e-Prints

The views in this paper are entirely those of the author expressed under Air University principles of academic freedom and do not reflect the views of the School of Advanced Airpower Studies, Air University, the U.S. Air Force, or the Department of Defense. In accordance with Air Force Regulation 110-8, it is not copyrighted, but is the property of the United States Government.

Major Harold E. "Hal" Bullock is a communications/computer systems officer with a broad background including space operations, strategic intelligence, and tactical communications. He has also served under both Navy and Marine commands. He has served with Det 2, Space Division (L. B. Johnson Space Center, Houston, TX), the Joint National Intelligence Development Staff (NAVINTCOM, Washington, DC), and the 2d Combat Communications Group, Patrick AFB, FL. He holds Bachelor's and Master's degrees in Computer Science and is a graduate of Squadron Officer School, the Marine Corps Command and Control Systems Course, and Air Command and Staff College.

So many people helped me with the research and production of this thesis it is difficult to know where to begin. Understanding the risk of leaving out someone, I want to acknowledge those who contributed the most.

First, I would like to thank my wife, Pam and children for putting up with my odd hours, domination of the kitchen table, and endless piles of "primary evidence" scattered around the house. Pam even read my thesis without complaint, which qualifies her for sainthood. Without the support of my family, I would never have completed this. I also want to thank my thesis advisor, Dr. Jim Corum (Maj, USAR), whose constant caution "not to worry about it" went largely unheeded. His guidance was invaluable in producing and editing the final product.

Many others helped me in the basic research process. Thanks to Lt Col Pete Faber, USAF, Mr. Al Lopes, and Comandante Fernando Ordóñez-Valázquez, Peruvian Air Force, for their translation skills. I am also grateful for the endless hours Dr. Bob Wright of the Center for Military History spent answering questions during numerous telephone interviews. His insight and observations permeate the work. Likewise, the contributions of Capt David "Scotty" Dawson, USMC, were essential for a complete picture of the operations in Somalia. Lastly, the interviews granted by MGen Zinni, USMC; Col Anderson, USMC; Mr. Walter Clarke, State Department; and Lt Cols Martinez, Borchini, and Eaton, USA were all extremely helpful in chasing down the many issues surrounding our adventures in Somalia during UNITAF and UNISOM II.

Col Steven Butler, USA (Ret.) provided valuable first-hand knowledge on the Dominican Republic crisis. His observation and scholarship made my research go much more smoothly than I expected. I am also indebted to Col Guilson-Lopez, Brazilian Army, for providing me a rare copy of the Brazilian after-action report on the Dominican Republic. This report provided hard-to-get information on the Latin American point of view.

For general information on UN peace operations, I am indebted to the UN military advisors who took several hours from their busy schedules to answer my questions during my visit. I also greatly appreciate the contributions of the many US and international officers who responded to my peace enforcement questionnaire. There was no substitute for their first-hand knowledge.

The United States has been involved in peace enforcement operations for many years. In that time we have learned some lessons. Unfortunately, we continue to repeat many of the same mistakes. Sometimes we have forgotten hard-learned lessons, and sometimes we never learned from our earlier experiences. The Dominican Republic deployment of 1965-66 and recent experiences under the Unified Task Force, Somalia (UNITAF) and United Nations Operations, Somalia II (UNOSOM II) are representative peace enforcement operations. This paper will examine which lessons we learned from these operations, which lessons we learned and lost, and which lessons we seemingly ignored.

Focusing on command and control, the issues can be loosely grouped into categories of force and command structure, political impacts, and interoperability. In force and command structure, the US has still not come to grips with the difficulties of operating in a multinational coalition under international (e.g. UN) control. The problems of dual lines of control and Byzantine command structures plagued both the Dominican and Somalia operations. The ability to integrate humanitarian relief and nation-building forces effectively into the overall structure seems to have deteriorated rather than improved. Stand-by, earmarked forces, combined exercises (including non-military agencies), and stronger civil-military integration cells could help mitigate difficulties, but need to be pursued more vigorously.

Within the area of political impacts on peace enforcement operations, the subject of prisoners is still a difficult area with little progress in the last 25 years. Nations have also continued to place political constraints on their forces limiting them both geographically and operationally. In both cases, these problems can not be eliminated, but must be compensated for. Rules of engagement also affect operations, and some progress has been made in this area. ROE for Somalia were aggressively designed from the bottom up and were well-received. Work remains in the area of non-lethal force, but the US has made definite improvements in the process.

Finally, for a multinational peace enforcement operation to succeed, interoperability must be achieved. Equipment and logistics provide the most visible area of concern; and problems remain with equipment, fuels, and transportation, and related skills. Difficulties among the geographically and culturally diverse forces in the Dominican Republic were repeated in Somalia. A related area, training and doctrine interoperability, seems less problematic. Though problems remain, the biggest challenge is underestimating the quality of the forces at first and overestimating their qualities later in protracted operations. Cultural and language interoperability overarches the other concerns and provides our biggest continuing challenge. Americans tend to be ethnocentric and mono-lingual, and our military system has done little to counter the associated problems. Education and training must be emphasized, and once multilingual and cultural skills are attained, these attributes must be tracked and available to both planners and commanders. Multilateral interoperability remains a challenge.

Overall, to improve command and control for multinational peace operations, we must look at past experiences, recognize the recurring problems, and aggressively attack the roots of those problems. In this paper, I have identified some of the "stickiest wickets" from the Dominican Republic and Somalia and suggested some solutions. The solutions are not simple or easy, but if we ignore the problems, they will certainly not simply "go away."

Disclaimer ii

Abstract iv

About the Author vi

Acknowledgments vii

Chapter 1: Introduction 1

Chapter 2: Command and Force Structure 4

Chapter 3: Political Impacts on Operations 35

Chapter 4: Interoperability Issues 69

Chapter 5: Conclusion 100

BIBLIOGRAPHY 108

Figure 1 Cartoon Satirizing UN Command and Control.................................................ix

Figure 2 Map Showing Humanitarian Relief Sectors.....................................................17

Figure 3 Weapon Authorization ID Card......................................................................27

Figure 4 ROE Pocket Card Issued for Operation Restore Hope.....................................48

Although peacekeeping, and more recently, peace enforcement have attracted much attention in both the military and the media, the United States is no stranger to either. We have participated in peacekeeping and peace enforcement missions at least as long ago as the Dominican Republic intervention of 1965-1966. If recent missions in Somalia and the former Yugoslavia are any indication, the U.S. will likely continue as a key participant for the foreseeable future. As a recent Army briefing phrased it, the "U.S. will participate in both peacekeeping and peace enforcement operations in the coming decade--like it or not!" By mid-1993 the United Nations had 80,000 personnel involved in peace operations across some 17 missions, and the 1993 UN peacekeeping budget totaled nearly $3 billion. Recent operations have shown that we learned some lessons from the past and improved in certain areas, while in other areas we either forgot or ignored what we learned in previous operations. I will examine issues which affect our ability to conduct peace enforcement missions successfully. While the information presented may have wider application to peacekeeping or general coalition warfare, my intent is to focus strictly on multinational peace enforcement operations.

Peace enforcement is fundamentally different from peacekeeping. Peacekeeping operations are conducted with consent of all parties and are designed to supervise, not enforce, a truce already in place. These operations are described and authorized in Chapter 6 of the United Nations Charter. Peace enforcement is much more demanding; it seeks to impose a peace with or without the consent of the belligerents. These operations are authorized in Chapter 7 of the United Nations charter. Peace enforcement usually includes humanitarian and nation-building activities performed in a non-permissive environment. I address peace enforcement rather than peacekeeping because of its greater military demands and because of the commensurably greater political sensitivities and constraints.

Specifically, this paper will focus on what I categorize broadly as command and control (C2). Under this umbrella I will examine issues relating to command structure and chain of command, political impacts on operational performance, and key interoperability areas. Although I deal specifically with interoperability issues in Chapter 4, the idea of interoperability is pervasive. Whether in command structure, civil-military relationships, or national political impacts on operations, multinational interoperability is an overarching consideration.

I will examine these areas in light of past and current actions using primarily the 1965 Organization of American States (OAS) operations in the Dominican Republic and recent US and United Nations (UN) actions in Somalia (Unified Task Force, Somalia (UNITAF) and United Nations Operations, Somali II (UNOSOM II), respectively). Illustrations from other operations will be used as appropriate. While not presented as complete case studies, these operations allow me to illustrate certain recurring themes.

These two operations are ideal for examining C2 for peace enforcement. First, they give us a 25 year window to see what was learned, what was learned and forgotten, and what was never learned. Secondly they have many common traits. Both included a number of mission changes and combinations including peace enforcement, peacekeeping, humanitarian intervention, and nation building. Both included US unilateral action and multilateral action under an organizational sponsor. In the Dominican Republic, this sponsor was the OAS, while in Somalia it was the UN. As in most peace operations, both were highly politicized and controversial, and are excellent laboratories for our discussion of multi-lateral C2.

Of course it is fallacious to think of any operation entirely in the light of the past. Each mission is unique with its own ethnic, political, military, and logistical aspects, but certain factors seem to have a continuing relevance. If we are aware of past failures and successes, we increase our chances of successfully accomplishing the mission.

Although much could be written about the grand strategy of peace enforcement operations, I will focus on operational issues. I specifically want to shed light on problems encountered by military members executing the military mission. While I will highlight problems and discuss possible solutions, my intent is not to lay blame, but to capture and preserve knowledge so hard-won and so easily lost.

How do we measure overall C2 effectiveness? While difficult to statistically quantify, success requires forces to work effectively and efficiently within the political framework to accomplish political objectives while maximizing productivity of resources, both human and material. The following chapters systematically examine how command structure, political influences, and interoperability have affected the overall effectiveness of command and control in peace enforcement missions.

Perhaps the most important factor in a peace enforcement mission's success or failure is its command and force structure. Some past multi-lateral peace operations have been performed under the authority of regional organizations, but most (since 1945) have been under the UN. Some have been single-nation led multi-lateral activities under the authorization (but not control) of the UN. A significant debate is currently underway with regard to the role regional organizations should play in peace operations. Some have suggested a greater role for regional organizations such as the European Community or NATO, citing the vested interest in peace such organizations have within their region. Others have pointed out the weaknesses and limitations of regional organizations. Still, there appears to be a role for both UN and regional interests. This paper's two cases provide insight into this choice.

Both the Dominican and Somalia involvements are examples of multinational operations under international control which evolved from unilateral US actions. The Dominican Republic intervention transferred to OAS control, while the Somalia mission was taken from UN control, operated as a US-led coalition (though under UN mandate), and ultimately returned to UN control.

In the sections below, I address the issues involved in supra-national command structures, discuss the dual lines of control in multinational forces, examine the decisions which guided the grouping and subordination of contingents, and examine the place of humanitarian organizations within the force structure. After analyzing our two example operations, I will offer some observations and recommendations regarding multinational force structures under the control of the UN or designated regional organization.

The Dominican Republic intervention was characterized by its commander (later deputy commander), Lt Gen Bruce Palmer, as "far more than just peace-keeping--it was peace-making" (emphasis in original). This peace-making mission--now termed "peace enforcement"--required endorsement and subsequent control by the OAS to remain politically viable in the international community. In order to be successful, the local command had to be both responsive to OAS direction and efficiently accommodate force contingents from the US, Brazil, Honduras, Paraguay, Nicaragua, and Costa Rica.

In the Dominican Republic crisis, the OAS was chosen over the UN for pragmatic reasons. In 1965, the world was near a height in Cold-War tension. Against the backdrop of the Cuban missile crisis, and with the Communist revolution in nearby Cuba only a few years in the past, the prospect for another Communist takeover seemed likely. Since the UN Security Council included both China and the Soviet Union, intervention with the purpose of stopping a Communist revolution was unlikely. President Johnson wanted a multi-lateral effort to show widespread support for intervention, but clearly wanted to keep the affair hemispheric. The advantage in using the OAS was US control. The OAS was a smaller group with largely anti-Communist views and the US was clearly the leading nation. Dominican factions could also cooperate with the OAS without seeming to buckle under to US pressure. Interim Dominican leader General Imbert's first act of conciliation was to announce "full support of the OAS and proposed OAS-sponsored elections...." He had previously been intransigent in dealing with the US as a unitary actor.

A relatively sympathetic OAS allowed the US to cover the mission with a peacekeeping cloak while pursuing an anti-Communist agenda. As Lt Gen Palmer put it,

The OAS provided a way to accomplish the missions of ensuring a stable, democratic government in the Dominican Republic (a goal most Latin governments shared), protecting the lives of US and other foreign nationals, and rebuilding the nation's infrastructure without having to work through a divided United Nations. On the other hand, the OAS involvement carried some liabilities. Against the recommendations of US Ambassador to the Dominican Republic, W. Tapley Bennett and Lt Gen Bruce Palmer, the OAS insisted all forces in the Dominican Republic come under OAS command and that those forces (the Inter-American Peace Force--IAPF) be commanded by a Latin American general. Brazilian General Hugo Panasco Alvim took command of the IAPF, and Palmer became his deputy.

General Wheeler, Chairman of the Joint Chiefs of Staff tried to reassure Palmer by telling him the IAPF concept was simply a way to "legitimize our activities in world opinion by identifying them with the OAS." But the actual impotence of the OAS was not that clear. Certainly the OAS shared the vision of a stable, democratic Dominican government, but Gen Alvim was no figurehead. By Palmer's own account, he was a good commander and his own man, though the two leaders were usually able to reach a mutual agreement. While the US was driving overall policy, the OAS was fully engaged and refused to be a front for the United States. President Johnson was forced to walk a fine line between using overwhelming military pressure to stop a Communist takeover and bowing to OAS pressure for peaceful resolution through negotiation and compromise. After violent demonstrations throughout Latin America and OAS protests in the wake of 82nd Airborne Division deployments, Johnson feared further landings would jeopardize his effort to transform the US venture into a multinational OAS enterprise.

Meanwhile, the UN was not satisfied to remain a disinterested observer in the crisis. After an effort by the Soviet Union in the UN Security Council to derail US efforts and a long debate over OAS jurisdiction, the subject was seemingly settled by a personal letter from OAS President Mora to UN Secretary General U Thant explaining the crisis would be handled in-house, under the authority of the Organization for American States. Further pressure for direct UN action were subject to US veto on the Security Council.

This did not, however, prevent continuing UN involvement, and through official observer teams, the UN continued to make their presence felt throughout the crisis. The observers were characterized as "left-leaning" by Palmer, who accused them of rushing to send back the rebel ("Communist") side of every incident to New York without bothering to get the OAS side of the story. In order to counter this UN interference, Gen Alvim, the Brazilian commander of the OAS forces felt it occasionally necessary to show the UN that the OAS was in total control. On one occasion after the IAPF captured a rebel-held area in Santo Domingo, the UN Observer Team demanded its return to the rebels; Alvim refused both for tactical reasons and to show the UN it did not control the IAPF.

The transition from US unilateral to OAS/IAPF command was eased by the fact that most high-ranking Latin officers had trained in the US and were familiar with US command structures. By assembling a staff made up of Latin Americans in several key staff billets with US deputies while other key staff were filled by US officers with Latin deputies, the transition went fairly smoothly. Most American scholars attribute this to real control remaining in the hands of Palmer and Ambassador Bennett.

OAS oversight worked fairly well for the Dominican intervention. In light of the evolution of the world situation over the last 30 years, the feasibility of control by regional organizations must be re-evaluated. I will return to this issue in the chapter summary.

The presence of dual lines of control complicates multinational operations and particularly those under control of an international organization. The OAS operation in the Dominican Republic was not free of this phenomenon. Sovereignty and national interests are of prime importance to any troop contributor, and the seven nations who contributed forces to the IAPF were no exceptions. In multinational operations–and in peace operations in particular–leaders must handle conflicting guidance from various national command authorities and either resolve the conflicts or accommodate them.

In fact, dual lines of control were institutionalized in the authorizing document called "The Act Establishing the Inter-American Force." Though forces remained under national command, they would

During the intervention, conflicting guidance from multiple official and unofficial command lines was frequent. General Palmer, deputy IAPF commander, characterized his commander, Brazilian Gen Alvim as "a good leader and commander" who "ran the show" and definitely "not a rubber stamp." Nonetheless, Alvim ran into trouble when he tried to exercise direct control over US forces. When faced with differences between Alvim's desires and US intentions, the US answered with foot-dragging.

General Palmer used the enormous US leverage gained by the one-sided troop ratio and logistics support to ensure the June 1965, IAPF regulations gave him wide latitude to act in the name of the commander. According to Palmer, he defined his own duties as deputy commander and used his influence and that of the US to work out any differences of opinion with Gen Alvim. Meanwhile Ambassador Ellsworth Bunker, who headed the three member "Ad-hoc Committee" which provided OAS guidance, used his influence similarly.

Gen Palmer and the US were aware Latin American contingents (including Gen Alvim, himself) were in regular communication with their respective national command authorities. In the days before cellular telephones and routine satellite communications, these conversations were routed through the US-monitored telephone system.

The conflicts in national guidance also affected the US ability to use Latin troops. According to one US officer present, US forces were required to perform some actions unilaterally that the Latin Americans would not participate in. Despite these differences in national guidance, the forces seemed to work in surprising harmony. Palmer insisted that even though there were times he didn't agree with Gen Alvim's desires, "they always worked it out."

This apparently amicable relationship can be attributed to two factors. First, the political guidance of the OAS was highly influenced by the US. Secondly the two leaders had compatible personalities. The men respected each other as professionals; although they disagreed on many occasions, the common respect they felt for each other helped smooth their differences. In the final analysis military operations often succeed or fail on the basis of personalities. In this case Palmer and Alvim worked well together, but it was hardly something that could have been planned.

The IAPF command was apparently successful in working through and around conflicting national guidance to reach the desired end-state. Within the OAS, the vision of a free and democratic--or at least non-communist--Dominican Republic was shared by the troop contributing states and made a common military goal possible. Even so, Gen Palmer was surprisingly vehement when he said the operation was the first to put US troops under command of the OAS and should "never again" be permitted to happen. Despite Gen Palmer's views, pressure continues for US troops to be placed under international command.

Grouping and Subordination

One of the most challenging tasks in creating any multinational command is structuring the force. When grouping and subordinating various national contingents, diverse units must be organized and used to maximize their capabilities while accommodating political restrictions or requirements. Frequently, contingent subordination and grouping is not decided by the combined force commander; he usually has the organization thrust upon him. This was certainly true of the OAS force in the Dominican Republic.

The overall guidance for the structure of the IAPF came through The Act Establishing the Inter-American Force. It outlined basic structure and put operational command of all troops under the OAS, through the IAPF Commander. The act further outlined a combined staff containing members from each contributing force. Although the OAS authorized the commander to issue "Force Regulations" binding on all members of the force, it is unclear how he could enforce such regulations since UCMJ authority (or its equivalent) was not surrendered by the various nations.

In practice, the OAS guidance omitted many details of the structure. According to US Marine Corps historians, the commanders of each contingent were to work out directly among themselves and with the OAS the unified command structure and force composition. However, Palmer and Ambassador Bennett were to "prepare recommendations as to [the] structure and functioning of [the] Unified Command and submit these to Washington for approval before commencing discussions with [Latin American] Force Commanders or the OAS Committee."

In the early days of Operation Power Pack and Operation Press Ahead (the designation for the deployment of US and Latin contingents, respectively), the composition was not at all settled. After the final OAS vote on 6 May authorizing the deployment, Guatemala, Argentina, and Venezuela agreed to send contingents, but later backed out. Many of the arriving Latin contingents gave little formal warning to US officials in Santo Domingo, including Gen Palmer, who had to accommodate the troops until a 22 May resolution could be passed, designating Brazil to provide the IAPF commander. Palmer's designated successor, General Alvim, finally assumed command on 29 May 1965.

Ultimately, the force structure evolved along cultural lines. The troops from Paraguay, Honduras, and Nicaragua, along with the Costa Rican police detachment were grouped together into the "Fraternity Battalion." This battalion and the Brazilian forces were then formed into the "Latin American Brigade" under command of Col Carlos de Meira-Mattos of Brazil. The US and Latin forces rarely worked together. The Latin units were assigned independent missions and given "sectors of the perimeter" to man. The number of trained, Spanish and Portuguese-speaking liaison officers needed to effect lower level coordination made combined operations below brigade level impractical.

Effectively, all Latin Americans came under the control of the IAPF commander, while the US Forces remained substantially under the command of Gen Palmer. As stated above, since the goals of the OAS and United States were close, and the personalities of the two senior leaders were compatible, this arrangement seemed to work. In a case where either the personalities or the politics are decidedly divergent, such success would be unlikely.

Humanitarian relief organizations (HRO) must be included in any discussion of the force structure. While not formally within the chain of command, humanitarian relief organizations were integral to the success of the Dominican intervention. As with almost every peace operation, the OAS mission included a significant humanitarian relief component.

Although there was no formally designated Civil-Military Operations Center (CMOC) in the Dominican Republic, the OAS did have a formal mechanism for orchestrating the various relief organizations' efforts. The OAS Ad Hoc Committee--led by Ambassador Ellsworth Bunker--provided on-scene strategic guidance to the IAPF. The OAS Coordinating Committee, a civilian committee under the Ad Hoc Committee, was responsible for all humanitarian efforts. The Coordinating Committee, composed of representatives of HROs active in the relief effort, met daily for the first few weeks of the crisis to sort out missions and priorities. The committee coordinated the feeding of over 40,000 families, orchestrated medical care in cooperation with the Pan American Health Organization and the Red Cross, and was involved in civic efforts as diverse as water purification, insect control, garbage collection, and burials.

Although the OAS formal report does not mention military involvement in relief efforts, such missions occupied much of their daily routine. Marines guarded USAID workers and protected OAS and HRO food stocks from looters. Army helicopters transported Inter-American Human Rights Commission authorities, and distributing food occupied a significant portion of the IAPF's time. The daily coordination between the HROs and the military was largely accomplished through US Army civil affairs teams from the 42nd Civil Affairs (CA) Company. They worked with USAID and CARE to open elementary schools and with private agencies and USAID to coordinate the food distribution program However, one of the military's primary lessons was to turn operations over to "normal welfare agencies" as soon as possible.

One of the problems cited by the 42nd CA Company demonstrates that civil-military coordination was not all it should have been. Often, once the situation was safe enough for civilian relief agencies and Dominican authorities to resume services in a given area, they did so without informing military relief workers, thereby duplicating effort unnecessarily. Closer civil-military relationships would have significantly increased efficiency. Nevertheless, the overall relief mission can be rated as very successful.

Relief agencies also helped with the military mission. Although later described by Gen Palmer as "pathetically" naive in political outlook, he credited the Peace Corps with helping the IAPF deal with the left-leaning Dominican Catholic Church which held great influence in Constitutionalist-held areas. In another example, the Red Cross brokered the first successful cease-fire (after failures by US and OAS diplomats) on 21 May 1965.

The biggest complaint by Gen Palmer about HRO worker seemed to be with their political views. Many relief workers held liberal views and sympathized with the anti-government rebels. The military favored the Dominican government and army--the opposing faction. This tension is visible in the exasperation of Gen Palmer who described Peace Corps members as "basically good people responding to a lot of suffering, but [who] didn't understand why the US was there" and didn't understand they would all be dead if the military had not arrived.

Despite this tension, the OAS Coordinating Committee, civil affairs specialists, infantry soldiers, and Marines seemed to work well together. As a result, the Dominican infrastructure was back in working order by the time elections were held the following year. The lessons are basic: Including humanitarian efforts requires cooperation and flexibility (politically and structurally.) Diverse humanitarian agencies must also be willing to work together and submit to an oversight committee if relief efforts are to have any hope of success. In the Dominican Republic, this will to cooperate existed, though it was strained at times. Planning to incorporate relief agencies in the force structure paid off. Starvation and disease epidemics were averted, the infrastructure was reconstructed, and the shaky Dominican government which emerged in 1966 had a viable society to govern. In a peace enforcement operation, that can be counted as real success.

Reviewing recent operations in Somalia, the structure of that multinational peace enforcement operation presents additional, important lessons. The force and command structure in Somalia was unusual in that the mission flowed from a traditional, UN Chapter 6 peacekeeping operation (UNOSOM I) into a US-led coalition under Chapter 7 (peace enforcement) UN sanction (UNITAF), to a UN-controlled Chapter 7 operation (UNOSOM II). I will limit my discussion to the force structure under UNITAF and its transition to UNOSOM II. As in OAS operations in the Dominican Republic, the command structure had to facilitate responsiveness to the governing political authority and efficiently accommodate a wide range of multinational forces.

UNITAF took over from UNOSOM I, an organization characterized by a senior UNITAF commander as "incapable of executing the Somalia mission and virtually immobilized by UN bureaucracy and Chapter 6 rules of engagement," and this "immobilization" was an impetus for the US-led UNITAF. The United Nations needed a self-sustaining, overwhelming force to intimidate the armed factions in Somalia and allow the humanitarian relief to reach the starving. This goal was endorsed by many Americans, and on the heels of Desert Storm, America seemed ideally positioned to perform the mission. Even within the coalition force it demanded, the US could provide the sorely needed decisive command. The US deliberately limited its mandate to providing security for the relief effort and the coalition members themselves and sought an early end to military involvement. UN Security Council resolution 794--crafted largely by the US--laid out very limited objectives to be reached leaving the nation-building to the UN.

In the Dominican Republic, a regional organization took over and performed the role of peace enforcement and nation building. Would that have worked in Somalia? The short answer is no, for a number of reasons. One candidate, the Economic Community of West African States (ECOWAS), is financially limited and would have been unable to field or maintain the forces necessary for the operation even if it could have reached political consensus on the mission. ECOWAS action in Liberia–dominated by Nigeria–has been particularly bloody, and overall performance has received mixed reviews at best. Other regional organizations have equally weak or non-existent military capabilities. Though the UN Charter does provide for the involvement of regional organizations under UN sanction, no organization in the region could do the job.

A major argument against regional organizations involves the issue of impartiality. Hostility and "bad blood" go back centuries in the Horn of Africa region. Some Somalis still talk of Greater Somaliland, and though relations have recently improved with Kenya and Ethiopia, most Somalis feel racially and ethnically superior to their neighbors who would likely have been ill-received as peace enforcers. The Somalis' racial superiority complex does not extend to white Europeans and Americans whom they consider equals.

Though a regional organization was unable to assume the entire mission, they have had a continuing role in the overall peace mission. General Zinni, UNITAF J-3, cited the OAU involvement as arbiters and particularly highlighted the contributions of Ethiopia and Eritrea in setting up dialog among warring clan leaders. A State Department official singled out the contributions of the Horn of Africa Committee for similar diplomatic efforts. Regional organizations have a vested interest in seeing peace in their region. Even hostile neighbors are interested in keeping regional conflict contained and in stemming the economically devastating flow of refugees caused by an intrastate conflict. In Somalia, the regional organizations, weak though they are, are making an important and continuing contribution. As one military advisor at the UN recently put it, "regional organizations can play a role--and you certainly don't want them against you."

Regional organizations could not, however, perform the total mission, and the US chose not to remain indefinitely engaged. The UN was the only long term option, eventually taking over in May 1993. However, some insist the deck was stacked against UNOSOM II from the beginning. The US was anxious to leave, feeling they had accomplished their limited objectives, but the UN was far from ready to receive the baton. The US, which had planned for a five or six week mission when it agreed to head UNITAF, was ready to turn over the command by mid-January. As the UNITAF commander was able to assign humanitarian relief sectors (HRS) to non-US forces, he began--with the approval of CINCCENT and the JCS--to reduce US forces and redeploy them to their home bases. (See Figure 2.) In his words, Gen Johnston "knew his mission had been accomplished" and was waiting for the UN to assume the greater mission formally. However, Resolution 814 authorizing UNOSOM II was not passed until late March, and the first staff members were not in place and ready to start the transition until April.

To say the joint task force staff was frustrated with the process is an understatement. Already far behind their expected departure, Gen Johnston agreed to a six-week transition with dual staffs, but most of the UNOSOM II staff had still not arrived even by mid-May. He voiced frustration, saying, "the UN has known for eight months they're coming to Somalia and there's still no game plan." He later termed the take-over "inept." Had it not been for the quick addition of a handful of top-quality US Army officers to the UNOSOM II staff, the hand-off would have been even more difficult.

Other officials saw it differently. Walter Clarke, deputy to the US Special Envoy to Somalia and Col Bruce Clarke, US Army expert on end-state planning, agreed the US had selected success criteria in Somalia which were hopelessly narrow. Even when met, these criteria were an insufficient starting point for UNOSOM II to have any hope of success. The US refused to disarm the factions unless they presented a direct threat to either relief workers or the UNITAF forces, themselves. Nor did the US want to undertake any sort of systematic nation-building. This was left to UNOSOM II which had neither the resources, nor the unity of effort to do the job. An Australian military advisor to the UN pointed out the ambiguity of Resolution 814 which some nations, such as Italy, took to be a peaceful mandate and who refused to participate in the subsequent military operations to pacify warring clans.

All seem to agree that the hand-over to UNOSOM II was muddled and incomplete. The UN was under enormous pressure to assume an unfamiliar Chapter 7 role and the US commander was under pressure to terminate what he saw as a successful UNITAF mission. Regardless of blame, clearly the UN was unready to assume command of the mission long after US forces were ready to transfer it to them. The major problems can be loosely grouped into the categories of organization and policy.

Organizationally, the UN lacked a sufficient C3 structure--either in New York or in a deployable staff element. Both UN Ambassador Madeleine Albright and President Clinton have cited the need for a true "Command Center" in the New York headquarters, staffed by competent, experienced staff who can make command decisions. The fact that, until recently, no one staffed the UN peacekeeping office past normal duty hours has been well publicized. Saying the UN closes down after 5 P.M. is an exaggeration, but it is true that getting a decision after hours is still difficult.

General Zinni among others has said that the UN needs a standing or stand-by deployable headquarters staff which would perform much like a CINC's staff with similar authority to command and control UN military operations Such a staff should have immediate access to the command center in New York, and in turn, direct access to the Secretary General This would have eliminated or significantly reduced the waiting to "spin up" the UNOSOM II headquarters. Instead, the C3 transition plan was still being created after the JTF transition to UNOSOM II had occurred. A deployable headquarters should also be backed by stand-by forces earmarked by their nations for UN deployment. These forces would regularly exercise with the UN staff and with each other in realistic scenarios. Mrs. Albright listed the lack of such a robust military force as one of the failures of the UN in Somalia.

Procedurally, command lines are still very muddled. While sovereignty dictates and the UN Charter recognizes nations will retain authority over their deployed forces, a more streamlined command methodology must be found. One of the biggest problems with UNOSOM II was the three chains of command to the US forces in theater.

UN commander, General Bir, held some authority for action. General Montgomery, Bir's deputy and senior US officer, answered to both Bir and to Admiral Howe, the Secretary General's Special Representative. While Admiral Howe was a UN official, he was also an American hand-picked for the job. When Bir and Howe disagreed, it put Gen Montgomery in a difficult spot. To make matters even worse, after the ambush of the Pakistani's on 5 June and subsequent decision by Howe to go to offensive operations, all missions by the Army Quick Reaction Force (QRF) in Somalia had to be approved by USCENTCOM headquarters in Florida. The QRF was not actually under UN control, but was to be available on-request from UNOSOM II, through Gen Montgomery. However, after 5 June, Montgomery was basically cut out of the QRF control. Thus, US forces had to contend with three heads: Gen Bir, Admiral Howe, and USCENTCOM. The added complication of the contingent forces' national commands made the entire operation almost hopeless from a command and control perspective.

In fact, after the shooting began in earnest, General Bir's control in Mogadishu was severely tested. The Italians, Pakistanis, and Moroccans drew into defensive postures and refused to man road blocks, patrol, or operate at night. The Pakistanis even went as far as to make "back room deals" with Gen Aideed's forces for safe passage through UNOSOM areas.

Even during UNITAF, dual lines of command caused significant problems for Lt Gen Johnston. While some contingents such as the Australians and Canadians had fairly broad authority from their governments, others like Saudi Arabia had to check with their national command authority before each action. This forced the JTF headquarters to employ the Saudis only in missions which could be pre-planned. Other nations were bound by their governments to "non-heroic" missions and could only be used with great restrictions on their movement.

Dual lines of control are an inherent problem of multinational operations, but to minimize the impact, the overall scope of the operation must be outlined to the contributors and agreed to in advance. Contributing nations should openly outline employment restrictions before the operation is committed, and the UN headquarters must be willing to set certain requirements for participation. Furthermore, dramatically changing the operation's mandate--even if unavoidable--requires consultation with the troop supplying governments. This type of negotiation will probably require a significant amount of time which must be expected. Standby forces, a standing headquarters staff, and combined UN exercises will go far to minimize confusion and shorten the planning cycle. But when it is time to commit troops, the employment negotiations will still take time. Anyone expecting the UN (or any other multinational body) to work quickly is due for disappointment.

With all the caveats on how the various contingents could be employed, it was even more important to organize them to maximize their capabilities within political limitations. In UNITAF, the task of organizing and subordinating the units became largely a negotiation between the supplying governments and the JTF staff. This was especially trying since, in the words of Lt Gen Johnston, "the politicians, diplomats did not...really have a game plan for who the coalition members are [sic] going to be." By the time UNOSOM II took over, the relationship was largely geographic with each humanitarian relief sector (HRS) under a contingent force and each HRS reporting directly to the UNOSOM II staff. It is useful to look at the UNITAF model because most lateral relationships among contingents were worked out in that phase, and no contingents were added for UNOSOM II.

The troop contributors were initially divided into two basic groups: those who were self-sustaining and able to operate independently, and the contingents who were either too small to operate alone or had inadequate logistics to do so. For some larger units and all the smaller ones, the initial screening for assignment location, subordination, and capability fell to an ad hoc unit called the Coalition Forces Support Team (CFST). The CFST began as a totally Marine-staffed unit, with a few Army members added later to give it a joint flavor. The CFST worked with advanced teams or national headquarters to establish the logistical, operational, and geographic parameters under which a contingent force would operate. Based on what they would do (politically) and what they could do (logistically and militarily), the units were assigned. A key part of the assignment process concerned unit subordination. Some contingents–the Italians in particular–insisted they work directly for the JTF commander, while others were satisfied to be employed as seemed most appropriate.

All brigade-size or larger units were placed either directly under the JTF commander or the ARFOR commander, MGen Arnold. Italian, French, and Canadian forces reported directly to Johnston, while the Australian, and later Belgian and Moroccan, forces worked for Arnold. Even some of the larger national contingents had severe restrictions. The Belgians, who had an internal agenda to work with one of the HROs, Doctors Without Borders, insisted they be assigned to Kismayo. Nations who provided theater airlift loosely subordinated their airlift forces under the US Air Force, while nations who could only service their own contingents kept their transport under national command.

Most smaller units were unable to leave the area immediately around the airport, either because of logistics shortfalls (for instance lack of vehicles), or unwillingness to work in the riskier areas. These limitations caused them to fall under a succession of battalion commanders, eventually ending with Lt Col Lesnowicz, commander of 3rd Battalion, 11th Marines. Col Lesnowicz ultimately commanded a collection of units equivalent to a reinforced mechanized regiment (5 battalion equivalents and nine nationalities). The happenstance manner in which Lt Col Lesnowicz came to command this diverse assembly makes it even more impressive that it seemed to work so smoothly with so few incidents. When one of Lesnowicz's superiors apologized for "dumping" all the small, third-world units on him, Lesnowicz bristled, replying that he was honored to have those units under his command.

In addition to the issue of the subordination and geographic assignment of the contingents, just getting them to Somalia was a challenge. The US was tasked–often on very short notice–to airlift most contingents. However, since UN airlift requirements were factored into the time-phased force deployment data (TPFDD), US airlift was not impacted by the foreign demands. Air Mobility Command (AMC) did get welcome assistance from British and New Zealand airlifters, but most of the burden fell to US Air Force C-5's and C-141's. According to AMC officials, their biggest overall challenge was lack of fuel and ramp space at Mogadishu.

In peace operations, humanitarian efforts play a major role. Resources and organizations outside the normal command structure should be used efficiently to meet the inevitable humanitarian needs.

Even more than in the Dominican Republic crisis, peace enforcement in Somalia was overshadowed by the work of the humanitarian relief sector. MGen Zinni, UNITAF J-3, put the number of HROs at 60 with 518 separate facilities, while the Center for Army Lessons Learned puts the number of HROs at more than 80. To improve the efficiency of the relief agencies, the UN set up a humanitarian operations center (HOC). Its stated purpose was to coordinate the efforts of all HROs, reduce duplication of effort, and provide a forum for discussion and resolution of common problems. Recognizing the difficult mission of the HOC, the UN hired CARE president, Phil Johnson, to head it. Referred to as the "backbone of the HOC," Johnson was an aggressive leader who did his best to coordinate overall strategy for the HROs. To augment the HOC, a civil military operations center (CMOC) was set up to coordinate military convoy-escort duties and serve as a liaison between the armed forces and the HROs. Unfortunately, illness forced Johnson to leave Somalia, and his absence created a tremendous void no one was willing to fill. According to Gen Zinni, the CMOC was forced to step in and take over most of the HOC's coordination and integration function.

Fortunately, recent experience in northern Iraq had given USCENTCOM, and Gen Zinni in particular, valuable experience in operating a CMOC. Col Kevin Kennedy, Operation Provide Relief chief of staff also brought valuable experience to the table. Because of his familiarity with the area and the HROs, he was selected to head the JTF CMOC. Still, the unfamiliarity of the average military person with the HOC and vice versa created considerable confusion. While not entirely avoidable, this unfamiliarity and distrust could have been reduced by including HROs in previous exercises. An exercise held by the First Marine Expeditionary Force (I MEF) using an Ethiopian scenario excluded all relief agencies, even though originally proposed by the exercise creators.

In addition to the military's ignorance, the HROs were unfamiliar with, and in many cases, unaware of other HROs doing similar jobs. These organizations, staffed in many cases with idealistic personnel, resented regimentation and often distrusted the military. While the CMOC did a good job as a clearinghouse for information, it was not authoritative and could not compel cooperation. As one Army battalion commander put it, "The [HROs] did a lot of good things, but [were] not under control."

Even coordination between UN organizations such as the UN High Commission on Refugees, the World Food Program, and the UN Development Program and the military left much to be desired. The coordination was so bad that when these organizations were restricted for security reasons from entering certain zones of Mogadishu, the organizations continued to carry out their tasks, totally unaware of their restrictions or the danger.

The military quickly recognized the need to organize relief, but made the mistake of mirror-imaging. Treating the situation much like an intelligence problem, they made it an essential element of information to find the mandate, political philosophy, and chain of command for each group. Unfortunately, they found that many groups were splintered and had no recognizable chain of command. Many of the organizations were manned by "green" recruits even more confused about their environment and mission than was the military.

There were, however, some bright spots. The idea of using an intelligence framework for gathering information on HROs had some merit. BGen Lawson Magruder, deputy ARFOR commander, recommended getting the HRO "order of battle" to find out about capabilities, personalities, and areas of responsibility. He felt this information to be of primary importance in a peace enforcement mission like Somalia. The 10th Mountain Division learned during hurricane Andrew relief to set up the military boundaries to correspond to civilian agency zones. This at least limited the number of HROs a given military unit had to work with. The best relationships were built when a military unit could build a one-to-one relationship with an HRO and form a sense of mutual understanding.

Another positive lesson was the use of the Army's Movement Control Center (MCC), a unit of the Corps Support Command. Regardless of political or religious orientation, all HROs were interested in effectively moving their relief convoys. The MCC approaches directly enhanced HRO aid distribution, and their services were soon highly sought after. The MCC served as a "foot in the door" for further cooperation between civilian and military organizations. Marine Forces set up their own CMOC, and though originally strictly limiting their focus to convoy protection, quickly expanded their role. They began distributing food, and even set up civil (Somali) police forces and organized local government councils. They worked alongside the HROs in an effort not only to help the people of Somalia, but also to lessen the threat to themselves by becoming "valuable" to the Somalis.

The humanitarian effort was plagued by the need for protection. Though the US forces agreed to protect convoys, they were unable to provide continuous protection for the multitude of workers and installations throughout Somalia. Walter Clarke, deputy to the US Special Envoy, attributed this lack of military protection to Gen Johnston's reluctance to enlarge the mission. He specifically cited a refusal in March 1993 to deal with direct threats by Somali "thugs" to kill ICRC and CARE officials if they would not pay $500,000 in "back wages." Most military officials, however, attribute the lack of comprehensive security to UNITAF's narrow mandate and the impracticality of guarding the vast number of personnel and installations within the HRO establishment.

In any case, the HROs took matters into their own hands, perpetuating a necessary evil which continually plagued both UNITAF and UNOSOM II. This was the use of Somali "bodyguards." Gen Johnston phrased the dilemma directly, "How do you disarm [Somalia] without disarming the [non-governmental organizations] or providing a security umbrella?" To answer the problem of controlling the armed Somali HRO guards, the military came up with a series of weapon-authorization cards. After a failed first attempt using generic "pink cards," they settled on blue ID cards carrying photos of an HRO worker (not a Somali). (See Figure 3.) The worker with the card could vouch for any guards in his or her company. While this solved the problem of independent, gun-carrying Somalis claiming to be "guards," many problems remained. Though the cards limited guards to small arms, many HRO guards continued to carry heavy weapons such as crew-served machine guns. These were routinely confiscated. It is easy to speculate that at least some workers, already fearful for their lives, were pressured into being "guarded" and served as unwilling covers for Somali bandits.

One aid worker, Sean Devereaux, of Irish Concern, tried to demilitarize the HROs in Kismayo. He formed a kind of "board of directors" of the local relief agencies and through force of personality and determination, helped organize and coordinate their efforts. He even began to have some success in convincing local relief agencies to fire their Somali "protectors," but was killed by his own former guards for his efforts. The incident sent a chill through the agencies, who quickly re-hired their guards. His loss set cooperative efforts back weeks.

The lack of overall coordination among humanitarian agencies and between HROs and the military significantly limited the overall success of UNITAF and later, UNOSOM II. Although one commander described the CMOC as "standing-room only every morning," packed with relief workers, it really had no ability to coordinate or control. While its function as a clearinghouse for information was good, the CMOC and humanitarian operations center it supports must receive more attention in the future and become a true relief coordination center. The relief and nation-building efforts in a peace enforcement mission are more critical to lasting peace than any effort by the military to force a halt to fighting.

In studying the OAS command and force structure and that of UNITAF and UNOSOM II in Somalia, some recurring patterns emerge we can use to improve the chances for success in future peace enforcement operations. In order to have an effective multinational command, a viable political and military command structure is essential. One way to achieve this is to institute an organizationally sanctioned, US-led command structure to act as proxy to the multinational organization–the UNITAF model. However, this has begun breeding resentment among some non-US officers who fail to understand why the US "always has to be in charge."

Another option uses a true, integrated UN command, incorporating both political and military leadership. UNOSOM II was supposed to be such a command, but its failures point to the need for a standing UN military command headquarters staff with access to senior political leadership. As mentioned earlier, this command staff would need to be backed by designated, stand-by force contingents from troop-supplying nations, and these forces would need to exercise realistically with the headquarters staff on a regular basis. The UN is taking initial steps in this regard and has established a 24-hour situation center to maintain a link with field operations. They have also taken steps to expand the staff of the Department of Peacekeeping Operations, adding additional political and military officers. These are good first steps, but to have any credibility as a command and control center, the other steps I've outlined need to be accomplished as well.

A third structural model employs a regional organization to command the peace operation. While successful in the Dominican Republic, this was a special case unlikely to be repeated in the future. First, the United States was a member of the OAS and was able to use Cold War pressures to achieve unity in a small regional organization. The US dominated the organization militarily and politically, and its position in the UN Security Council allowed it to block interference by the UN. In the post Cold War world, the UN is increasingly the sole empowering agency for peace operations. OAS sponsorship for peace operations has at least twice been rejected since 1965--in Nicaragua in 1979, and in Central America in 1990. While regional organizations should and will play a part in peace enforcement missions, it will likely be either in a diplomatic role or as a "sub-contractor" to the UN. NATO's role in former Yugoslavia is a pertinent example. For a regional organization to be useful, its political goals must be compatible with the UN, especially if a UN resolution is in effect. While accountability to the UN might be loose, it will be real.

Having a capable command does not totally eliminate the problem of dual lines of control from the UN and from the contingent's national government. It does, however, provide a framework for working through many issues of sovereignty before they become a life-or-death matter. US doctrine does little to resolve this dilemma, when on the same page of FM 100-23 (Peace Operations), it says both "the UN commander normally has operational control" and "the US commander will retain full command authority." It goes on to say that the command relationship between US forces and international organizations will be "subject to mutual resolution." True, but hardly helpful. The only way to work through these command authority issues is by creating a true UN command structure and conducting exercises to refine it.

Issues surrounding sovereignty and the reluctance of nations to create a supranational command make the creation of a UN military headquarters difficult. Even if a standing UN headquarters could be implemented, there would probably still be strong resistance to turning full command of national forces over to UN command. Even so, the creation of a standing headquarters unit would eliminate much of the time currently required to name a commander and assemble and train a staff each time a crisis arises.

The final composition of such a UN command headquarters would have to be somewhat flexible, of course. Col Colmenares, a veteran of UN Operations in El Salvador (ONUSAL), points out that the need to preclude officers from holding important positions who are from countries with "profound differences" with other countries in the force. Colmenares also maintains the staff should be truly representative of the contingents involved, unlike the one-sided UNITAF model. Despite the colonel's concern, a force composed from predominantly developing nations is unlikely to have sufficient large-scale logistics or command experience to successfully conduct or control a large operation.

In coordinating the humanitarian relief efforts, both the Dominican and Somalia operations highlight the need for closer relations among humanitarian agencies, the sponsoring political organization (UN or OAS), and the military. Civil affairs personnel played a large role in the Dominican Republic in trying to maintain this link. While not entirely successful, they and the OAS Coordinating Committee had significant success in coordinating the relief and rebuilding effort. Official agencies such as USAID were also very helpful in the Dominican Republic. Gen Zinni has stressed the need for USAID and their Office of Foreign Disaster Assistance to play a larger role. Their Disaster Assistance Response Teams (DART) are a big help to any agency and should have a major part in the CMOC and in helping the operations and plans directorates as a special staff.

One way to bring about closer HRO-military relations has already been mentioned--their involvement in joint and combined exercises. To build trust, these exercises may initially be strictly humanitarian relief scenarios--for example a hurricane or flood relief mission. Later, peace enforcement exercises should be included. The military cannot expect to work as well with agencies and individuals who have had no contact with the military until a real situation develops. One imaginative measure the Dutch have taken is to create a one-week course in their UN Training Centre for journalists. This helps foster understanding between the two communities and is popular among journalists. A similar training course in our professional schools for humanitarian agencies should pay big dividends.

To improve the command and force structure for the next peace enforcement mission we must begin now. The answers outlined here are not comprehensive, but are a starting point. Peace enforcement is unfortunately a growth industry. We cannot afford to continue doing business as usual if we expect success in the next mission.

Although Clausewitz reminds us that war is fundamentally political, military officers continue to be surprised and frustrated by political intervention in military operations. If war in general is political, we should expect coalition peace enforcement operations to be doubly so. In peace enforcement missions, the objective is not to defeat an enemy, though a faction may have to be forcibly restrained. The objective of peace enforcement is to reconcile the warring sides to a lasting agreement. Gen Palmer, commander of US forces in the Dominican Republic, said his mission was at least as much political as military. So it is in every peace enforcement operation.

Peace enforcement missions usually require intervention in one or more nations' "internal affairs" with a third-party force. These interventions are frequently multinational affairs with each participating nation having its own agenda. Even under UN or regional authority, troop contributors will have different views of the intervention and different objectives. Complicating matters further, the sponsoring organization will have its own agenda which may not mesh with that of any individual participant. These factors combine to make peace enforcement operations especially subject to political and diplomatic considerations, with significant impact on the associated military options.

In the following sections I will examine some of the political aspects of peace enforcement operations in the Dominican Republic and in Somalia. In particular, I will focus on rules of engagement, "prisoners of war," national restrictions on military roles, the need for political awareness in the military, and the necessity to find an "enemy." After looking at these operations, I will offer some general observations and recommendations for future peace enforcement actions.

In the Dominican Republic intervention, the ROE had to be understood at all levels and reflect the political objectives of the OAS and its member states. Balancing this was the need to keep ROE from becoming a hindrance to operations or leading to needless casualties.

In Gen Palmer's oral history, he claims no rules of engagement were prescribed for operations in the Dominican Republic, so "we developed our own." Even if technically true, the ROE were certainly under scrutiny and were consistently tightened by Washington. Lawrence Yates cites State Department telegrams to the embassy in Santo Domingo carrying orders from the President to use no more force than necessary and specifying particular areas and actions which could be supported with US troops. When the Latin American troops arrived, they accepted the same rules of engagement and promulgated them in their operations plans. As expressed at lower echelons, the ROE, even early in the operation (5 May 1965), were tight indeed. Orders issued by the commander of 1st Battalion, 6th Marines specify that "Marines were permitted to fire only when they had been fired upon directly and had a clearly identifiable target which could be hit, preferably with one round." The Marine order also limited such responses to small arms. No mortars, cannon, or crew-serve weapons were to be used.

The ROE were understood by the American marines and soldiers, though they were not liked. Well-disciplined, they seldom violated the orders. In order to avoid any chance of injudicious application of force, Palmer denied the decision to fire to individuals or their immediate superiors. Only platoon leaders or higher could authorize return fire except in extreme situations. Concerned about over-stepping their bounds, the Marines forbid even the application of riot control agents without direct approval by a battalion commander. By 28 May, the Marines were forbidden to carry more than 15 rounds of rifle ammunition (10 for most units) or 5 rounds of pistol ammunition.,,

While American forces honored the letter of the ROE, the Latin Americans–officially under the same rules–interpreted them much more loosely, and often violated them freely with little resulting criticism. At one point, the Latin American Brigade came under sniper fire, abandoned the ROE, and started a prolonged fire fight with Constitutionalist forces. Gen Palmer later characterized their response as "trigger happiness." He singled out the Hondurans' lax fire discipline, saying they "loved to throw hand grenades like popcorn."

Though strict from the start, ROE continued to tighten as diplomatic efforts to solve the crisis intensified. In general terms, the ROE began as "don't fire unless fired upon." This was enforced with increasingly strict discipline. Eventually, as political factors became more important, OAS forces were forbidden from capturing and holding rebel positions. According to Palmer, "the US decision to seek a negotiated settlement required forces to not return fire–to shoot only when absolutely necessary to protect their lives or positions." Rules for conducting checkpoints and roadblocks also seemed to change frequently. Marines spoke of information from "higher authority" which "constantly changed as to who could pass checkpoints, what passes/credentials were to be honored..."

Ultimately, the ROE passed through more constrictions, ending in a "prohibition against firing unless one's position was in imminent danger of being overrun." This increasingly strict ROE led to dangerous situations for the IAPF. There are numerous cases of direct danger to OAS forces due to strict ROE, but one serves to illustrate the problem.

During a morning clearing operation on 30 April 1965, Company I, 3rd Battalion, 6th Marines came under heavy fire from a crew-served machine gun and numerous snipers in buildings near the airport. With his men pinned down, the company commander requested permission to use a bazooka (3.5" rockets) to clear the buildings. His request went beyond normal ROE, so it had to go up the chain of command–eventually all the way to President Johnson who ultimately authorized the use of rockets. However, this authorization did not come for several hours. Meanwhile, the company was forced to clear the buildings from the inside using a series of squad rushes. In multiple incidents, they suffered one marine killed and eight more wounded before authorization to use the rockets came. Using the bazooka, Company I quickly reduced the remaining positions with no other losses.

ROE must be appropriate to the military and political situation, and clearly communicated to forces at all levels in the chain of command. The rules should not place military men and women in untenable situations. Where such rules are dictated, the commanders on the scene have an obligation to protest vigorously and the authorities in Washington or New York have an obligation to listen.

One of the stickiest problems in peace enforcement operations remains the issue of prisoners. This includes both what to do when our forces are taken captive, and what we do when peace enforcers take prisoners, themselves. In the Dominican Republic, OAS forces were under constant threat of capture. In some cases, soldiers on patrol strayed out of OAS-controlled areas and were taken captive. Sometimes they were returned unharmed, but on other occasions, bodies which had been "found" were brought to OAS checkpoints. On at least one incident, direct negotiations between civilian OAS representatives in Santo Domingo and rebel forces yielded the release of two unwounded Marines after a third was shot and killed. Marine command diaries refer to the two recovered Marines as POW's, even though there was no declared war and the Geneva convention was not invoked. The subject of how to handle POWs never appears in Gen Linnell's staff meeting notes of the period, so if there was any general policy about how do deal with POW's it is not recorded.

The subject of enemy prisoners of war (EPW's) is more fully covered in the records. The first prisoners were taken soon after the 3rd Brigade of the 82nd Airborne Division arrived in country. The prisoners were officially referred to as detainees to avoid legal implications, since the operation was a peace enforcement action, not a war. According to Yates, prisoners were taken "in bunches" and because of inadequate facilities and lack of military police (MPs), the first "batch" was turned over to Dominican Police. Since the Dominican Police were an agent on the opposing faction, this was like letting the fox guard the hen house. The Dominican Police simply rounded up the prisoners in a courtyard and shot them, earning a black eye for not only the interim Dominican government, but the US as well. As a result, the Army stepped up efforts to bring in adequate numbers of military police, and US forces were instructed not to turn over any more prisoners or their weapons to the Dominican Police. The "detainees" or "line crossers" were now interviewed by counter-intelligence linguists and photographed. Counter-intelligence and civil affairs personnel would screen "line crossers" and "select detainees to be evacuated." (Emphasis added.)

In hindsight, the shortage of MPs seems like poor planning followed by bad judgment in turning prisoners over to their local opponents. It seems the US went to the Dominican Republic expecting to fight a war. Initial intelligence said the US would be fighting 25,000 armor-equipped "Communists." Even so, the force packages did not provide for MP/POW handling troops, instead specifying only combat forces. Furthermore, checkpoints were jointly manned at that time using US forces and loyalist Dominican Army and Police personnel. Clearly, the US was not neutral, and this blinded them to the danger of turning over members of one faction in the civil war to the other faction. Politically, this incident provided part of the impetus to push US and OAS forces to a more neutral stance. I will return to the need for neutrality later in this section.

One of the most direct ways political considerations affect military operations is through constraints placed by national command authorities on their forces. As in every peace operation, the US, and later IAPF, forces were significantly restricted. Within their constraints, commanders had to employ forces to maximize their effectiveness. In a conflict where distinguishing combatant from civilian was nearly impossible and where a purely "military solution" was rejected almost immediately, troop commanders were in a particularly difficult situation. However, experience shows that in peace operations, this is the normal situation, not an aberration.

Despite the precarious diplomatic balance the US and the OAS were attempting to strike in reconciling the factions, the leaders–at least in the early months–saw a war which "could be won." In late April, only a few weeks after US troops had landed, Ambassador Bennett and key military counterparts, MGen York and Commodore Dare, thought a showdown with the Constitutionalist forces likely. They deployed their troops accordingly, paying scant attention to OAS efforts to mediate a cease fire. Though York, as ground forces commander, witnessed a cease-fire agreement signed by the ruling Junta and the Constitutionalist leaders, he did not recognize the agreement. After arriving, Lt Gen Palmer also repudiated any such agreement feeling it put US forces in an untenable situation.

As the situation unfolded, control from Washington (both US Government and OAS) tightened. Though General Wheeler, the Chairman of the Joint Chiefs, decried field commanders "having their hands tied by....theorists at higher headquarters," the hand-tying continued. According to the Marine History report, the operation was governed by "close civilian and military control at the highest levels of government in Washington."

What was the effect of this control? In many cases the IAPF was restricted both geographically and operationally from conducting operations as they would have preferred. While the OAS ambassadors strove to bring the factions to the bargaining table, York, and later Palmer, fought for increased troop numbers. They also argued for the need to split the rebel forces and unite the geographically separated Army and Marine controlled areas. President Johnson tried to accommodate both sides, and as normal in such compromises, pleased neither. Later, on 15 June 1965, the Latin American Brigade and the 82nd Airborne attacked and pushed the rebel (Constitutionalist) forces almost into the river. A military victory seemed imminent which would put the ruling Dominican military junta in sole command of Santo Domingo. But, just as the final push was about to begin, Gen Palmer, under orders from Washington, overruled Gen Alvim and ordered York to stop his troops (the 82nd). Washington felt strongly that a diplomatic solution, not a military one must resolve the Dominican conflict. As quoted by Yates, General Palmer said neither York nor Alvim understood the rationale and "didn't like it worth a damn." This was the last time a military solution by IAPF troops was attempted.

The case of Radio Santo Domingo further demonstrates the political restrictions. The creation of an International Security Zone had been approved by Washington and occupied by OAS troops. This zone split the city into a southern section predominantly occupied by the interim government and a northern area controlled by Constitutionalist forces. On the recommendation of the MAAG group commander, Colonel Joe Quilty, the zone left the primary radio station in the hands of the Constitutionalists. Although the station was used to broadcast anti-OAS and anti-US propaganda, it remained beyond IAPF control. The military, and later the CIA, demanded its removal; but Washington forbade any moves to silence the station. Dominican government forces eventually captured it in the last two weeks of May 1965. Palmer later said the OAS forces had been in no way able to interfere with the Dominican government operation and did not want to become the "ham in a ham sandwich" between the two sides, but it is clear the OAS shed no tears over the rebels' loss of Radio Santo Domingo. In fact, the official Brazilian history repudiates Palmer by recording that Latin units actually assisted the Dominicans during the operation. Thus, this "thorn in the side" was removed by proxy.

According to the Marines, the "political restrictions placed military commanders at a tactical disadvantage." The clear mission of evacuating US and foreign national civilians became clouded by later US and OAS policies. Military commanders "had to be prepared to take on the trappings of a diplomat." Instead of the clear military solution they thought could be achieved, generals were placed in the uncomfortable situation of mediating a national dispute alongside career politicians and diplomats.

The rush to de-fuse the crisis and quickly reduce military forces led to withdrawals based on numbers of servicemen, not units. This caused a breakdown in "tactical unity" among deployed Marines and left "major units in a disorganized state, unavailable for deployment elsewhere if the need arose." The decision was taken without consultation with the military and once made, was apparently too sensitive to change.

Despite the problems such political oversight and restrictions caused and the apparent incongruity between military and political aims in the Dominican Republic, General Palmer realized a political solution would be in the US's best interests. He worked diligently to mold force application to the diplomats' designs. While he continued trying to shape political decisions to military realities, he remained the soldier-statesman, seldom losing sight of the political nature of the crisis.

If political objectives are foremost in peace enforcement operations, did individuals at all levels of command understand the political objectives and corresponding restrictions? Was such understanding at the lower levels of the force really necessary? According to Gen Palmer, the political "facts of life" must be clear to everyone from commanding general to the private manning the roadblock if a peace operation was to be successful. In the Dominican crisis, political awareness began at the military's highest level. Gen Wheeler advised Palmer at the onset to "get close to the ambassador," adding "this is as much political as it is military." Palmer later said it was very good advice. Clearly, he saw the inevitability of Washington taking direct control of the military situation in the Dominican Republic. Still, he was never politically passive. On one occasion, the OAS wanted to dismiss conservative, senior Dominican Army officers–a demand from Col Caamaño (leader of the Constitutionalist forces.) Palmer told the OAS he would support such a dismissal if ordered, but warned that Gen Alvim and the Latin American Brigade would not and, if pushed, the IAPF would probably unravel. Ambassador Bunker accepted Palmer's advice, and the OAS dropped their support for Caamaño's demand.

Palmer felt his political awareness must be transmitted to his troops. In his oral history he talked at length of the importance of leveling with the soldiers. He said this "leveling" had to begin with the senior officers who should, in turn, educated their company grade officers and NCOs. He felt his troops "understood that Dominican situation probably better than Ambassador Bunker did." Political naiveté among leadership and staff could have extreme consequences. Palmer noted that if the staff and commanders were not kept politically informed, they would probably do "something dumb" or let the overall commander do "something dumb." He apparently spoke from experience.

Gen Alvim suffered a similar reluctance to submit totally to political oversight. His stubbornness eventually provoked a showdown. In early January 1966, during a coup attempt by right-wing Dominican military officers, Alvim refused to support the provisional Dominican government. At OAS Ambassador Bunker's insistence and "in light of his political unreliability," Alvim was quietly recalled by Brazil.

If political education for military forces was important, how did this education take place? In addition to the informal indoctrination by troop leaders to the NCOs and privates already mentioned, each soldier was issued a copy of President Johnson's speech justifying US intervention. They were also issued a fact sheet which outlined the main objectives of the operation: protection for and evacuation of US citizens and foreign nationals, initiation of humanitarian programs, and prevention of a Communist victory.

Palmer felt that in a peace enforcement mission, tactical leaders' initiative should be dramatically curtailed–a contrast to normal war-time operations. Instead of allowing local initiative, "the senior commander must translate political guidance into specific tactical guidance for subordinates." Palmer placed emphasis on the political awareness of his troops, but also required the guidance issued by commanders to reflect political objectives, leaving little room for misunderstanding.

Despite his emphasis on political awareness, all was not as rosy as Palmer may have thought. According to Yates' research, US soldiers in the Dominican Republic were, by their own admission, uneducated on political-military operations and confused by political control of military operations. In Yates' words, "Their only training came through painful experience." The difference between commanders' perception and the actual education received by the troops points to a need for more formalized political indoctrination. If such awareness was as important as Palmer asserted, it is hard to understand the haphazard and informal education process. Despite Palmer's own assertion that US forces understand the political environment, they remained largely ignorant.

Perhaps the largest political-military disconnect in the Dominican intervention was over the issue of who was the enemy and the extent of IAPF neutrality. In the first days of US involvement, President Johnson addressed the nation, and in his opening remarks, said, "Let me also make clear tonight that we support no single man or any single group of men in the DR [Dominican Republic]." The seeds of later confusion were sown when later in the same address, he went on to explain, "Our goal,...is to help prevent another Communist state in this hemisphere." Palmer admits in his oral history that in the beginning, his mission was to "support whatever government the ambassador could scrape together," but in later periodic reports he said US forces were engaged in stability operations whose goal was neither "to maintain the status quo,...nor to support any particular faction or political group." (Emphasis added.) In both Johnson and Palmer's statements the term "neutrality" seemed quite flexible.

The subject of when, if ever, true neutrality was enforced in dealings with the two factions remains cloudy. Palmer designated 16 May as the date from which neutrality was enforced. He cites the order on that date to "neutralize" the Dominican Navy and Air Force. Yates dates true even-handedness from the Red Cross negotiated peace-settlement of 21 May 1965. However, as late as 27 May, armed Dominican government soldiers were allowed to freely traverse OAS lines, while "Rebels" had no such freedom. Even on 30 May, a Marine Corps "enemy situation map" depicted only "rebel" territory. As the situation developed, US forces accepted less assistance from government troops, but throughout the operation, the Loyalist faction received more favorable treatment than the Constitutionalists.

In the first month of operations, neutrality was not even cosmetic. When the Constitutionalist "congress" elected Caamaño president, US officials ignored the event, continuing to seek a "more suitable government." Eventually Gen Imbert, US choice to lead the Government of National Reconciliation, was sworn in on 7 May. US forces specifically sought Dominican soldiers to help at roadblocks and checkpoints, and two platoons of Dominican Army soldiers were attached to US units. A National Police liaison officer, Capt Daniel Moore, was also attached to the 6th Marine Regiment for the extent of their deployment. One of Moore's prime services was guiding Marine movements to avoid rebel areas. In Operation Forward March, a movement to straighten US lines in the International Security Zone, US forces coordinated extensively with the Dominican Army and National Police.

While the policy of favoritism was officially endorsed through such actions, it was reinforced by informal factors. Inadequately briefed soldiers assumed their mission was to "kill Commies" and assumed the Communist-influenced Constitutionalists (rebels) were their enemy. The characterization of rebels as "Communist" was widespread and appears frequently in official diaries and logs. As early as 2 May, battalion commanders discussed the danger of "immediate threat from dissident elements...," painting the Constitutionalists as the enemy.

As General Palmer pointed out, this portrayal of the leftists as the "enemy" was encouraged by Constitutionalist aggression. As he phrased it, "...as far as our troops were concerned, one side was shooting at them, and the other wasn't. And the guys that were shooting at them–it was hard to tell the troops that they weren't 'enemy.'" Later rebel actions reinforced such views. They were discovered using ambulances to transport arms and troops, sometimes concealing them in coffins. Shoe shine boys were found selling poisoned food and soda and hiding weapons in their boxes.

Latin American IAPF forces shared this perception. Brazilian documents talk of forming up lines opposite the Constitutionalist area–termed "Rebel Zones"–from which nightly mortar attacks would rain down on their positions. Col Meira-Mattos describes how Constitutionalists would cross the lines during the day as civilians, reconnoiter the Latin positions, then attack that night. But, considering that the Latin American Brigade was reinforced by two Dominican Army tank platoons , it is understandable that the Constitutionalists would fail to acknowledge OAS "neutrality."

The identification of the leftist, Constitutionalist faction as the "enemy" and the Dominican government (in any of several incarnations) as the "friendly side" may have been inevitable in the Cold War 1960's. Whether the Rebels attacked OAS troops because they sided with the opposing faction, or whether the OAS troops sided with the government because of the Rebel attacks is unclear. What is clear is that real peace was not possible until the OAS took a more neutral stand.

As in the Dominican Republic intervention, political considerations governed Somalia peace enforcement operations. Unlike the Dominican Republic, the ROE in Somalia, reflected political objectives without running counter to operational requirements or leading to avoidable casualties. The basic principle of the Somalia ROE was that a soldier could fire to defend himself if threatened. Unlike the Dominican Republic, forces did not have to be fired upon before they could defend themselves. This principle remained in effect throughout the mission and was later accepted by UNOSOM II.

The ROE were formulated at USCENTCOM headquarters with recommendations from the 10th Mountain Division and I MEF judge advocates. With additional input from the command's State Department representative, CENTCOM's judge advocate completed the ROE, and after a series of iterations between CENTCOM and the Office of the Joint Chiefs of Staff (OJCS), they were approved. After approval, incremental adjustments in the implementation were made based on inputs from the field commander, Gen Johnston, and UNITAF's judge advocate, Col Lorenz. Learning from the 1983 Lebanon operations, the military were proactive in defining the ROE for Somalia, then getting them blessed by higher authority, rather than waiting to have ROE imposed.

Almost every nation supplying forces to UNITAF adopted the US ROE. Most were pleased with the less stringent ROE. In particular, the French and Canadians, who had suffered under highly restrictive ROE in Bosnia and Cambodia, were relieved by their freedom under UNITAF rules. Even though the coalition used the same ROE, force contingents had various interpretations of the rules. These were usually minor differences driven by national policy or training level. To standardize, the JTF headquarters employed a "coalition force coordination cell," to coordinate ROE and commander's intent among the force contributors. UNITAF also established a briefing, given by the staff attorney in cooperation with the liaison cell, to resolve any interpretation difficulties. To ensure the rules were clear and understood by the average soldier, they were reduced by Col Lorenz to simple language, put on cards, and passed through the various chains of command to the lowest level. (See Figure 4.) For continued adherence and to ensure new arrivals were informed, UNITAF scheduled periodic ROE refresher classes.

The ROE process, though well thought-out, was not without problems. The biggest shortcoming cited by those I interviewed was the lack of guidance on implementing non-lethal force. The rules were clear about deadly force, but if an individual soldier decided such force was unwarranted, he had few officially-sanctioned tools to implement "the minimum force necessary." A number of informal solutions were developed to fill this void. One was a large tent stake–sometimes referred to as a "Somali-be-good-stick"–used as a make-shift billy-club. The "sticks" served well since the normal tool for internal discipline among Somalis is a large, ornate stick wielded freely by a tribal elder. In situations where a bullet was too much, tent stakes often served well. Tent stakes were also affixed, pointing outward, to vehicles as make-shift pikes to ward off attackers in a style reminiscent of the "Road Warrior" movie vehicles. Perhaps the best known non-lethal force tool was cayenne pepper spray used in lieu of tear gas for riot control. However, the cayenne spray was not available until well into the operation.

Incidentally, the dilemma over applying non-lethal force was most acute among Western contingents. Most African and Arab contingents quickly arrived at a "proportional force" policy. When Moroccan forces were attacked by rock-throwing teenagers, they simply began throwing rocks back at the assailants. The "children" soon learned not to attack the Moroccan vehicles, though they continued to attack US vehicles. Eventually, MGen Arnold, ARFOR commander did an ROE review covering how and when to apply lethal and non-lethal force and prescribing the use of Somali elders in dealing with attacks whenever possible. Non-deadly force application was taught using anecdotal evidence and hypothetical scenarios. This additional training helped significantly in day-to-day work among the people.

The other major issue with ROE lay not in the rules, but in the ability of the contingents to execute them. The UNITAF forces were, by and large, good soldiers with highly developed combat skills. These soldiers often restricted themselves, refusing to fire even when the ROE permitted it because they were unsure of their target or unconvinced that deadly force was necessary. Members of the Joint History Team attributed this restraint to both "good soldier skills" and high ethical standards.

As UN-sanctioned UNITAF gave way to UN-controlled UNOSOM II, the official ROE did not change, but the quality of the troops who replaced their countries' "elite" units were, on average, less well trained and disciplined. For instance the Pakistani 6th Battalion, Punjab Regiment, highly respected for their competence, was later augmented by other units in the Pakistani brigade which were inferior, even by the assessment of the Pakistanis. UNOSOM II troops were supposed to provide security for the continuing relief effort–reinforced by the US Army QRF. However, most UNOSOM II forces were oriented more toward logistics and combat support. When the mission turned into offensive operations in June, these forces were much less able to operate effectively in a combat environment and to execute the ROE with the same reserve and accuracy as the UNITAF forces.

Even before the shift to UNOSOM II, ROE implementation was being gradually tightened. The rules themselves remained constant, but they were interpreted with increasing strictness and were tailored to specific locations. For instance, in Mogadishu, the forces began using highly trained snipers as a kind of direct fire support. If attacked or in danger, patrols would often relay the situation to over-watching snipers who could apply force more accurately than the patrol. UNOSOM II forces were unequipped or psychologically unprepared to implement this type of sophisticated reaction. The tragedy of 5 June when 24 Pakistani soldiers were killed in ambush was not the fault of over-restrictive ROE, but of faulty intelligence and poor training.

Despite being more strictly interpreted, at least through the end of UNITAF, the ROE were flexible and appropriate to the mission. Unlike the Dominican intervention and many other peace operations, UNITAF provided a well-defined feedback loop so ROE could be adjusted or reinterpreted in a timely fashion, based on the local commander's needs. Staff judge advocates (SJA) with special ROE training walked the streets of Mogadishu, and the lawyer became a very real and highly valued staff member. The combination of a tight ROE feedback loop back to CENTCOM and the Joint Chiefs of Staff, and the on-the-street work of the SJAs kept the rules realistic. All in all, the rules of engagement were one of the bright spots of Operation Restore Hope.

While the ROE process was much improved, a perennial problem of peace operations remained for both UNITAF and UNOSOM II–the problem of prisoners. Again, this topic is two-sided including both the prisoners captured by UNITAF and UNOSOM forces and coalition troops captured by the Somalis. I will begin with the far more common dilemma of Somalis captured by coalition forces.

Somalia was not the site of and international armed conflict, and it had not even the semblance of a civilian government. Therefore, neither the Geneva conventions nor civil law could be applied. Some of the resulting problems have been mentioned. How do you tell combatants from criminals? Should a Somali captured while attacking a patrol be treated differently from a Somali caught breaking into a warehouse? These questions, though addressed in the legal annexes, were never really resolved and continued to plague the soldiers and Marines in Somalia.

There is some indication that detainees were planned for by the lower echelons, but overall implementation was inconsistent. In the absence of actual war, all prisoners had to be treated as criminals. Still, an informal distinction was made between criminals who would have been EPW's in a war and those who would have been civil criminals. ARFOR headquarters established a detainee compound, but it was limited to handling those who had fired on peace enforcement troops or their charges. Without civil police, no option was available for pure criminals. When possible, coalition forces turned them over to clan elders, but this practice was spotty and normally not viable. In most cases, a Somali who had committed a crime such as theft against a protected HRO compound or military storage area were held for a few hours and released. The futility of arresting Somalis soon became obvious.

In cases where Somali on Somali crime was witnessed, the forces had even fewer options. Before the Somali police force was reestablished, military forces had no choice but to either ignore the violence, or after stopping a crime, release the assailant. In one case an Army team was able to stop a gang-rape committed in plain view, but had nowhere to take the offenders. Over and over through my research, those I interviewed raised the same issue: they had no method for dealing with criminals.

After the process began to re-establish local police forces–not yet complete at the time of this writing–the troops had some outlet. Still, re-constituting the police merely postponed the problem since no judicial or penal systems existed. In some cases the frustration of local commanders led them to skirt the edge of international law in establishing local police forces and judicial systems. Marine forces in Baidoa established an auxiliary police force to deal with local criminals. For legal reasons, the force was termed "auxiliary police" or "security force" since US forces are forbidden from establishing, training, or materially aiding foreign police forces. A similar effort was successful in Baderra using former Somali police officers. Despite isolated success stories, neither the US military, nor its coalition partners were in any position to reinstitute the native rule of law.

In fact, US forces and some western allies were nearly alone in detaining prisoners. Many contingents, especially those from less developed areas, saw immediate punishment as more productive than incarceration. This treatment was familiar to Somalis who had previously dealt with crime in much the same way. After a while, criminal acts tended to decrease in the areas controlled by these contingents and increase in the areas controlled by the US and other, more restrained contingents. In the long run, almost all criminals (including those who attacked coalition forces or HROs) were released to commit further acts of violence. The question of how to handle detainees in an intrastate peace enforcement situation remains unanswered.

For contingent members captured by Somalis, the issue was not much clearer. Though seldom discussed publicly, the capture of peace enforcers was considered and planned for at the operational level. According to Lt Col Eaton, ARFOR G-3, POWs were prevented during UNITAF through extraordinary measures. No helicopters or vehicles were allowed to travel alone, and a reaction platoon or company was always on-call. Wherever possible, escort was increased and exposure decreased. Nevertheless, the subject of what to do once a coalition member was captured was not discussed. According to Col Anderson, "The military leadership was warned [prisoners or catastrophic casualties] would happen, but neither the State Department nor military leadership wanted to deal with it." Beyond taking measures to prevent such an occurrence, no further guidance was issued. Lt Col Martinez, ground component commander of the UNOSOM II Quick Reaction Force (QRF) and Lt Col Borchini, 8th Psyops Battalion commander, agreed that preparation for POWs broke down at the tactical level. Once prisoners were taken–including the celebrated capture of Army Chief Warrant Officer Michael Durant–the mission became "very emotional." After the casualties suffered on June 5th, most contingents drew back into a "fortress stance," effectively returning the streets of Mogadishu to the warlords.

All those interviewed for this study agreed the capture of a US or coalition soldier was inevitable. However, neither the US nor the UN publicly discussed the possibility, and this silence only magnified the public shock and outcry once the inevitable happened. Ambassador Borg-Olivier agreed that when United Nations forces suffer POWs, the UN and troop contributors must share the responsibility. Still, no clear-cut solution exists, and the issue is aggravated by the fact that the belligerents are not party to, aware of, or even care about international obligations under treaties or agreements on the treatment of prisoners.

Perhaps the UN should have taken more aggressive action to prevent the ambushes and captures, but in any significant conflict, prisoners must be anticipated. One thing that can be done is to prepare the public for that likelihood and assure military coalition members of a professional and uniform response to the capture of a coalition soldier. Of course, publicizing the possibility of casualties and prisoners may squelch public support for an operation, but if public support is marginal, the contingent should reconsider involvement.

Another impact on coalition operations in Somalia was the political constraint troop providers placed on their forces. To a much greater extent than in the Dominican Republic, national leaders put limits on the places a force could be used and the types of missions it could perform. Such constraints should be expected; multinational forces must be employed to maximize their utility and minimize the overall impact of these restrictions.

In general, every member of the coalition force negotiated either mission or geography and often both. The coalition nations fell loosely into three groups. The first gave general guidance and broad latitude to their local commanders and included the Canadians and Australians. These contingents were practically unlimited in terms of the actions they could perform. The second group included forces who required very little support, but were somewhat restricted by their national command authority either geographically or operationally. This group included the Pakistani and Moroccan forces. The final group included both large and small contingents who were very constrained by either the type of mission they could accept, their area of responsibility, or both. This group included most Arab contingents, the Italians, Germans, and Belgians.

Many of the smaller contingents were willing to go beyond their initial mandate, and if asked, would petition their governments for wider latitude and a bigger role. This was particularly true of the African contingents who were aware they were establishing their international credentials. On the other extreme, some national contingents did as little as possible and just had to be "worked around."

Among countries with specific geographic requirements, the Belgians refused to work anywhere other than in Kismayo where Belgian HRO relief efforts were concentrated. The Italians insisted on northern Mogadishu, even though this split the city in an inconvenient way. Small contingents had geographic requirements as well. Most wanted to stay on the airfield away from the more dangerous missions and near their logistics. MGen Arnold took pains to encourage and praise these units in order to draw them out and give them more ambitious roles. Often the ARFOR G-3 had to work directly with a small contingent's operations officer to personally "set things up," so they could move off the airfield.

While this method worked for contingents with some degree of local authority, others with less authority, or who were uninterested in broadening their mandate, simply stayed put. In some cases, the JTF commander was able to accommodate the contingent's wishes, and the overall UNITAF mission suffered little. Belgian forces, for example, stayed in Kismayo, but were quite effective in getting the aid through. However, when the nation building expanded under UNOSOM II, these contingents were often unwilling to move out to fully occupy their assigned humanitarian relief sector (HRS). Again, the Belgian contingent is a good example. While doing an excellent job within the city, they practically ignored the surrounding countryside.

Compounding the geographic constraints were mission restrictions borne out of a need to reduce or eliminate risk. Perhaps the most extreme example of this was the German contingent which could not accept any possibility of casualties. According to the Joint History Team, this "humiliated" the German commander. The United States was also very "risk conscious" and frequently canceled missions if, during the planning phase, it did not appear to be a sure success. Often coalitions were more afraid of failure than of casualties.

After the UNOSOM II mission turned offensive, many nations' risk tolerance level dropped. In particular, the Pakistani, Moroccan, and Italian contingents drew back into extreme defensive postures. Col Anderson attributed this to the use of peacekeeping troops for a peace enforcement mission. When UNOSOM II took over, most combat troops were replaced by peacekeepers and logisticians. These troops, equipped primarily for humanitarian missions, were suddenly thrust into the role of combat troops. Many nations simply refused to participate. Some UN military advisors attributed this problem to the varying interpretations of UN Resolution 814 which authorized the rebuilding of Somalia as a nation. While the US saw the need to "marginalize" the warlords, the Italians saw a strictly humanitarian UNOSOM II mission. The purposefully vague resolution caused a failure in translating UN (strategic) guidance into operational and tactical guidance.

Ethnic and religious factors also limited coalition members. Initially UNITAF officials were concerned about the placement of Italians in Mogadishu because of their colonial past there. Rather than causing a conflict as feared, the Italians instead appeared to become friendly with the Ali Mahdi faction. Their seeming favoritism had the potential for polarizing the main factions, and Johnston was obliged to consider this in the plan to stabilize Mogadishu.

Arab members were sometimes constrained by both culture and history. While the UAE saw the Somali mission as a historical obligation, the Egyptian history of disharmony in Somalia led them to severely limit their operations. This was compounded by the fact that Boutros-Ghali was Egyptian and unpopular with the Somali people. Ramadan also affected Arab peace operations since Moslems from the Arabian peninsula refrain from eating and drinking (even water) during the day–a big limitation in an area like Somalia. While no single limitation was a "show-stopper," they all had to be thought through and accounted for.

The Somali people had a strong sense of cultural and racial superiority over all other African and many non-African, developing countries. They would tolerate white Americans or Western Europeans whom they considered near-equals, but balked at the idea of Black-African troops patrolling their towns or distributing relief supplies. In other areas the Somalis were equally resentful of former-colonial powers like France or Italy. This combined to make HRS allocation quite a challenge.

Meeting the combination of contingent requirements and constraints was very challenging for UNITAF, but within the limited mission, inspired leaders like Johnston and Arnold proved equal to the task. Some countries were used to their maximum capabilities while others were simply worked around. Often various contingents even competed to see who could do the most as a way to boost their national image. However, when UNOSOM II took over, their much larger mandate, reduced forces, national constraints, and lack of unity took a tremendous toll. Eventually the broadened, more militant mission caused Gen Bir, in the words of one battalion commander, "to lose complete control for a while."

In all coalition operations national command authorities restrict their forces. If the command has a narrow mission and sufficient unity of purpose the contingents can work within their limits and still accomplish the mission. However, if as in UNOSOM II, the operation suffers under divergent command authority, a greatly enlarged mandate, and a simultaneous reduction in forces, both in quality and quantity, failure seems almost inevitable.

In Somalia, the UNITAF commander saw both his political and military objectives to be very narrow. According to Walter Clarke, State Department deputy to the US Special Envoy, General Johnston did not want to see that, after the first few weeks, his limited goal was untenable. In Clarke's words, Johnston refused to accept his implied authority in a nation without a government; the narrow political blinders Johnston and his senior military authorities wore doomed UNOSOM II from the very beginning. While an extreme view of the commanders' political myopia, Mr. Clarke is by no means alone. Conversely, others saw Johnston as extremely politically aware, careful to maintain the exact scope envisioned by President Bush when he committed US forces in November 1992. Regardless of perspective, all agree Johnston had a political vision for success in Somalia which matched his military strategy. But what of the personnel below him? Did they have political insight and an overall understanding of the mission in Somalia? Is such knowledge necessary?

As Col Nordbø of the Norwegian UN delegation put it, "One little [political mistake] can ruin the whole thing." To minimize these mistakes, he maintained even privates should understand the political situation since they will make many critical, politically-charged decisions. Orders can't be given to cover all circumstances, and time will slip away while senior officers are brought in. Other advisors had mixed responses, but all supported the idea that at least battalion commanders and above should be well versed in both political and military objectives. MGen Zinni agreed political education is extremely important for all field grade officers, but that only cultural awareness and an understanding of the military mission should be emphasized at lower levels. Tempering this still further, Col Gray of the New Zealand delegation cautioned that "impartiality can be a victim of political knowledge."

Despite the opinions of Generals Zinni and some UN military advisors that political decisions are limited to field grade and senior officers, choices requiring an understanding of the overall political objectives were often made by junior officers in Somalia. In Afgoye, an MP company commander was given the task of "cleaning up the town." Beyond running the bandits off the streets, he had to identify responsible citizens, form local government councils, and restart the town's civil organization. This captain served both as mayor's advisor and civil service coordinator until local authorities could take over. Because of his talent and overall understanding of the political dimension of his actions, he was successful.

Some insist that a basic understanding of the political goals should go all the way to the "PFC-level." Citing the old adage that all military men are "ambassadors in uniform," these officers remind us that knowledge is a force multiplier. A soldier who understands the objectives and the environment doesn't have to be continually motivated. Among the soldiers in the cities there was a periodic, ground-swell chorus of "why are we here?" Few Somalis within the cities were openly starving, and many soldiers began to be confused about the deployment's overall purpose. Thinking commanders rotated soldiers out into the countryside to see the starvation and the differences made by the feeding stations and relief convoys. After seeing the end-result of their work, they better understood the "big picture."

Although many officers and enlisted men had some grasp of the political situation and overall objectives, most officials and military leaders simply "understood that they didn't understand" the political dimension. Dr. Bob Wright of the Joint Historical Team interviewed dozens of officers and diplomats in Somalia from Ambassador Oakly and Gen Johnston to junior officers and NCOs. He insists that most officers felt they knew neither the "end-state" nor their purpose in the grand scheme.

On another level, many officers were ignorant of the politics of coalition operations. Many mid-grade and senior US officers were ignorant of the position and power some of their foreign counterparts wielded. When they saw a Turkish or Moroccan major, they were likely to treat him with the same deference they would an American of equal rank–which is to say very little. Most did not understand that foreign commanders were hand-picked, highly educated, direct representatives of their governments. These "mere" majors and lieutenant colonels often reported nightly to their national command authorities. The implications for political and diplomatic relations are sobering and vividly demonstrate the urgent need for political-military advisors to the commander.

Do soldiers in a multi-national peace enforcement need to understand the overall mission? I believe they do. They may not need the knowledge of an ambassador or JTF commander, but should know the general political and military objectives and his place in the overall plan. Cultural, political, and mission training are all essential to a successful operation. The soldier really is "an ambassador in uniform."

The most controversial decision of our Somalia experience was our decision to abandon neutrality and pursue a specific enemy. Unlike the Dominican Republic where the coalition began the operation very partisan in outlook before moving toward even-handedness, the UNOSOM II force abandoned UNITAF's impartial stance to pursue a single faction.

A neutral position does not imply pacifism. Under UNITAF, the forces were impartial, yet sided against lawlessness and banditry. As one officer, put it, "everyone was a bad guy." It was not personalized. A good example was the US and Belgian response to the violence between rival warlords, "Col" Jess and "Gen" Morgan. UNITAF forces went to extremes to be even-handed, yet forced a cease-fire. Senior officers and diplomats worked hard to keep open a dialogue with all parties. Though UNITAF tried to stop the fighting, they did so without eliminating contenders for power.

Once UNOSOM II took over, though, their charter to rebuild Somalia required a stable Somali government. Mohammed Aideed, a key contender for national leadership, was identified by Admiral Howe, as part of the problem, not part of the solution, and backed by the US and UN, Howe decided to "marginalize Aideed." Beginning as an effort to reduce his influence, it ended in an all-out attempt to capture him and his senior leadership.

The first solid step toward personalizing the conflict and "marginalizing" him was the attack on Aideed's radio station. Although the station had been spewing forth a steady torrent of anti-American and anti-UN propaganda, it had been successfully countered through a US counter-propaganda campaign. Seeing a peaceful approach as more credible, UNITAF and Ambassador Oakly took the stand that it was Aideed's right to broadcast whatever he wished. The US simply fought him with the truth on their own radio station. Their highly successful approach included positive stories using extensive interviews with local Somalis to counter Aideed's lies, often using members of Aideed's own clan to refute him.

After officially targeting Aideed, though, the war quickly escalated. In addition to Howe's focus on Aideed, UN Secretary General Boutros-Ghali continued to make sweeping statements about completely disarming the country. This threatened every warlord, but put special stress on Aideed because of his already high visibility in Mogadishu. Another factor accelerating the rush to open war was the loss of corporate knowledge and experience after the change of command to UNOSOM II and from Ambassador Oakly to Admiral Howe. The final catalyst for the open war against Aideed was his 5 June ambush of the Pakistani soldiers, which pushed UNOSOM II irrevocably into bloody confrontation.

But, according to MGen Zinni, the die was cast much earlier. Our attempt to impose a total solution required a clear enemy. When faced with an abstract enemy like lawlessness or brutality, Americans tend to take sides or create a visible enemy. While UNITAF could afford to remain distant, UNOSOM II's ambitious mandate made this approach much more difficult. Was it inevitable? Zinni says yes. Our desire to impose a US/UN solution based on Jeffersonian democracy led UNOSOM II to be too ambitious and dictatorial in its "solution." While in Somalia, "half a glass may be the best we can hope for" our focus on ideal democracy pushed us to reject an unsatisfying Somali solution and ultimately led to failure.

In hindsight, almost everyone sees singling out Aideed as our enemy to be an overall liability. In the words of end-state planning authority, Col Bruce Clarke, "Even in Chapter 7 ops,....you must maintain neutrality if you are to avoid becoming involved on the side of a given faction and remain effective." Col Clarke adds that the fatally flawed, limited vision of UNITAF left UNISOM II in a position where neutrality was much more difficult. Col Toralv Nordbø said it more directly. "UNOSOM II made two big mistakes. The first was to go after Aideed. The second was not to get him."

In both the Dominican Republic and Somalia, political considerations affected almost every decision and shaped each military movement. Rules of engagement were critical in both missions. In the Dominican Republic, the imposed, extremely restrictive ROE cost lives, but by the Somalia the process had greatly improved. The ROE were bottom-up rather than top-down, and although interpretation tightened, the rules themselves were constant. FM 100-23 warns that ROE may not provide detailed guidance; commanders must exercise judgment. In the Dominican crisis, this judgment was intentionally removed. In Somalia, the latitude not only to commanders, but to junior enlisted men, was restored.

Even with the wider latitude in Somalia, a few major ROE problems were noted. One involved non-lethal force. Planners must devote more attention to proper response when a bullet is too much. Several UN military advisors warned of the US tendency to see things as "black and white." We must, they insisted, learn to recognize the various shades of gray in peace operations. According to them, ROE should be "broad policy guidance" which should guide, but not define "orders for opening fire."

A second problem involved inability by some to carry out the ROE. An obvious, but overarching principle is that ROE must be interpreted and implemented uniformly among contingents. This was true neither in the Dominican Republic nor in Somalia. Although each country agreed to the same rules, skill levels and cultural background led contingents to implement them in widely different manners. ONUSAL veteran, Col Colmenares, insists the "highest standards for selecting troop commanders at all levels and continuity in employment policies" (emphasis in original) are essential to peace operations ROE. While FM 100-23's caution that multilateral force contingents will likely operate with different ROE may be exaggerated, ROE will certainly be interpreted differently. Combined exercises, scenario-based education, and high-quality forces are the only hedges against this problem.

Prisoners remain a volatile and difficult issue. The international legal system has few provisions for "detainees" or prisoners in a peace enforcement action. NORDIC UN training manuals prescribe the quick release without prejudice of any prisoners taken. While seemingly foolish or counter-productive, forces in the Dominican Republic and Somalia were compelled to similar practices. In the Dominican Republic, the local police were party to the conflict, and in Somalia there were no civil authorities.

One possible answer to this issue is training and supervision for civil police by some respected international agency as part of any peace operation where national police are either unprepared or undependable. Efforts to rebuild the police system in Panama after Operation Just Cause and in El Salvador may provide some guidance. However, in any such effort, we must remember that it can not be done overnight. Such efforts are time and labor intensive.

The jury is still out on the extent of formal political training needed for peace enforcers. Most agree that formal training for mid-grade and senior officers is highly desirable, and many propose orientation and mission-specific training for junior officers and enlisted men. At the highest levels, FM 100-23 recommends a Combatant Commander's Advisory Committee to link theater strategy to national policy goals and tie these to the specific objectives of the State Department and individual ambassadors. During UNITAF, Ambassadors Oakly and Albright and General Johnston seemed to have a very tight and amicable relationship. However, during the transition to UNOSOM II and afterward, the diplomatic and military coordination deteriorated dramatically.

Even among low-ranking personnel, evidence shows that frequently the roles of peace enforcer and diplomat merge. From a Marine lieutenant who defuses a riot in Santo Domingo to an MP Captain who sets up a local government in Afgoye, the need for diplomatic and political awareness is evident. Although "good soldier skills" were cited as most critical, the needs don't stop there. At the very least, pre-deployment training should stress overall understanding of the mission, and emphasize negotiation and mediation skills. PME courses should reinforce national strategy and cover the interrelationships among national and international bodies and these agencies' interfaces with the military. In the future, we should not deploy peace enforcer