While the execution of the Haiti operation was generally successful, interagency planning was described by participants as slow and disjointed, and, until May 1994, lacked clear political guidance. While civilian agencies were developing a "comprehensive political-military plan," major players continued to disagree on the goals until the final weeks prior to launching the mission. Without appropriated resources, agencies could only contemplate what they would do. This process and security restrictions resulted in considerable frustration at the operational level where military planners found themselves preparing for civil-military operations without being able to talk with their civilian counterparts. It further contributed to initial confusion on the ground in Haiti when the military units expected to find civilian agencies ready to begin operations from the outset, and civilian agencies found their familiar Washington-based military counterparts deployed without leaving clear points of contact for further coordination. Table 4 summarizes the interagency planning problems identified in the workshop.
Workshop participants recognized that the dramatic changes in U.S. Haiti policy over nearly eleven months were the primary cause for the confusion. The U.S. approach to Haiti changed from July to October 1993 when a small peace-keeping force was thought to be sufficient, to the decision, in May 1994, to prepare for the use of force, and, finally, to the decision to actually use force, which emerged between July and August 1994. Between October 1993 and May 1994, however, there was no agreement on whether military force would be used at all, although, according to one workshop participant, "in retrospect, we were probably committed to use force in January [1994, when General Cedras rejected the U.S. ultimatum to depart before sanctions were toughened]." Nevertheless, security required to keep the military planning secret constrained operators who needed to deal with each other.
Workshop participants also felt that inter-agency differences in approaching the Haiti question contributed to planning confusion. Publicly, U.S. policy was to use diplomatic means to accomplish the departure of the Haitian military and the return of Aristide. Within the Administration, the NSC favored use of force, while DoD remained negative. USACOM began planning for a military contingency while DoD's civilian leader-ship remained in denial.
Interagency planning was frustrated until May 1994, when President Clinton announced the Administration's policy change to consider the use of force. USACOM's planning until that time had been tightly compartmentalized and confined to the military operation. USACOM planners knew that they needed to coordinate with civilian agencies, but they were precluded from doing so by security concerns. The 18th ABN Corps' planning for an opposed entry (Plan 2370) began in May, while the permissive entry plan (Plan 2380) began development in July at 10th Mountain Division at Ft. Drum, New York. Compartmentalization of these two planning processes was, according to one senior workshop participant, "carried on far too long - until the final weeks. This led to pointless delays," and contributed to incomplete interagency coordination. Players in different agencies knew that they should be coordinating. "Toward the end of the summer, when it looked like a go for the military option, planning to integrate the military and civilian factions started." One senior player noted: "We tried to do the inter- agency coordination on September 12, but it was a disaster." There were too many people for real candor. "People just recited what they were doing." A senior military officer expressed alarm, reportedly observing, "This is the kind of planning that gets people killed."
Several other factors besides the evolving policy definition contributed to planning confusion. The workshop identified a serious failure of communication between strategic, operational, and tactical levels of the operations. Strategic planning took place in the Executive Committee (Ex-Com) which began operation in May under NSC leader-ship and included the Departments of State, Defense, Justice, and Treasury, the CIA, and AID. USACOM, which was doing the operational planning, was usually represented by the Joint Staff in these meetings, despite the fact that USACOM and The Joint Staff were perceived by some to be engaged in a tug of war over who would lead the process. USACOM usually got its reports of what happened at Ex-Com meetings from the Joint Staff. Other participants on occasion had different understandings of what had happened and who was to do what.
Serious operational level planning by different agencies began in June when interagency working groups and JTF-level coordination among the military units were established under USACOM. OSD SOLIC (Special Operations/Low Intensity Conflict) had taken initial steps to establish a CMOC (Civil-Military Operations Center) earlier in the spring of 1994, responding to Secretary Perry's instruction to begin interagency planning, but had been frustrated by DoD's continuing reluctance to become involved in Haiti. AID's Office of Disaster Assistance (OFDA), the key link to NGOs and PVOs, began working with SOLIC during the summer and worked out plans for a major surge of delivery of food, fuel, medicines, and other relief supplies after the U.S. assumed control. AID's Office of Transition Initiatives began serious planning for its participation in May, after the President's policy announcement. While these different agencies were all planning their own participation in the Haiti operation, there was little operational level coordination between agencies. Until the 10th Mountain Division began developing the permissive entry plan in July, there was little coordination between military and civilian agencies. In particular, there was little interaction between Washington-based agencies and ongoing planning by the military outside Washington, although the OFDA team was at Ft. Drum for ten days before deployment. Interagency coordination at the tactical level did not take place until troops were on the ground in Haiti.
In sum, interagency dialogue was adequate at the strategic planning level, particularly once all relevant players were in attendance, but inter-agency discussions were not carried through to the operational level and linkages between the strategic and operational levels were deficient. While strategic planning took place under NSC leadership, concrete decisions were postponed to the last minute, so policy guidance could not be communicated effectively to the operational level commanders. The latter felt that they lacked the go-ahead to develop an integrated OpPlan with clear, attainable objectives, and adequate lead time to complete the planning coordination pro-cess. Moreover, there had never been an integrated civilian-military operational plan, and the civilian agencies were unfamiliar with the concept and with the idea of establishing precise lines of command and control and timelines for execution of projects. The first-ever attempt at a political-military operational plan undertaken by State, AID, Justice, Treasury, etc. during the late summer improved comprehension substantially but was a far cry from the clarity and rigor of military planning.
The operational planning disconnects caused follow-on problems. The military operation itself went smoothly, and the shift from opposed to permissive entry was accomplished with relative ease. Nevertheless, a number of complications occurred on the ground, largely because of the incomplete interagency coordination and other factors that included military and civilian organizations' mutual ignorance of counterpart culture and capabilities. Finally, the civilian agencies, for the most part, were unable to surge in capacity to meet the additional effort needed to plan and conduct the operation. Each of these issues is discussed in greater detail below.
Because there were two OpPlans, the logistic support to the operation was initially chaotic, and the integration of civilian agency support was delayed and cumbersome. For example, aircraft flow to Haiti required merging the TPFDD (Time Phased Force and Deployment Data) of the two operational plans. The merging was problematic and sometimes confused, according to workshop participants. Deploying the 10th Mountain Division aboard the carrier solved some of the problems, because its organic equipment accompanied the force. For others, it was more confusing. The Coast Guard was only included in Plan 2380. "We expected the 10th Mountain Division to take over early, but it didn't for about 30-45 days. We went in after only five days, and the 18th ABN Corps didn't know what we were doing. Our equipment seemed to have been thrown onto whatever ship happened by." The problems were sorted out after about ten days with the Coast Guard setting up its own logistics base in Guantanamo and using its own cutters to bring in supplies.
In another example, AID officials found that they could not get transport to Haiti. Their DoD points of contact deployed with the operation and left them with no ready Washington-based, working-level access to the DoD mobilization. Military transportation to Haiti for the civilian agency personnel was not available initially both because it was not planned and because the civilians did not know how to access the system. One workshop participant also noted that "with the current TPFDD and the way it flows, there are no slots to add interagency logistics."
A third set of problems resulted from failures of communication and coordination between civilian and military organizations responsible for rebuilding the Haiti infrastructure. Organizational and "cultural" differences between civilian and military organizations explained a lot of the problems, but the bottom line was that interagency operational-level coordination was incomplete in the preparation phase. For example, in Cap- Haitien, the land and water force commanders (10th Mountain Division and Coast Guard) coordinated responsibilities closely. "We had our tents pitched next to each other, but the third tent was missing - the civilian USAID tent." "There was no one to answer our questions about civilian assistance capabilities for 30 days into the operation." As commanders, we knew "we were going into a `fourth world' nation, but we didn't know the limits of our civilian agencies." "We were ignorant about what the other agencies were doing." Workshop members thought that a civilian agency capabilities presentation would have been helpful in the preparatory phase.
U.S. military planners were surprised that their civilian counterparts were not immediately ready with nation-building programs. Development planners were upset that the military refused to accept responsibility for civic action and nation-building efforts at the outset, although that policy had been determined at the strategic level. One workshop member noted that three assumptions underlined military planning for the Haiti operation, and none of them was correct. The first assumption was that lifting the embargo would result in an immediate inflow of money; the second was that NGOs and PVOs would immediately undertake a massive nation-building activity, and finally, that money would flow once the U.S. was on the ground. While these assumptions were clearly too optimistic, they reflect a view of the unfolding of the post-entry interagency process in which the military expected civilian agencies to respond to the operation as if they operated like the military itself - with a fully-planned implementation.
A senior workshop participant noted that a central problem of peace operations is that the U.S. government does not presently have a doctrine to integrate the military segment of an operation with the civilian assistance side. "Economic assistance operations are not real time because USAID contracts everything out." To begin the contracting process, AID requires both a final policy decision and budget authority, neither of which is often available until the last minute, or, often, until after the operation has begun. More-over, although they had operational staff on the ground in Haiti before the military operation, civilian agencies did not augment their in-country staffs adequately once the operation commenced. Unlike the military, civilian agencies do not have a reserve of personnel that can be sent into a crisis situation. In Haiti, they did not have enough personnel on the ground to coordinate effectively with the military command centers or with military units in the field.
Military-nongovernmental organization coordination incomplete
In addition to failure to understand the development agencies' lack of surge capabilities, the workshop recognized that the military and civilian private voluntary organizations (PVOs) have not learned to work together, particularly in developmental programs. Doing so is constrained, inter alia, by a real suspicion of one another. One military participant observed, "we were viewed with complete suspicion by the PVOs because they were concerned about compromising their neutrality. That point of view took us a while to understand. We finally made some progress at the grass roots level. We assisted them when they asked for something or told us what they needed." Military participants recognized that many more civil affairs personnel had been needed in Haiti, but in the post-Somalia environment, "this was hampered by DoD fears about doing nation-building." An exchange between an AID representative and a Special Forces representative highlighted the confused expectations that military and civilians had of each other. The AID representative clearly had expected Special Forces to fulfill civic action missions, but was corrected: "We didn't have a civic action mission." At the outset, Special Forces personnel were in Haiti "to keep order" was the feeling on the military side. AID did not recognize, until they arrived in country, that they would not be working with military counterparts on developmental activities.
However, there was much better cooperation on the relief side, with the military observing, coordinating, and, on occasion, assisting with the delivery of relief supplies in coordination with AID/OFDA. Moreover, U.S. Army Reserve civil affairs officers took over key roles in almost every Haitian ministry, cataloguing what assets were available, trying to get activities underway, and to some degree informally directing activities.
There were conflicting expectations on both sides because of incomplete coordination at the operational and tactical levels. Once the effective collapse of Haitian government institutions was apparent, Special Operations units acquired additional responsibilities of a civil affairs nature throughout the countryside, and for many months constituted almost the only civil administration. After elections, mayors gradually assumed some responsibility, and AID/OTI contractors and NGO/PVO personnel became active. Special Forces personnel remained until early 1996.
As many of the above examples indicate, the absence of detailed prior operational planning and coordination was offset in many respects by the initiative of the tactical level people on the ground, both military and civilian, who sought out and found their counterparts and conducted effective ad hoc coordination.
The workshop participants concluded that several other organizational and operational factors affected coordination across agencies. First, the military were actually the last players to arrive on scene, not the first, as they are accustomed to being. Civilian agencies, including UN advance teams, had been in-country for some time. Some of the PVOs had been established in Haiti for years.
The question of "who is in charge" emerged with the military looking for a clear chain of command. Because the military and civilians were not adept at working with one another, there was confusion about roles and responsibilities. Unity of command is a military axiom. When no one seems to be in charge the military will naturally take command. Throughout the planning process, the military, clearly a dependent variable in the broad political game, was frustrated by the absence of a clear decision-making hierarchy and by the delays in decision-making. This carried over into the field implementation.
In Haiti, the U.S. military operation was only one leg of a triad that included restoring democratic institutions and rebuilding the ravaged economy. While the military was essential in providing internal stability, it was largely irrelevant to the other activities, especially since forces were directed not to conduct nation-building activities. Military participants at the workshop lamented that "there was no one in charge of the over-all operation." They perceived a need for an operational level commander who would coordinate and direct all the agencies and forces involved. Some believed that this should be the Force Commander, others felt that it should be the Ambassador. Still others disagreed, citing the fact that the Ambassador is usually not staffed to handle such an operation, and in many cases there wouldn't be an Ambassador in-country (Haiti was considered an exception rather than the rule).
In Haiti, the Ambassador was in charge of all civilian operations while the Force Commander was in charge of all military operations. They coordinated closely, but did not establish a com-bined "war room" to provide follow up. This was in part due to the fact that State and AID missions were not augmented to handle the operation.
U.S. policy was to preserve the illusion of a Haitian government, and U.S. forces operated under the concept that they should support indigenous institutions, not supplant them. The U.S. did not enter Haiti as an occupation force. Some of the most serious challenges to this operational concept came in the area of providing local security. One set of problems occurred because of confusion over how to deal with the Haiti security forces during the 45-day transition period that resulted from the Carter negotiations. Pre-operation planning recognized that a police force would be needed_the details were coordi-nated in August_but the planning did not provide adequate follow-up for the security force. Neither Plan 2370 nor 2380 envisioned a period during which General Cedras and his cronies would remain in office after U.S. forces entered Haiti. Plan 2370 was intended to oust them and Plan 2380 assumed that they had left. The situation on the ground differed from both plans. In particular, the complete collapse of local security and police forces was not anticipated. A power vacuum ensued which led to a series of problems with the FAd'H, and the assumption of security responsi-bility by the U.S. One participant observed: "We need the capacity to provide policing. We don't have the doctrine for this. This issue was debated before the operation was launched and left to be resolved as the operation unfolded."
In Haiti, the FAd'H "owned" the police, so when the army was "taken down," the police disappeared as well. Aristide made it clear from the outset that he didn't want to work with a remnant FAd'H, but wanted a fully revamped police force. There wasn't time to accomplish this. The initial interim solution was to develop a 3500-man police force from the former FAd'H. This number later was reduced to 1500. The program for building a police force was a long-term (five-year) program which had begun in 1991 under ICITAP (the Department of Justice's Interna-tional Criminal Investigative Training Assistance Program), but which ceased operation after Aristide's overthrow. ICITAP had to start its police training program over, and it was reluctant to rush the process. The first plan was to build an interim force in 18 months. By February 1996, when the UN mandate ends and a new president is elected, plans are to have in place a 7000-man police force consisting of the 3000-man Interim Police Force drawn from former FAd'H and police (widely regarded as "worthless") and 4000 new trainees. To accomplish the latter will require doubling the number of police academy graduates and addi-tional funding from the U.S. While the remaining U.S. military under UN authority will pull out in February 1996, DFTs (Deployments for Training) and MTTs (Mobile Training Teams) may be able to augment the continuing international effort.
A second set of problems emerged over different concepts for neutralizing the Haitian army itself. The de-mobilized Haitian military were, potentially, a dangerous and disruptive force. Plan- ning how to handle them evolved through several stages. A review of the sequence of events that led to the dissolution of the FAd'H makes it clear that they received inconsistent and ambiguous messages. Some elements of the U.S. military were perceived to support the idea of retaining a minimal force army, a position that was contrary to that held by civilian agencies and Aristide, but which had not been clearly resolved in the pre-operation interagency discussions. Mixed mes-sages were sent, with approaches complicated by the 45-day transition. The initial plan involved employing some FAd'H as interim police, and later as the source of an interim, 1,500-man force envisioned as the follow-on to the FAd'H. The FAd'H's rapid disintegration made this plan impractical. By January 1995, it was obvious that the Haitians wanted the FAd'H dissolved (as Aristide had insisted all along, although he maintained publicly that it was a "constitutional matter"). In the end, a potentially volatile situation was controlled, and the FAd'H was "slow rolled" out of existence.
The decision to dissolve the FAd'H presented another complication. Without a new military organization, without acceptance as a police force, and without any other occupation, the former soldiers were likely to become bandits. The solution was to provide job training (including computer skills) as a way to break the culture and to give them some useful skills. But, as one skeptical workshop participant commented: "It remains to be seen if they come out as computer technicians or educated bandits."
The rapid collapse of the Haitian security forces and confusion among U.S. forces about the Haitian security situation contributed to the first month's confusion over the role of the U.S. forces in maintaining order in Haiti and the rules of engagement (ROE) under which they operated.
While the ROE for the Haiti operation were consistent among the forces involved, ROE inter-pretation clearly was not. Some of the differences can be explained by the different operational con- cepts developed in the two separate plans that evolved prior to entry into Haiti. Perceptions influenced interpretation of the ROE, as well. The 10th Mountain Division, last deployed in Somalia, was focused on protection of its own force from hostile fire. The U.S. Marine Corps forces in the Cap-Haitien region had an aggressive interpre-tation as evidenced by the firefight with the FAd'H on September 24th.
Under the assumption that U.S. forces were supporting the Haitian government, and, following previous experience in which occupation forces had used local police forces to maintain neigh-borhood security, the U.S. Army forces in Port-au- Prince were initially instructed not to become involved in Haitian law and order, and stood by while the FAd'H beat innocent civilians celebrating the arrival of the U.S. troops. The television cover- age of these events likely caused the rapid change in this posture. Workshop participants noted that no change in the ROE itself occurred, but a new interpretation of the original ROE emerged from the September 20 events.
The collapse of Haitian government authority during the 45-day transition period required that the U.S. assume many of the responsibilities of an occupation force, while preserving the surface facade of support to a non-existent Haitian govern-ment. In the countryside, the U.S. Special Forces operated with considerable autonomy appointing and supporting local officials like police, judges, and mayors. In the ports, the U.S. Coast Guard took charge, fixed the navigation aides, and ran the harbor operations. In one case, a corrupt official (the Port Director in Port-au-Prince) was summarily removed when he refused to stop his practice of demanding bribes to allow ships to enter the port.
Return to the NDU Home Page
Return to the INSS Home Page
Return to the ACTIS Home Page