A Review of the September 2005 Shooting Incident Involving
the Federal Bureau of Investigation and Filiberto Ojeda Ríos

August 2006
Office of the Inspector General

The OIG reviewed the FBI’s attempted arrest of Ojeda and found that the shot that killed Ojeda was fired in compliance with the DOJ Deadly Force policy. We found that the FBI delayed entering the house after Ojeda was shot for legitimate reasons of agent safety, not because the FBI wanted to allow Ojeda to bleed to death. However, our detailed review found several problems in the conduct of the FBI operation. For example, we concluded that the emergency daylight assault that preceded the exchange of gunfire was extremely dangerous to the agents and not the best choice available. Similarly, while we concluded that the FBI’s decisions regarding the entry of the residence were not improper, we determined that some of the decisions were based on inaccurate or incomplete information about what was happening at the scene. We also found deficiencies in the FBI’s preparations for the operation. For example, we determined that the FBI failed to adequately prepare for the possibility that negotiators would be needed. In this chapter, we make ten recommendations stemming from these and other aspects of the operation. Most of our recommendations are intended to highlight lessons that we believe should inform the planning and related training of future FBI operations. Other recommendations address specific policy or tactical issues. I.   Recommendations Regarding Compliance with the DOJ Deadly Force Policy and Related Issues Recommendation No. 1:   Conduct an inquiry relating to the three rounds fired by an unidentified FBI agent and the two unreported rounds fired by SA George. Several FBI agents reported that they perceived that several shots came from inside the house through the front door during the initial exchange of gunfire, a perception that contributed to their belief that there was more than one weapon being fired from inside the house. The Puerto Rico Institute of Forensic Sciences found, however, that the three bullet holes in the front door all were made by shots from outside the house. Because there were no bullets or bullet fragments found in the house that could be attributed to these particular holes and impacts, it was impossible to determine from the forensic evidence which of the FBI agents fired the three rounds through the front door. When we interviewed all of the agents who recalled firing their weapons, none reported having fired any rounds at or through the front door. Based upon the available forensic evidence and testimony, we determined the agents who we believe were in a position to have fired these shots. However, these agents declined to provide voluntary follow-up interviews to the OIG. Because we were unable to determine which agent fired shots through the front door or to determine whether the agent was targeting a particular threat in firing these shots, we could not conclude whether they were fired in compliance with the Deadly Force Policy. In addition, the Forensic Institute recovered outside the residence two .223 shells that did not match any of the weapons carried by the eight FBI agents who recalled firing during the operation. The Institute subsequently matched these shells to the weapon carried by SA George, who did not tell us in his interview that he had fired his weapon. The trajectories and impact points of these two rounds are unknown, although neither round struck Ojeda. Because George, through counsel, declined our request for a follow-up interview, we also do not know whether he fired these two rounds intentionally or how they were targeted. We recommend that after criminal investigations into this incident have been concluded, the FBI conduct an inquiry relating to the three rounds fired by an unidentified FBI agent through the front door of the residence and the two unreported rounds fired from SA George’s weapon, in order to determine whether these rounds were fired in compliance with the Deadly Force Policy. Recommendation No. 2:   Review the use of flash bangs in outdoor operations. We concluded that Ojeda opened fire on the FBI agents as they attempted to approach and enter the residence before any agents discharged their weapons at him or at the residence. The evidence did not support the allegation made by Ojeda’s wife in public speeches and media interviews subsequent to Ojeda’s death that the FBI fired first. We observed, however, that it was possible that Ojeda and his wife got the impression that the FBI opened fire first as the result of the detonation of a flash bang by one of the sniper-observers at the moment the FBI vehicle pulled up to the house. Our experts commented that using a flash bang outdoors to create a distraction is not nearly as effective as using it indoors, where the noise, pressure waves, and extremely bright light all work to the advantage of the arrest team. They also believed that using a flash bang outdoors creates a firecracker effect and risks alerting the subject rather than distracting or confusing him. We recommend that the FBI review the use of flash bangs in outdoor environments and under circumstances in which their use could have the unintended effect of alerting the subject or providing the mistaken impression that the FBI is opening fire on a subject before seeking his surrender. Recommendation No. 3:   Adopt a “standard load” procedure for HRT agents. In Chapter Five, we explained that there is no “standard load” required for HRT weapons that would have permitted the FBI to establish precisely how many shells had been fired from each weapon by determining how many magazines had been spent and how many rounds were left in the unspent magazines. As a result, the number of rounds fired by each agent had to be reconstructed by relying on the agent’s recollection and on the ability of the Puerto Rico Institute of Forensic Sciences to correctly match each spent .223 shell found at the scene to a particular HRT weapon. Although the forensic evidence was consistent with the agent statements in most significant respects, if “standard load” procedures were in place there would have been additional, reliable evidence regarding the number of rounds fired by each agent. We therefore recommend that HRT adopt a standard load procedure that would enable accurate post-incident accounting of the number of rounds fired by each agent, and that other components of the FBI adopt similar procedures to the extent they are not already in place. II.   Recommendations Regarding the Decision to Conduct an Emergency Daylight Assault Recommendation No. 4:   Adequately consider all available options in emergency situations as time permits. We examined whether other options were available to the FBI following the reported compromise of the sniper-observers near Ojeda’s residence, and whether the FBI commanders in Puerto Rico gave adequate consideration to alternatives other than a daylight emergency assault. We, along with our experts, concluded that a surround and call-out strategy was an available, superior option and that the FBI had sufficient information regarding the exterior of the residence to adopt it, or at least consider it seriously. We also reviewed the strategy of extracting the sniper-observer team as a possible alternative to an immediate course of action that posed foreseeable and significant risks to the agents. SAC Fraticelli told the OIG that he suggested a surround and call-out option when the compromise was reported by the sniper-observer team. However, HRT Deputy Commander Steve and his subordinates from the HRT’s Operations and Training Unit stated that they assumed, even after the reported compromise of the sniper-observers, that Fraticelli still would not consider this option because it might result in the barricaded subject situation he wanted to avoid. We also found that the extraction option was given only cursory consideration. HRT Deputy Commander Steve told the OIG that after the sniper-observers reported the compromise, no consideration was given to options other than a direct assault on the residence to arrest Ojeda. We were troubled by the lack of consideration given to alternative courses of action. Despite the change in circumstances that dramatically increased the risk associated with an assault of the residence – that is, assaulting the front of the residence in broad daylight instead of surreptitiously at night – the commanders chose not to deviate from an approach that was established early in the planning process for the arrest operation and that was based on conditions that no longer existed. We believe this case highlights the importance of adequately considering options in emergency situations, including revisiting options that may have been discarded in the initial plan. In addition, operational commanders should be prepared to consider how changed conditions affect the viability of their original plans in light of changed circumstances. Recommendation No. 5:   Enhance the Spanish language capability of HRT’s sniper-observer teams. In Chapter Six, we examined the circumstances of the reported compromise of the HRT sniper-observer team that lead to the decision to conduct a daylight emergency assault. We found it significant that SAC Fraticelli and other agents in the San Juan FBI told us that they now believe that the sniper-observers were not in fact compromised and that the people who were overheard conversing in Spanish near the sniper-observers were discussing matters unrelated to the FBI’s presence. Because none of the sniper-observers sent to the scene spoke Spanish, they were unable to understand the substance of the conversation that they overheard or to determine whether their presence was in fact compromised. In light of the prevalence of Spanish as the primary language for most residents in Hormigueros, including Ojeda and his wife, and the need to identify Ojeda, we believe the FBI should have anticipated that the sniper-observers might need to understand conversations in Spanish. It therefore would have been useful to have had some Spanish-speaking agents on the sniper-observer team. We recommend that HRT consider this need in future operations and enhance the Spanish language capability of its sniper-observer teams. III.   Recommendations Regarding the FBI’s Entry Decisions Recommendation No. 6:   Ensure that, if decision-making in critical incidents is assigned to FBI Headquarters, there is adequate communication between the field and Headquarters concerning the situation at the scene and the decisions that are made in Headquarters. In Chapter Seven, we highlighted several significant examples of the consequences associated with FBI Headquarters’ involvement in the incident after the emergency assault failed. With respect to Bald’s decision to require CTD approval for any entry decision, we found that this decision was based primarily on a perception that Fraticelli was “overwhelmed” by the situation. However, we determined that information reported to Bald that he said confirmed his concerns about Fraticelli’s command did not necessarily reflect what was actually happening at the scene. Bald told us that the report of an HRT agent getting close enough to Ojeda’s residence to look inside through a window suggested to him that HRT was making decisions independent of Fraticelli. But we concluded that this incident did not occur and noted that even if the reporting actually related to the limited breach of the residence, the activity was taken with the SAC’s approval. Bald was also told that the HRT Commander made a statement that HRT had the lead at the scene. This statement, which suggested a chain of command that violated FBI policy, reinforced Bald’s concerns because he knew that the chain of command placed CTD, through the SAC, in charge. Yet, we determined that the HRT Commander’s statement – made from Quantico, Virginia – did not reflect the situation at the scene in Puerto Rico, where HRT was cautiously preparing for a nighttime entry under the SAC’s authority. The consequences of FBI Headquarters’ involvement were also evident in the decision to delay entry until the next day. As, we explained in Chapter Seven, Headquarters was operating throughout the evening of September 23 under the assumption that there was a second armed subject in the house and on the belief that Ojeda may have been wounded but that his condition was essentially unknown. By contrast, the HRT agents at the scene who were closest to the action told us they were virtually certain as time passed that Ojeda had been killed or very seriously wounded. The agents’ evolving perception of the lessening of the threat was never communicated to officials at FBI Headquarters. FBI Headquarters’ involvement apparently also affected the clarity of communications between SAC Fraticelli and HRT Deputy Commander Steve concerning CTD’s rejection of the nighttime entry. As we described in Chapter Seven, while CTD’s final decision was conveyed to Fraticelli sometime before 9:00 p.m., the decision apparently was not shared with Steve at that point because HRT continued to draft a nighttime entry plan, and the HRT agents continued to prepare to enter the residence that evening. Steve only learned of CTD’s decision much later, when he and Fraticelli called DAD Lewis to try to persuade him that HRT should conduct a nighttime entry. When Fraticelli told Steve that Lewis said there would be no entry, the decision was conveyed for the first time to the HRT agents at the scene – at 11:33 p.m. We found this lack of communication between Fraticelli and Steve troubling and believe it was in part a consequence of adding another level of management to the operation. In our view, these examples demonstrate how the quality of decision-making – both the substance and the process – can be affected by managers’ remoteness from the scene of a crisis incident. Through regular and continuous communication, FBI managers must protect against the information disconnects we identified in this case. We recognize that FBI Headquarters’ involvement brings significant operational experience to crisis incidents and can provide perspective regarding how decisions in one incident might impact FBI operations more broadly. Headquarters’ involvement can also bring a measure of deliberation and caution to crisis incidents that might not otherwise exist. In response to the Ojeda operation, we believe the FBI should carefully consider the conditions under which it will assume control over a crisis incident. The FBI should also evaluate how it ensures that adequate information flows to the Headquarters officials who must approve operational decisions, and how the FBI will ensure continuous communication between FBI Headquarters and the on-scene commanders. Recommendation No. 7:   Ensure that the apparent miscommunication between the HRT Commander and Deputy Commander and the lack of adherence to the proper chain of command are not repeated. Following the FBI’s limited breach of the gated door to Ojeda’s house, HRT Commander Craig, who was in Quantico, Virginia, told us that he gave an order to HRT Deputy Commander Steve in Puerto Rico that HRT not enter the residence. However, Steve told us that Craig did not give him any orders at all during their conversations on September 23. This discrepancy was troubling. Steve either misinterpreted or ignored what his superior considered an order, indicating a significant lapse in either communication or command. But just as problematic was Craig’s issuance of any tactical order at all. Although Craig was Steve’s superior as the CIRG ASAC for the Tactical Operations Branch, Craig was not the tactical advisor to SAC Fraticelli, who was the commander for this operation. Under FBI policies that placed the SAC in charge of the operation, Craig did not have authority to make tactical decisions at that time. Chain-of-command responsibilities are set forth in the FBI’s Crisis Management Program guidelines and were clearly defined in the CONOP for this operation, which identified Fraticelli as the on-scene commander. We therefore recommend that CIRG and HRT management review the chain of command discrepancies that Craig’s order demonstrated, as well as the apparent miscommunication between Craig and Steve, and take appropriate steps to ensure this situation is avoided in future HRT operations. Recommendation No. 8:  Provide guidance regarding the exigent circumstances under which operational plans from the field may be presented to Headquarters for approval orally rather than in writing. When AD Hulon called SAC Fraticelli at 8:05 p.m. to tell him that CTD must approve any entry plan, Hulon initially left open the possibility of a nighttime entry and told Fraticelli and the HRT Deputy Commander that the proposal should be put in writing and sent to Headquarters for review.  Hulon told us that he asked for the plan in writing because he considered it standard procedure under the circumstances and because it would allow FBI Headquarters to carefully review the proposed action. We questioned Hulon’s requirement that the proposal be put in writing and believed it was unduly bureaucratic in light of the circumstances. We also noted that Bald told us written plans should be done if time permits, but that there would have been nothing wrong in this case with HRT presenting the plan orally to the SAC and then to Hulon.  While we believe Hulon still would have rejected the plan, we also believe presenting it orally would have accelerated the decision and avoided the disconnect we described between Fraticelli, the HRT Deputy Commander, and the agents at the scene about whether a nighttime entry was going to be approved.  As we explained in Chapter Three, CTD’s decision was not conveyed to the agents at the scene until 11:33 p.m., several hours after Hulon had decided that the entry would be delayed until the next morning. FBI policy allows the use of an oral briefing in lieu of a written arrest plan “in exigent circumstances.” We recommend that the FBI consider providing further, more specific guidance regarding what circumstances may be considered “exigent.” IV.   Recommendations Regarding the FBI’s Preparations for and Conduct of Negotiations Recommendation No. 9:   Adequately assess whether an arrest operation could result in a scenario requiring negotiations. In Chapter Eight, we examined HRT’s decision not to deploy to Puerto Rico with a CIRG negotiation team from the Crisis Negotiation Unit. We concluded that this decision reflected inadequate consideration of a barricaded subject scenario that could result from any attempt to arrest Ojeda. The lack of a negotiation team was caused by the HRT Commander’s narrow view of the goal of the operation and skepticism about the operation’s chances for success, and the Deputy Commander’s understanding that the SAC wanted to avoid a barricade or hostage scenario. We found unpersuasive these explanations for the decision not to include a negotiation team. HRT recognized during its planning for the operation that Ojeda likely would violently resist apprehension, knew that he had previously created a stand-off situation with HRT, and understood that the operation was being conducted in a challenging environment where compromise and the loss of the element of surprise were distinct possibilities. We do not believe that skepticism about an operation’s chances for success or a SAC’s tactical preferences were adequate reasons not to prepare for the foreseeable possibility of negotiations. We also found that applicable FBI guidelines supported the deployment of negotiators with HRT under the circumstances in this case. The Manual of Investigation Operations and Guidelines states that negotiators should deploy with field office SWAT teams “if and when” the potential exists for the use of negotiation resources. The FBI Critical Incident Handbook states, “[i]nvolve the [Crisis Negotiation Team] in the planning stages of a high-risk situation where negotiation may be required.” HRT failed to adhere to this guidance. It is important to anticipate and prepare for contingencies that may reflect lesser preferred outcomes but that may nevertheless occur for reasons outside the control of the arresting agents – in this case, compromise of the sniper-observers and the possibility that a barricaded subject scenario might arise. The FBI should use this case in future operations planning and training to illustrate the importance of carefully considering whether negotiators should be deployed under circumstances where a scenario requiring negotiations is reasonably foreseeable, even if undesirable. Recommendation No. 10:   Ensure that negotiators are integrated into the tactical planning where there is a potential need for negotiations. Recognizing that negotiators might be needed in an operation is only the first step in planning for their potential use. Effective negotiation requires that the negotiators be integrated into the tactical planning of an operation. The FBI Critical Incident Handbook states that “[n]egotiation and tactical strategies should complement/parallel each other. Utilize each in synchronization to affect the safest outcome as possible for law enforcement personnel.” Effective negotiation also relies on organizing the negotiators to enhance communication with the subject and provide expert guidance to the on-scene commander. Under FBI guidelines, this is accomplished with three negotiators:   a primary and a coach who work together at the scene as a team, and a crisis negotiation coordinator who is co-located with the on-scene commander to serve as the negotiations advisor. In this case, while Fraticelli anticipated the possible need for negotiators and arranged for two San Juan FBI negotiators to be available, he did not integrate the negotiators into the tactical planning for the operation. HRT shares responsibility for this deficiency, because even though HRT Deputy Commander Steve told us that he knew San Juan FBI negotiators would be available if needed, he did not take any steps to integrate them into HRT’s tactical planning. In our view, the consequence of this lack of integration was that once negotiations were needed, the negotiators were not organized and did not function as FBI guidelines instruct. Specifically, only one negotiator was allowed at the scene and the crisis negotiation coordinator was not at the Command Post with Fraticelli to serve as an advisor. As we discussed in Chapter Eight, Fraticelli might have handled Ojeda’s demand for a reporter differently if FBI guidelines had been followed. While we could not find that the outcome in Puerto Rico would have changed if negotiators were involved in the planning of the operation, we believe this case highlights the importance of adhering to the guidelines the FBI has developed to make negotiations an effective tool for resolving crisis situations. The FBI should use this case to reinforce the importance of integrating negotiators into the tactical planning of operations where negotiations might be needed.