a timely manner.

Consequently, even before their arrival the additional 3000 troops were approved. The announcement was made before Sahnoun had been informed, let alone consulted. Not only was it unlikely to be accepted by the warlords, but it jeopardized the agreement reached on the initial 500. Aidid said:

The decision was taken without consulting me…What I accepted was 500 soldiers to secure humanitarian relief from the airport to distribution points in Mogadishu. But these 500 Pakistanis chose to stay at the airport and therefore never did what they came for. I found even this difficult to accept since I wanted Somalis to be in charge of airports as well as borders, for the sake of Somali sovereignty. [27]

Needless to say, the deployment of the 500 troops was doomed. They began arriving on 14-15 September 1992 and were prevented from deploying at the port or securing the airport by the faction in control of each area. Only by early October had all 500 arrived, but they did not manage to deploy beyond the airport. No other troops were forthcoming.

The security conditions worsened dramatically as the famine reached its height in October/November 1992. It was felt that in the absence of the increased security units, the 6 October “100-Day Action Program for Accelerated Humanitarian Assistance for Somalia” would be ineffective. Of 100,000MT of food to be delivered between 12 October and 3 December, WFP only delivered 18,900, of which an uncertain amount actually reached the victim population. The thrust of the $82.7 million plan included: a massive infusion of food aid; the aggressive expansion of supplementary feeding; the provision of basic health services and a mass measles vaccination campaign; the provision of clean water, sanitation and hygiene; the provision of shelter materials, including blankets and clothes; the delivery of seeds, tools and animal vaccines with food rations; the prevention of further refugee outflows and encouraging repatriation; and the strengthening of Somali civil society at the national, regional and local level. All of this required swift delivery and freedom of movement.

However, with a peacekeeping mandate, the increased number of troops was unlikely to make a fundamental difference to the security of humanitarian activities. The mob had run amok and they were not about to facilitate the passage of defensively-postured troops. Instead, there needed to be a perception of integrating social and political factors on the ground with political authority above the internecine anarchy. The perception of peacekeeping between entities could not function as an overall approach in anarchical conditions without being torn apart by those conditions.

The role of diplomatic tools, including negotiation and traditional peacekeeping, is critical but these are only one instrument in an arsenal and need to be employed as part of an integrated approach. Negotiating with factions is a necessary part of easing the conduct of operations. For instance, not only informing factions in control of the territory which a humanitarian convoy will cross, but even travelling with representatives of these factions will ease passage across road blocks. But this robust diplomatic activity must be matched by means of functioning independently of the will of the warlords, including operating with a single political objective, with means of using force in a limited manner at only particular moments, as well as a capacity to police the Somali-UN interface. This combination of legitimacy and effectiveness in turn would foster more consent locally and strengthen the diplomatic front, in turn lessening the need to employ other instruments.

IV. UNITAF: The Military Instrument

1. Humanitarian Enforcement

By November 1992, the two principal problems in Somalia were perceived at the UN to be famine and weapons. Only a trickle of assistance was getting through as security conditions worsened. Any effort at a political process had ceased. The streets of Mogadishu were dominated by “technicals”, or vehicles on which an automatic weapon had been mounted. UNOSOM observers in Mogadishu were hijacked and robbed by troops loyal to Ali Mahdi. The Pakistani troops finally secured the airport on 10 November, but came under heavy fire from Aidid’s forces. A WFP cargo ship was shelled in the port on 24 November. Relief agencies were paying protection money and bribes to guarantee the safety of their some-400 civilian personnel. The Security Council and the Secretary-General agreed the situation had become “intolerable” and that resort to use of force under Chapter VII to deliver humanitarian assistance should be considered.

On 25 November, US Secretary of State Lawrence Eagleburger visited Boutros-Ghali and informed him of the readiness of the United States to “take the lead” in an enforcement operation. The National Security Council (NSC) had decided to intervene on 21 November “because of the scale of human disaster and the realization that the United States was the only nation perceived by Somalis and by the regional states as being in a position to maintain neutrality and with the ability to launch the necessary large-scale aid operation” [28]. The United States had hardly been a neutral player in Somalia during the Cold War, but in comparison to the local perception of the UN at the time the US was welcomed, and it did have the means to do the job. Aidid put it into perspective:

Boutros-Ghali had his hidden agenda: He wanted Somalia to become a UN trusteeship and eventually to restore Siad Barre…Boutros-Ghali was also responsible for the delay of humanitarian aid in order to strengthen the demand for trusteeship. This ended, however, with a decision to establish UNITAF. This decision was unilateral. Initially, therefore, SNA was against it; but because of the humanitarian needs, and the fact that we did not expect the US to have colonial interests in Somalia and that country’s democratic history, we accepted UNITAF. Statements by the US President in support of this also contributed to our decision, since we were not against humanitarian assistance. We therefore accepted the US force with pleasure. I met with Ambassador Robert B. Oakley former US ambassador to Somalia and special presidential envoy, who said that the US came as a friendly force. I also said that on radio because Oakley emphasized that UNITAF would not interfere in internal affairs. I therefore thought that Boutros-Ghali would not be able to fulfil his hidden agenda. [29]

In addition to this, the “CNN factor”, and the grim picture of Somali famine and war on living room television sets finally took its toll-although much more slowly than in the manner it instantly destroyed support for the operation only one year later.

Also, the US Republican administration had just been defeated at the polls in October, and whether or not there was a desire to leave the new Democratic administration with an imbroglio, there was a desire amongst some Republican appointees, such as Andrew S. Natsios, then Assistant Administrator of the US Agency for International Development (AID) and President George Bush’s special coordinator of Somali relief, to do a last good thing [30]. He described a meeting in the second week of November 1992 held in the Oval Office where Bush ordered a policy review to determine how to halt the famine [31]. A plan had been devised before the Security Council decided the matter.

Furthermore, the Pentagon, the State Department and the NSC were reassessing their priorities and restructuring their offices after Bush’s speech at the UN General Assembly in September outlined a new US commitment to UN operations. In May 1992, few officials in all three anticipated the eventuality of US participation in UN operations, including the prospect of UN command and control of US troops. There were good reasons and bad reasons for this. The bad reasons had to do with the Vietnam syndrome and Weinberger principles: the avoidance of limitations on the use of force. The good reason was that there was a realization of an incapacity in the UN system to command the kinds of assets that the US could provide. The principal area of reluctance seemed to be at the 3-star general level and above. The younger generation of leadership below the 2-star level was more forthcoming about UN operations, and this generation was to be entirely alienated by the Somalia experience [32].

After Bush’s speech a specifically designated office was opened in the Pentagon for peacekeeping, eventually with its own Deputy Assistant Secretary of Defence. The State Department later established its own office for peacekeeping. Military bases were assessed for their suitability as peacekeeping training centres. Prospects for a US-organized military planning cell at UN headquarters were forwarded to the UN secretariat, and smartly rejected by the UN’s “not-invented-here syndrome”. Above all, there was consideration of accepting UN command of non-combat troops. The US was the only country in the world that had not accepted UN command of its forces and that the most powerful country in the world was even considering this subordination ought to have been treated with the utmost care as a critical building block in international organization. Instead, the Secretary-General pushed it too far, too quickly.

On 25 November 1992, the Secretary-General proposed five options to the Security Council. First, continue and intensify efforts to deploy UNOSOM I in accordance with its existing mandate and authorized strength of 4200. This he dismissed as inadequate. Second was the complete withdrawal of military assistance to humanitarian relief, abandoning the agencies to their own security arrangements. He rejected this and argued that wider rules of engagement (ROE) could help break through into Somalia. The other options were at the opposite extreme, proposing to put an end to violence against the relief effort. The third option was a deterrence proposition, a “show of force” in Mogadishu to discourage warlords and bandits. The Secretary-General contended that to have the desired deterrent effect, the operation would have to be countrywide. In any case, while it was an old peacekeeping dictum that a show of force can avoid the need to use force, it was also true that to be credible there had to be a willingness and competence to use force and threaten it on an on-going basis. This would have raised questions about the kind and degree of force, and the tenure of its presence.

The Secretary-General favoured the fourth and fifth options: a “countrywide enforcement action” authorized by the Security Council under Chapter VII. Under the fifth option, the operation would be commanded and controlled by the United Nations. While the Secretary-General would have preferred this, he conceded that the UN did not have the capability to command and control an enforcement operation of the size required. Under the fourth option, this could be undertaken by member states responsible directly to the Security Council. In this regard, he mentioned the US offer to organize and lead such an operation. The Secretary-General emphasized that, unlike the Gulf, enforcement by member states would have to be defined and limited in time, in order to prepare for the return to peacekeeping and to another vague and ill-defined concept from An Agenda for Peace, post-conflict peace-building.

On 3 December 1992, the Security Council adopted Resolution 794 which explicitly authorized the Secretary-General, the United States and other troop-contributing countries to “use all necessary means to establish a secure environment for humanitarian relief operations in Somalia”. While use of the language of Article 42 but only express reference to Chapter VII as a whole was reminiscent of the Gulf War formulation, unlike Resolution 678 this time the Council identified more clearly the task to be accomplished and referred directly to the United States, which it had avoided two years earlier. Furthermore, troops were to be under UN auspices and command and control was regarded as “unified”. Troop-contributors were to coordinate their activities with the UN.

The Security Council asserted its authority over the operation through monitoring and reporting procedures. It established an ad hoc committee to supervise the implementation of the resolution; requested reports from troop-contributors within 15 days of their participation, and periodically thereafter; and detailed a UNOSOM liaison unit at the unified command field headquarters. The Security Council emphasized the limited nature of the mandate by requiring the Secretary-General to draft a revision plan and mandate for UNOSOM on UNITAF’s withdrawal. While the Council decided that the deployment of UNOSOM’s authorized strength should proceed at the Secretary-General’s discretion, the deployment of UNITAF effectively halted UNOSOM I deployments until a new plan had been drafted. But to avoid the slow deployment of UNOSOM II, the UN should have strengthened its forces gradually to better position itself for UNITAF contingent withdrawals.

As a result of this “unified” arrangement, juridically originating in the Security Council as a single source of authority, two chains of command developed: one for the US-led coalition and the other for the UN under the Secretary-General. The latter was virtually moribund in comparison to the powerful coalition and effectively a relationship emerged between the US leadership in Washington, including its operational command in the field, and the UN Secretary-General.

On 4 December 1992, President Bush ordered the deployment of “Operation Restore Hope”. UNITAF was to include 28,000 US personnel under the command of General Robert Johnston and more than 10,000 troops pledged by 20 other countries [33]. The deployment plan focused on the worst areas of the famine, in south and central Somalia, and was constrained by Bush’s pledge to bring the troops home by Inauguration Day in January 1993. Even excluding the northern half of the country, Pentagon planners estimated that the four-stage operation would require at least 60 days to implement. With less than a month of planning behind them, the first US units arrived in Mogadishu on 9 December. They were joined by the French Foreign Legion from Djibouti and forces were expected from Belgium, Canada, Egypt, Italy, Saudi Arabia and Turkey. With the arrival of international forces, some Somali armed groups left Mogadishu for the countryside. Rather than circumventing conditions in Mogadishu and developing stable conditions outside, as Sahnoun’s “regionalization” had tried to do, this served to spread the conflict in Mogadishu to the regions and exacerbated security conditions in the countryside.

According to the second and third stages of the plan, US and allied forces spread out from the airport and harbour of Mogadishu to the other seven population centres. They secured the airstrip at Baledogle, halfway to Baidoa, and then pushed on north and west to Baidoa – 160 miles from Mogadishu – Oddur, Belet Uen and Gialalassi. Then moving southwards, they took control of Kismayo and Bardera and secured a land corridor between Bardera and Baidoa. The idea was to establish food distribution centres at each site and guarantee the delivery of large quantities of food in order to eliminate looting and hoarding, the use of food as a weapon, and therefore weakening the power of the warlords. After all, when US troops arrived on 14 December in Baidoa, what had come to be known in the media as “the city of death”, some 500 civilians were dying daily from starvation yet stores controlled by the factions were found filled with food. In a final stage, it was envisioned that the delivery of humanitarian relief and the responsibility for security would be transferred to a UN peacekeeping force. This transition had to wait more than 60 days.

In the meantime, the territorial extent of UNITAF’s deployment in 40% of the country set a pattern that UNOSOM II never managed to deploy beyond. Despite demands by the Secretary-General to deploy throughout the country, [34] and to address the humanitarian needs and security concerns in the north, Sahnoun’s notion of dividing the entire country into four regions was discarded. Instead, nine Humanitarian Relief Sectors (HRS) were established according to military imperatives dictated by response needs of the famine around Kismayo, Bardera, Oddur, Baidoa, Baledogle, Gialalassi, Belet Uen, Mogadishu and Marka. Mogadishu was chosen as the headquarters because of the size of the port. The only other comparable deep-water port nearby was at Kismayo, but it was not big enough. The net result was to centralize operations again in Mogadishu, with the disadvantages of being caught between the struggle of the two principal warlords, Mahdi and Aidid.

Nevertheless, the 9 December amphibious landing in the capital, in the bright glare of the international media, which had completed its own landing earlier, was unopposed. Baidoa was reached five days later and by 20 December, a 20-truck relief convoy arrived and marked the opening of the first of six major relief supply routes. With Belgian, Canadian, French and Italian troops, US forces secured Belet Uen on 28 December. While UNITAF troops were authorized to use force under Chapter VII, there was still no concept of operations that dictated rules of engagement. But this did not pose the contentious problem. UNITAF exercised its powers restrictively, disarming only those Somalis threatening its forces or preventing the primary objective of delivering food. Although by 18 December, given the continuation of fighting and looting in Mogadishu and in areas in the countryside not under UNITAF control, the US began considering the confiscation of weapons not voluntarily surrendered for cash or compensation.

2. A “Secure Environment”

The principal point of contention that arose between the US and the UN Secretary-General, was the scope of disarmament. The disagreement effectively centered around the mandate of Resolution 794: what constituted a “secure environment for humanitarian relief”? Did it mean, restrictively, to protect corridors for specific deliveries of assistance to end the famine? Or more broadly, did it mean creating generally secure conditions in which the UN could freely operate? In the latter case, the US would have to have fundamentally altered the environment of anarchy in Somalia, which would have entailed not only breaking the vicious cycle of violence and famine, but also addressing the root cause of anarchy through a process of national reconciliation and political reconstitution. The US had been prepared to respond to the symptom of famine, the most dramatically tragic leg of the vicious cycle. It had not intended and in the field was not prepared to pacify Somalia.

Boutros-Ghali, however, argued that the coalition should not withdraw before controlling heavy weapons and disarming lawless gangs. The United States refused to endanger prospects of a speedy withdrawal with “mission creep” and the spectre of getting bogged down in precarious anarchical conditions. It called for the UN to deploy a strong enough force with broad enough rules of engagement to protect humanitarian activities under the existing security conditions. To have called for this was to misunderstand the capacity of the UN and its inability to manage or finance such a force, which the Secretary-General was quick to point out. This indicated, at this early stage, that the US had genuinely unrealistic expectations that the UN could replace it in the field. This mistake was fatal for UNOSOM II and for the US, which not only would be unable to withdraw, but would be drawn more deeply into Somalia under a much more complex mandate. The disagreement over the issue of disarmament exacerbated this by delaying the development of a transition plan.

Boutros-Ghali’s sentiments were understandable: the opportunity for the UN to be something more than a symbolic guarantor of agreements was irresistible. With a powerful force deployed under a joint mandate and the opportunity to employ that force in pursuit of international interests, the limitations of disarmament were overshadowed. If Boutros-Ghali’s demands made some sense, it was with regard to their timing since at that moment there was a unique window of opportunity. Aidid had been intransigent consistently until UNITAF; his receptivity meant that this was the best chance to convince his forces or the majority of other Somalis generally to disarm. The US strength evoked fear, but also confidence in its capability. There was also a certain amount of local trust in the US based on a number of bridges between the two countries, including amongst Somalis that had been trained by the US military or in US academia. Both Aidid and Mahdi had been educated in the US and each had a son in the armed forces: Mahdi’s son was on active service in the US Army; Aidid’s son was called up from the reserves of the US Marines, after he volunteered his Somali linguistic capability, and served for two months in UNITAF. Even if there was not a willingness to disarm, at least an openness to the possibility that it would take place could be sensed in Mogadishu as the cost of an AK-47 in the market rapidly declined in December from US$200 to US$100.

A decisive deployment could have been coupled with enough leverage and confidence to pursue disarmament, but both were lost. The passive manner of operating after UNITAF’s deployment diminished the impact of powerful forces as a credible threat and Somalis then felt both free and compelled for survival to continue to live by the gun. At the other extreme, the confidence of Aidid was lost as he began to realize in January 1993 that the US was at least cooperating with his nemesis, the UN, and beginning to conduct tasks beyond the delivery of humanitarian assistance. This resuscitated Aidid’s intransigence, who turned against the US as much as he had been against the UN or even other warlords. By January, the window of opportunity for disarmament was closed for good.

Even if the US had taken advantage of this window, what impact would this have had? Since they had only a few weeks in December to start a disarmament campaign decisively, US forces would have had to be prepared to start as soon as they landed, in the manner that they approached implementation of a strategy for delivering food. Disarmament had not been planned for and it was not a task that Bush and his leadership had envisioned when they offered to lead an enforcement operation. To take advantage of the timing, disarmament would have had to be approved in November; arguing about it in December was already too late.

Furthermore, a disarmament effort would need to seal all borders by air, sea and land, in cooperation with other states, a commitment which was not forthcoming. It would have had to be conducted across the entire country simultaneously. To have disarmed any one part of the country before the other would have rendered it susceptible to attack by neighbouring areas still armed. As it was, sometimes weapons confiscated from individuals had to be returned because it was their only defence in a hostile environment. Similar problems existed for NGOs when UNITAF from time to time confiscated their guards’ arms. Typically, the arms of the weak, the easiest to disarm, were confiscated first, worsening the imbalance of power in favour of the warlords. Given their power, Aidid and Mahdi would be less likely than others to be disarmed, consequently worsening conditions of conflict and anarchy.

Successful disarmament did not require the removal of every weapon in the country; sufficient disarmament could have been conducted to weaken the warlords enough to make them more reliant on a process of political reconstruction. This meant disarmament would have functioned best as one part of an overall strategy, but such a framework did not accompany the deployment of UNITAF. Some senior US officers in UNITAF felt that to pursue disarmament in December 1992 would have meant losing another 100,000 civilians to the famine because of the drain on resources it would have caused. But, they further argued, after having served at senior levels in UNOSOM II and witnessed the worsening of security conditions, that the loss of life would have been worth it if it helped resolve the conflict and prevented the cost of violence in the long term.

However, since weapons were only a symptom, disarmament could not have been sustainable by itself. The Belgians were the most successful at demobilization and managed to create a “weapons-free” environment in Kismayo. But it was not a clean process conducted through legitimate leverage and local confidence. Instead, they seem to have used force abusively, beating and shooting violators. They were charged with colonial behaviour and trials took place on their return to Belgium. As soon as they withdrew, the weapons were back in Kismayo. Disarmament needed to be matched by the development of social and political conditions that could replace the need to rely on arms; the rule of law needed to replace the law of the jungle. Not only did the violent conflict of warlords need to be transformed into political competition, but the need for the private individual to rely on weapons for survival in the context of anarchy also had to be addressed. In the absence of a rule of law and order, weapons had replaced the Xeer as social fabric. The persuasiveness of arms was such that some US soldiers considered disarmament in Somalia would have been as difficult as making the United States weapons-free.

If disarmament was to have been conducted as part of a larger political framework, approaches other than cleansing the country of weaponry may have been employed. Various forms of control could be envisioned, such as concentrating weapons confiscated from uncontrolled elements and transferring them to newly constituted, accountable military and police forces.

The scope of this kind of political activity, as obviously necessary as it was, was entirely beyond the scope of UNITAF’s activities. Instead, diplomatic activity was conducted by Oakley on one front while UNITAF entered something of a second phase. As clashes continued and sniping worsened in Mogadishu, US troops began seizing and destroying arms caches. They relied on tips and random sweeps of the city. By mid-February they had collected: 1.27 million rounds of light ammunition; more than 129,000 rounds of heavy ammunition; 2255 small arms; and 636 heavy weapons, including tanks, mortars, grenades, rocket and missile launchers, and surface-to-air missiles [35].

UNITAF began the germ of a shift from relief to rehabilitation. US political officers began coordinating town meetings in Mogadishu, Baidoa, Bardera and Kismayo, in an attempt to foster Somali municipal institutions. US troops began clearing streets and restoring water supplies. Most significantly, the US began organizing local police forces, intending to deploy some 3500 in Mogadishu. By the end of January 1993, some 300 unarmed officers were deployed in Mogadishu and another 300 in Kismayo. They were beginning to enjoy a certain amount of success in the presence of UNITAF forces. However, all of these developments could not be sustainable because they occurred in a vacuum. Necessarily, they largely collapsed during the transition from UNITAF to UNOSOM II, not only because of the absence of a political authority to ensure their direction and continuation, but also because the preoccupation in the transition with internal, operational organization meant shifting focus away from surrounding environmental conditions.

Another consequence of operating in a political and law and order vacuum concerns legal constraints on the behaviour of UNITAF forces. While the problem of the lack of clear operational guidelines for “peace-enforcement” in internal conflicts was mitigated somewhat by the restrictiveness of the scope of operations and by the desire to avoid a disarmament imbroglio, there was no context in which to address abuses of force or unlawful behaviour. In addition to the authorizing mandate, UNITAF was governed by international humanitarian law and the national law governing each contingent. Both Belgium and Canada held trials for the behaviour of their troops. A US soldier was convicted of aggravated assault of two Somali civilians by a court martial held by the US forces in Somalia.

National responses to violations of law, however, are an inconsistent means of filling a law and order vacuum in anarchical conditions. This was a problem that plagued UNOSOM II. During UNITAF, there was insufficient investigation of incidents in which the shooting of Somali civilians was mistaken, negligent or in fact criminal. [36] The failure to record Somali casualty figures was itself disturbing in both UNITAF and UNOSOM II. Above all, the need to exercise law and order functions locally, including the arrest and detention of local Somalis, was particularly problematic in the absence of a political framework whose authority provided for means of prosecution, and therefore the ability to prefer charges on detainees. Arrests without charges constitute unlawful detention. This issue remained unresolved and failed to be accounted for in the planning for UNOSOM II.

On the diplomatic front, Oakley continued the habit of negotiating between the powerful factions. He arranged several meetings between Mahdi and Aidid, who on 28 December together led a peace march along the Green Line. But efforts excluded other factions, such as Morgan in Bardera and Jess in Kismayo. The US was still positively perceived during President Bush’s one day visit on New Year’s Eve, in comparison to the hostile reception from angry Somalis during Boutros-Ghali’s visit on 3 January 1993. The UN-sponsored Informal Preparatory Meeting on Somali Political Reconciliation, under the Secretary-General’s chairmanship, was convened in Addis Ababa 4-15 January. It was attended by the Secretaries-General of the League of Arab States (LAS), the Organization of African Unity (OAU), the Organization of the Islamic Conference (OIC), the Chairman of the Standing Committee of the Countries of the Horn, and the representative of the current Chairman of the Movement of Non-Aligned Countries, as well as 14 factional leaders who agreed on an immediate and binding country-wide cease fire; the concentration of heavy weapons and disarmament of militia under UNOSOM and UNITAF supervision; and the convening of a national reconciliation conference on 15 March. [37]

These agreements would ultimately prove fruitless, and in fact set the stage for succeeding months: it was clear from this point that positions were diverging and conflict proved inevitable. Aidid’s alienation was becoming clearer. Rather than fostering a cease fire, this diplomatic effort, in the absence of an accountable centre of gravity, was in fact delimiting the trenches in an evolving contest between Aidid and the international community, including amongst factions supporting one side, the other, or neither. A political front was developing between Aidid and the US and the UN. This would proceed irreconcilably and the 15 March meeting would serve as a political battlefield in which both sides jockeyed for the initiative to displace the other. In the context of this diminishing consent, the closing of the window of opportunity to disarm, as well as a reduced capacity to disarm by force, the disarmament provisions were unrealistic, and worse, later constituted the casus belli in June. As one UNOSOM official stated, “We failed in January-March 1993 to even conceptualize what political framework would work or would be accepted.”

Humanitarian conditions improved dramatically. By 20 January, 40,000 tons of relief supplies had passed through Mogadishu, including 8000 tons of medicines, seeds, tools and veterinary supplies. The port of Kismayo had been reopened. Road convoys had delivered 25,000 tons of food, the remainder delivered by air. Humanitarian Operation Centers had been established in the main population centres to coordinate UNITAF, UN and NGO relief activities.

During the UNITAF phase, there was fruitful cooperation. US soldiers admitted that when they arrived in Mogadishu they had no idea what “ICRC” or “USAID” stood for, but the novelty of their mission, of humanitarian assistance, meant they were willing to commit their resources to a supportive role as much as providing overall leadership of the plan. There was a common goal by force of circumstances amongst NGOs and the military. For their part, NGOs felt UNITAF forces worked hard, day and night, and the abundance of resources and minimal bureaucracy meant transportation and escorts were readily available. This assisted cooperation, which, by the transition phase, had become strained. The arrival of UN bureaucracy and fewer available resources restricted good will.

This process became increasingly noticeable as US troops accelerated their withdrawal in February 1993. The humanitarian mission had been accomplished and US planners were frustrated at the slow pace a transition plan was developing, eager as they were to hand back the larger security issues to the UN. From its high in mid-January of 25,800, the US contingent dropped by the end of February to 16,000. Non-US troops increased from 10,000 in early January to 15,000.

3. Coalition Leadership

The US style of coalition-building began to give way to UN means of choosing troop contributors. When building a coalition, the US considers the type and quality of the assets offered and the variety of geographic representation only second. The UN, on the other hand, requires first that a force represents as wide a geographic spectrum as possible. This is not so much a criterion of legitimacy, particularly since a UN force is deployed with a collective mandate from the Security Council; rather, it is a vestige of traditional peacekeeping which acts like a trigger mechanism for the influence the participating nations may bring to bear on recalcitrant belligerents outside the local environment. Therefore, the more participants from as many regions as possible, the more favourable can be made the environment in which the force functions.

While a geographic spectrum continued to be a requirement theoretically in more complex operations, the UN was increasingly more reliant on whoever was willing to contribute the required assets. Unlike the US, the UN is more a beggar than a chooser. This fosters the phenomenon of poorly equipped contingents arriving in the field reliant on non-existent UN logistics and is one reason for slow, weak and fragmented deployments.

When building a coalition, not only is the US demanding in terms of the suitability of the assets offered by a nation and a contingent’s operational preparedness, but it actually convenes, prior to deployment, the commanders that will serve in the field. The US does not accept merely representatives of the participating states, only the field personnel that will have to function together are permitted and required to meet. This creates an organizational centre of gravity which the US dominates and establishes the means of interoperability between the contingents, so that this is not debated after the force arrives in the field and does not delay implementation of the mission. US officers on the ground, although operating in a loose relationship with other contingents, nevertheless can direct them and regulate their fulfilment of instructions from the overall US command. In this manner, minimum standards can be maintained and a single objective achieved.

When time came for the UN to assume command and control in Somalia, it could not possibly have provided the kind of glue or create the centre of gravity that the US provided in UNITAF. As a tendency, the UN does not convene either its military or civilian headquarters staff or the contingent commanders prior to deployment. Regardless of the mandate, the powers of the contingents and concept of operations of the mission are interpreted in national capitals and the interoperability of national forces is left to be determined after deployment in the field. Often interoperability is never achieved and dysfunction results.

As fine an officer as the UNOSOM II Force Commander, Lieutenant-General Çevik Bir, may have been, US planners felt he did not understand how to command such a disparate, multinational operation. They claimed he was not forceful enough in demanding from contingents compliance with the UN mandate. In fact, when Major-General Thomas M. Montgomery, the commander of US forces in Somalia and Deputy Force Commander of UNOSOM II, tried to do so, the contingents complained to Bir of US dominance. When it became clear to the US that the UN would not be able to replace it in the field to the degree required, the US responded to fill the vacuum, keeping it engaged in the field to a greater extent than it expected or wanted.

While this may have been the only option under the circumstances, not only did it mean an overwhelmingly dominant position for the US in the UN force, thus rendering a fully integrated UN operation on paper more of a joint operation with US forces operating separately. It also meant that since the centre of gravity had developed as a result of a military operation, UNOSOM II and its first operational plan (OPLAN 1) would be dominated by military imperatives. Not only were the numbers in the whole mission, despite its multiple objectives, almost entirely military, but just as the UN could not field a robust military leadership, it fielded an even weaker political staff.

The political mandate was novel and more experimental than were military deployments, and therefore could not provide overall direction or leadership. UNOSOM II would be inescapably a victim of UNITAF, not only because UNITAF closed a political window of opportunity, even if unintentionally, but because UNOSOM II would be constructed in the field around the remaining core of UNITAF. This was inevitable given the conventional manner of UN deployment, which usually arrives without there being any such core and usually succumbs to the will of intransigent factions.

The dominance of military leadership and the uncontrollable, non-sustainable consequences of this, indicated the limitations of military force. In the same way that diplomacy cannot provide an overall framework in internal conflicts, a peace process cannot be conceived as part of a military framework. The military is most effective when it has clear objectives devised by a political authority and is employed in this limited manner, as it was in the first UNITAF phase of delivering humanitarian assistance. When this task or set of objectives is completed, the military requires on-going political direction; if it is left without this and becomes the principal decision-making authority in the field, its decisions are likely to be based on military imperatives, which are characteristically short term and limited in scope.

Furthermore, political direction from home capitals or UN headquarters is not sufficient. Between the first and second phases of UNITAF a new US President took office, necessarily delaying political direction as existing policy was affirmed or altered. Also, decisions that are quite rational in the bureaucratic context in which they are made, in ministries of defence or foreign affairs, may be simply irrelevant to rapidly evolving ground conditions. This time gap between decision-making and implementation may be too great, and the variety of domestic considerations, including available resources and national public opinion, may skew the nature of the response to the objectively identifiable requirements in the field. These will be inevitable factors affecting the deployment of national forces initially, as well as throughout the period they are away from home. However, that they should dictate the decisions made for issues that arise daily in an operational area is not tenable.

Far away political direction may give a military field commander the freedom to complete a mandated task, and this is fine if the task is clearly circumscribed. Thereafter, an undue burden is placed on the commander, whose troops must be either withdrawn, given another clearly limited task or, if more complex and rapidly shifting goals are to be identified and implemented, given political direction in the field. The diplomatic efforts by Oakley throughout UNITAF would not be adequate. Both needed to be part of another framework, not independent. Nor could diplomacy in tandem with military troops respond to conditions in Somalia. Overall political direction had to encompass both in an on-going framework to ensure sustainable results from a middle road between diplomacy and military force, although including either or both as required in a limited context.

V. UNOSOM II: A Flawed Concept

1. The Operational Logic

UNOSOM II was authorized under Chapter VII of the UN Charter, but the premise of its presence in the country was conceived as one of “assistance”. “Assistance” implied local consent; enforcement powers meant the opposite. The conditions in Somalia were anarchy and something much more than “assistance” was needed; so much so that UNOSOM II could not resist doing more. But it did so intermittently because it bounced back and forth between its mandate and ground needs, almost unconsciously. Ultimately, it did neither adequately. It had to conduct some political and law and order tasks independently of Somalis since the concept of “assistance” to the Somali people, in the absence of Somali institutions, did not indicate how the UN would carry out tasks in an “assistance” manner. And when UNOSOM II in its later phases finally began to help build Somali institutions they had to do so in more of a “control” manner than “assistance” precisely because of the weakness of these institutions. That they had been fostered by the UN in the first place without any existing basis meant that the UN had effectively the power in Somalia of a governorship-in-trust. Yet the refusal to acknowledge this and design a comparable mission framework meant that the mix of all these premises fragmented UNOSOM II’s directional capability.

The variety and number of tasks assigned to UNOSOM II under the mandate were well beyond an “assistance” concept. The need to conduct some of these tasks on the one hand and the perception of having to maintain an arm’s length merely “assisting” Somalia on the other, yet utilizing massive force as a response to the resulting dysfunction, caused the mission to be reactive. It quickly lost sight of the root problems of Somalia and the ultimate purpose of the mission as intermediate security concerns dominated.

In the Secretary-General’s report of 3 March 1993,38 concerning recommendations for the transition from UNITAF to UNOSOM II and the latter’s concept of operations, all of these inconsistencies were apparent. In the Secretary-General’s opinion, “a secure environment” had not been created in Somalia, despite improvements in security and humanitarian conditions. Despite the reason for this being that the root causes of violence had not been tackled, he too focused on the resulting conflict disproportionately. He affirmed that there was still no effective functioning government in the country, no organized civilian police force and no disciplined national armed force. The security threat to the UN, its agencies, UNITAF, the ICRC and NGOs was still high in Mogadishu and elsewhere. Furthermore, there were no international deployments in the northern 60% of Somalia or along the Kenyan border, where security continued to be a matter of grave concern.

The Secretary-General concluded, therefore, that UNOSOM II should be authorized under Chapter VII and endowed with enforcement powers in order to establish a secure environment throughout Somalia. Here a secure environment meant not only the short-term UNITAF approach, but something perpetual and sustainable, something that Somalis could take over. But overshadowing the requirements implicit in creating a perpetually secure environment was the call for enforcement. While such powers were necessary, they had to be employed strictly to underwrite political objectives of attending to the root causes. This meant that UNOSOM II had to behave as an independent political authority with the military and other civilian components, humanitarian and diplomatic, implementing its decisions. But the lack of this disciplined linkage of military force to specific political goals resulted in each aspect of the operation functioning in a certain degree of isolation, exacerbated by the disparity in size and structural development between the military and civilian leadership and personnel.

This in turn resulted in effective military leadership of political strategy, partly facilitated by the Secretary-General’s imperative of ensuring that UNOSOM II was an enforcement operation. It was to be the first UN-commanded operation of its kind and a part of the experimental euphoria of the time. But the drive for an enforcement capability still obscured the need to develop the third option, the middle road in use of force, the original context of a “peace-enforcement” concept of operations, which was not enforcement at all costs. Simply connecting enforcement powers and a political mandate could not produce the concept needed.

2. Political Transformation

The two approaches relied on by Boutros-Ghali to tackle the Somali apocalyptic cycle were military disarmament and political reconciliation. The mandate would empower UNOSOM II to provide assistance to the Somali people in rebuilding their economy and social and political life, re-establishing the country’s institutional structure, achieving national political reconciliation, recreating a Somali state based on democratic governance and rehabilitating the country’s economy and infrastructure. These were intensely social and political tasks addressing the heart of the cycle, but they were still being approached with military and diplomatic instruments, through disarmament and reconciliation. If these social and political tasks were to be accomplished, there had to be political transformation, not merely reconciliation between factions. Could international bureaucracies be honest about or even realize what state-building entailed, being as they are the product of that process, yet entirely outside it? Bureaucracies maintain the status quo but are unused to creating it.

How could Somalis be unified without a historical culture of unity, and with their only experience of centralization having gone terribly wrong? How could this unity be achieved in the context of the perpetually fragmenting forces of anarchy driven by powerful factions, requiring somehow to reverse this cyclical inertia with little means? How could this be achieved given the persistence of social divisions along clan lines, which was one of the few social reference points left, rooted in historical lineage? As separate from clans the evolution of factions had become, there were nevertheless clan linkages, symbolically, historically and factually to factional members. And under whose leadership would this unity be centred, given that most of the middle class intellectual élites had left Somalia for neighbouring countries or had gone farther abroad, to Europe and North America, and refused to return despite repeated requests from UNITAF and UNOSOM II?

Could a third party play such a role? With an “assistance” mandate the UN certainly would be unprepared to play “statesman”. Yet the tasks it assumed were precisely those that a statesman would have to contend with. This issue is at the core of appreciating the role that an international political authority, a governorship-in-trust, would have to fill, either itself or by fostering a local equivalent. But how?

First, there had to be a decisive, powerful deployment, quickly and comprehensively. Initially, it would have to cover the trouble areas and ultimately the entire country, but not at the cost of becoming a vulnerable, thin blue line. If a force arrives in strength and is in control of ground conditions, not submitting itself to the direction of the existing inertia, it can win the confidence of the local population which will regard it as an effective authority. Furthermore, if an operation arrives forcefully, its appearance of strength can avoid the use of force. For example, the powerful arrival and deployment of UNITAF caused something of a factional retreat, yet the restrictive exercise of its powers led to the renewal of a factional offensive.

Deployments can shift the balance of power locally and must be reinforced if local intransigence is to be controlled, factions displaced, and the momentum of the cycle of violence reversed. The element of surprise and the principle of seizing the moment and snatching the initiative are as important politically as they are in conventional warfare. The US strategic tendency of overwhelming situations would be best employed at this beginning point, but it cannot be maintained as the principal strategy thereafter. An anathema to the US military is the idea of being a policeman one day, a soldier the next, and a policeman the day after that. The problem is precisely that because of the US black and white options for force, using either none or alternatively all; and once all force is used, it is difficult to go back to using none. In fact, this is precisely the job of a civilian policeman on the streets, whose presence is relied on most of the time, but who uses force at other times, then must return to a passive role.

Military forces must be prepared to behave as policemen, as well as soldiers, if the centre of gravity created by a decisive deployment is to be consolidated. This is because of the second requirement of responding to a social and political malaise, namely the establishment of a political presence and the need for subordinated assets to implement political decisions independently of local conditions. “Independent” does not mean making decisions in an organizational vacuum in a headquarters isolated from local conditions; merely, they must not be dictated by the will of the factions, the very problem the operation has come to contend with. On the contrary, as independent as the international authority must be, there must also be as many contacts and as much joint decision-making with local players as possible.

If there is to be genuinely a transition in a country from one set of conditions, such as anarchy, conflict and famine, to another set of conditions, resulting in order, peace and health, and whether a UN mission is a supervisor of this or a catalyst for it, the operation must secure the reigns of power first. How else can the power that is driving the existing problems be transferred to a new authority? The flaw of UN transitional processes in the past is that they do not manage to secure power first, even if sometimes this has been done on paper juridically, as in Cambodia. Consequently, missing this step leads to either a transformation that occurs by itself, the continuation of conditions as they had been before, or a slight realignment of this in which the international community permits itself the luxury of treating the existing power structure as legitimate, having been rubber-stamped by a UN operation.

Then the question arises pertaining to the transfer of this centre of gravity to a local authority. It may be that local individuals will play a greater and greater role in the international operation’s political, legal and security institutions, thus creating a trained cadre that can inherit the shell of each institution as the international personnel disengage from it. However, while these institutions may be appropriate for the tasks of the international operation and a consequence of member states’ cultural style of governance, they may not be ideally suited for the country in question. On the other hand, they may serve as a means to an end, which displace fragmented factions and can be transferred subsequently into something more indigenous. This may not be possible if the logic varies too greatly between the internationally established institutions and local cultural traditions. In fact, another collapse, of newly established but alien and rootless institutions, may occur. In Somalia, had the logic of the centralized and artificial European state concept failed, or had it been simply mismanaged? In the absence of the indigenous Xeer, was there an alternative to state institutions? Simply, what was the ideal basis for institution-building in Somalia?

Determining the basis, source and structural logic of a new authority was very much an anthropological project and operations of the type and magnitude of UNOSOM II require strong components with anthropological, sociological, psychological and area expertise. Was democracy, for instance, exportable to Somalia? Being a Western imperative, what form could it take in Somalia? One limitation of elections, the ultimate and ideal objective of the UNOSOM II political process, was that they would not have been necessarily conceived by Somalis to have a winner and a loser and transfer of power. Winning an election under the circumstances of anarchy, would be one factor in a balance of power that evolved by itself, as it had been in Cambodia.

If there was any doubt about this, it was expressed very clearly in Aidid’s book [39]. He described democracy historically in Somalia over 4000 years as a system in which all parts of the society participated, regardless of leadership. In the future, he said, there should be the ideal of “Autonomous Democracy”: “We have to realize that the model of healthy and functional democracy that we are planning to adopt in Somalia is such that every one will have perfect autonomy and satisfaction of serving the nation whole-heartedly, and no one will be able to exploit the people by becoming President, Prime Minister or a big boss.” This meant that “After the fair elections to the Parliament, a truly national government should be formed by inviting members from all the national parties and not just by the members of the majority party…Therefore, the idea of the formation of National Government inviting members from all the national parties should satisfy all the voters – no matter which particular party one has voted for”. By extrapolation, then, regardless of who won an election, Aidid would be part of the government.

If elections were not chosen as the basis for a transfer of power, what alternatives could there be? A constituent assembly of clans and other social group leaders, with or without the factions? Somalia was no longer pre-colonial or post-colonial, but something else in the absence of the Xeer, an alternative social fabric or effective alien or indigenous institutions. Whatever the most sustainable model of governance it would have to be devised in the context of an international caretaker administration of the territory. Without this, any ideal model could not be put in place, prevented as it would be by the tyranny of the factions threatening any transition.

Used to the luxury and habit of operating at arms length, affordable on a battlefield, troops in UNITAF and UNOSOM II were insufficiently briefed on local social and political phenomena. There tended to be in the field the perception that Somalis had lost their value of human life; infinitely more important was the currency value of a plank of wood, a piece of corrugated sheet metal, livestock or a camel. Soldiers witnessed killing as a means of dispute settlement at mundane levels of disagreement. Killing had become a normalized form of social behaviour. Intense aggression in anarchical conditions, that favoured whoever was the physically stronger in the case, led international troops to conclude “The Somalis don’t want our help.” Their view of Somalis was similar to Burton’s nearly a century and a half earlier: They are “constant in nothing but inconstancy – soft, merry, and affectionate souls, they pass without any apparent transition into a state of fury, when they are capable of terrible atrocities” [40].

This fed a behaviour and perception of opposition and exclusion, rather than integration between international forces and local Somalis, and set a psychological stage for conflict. It was expressed by the high walls around the airport and UN compound in Mogadishu behind which UNOSOM II kept. In another miniature vicious cycle, the more hostility there was between Somalis and the UN, the less freedom of movement international troops had in Mogadishu; the less freedom of movement there was, the greater the gulf between the two grew. This narrowed the social and psychological basis for communication and established an environment ripe for misunderstanding.

3. The Blueprint

Instead of this complex social and political approach to Somalia, the core of the operational concept outlined in Boutros-Ghali’s 3 March report was a series of military tasks [41]. When political reconstruction did begin, it was too little too late, and ill-defined, resulting in political failure and military retreat. If there was to be any attempt at clarity, it was in the areas of disagreement between the US and the UN Secretary-General during UNITAF, along the lines of:

  1. monitoring that all factions continue to respect the cessation of hostilities and other agreements to which they have consented;
  2. preventing any resumption of violence and, if necessary, taking appropriate action against any faction that violates or threatens to violate the cessation of hostilities;
  3. maintaining control of the heavy weapons of the organized factions which will have been brought under international control pending their eventual destruction or transfer to a newly constituted national army;
  4. seizing the small arms of all unauthorized armed elements and assisting in the registration and security of such arms;
  5. securing or maintaining security at all ports, airports and lines of communications required for the delivery of humanitarian assistance;
  6. protecting the personnel, installations and equipment of the United Nations and its agencies, ICRC as well as NGOs, and taking such forceful action as may be required to neutralize armed elements that attack, or threaten to attack, such facilities and personnel, pending the establishment of a new Somali police force which can assume this responsibility;
  7. continuing the programme for mine-clearing in the most afflicted areas;
  8. assisting the repatriation of refugees and displaced persons within Somalia;
  9. carrying out such other functions as may be authorized by the Security Council.

The report stated that further to the disarmament provisions of the 8 January Addis Ababa agreements, a planning committee composed of senior officers from UNITAF and UNOSOM had developed a “Somalia cease-fire disarmament concept”. This required the establishment of cantonment sites, for storage of heavy weapons, as well as transition sites for temporary accommodation of factional forces while they turned in their small arms, registered for future governmental and non-governmental support and received training for eventual reintegration into civilian life. Cantonment and transition sites would be separated from each other to prevent any possibility of factions or groups seizing the heavy weapons. Those failing to comply with timetables or other modalities of the disarmament process would have their weapons and equipment confiscated or destroyed.

UNOSOM II was to be conducted in four phases of military operations. Transition from phase to phase was supposed to be dictated by parallel political reconciliation efforts and rehabilitation programmes. Heavy military activity in the beginning would gradually decrease as civil authority increased in activity, in a seamless, linked evolution. But the stages were defined by the degree of military activity and there were no criteria for the political tasks. As such, the political process could not develop meaningfully. It was disconnected from and overwhelmed by the military, and this prevented devolution of efforts from military to civilian tasks.

This was a flawed conception. Translated it meant international military activity decreased as Somali civil authority grew; UN civil authority, in this scenario, could serve only to negotiate amongst Somalis to foster local authority. The UN political office was not large enough or conceptually structured to serve as a link or the glue for the progression from phase to phase. It was very much a vestige of the diplomatic activity of UNOSOM I that approached reconciliation through negotiation rather than by institution-building. Diplomacy could not establish Somali authority.

Figure Three: UNOSOM II Balance of Military and Political Tasks (not included)

Phase I concentrated on the transition of operational control from UNITAF. Military support to relief activities and the disarming of factions would continue throughout the transition. The Secretary-General set the date of transfer at 1 May 1993, but it finally took place on 4 May. Phase II focused on consolidating UN operational control and would conclude when UNOSOM II had deployed and was operating effectively throughout Somalia and the border regions. In Phase III, major efforts would be made to reduce UNOSOM II’s military activity and assist civil authorities in exercising greater responsibility. This phase would end when a Somali national police force became operational and major UN military operations were no longer required. Phase IV concerned redeployment or reduction of UNOSOM II forces.

Although the central goal of UNOSOM II was to assist the people of Somalia to create and maintain order and new institutions for their own governance, any degree of detail in the 3 March report was devoted to the military tasks of phases I and II. Only one broadly worded paragraph pertained to the critical Phase III that was the purpose of the exercise. The standard of measurement for the end of this phase was itself a military imperative. Since a functioning police force was the exit strategy, the establishment of police forces dominated the prioritization of institution-building. This was true at the end of UNITAF and after the combat phase. By mid-May it was reported that in Mogadishu 2840 police were directing traffic, controlling crowds and protecting feeding centres; while another 2000 performed similar duties in other cities.

But establishing police forces in a vacuum could not be sustained, particularly in spring 1993. There was no government or court system of prosecution, nor any law for the police to enforce. Eventually the 1960 Constitution and the 1962 Criminal Procedure and Penal Codes were resuscitated and mixed with Islamic shariah law, but this was utilized more by the fledgling courts from the end of 1993 than by the police before. To link functioning police forces, without an established civil authority, to the reduction of military operations was entirely artificial. There would be violence for as long as there was no political authority, and probably afterwards as well. A police force did not imply the existence of a political authority, as much as it needed one, nor did it imply political or diplomatic reconciliation.

With regard to military details, the Secretary-General called for a military component of 20,000 [42] all ranks to carry out the assigned tasks and an additional 8000 personnel to provide the logistic support, which would be the first US troops to serve under UN command, albeit in a non-combat role. According to these figures, of a force of 28,000, about 29% would support the active work of about 71%. But the environment turned out to be more hostile than such a plan was prepared for. For instance, it was estimated that for a contingent to be self-sufficient in the field it required a minimum of 1400 personnel to provide the means of survival in Somalia. At the height of its strength in late 1993, the German contingent deployed at Belet Uen numbered about 1700, and it was a logistics unit that was supposed to support the Indian contingent of 5000. This meant that it was taking some 83% of the German contingent to maintain its own livelihood in Somalia, leaving a bit more than 17% of the contingent to do its substantive job. This was indicative of a mission that spent more time surviving than saving the country. In addition to the UN force of 28,000, the US agreed to provide a tactical Quick Reaction Force (QRF) of 1000 troops positioned offshore.

The Secretary-General further indicated the capability required by combat forces. These included: patrolling and close-combat; information-gathering and interpretation; indirect force; anti-armour fire; all weather night and day operations; casualty evacuation; tactical communications; and air support (fire power and transport). In contrast, a civilian staff of approximately 2800 was proposed, though their tasks and distribution were not indicated, and few ever reached the field.

The report outlined UNOSOM II’s conventional peacekeeping chain of command, in which a civilian SRSG is supposed to have supreme authority in the field. Admiral Howe was appointed SRSG on 5 March 1993 with the broad writ of overseeing the transition from UNITAF to UNOSOM II, and continuing the tasks of promoting political reconciliation, coordinating humanitarian assistance and paving the way for rehabilitation and reconstruction of the country.

The Secretary-General did note that the deployment of UNOSOM II would not be subject to the agreement of any local factional leaders; yet he also reiterated his belief that:

Notwithstanding the compelling necessity for authority to use enforcement measures as appropriate,…the political will to achieve security, reconciliation and peace must spring from the Somalis themselves. Even if it is authorized to resort to forceful action in certain circumstances, UNOSOM II cannot and must not be expected to substitute itself for the Somali people. Nor can or should it use its authority to impose one or another system of governmental organization. It may and should, however, be in a position to press for the observance of United Nations standards of human rights and justice. [43]

This perception was contradictory: If Somalis could be expected to “do it themselves”, then what was the need for a large and intrusive operation? Establishing an interim authority is not a substitute for the Somali people but a means of transferring power from factional tyranny to the local population. A mission like UNOSOM II could not remain aloof from Somalis in the manner suggested by the Secretary-General.

The 3 March plan was approved on 26 March by Security Council Resolution 814. Acting under Chapter VII, the Council decided to expand the size and mandate of UNOSOM in accordance with the Secretary-General’s recommendations for a period extending to 31 October 1993. Emphasizing the security aspects of the plan first and foremost, the resolution demanded that all Somali parties comply fully with the commitments they had undertaken, and in particular with the Agreement on Implementing the Cease-fire and on Modalities of Disarmament, [44] and that they ensure the safety of the personnel of all organizations engaged in humanitarian and other assistance to Somalia. It called for all, and particularly neighbouring, states to cooperate in the implementation of the arms embargo under Resolution 733 of January 1992.

Thereafter, the considerably more complicated provisions of the plan were approved in a broad brushstroke. The Council requested the Secretary-General, through his Special Representative, and with assistance from all relevant United Nations entities, offices and specialized agencies, to provide humanitarian and other assistance to the people of Somalia in rehabilitating their political institutions and economy and promoting political settlement and national reconciliation. Such assistance would include economic relief and rehabilitation of Somalia, the repatriation of refugees and displaced persons within Somalia, the re-establishment of national and regional institutions and civil administration in the entire country, the re-establishment of Somali police, mine-clearance and public information activities in support of the UN activities in Somalia.

While there was a military force to conduct the military tasks, the civilian components required for each of these tasks to be accomplished were not adequately identified. A small civilian headquarters staff under the SRSG would have to be subdivided into small offices with only a few individuals each to manage a process more complex than were the military tasks. If any part of the huge military body lent on the spindly civilian crutch, it would have snapped.

In the meantime, since the October 1992 100-Day Plan had expired on 19 January and security conditions improved, the UN introduced a new “Relief and Rehabilitation Programme” for Somalia on 4 March. It reflected a shift in emphasis from emergency food aid to longer-term reconstruction. Despite improvements in humanitarian conditions, in southern and central Somalia people were still destitute and dependent on relief food assistance. Measles, diarrhoea and other infections continued to take a heavy toll, particularly on small children. Lack of access to clean water sources and poor sanitation presented major health threats. In the 3 March report, the Secretary-General identified three challenges facing the UN in 1993: facilitating the voluntary return of approximately 300,000 refugees and internally displaced persons; providing jobs and work for the many millions of unemployed Somalis, including members of armed gangs, militias and various private armies; and helping the Somalis in rebuilding their society and rehabilitating the decayed infrastructure.

Devised in cooperation with Somalis, UN agencies, the ICRC and NGOs, the plan was expected to cost $166.5 million (in contrast to the $1.55 billion it was conservatively estimated UNOSOM II would cost over 12 months). The Programme covered the period from March to December 1993 and focused on activities in ten priority areas: re-establishment of local administrative capacity; re-establishment of national and local police forces; support services for women, particularly those victimized by violence and trauma; return of some 300,000 refugees and over 1 million displaced persons within Somalia; development of a food security system; establishment of a basic health care system; increasing the availability of potable water and of sanitation; expansion of agriculture and enhancement of livestock; work opportunities for the unemployed; and re-establishment of primary education and vocational training.

The Programme was adopted at the United Nations Conference on Humanitarian Assistance to Somalia, held 11-13 March 1993 in Addis Ababa, under the chairmanship of the UN Under-Secretary-General for Humanitarian Affairs. It was attended by some 190 Somali representatives, as well as senior representatives of donor governments, international agencies, regional organizations and NGOs. Donors pledged $130 million towards the cost of the plan and it was expected the remainder would be forthcoming as implementation gained momentum. But just as the October 100-Day Plan was disrupted by insecurity and the deployment of UNITAF, this Programme was soon to be disrupted by UNOSOM II’s combat operations, which effectively ceased humanitarian activities for many weeks.

4. The Addis Ababa Agreement

The stage for conflict crystallized at the Conference on National Reconciliation in Somalia, convened in Addis Ababa 15-28 March 1993. This was further to the January agreement at the Informal Preparatory Meeting in Addis Ababa. The Conference was chaired by the Secretary-General’s Deputy Special Representative for Somalia, Ambassador Lansana Kouyate of Guinea, and attended by the leaders of 15 Somali political movements, as well as representatives of the regional and other arrangements that had attended the January meeting – the LAS, the OAU, the OIC, the Standing Committee of the Countries of the Horn and the Non-Aligned Movement. After two weeks of intensive and intermittent negotiations, the leaders of all 15 Somali political groups signed on 27 March the Agreement of the First Session of the Conference of National Reconciliation in Somalia [45]. At the closing session of the Conference on 28 March, the Agreement was unanimously endorsed by all the participants, including representatives of women’s and community organizations, as well as elders and scholars.

The Agreement comprised four parts: disarmament and security, rehabilitation and reconstruction, restoration of property and settlement of disputes, and the establishment of a Transitional National Council (TNC) vested with administrative and executive authority. The parties agreed to consolidate and carry forward advances in peace, security and dialogue made since the beginning of 1993. They reaffirmed their commitment to comply fully with the cease-fire agreement that was signed in Addis Ababa in January, including the handing over of all weapons and ammunition to UNITAF and UNOSOM II. There was to be disarmament of the factional militias within 90 days.

This was an unrealistic proposition. Since UNITAF had disarmed only sporadically, as a matter of convenience rather than as a concerted programme, there was no institutional framework UNOSOM II could inherit, such as coherent procedures, identified cantonment sites or generally agreed target locations or elements of factions to be disarmed. All of this would have to be established as UNOSOM II was deploying, acclimatizing to Somalia and organizing internally, which meant it would approach disarmament not as a comprehensive programme as it was supposed to but more arbitrarily and haphazardly.

Furthermore, there was too great a gap between the nominal agreement in Addis Ababa and genuine consensus between the factions, and too little understanding between the factions and the UN. Since there had been a difficult process of negotiation, characteristically broad terms were drafted to achieve consensus, without clarifying details that might have prevented any kind of agreement. But this was necessary if implementation was not to be disputed on the ground. This is the danger of the diplomatic approach: to achieve agreement in the conference hall, terms tend to be broadly written and capable of multiple interpretation. While this was tolerable for traditional peacekeeping, it has proved crippling in complex, internal conflicts.

Different expectations of accepted terms lead to confrontation and conflict on the ground, usually at the cost of the UN’s effectiveness, such as in Cambodia, or to violent conflict if the UN makes a stand, as it was preparing to do in Somalia. Of the 15 “political movements” at Addis Ababa, only 4 or 5 were very influential, such as Aidid’s SNA, Morgan’s SNF, Jess’s SPM, and the SSDF allied with Mahdi. Each felt they could interpret the Agreement to their advantage on the ground. The balance of forces in the field would dictate the actual details of the terms accepted; and this would be determined by physical strength, in which all had self-confidence since all continued to struggle and expected to defeat the others. Also, as one UN official observed at Addis Ababa, “It is a Somali habit: they are friendly together abroad, but at home they realize they are worse off then they thought when they made an agreement. So conflict ensues.” This further limits the reliability of contractual arrangements relied on by the UN in its operational calculations, which tend towards best-case-scenario preparations in worse-case-scenario conditions.

There was a difference of opinion about how to approach the factions after the conference between UNOSOM II political officers and other civilian and military officials. Some felt that the importance of Addis Ababa was to give UNOSOM II a local mandate that matched its international mandate of disarmament and national reconciliation – a form of consent that had not been generally obtainable. But it was artificial consent. Nominally, UNOSOM II could claim some legitimacy in its actions on the basis of Somali acceptance of its mandate, but this was not enough.

Political officers recognized the limitations of the Addis Ababa Conference. In hind-sight they posed the question about the right moment to bring all the people together, both factional leaders and other groups, including elders. One UNOSOM II official described the need for a political centre of gravity before local reconciliation should proceed: “There is no point holding a conference for the sake of a conference. The Addis Ababa Conference was meaningless as a conference, regardless of the good will…There cannot be agreement in Somalia without a conference, but the pieces must be in place first and then a conference can formalize what is in place. Therefore the conference must be well-prepared.” Speaking in early 1994, the official added, “It would be disastrous if Somalia had another failed national reconciliation conference. If the people are let down again, the preparations for war already in motion will be exacerbated.”

At UN headquarters in Mogadishu, there were opposing views on disarmament through diplomacy and disarmament by force. Disarmament in 90 days was not possible. Seasoned UN officials with experience rooted in traditional operations were used to presenting the UN view to the parties and then proceeding according to agreements that could be reached or disagreements that limited options. In this manner, accountability of parties and terms of agreements evolved. The opposing view was held by other civilian officials, the many US advisors from the Pentagon and the NSC, and the military: If the Somalis did not live up to their commitments, it was felt, then “we can force them to do it”. The diplomatic view charged such attitudes as being the “arrogance of power”.

The enforcement view, with which the UN Secretary-General was firmly associated, was by far the dominant, not least because of the greater number and influence of those who held that view. But the use of force to respond to diplomatic gaps in anarchical conditions was an ineffective approach. Concentrated force could not be aimed simultaneously in the different directions necessary. It was a recipe for confrontation and inevitable conflict.

The so-called political process suffered the same gaps and flaws of an artificial agreement. At Addis Ababa there was a strategic manoeuvring between factions and the UN to create a framework which could be employed to their respective advantages. The TNC would consist of 74 members, with two men and one woman from each of the 18 regions of the country, one from each of the 15 political movements, and 5 from Mogadishu. This was to serve as an interim government before elections could be held in 1995. The Agreement referred to the drafting of a provisional constitution, or “charter”, and the establishment of an independent judiciary. Most significantly, arrangements were made for autonomous councils on the regional and local levels – a revival of the “regionalization” strategy.

Competition between the factions and the UN focused on the timing, location and direction of political momentum. The factions wanted to create a TNC first, as soon as possible, because this gave them primacy. They were united on this and perceived the TNC as one more battlefield. This was a top-down approach, while the UN, on the other hand, jockeyed for a bottom-up approach. According to a UNOSOM II political officer, “At Addis we planned to establish the District Councils (DC) and Regional Councils (RC) only; and only after that was complete, then on the basis of the DCs and RCs and the experience of this, create the TNC.” If the TNC was to be vested with administrative and executive authority, and if the UN was trying to control the process of establishing executive authority, then this was a clear affirmation that the UN could not merely “assist” but was assuming executive authority as a governor-in-trust.

Rather than establishing and controlling authority, the UN began by negotiating and turned the issue into a political confrontation with the factions instead of relegating it to an administrative level. This turned the question of “how” to maintain the political centre of gravity under international control to “whether” it would be under international or factional control. The strategy was to displace the factions by establishing an alternative authority on the basis of the Somali people, emerging from the elders, women, the social institutions rooted in the clans or any other constituency not assimilated in the factional competition. But building an authority on the basis of such disparate fragments with only few diplomatic resources could not compete with the power centres of the factions. As it happened, the factions established the TNC prematurely, though, predictably, continued to fight, rendering the TNC a fiction but out of the control of the UN. As one official stated, “The TNC was not a consequence of the UN at that point.” The UN focused on the DCs and RCs, which evolved painfully slowly. It had lost the initiative to create a political centre of gravity and was left with military force as its only response.

The battlefield was being delimited: Aidid had made something of a miscalculation at Addis Ababa. He had conceded to the Agreement because he felt, not only could he compete for control of the top-down TNC process as part of the overall factional conflict, but that he could out-manoeuvre the UN in the countryside by controlling the bottom-up process as well. This meant there were several fronts emerging on which Aidid and the UN would face each other irreconcilably, all of which would merge into an unavoidable violent confrontation. There was insufficient detail regarding disarmament, and opportunity for misunderstanding and disagreement; there would be disagreement about the primacy of the TNC and the process of regionalization; and competition for the latter.

Aidid’s intransigence evolved progressively within one month of UNITAF’s arrival. With each milestone – the second phase of UNITAF, the January Informal Meeting and the March Conference – the gap widened between Aidid and the UN and the US. The competition for the regions had begun under UNITAF and Aidid had prevented UNITAF’s second phase from making any significant impact in this regard, but as UNOSOM II’s focus turned explicitly on this, the competition became more acute and volatile. Aidid described from his point of view how this developed:

UNITAF extended its operation after one month by General Johnston, who became more involved in Somalian affairs. The UN and the US started to call the leaders of the liberation movements “warlords” and only wanted to talk with the elders. I met with Oakley and asked what was the interest of the UN and US in Somalia, but did not get an answer. What became evident was that the UN and the US had converging interests. The humanitarian mission was transformed into a military operation. The UN and the US both understood that the SNA was the only obstacle in their efforts to take control of Somalia. At that time SNA controlled 11 out of 18 regions of Somalia, and 75% of the population. But the UN used lack of government and local authority structures as an excuse for wanting to take control. A political gap existed, and the UN took advantage of this by contacting the elders and playing on them. Thereby they ruined my attempt to build local structures. SNA’s programme was regional autonomy, and the people were happy with this. [46]

He added that UNITAF behaved rapaciously and lost the support of the population.

They began to disarm in December without our agreement. They forcefully inspected houses, they looted and they raped. SNA demanded to be present when they made their inspections, but the US ignored this. Many claims were filed against UNITAF, but nothing happened. In February 1993 the biggest demonstration in Somalian history took place against UNITAF. This was not adequately mentioned in international media…UNITAF forces were isolated for 3 days by women and children due to their misbehaviour.

By the time of the Addis Ababa Conference in March, Aidid accepted the provision in the Agreement stipulating, as he put it, “that before the TNC could be established, there had to be peace in Somali communities.” He felt he could do this or at least prevent the UN from doing it first. His charge that UNITAF was against the political process, because it “would effectively re-establish Somali sovereignty”, reflected the essential competition between the UN and himself. These prevailing conditions meant the operation was over before it had begun, and the events just had to play themselves out, logically and inevitably. There was nothing apparent that could alter at this stage the trajectory and momentum of interconnected factors.

VI. UNOSOM II: A Political Failure

The overall experience of UNOSOM II can be divided into six phases, unrelated to the four phases envisioned in the planning of the operation. Following a first, transitional, phase, a second phase of reception and consolidation lasted from the 4 May transfer to the 5 June ambush of Pakistani troops in Mogadishu. This sparked on 6 June the third phase, of combat operations, which continued until the deployment of a company of 400 élite US Ranger special forces on 31 August. Thereafter, a fourth phase focused on hunting Aidid until the failed Ranger raid of 3 October, which led to a fifth phase of strategic reset, including the US decision to withdraw. The final posture by March 1994 was to return to the original UNOSOM II regional strategy, but restricted to the southern half of the country, under the second operational plan (OPLAN 2).

1. Deployment

A transitional phase had effectively begun by the beginning of March 1993, as the US

became impatient about withdrawal, and ended on 4 May 1993, when US forces formally handed Somalia back to the UN. There was supposed to be a seamless interface in the transition from UNITAF to UNOSOM II. It was to be a process of painting UNITAF blue. The principle was a piece-by-piece transfer of authority to UNOSOM. When UNOSOM was in place after a general transfer, the US would pass the hat of command to the UN. The seamless strategy had been a US proposition which the UN accepted. There was the belief that the population should not even notice the difference on the ground. This was a critical mistake: the transition was from one mandate to another, from one mission to another, and from one flag to another. The shift should have been dramatically marked for the population and NGOs. There was a much wider mandate with many more tasks. The political nature of the mission and the force was altering. However, in the minds of the population it was the same force, particularly since the flags did not change significantly. Although it was never explicitly stated, much more would have to be accomplished with far fewer resources, and the effects of this caused frustration locally and amongst NGOs. This gave the impression of a failing mission rather than the slow start of a new operation.

The transfer of command occurred without incident or ceremony, but it had been rushed: UNOSOM II had only 25% of its staff, no computers and a low level of organization. On 3 May, the last 340 US troops attached to UNITAF had departed, leaving behind 3625 US servicemen in logistics, communications and other support roles, awaiting replacements from other contributing nations and civilian contractors. It was expected that civilian contractors would be the primary means of support whenever possible. The US QRF, which would remain “indefinitely”, included 1381 troops, officially counted among the 21, 521 UNOSOM II personnel already deployed. It was expected that the operation would remain in Somalia for two years. Peaceful disarmament proceeded as rifles, submachine guns, grenades and other weapons were confiscated. The ability of UNITAF to enter and exit Somalia quickly and pass control to the UN was positively received on the US homefront. “Somalia became the litmus test for determining whether the U.S. could “help keep peace in the world” without having to do it all by itself.” [47]

The UNITAF HRSs were realigned into five Areas of Responsibility (AOR):

  1. AOR Kismayo combined HRSs Kismayo and Bardera, and covered the area from Kismayo on the coast, along the Kenyan border to the Ethiopian border north of Bardera;
  2. AOR Baidoa combined HRSs Oddur, Baidoa and Baledogle, and extended from the Ethiopian border north of Oddur, south past Baidoa and Baledogle, to an area Northwest of Mogadishu;
  3. AOR Marka-Mogadishu combined HRSs Marka and Mogadishu, and covered the coastal area and inland countryside between Kismayo and Mogadishu, including Marka and half of Mogadishu;
  4. AOR Gialalassi subsumed HRS Gialalassi and extended from the other half of Mogadishu northwards to Buulo Barde; and finally
  5. AOR Belet Uen subsumed HRS Belet Uen and extended from Buulo Barde north to the Ethiopian border. This represented the extent of deployment into the Somali hinterland.

Just as the HRSs had been dictated by military responses to the famine, the new AORs did not correspond to Somali administrative regions.

Figure Four: OPLAN 1 Areas of Responsibility (not included)

According to OPLAN 1, the realignment was “in anticipation of the new mission and changes in available forces”. The”new mission” was interpreted militarily and deployment considered exclusively a military issue, inevitably because of the number of troops in comparison to the civilian staff in the area and the domination of military planning in Mogadishu for UNOSOM II. While UNOSOM II may have had a political mandate, OPLAN 1 was a military plan and there was no corresponding political operational plan other than the 3 March Secretary-General’s report. Consequently the AORs reflected tasks based on the military imperatives of the mandate and the responsibilities of new contingents or UNITAF redeployments. This was implicitly supported by the Secretary-General’s desire to deploy beyond the 40% of the country UNITAF had occupied and his concern for the short-term security environment.

The AORs’ disregard for the political administrative boundaries of the Somali regions indicated the domination of military imperatives at the cost of the political mission. This reflected a perceptual framework of the outsider looking in fulfilling priorities detached from the wider range of ground requirements, rather than focusing on internal reconstruction. This necessitated the outsider behaving as an insider, though sufficiently independently to move in a new direction away from anarchical conditions. The military could have deployed proactively and independently of local conditions and could have snatched the initiative in each area of deployment. Instead, functioning independently of the institutions to be built resulted in reacting to the conditions of anarchy, and in turn the military were drawn into those conditions rather than managing to displace them.

The territorial integrity of the regional boundaries should have been maintained and military contingents deployed should have been responsible to a UN civilian official with overall authority for that area. How else could political and judicial institutions be rebuilt unless military, political and humanitarian assets were orchestrated at the local level for each administrative region? There would be no structural form on which indigenous authority could be grafted and fostered. How could police forces be adequately established to provide an exit strategy for the military if they were propped up in pockets related more to the idiosyncrasies of the contingent located in the area than a local authority with a jurisdictional area? During the second phase, fragments of police forces had been established in some population centres, but this process had a ceiling in the absence either of a country-wide, standardized programme or judicial, political and administrative institutions established at a comparable rate, including the promulgation or resuscitation of laws to be enforced.

The tendency, therefore, was for police forces to be established by the military without any geographic sense, and not to be trained by civilian police officers. Local institutions could not take root if the spatial framework of UNOSOM II was dictated by overarching military goals and not by identifiable jurisdictional areas. The inconsistency between the boundaries of the AORs and the Somali regions contradicted the intended phases of increasing transfer of tasks from the military contingents to civilian institutions. Civilian authority, whether UN or Somali, required military security to develop under the circumstances. Yet the dominating consequences of military deployment undermined the conditions for establishing civilian authority.

An election, for instance, could not have been possible without a realignment of operational boundaries in order to match the regions. The same mistake had been made in Cambodia in which initially sectors were established by military imperatives, according to the location and strengths of the factions. These had to be realigned in conformity with the provincial boundaries of the country to facilitate the organization and conduct of elections. This lesson had not been learnt in the intervening year.

During the second phase, UNOSOM II had been drawing troops away from Mogadishu as UNITAF contingents withdrew from their areas in the countryside. For instance, the Canadians were leaving Belet Uen and the 5000 Indians that had been promised had not yet arrived. This process would be reversed after 5 June, when troops were brought back to Mogadishu for the combat phase, ending any possibility of deploying into the northern 60% of the country. Throughout, Mogadishu remained the central focus of operations, with all the disadvantages this implied contrary to the logic of regionalization. One-third of UNITAF had been deployed in the city, but more than one-half of UNOSOM II was concentrated at the airport and the UN compound with access only to Mahdi’s area of control in northern Mogadishu. By the end of the year, there were 16 nations deployed in the capital and 8 outside; of 24, 199 troops in early 1994, some 13, 431 were in Mogadishu and 10,768 were in the countryside.

Military deployments both in Mogadishu and the countryside tended to be in small pockets, in defensive bases. In Mogadishu, troops were deployed only on Mahdi’s side of the Green Line and located at the airport, the UN compound and adjacent checkpoints. The briefing map on the wall of UNOSOM II’s operations room at headquarters had on it marked in blue areas of local support; it had two isolated blobs of blue and the rest was colourless. The environment in Mogadishu and outside in the country was hostile, so forces only patrolled narrow corridors and left the actual control of the territory to the factions. In Mogadishu freedom of movement of the military was limited to two routes: between the compound and the port area and to restricted areas only on intelligence missions. There tended to be more ease of movement for NGOs with local armed guards since Somalis shot more at each other or at uniforms. The private contractor providing catering facilities for UNOSOM II, Morres Catering, was owned and operated by an Australian who had provided similar services to the UN in Cambodia. He used two armoured personnel carriers (APC) of his own and Khmer guards, but this degree of protection did not save his son from being executed at the seaport in revenge for firing five female Somali workers.

Territorially the UN had become part of the landscape of the factions. This was symbolized by the fact that the actual UNOSOM II headquarters building just behind the compound wall sat directly on the Green Line facing SNA territory. It constituted, curiously, both the rear and the front. That is representative of internal conflicts generally where the battlefield is not a distant no man’s land but is everywhere. Therefore a priority of deployment should have been to control the territory as a whole. How else could it be administered and authority transferred? This could only have been possible if the deployment had been conceived as part of an administration of the territory, which it was not. And not to do so meant losing the initiative on the ground.

In transitional arrangements the role of territory is critical because it is the physical expression of the jurisdiction of a political authority. To fail to control the territory and proclaim that jurisdiction was to fail to have any authority. And UNOSOM II failed to establish its political authority. Instead, it could respond to hostility imperfectly by remaining calmly behind defensive walls or bombing inefficiently and destructively. The area in front of the headquarters, the size of a city block, was not just reduced to rubble, but to pebbles and then cleared. It was flat and empty. But what about the next block, and the block after that?

The neat lines on paper of the AORs did not accurately reflect actual deployments or represent UN control. They were the operational limits of specific contingents, but quite another map that had been drawn by the factional balance of forces defied any AOR boundaries which were not expressed on the ground in any meaningful sense. Yet another map of clan habitation existed by itself. Separate from all of these were the political-administrative boundaries of the Somali regions, which were the most important for UNOSOM II’s endstate goals, but which had no physical representation on the ground at all.

As late as January 1994, the AORs were realigned again as a consequence of the strategic reset. The new operational boundaries were further disconnected from the political boundaries. The AORs became larger outside Mogadishu and were turned into meaningless expanses with arbitrary deployments at population centres, dictated less by the logic of a plan than the politics of inter-contingent relations. This had been progressively the case from the task-specific HRSs to a new, but less clear mission under OPLAN 1 and then a strategic reset that abandoned some of the logic and the tasks of the original UNOSOM II security imperatives under OPLAN 2. In Mogadishu, by contrast, the focus of deployment had narrowed to such an extent that the majority of contributing nations were planned for inclusion in the single Mogadishu/Medina Command. AOR Marka-Mogadishu was divided into AOR Marka and AOR Afgooye, which included northern and western Mogadishu. AORs Baidoa and Kismayo were joined together as AOR Baidoa-Kismayo, and AORs Belet Uen and Gialalassi were joined as AOR Balad. Disregarding the political boundaries was the continuation of a mistake at a time when more reliance than before was placed on the development of regional and district administration. Yet, military operational areas were unrelated to the UNOSOM II civilian zones being established in pursuance of the regionalization strategy.

There were two particular factors that affected redeployments in Somalia. First, not all contingents had the same capability: they varied in numbers, kinds of equipment and styles of operation. Their logistical limitation affected their range of operations and degree of self-sufficiency. Developing nations tended to rely on UN logistics while Western nations were better equipped for self-sufficiency.

Second, by far the most debilitating factor was the unwillingness of contingents to move. It was unrealistic for a contingent deployed in the field not to expect to be moved, yet many did not and refused to redeploy. In addition to operating under a variety of guidelines and restrictions, contingents deployed with different motives which were reflected in their geographic preference. The Indians and the Koreans would not go to Mogadishu. Others would not go elsewhere. Security for some was defined by the distance from the airport; others felt the further they were from Mogadishu, the safer it was.

Therefore, UNOSOM II was left with units refusing to move from certain areas. The Force Commander had to be prepared to ask the government of a contingent to get troops to move. Some contingent commanders were willing to ask their governments on behalf of the Force Commander. In such a case, a subordinated contingent commander was in a position to participate in decision-making above the Force Commander’s head. This was hardly a flexible arrangement able to meet rapidly changing ground conditions in which warlords not only controlled the territory but could respond to evolving situations more quickly and effectively. This would have disastrous consequences during the combat phase when contingents were attacked and assistance was slow to come.

2. Civilian Organization

Technically, UNOSOM II was divided structurally between civilian and military affairs and staffed respectively by a Chief Civilian Personnel Officer (CCPO) and a Chief Military Personnel Officer (CMPO). Despite its organizational purpose, the distinction was of minimal consequence: although the SRSG as a civilian had ultimate authority in the field, his minimalist staff was overwhelmed by what in UN terms was an enormously powerful military command.

The civilian personnel were recruited on an international basis from outside or within the UN system by the field operations division of the Department of Peace-Keeping Operations (DPKO). As late as February 1994 there were still only 400 civilians in the operation of the 2800 expected. These were divided evenly into “substantive” and “administrative” categories. The latter provided various services for both the military and civilian staff, including catering, building management, transportation, computers, local staff and other necessary support.

If UNOSOM II was to succeed, the burden of the operation should have rested on the substantive division of the civilian staff. Whatever the problem of military deployment, the civilian staff deployments were of a different dimension of dysfunction. The civilian body virtually did not exist and what did reach the field was stuck at headquarters in Mogadishu. Without a staff the aims of “peace-building” and rehabilitation and a final political solution including the marginalization of the factions were quixotic. The absence of civilian staff is a perennial problem of UN operations. Despite desperate needs in the field, human resources officials at headquarters in New York create a bottleneck preventing resources from outside the UN system from reaching empty offices and meeting requirements in perpetually short-staffed missions. It may be less expensive to rearrange and deploy internal UN staff to the field, but the net result is that civilian components are left empty-handed in hostile environments, or in environments that become hostile because of a UN failure to meet basic local expectations.

This blockage is sometimes breached when the field headquarters requests the appointment of a specific individual. Since component officials are best qualified to identify needed expertise, this means dividing the time of small civilian resources between seeking staff and fulfilling their mandated substantive tasks. It also implies that staffing is done well after the date of deployment and into the substantive phases of the operation. In this manner, all political initiatives are lost on the ground and by the time additional staff arrives, there is a confrontational pattern between the UN and local factions that becomes a stalemate the UN tends not to manage to break.

The civilian staff seemed not a priority at all. One reason for this was the vague political objectives of the mission. Without a clear identification of tasks that defined the political goals, little foresight anticipated the civilian staffing requirements. No one had thought seriously how to conduct a political operation, nor wanted to admit what it really meant – calling it “assistance” when in fact it was governorship. Given the security imperatives, a military OPLAN was the centre of attention. To which tasks were civilian strengths being matched? Without answers to this question, civilian staffing was quite arbitrary and infectiously slow. And since desperately few civilians were in the field, this in turn retarded the development of an operational concept.

Instead, the sub-culture of the civilians that did reach Mogadishu dictated their approach to the problems in Somalia. Diplomats emphasized the need for negotiations; lawyers were frustrated by the absence of judicial institutions or understanding of them in an inhibiting UN bureaucracy; specialists in public affairs were frustrated by the tendency for propaganda by the military face of the operation; and humanitarian affairs specialists continued to emphasize the primacy of assistance. There was no means to bring these disparate elements into any coherent and comprehensive approach to the problems of Somalia, or to develop common goals for each. A disunited front prevented UNOSOM II from displacing any faction, let alone all the factions as a whole.

The substantive personnel were organized according to several offices and divisions, including the Office of the SRSG; the Political Affairs Division; the Office of the Political Advisor to the SRSG in Nairobi; the Office of the Legal Counsel; an Information Office; the Public Affairs Division; the Media and Spokesman’s Office; the Demobilization and Demining Division; the Humanitarian Affairs Division; a Security Office; and the Justice Division.

There was no electoral component, even though an election was not only felt in UNOSOM II to be the ideal goal but in fact it was considered part of the exit strategy, implicit in the democratizing mandate and the end-state of the Addis Ababa March Agreement. While it was premature to consider an election in the early phases of UNOSOM II, aside from the poverty of the idea of an election altogether, it was nevertheless the time to plan for it. Had elections been seriously planned for, it would have put more pressure on staffing for political tasks in order to create a framework and conditions for registration and balloting. Not to plan and prepare for an election at the earliest stages of UNOSOM II was tantamount to deleting it as a potential goal altogether.

The military Force Commander was hired as a civilian in the Office of the SRSG. While the military reported to him, this did not subordinate the military effectively, even if it did juridically, to the civilian authority of the SRSG nor integrate in a balanced way military and civilian affairs. In fact, it served as a channel by which the military weight subsumed both civilian authority and civilian affairs. Both were dragged in the direction military forces evolved according to ground conditions, which not only led to a combat phase but lost its way in it. That the SRSG was himself a military officer did not prevent him from maintaining the independence of the office as a civilian office. But this separateness was simply overshadowed by the mass of military resources surrounding the office, the inertia of a military mission preceding UNOSOM II and the military imperatives of the combat phase.

Civilian personnel tended not to have problems with the military in principle. There was accommodation with those contingents that reported to the Force Commander. However, they felt that the separate chain of command of the US forces caused friction. There was the feeling that the Americans were used to a different standard of resources and although they had brought their own equipment, they wanted more. On the one hand, it was a historical tendency of the US war machine to have larger numbers of resources than were required to accomplish a task. On the other hand, whatever the US shortcomings, the UN approach to staffing was fundamentally dysfunctional. While it is a historical tendency of some national militaries, such as Britain’s, to do the job with less than is required, it is a UN tendency to have so few resources as to prevent the job being done.

The Justice Division was further divided into a Police Academy, in which police advisors were to train Somali police, and prisons and judiciary subdivisions. During the strategic reset, as more reliance was placed on the police to facilitate UNOSOM II military disengagement, the Director of the Division, Chief A.A. Adedeji, devised a new plan in which six subdivisions were created for the Police; the Judiciary; Corrections; Juvenile Justice; Crime Prevention; and Human Rights. There were very serious staffing difficulties and after a combat phase and the degree of danger both in Mogadishu and the regions, it became extremely challenging to find qualified individuals willing to commit to a 6-month contract in Somalia. Reliance on the Justice Division to replace the combat phase and facilitate withdrawal, yet with the lack of staff to accomplish even a fraction of the necessary tasks, meant there would not be a transfer of any kind, just withdrawal that amounted to retreat or abandonment. This could have been foretold in the negligent staffing practice before and at the start of the operation.

UNOSOM II disengagement from combat operations turned the Justice Division into cannon fodder, placing it on the front-line of the mission’s objectives, yet not supporting its talented officials with adequate means to accomplish the breadth of the tasks they had had to assume. This reflected a lurch from the military emphasis to the originally intended “peace-building” objectives, yet without a clear concept of operations and therefore insufficient preparation and support. A new set of tasks would have to be approached in isolation in the wake of warfare. There was no overall glue or framework. The mission seemed to decompose into threads that eventually would be consumed by intensified anarchical conditions.

The staff of the Justice Division tended not to have come from within the UN system and frustrated with their lot they concluded that the best way for those in the system to survive was to accomplish nothing. To do otherwise seemed to create too many enemies, in an organization in which survival and advancement were best served by exercising influence over bureaucratic territory as a fact. The marked exceptions to this bureaucratic sub-culture were specific seasoned UN officials, such as in the Political Division and the Office of the SRSG. Such good people were too few in the paucity of civilian staff.

The opposing view to the Justice Division was held by the Office of the Legal Counsel located across the hall from it in the UNOSOM II civilian headquarters building. While the Justice Division focused on reconstitution and rehabilitation of the Somali judicial infrastructure, the Legal Counsel fulfilled the position of the legal advisor to the executive. It was bureaucratically logical for the office to legally justify or obfuscate the facts of operational mistakes. It was composed of loyal officials from within the system whose task it was to apply the law not objectively but according to political convenience. Their poor calculations, however, proved not only embarrassing for the operation but were counterproductive for the executive they served. For instance, the legal construction used to justify long-term detentions relied on the broad powers under Security Council Resolution 814 and specifically paragraph 5 of Security Council Resolution 837 of 6 June 1993, authorizing the Secretary-General to investigate, arrest, detain for prosecution, trial and punishment those responsible for armed attacks against UNOSOM II. Specific detentions were authorized by directives of the SRSG referring to these provisions. Following the detention on 21 September 1993 of Osman Hassan Ali (Atto), the SNA number two, David Ijalaye, Senior Legal Advisor to the SRSG, described the detention as a “preventive measure”. [48] However, in the absence of a means of prosecution and due process, this was rejected as a contravention of Article 9 of the International Covenant on Civil and Political Rights by the Independent Jurist, The Honourable Enoch Dumbutshena, former Chief Justice of Zimbabwe, dispatched by the Secretary-General to investigate the grounds for detention [49].

Furthermore, another construction used to respond to violations of humanitarian laws of war by UNOSOM II was that the UN as an international organization was not a party to the Geneva Conventions and therefore not bound by its provisions. This perverted argument contradicted the UN’s own standard setting in the past. It was a contravention of the UN Charter: the powers of the Security Council are circumscribed under Article 24(2) by the purposes and principles of the UN, including respect for human rights and fundamental freedoms under Article 1(3). Also, the Geneva Conventions are universally subscribed to and are considered customary international law. That they are generally applicable without any derogation being permitted has been explicitly affirmed by the International Tribunal for the Former Yugoslavia. The construction put forward was resoundingly criticized [50]. The Office of the Legal Counsel failed to ensure that justifications for executive actions were in accordance with universally recognized humanitarian and human rights standards.

As part of the return to a strategy of regionalization after the end of combat operations, a proto-civilian administration of Somalia started as Zone Offices began to be established. The country was divided into 6 Zones and 18 Subzones. Each Zone was headed by a Zone Director, who was also responsible for each Subzone. There was supposed to be a representative of each substantive component in each Zone. In the field, however, there were a number of missing personnel due to lack of appointments and a number of offices, such as Belet Uen, which had to be evacuated as a result of attacks by local factions. Zone Offices were under siege well after 1993.

Each component representative was under the authority of both the component head in Mogadishu and the Zone Director. It was conceptually ambiguous who was in charge, but effectively component representatives would be more responsive to the component office in Mogadishu since that is where substantive decisions were being made for the most part. This fragmented the functioning of the zones since each representative had a different set of instructions, different tasks and a different agenda. Zone Directors behaved more as component representatives, reporting to their substantive component headquarters in Mogadishu, the Political Division, and attempting to respond to its goals rather than uniting all component representatives and responding to immediate, local ground requirements.

The problem was particularly acute in the relations between the Zone Director and the military contingents deployed in the area. The civilian representatives were reliant on the security framework provided by the military, which meant that the space of zone operations was dictated by the physical boundaries of military operations and not the administrative boundaries of the zones, Somali regions or even AORs. Also, the coherence of military contingents was much greater than the nascent, fragmented civilian organization, and therefore although the Zone Director may have had authority over the zone as a whole, in fact he was effectively subordinated to the military and its imperatives, given the greater influence over the environment it could wield. If a Zone Office was attacked, that the Director was not more than a visitor and guest of the military deployed locally was illustrated dramatically as he sought sanctuary amongst the troops. This was not the way to replace military responsibilities with regionalization.

The idea should have been to give the Zone Director decision-making authority over all the local civilian and military staff and personnel. Local strategies needed to be devised jointly in a single political body for the Zone. Rather than each component responding to the component office in Mogadishu, UNOSOM II Headquarters needed to be jointly organized under a single political decision-making authority which devised a single strategy in which the military and civilian components as a whole were orchestrated. From this single strategic source, political goals would be communicated to each Zone Director, who would then orchestrate the implementation of a single policy jointly through component representatives. This is at the basis of any concept of interoperability. This does not imply complete severance of links between contingents, component representatives and Headquarters; on the contrary, these links should serve to underwrite effective implementation of policy, and as subordinate and administrative they should not develop independent policies but provide means to implement political strategies and tactics.

3. Military Force Organization

UNOSOM II military headquarters was divided into seven branches: (U1) Personnel; (U2) Intelligence; (U3) Operations; (U4) Logistics; (U5) Civilian-Military Affairs; (U6) Signals; and (U7) Engineers. Each was headed by a Chief Officer-in-Charge with the rank of Colonel. This was an essentially North Atlantic Treaty Organization (NATO) structure, although the NATO nomenclature with “S” rather than the “U” designation had been flatly discarded by the Pakistani Commanding Officer, Brigadier-General Imtiaz Shaheen. The branches fell under the responsibility of the Chief of Staff (COS), who in turn was responsible to the Deputy Force Commander (DFC) and Force Commander (FC). National contingents were individually responsible to the DFC and FC.

Figure Five: UNOSOM II Force Headquarters Organization (not included)

In addition to the branches, there were “Special Establishments” responsible to various offices in the high command. Under the COS there was a mixed civilian and military Public Affairs Office; the Office of the SGS; the Office of the Force Provost Marshal (FPM); and an entirely military Legal Cell with a maximum capacity of six. A Force Information Unit was responsible to the Chief Military Information Officer (CMIO) and included: (i) a Human Information Unit, and in turn an Investigation Section; (ii) a Signals Information Section; (iii) an Imagery Information Section; and (iv) Operations Planning, which included in turn an Analysis Section. There was under the Operations Branch a Joint Military-Civilian Committee (JMCC). The Logistics Branch included Movement Control, a Supply Unit and the Postal Service. The Signals Branch had a Composite Signal Support Unit. Under the Camp Commandant was a Turkish Defence Company with: (i) a Signals Section; (ii) a Supply Platoon; (iii) an Engineers Section; (iv) a Mechanised Infantry Platoon; (v) a Medical Section; (vi) a Weapons Platoon; and (vii) a Transport Platoon. There was also a Norwegian Headquarters Company with: (i) a Logistics Platoon; (ii) a Transport Platoon; (iii) a Headquarters Service Platoon; (iv) a Military Police Access and Security Platoon; and (v) a Catering Platoon.

Figure Six: UNOSOM II Branches and Special Establishments (not included)

The CMPO included sections for Personnel Administration and Manning; Security and Discipline; Welfare; and later Protocol. It was the responsibility of the CMPO to keep casualty figures of the military while civilians kept their own casualty records. It was the responsibility of each contingent to finally identify and send home remains of their nationals, but there was a division of responsibility with the centre in which UNOSOM II mortuary facilities served as a joint “clearing house”. Unfortunately, no figures were kept of Somali casualties, although the policy of OPLAN 1 stated that “UNOSOM II will be the executive agent for mortuary affairs support to coalition forces and Somali Nationals who die as a result of hostile or non-hostile actions involving UNOSOM II forces” [51].

These were the first mortuary facilities of their kind attached to a UN operation. They were provided by the US as a service since the UN did not have the experience in managing casualties on a large scale. Located on the edge of the airport next to the sea, the facility had stacks of metal coffins strapped together in blocks of 25. It was a new sight for UN operations, whose planners had not come to grips with the inevitable cost of dangerous missions. Denial of the cost of casualties on the part of the UN and the endemic US perception of “enemy” and “ally”, led to the exclusion of accounting for Somali casualties and resentment on the part of Somalis.

The CMPO had liaison officers from each contingent that were supposed to supply the centre with relevant information. However, most decisions were made between New York and U3 Operations. Therefore, personnel priorities were not set but only implemented by the CMPO. Whether priorities were misjudged or merely mishandled, as a result of painfully slow personnel deployments characteristic of UN staffing in New York or reluctant national capitals who chose to send whom they wished rather than responding to what was needed, key parts of military staffing were under-represented. There was so much concern with numbers and getting troops out into the field, redeploying them, and keeping them supplied that those parts of the military organization that could best support an overall transitional process at the political interface tended to be weak links in the chain – perhaps also because of their subordination in the military subculture to the prestige of fighting units.

These tended to be cost-effective, specialized and professional units that ought to be emphasized over the focus on more and more troops. For instance, the Force Provost Marshal Company was authorized to have a strength of 22, but only had 14, some 64% of its capacity. As hostilities intensified UNOSOM II arrests of Somalis increased, and as the implications of this increased in significance military policing and legal offices could have played a much more significant role than they did, linking the use of force to the logic of law and order rather than reacting to combat imperatives.

Following the late 1993 strategic reset, under OPLAN 2 there was a general reduction by 15% of personnel. Some positions were cancelled while in others individuals were replaced with new arrivals. There was a large-scale change-over in March 1994 in which there was a rotation of all the chiefs of branches. Indians replaced the US Chief of Operations and the Belgian Chief of Signals. Pakistanis replaced the Italian CMPO and French Logistics Chief. A Malaysian assumed control of Humanitarian Affairs and the Italian Chief of Intelligence was replaced by another country.

As the combat phase intensified, on 11 August 1993 U5 Civilian-Military Affairs was reduced and U3 Operations was expanded, diminishing further the critical means of fostering popular support and forcing greater reliance on military force. U5 used to be composed of Public Affairs; Humanitarian Affairs; Civil Liaison; and Legal Affairs. After 11 August, it was reduced to include only Plans/Operations and Civil Liaison. Its focus turned entirely onto military issues. U3 initially included Plans; Current Operations; Future Operations; Air; and Movement Control. After 11 August it was expanded to include the JMCC; Psychological Operations; Future Operations; Naval Operations; Current Operations; Plans; Air Operations; DIS/CF; Air Field Operations; and Liaison. This also reflected greater focus on Mogadishu.

Separate from the UNOSOM II force organization was the Quick Reaction Force under US command. Howe had requested a QRF and the US agreed to provide it. According to OPLAN 1, its tasks included:

  1. On order, conduct military operations in response to hostile threat and attacks that exceed UNOSOM II military force capabilities and assist in military-oriented operations that are beyond the capabilities of UNOSOM II military forces.
  2. Be prepared to conduct military operations as an initial entry force to facilitate expansion of security throughout Central and Northern Somalia.
  3. On order, provide reaction force in support of contingency operations. Priority for planning and execution is AOR Kismayo and Mogadishu City followed by remaining AORs. Conduct necessary planning for full range of contingency operations in Somalia.

Despite its priorities regarding the two main ports, Kismayo and Mogadishu, the QRF was designed for the Somali countryside and not for combat operations in Mogadishu, which had been relatively quiet during the planning stage. US officers in Mogadishu felt this had been bad planning since in their opinion the QRF could not contend with combat operations in Mogadishu. It was composed of Blackhawk helicopters; humvees; APCs; and light machine guns. US officers had found it difficult to convince either the UN or Washington that tanks were needed, because, they said, no one appreciated how the conflict would escalate. They claimed AC 130s and attack helicopters could not reach SNA areas. Secretary of Defence Les Aspin was reluctant to ask for tanks as the combat phase escalated since military resources had to be balanced.

But were tanks the answer? They too would be more effective in open areas unless there was an intention to destroy large parts of Mogadishu, which would have been counterproductive. Again, the problem was not the available force but the need for a better, limited approach to force that emphasized environmental security rather than relied on powerful force after the fact. For instance, helicopters were shot down with RPGs. The US thought this could be done only with launchers, but the Somalis used 3 point cross-referencing and did it. They launched attacks with 100 grenades at a time. They had heavy 12 point machine guns, 60mm mortars, 82mm mortars and 120mm mortars, which they did not use. Also, it was rumoured that Stinger missiles were present, though none were fired. Tanks could not have been relied on to neutralize such means of attack. A better investment would have been made in intelligence through contact with Somalis; and better still would have been a secure environment that was understood to mean physical control of the whole city and neutralizing potential attacks on the basis of information.

Under OPLAN 2, the QRF was placed under the UN Force Commander’s command and control. Each Brigade was to provide its own QRF. There was a problem of some contingents refusing to serve under other contingents, so there was a need for an international Divisional Headquarters. If there was to be more combat, however, the Force Commander’s Headquarters needed to be both the staff and tactical headquarters, and that was a problem. So the Force Commander commanded through the Brigade Commanders. The new QRF was smaller and lighter: it was composed of Malaysians, Pakistanis and Zimbabweans and included a mechanized company, a tank troop, a light company and a headquarters battalion.

4. Command and Control

The command relationships for UNOSOM II were largely devised by the US in order to permit on the one hand withdrawal of US forces by transferring leadership from the US to the UN, and on the other to balance the inevitable need for US involvement, at least in the initial transitional phases of UNOSOM II, and the US desire to avoid submitting US forces to UN command as much as possible. In a genuine sub-contracted operation in which the UN decides to appoint an agent to fulfil a task on its behalf, there is a separation between political control by a UN body, such as the Security Council, with the Secretary-General and the secretariat as an intermediary link, and military command by an individual state or a group of states. This relationship has not been adequately defined, but an accountable sub-contract should include at least three conditions. First, it should be clear that the state or group of states is acting on behalf of the UN and that the link between the two is direct. Second, given that the command of the operation may be not functionally part of the UN, instructions to its agent – the appointed force – must be clear, specific and incontestable. Third, the agent must be directly responsible to the authority of the UN. If UNITAF had been an ideal sub-contract, the axis of command and control would have turned on the separation of UN political control and coalition command.

UNOSOM II, however, was not a sub-contract nor an exclusively military operation in design. In fact, it represented a shift from something of a sub-contract to an integrated UN operation. Therefore, the cost of US withdrawal for the Pentagon was to devise some kind of arrangement that could be acceptable to both the US and the UN. It turned out to be a compromise and not entirely palatable to either. In the development of a model, the axis shifted from a horizontal position separating political control above and military command below, to a vertical position separating UN and US military command side by side. This reflected a shift from civilian political to military concepts of command and control [52]. It effectively de-emphasized political direction of military forces, further to the US Weinberger principles which dictated military independence in the field. And although the UN Force Commander was responsible to the UN SRSG, each military contingent was separately linked to its home capital, therefore obscuring any political role on either the side of the US or the UN.

The question arose as to the relationship between the UN and US chains of command. Focus shifted from strategic command and control relationships to operational and tactical relationships. A model was determined by dividing responsibilities at this lower level. This model would come to be reflected in the US Army Field Manual being developed throughout 1993 for US involvement in peace operations [53]. It subsequently entered the literature, outside the specific context in which it was devised, as an obvious distinction to be made for US-UN cooperation.

The UN chain of command reflected in form that of a conventional, integrated peacekeeping operation. Brigade Commanders reported directly to the UNOSOM II Force Commander, who in turn reported directly to the SRSG. The SRSG was responsible to the Secretary-General, who in turn reported to the Security Council, the mandating authority. However, OPLAN 1 stated in its Annex J on “Command Relationship”: that “The relationship between COMUNOSOM II (UNOSOM II Command) and assigned coalition forces, unless modified by the coalition government and/or the UN, will be the same as the relationship between COMUNITAF (UNITAF Command) and these coalition forces.” Annex J had been drafted by the US Chief Operations Officer (COO) and reference to the term “coalition” clearly reflected this.

However, there is a difference between a coalition and a “UN operation”. A UN operation is “collectively” mandated and controlled. This means that although it has been mandated by and is responsible to only 15 nations on the Security Council, because of the position of the Council in the UN system, the implication is that the operation is an instrument authorized by and in the service of the interests of the international community as a whole. In other words, it represents something more than the sum of the nations participating in the operation in the field or in the decision-making process. “Collective”, therefore, refers to a position in law.

“Coalition”, by contrast, merely refers to cooperation and a venture in which more than one nation is participating; it is an operational characteristic. A coalition is not something more than the sum of its parts. It refers most specifically to the US strategy of spreading responsibility in a military venture, used in the Gulf War and then in UNITAF. It was developed both for practicality, to divide the cost of resources amongst a number of nations, and also for legitimacy to ensure widespread support politically for a venture – which needs to be maintained even if it might contravene law. It is politically a step away from unilateralism, but it is still possible for a coalition to be politically supported and not legally legitimate if there is a sufficient constellation of interests in contravening law, such as in the Gulf War, during which unlawful aspects of enforcement were overshadowed by the lawful right on the part of the international community to use force to reverse aggression.

The position of a military coalition in law is the same as a unilateral force or an alliance such as NATO. Consequently, it may use force only in self-defence under Article 51 of the UN Charter. It is possible for a coalition operation to be a legitimate collective action, but only if it is a genuine agent of the UN and mandated by a UN body, such as the Security Council or the General Assembly. This would have to be a genuine sub-contract arrangement and not a coalition operating under a vague, fig-leaf resolution of the Security Council. As a genuine agent, acting as an instrument of the international community as a whole and in international interests, it would fly the UN flag. Coalitions in the Gulf and in Somalia did not, since the gap between a collective position in law and the operational imperatives of coalition warfare have not been bridged.

Another, perhaps more fatal, distinction between a UN operation and a coalition has to do with relations between contingents and Force Commanders. The critical mistake of OPLAN 1 was to perceive command relations for UN operations to be the same as in coalitions. This led directly to the fragmentation of UNOSOM II. Despite the proverbial tendency of contingents in both peacekeeping and more challenging types of operations to maintain separate communications links with and to continue to receive instructions from home capitals, even after subordination to a collective UN command structure, in a UN operation contingents are supposed to be strictly under the authority of the Force Commander. They are not meant to behave independently at all, except in the internal workings of the contingent as part of the natural course of implementing the Force Commander’s instructions.

Although OPLAN 1 stated that Brigade Commanders were supposed to report to the UN Force Commander, placing this procedure in the context of a coalition framework effectively widened the gap between Force Commander and contingents. The reason for this is that in US-led coalitions it is possible for contingents to behave independently to a certain degree without adversely affecting the coherence of the operation because the strength of the US, as a kind of centre of gravity, acts like a glue for the coalition as a whole. It is likely in any case that contingents will be approaching a coalition with a certain amount of independence since each nation is participating in a coalition as an independent member of a cooperative venture, as opposed to the UN ideal of an integrated and collective command structure. In the practice of US coalition-building there is an effective subordination of participating contingents to US leadership, but the relationships are still much looser in design than they are in a UN command structure.

Therefore, in the transition from US to UN leadership, what was already a fragile structure in conventional peacekeeping operations would become uncontrollable for the UN Force Commander once the problem was exacerbated with the addition of a coalition looseness in the wake of UNITAF. The degree of independence operationally permissible for contingents in UNITAF, and the removal of the binding influence of the US, meant the contingents could not be tamed by the new UNOSOM II Force Commander. The gap was never bridged. The attempt to transfer a coalition to a UN command structure without devising a means to further integrate those contingents in the manner necessary for a collective command structure, resulted in the existing coalition fragmenting further: like satellites revolving around a large centre with a powerful gravitational pull that is suddenly replaced with a much smaller body with a pull weaker than any individual satellite, UNOSOM II contingents floated in their independent directions. In answer to the question of whether the US really expected the UN to be able to fill the gap that would be left by the US in the transition, a senior US military official stated, “Yes. It was a bad call.”

The role that the US had designed for itself in UNOSOM II was insufficient to prevent this break-up; instead, it had become like a large satellite moving in its own direction, to which gravitated a smaller UN body, almost like a satellite itself, leaving national contingents very much on their own. The US could retain not just effective independence, but juridical independence as well from the UN command structure, unlike all other contingents which were juridically subordinate as part of the collective. The US could at the same time remain part of the operation by designating dual US-UN command roles for two of its officers: the Commander of US Forces in Somalia was designated also as the Deputy Force Commander of UNOSOM II; and the Deputy Commander of US Forces in Somalia was designated also as the Commander of UNOSOM II Logistics Forces Command as well.

Furthermore, the US Commander-in-Chief of Central Command in the United States retained command of all US Forces assigned to UNOSOM II and would assign operational control (OPCON) to the Commander of US Forces in Somalia. When directed, the Commander of US Forces in Somalia would assume tactical control (TACON) of the US Quick Reaction Force. Consequently, while it could be said that US personnel participated as part of a UN structure, this was also something of a fiction since a US commander was always in control, and ultimate command rested with the US and not the UN Force Commander. It was rather as if a US unit was attached by its side to a UN operation, but its top was attached by a string, through the link of a US commander, to coexisting but independent US forces. There was further still another US link above the UN Force Commander and that was the American SRSG, Admiral Howe.

Figure Seven: UN and US Command and Control Relationships (not included)

There was a period when the US was running operations on a US-only basis. In the case of logistics, for instance, US decisions would have priority, especially when deciding whether or not to put troops at risk. The US Commander could override any decision of the UN Force Commander, since the two hats of the US Commander were as the UN Force Commander’s direct subordinate and indirect superior. The ultimate delineation of command and control for UNSOM II, then, was overall US command and a kind of UN operational control. Not only was this not an integrated command structure, it was in fact the reverse of a legitimate sub-contract in which a single nation or coalition command was under the direction of overall UN political control. However, what prevented this from being effectively a unilateral operation was the existence of a collective mandate, the direct role of the UN Secretary-General in operational decision-making and the participation of other nations subordinated on paper to a collectively accountable UN Force Commander. The command and control failure was to retain a coalition approach to UNOSOM II under the prevailing conditions.

The tribalism that resulted proved operationally dysfunctional. In a US-led coalition, at least some common procedures are agreed on by participating nations prior to deployment. Even if operational concepts of contingents were not thoroughly harmonized, the sheer weight of US leadership ensured the effectiveness of the operation. So, for instance, under UNITAF, contingents were deployed to different geographic areas with vastly different leadership styles. Although this tended to alienate the local population, UNITAF had short-term goals so a patch-work impact locally was not critical. Such an inconsistent style would be quite damaging in long-term reconstruction activities, however, given the need for regional harmonization to ensure national unity. Operational styles varied from the Belgians, for instance, whose roughness scared Somalis so much that they used to take their clothes off to show they were not carrying weapons, to the Italians who mingled with the local population and were seen as friends.

These variations would be fatal for UNOSOM II. As the heavy hand of the US was increasingly disliked by Somalis in the combat phase, the Germans were positively regarded partly because the logistic resources they had brought for the Indian contingent were being given to local Somalis. But this difference in behaviour also reflected different political agendas and perceptions of the nature of the operation. Many participating nations did not appreciate what Resolution 814 meant or what it would take to implement, and they disagreed on its meaning. So there were always different responses to the Force Commander’s orders: some followed his command, others refused to comply, and some contingents complied only after getting approval from their home capitals. What was a conventional problem in traditional peacekeeping operations was exacerbated as the operational environment became more dangerous. And while this insubordination was not critical in the less complex peacekeeping buffer zone, it proved crippling as unity of command became necessary to respond to quickly evolving, life-threatening conditions.

This was no longer a question of differences about interpreting a gray mandate, but a black and white difference of opinion about the use of force. So distinct was the disagreement amongst contingents that they could all label themselves as a “Chapter VI” or “Chapter VII” nation. This shorthand was used to differentiate those contingents that had arrived in Somalia prepared to use force proactively from those that were not prepared to use force except in self-defence, even though in legal terms this was a simplistic distinction between Chapter VI and VII of the UN Charter to the point of inaccuracy. Although the operational mandate of UNOSOM II was under Chapter VII, each contingent interpreted this differently according to their political and military culture. The Germans based their interpretations on their constitutional limitations at the time. Some countries only came to the field based on promises that they would be protected; to ensure the participation of certain nations, apparently the US had promised this to them. There was a long process of clarification which took place on the ground.

Battalions had been informed about the nature of UNOSOM II, but it also evolved over time. As it did, headquarters pushed the United Arab Emirates (UAE), Saudi Arabia, Kuwait and the Germans to change, but they refused. For instance, in August 1993 the Saudis were ordered to protect a certain perimeter, but they replied that this was outside their understanding of Chapter VI. The Germans interpreted Chapter VI to mean protection of their own camp in Belet Uen, but not anything outside. So the Italians, a Chapter VII contingent, had to protect them, NGOs and local civilians in the area. Furthermore, the Australians had their own directives from Canberra. In fact, each country probably had similar instructions from home capitals, but there was no summary of all of these documents and the Force Commander did not have a collection of them. He discovered each operational concept by trial and error, more by surprise than by being informed.

5. Rules of Engagement

This kind of institutionalization of separate operating procedures permitted by the persisting coalition formula defied the official UNOSOM II rules of engagement . Chapter VI contingents devised their own rules, maintained traditional peacekeeping ROE or simply refused to accept the OPLAN 1 ROE. In addition to this, there ought to have been the problem of interpreting the meaning of the OPLAN 1 ROE. The US drafted them and on the surface they were reasonably restrictive. But the US interpreted the ROE according to their heavy-handed use of force sub-culture, which quickly escalated into heavy force. US military criticism of the UNOSOM II experience emerged later, claiming that more force used earlier would have yielded greater success, rather than recognizing the need for a limited force concept, for built-up areas such as Mogadishu, that could be relied on for guidance in the interpretation of the ROE. For Chapter VII contingents, there was a tendency not to identify some general limited force concept but to determine how best to cooperate or avoid cooperation with the US interpretation of the ROE that dominated UNOSOM II.

According to OPLAN 1, UNOSOM II personnel could use deadly force: (a) to defend themselves, other UN lives, or persons and areas under their protection against hostile acts or hostile intent; or (b) to resist attempts by forceful means to prevent the Force from discharging its duties. “Hostile Acts” was defined as: “the use of force against UNOSOM II personnel or mission-essential property, or against personnel in an area under UNOSOM II responsibility.” “Hostile Intent” meant “the threat of imminent use of force against UNOSOM II Forces or other persons in those areas under the control of UNOSOM II.” Whenever practicable, a challenge was to be given before using deadly force, either by shouting in English or Somali, “UN, Stop or I Fire”, or by firing warning shots in the air. Furthermore, the principles for the use of force included the prohibition of any action which could reasonably be expected to cause excessive collateral damage; the forbidding of reprisals; and the use of minimum force at all times. “Minimum Force” was understood to mean “the minimum authorized degree of force which is necessary, reasonable and lawful in the circumstances”.

In essence, this resembled the conventional peacekeeping rules of engagement, which restricted the use of force to self-defence, although self-defence as interpreted to mean not just defence of the person of the peacekeeper but of the mission objectives as well. However, the interpretation of this latter provision regarding the scope of the use of force beyond the defence of the person distinguished UNOSOM II from peacekeeping. Specific rules regarding this were identified:

  1. UNOSOM Forces may use deadly force in response to a hostile act or when there is clear evidence of hostile intent.
  2. Crew-served weapons are considered a threat to UNOSOM Forces and the relief effort whether or not the crew demonstrates hostile intent. Commanders are authorized to use all necessary force to confiscate and demilitarize crew-served weapons in their area of operations.
  3. Within those areas under the control of UNOSOM Forces armed individuals may be considered a threat to UNOSOM and the relief effort whether or not the individual demonstrates hostile intent. Commanders are authorized to use all necessary force to disarm and demilitarize groups or individuals in those areas under the control of UNOSOM. Absent a hostile or criminal act, individuals and associated vehicles will be released after any weapons are removed/demilitarized.
  4. If UNOSOM Forces are attacked or threatened by unarmed hostile elements, mobs and/or rioters, UNOSOM Forces are authorized to employ reasonable minimum force to repel the attacks or threats. UNOSOM Forces may also employ the following procedures: verbal warnings to demonstrators, shows of force including use of riot control formations, and warning shots.
  5. UNATTENDED MEANS OF FORCE. Unattended means of force, including booby traps, mines, and trip guns, are not authorized.
  6. DETENTION OF PERSONNEL. Personnel who interfere with the accomplishment of the mission or who otherwise use or threaten deadly force against UNOSOM, U.N. or relief material, distribution sites, or convoys may be detained. Persons who commit criminal acts in areas under the control of U.N. Forces may likewise be detained. Detained personnel will be evacuated to a designated location for turn-over to military police. [54]

In contrast to OPLAN 1, the German contingent issued to its troops a pink card with its own ROE. According to this, every soldier of the German contingent had the right to defend himself at anytime, anywhere against an attack. He could also defend against attacks which were directed towards other soldiers, civilians, civilian personnel, materials and facilities of the German contingent, or people and facilities under the German contingent’s protection. However, the German soldier was not allowed to participate in enforcement action of UNOSOM II. Nor was he permitted to intervene in confrontations between local gangs or Somali ethnic rivals. Using force was not permitted for tasks such as delivering assistance. Mines, booby traps or similar exploding traps were not allowed, even for securing the German camp.

On the use of force in self-defence, the German ROE were specific. The use of a firearm was a valid means of defending against an attack only if the use of milder means, such as pushing back, hitting with a stick, fists, or rifle butts, did not promise to be successful because of the intensity of the attack. It was not permissible to use more force than necessary for an effective defence. Self-defence, it said, was not a punitive action, and not only were reprisals forbidden, but they were punishable. Also, the soldier could not fire a weapon at an escaping criminal, such as a thief.

In the event of an attack, warning calls or shots had to precede the firing of a weapon. While the warning call was the same as in OPLAN 1, “United Nations, Stop or I Fire!”, the German ROE insisted that this be shouted in Somali first and only then repeated in English if there was no reaction. Then, only after warning shots, was it justifiable to fire in self-defence. The attack had to be immediate and directly endangering the body and life of a soldier or third party. It described the example of an attacker aiming a weapon at the soldier in a surprising manner. After the attack was over, the German soldiers had to provide medical assistance to the wounded assailant. After apprehension, an attacker’s weapon was to be confiscated and immediately handed to the local UN support group of the Somali police or the UN military police.

Despite these restrictive rules, however, there was a feeling amongst young German officers, frustrated by their limitations, that if necessary they would have violated these principles and behaved more aggressively if they personally felt the situation called for it, regardless of the consequences. Such a reaction, therefore, would not have been according to any particular ROE, either German or UN, and would have been a personal judgement. Having said this, on one occasion a Somali was shot when unlawfully entering the German camp, and one of the senior officers described the lack of psychological preparation the younger officers and men had for such incidents resulting in dead bodies.

As the combat phase proceeded, the difference between various sets of ROE widened. US military interpretations demanded a greater use of force and Chapter VI contingents entrenched their positions against participating. Paradoxically, it seemed rather than an international force uniting Somalia, Somali anarchy burst apart the UN mission.

6. Political Failure

One official of the Political Division arrived in Somalia before the deployment of UNOSOM I, as part of the Technical Mission following James Jonah’s 3 March 1992 cease fire, and remained throughout UNOSOM II. He said that during the period of UNITAF, UN political officers and the military worked together closely. For instance, “We met many times and we succeeded in convening the March (1993 Addis Ababa) Conference because of UNITAF cooperation. When we discussed with Aidid the idea of the conference, we reached a deadlock because of Morgan’s aggressive actions in Kismayo. We went to the military and asked them to bomb Morgan’s military equipment to prove to Aidid we did not support Morgan.”

“Then a surprising thing occurred,” he added, “on 7 May (1993) UNOSOM II took over and the Political Division was completely in the dark. Between 7 May and 9 October 1993 there were no relations between the political and military side of UNOSOM.” Only after 9 October did the two begin to cooperate, with the Political Division negotiating between the SNA and UNOSOM II military regarding such issues as troop redeployments. During the period 7 May – 9 October, the Political Division spent much of their time in the countryside trying to establish Regional Councils and District Councils. As late as February 1994 few had been established and were functioning effectively. And although they were ready to form a court in every district, the Political Division was holding back waiting for the rest of the country to mature – a perceptual approach which seemed to defeat the purpose of displacing the factions, which required controlling the political initiative and not reacting to events.

Paradoxically, it was the objective of the strategy of regionalization to decentralize power under decentralized anarchical conditions. But this was only half the picture: there needed to be a process of reconstituting a centre, but focus on undermining the factions served instead to fragment the country more and the inability to move beyond subverting factions meant the start of a process of ultimate centralization was never reached. The factions remained in control of this, and by their own conflict also prevented ultimate centralization. The viability of the Regional and District Councils depended on significant assistance, and this was rapidly declining due to donor fatigue. And even if the military and the Political Division cooperated in a bottom-up approach, to succeed there would have to have been a well-orchestrated political strategy of subversion of the factions with a clear goal of establishing a new authority, and not merely the logic of elections to save the UN face and exit.

Such a strategy did not exist in which the military could have been employed judiciously towards a sustainable end. A political strategy might have conflicted with a military strategy if the two had been devised in isolation, separate from one another – not unlikely in UN operations. However, the lack of political-military cooperation was not the result of a conflict between a military and political operational concept. In fact, concepts or the interpretation of the UNOSOM II mandate had never been discussed between political officers and the military. Although the Political Division remained in contact with Admiral Howe, there were never any round tables convened between components. The first such meeting occurred on 9 October 1993 when 6 principals met: Howe, his Deputy, the Force Commander, the Chief of Staff, the Political Division, and the Division of Humanitarian Affairs. Therefore, the military decision-making that dominated the mission simply disregarded the notion of operational political input from the field, and at most looked to political masters in Washington or New York. This ultimately led to a combat phase in the first place, and its lack of success subsequently.

During the phase of reception and consolidation, the increasing political isolation of Aidid led to a spark that ignited combat operations [55]. Since the Gulf War was the model to be followed in the minds of the military-dominated leadership of UNOSOM II, the intent was to force an opponent to submission. Most had not served in peacekeeping missions and were not concerned with the issue of impartiality. At the other extreme was the paralytic approach of seasoned UN officials who felt there should be exclusive reliance on negotiation with the factions. In fact, the notion of impartiality had to be reconceived as no longer pertaining to the parties, whose lack of clear consent would frustrate an operation thus reliant, but as a reference to the integrity with which a mandate would be implemented. Impartiality could not be merely disregarded in a military operation that concluded alliances with some and identified others as the enemy. Nor could there be submission to the will of the warlords. Instead, there needed to be a clear political agenda which an operation was underwriting with at least tacit support of local factions. So that if a factional element violated the overall framework, a forceful response against that faction would be regarded not as favouring one side or another, but maintaining the integrity of the mandate.

However, there was a sense in the military hierarchy of distaste for dealing with warlords and criminals. They had caused hundreds of thousands of deaths through starvation and war. UN officials argued that the factions existed as a well-armed fact and therefore had to be dealt with; and by “dealing with” they meant negotiations. And, furthermore, there had been an agreement with the factions at Addis Ababa. But the feeling in the military was that it was a bad agreement signed by criminals who should be marginalized. They were more specific: They did not like Morgan, but they hated Aidid more. Aidid’s close ally, Jess, occupied Kismayu after Siad Barre’s troops were routed by Aidid’s forces following Barre’s last bid to retake the capital in February 1992. Jess remained in Kismayu until February 1993, when Morgan infiltrated past the Belgian contingent and drove Jess out. UNITAF accepted this new status quo. But on 7 May 1993, Jess attacked Kisamyu in an attempt to throw Morgan out. The Belgians effectively repelled this and claimed the area to be “weapons free”.

By 10-11 May, Aidid’s radio station was charging UNOSOM II with attempting to be a “trusteeship”, not just because of the Kismayu incident, but also because of the process underway by the Justice Division of selecting judges for the nascent judicial system. Mogadishu was Hawiye territory and there was no consideration of the resentment that resulted from appointing an Issaq northerner or Darod southerner as judge. Appointments were arbitrary, partly as the result of ignorance and partly as the result of the arrogance of not wanting to take sensitivities into account.

On 24 May, Aidid’s spokesman, Hassan Awali, asked to meet with the Office of the SRSG and did so for 1 1/2 hours. He revealed to the UN civilian official in the meeting a UNOSOM II document in his possession stating that SNA judges were not permitted to enter Mogadishu’s barely functioning court house, and that only UNOSOM II judges were permitted to enter. The civilian official had not been aware of the document and was disturbed by it. Nevertheless, he explained UNOSOM II’s plans for the courts, prisons and police in pursuance of Resolution 814. Ali answered: “If we knew this we would never have agreed – but at least we would have known.”

In his testimony before Congress, Tom Farer, Professor of law and international relations at American University and a legal consultant to the SRSG in the summer of 1993, stated that: “the failure of Admiral Howe and his associates to discuss with Aidid and other faction leaders their plans for reestablishing the judiciary and to strive for a consensus approach was not absentminded; it was rather a calculated tactic in the service of a grand strategy.” Radio Mogadishu’s condemnation of the judicial selection process became increasingly vociferous. “To UNOSOM’s principal strategists, this response seemed another sign that they were moving in the right direction, a direction away from warlord power toward a broadly-based popular government ruling with the consent of most of the governed.”

In order to gain political advantage by taking control of the process of bottom-up regionalization, Aidid proposed to UNOSOM II in early May that they provide the resources and logistic support for a reconciliation conference to be convened at Galcayo, in central Somalia. It was potentially the most explosive area next to Kismayu, and eventually did explode. To out-manoeuvre Aidid, a factional opponent, Mohamed Abshir Mussa, the former Commander of the Somali National Police during the 1960s and, according to Farer, “who, in the view of US diplomats in Mogadishu, wore the country’s largest white hat”, suggested that the issue of Kismayu should be placed on the agenda of the meeting. UN officials in the Office of the SRSG argued that there should be one issue, Galcayo, and then the other could be dealt with, but the UN leadership wanted both. One UN official said, “I didn’t understand it was our position to dilute Aidid: he would have had less to say about Kismayu as it was an intra-Darod dispute.”

This was in contrast to Galcayo where Aidid’s proposed rapprochement between his Habr-Gedir sub-clan and the Majertin sub-clan of Abdullahi Yusuf, an opponent of Abshir, could also have served to strengthen an alliance between Aidid and Yusuf against the UN and its clients. As such, the strength of both would have increased significantly, creating the strongest alliance of warlords in the country. Therefore, the UN leadership tried to expand the scope of the conference in order to marginalize Aidid. “I realized this,” said one UN official, “when I read a speech of the SRSG in which he referred to inviting the first President of Somalia, Ali Mahdi.” In addition to calling for the discussion of Kismayu, there was an attempt to expand the participation of the Galcayo meeting as much as possible.

This manoeuvring led to a breakdown between Aidid and UNOSOM II over the arrangements for the conference. Not only was there disagreement about the guest list, but about who has host, the venue, security issues and publicity for the conference. A variety of unimportant details were the battleground for dominance of the conference. For instance, the UN tried to produce more flyers for the meeting than the Somalis. The SNA put up 1000 in a hall and the UN said this was too many and that they had to come down. This was a direct move against Aidid.

Ultimately, the opening session was never convened. Aidid and Yusuf concluded their alliance in hotel room meetings; the UN cut all assistance for the effort; and Abshir was excluded from the rapprochement, although he had been courted at one point in the early proposals for the conference. According to Farer:

this cluster of events constituted a failure on the part of Aideed to realize expected dividends. And that coming on the heels of other unpromising developments, it did nothing to reduce if it did not accelerate the momentum of his decline. From its perspective, this failure was the penultimate straw, that it underscored the precipitous decline in his fortunes and prepared him to conclude that only by a shocking bold move could he evade the point of no return.

Two linked issues proved to be the final straw: whether or not to take control of Aidid’s radio station and the effective control of heavy weapons as part of the disarmament and demobilization mandate of UNOSOM II. Aidid had declared that the radio was a national asset but that it would only be returned after the establishment of the TNC. It was in fact a key asset of Aidid’s. On 31 May, Abshir and 10 other anti-Aidid factions proposed to the SRSG to seize the station. Because of the tone of Radio Mogadishu’s anti-UN sentiments, UNOSOM II had complained to Aidid. Internally, some UN officials said the station had to be taken and the Office of the SRSG was asked to formulate a justification, as well as to consider justifying taking the court building. The Office of the SRSG found it difficult to find a relevant provision in Resolution 814. This discussion was soon known to be taking place and Aidid was probably aware of it.

On 4 June, a warning was issued to SNA that UNOSOM II would inspect 5 weapons storage sites the next day. This was delivered in the form of letters by US Colonels Byrne, and Kevin McGovern, the deputy officer in charge of U2 Intelligence. According to the SNA, [56] letters were brought addressed to someone not concerned with security issues for an unexplained reason. They arrived at 4 p.m. on a Friday, the Islamic day of rest, and would not be read until Monday; the inspections were to take place on Saturday. The US colonels told Abdi Hassan Awali to just take them. The SNA said there was no date or place mentioned; the colonels indicated only that there would be an inspection of some areas where arms were cantoned and that the SNA would be warned of this closer to the time. According to General Montgomery, the letters were personally received by Abdi Qaibdid, the SNA number three, although he tried to refuse acceptance of the letters.

Civilian officials in the Office of the SRSG had not been informed of this warning, nor had the Political Division. They found out that something was occurring when they heard gunfire all around the UN Headquarters at 10:30 Saturday morning, 5 June, when there were 3 attacks. They were told this was a reaction to inspections. One site being inspected was the radio station, authorized weapons storage site (AWSS) #3. It had been registered as a weapons site by UNITAF, although it was empty but for a few rounds of ammunition. It was destroyed on 6 June. AWSS #5 was also a weapons site inspected on 5 June and more weapons were found than had been there in February, the date of the previous inspection.

These inspections began at an extremely tense time and the perception of Somalis was that they were intended to be disarmament and not merely inspections. Aidid may have feared UN control of those heavy weapons he had not hidden or that if UNOSOM II discovered the discrepancies at AWSS #3, they might seize it and the radio station, and therefore, says Farer, “decided on the desperate gamble of an attack”. In fact, the operation had been intended by UNOSOM II to seize the radio station and as such the question arises as to the why the UN was not prepared to deal with such a potential response from Aidid. Was it a miscalculation? Whatever the SNA claims about the letters, they were well prepared for the arrival of the Pakistani inspectors.

The inspection began at 07:00 on 5 June. The SNA claimed that this was led by the US, who then slipped out and left the Pakistani troops at AWSS #3. According to the SNA, this ultimately ended in 75 Somali deaths, in addition to the 24 Pakistani casualties. According to UN witnesses that survived, women and children convened to stop the inspections with knives and machetes, and were incited to violence by SNA sharpshooters firing into the crowd. According to General Montgomery, “these people were badly taken apart…They were overwhelmed by women. It was well planned. Everyone know who did it.” [57] It took 3 hours before any assistance reached the Pakistanis.

An SNA representative tried to contact UNOSOM II, but with little success. He managed to contact a political officer who went to see Aidid at 13:30 on 5 June. Aidid conveyed the message to him that he wanted things calmed down, but on two conditions. First, he wanted regular SNA-UNOSOM II consultations, and second, a cessation of hostilities. Both were unacceptable to UNOSOM II. A written response was drafted for the SRSG. Awali came and read the response and said he would not take it back to Aidid. Instead, he wanted to arrange a meeting between Aidid and the Office of the SRSG. This never took place.

On 6 June, the Security Council passed Resolution 837, a kind of arrest warrant for Aidid. While the Resolution referred to the SNA, it did not expressly mention Aidid; his name had been removed in later drafts of the text. The Resolution reaffirmed “that the Secretary-General is authorized under resolution 814 (1993) to take all necessary measures against all those responsible for the armed attacks…, including against those responsible for publicly inciting such attacks, to establish the effective authority of UNOSOM II throughout Somalia, including to secure the investigation of their actions and their arrest and detention for prosecution, trial and punishment”. It also urged “Member States to contribute, on an emergency basis, military support and transportation, including armoured personnel carriers, tanks and attack helicopters, to provide UNOSOM II the capability appropriately to confront and deter armed attacks directed against it in the accomplishment of its mandate”.

UNOSOM II civilian officials had provided background information for the resolution, but they understood that the US had drafted it, probably in the United States. The Office of the SRSG was outside this process. However, it was the understanding of the military leadership that the SRSG had input. Both Generals Bir and Montgomery claimed they did not – although at least Montgomery felt he would have done the same. The decision was taken at a completely different level. It was the feeling of UN civilian officials that if there was a decision to go to war, then as much mileage as possible should have been extracted before the use of force. There was a feeling that Aidid could still be pressured to disarm.

This was one of two views at this critical juncture that represented the two extremes underlying the logic of the operation. At the one extreme was the diplomatic view of negotiations. At the other extreme was the feeling that it was difficult to turn the military backs on the ambush and negotiate. There had to be a response. But there seemed not to be a middle ground, a limited response possibility. The black and white options of doing nothing or turning on a tap of uncontrolled force were starkly apparent. As limited as US troops felt they were, in the absence of a limited force concept, there was only a sledge-hammer available – and the US was used to using more than just one.

Given the political need in New York to respond to the Pakistani tragedy, the Security Council authorization was in place; now all that was needed was an operational green flag to signal the charge. This came on 8 June in a meeting in Mogadishu to discuss military action. On the civilian side, Howe, April Gillespie, then acting as an advisor to the UN on the issue and also a Gulf War transplant, and another individual were present. Of the military, the Force Commander and representatives of the US military were present. None of the rest of the SRSG’s civilian staff nor the Political Division were invited. The SRSG resisted the option for military action at that time, but the rest of the Americans pressed for action. Howe’s civilian staff, when they later heard of his stance, remained loyal to him because of this. However, the meeting continued for 9-10 hours. Gradually the SRSG was broken down and forced to change his position, until he ultimately accepted the military option, especially since the Secretary-General also wanted this.

On 10 June, the military began large-scale preparations. They relocated civilians in new headquarters or flew them massively to Nairobi. Only 40-50 civilians were left and most of these were technical personnel. The Political Division was entirely marginalized. UNOSOM II attacked on 12 June and held the initiative for about 3 weeks before losing it to Aidid. Between 12-14 June, a number of SNA storage sites were destroyed by AC 130s and helicopter gunships. On 17 June, there was an air and ground attack on an SNA enclave on Afgoye Road. Pakistanis were ambushed on 30 June and the Italians on 2 July. There was a major US operation against the Abdi House on 12 July. On 8 August, in the port area a command-detonated mine was used against US military police. During this period, the military conducted all actions and made all decisions, and the military were dominated by the US: they had the people, the plans and they made the policy.

The UN had called for special forces form the beginning of the action, but Resolution 837 was a major strategic shift in 24 hours. Nothing came until the Rangers were deployed on 31 August. One reason for the delay was that the Secretary-General wanted special forces from other countries, such as the UK and France, and not just the US. When the Rangers arrived, another dimension was added to the problems of command and control since they were under the command of another general, although Montgomery understood he would have operational control.

With the arrival of the Rangers there was a shift in focus from destroying SNA assets to capturing Aidid. The Rangers had been launched to get Aidid. In September, the conflict intensified: the UN compound was hit with mortars and the walls were hot from RPGs; there were ambushes and the roads were dangerous. Early in September the Nigerian Battalion was redeploying from Belet Uen to replace the Italians in Mogadishu after their falling out with the US. The Italians completed their redeployment by 15 September, but in the meantime, on 5 September, at Strong Point 49 the Nigerians were ambushed and lost 9 killed and 1 captured. There was another incident on 9 September on 21 October Road.

In early September, the Belgians and the French announced their withdrawal from Somalia. This destroyed finally the original UNOSOM II mission in which operations in the south would expand into central Somalia. Expansion had required the Indians, who only began to arrive in September and were not in Somalia until October, due to their disapproval of the hunt for Aidid.

The combat phases reached a climax. On 21 September, the senior SNA officer Atto was captured, but the same day Pakistani APCs were ambushed on the Afgoye Road. On 25 September, the first US helicopter was downed. On 3 October, the Rangers raided the Olympic Hotel but did not capture Aidid. They had finished the raid, but in trying to save their troops, more helicopters went down. The Pakistanis and the Indians helped the QRF get to some of the troops left alive. Perhaps done so for intelligence purposes, UNOSOM II was not informed of the Rangers’ activities and therefore arrived late. The Rangers were in a trap, were too few to extricate themselves and it was difficult to help them. This has resulted in a perception in the US military now that US troops should only ever rely on Americans for help.

Just as the CNN factor had in part led to US involvement in Somalia, so too it ended it instantly when the images were broadcast on Sunday night of an American soldier’s body being dragged through the streets of Mogadishu by Somalis playing. The carcass of a political failure in the field was directly linked to political sustainability on the home front. Without any meaningful national debate, the single question posed, most forcefully by Senators Byrd and Dole on Monday morning, “Why are we there?” was enough to effectively end one of the most challenging UN operations in history. These 4 words, which seemed then to ring much more loudly than the demand one year earlier in the face of famine that “we must do something!” led to abandonment of Somalia by the international community. On 7 October, the US President, Bill Clinton, addressed the nation and announced that US troops would withdraw by 31 March 1994. He concluded that “The solution to Somalia’s problems is not a military one, it is political” – and this implied that “political” meant exclusively “national,” but the problem instead was that “international” needed to be political.

VIII. Conclusions

  1. The departure of UN forces from Somalia resembled more a retreat than a withdrawal. The reasons for this are being lost or forgotten. As much printed material as there is available on Somalia, in UN reports, press coverage, and the cottage industry of secondary source publications in academic and practitioners’ journals, only a limited amount has reflected the internal strengths and pitfalls of US and UN operations in the field. The net result is that the wrong lessons have been learnt from the watershed experience of the second generation of UN operations over the last six years. United Nations officials were left feeling that the use of military force should be avoided in peace operations, since the degree of destruction in Somalia was not matched by achievement of many objectives. The United States concluded that failure was due to the fact that not enough force was used. The Secretary-General returned to the black and white options of either defensive peacekeeping or high-intensity enforcement. These had initially led to the need for a limited use of force concept that failed to be adequately tested in Somalia.
  2. The phases of international involvement in Somalia roughly correspond to the three eras of UN operations generally. UNOSOM I in 1992 was based on the traditional peacekeeping concept. This was the diplomatic use of troops. The military force employed during UNITAF in late 1992 and early 1993 reflected the attempt to enhance the military competence of the UN in its second generation. UNOSOM II after spring 1993, albeit shifting between these two models, needed a strong political capability to succeed. The first two are blunt instruments and have their place. By themselves, neither could have responded to the conditions in Somalia. Aspects of both needed to be employed as part of a comprehensive strategy to respond to the social and political demands of Somali anarchy.
  3. The mandate of UNOSOM II was not “enforcement” in the manner of reversing the acts of an international state aggressor. Instead, it was perceived to be “peace-enforcement,” which was something less, but it was not clear what. The term “peace-enforcement” had a very specific origin. It appeared in June 1992 for the first time in paragraph 44 of Boutros-Ghali’s An Agenda for Peace. This was a proposal more for a mechanism than a concept, namely “peace-enforcement units”. It recognized that the black and white options of non-military sanctions and the massive use of force were inadequate. The UN, it said, had been called on to restore cease-fires that had broken down and this required something more than what the secretariat had at its disposal. Its ill-definition in An Agenda for Peace has led to its confusion with large-scale, conventionally conceived enforcement.
  4. Anarchy was a principal characteristic of Somali social organization before the colonial powers’ scramble for Africa reached the Horn. That is, if anarchy is understood to mean the absence of a central government. But there was a fundamental difference between the chaotic disorder and uncertainty that followed the collapse of the Somali Republic in January 1991 and the lack of centralized authority in pre-colonial times. The successors to the Somali state were not clans and sub-clans, which were reverted to by the population as social institutions, but political factions. These armed groups, which commanded some resources and managed to control limited territorial areas, may have been divided along clan or sub-clan lines initially, but their perpetual fragmentation led to alliances of convenience that crossed clan lines. Established and led by what has been referred to in the Somali context as “warlords” and their cronies who galvanized support through clan allegiance, factions became the repositories of power by the force of arms. The evolution of factional violence did not proceed according to clan loyalty, but to the vicissitudes of uncontrolled warlords. This was only a recent phenomenon in Somali history.
  5. Despite linguistic and cultural homogeneity, there was no pre-colonial collective consciousness amongst Somalis. They did not identify with each other regardless of origin, but differentiated themselves according to their patrilineal genealogical origins. At the time, there prevailed a kind of pan-Somali code of conduct called the Xeer. This was a self-regulating system in which one was prevented from dominating others. As part of the social contract, the source of individual authority was not the capacity to control others, but the capability to ensure others were not dominated. Successful production and not control of centralized surplus justified a role in authority. Under colonial administration, a new set of relations developed. The effort of Somalis to replace an alien master of centralized government and surplus resources displaced the indigenous social fabric. When the state finally collapsed in 1991, there was no check or balance on power to restrict the behaviour of factions.
  6. Somalia in 1992 resembled an apocalyptic image in which the four horsemen emerged in something of a vicious cycle. As a result of anarchy, war and famine and death plagued the country. Anarchical conditions perpetuated civil war and caused in turn increasing fragmentation, which in its turn caused more competition and conflict. The destruction of the means of food production and the willful use of food as a weapon led to famine. The conditions of internecine violence that led to the famine also limited the access of international humanitarian assistance. This intensified the famine and exacerbated in turn the security conditions that led to it as fighting over power and fighting over food for survival became increasingly connected. Even the limited control over their members that factional warlords exerted at any given time could not prevent self-interested looting by individuals. By mid-1992, the entire population was in a life-threatening position. Somalia was described as “the single worst humanitarian crisis in the world”.
  7. The international community responded to this apocalypse slowly and in a fragmented manner. The late response alienated Somalis and turned them against the UN even before its arrival. They felt the international community had abandoned them. This was a feeling that later turned into outright opposition to what was perceived as a partisan effort. The UN never managed to transcend its alliance with Mahdi and its alienation of Aidid. The divergence only grew wider and wider and the UN effectively served to drive deeper a wedge between the two. Peacekeeping between factions in internal conflicts is to further fragment a collapsed state.
  8. Sahnoun tried to foster coordination between the different political kingdoms of the UN and a local political process. He attempted to address the source of the problem, the anarchy. Sahnoun pursued a strategy of decentralization, or “regionalization,” to displace the warlords. To be effective in such anarchical conditions, the UN had to provide a centre of gravity around which alternate leadership could grow and strengthen. However, in the absence of local physical infrastructure, any UN assistance to mitigate this or a UN political infrastructure, the process was too fragile and easily disrupted by the warlords. Also, Sahnoun lost the support of the Secretary-General.
  9. Throughout this flawed and fragile political process by far the greater focus at UN headquarters in New York was on the symptoms of insecurity and famine. The overall flaw of UNOSOM I was to attempt to apply the wrong instrument, the peacekeeping model, at the wrong time. In a diplomatic fashion, this treated internal factions as state territorial entities. The bureaucratic habit of the UN, seeking to function with familiar institutions, in the absence of state officials, gravitated towards the factions and warlords. The warlords were at least identifiable to some degree, a critical requirement in a diplomatic culture. But UN officials behaved with the factions as they would with sovereign governments. Therefore a primary objective when the UN became apprised of the Somali situation was to achieve a cease fire. But the limitation of this could have been anticipated: when factional fighting subsided, it was replaced by general lawlessness, banditry and looting.
  10. By November 1992, the two principal problems in Somalia were perceived at the UN to be famine and weapons. Only a trickle of assistance was getting through as security conditions worsened. On 3 December 1992, the Security Council explicitly authorized the Secretary-General, the United States and other troop-contributing countries to “use all necessary means to establish a secure environment for humanitarian relief operations in Somalia”. The idea was to establish food distribution sites at each population centre and guarantee the delivery of large quantities of food in order to eliminate looting and hoarding, the use of food as a weapon, and therefore weaken the power of the warlords. A principal point of contention arose between the US and the UN Secretary-General regarding the scope of disarmament. What constituted a “secure environment for humanitarian relief?” Did it mean, restrictively, to protect corridors for specific deliveries of assistance to end the famine? Or more broadly, did it mean creating generally secure conditions in which the UN could freely operate? In the latter case, the US would have had fundamentally altered the environment of anarchy in Somalia, which would have entailed not only breaking the vicious cycle of violence and famine, but also addressing the root cause of anarchy through a process of national reconciliation and political reconstitution. The US had been prepared to respond to the symptom of famine, the most dramatically tragic leg of the vicious cycle. It had not intended and in the field was not prepared to pacify Somalia.
  11. Humanitarian conditions improved dramatically. When time came for the UN to assume command and control in Somalia, it could not maintain the centre of gravity that the US established under UNITAF. The US responded to fill the vacuum, keeping it engaged in the field to a greater extent than it expected or wanted. While this may have been the only option under the circumstances, not only did it mean an overwhelmingly dominant position for the US in the UN force, thus rendering a fully integrated UN operation on paper more of a joint operation with US forces operating separately. It also meant that since the centre of gravity had developed as a result of a military operation, UNOSOM II would be dominated by military imperatives. The dominance of military leadership and the uncontrollable, non-sustainable consequences of this, illustrated the limitations of military force. A peace process cannot be conceived as part of a military framework. The military is most effective when it has clear objectives devised by a political authority and is employed in a limited manner. When this task or set of objectives are completed, the military require on-going political direction; if it is left without this and becomes the principal decision-making authority in the field, its decisions are short term and limited in scope.
  12. UNOSOM II was authorized under Chapter VII of the UN Charter, but the premise of its presence in the country was conceived as “assistance”. “Assistance” implied local consent; enforcement powers meant the opposite. The conditions in Somalia were anarchy and something much more than “assistance” was needed; so much so that UNOSOM II could not resist doing more. But it did so intermittently because it bounced back and forth between its mandate and ground needs, almost unconsciously. Ultimately, it did neither adequately. It had to conduct some political and law and order tasks independently of Somalis. When UNOSOM II in its later phases finally began to help build Somali institutions they had to do so in more of a “control” manner than “assistance”, precisely because of the weakness of these institutions. That they had been fostered by the UN in the first place without any existing basis meant that the UN had effectively the power in Somalia of a governorship-in-trust. Yet the refusal to acknowledge this and design a comparable mission framework meant that the mix of all these premises fragmented UNOSOM II’s directional capability.
  13. The two approaches relied on by Boutros-Ghali to tackle the Somali apocalyptic cycle were military disarmament and political reconciliation. The mandate would empower UNOSOM II to provide assistance to the Somali people in rebuilding their economy and social and political life, re-establishing the country’s institutional structure, achieving national political reconciliation, recreating a Somali state based on democratic governance and rehabilitating the country’s economy and infrastructure. These were intensely social and political tasks addressing the heart of the cycle, but they were still being approached with military and diplomatic instruments.
  14. If these social and political tasks were to be accomplished, there had to be political transformation, not merely reconciliation between factions. This required a decisive, powerful deployment, quickly and comprehensively. Then military forces needed to be prepared to behave as policemen, as well as soldiers, if the centre of gravity created by a decisive deployment was to be consolidated. If there is to be genuinely a transition in a country from one set of conditions, such as anarchy, conflict and famine, to another set of conditions, resulting in order, peace and health, a UN mission must secure the reigns of power first. Determining the basis, source and structural logic of a new authority is an anthropological project and operations of the type and magnitude of UNOSOM II require strong components with anthropological, sociological, psychological and area expertise. Used to the luxury and habit of operating at arms length, affordable on a battlefield, troops in UNITAF and UNOSOM II were insufficiently briefed on local social and political phenomena. This fed a behaviour and perception of opposition and exclusion, rather than integration between international forces and local Somalis. This set a psychological stage for conflict. Instead of this complex social and political approach to Somalia, the core of UNOSOM II’s operational concept was a series of military tasks. When political reconstruction did begin, it was too little too late, and ill-defined, resulting in political failure and military retreat.
  15. The battlefield was being delimited: Aidid had made something of a miscalculation at the March 1993 Addis Ababa conference. He had conceded to the Agreement because he felt not only could he compete for control of the top-down political process as part of the overall factional conflict, but that he could out-manoeuvre the UN in the countryside by controlling the bottom-up process as well. This meant there were several fronts emerging on which Aidid and the UN would face each other irreconcilably, all of which would merge into a violent confrontation. There was insufficient detail regarding disarmament, and the opportunity for misunderstanding and disagreement; there would be disagreement about the primacy of a nation-wide authority and the process of regionalization; and competition for the latter.
  16. There was supposed to be a seamless interface in the transition from UNITAF to UNOSOM II. It was to be a process of painting UNITAF blue. The principle was a piece-by-piece transfer of authority to UNOSOM. When UNOSOM was in place after a general transfer, the US would pass the hat of command to the UN. There was the belief that the population should not even notice the difference on the ground. This was a critical mistake: the transition was from one mandate to another, from one mission to another, and from one flag to another. The shift should have been dramatically marked for the population and NGOs. There was a much wider mandate with many more tasks. The political nature of the mission and the force was altering. Much more would have to be accomplished with far fewer resources, and the effects of this caused frustration locally and amongst NGOs. This gave the impression of a failing mission rather than the slow start of a new operation.
  17. If UNOSOM II was to succeed, the burden of the operation should have rested on the substantive division of the civilian staff. The civilian body virtually did not exist and what did reach the field was stuck at headquarters in Mogadishu. Without a staff the aims of “peace-building” and rehabilitation and a final political solution including the marginalization of the factions were quixotic. The civilian staff seemed not a priority at all. One reason for this was the vague political objectives of the mission. Without a clear identification of tasks that defined the political goals, little foresight anticipated the civilian staffing requirements. No one had thought seriously how to conduct a political operation, nor wanted to admit what it really meant. Civilian staffing was quite arbitrary and infectiously slow. This in turn retarded the development of a non-military operational concept.
  18. The degree of independence operationally permissible for contingents in UNITAF, and the removal of the binding influence of the US, meant the contingents could not be tamed by the new UNOSOM II Force Commander. The gap was never bridged. The attempt to transfer a coalition to a UN command structure without devising a means to further integrate those contingents in the manner necessary for a collective command structure, resulted in the existing coalition fragmenting further. The role that the US had designed for itself in UNOSOM II was insufficient to prevent this break-up. The US retained both effective and juridical independence from the UN command structure, unlike all other contingents which were juridically subordinate as part of the collective. The US could at the same time remain part of the operation by designating dual US-UN command roles for two of its officers: the Commander of US Forces in Somalia was designated also as the Deputy Force Commander of UNOSOM II; and the Deputy Commander of US Forces in Somalia was designated also as the Commander of UNOSOM II Logistics Forces Command as well.
  19. Separate operating procedures permitted by the persisting coalition formula defied the official UNOSOM II rules of engagement . Chapter VI contingents devised their own rules, maintained traditional peacekeeping principles or simply refused to accept the UNOSOM II rules. In addition to this, there ought to have been the problem of interpreting the meaning of the UN rules. The US drafted them and on the surface they were reasonably restrictive. But the US interpreted the rules according to their heavy-handed use of force sub-culture, which quickly escalated into heavy force. US military criticism of the UNOSOM II experience emerged later, claiming that more force used earlier would have yielded greater success, rather than recognizing the need for a limited force concept, for built-up areas such as Mogadishu, that could be relied on for guidance in the interpretation of the rules. For Chapter VII contingents, there was a tendency not to identify some general limited force concept but to determine how best to cooperate or avoid cooperation with the US interpretation. As the combat phase proceeded, the difference between various sets of rules widened. US military interpretations demanded a greater use of force and Chapter VI contingents entrenched their positions against participating.
  20. UNOSOM II needed to reconstitute a centre, but focus on undermining the factions served instead to fragment the country more and prevent the process of ultimate centralization. The factions remained in control of this, and by their own conflict also prevented centralization of authority. Even if UNOSOM II military and civilians cooperated in a bottom-up approach, to succeed there would have to have been a well-orchestrated political strategy of subversion of the factions with a clear goal of establishing a new authority, and not merely the logic of elections to save the UN face and exit. Such a strategy did not exist in which the military could have been employed judiciously towards a sustainable end. Concepts or the interpretation of the UNOSOM II mandate had never been discussed between political officers and the military. Therefore, the military decision-making that dominated the mission simply disregarded the notion of operational political input from the field, and at most looked to political masters in Washington or New York. This ultimately led to a combat phase in the first place, and its lack of success subsequently.

Notes

1. Financial Times, 28 February 1995.

2. F.H. Hinsley, Power and the Pursuit of Peace (Cambridge: Cambridge University Press, 1963).

3. Cf. Ramsey Clark, The Fire This Time: U.S. War Crimes In the Gulf (New York: Thunder’s Mouth Press, 1992).

4. “Supplement to an Agenda for Peace: Position Paper of the Secretary-General on the Occasion of the Fiftieth Anniversary of the United Nations,” UN Doc. A/50/60 and S/1995/1 of 3 January 1995, para. 1.

5. Ibid., para. 36.

6. Ibid., para. 8-22 and 33-36.

7. Ibid., para. 35.

8. Ibid., para. 77-80.

9. I.M. Lewis, “Misunderstanding the Somali crisis,” Anthropology Today, Vol. 9, No.4, August 1993, p. 1. Cf. his classic works on Somali social organization: A Pastoral Democracy (New York: Holmes and Meier, 1982) and A Modern History of Somalia (Boulder, CO: Westview Press, 1988).

10. Abdi Ismail Samatar, “Destruction of State and Society in Somalia: Beyond the Tribal Convention,” The Journal of Modern African Studies, Vol. 30, No. 4 (1992), p. 625.

11. Captain Sir Richard F. Burton, First Footsteps in East Africa (New York: Dover Publications, Inc., 1987; first edition: London, 1856), Vol. I, p. 88.

12. Cited in Andrew S. Natsios, “Food Through Force: Humanitarian Intervention and U.S. Policy,” The Washington Quarterly, Vol. 17, No. 1, Winter 1994, p. 136.

13. This and the following quotes, Samatar, p. 630 et seq.

14. Ibid., pp. 632-633.

15. Samatar, p. 633; and see further pp. 634-638.

16. Ibid., p. 635.

17. For a literary but particularly illustrative account of this process, see Nuruddin Farah’s trilogy, Variations on the Theme of an African Dictatorship.

18. See further, “Somalia beyond the Warlords: The Need for a Verdict on Human Rights Abuses,” Africa Watch, Vol. V, No. 2, 7 March 1993, pp. 5-8; and “Somalia: Update on a disaster – proposals for human rights,” Amnesty International Doc. AFR 52/01/93, 30 April 1993, pp. 5-8.

19. Keesing’s Record of World Events, July 1992, p. 38992.

20. Personal interview with General Mohammed Farah Aidid, Serena Hotel, Nairobi,

20 February 1994.

21. Assessments of factional strengths varied greatly: Aidid claimed to have more than 30,000 “troops” while UNOSOM II intelligence believed his faction numbered up to 2000 and Mahdi’s forces 1000-1500.

22. Personal interview with US Admiral Jonathan T. Howe, UNOSOM II Headquarters, Mogadishu, 10 February 1994.

23. Sahnoun’s quotes in this section are from: Mohamed Sahnoun, Somalia: The Missed Opportunities (Washington, DC: United States Institute of Peace, 1994), pp. 25-27 and 37-40.

24. Personal interview with Ali Mahdi Mohamed, north Mogadishu, 10 February 1994.

25. Cf. the assessment of the wet-food programme in Netherlands’ Development Cooperation, Operations Review Unit, Humanitarian Aid to Somalia (The Hague: Ministry of Foreign Affairs, 1994), p. 142.

26. UN Doc. S/23829 of 21 April 1992, para. 27.

27. Interview with Aidid, 20 February 1994.

28. Helen Chapin Metz, Ed., Somalia: A Country Study, 4th ed. (Washington: Federal Research Division, Library of Congress, 1993), p. xxxiv.

29. Interview with Aidid, 20 February 1994.

30. Personal meeting with Andrew S. Natsios, Providence, RI, November 1992.

31. Cf. Natsios, “Food Through Force,” p. 144.

32. Cf. for instance on “the American way to fight,” John F. Hillen III, “Peacekeeping is Hell: America Unlearns the Lessons of Vietnam,” Policy Review, Fall 1993, pp. 36-39.

33. In addition to the United States, UNITAF eventually included military units from: Australia, Belgium, Botswana, Canada, Egypt, France, Germany, Greece, India, Italy, Kuwait, Morocco, New Zealand, Nigeria, Norway, Pakistan, Saudi Arabia, Sweden, Tunisia, Turkey, United Arab Emirates, United Kingdom and Zimbabwe.

34. UN Doc. S/24992 of 19 December 1992, para. 29.

35. New York Times, 16 February 1993.

36. Cf. the incidents referred to by Africa Watch, 7 March 1993, pp. 17-18.

37. See further UN Doc. S/25168 of 26 January 1993.

38. UN Doc. S/25354 of 3 March 1993.

39. Brig. Gen. Mohammed Farah Aidid, “Democracy in Somalia – its Roots and its Future Scenario,” in Mohammed Farah Aidid and Satya Pal Ruhela, Eds., The Preferred Future Development in Somalia (New Delhi: Vikas Publishing House Pvt Ltd., 1993), pp. 12-25. The following quotes are from pp. 24 and 20, respectively.

40. Burton, pp. 77-78.

41. S/25354, para. 56-69.

42. The list of troop-contributors for UNOSOM II recommended by the Secretary-General and approved by the Security Council included: Argentina, Australia, Bangladesh, Belgium, Botswana, Egypt, France, Germany, Greece, Hungary, India, Indonesia, Ireland, Italy, Jordan, Malaysia, Morocco, Namibia, New Zealand, Nigeria, Norway, Pakistan, Republic of Korea, Romania, Saudi Arabia, Sweden, Tunisia, Turkey, Uganda, United Arab Emirates, United States, Zambia and Zimbabwe.

43. UN Doc. S/25354, para. 92.

44. See further, UN Doc. S/25168 of 26 January 1993, Annex III.

45. Cf. UN Doc. S/26317 of 17 August 1993, para. 23-42.

46. Interview with Aidid, 20 February 1994.

47. Newsweek, 21 January 1993, as cited in John Tessitore and Susan Woolfson, Eds., A Global Agenda: Issues Before the 48th General Assembly of the United Nations (New York: University Press of America, Inc., 1993), p. 85.

48. UNOSOM II Office of the Spokesman, UNOSOM II Weekly Review, Issue #2, 30 September 1993, p. 5.

49. The Hon. Enoch Dumbutshena, “Report on Eight (8) Somali Detainees,” January 1994. The report recommended the release of SNA members or those suspected of supporting SNA activities held in detention for several months: Osman Hassan Ali (Atto); Omar Salad; Mohamed Hassan Awale; Abdikarim Duale Gelle; Mohamud Salad Farah; Ali Abdualla Abdi; Ali Hussein Mohamoud Isse; and Mohamed Mohamoud Afrah Awale.

50. Amnesty International, “Peace-keeping and human rights,” January 1994, AI Index: IOR 40/01/94, pp. 29 et seq.

51. Appendix 2, Annex D, p. 1.

52. On distinguishing military and civilian conceptions of command and control, see Thomas P. Coakley, Command and Control for War and Peace (Washington, DC: National Defense University Press, 1991), pp. 34-38.

53. See specifically, US Army FM 100-23, “Peace Operations”, Version 6, January 1994.

54. Appendix 6 to Annex C, “Operations”, UNOSOM II OPLAN 1, pp. C-6-1, C-6-2.

55. Compare the following with Tom Farer’s testimony before the Committee on Armed Services of the US House of Representatives on 14 October 1993: “United States Military Participation in United Nations Operations in Somalia: Roots of the Conflict with General Mohamed Farah Aideed and a Basis for Accommodation and Renewed Progress.” Following quotes in this section are from pp. 13-23 of a printed manuscript. Much of the account in this section is drawn from interviews with SNA officials and individuals from UNOSOM II’s Office of the SRSG.

56. Interview with Mohamed Hassan Awali, Serena Hotel, Nairobi, 19 February 1994.

57. Interview with General Montgomery.

Tagged with:
 

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>