Report of the Department of the Treasury on the Bureau of Alcohol, Tobacco, and Firearms Investigation of Vernon Wayne Howell Also Known as David Koresh September 1993/Part 2 ("Analysis")

ATF Properly Initiated an Investigation of Koresh
Since ATF's repulsed effort to search the Branch Davidian Compound, some members of the public and the media have questioned the propriety of ATF's decision to initiate an investigation of Koresh and his followers. Questions have been raised as to whether the cult members were justifiably suspected of violating any applicable federal laws. Others have conceded that Koresh was violating the law but have suggested that the violations should have been ignored:


 * What were the Davidians doing to provoke [the raid]? Probably they were converting semiautomatic rifles to full auto. That is certainly a crime; even possessing the capability to convert them is a crime. But down here in the Fifty Caliber Belt this particular crime is usually treated about as seriously as spitting on the sidewalk (Larry McMurtry, "Return to Waco," The New Republic, June 7, 1993, page 16).

Some also have expressed concern about whether ATF was motivated inappropriately to focus a federal firearms investigation of Koresh because of its concerns about his alleged sexual abuse of children and his polygamy. Others have asked whether ATF selected Koresh improperly for investigation because of his nontraditional religious beliefs and practices. One columnist, for instance, asked shortly after the Compound burned to the ground:


 * Who, exactly, were the Davidians bothering? The administration says they were hoarding guns. How un-American and how un-Texan. May we expect the administration to lay siege now to the National Rifle Association? The Davidians were also said to be abusing children. A graver charge, but not a charge that [warrants the ATF's or the FBI's actions]. (Leon Wieseltier, "The True Fire," The New Republic, May 17, 1993, page 25).

A Washington, D.C., newspaper columnist raised similar concerns a few weeks before the Compound burned:


 * No government official has yet explained what crime was being committed by the dimwits of something called the Branch Davidians ... [W]hat provoked this show of force from the crack troops of the Bureau of Alcohol, Tobacco, and Firearms? Was someone caught smoking in a restricted area? Were the faithful of the Rev. David Koresh distilling ardent spirits in an illegal still for one of his holy rites? Was there an illicit drug on the premise or did someone have a shotgun that ran afoul of government standards? ... If Americans cannot live the life of the rugged--albeit somewhat loony--individualist in the vast reaches of the great West, where can they live normal American lives? (R. Emmett Tyrell, Jr., "Crystals in the Waco Crucible," The Washington Times, April 4, 1993, page B1).

These criticisms are not supported by the evidence. A review of the investigation makes it clear that the ATF inquiry into the activities of Koresh and his followers was consistent with the agency's congressional mandate to enforce federal laws regulating the possession and manufacture of automatic weapons and explosive devices. Indeed, ATF would have been remiss if it had permitted considerations of religious freedom to insulate the Branch Davidians from such an investigation.

At the outset, it should be emphasized that ATF focused on the Branch Davidians only after it was asked to do so by local law enforcement authorities who had been scrutinizing the conduct of cult members. Since the Roden shoot-out in 1987 and the return of a substantial cache of weapons and ammunition to the cult, the sheriff's department had watched Koresh add to his arsenal and had heard reports of other behavior on the Compound that generated concern, including the construction of an underground firing range and bunker and a tight cluster of ramshackle buildings that was beginning to resemble a fortress. Although neighbors feared cult members and complained about gunfire at the Compound, the local investigation developed slowly.

In late May 1992, however, the sheriff's department received concrete evidence from UPS that Koresh was receiving substantial shipments of weapons components. In addition, the sheriff's department received reliable reports that Koresh was purchasing large quantities of semiautomatic rifles from local arms dealers. Recognizing that the investigation of a large-scale firearms case involving a close-knit group such as the Branch Davidians would be a substantial undertaking, the sheriff's department sought the assistance of ATF, which has expertise in firearms investigations, as well as a tradition of working closely with local law enforcement agencies.

Even after ATF assistance was sought, ATF agents did not formally open an investigation until making a preliminary determination that federal crimes were being committed. Special Agent Aguilera debriefed local officials and then searched national firearms registries. When Aguilera learned that neither Koresh nor his followers were registered owners of any lawful machineguns, but nevertheless were receiving shipments of machinegun parts and paying for them with large quantities of cash, he reasonably grew suspicious that they were engaged in the illegal manufacture of machineguns. Upon learning that Koresh had received a shipment of materials used to manufacture explosives, specifically grenades, Aguilera formally opened an ATF case in early June 1992.

While some have suggested that ATF targeted Koresh because of his religious beliefs and life-style, the Review has found no evidence of any such motivation. Indeed, ATF recognized early the delicacy of an investigation of such an unorthodox community. Aguilera's supervisors appropriately classified the case as "sensitive," thus ensuring greater supervisory scrutiny of a case that was perceived at the outset to have the potential for raising thorny religious issues as well as difficult safety issues, particularly regarding the women and children living at the Compound.

Whatever controversy there might be about the types of weapons American citizens should be permitted to maintain, federal laws draw a definite line at fully automatic guns and explosive devices such as grenades, which are thought to be more suited for battlefield use than any other purpose. That a private individual has access to a single unlawful machinegun must be cause for federal concern. Where a group is found to be stockpiling many such weapons and to be developing the capability to manufacture many more, ATF must pursue the case. And while the group's religious beliefs should not be cause for targeting it, neither should the beliefs insulate the group from federal scrutiny.

Evidence Developed by ATF's Investigation Warranted Application for and Issuance of Search and Arrest Warrants
Some media accounts have asked whether ATF's investigation unearthed sufficient evidence to support either the ATF application for an arrest warrant for Koresh and related search warrants or Magistrate-Judge Dennis Green's issuance of those warrants. According to The Washington Times, for example:

One Washington lawyer, considered a weapons expert, said a review of the affidavit, written by ATF agent Davy Aguilera, shows there was no probable cause to arrest Koresh or search the property. The lawyer, who asked not to be identified, said the ATF search appeared to be based on a desire "to punish Koresh for showing disrespect" to the ATF. (Jerry Seper, "Affidavit to Search Waco Site Criticized," The Washington Times, September 2, 1993, page 6).

See also James L. Pate, "Waco's Defective Warrants," Soldier of Fortune, August 1993, page 46, which states, "The original affidavit does not show probable cause. Probable cause did not exist." The evidence that ATF presented in support of the warrant applications, however, plainly showed there was probable cause--as that term is commonly understood --that numerous federal criminal violations were being committed by Koresh and his followers.

Aguilera faced two significant obstacles in his investigation. First, he had to overcome the largely antisocial, isolated routine of Compound residents. As a rule, residents never spoke to outsiders about Compound activities and harbored deep suspicions of law enforcement personnel. Second, Aguilera wisely sought to keep his investigation a secret from Koresh and his followers in order to ensure strategic and tactical flexibility in case search or arrest warrants needed to be served. Aguilera sharply circumscribed his inquiries about Koresh to third parties, including arms dealers and former cult members, for fear of alerting the Branch Davidians that they were under scrutiny.

Still, by late February 1993, Aguilera had amassed an impressive amount of evidence that Koresh was unlawfully possessing and manufacturing machineguns and explosive devices and that he was unlawfully storing substantial quantities of black powder. At the outset, Aguilera knew that Koresh was receiving shipments of M-16 parts and materials used to make explosives, among other firearms materials. Because neither Koresh nor any of his followers were registered owners of any M-16 machineguns, indeed, of any machineguns at all, Aguilera could fairly infer that Koresh was purchasing the M-16 parts to convert AR-15 semiautomatic rifles into machineguns illegally. His suspicions were confirmed when he learned that Henry McMahon had sold about 90 AR-15 lower receivers to Koresh and that McMahon tried unsuccessfully to conceal the bulk of those sales and then to mislead Aguilera about the identity of the purchaser. Aguilera also discovered that Koresh had purchased AR-15s and AR-15 upper receivers from several other sources.

Once stocked with AR-15 receivers and M-16 parts, Koresh needed only a metal lathe and milling machine to make more than 100 machineguns. Reports from a number of sources soon made it clear that Koresh possessed both machines at the Compound and that he had experienced operators, including a mechanical engineer, who were designing fully automatic weapons for Koresh and manufacturing them on the premises. Former cult members related how Koresh seemed to be obsessed with the manufacture of machineguns, especially ones of high caliber.

They reported that they had seen automatic weapons on the Compound and that a fully automatic AK-47 had been passed around during one of Koresh's "Bible study" sessions. These statements found corroboration in the reports from neighbors of automatic gunfire erupting from the Compound and the construction of an underground firing range replete with objects riddled with bullet holes.

Evaluations from ATF experts gave Aguilera reason to believe reports that Koresh was manufacturing illegal explosive devices at the Compound. Indeed, ATF's explosives expert determined that several items Koresh had received, including gunpowder and igniter cord, were themselves explosives requiring proper registration and storage--neither of which Koresh provided. The expert also determined that those items, together with the inert grenade shells Koresh had received by UPS, probably were being used to manufacture explosive grenades. Aguilera knew from interviews of former cult members that Koresh had often expressed a keen interest in making live grenades and that grenades had been seen at the Compound. Hence, he had probable cause to believe a search would uncover unlawfully manufactured and maintained explosives.

The intelligence that Aguilera gathered from former cult members and others who had dealt with the Branch Davidians was corroborated further once Special Agent Rodriguez began to visit the Compound in his undercover capacity in late January. Koresh impressed the undercover agent with his technical knowledge of how rifles can be made fully automatic and with his familiarity with the laws regulating the conversion of weapons into machineguns.

By the conclusion of his investigation, Aguilera had more than enough reason to believe that materials and equipment used to make machineguns and explosive devices, as well as the weapons themselves, would be found on the Compound and that Koresh had violated federal laws regulating the possession and manufacture of machineguns and explosive materials and the storage of explosive materials. The ATF decision to seek warrants on the basis of this evidence was thus entirely proper, and it is not surprising that the U.S. Attorney's Office applied for those warrants, and the magistrate-judge promptly approved them.

Some critics of ATF's enforcement actions have questioned why ATF felt compelled to take any action at all at the conclusion of Aguilera's investigation even assuming there had been probable cause to believe federal violations had taken place. Any suggestion that Koresh should have been left to produce and stockpile machineguns and explosives, however, is without merit. The weapons and explosive violations disclosed by the investigation went to the heart of ATF's mission, as defined by duly enacted statutes, and fell squarely within the range of unlawful conduct the agency routinely investigates. Moreover, the information uncovered by Aguilera indicated that Koresh and his followers posed a far greater threat to society than might be posed by an individual who quietly keeps an illegal weapon or even a collection of such weapons. (See Appendix G.)

Of greatest concern to Aguilera was the evidence that Koresh had a propensity toward violence and intimidation. Indeed, Koresh's control of the Compound originated with his triumphant gunfight with Roden, which only was ended by armed deputies who "got the drop" on Koresh before he and his followers could "finish off' the pinned-down and defeated Roden. Furthermore, not only did armed guards receive UPS deliveries, but also they were reported to have been given standing orders by Koresh to shoot any "intruders." On one occasion, the guards opened fire on a newspaper delivery person. Koresh's pronouncements that his time was coming and, that when it did, the Los Angeles riots would pale in comparison also marked him as someone ready to use the machineguns and grenades he was stockpiling.

The veracity of Sparks' account of Koresh's statement about the coming of "his time" and the Los Angeles riots has been challenged by many media sources. Its inclusion in the federal affidavit used to obtain the warrant to search the Compound has been cited as the leading example of how the affidavit was riddled with errors. According to doubters, Sparks last visited Koresh before the Los Angeles riots took place; therefore, her recollection of Koresh's statement must have been faulty. Jerry Seper, who wrote "Affidavit to Search Waco Site Criticized," The Washington Times, September 2, 1993, page 6, for instance, reported the following:


 * The affidavit ... purported to document a conversation Koresh had with an investigator from the Texas Department of [Protective and Regulatory Services]. It quoted the cult leader as making a threat of a fiery compound battle that would make the Los Angeles riots 'pale in comparison.' But Koresh met with Sparks on April 6, three weeks before the riots began.

The purported comment is an example of the affidavit's apparent errors, which have caused some critics to suggest it was designed to justify the disastrous assault by ATF rather than portray conditions inside the Compound accurately.

See also Daniel Wattenberg, "Gunning for Koresh," The American Spectator, August 1993, page 32, which reported the following:


 * [T]he affidavit is wrong. Koresh did not tell Sparks on her visit to the Compound that 'the riots in Los Angeles would pale in comparison' to his self-revelation in Waco. Unless the man really was a prophet, he could not have told her this. The Los Angeles riots broke out on April 29, 1992, more than three weeks after Sparks had last visited Koresh.

The ATF affidavit, which unfortunately discusses only two visits by Sparks to the Compound, one in late February 1992 and one on April 6, 1992, indeed has been the source of this understandable media bewilderment. The source of this confusion lies not in any lack of candor by Sparks, but in the failure of the ATF affidavit to make clear that Sparks' information was the product of more than two visits to the Compound.

The Waco Administrative Review has determined that Sparks, together with other investigators, visited the Compound on at least three occasions: February 27, 1992; April 6, 1992; and April 30, 1992. In addition, she spoke by telephone with Koresh on many occasions, both before and after the April 30 visit. The Los Angeles riots began April 29, 1992, shortly after a Simi Valley jury returned its verdict in the "Rodney King case." The riots were the subject of Ted Koppel's "Nightline" broadcast that evening and were front page news across the nation the next day--April 30, 1992--the last day on which Sparks made a contemporaneous record for a visit with Koresh at the Compound. See Seth Mydans, "The Police Verdict: Los Angeles Policemen Acquitted in Taped Beating," The New York Times, April 30, 1992, page 1. See also Linda Deutsch, "Officers Cleared in Beating," Waco Tribune-Herald, April 30, 1992, page 1.

Obviously, the timing of Sparks' visit to the Compound relative to the riots powerfully corroborates her account; Koresh naturally would rely on such a recent event to help Sparks visualize what he planned for when "his time" came. In addition to Sparks' recollection of the April 30 encounter, her fellow investigators corroborate both the timing and the substance of Koresh's remarks--and the case file for the Compound documents the April 30 visit.

Perhaps most troubling, in light of his collection of weapons and his threatening rhetoric, was Koresh's apocalyptic theology and his renaming the Compound "Ranch Apocalypse." Although Koresh might simply have been preparing to defend himself against an apocalyptic onslaught, ATF justifiably feared that Koresh might soon have been inspired to turn his arsenal against the community of nonbelievers. In fact, the Review has learned that well before the ATF action on February 28, Koresh made plans for just such an event. He told his followers that soon they would go out into the world, turn their weapons on individual members of the public, and kill those who did not say they were believers. As he explained to his followers, "you can't die for God if you can't kill for God." Koresh later cancelled the planned action, telling his followers that it had been a test of their loyalty to him.

The extraordinary discipline that Koresh imposed on his followers, which enabled him, for example, to obtain all of their assets and to establish exclusive sexual relationships with the Compound's female residents, while not itself cause for ATF intervention, made him far more threatening than a lone individual who had a liking for illegal weapons. The Compound became a rural fortress, often patrolled by armed guards, in which Koresh's word--or the word that Koresh purported to extrapolate from the Scriptures--was the only law. And the accounts by former cult members, including an abused child, that Koresh was sexually abusing minors made it clear that Koresh believed he was beyond society's laws. Were Koresh to decide to turn his weapons on society, he would have devotees to follow him, and they would be equipped with weapons that could inflict serious damage.

In the wake of the tragic consequences of the ATF raid on February 28, 1993, and the evidence discovered at the Compound after it burned down on April 19, 1993, it is no longer necessary to speculate on the threat that Koresh and his followers posed. On February 28, as an agent finished reciting ATF's judicial authorization to enter the Compound, the Branch Davidians responded with volleys of deadly fire, using the weapons they had been stockpiling for so long. Some of these weapons were found later in the ruins of the Compound after April 19, including well over 200 firearms, dozens of unlawful machineguns, and numerous prohibited grenades and grenade components. (See Figures 36 through 39. )

In short, the ATF investigation of Koresh and his followers, although posing difficult investigatory challenges for ATF, was an appropriate response to a dangerous situation. In light of the information presented by local authorities, it would have been irresponsible for ATF not to have initiated an investigation and similarly irresponsible for ATF not to have pursued Koresh once Aguilera's investigation showed there was probable cause to do so. ATF's willingness to rise to this difficult challenge can only be commended.

The question remains, however, whether ATF selected the appropriate enforcement option when it decided to forcibly execute search and arrest warrants at the Branch Davidian Compound. The following section of this Report analyzes the process that led to the decision to attempt a raid and the development of the raid plan itself. Additionally, the Treasury-Justice forward-looking review focuses on the broader questions surrounding law enforcement's interaction with nontraditional groups like the Branch Davidians.


 * Figure 36: Photograph of "pineapple" type grenade casings.
 * Figure 37: Photograph of arms bunker with arsenal of assorted weapons.
 * Figure 38: Photograph of remains of assault rifle.
 * Figure 39: Photograph of remains of assault rifle.

Introduction
Any analysis of the plan for the February 28 raid and the planning process that preceded it must recognize that the plan was never actually followed on raid day. In particular, although the plan was entirely dependent on the element of surprise, Chojnacki, the Incident Commander, and Sarabyn, the Tactical Coordinator, went forward despite learning from Rodriguez, the undercover agent, that Koresh had learned something about the coming raid. Failing to recognize the importance of this information, the commanders ignored a critical pre-condition of the plan--the presence of the men in the pit, separated from the weapons--and thus left ATF agents highly vulnerable to attack by the Branch Davidians. See Murphy at B-106; Kolman at B-62; Sobocienski at B-132. The analysis that follows, therefore, addresses a tactical plan that was never implemented.

Nevertheless, even if improvements in the tactical plan might not have averted the tragedy at the Branch Davidian Compound, an analysis of the planning process and the plan itself is still warranted. Most of the Review's tactical experts agree that the plan had a reasonable chance of success if all of the planners' major factual assumptions had been correct. See Murphy at B-104; Ishimoto Executive Summary; Sobocienski at B-131. If the men in the Compound could have been counted on to be working in the pit, separated from the weapons reportedly locked away in the "arms room," and if ATF agents could have driven up to the Compound without its residents knowing of the operation until it was too late to offer effective resistance, the warrants might well have been executed with a minimum of disruption and without loss of life. But the caveat here is crucial, for many of the planners' assumptions were just that--expectations based on too little information about the Branch Davidians.

The problems here rest as much in the planning process as in the plan itself. Not only were the planners, led by Sarabyn, too quick in concluding that a massive mid-morning raid was the best possible enforcement option, but they chose a plan whose window of opportunity might have been far smaller than they realized. The planners also failed to prepare for contingencies that would arise if that window were missed. Against a target as formidable as Koresh, such errors exposed ATF to grievous consequences.

Responsibility for these flaws cannot simply be placed at the feet of those who did the actual planning. Those charged with this mission devoted considerable time and energy to devising a safe and successful operation. They lacked, however, the training, experience, and institutional support that were demanded by the extraordinary operation they were planning, which was qualitatively, as well as quantitatively, different from the many smaller enforcement actions each had led successfully in the past. See Ishimoto at B-15; Paschall at B-109; Murphy at B-95; Kolman at B-63; Morrison at B-88. ATF's management never addressed these deficiencies by giving the planners a supportive structure to supplement their own experiences. In addition, ATF's upper management did not actively oversee the development of the tactical plan, even though it involved the mobilization of more than 100 agents--the largest law enforcement effort ever mounted by ATF and one of the largest in the history of civilian law enforcement. See Morrison at B-87.

The Decision to Execute the Warrants by "Raiding" the Compound Was Made Before Other Options Were Fully Exhausted
Before reviewing the development of the raid plan devised for the Branch Davidian Compound, this section begins by looking at the decision to have a "dynamic entry" as opposed to some other type of enforcement action. Based on the limited information available when ATF selected the raid option, the Review simply is not in a position to say conclusively whether ATF's decision to rely exclusively on a raid--instead of, for instance, a scheme to arrest Koresh while he was away from the Compound or to establish a perimeter around the Compound and negotiate--was well founded. However, the Review does not believe that the ATF planners were in a position to make such a judgment either, and they should have been. A massive raid against a heavily armed, disciplined, and well positioned group will always, however cunningly planned, be a risky operation, especially when children and other innocent persons are present. If less risky alternatives that can achieve the same ends are possible, they ought to be pursued. ATF did not adequately pursue these options. The agency's failure to gather the information needed to assess the chances of such alternatives succeeding made its rejection of them, and its choice of the raid option, far too premature. The Decision to Use Force When Executing the Warrants

The threshold issue presented to ATF was whether any force would be needed to execute the arrest warrant for Koresh and the search warrant for the Compound. Some have suggested that, having obtained such warrants, ATF should simply have asked Koresh to surrender himself and his weapons or asked him for free passage into the Compound so that ATF could conduct a search for unlawful firearms and explosives. Claiming that Koresh had surrendered previously to local law enforcement authorities after his shoot-out with George Roden in 1987, critics have argued that ATF's decision to use force made a violent confrontation inevitable and played into what they have characterized as Koresh's apocalyptic vision of a final battle between his army and law enforcement agents, whom he called the "Assyrians." The Review finds no basis for these criticisms and believes that the decision not to rely on Koresh's goodwill was entirely appropriate and rested on valid considerations.

There was, in fact, no evidence that Koresh was prepared to submit to law enforcement authorities or that he had done so in the past. His surrender in the Roden shoot-out occurred only after deputies of the McLennan County Sheriff's Department "got the drop" on him while he and his followers were busy training their weapons on Roden, who was pinned down behind a tree at the Compound. Based on the information developed during the course of Special Agent Aguilera's investigation--which showed Koresh's propensity toward violence, his use of armed guards, and his control of a massive arsenal of automatic and semiautomatic weapons--the ATF planners reasonably concluded that a polite request to search the Compound without readiness to use force would have been foolhardy and irresponsible. This conclusion could only have been confirmed once Special Agent Rodriguez began his contacts with Koresh and learned of the latter's disdain for the firearms laws and hatred for those charged with their enforcement. While concern about Koresh's apocalyptic vision should have suggested using an enforcement approach that afforded an opportunity to first ask for voluntary compliance and to avoid an initial, potentially provocative show of force, it was not a valid reason for ATF to forsake its law enforcement responsibilities. Intelligence Failures and the Failure to Try to Arrest Koresh Off the Compound Followed by an Effort to Execute the Search Warrants

Having understandably decided not to rely solely on Koresh's voluntary compliance with the warrants, ATF tactical planners initially focused their attention on arresting Koresh while he was away from the Compound, either by luring him off or by waiting until he had left it on his own accord. Koresh's followers, the planners believed, had become so accustomed to relying on his leadership and guidance that they would be far less likely to resist ATF in any organized way if Koresh could be removed from the scene; this advantage would be significant whether ATF established a siege of the Compound or conducted a raid.

It is now impossible to know whether ATF's execution of the search warrant would have been aided by arresting Koresh while he was away from the Compound. Still, the planners' reasoning on this score makes sense, and their consideration of this option indicates an effort to minimize the risk to agents and Branch Davidians. That effort, however, was not sufficient, because it was abandoned prematurely, without adequate exploration of its feasibility.

As early as their first tactical planning meeting in December 1992, the planners concluded that Koresh virtually never left the Compound, therefore, they thought they would not be able to lure him away from it. And they remained convinced of this when the undercover house was established in early January 1993. The Review, however, has not been able to identify the basis for this early conclusion. For example, no written ATF report addresses the issue before this meeting, and aside from Joyce Sparks, who had only visited the Compound a few times and told Aguilera she thought Koresh did not leave often, the Review was unable to identify a reliable source for this common assumption among the planners. Certainly, there is no evidence that any of the tactical planners evaluated either the source or substance of the initial intelligence on this point.

The establishment of the undercover house should have given the planners an opportunity to test the validity of their assumption, by having the undercover agents monitor Koresh's comings and goings. But this did not happen. Although reports from the undercover house seemed to confirm the earlier "intelligence" because the agents indicated that they never saw Koresh leave the Compound, this information was unreliable.

The defects in the intelligence relating to Koresh's movements are particularly significant, because they are symptomatic of the problems that afflicted the undercover house operation. To be successful, an intelligence operation must be able to develop adequate and reliable information, disseminate that information to the appropriate supervisors, and ensure that those supervisors recognize the meaning and limitations of that information. See Murphy at B-100; Ishimoto at B-17; Kolman at B-58; Morrison at B-87, 89; Paschall at B-109, 111. The undercover house operation fell short in all three of these areas.

From the outset, the production capabilities of the agents in the undercover house were crippled by ATF's failure to give them a comprehensive idea of what the planners needed. The agents also lacked the supervision and feedback needed to ensure that they performed their mission. "The organization of the undercover house and its activities was marked by no clear chain of command or direction of their actions." See Ishimoto at B-17. Generalized surveillance often serves an important function, especially during the investigative stages of a case, when agents begin to explore the nature of a target's activities. But by the time the undercover house was established, ATF's tactical planners had a number of specific questions that they either wanted answered or were assuming already had been answered. Instead of being told what these questions were, the agents in the undercover house were told only to look for certain routines at the Compound. Without clear objectives, supervision, or feedback, morale and performance began to deteriorate soon after the operation began, and the vigilance of the agents suffered accordingly.

One result of this intelligence production breakdown was that the agents in the undercover house could not tell reliably whether Koresh ever left the Compound, and they never took the additional measures necessary to find out. With Sarabyn's approval, the agents discontinued around-the-clock surveillance. And even when they were watching the Compound, the agents could not identify the occupants of cars seen leaving the Compound on the road in front of it and could not see whether anybody left the Compound on foot or in vehicles using the road or trails behind the Compound. With limited night-vision capability, the agents were unable to determine who left at night. In addition, several agents did not know, and no agents were certain, what make of car Koresh drove. Although some of the agents learned at various stages in the undercover operation that Koresh reportedly drove a Chevrolet Camaro, a number of other cars at the Compound were registered in Koresh's name because cult members gave all their possessions to him. Koresh, therefore, presumably might have driven any of the cars.

Dissemination of information was also a problem, because no adequate provisions were made for providing the raw intelligence that the agents in the undercover house were able to obtain to the tactical planners. Supervisors did not direct undercover house agents to keep comprehensive surveillance logs throughout the operation, and no arrangements were made for the agents to give oral briefings to the tactical planners or to an intermediary.

Although many of the tactical planners thought that surveillance was being supplemented with regular debriefings of Rodriguez about his contacts with Koresh and other cult members, little useful intelligence about Koresh's movements was actually gleaned from the agent's relatively few contacts. (See Figure 40.) However, salient intelligence was obtained on February 17, late in the planning process, when Koresh offhandedly told Rodriguez that he rarely went to town because the people there did not like him. Aguilera noted this significant statement in his reports. Even though the statement supports the belief that Koresh did not leave the Compound often, it contradicts the planners' view that he never left, and should have alerted them that from time to time, Koresh did leave the Compound.

ATF's mishandling of the intelligence regarding whether Koresh ever left the Compound resulted, in part, from the lack of a system to process intelligence. Rather than flowing to a single accountable person responsible for collection and analysis, intelligence swirled among many persons--none of whom sufficiently questioned its reliability. Overall intelligence collection and planning was not centrally managed. See Ishimoto at B-17. Because it was not treated with sufficient rigor, intelligence may well have been interpreted to conform to planning needs. If a single person had been responsible for the production and dissemination of tactical planning intelligence, the problems at the undercover house might have never arisen. If they had occurred, they could have been corrected or at least brought to the planners' attention.

The limitations of the intelligence operation were compounded because the tactical planners did not recognize those limitations. As a result, they overvalued the significance of the information they did receive. Thus, although the undercover agents' knowledge of Koresh's movements was thin at best, the tactical planners believed that they had confirmation that Koresh never left the Compound. Moreover, having surveilled the Compound and its surroundings only a few months earlier, the planners should have recognized the inherent limitations of monitoring Koresh's movements from the undercover house, since the rear of the Compound could not be seen from there. At the very least, Sarabyn, who had visited the undercover house, must have understood that Koresh could have left undetected from the rear of the compound. As a consequence of their belief that Koresh never left, the planners devoted little effort to developing a plan to lure him off the Compound. By the end of January, after attempting unsuccessfully to convince Joyce Sparks' superior to allow Sparks to request a meeting with Koresh at her office, ATF largely abandoned any effort to lure Koresh away from the Compound.


 * Figure 40: Diagram depicting Rodriguez's undercover contacts with the Compound.

Had more attention been paid to determining whether Koresh ever left the Compound, ATF's planners might have learned that he did in fact leave the Compound on at least two occasions while the undercover house was in operation and on several other occasions in late 1992 and early January 1993. This is not to say that he could have been intercepted on any of these trips or that ATF could have devised a plan that would have succeeded in luring Koresh away. But, given the planners' reasonable expectation that arresting Koresh away from the Compound would vastly reduce the risks attending any law enforcement action at that location, far more effort should have been made in this area. See Kolman at B-47, B-50. And ATF's failure to make such an effort must be attributed to management's failure to establish an effective intelligence operation.

A Siege With Koresh Present on the Compound

Regardless of whether Koresh could be arrested away from his followers, ATF still had to decide how it would execute the search warrant at the Compound. Initially, the tactical planners considered the siege option. In this scenario, agents would first ask those inside the Compound to honor the warrant. If access were denied, ATF would immediately establish a perimeter around the Compound and seal off its inhabitants until they relented and permitted the search to proceed. This approach would minimize the risk of a violent confrontation between ATF and the Branch Davidians and, even if violence erupted, would minimize the agents' exposure to gunfire from the Compound. These advantages led some planners to favor a siege over a dynamic entry plan even after they had surveyed the Compound and its surrounding area.

The planners ultimately rejected the siege option mainly because the intelligence obtained in January from former cult members highlighted the drawbacks of such an operation. Most significantly, several former cult members noted the distinct danger that Koresh would respond to a siege by leading his followers in a mass suicide. Even if no suicides occurred, the costs of a siege would be high. With their own source of well water, a three-month supply of military rations, and experience with the rigors of a rudimentary lifestyle, the Branch Davidians, former cult members believed, could withstand a long and arduous standoff. The planners were also concerned that a siege would give Koresh and his followers time to destroy evidence of their violations of federal firearms and explosives laws. Several tactical planners expressed concern that Koresh would outlast the patience of the American public and that they might be directed to raid the Compound after a lengthy stalemate. They feared that such a raid, against a prepared and fortified foe, would be far more dangerous than a surprise raid.

In retrospect, many of the tactical planners' concerns about a siege were validated, especially the fear of mass suicide and the appraisal that Koresh had the discipline and resources to withstand a siege for a prolonged period. See Kolman at B-48-49 That the planners were proven correct, however, does not necessarily validate the process by which they reached their conclusions.

Several of the planners told the Review that they assumed, in substance, that when dealing with a cult, mass suicide is a serious risk. Interviews certainly provided an opportunity to assess former cult members' credibility, emotional state, and objectivity, and a basis for determining whether their suicide predictions were reasonable. Nonetheless, before allowing the specter of mass suicide to deter them from pursuing the siege option--which they considered less risky to all involved--the planners should have sought assistance from psychologists and other experts who were better equipped to evaluate the accounts of the former Branch Davidians. Consultation with such experts could also have improved ATF's overall understanding of Koresh and his followers, including the group dynamics among the cult members inside the Compound and their extraordinary belief systems. Such an understanding, in turn, might have broadened ATF's consideration of enforcement options other than a raid, and heightened their appreciation of the dangers of raiding a group that apparently shared an apocalyptic theology.

There were many serious drawbacks to a siege, and ATF's tactical planners accurately perceived them. Because of these valid tactical concerns and perhaps because, like most law enforcement agencies, ATF had little experience with extended sieges, the tactical planners viewed a siege as a tough option. In the end, though, the chief reason the planners discarded the siege approach was their increasing optimism about the dynamic entry option. This optimism, unfortunately, was in large part based on faulty intelligence that made conditions seem much better for a raid than they really were. See Murphy at B 103; Morrison at B-87.

The Decision to Pursue a Raid Option and Develop a Raid Plan
The chief attraction of a raid scenario was that it offered the possibility of catching Koresh and his followers by surprise and avoided the risk of a protracted and costly standoff. The element of surprise seemed quite achievable to the planners, based on their flawed understanding of the daily routine in the Compound. If agents could sweep into the area at 10:00 a.m., they would find the Branch Davidian men working in the pit outside the Compound, without access to the weapons that Koresh kept under lock and key next to his bedroom. The men could be detained, the arsenal secured, and Koresh arrested. The assumptions on which this plan rested, however, relied on the same inadequately evaluated intelligence that led the planners to prematurely discount the possibility of Koresh ever leaving the Compound. As a result, the operation was far more vulnerable than its planners, especially Sarabyn, ever realized. And this vulnerability was increased by the failure of the planning process to produce a common understanding among the planners of what the operation's key assumptions were, and of the importance of surprise to the mission's success.

Had they appreciated the risks involved, the planners might have done far more to prepare for the possibility that ATF might not be able to surprise the Branch Davidians. As it was, they did virtually no contingency planning. See Murphy at B-104; Kolman at B-65, 36; Morrison at B-88; Ishimoto at B-15. And they failed to devise a structure that maximized the flow of intelligence to the key decisionmaker on raid day so that he could verify that Koresh was unaware of the impending raid before committing ATF agents to the front of the Compound.

Intelligence Failures
This section examines the plan's critical assumptions and the quality of the intelligence on which they rested.

The "Arms Room"

The tactical planners' conclusion that Koresh kept the weapons and explosives under lock and key in a room adjacent to his own was based almost exclusively on the statement of one former cult member, David Block. Block's actual statements to ATF agents, however, were far less definitive than the planners treated them; he had simply indicated that Koresh maintained control over the weapons in the neighboring room and that his permission was needed to possess one. Although Sarabyn and the other planners could have contacted Block and other former cult members to clarify this matter and learn more about the circumstances under which Koresh distributed weapons, no such effort was made. Moreover, the planners failed to consider how Block's prior relations with Koresh, and his decision to break away from the Branch Davidians at the Compound, might have affected the reliability of his statements. Although the planners knew Block had met with a self-described "deprogrammer," Rick Ross, they never had any substantive discussions with him concerning Block's objectivity about and perspective of Koresh and his followers. Nor did the planners pay appropriate attention to the fact that Block had left the Compound over six months earlier.

Even though the success of their plan might have turned on whether Block was right about the location and control of the Compound's weapons, ATF's planners simply began to treat the report that the arms were kept under lock and key in the arms room near Koresh's bedroom as established fact. When Sarabyn explained to Cavanaugh, who had not attended the late January meeting, why the planners had chosen to proceed with a raid rather than a siege, he noted that an important factor in the decision had been intelligence that the guns were kept under lock and key by Koresh. Sarabyn reported as fact his speculation that, because of an increasing paranoia, Koresh never distributed the guns for fear of a mutiny. Several other tactical planners inaccurately believed that Block's statement about the arms room was corroborated by several other former cult members.

Having fixed upon Block's statements about the storage routine for the weapons, the planners apparently ignored the rest of Block's information: that Koresh distributed AK-47s from time to time, that cult members would keep them under their beds on those occasions, and that Block was uncertain about whether they kept their guns loaded. They also discarded reports by other former cult members that they had seen weapons distributed around the Compound and that Koresh frequently conducted live-fire shooting practice. In particular, they did not pay sufficient attention to Block and Breault's assertions that they had left the cult in large part because of Koresh's insistence that they prepare to resist law enforcement authorities when the anticipated confrontation came. In fact, as law enforcement officials discovered after the Compound burned, powerful evidence indicates that arms were stored in other locations in the Compound in addition to the arms room. No Guards or Sentries

At the outset of the planning process the tactical planners--based on the statements of former cult members, including Block and Breault, the UPS delivery person, the sheriff's department, and Texas National Guard overflights--anticipated that there would be armed guards or sentries at the Compound. After the agents in the undercover house saw no such guards, however, the tactical planners concluded that Koresh had stopped taking this precaution. Why he might have done so is something they did not sufficiently consider. If Koresh's control of weapons and refusal to leave the Compound stemmed from paranoia or fear of law enforcement, as the planners appeared to think they did, it would have been rather odd for him to suddenly stop posting sentries. The planners never considered that Koresh might simply have repositioned his guards inside the Compound where they would not have been seen by the undercover agents.

It appears that sentries were indeed concealed within the Compound during the weeks leading up to the raid. Mark Spoon, who lived next door to the undercover house, has related to the Review that, during this period, Koresh or one of his "Mighty Men" often would telephone him when an unfamiliar car drove up the road toward the Compound, and inquire about the car's occupants. To know a car had approached on the road, a vigilant cult member would have had to have been posted in the Compound's tower; such lookouts were in place on raid day.

The Men in the Pit

The same intelligence processing failures that led planners to be confident that Koresh never left the Compound led them to be unduly confident that all or most of the Compound's men could be found in the pit at 10:00 a.m., every morning, except on Saturday, their sabbath. Even though most dynamic entries are executed shortly before dawn--when most suspects are likely to be asleep and caught by surprise--the planners' confidence that the men would be in the pit led them to give up the predawn advantage in exchange for finding the men in the pit at 10:00 a.m.

The mother of a cult member, who had visited the Compound for two days in November 1992, had some recollections of the work routine there. But the principal source of the planners' information about the men in the pit was the surveillance conducted by the agents in the undercover house. However, neither the visitors' observations nor the reports of these agents supported the tactical planners' assumptions about a predictable routine.

Certainly, the surveillance logs maintained by the agents at the undercover house provided no basis for such a conclusion. See Ishimoto at B-19. For four days of the operation, there is no record that any surveillance was conducted and, for two other days, the only entry is the notation "no activity." Indeed, between January 11 and January 29--during which time the tactical plan was drawn up--on the majority of those days the logs do not refer to the men working in the pit. On several other days, the logs indicate that there was no activity when it was rainy, implying that no one had been in the pit. Over the life of the undercover house operation, from January 11 through February 17, the surveillance logs refer to the men working in the pit on only 14 out of the 36 days for which surveillance was maintained. Although the undercover house ceased 24-hour surveillance efforts on January 19 and terminated surveillance activities altogether on February 17, many tactical planners erroneously believed that tight surveillance continued until the day of the raid, compounding their overvaluation of the intelligence. In fact, several of the tactical planners first learned about the limited nature of the surveillance coverage only after being interviewed by the Review. Since Sarabyn approved the cessation of 24-hour surveillance, responsibility for the other planners' ignorance on this point must be shouldered by him.

Even though some log entries refer to the men in the pit, the logs only reported sporadically how many men had worked, exactly when they began work, whether the men worked in the rain, or the degree to which activity in the pit was actually visible from the undercover house. In fact, the agents never could have seen how many men were actually in the pit, because their reports of work there were based on their observations of traffic between the front of the Compound and the pit, people near the pit passing supplies down into the pit, and construction noises emanating from the Compound. The absence of any view of the interior of the pit itself was significant because, as Sparks and others had informed ATF, Compound residents could enter and exit the pit through the buried bus, without ever having to walk above ground. And even if the agents had been able to look into the pit and count the men inside, they might not have known what proportion of the Compound's men were working, since they never knew how many men lived on the premises. While agents believed that approximately 75 people lived at the Compound, more than 125 people were there on February 28. On those days that the surveillance logs did indicate the number of men observed working in the pit, the number was never more than 13. Had the planners assessed the logs properly, they at least would have questioned their belief that almost all of the men residing at the Compound would be found in the pit.

Even though the surveillance logs were not the only source of intelligence from the undercover house, alternative avenues were not exploited sufficiently and hence did not compensate for the information lacking in writing. Oral communications were infrequent and informal. Aside from one occasion in mid-January when an undercover agent recalls speaking to Sarabyn about the routine of the men working in the pit, none of the agents in the undercover house recalls being debriefed by Sarabyn or any other tactical planner about the matter. Similarly, videos and photos of activities at the Compound were taken but never viewed by the raid planners.

Even though they were presented with insufficient information about the men in the pit, none of the tactical planners requested that particular attention be paid to this point. Instead, they assumed the existence of a predictable routine, based on the inadequate information they had, and then based their entire plan on that assumption. Indeed, the planners' misplaced confidence that virtually all the Compound's men could be found in the pit each morning might have caused their failure to take any measures to ensure that the presence of the men in the pit be confirmed before the raid went forward. No raid commander was charged on raid day with verifying this critical precondition for the operation's success. And when Cavanaugh, positioned in the undercover house on February 28, observed before the raid began that "all was quiet" at the Compound and did not see the men in the pit, he did not fully appreciate the enormous significance of this lack of activity. Certainly, Chojnacki and Sarabyn disregarded the importance of the condition, for they rushed to launch the raid before they expected the men to be in the pit.

The Discounting of Armed Resistance from Women in the Compound

The planners were able to base the raid plan on the presence of the men in the pit because they apparently assumed that the women would not use the weapons Koresh had stockpiled. Although the planners anticipated that one female cult member, a former police officer, might be armed with a handgun, they studiously ignored or discounted evidence that other women might also be prepared for armed resistance. The planners apparently gave little weight to Block's statement to Buford and Aguilera that Koresh issued rifles from time to time to at least five of the women. A photograph taken by undercover house agents in late January of a female aiming a rifle from the front door of the Compound was shown to Sarabyn, who did not share it with the other raid planners. Sarabyn apparently concluded that the rifle might be a BB gun and could not understand why the woman was aiming it. The picture, however, is of poor quality, and the type of rifle cannot be discerned. The failure to pursue this matter by enhancing the image or seeking more information from the undercover agents, is yet another example of how data inconsistent with the planners' assumptions often was shunted aside.

At one level, the intelligence failures that lessened the chances that ATF's tactical plan would succeed were management failures. The agents in the undercover house did not conduct effective surveillance or keep comprehensive records of what they did see and what they could not see. The planners did not alert the undercover house agents to their tactical intelligence needs or ask hard questions about the information they received. But it would be quite unfair simply to hold these individuals responsible for the breakdown here. The agents in the undercover house should have been supervised by someone attuned to the needs of the planners. And the planners, charged with engineering the biggest raid in ATF history, should not have been required to interpret raw intelligence. What was needed was a separate intelligence structure to ensure that usable, reliable information was funneled to the planners and that the planners knew the limitations of the data they received. See Ishimoto at B-17; Murphy at B-100; Kolman at B-59; Morrison at B-89. ATF's leadership at headquarters and in the Houston division must bear responsibility for allowing the operation to proceed without such a structure.

No Meaningful Contingency Planning

The same confidence that led ATF raid planners to discard intelligence inconsistent with the assumptions central to their plan might also have led them to do little to prepare for the possibility that conditions would not be right on raid day. See Kolman at B-14 and B-66. Perhaps they did not realize how fragile, because of its dependence on surprise, their plan really was. The absence of any contingency plan, other than to abort the raid before arrival at the front of the Compound, left the raid commanders with the stark choice between going forward or canceling an operation in which so much already had been invested. That failure also meant that when ATF agents encountered heavy gunfire upon their arrival at the Compound, most had little choice but to proceed with their mission, at great cost. See Morrison at B-88.

If there had been meaningful contingency planning for the possibility that ATF might lose the advantage of surprise before the cattle trailers arrived at the Compound, agents confronted with a forewarned target would still have been able to move into siege positions, securing a perimeter around the Compound. Compared with a surprise raid, a siege had marked disadvantages, all of which the planners recognized. Yet a siege was a preferable alternative when compared with a raid against a target that was ready and waiting. The planners should at least have explored using a siege as a recourse. The planners also did not prepare for the possibility that Koresh would try to break out of the Compound, alone or accompanied by his followers, if he learned about the raid. Indeed, there was not even a plan for postponing the raid, even though certain circumstances, such as a late-night Bible session or inclement weather, might cause the Compound's men to arrive at the pit later than 10:00 a.m. By failing to establish any alternatives to the raid plan, ATF's tactical planners contributed to the pressures on the raid commanders to go forward with the massive operation and to not let the training resources invested and the planning for the raid go for naught.

In addition, sufficient thought was not given to what ATF agents would do if they arrived in front of the Compound and were met with either an organized ambush or scattered pockets of armed resistance. If necessary, the cattle trailers could have made a detour before reaching the road leading to the Compound. But once the trailers went up the driveway and reached the front of the Compound, the agents reached a point of no return: The plan called for the agents to carry out their assignments, regardless of the resistance they encountered. Since the cattle trailers provided no protection and the grounds in front of the Compound afforded only limited cover, the raid planners saw no better option.

If the tactical planners had given sufficient thought to the level of firepower that Koresh and his followers could bring to bear on agents massed in front of the Compound, they might have done more to ensure that the raid would not go forward without the advantage of surprise. The raid commanders sent Rodriguez into the Compound to check conditions only because of the publication of the first part of the "Sinful Messiah" series. The plan produced by the tactical planners did not call for such an effort. In addition, had the plan incorporated verification of conditions by an undercover agent, measures could have been taken to close the time gap between the undercover agent's departure from the Compound and the arrival of the trailers. They also would have prepared some scheme to help agents withdraw from their vulnerable positions if they became pinned down by hostile fire. One method of extracting the agents might have been to send in Bradley Fighting Vehicles, which could have been positioned a short distance from the Compound, concealed on flatbed trucks. In any event, a reserve force of agents could have been deployed nearby. To the extent that these precautions would have taxed or exceeded ATF's resources, the agency might have reached out for assistance from other law enforcement authorities or reconsidered whether to conduct the raid. Because the planners did not explore any of these contingency options, ATF did not have a plan or the capacity to extract any agents, including wounded agents, from their exposed positions in front of the Compound.

Having failed to prepare for an ambush, the planners also failed to prepare for a stand-off. Given their fears that Koresh might lead his followers in a mass suicide if surrounded by agents, it is unfortunate that the planners did not heed Deputy Tactical Coordinator James Cavanaugh's repeated requests that ATF have a contingency plan to negotiate with those inside the Compound. See Kolman B-51-52. The raid commanders did not even arrange to have the telephone number for the Compound on the day of the raid. Cavanaugh was able to find the number written on a calendar in the undercover house after the shooting began. ATF was fortunate that Cavanaugh filled the planning gap and handled the crisis adeptly.

Among other contingency issues the planners left unresolved was whether agents would still seek to execute the warrant at the Mag Bag if ATF agents were repulsed at the Compound or, alternatively, whether they would set up a perimeter around the Mag Bag to contain any cult members inside on raid day. In large part because this issue was not considered, ATF left the Mag Bag unsecured for a brief period on February 28, and three cult members ended up maneuvering around ATF's flank, endangering the lives of many agents and eventually engaging in a deadly shootout with a group of ATF agents. See Kolman at B-64.

The absence of any contingency planning cannot be attributed entirely to the planners' confidence that conditions would be favorable and that ATF's advantage of surprise would be decisive. It also reflected the planners' lack of experience in orchestrating operations of this magnitude. Only Buford had been involved in the planning of an enforcement action of comparable size, the CSA siege. This siege was a set-piece encounter where the need for fallback positions was less critical than it was for the dynamic entry contemplated at Waco. Indeed, the fact that the CSA siege was in large measure a success may have led the planners to discount the likelihood that the action against the Branch Davidians would go awry. In addition, this success confirmed what the other SRT leaders knew from their own experiences leading countless smaller operations: Things might not go as planned, but ATF could still successfully achieve its objectives.

The result of these diverse factors was a tactical plan that did not contemplate any meaningful contingencies. And the training for the raid followed the lead of the planning. The sessions at Fort Hood concentrated almost exclusively on preparing the agents to enter the Compound quickly and secure the residents expeditiously. Little time was spent training the agents to withdraw from the Compound in an orderly manner if necessary. In light of this planning and training focus, it is not surprising that the mind-set of ATF's commanders on raid day was to go forward with the raid unless the Branch Davidians were seen preparing to ambush.

Given law enforcement's limited experience in operations of this magnitude, the failure of the planners to consider that their operation might go awry and prepare for that eventuality is tragic, but somewhat understandable. In contrast, the failure of ATF's national leadership to ensure that some contingency planning was done is simply unacceptable. The headquarters officials briefed on the plan certainly knew that the raid planners lacked experience with operations of this size, and they should have recognized the risks involved in the raid. Yet it does not appear that anyone in ATF's leadership asked the obvious questions beginning with "What happens if ..." and then directed that further planning be done to address those concerns. Management cannot be expected to know all the details of a field operation, but its job is to ask these hard questions and carefully consider the answers. Had ATF's top managers considered the implications of a plan that could leave a large force of agents stranded in front of a Compound containing heavily armed fighters, and could leave the agents no alternative but to fight their way out, February 28 might have ended differently. See Morrison at B-86-87.

Command and Control Flaws in the Raid Plan
Other deficiencies in the planning effort that likely contributed to the pressures felt by the raid commanders on February 28 rest with the command and control structure established for the operation and with the selection, placement, and use of command personnel. Overall command of an operation of this magnitude must be placed in the hands of commanders who have access to the information on which decisions to proceed or abort must be based, who have an understanding of that information, and who have a perspective from which they can make measured judgments on how to proceed. Furthermore, because ATF's raid commanders placed themselves in locations where calm deliberation was difficult and because they lacked appropriate intelligence support, the likelihood that these commanders would make the right decisions on raid day was reduced. See Kolman at B-62 63; Morrison at B-88.

The General Command Structure

The raid on the Branch Davidian Compound was the first ATF operation conducted within the framework of the National Response Plan (NRP). ATF generated the NRP after a number of multiple-SRT mobilizations in order to establish consistent policies and procedures for such efforts. The NRP was finalized only shortly before the events near Waco. Although the plan barely addressed critical issues such as how a major operation should be planned, it did establish a command structure for such actions and specify how positions within that structure were to be filled.


 * The NRP described the two main command positions in an operation:


 * (SAC)/Incident Commander--"in charge and responsible for operational and administrative control of critical incident management resources," to "determine the overall strategy for responding to and/or resolving a critical incident or operation" and to prepare a written operations plan; and


 * Tactical Coordinator--"[D]esignated by the SAC/Incident Commander to direct and control all tactical (operational) functions during a critical incident." "He/she will direct all SRTs assigned to the critical incident" and will "[s]upervise the development of specific tactics and procedures to support the SAC/Incident Commander's strategy for resolving the critical incident or completing the operation. These tactics and procedures will be subject to the SAC/Incident Commander's approval."

Because Chojnacki was the SAC for the affected ATF field division, the NRP mandated that he serve as Incident Commander for the raid, regardless of whether he had adequate tactical training and experience for this particular mission. Chojnacki had more than 27 years of law enforcement experience at the time of the raid and had participated in and directed countless search and arrest warrant executions. Other senior agents with ATF, however, had more relevant training and far greater experience in substantial tactical operations. He in turn chose Sarabyn to be Tactical Coordinator, guided by the NRP requirement that this position be filled by "an ASAC who has completed SRT training." Because agents from both the New Orleans SRT and the Dallas SRT would be involved in the raid, Pete Mastin, the New Orleans SAC, and James Cavanaugh, the Dallas ASAC, were designated Deputy Incident Commander and Deputy Tactical Coordinator, respectively, consistent with the NRP.

Although credit must be given to ATF for establishing a framework like the NRP so that structural issues would not have to be reconsidered every time a major operation needed to be planned, the command structure dictated by the NRP set ATF's planning for the raid off on the wrong foot. By assigning personnel to critical command positions on the basis of rank--rather than ability, experience, training, or knowledge of the case--the NRP created a chain of command that did not ensure that each position was filled by the most qualified individual. See Kolman at B-65-66; Ishimoto at B-13. Sarabyn, who had led the tactical planning team, had supervised this investigation and had been instrumental in drafting the NRP, had to report to Chojnacki, who lacked this background, and consequently deferred to Sarabyn on many critical issues. This deference blurred lines of responsibility. In turn, Sarabyn was effectively in charge of the tactical planning for the operation despite his lack of any large-scale tactical planning experience. Sarabyn had attended SRT training, but only as an observer. In any event, that training was designed to teach SRT members to perform as a team, and it did not focus on developing tactical leadership skills or planning capabilities for larger operations. Sarabyn was selected as the Tactical Coordinator for the operation not because of his expertise, but because the NRP required the position to be filled by a person who had received SRT training and was an ASAC or a higher ranking official. These requirements limited the field of candidates and excluded persons of lesser rank who had significantly more experience. Likewise, Cavanaugh, who was chosen as Deputy Tactical Coordinator because of his position in the ATF hierarchy, also lacked the familiarity with the operation needed to be an effective commander. Meanwhile, Buford, the only participant in the raid who had directly relevant experience, was relegated to joint command of one of the SRTs.

Command and Control on Raid Day
The command and control plan established for the raid near Waco accentuated the NRP's structural deficiencies by failing to place commanders where they could make informed, considered decisions and maintain control over the day's events.

Chojnacki, although in charge of the entire operation, placed himself in a helicopter during the critical phases of the operation--the final 30 to 40 minutes before the cattle trailers arrived at the Compound and at the outset of the firefight. As a result, he could not effectively communicate with either the other raid commanders or the SRT team leaders during this entire period. See Ishimoto at B-14; Kolman at B-63; Sobocienski at B-130. Similarly, Sarabyn, who could best evaluate the significance of events at the Compound, could not see the Compound during his 17-minute trip from the staging area. By riding with the cattle trailers, Sarabyn severely limited his ability to receive and process information. See Ishimoto at B-14. Furthermore, when Sarabyn arrived at the Compound, he was pinned down and was unable to change the SRT instructions in light of the markedly changed circumstances. The ramifications of this leadership breakdown were substantial. Sarabyn, for example, never had an opportunity to communicate with the New Orleans team members when they took heavy fire while trying to secure the arms room. A commander in the undercover house would have known from hearing the extensive gunfire that weapons had already been distributed and that the New Orleans team's objective was significantly less vital. It may have been difficult under any circumstances to divert the New Orleans team members quickly to another task once they were committed. However, Sarabyn's poor vantage point, from which he could not see the New Orleans team, together with the lack of any preraid contingency planning, effectively precluded any such change in direction.

The only commander placed at a vantage point that allowed him to maintain the kind of perspective over an operation so critical to effective command and control was Cavanaugh. Cavanaugh was in the best position to make the final decision as to whether the raid should go forward. From his post at the undercover house, he was able to maintain open communication lines with diverse elements in the field, and he could himself keep an eye on the Compound to watch for any changes in conditions. This is indeed the place where the raid's overall commander should have been. See Ishimoto at B-14; Kolman at B 63; Morrison at B-88. However, Cavanaugh was not the overall commander. He was simply given responsibility for monitoring the Compound and aborting the raid if he saw signs that the Branch Davidians were laying an ambush. And, through no fault of his own, he was not well equipped to perform even this limited role properly. He was not particularly familiar with the normal routines of the Compound or the tactical details of the raid plan. In addition, Cavanaugh held a lower rank in ATF's hierarchy than Chojnacki, and lacking Sarabyn's status as a ASAC in Chojnacki's division, Cavanaugh could not be expected to be aggressive about calling the operation off if necessary.

Even had ATF deployed its raid commanders to positions more conducive to deliberate thought and an exchange of views, they would still have needed access to information on which to base their decisions, and aid in assessing that raw information. Pursuant to the NRP, an agent was designated to serve as the intelligence coordinator, but the sole reason for this designation seems to have been a desire to comply with the NRP. The designated agent was told nothing about the investigation or the tactical planning until he arrived in Houston for a briefing in mid-February. When he arrived in Waco, a few days before the raid, he was assigned the job of writing an operations plan, which kept him occupied until raid day. He was specifically told not to worry about intelligence matters, because they purportedly had already been dealt with. Indeed, in the days before the raid, he had difficulty contacting any of the raid commanders and never conferred with them about the intelligence operation. On February 28, the intelligence coordinator spent the crucial time immediately before the agents got underway driving from the command post to the staging area to signal the agents' departure, rather than being available to process intelligence information.

Had provision been made for a knowledgeable and involved intelligence coordinator, familiar with the conditions needed for the raid to succeed and charged with ascertaining whether they were present, he or she would have been able to brief the raid commanders on the significance of Rodriguez's report from the Compound. The coordinator might also have focused the raid commanders' attention on the reports from the forward observers in the undercover house, who saw no activity at the Compound and no men in the pit, but noted the presence of many media representatives roving near the Compound.

An intelligence coordinator might also have recognized that the forward observers assigned to the operation were a vastly underutilized intelligence resource. The purpose of ATF's new forward observer program was to put agents in the field before a mission to gather and provide raid commanders with current intelligence, and thereafter to give defensive coverage for those agents executing warrants. But forward observers were not meaningfully represented in the raid planning sessions, and the plan produced reflected their lack of participation. Even though forward observers recommended that enough teams be deployed to cover the Compound's entire perimeter, particularly the vulnerable position of the New Orleans team, the plan called for forward observers to be deployed to only two areas--one near the hay barn and the other at the undercover house. And the observers assigned to the hay barn moved from there to their position less than an hour before the arrival of the SRTs: too late for them to provide any meaningful intelligence to the raid commanders, too late for them to avoid crossing paths with cult member David Jones on the road, and too late for them to see cult members advance out of the Compound and occupy concealed positions from which they fired once the shoot-out began.

ATF's neglect of the use of the forward observers was not just an intelligence failure. See Morrison at B-90. The ATF plan also did not establish a common understanding among the raid commanders and the forward observers regarding the rules of engagement for the forward observers. As a result, there was no coordination between the forward observers and the agents in the SRTs. Although a forward observer did initially fire a shot at a clear threat in a window, this lack of coordination led to a delay of several minutes before the forward observers directed fire at the many other Branch Davidians who were shooting at the agents attempting to execute the warrants. See Ishimoto at B-14; Kolman at B-51.

The planning failures in the Waco raid stemmed in large part from an assumption on the part of ATF's leadership and those given specific planning responsibilities that an operation involving more than 100 agents against an extremely well-armed group of hostile cult members was just like any other enforcement action, only bigger. Lacking experience or training in raids like the one contemplated against the Branch Davidians, the planners assumed that what had worked for them in so many smaller operations would work again. The result of ATF's failure to support and guide this dedicated and well-intentioned group was a plan that rested on unreliable intelligence and that made the agents sent against the Branch Davidian Compound far more vulnerable to ambush than they realized. Therefore, the agents were unprepared to deal with the ambush when it occurred.

Section Three: Media Impact on ATF's Branch Davidian Investigation
The media's interest in covering suspected criminal conduct and official responses to it will frequently be at odds with law enforcement's desire to have the advantage of surprise in its activities. However, the two sides generally accommodate each other partly out of necessity and partly out of each side's respect for the mission of the other. No such accommodation was reached at Waco. During their parallel investigations, both ATF and the media missed opportunities to take actions that might have averted the tragedy of February 28. Because the institutional tensions between law enforcement and the media are inevitable and perhaps necessary, this section underscores that there is more at stake than law enforcement's right to enforce federal criminal law and the media's right to get the story.

ATF's Efforts to Delay the Publication of the "Sinful Messiah" Series
Early in ATF's investigation of alleged criminal activity at the Branch Davidian Compound, Special Agent Davy Aguilera learned that the Tribune-Herald also was investigating Koresh. The two investigations continued on their separate courses for some time, with ATF trying to conceal the extent of its interest in the Branch Davidians. When Aguilera learned that former cult member Marc Breault was providing information to Mark England, a Tribune-Herald reporter, Aguilera asked Breault to stop speaking to the newspaper; it appears that Breault complied with this request. By January 1993, however, ATF's tactical planners began to fear that the Tribune-Herald's publication of its series about Koresh would interfere with, or at least complicate, the agency's plans to execute warrants at the Compound. It was thus on February 1 that Chuck Sarabyn and Earl Dunagan met with Barbara Elmore, the Tribune-Herald's managing editor, at the U.S. Attorney's Office in Waco, and asked her to delay publication of the Davidian "Sinful Messiah" series until ATF could complete its operation. In the course of this meeting with Elmore, and subsequent meetings with Tribune-Herald personnel, Sarabyn, Dunagan, and, later, Phillip Chojnacki, disclosed not only ATF's intent to take action against the Compound, but also the anticipated date of that action.

Nothing in the agency's formal guidelines at the time barred this kind of media contact or even addressed it. The question remains, however, whether ATF exercised good judgment in initiating contact with the Tribune-Herald. Informed opinion differs on this point. Indeed, one of the tactical experts consulted by the Review wrote that ATF's efforts to obtain press cooperation violated basic principles of operational security. See Kolman at B-76. On balance, though, the Review believes that ATF made a reasonable judgment call in deciding to contact the Tribune-Herald. The danger that the Tribune Herald would betray ATF's confidences might well have been outweighed by benefits to the agency if the newspaper's series could be delayed or, at the very least, if the publication dates were known. At the time ATF made its decision, the Tribune-Herald already had some knowledge of ATF's investigation and, in any event, ATF always retained the option to cancel or postpone the raid.

Given the resources the Tribune-Herald had committed to this investigation, ATF could hardly have requested the paper's forbearance in general terms. Nor could ATF reasonably ask the Tribune-Herald to withhold publication indefinitely. As a result, discussions between ATF and the paper were unlikely to be successful unless they reached some level of specificity.

Unfortunately, ATF did not adequately consider who should represent ATF in its negotiations or what strategy would best serve their desire to have the Tribune-Herald delay publication. ATF's media policy permits, and indeed encourages, local field offices to handle media issues without requiring headquarters involvement except when the national media are involved or ride-along requests for the media are concerned.

Chojnacki's selection of Sarabyn and Dunagan to represent ATF, though consistent with ATF guidelines, exposes a structural defect in the guidelines. Although Sarabyn had taken ATF's general media relations course required for all supervisors, neither he nor Dunagan had any specialized expertise in media relations. Nor, for that matter, did Chojnacki. A field office should not be permitted to initiate contact with media about ongoing criminal investigations without formal participation or approval by ATF headquarters officials, particularly in an operation of the scope at issue here.

The point is not one of hierarchy, but of experience. During the course of his dealings with the newspaper, Chojnacki did seek advice, though only informally, from Jack Killorin, ATF's director of public affairs. But neither Killorin nor anyone else in the headquarters public affairs office was told that Sarabyn and Dunagan already had met with the newspaper or that Chojnacki had invited Cox officials to observe raid training. ATF headquarters might not have approved the raid if it had known that ATF had revealed possible raid dates to the Tribune-Herald early in the negotiations. Moreover, learning that the ATF Houston division had given up so much information without extracting anything in return, might have led headquarters to become involved in the negotiations.

Even though relations between the media and law enforcement can on occasion be adversarial, accommodation can bring important benefits to each side. A law enforcement agency's ability to inform the public about its achievements and to deter future offenders may depend upon the cooperation of the press. The media's ability to cover stories about criminal activity and to be protected adequately while covering such stories may depend on their having a working relationship with a law enforcement agency. These working relationships develop over time and are often are based as much on personalities as on institutional needs. Here, there was no such relationship, no history of previous dealings between ATF and individual agents on the one hand and the Tribune-Herald and its parent organization on the other. ATF agents arrived in Waco and simply asked newspaper representatives to give up something of tremendous value to the paper--the opportunity to expose a local problem with independent, in-depth coverage. Were the paper to delay its series until after a raid had exposed the Compound's activities, the story would lose its exclusivity and its profitability. In exchange, the agents offered the paper security advice, advice that any law enforcement agency would give freely if asked and a chance to watch raid training at Fort Hood, an event with minimal news value.

Had Chojnacki entrusted the press negotiations to those in ATF with more experience in media relations, an arrangement that would have been more suitable to ATF and the Tribune-Herald might have been made. ATF's representative in these negotiations might also have been more attuned to the kinds of arguments more likely to persuade the newspaper, arguments specifying the harms that could flow from publication of the "Sinful Messiah" series, instead of vague talk about how Koresh might become agitated.

And even if those responsible for press relations at ATF could not present better arguments to convince the Tribune-Herald to delay publication, they may have recognized the benefits of seeking assistance from local law enforcement officials in these negotiations, officials with whom the newspaper had a working relationship. Here, ATF's failure even to explore coordination with local officials increased its vulnerability to local conditions. The local U.S. Attorney's Office, which was instrumental in setting up the first meeting between ATF agents and Tribune-Herald representatives, might also have played a larger role in the negotiations.

Chojnacki's fervent pitch to the Tribune-Herald on February 24 gave the newspaper confused signals. ATF wanted the paper to delay publication of the "Sinful Messiah" series because publication might alert Koresh that some sort of enforcement action was imminent. But Chojnacki, seeking to conceal the precise timing of the raid, gave no indication that ATF would be definitely launching a raid or that if there was to be a raid, it would come soon. Chojnacki in fact suggested that ATF might have to "go home" if he could not get a warrant.

If an ATF representative with more media relations experience and no critical role to play in the coming raid had been responsible for ATF's negotiations with the Tribune Herald, ATF might have pressed its case beyond the February 24 meeting, perhaps with executives at Cox Enterprises. Chojnacki, however, was understandably preoccupied with his responsibilities as overall raid commander. When his presentation at that meeting failed, Chojnacki decided not to negotiate further with the newspaper.

Media Activity Raid Day
Although the flaws in ATF's efforts to delay the Tribune-Herald series require attention if ATF is to be more effective in future negotiations with the media, it may be that no overtures, however skilled, would have convinced the newspaper to delay its series. What cannot be dismissed, however, is ATF's failure to consider how the emergence of the Branch Davidian Compound as a focus of media attention after publication of the first articles of the "Sinful Messiah" series could affect conditions in and around the Compound on raid day and to take precautions to minimize the possibility of media disruption.

By daybreak February 28, Tribune-Herald reporter Tommy Witherspoon's informant had alerted him to the timing of the raid. Tribune-Herald executives had seen helicopters landing at the Texas State Technical College airfield and had interpreted them correctly as a sign that an ATF raid was imminent. AMT ambulance dispatcher Darlene Helmstetter disclosed details about a pending law enforcement operation to her friend Dan Mullony, a cameraman at KWTX.

Based on what they deemed to be reliable information, KWTX and the Tribune Herald decided to send a total of 11 of their personnel (three from KWTX and eight from the Tribune-Herald) to the Compound vicinity to cover the raid. The reporters arrived at the scene early and travelled up and down the roads around the Compound as they prepared to cover the story. One of their number, KWTX cameraman Peeler, became lost, and, in asking for directions, unwittingly tipped a cult member that a raid was imminent. Another group of reporters went to a house directly across from the Compound and asked for permission to watch ATF's enforcement action, without taking any precautions to ensure that these neighbors would not in turn alert Koresh to the impending raid. Many media personnel used cellular phones--unsecure communication devices whose signals are capable of easy, although illegal, interception.

The foregoing actions, which were taken by representatives of news organizations aware that Koresh and his followers were suspected of stockpiling weapons and manufacturing illegal firearms and explosives, belie the claim recently made by the Society of Professional Journalist's Waco Task Force that both KWTX and the Tribune-Herald "took precautions to prevent any alerting of the Davidians." (Report at 6.) The extent of those precautions consisted only in using unmarked vehicles in the Compound's vicinity.

The Society of Professional Journalists' Waco Task Force makes another claim that bears mention here. According to its report, the Task Force "found no concrete evidence validating the accusations that journalists from the newspaper or the television station tipped off the Branch Davidians as to what was happening." (Report at 6.) In contrast to this claim, James Peeler has admitted to the Review that he told someone later identified as David Jones that a law enforcement action would soon take place at the Compound. It is undisputed that Jones took this information and alerted Koresh. But however tragic the results of his carelessness may have been, Peeler should not be made the scapegoat for the fact that Koresh learned of the raid. Given the extent of other obvious media activity in the area, had Koresh not learned of the raid from Peeler, he might just as easily have been placed on guard by that other activity.

The prospect of substantial media activity in the area, and the dangers such activity could pose to the raid, should have been clear to ATF's raid commanders. They knew that ATF had been telling a newspaper for some time that a raid was imminent. And they knew that the appearance of the first installment of the "Sinful Messiah" series the day before had trumpeted the offenses going unprosecuted at the Branch Davidian Compound. Finally, they knew the paper was considering Koresh's request to send a reporter into the Compound Saturday or Sunday. From just this information, ATF should have foreseen the possibility that media personnel, or mere gawkers, would be in the vicinity of the Compound on February 28. Indeed, for all the agency knew, one of the Tribune-Herald's rivals could have sent a reporter in to get Koresh's reaction to the "Sinful Messiah" series.

As to the danger posed by such media activity, ATF should have recognized that Koresh's reported hostility to strangers could only have been increased, especially as to some media personnel, in the wake of the "Sinful Messiah" series. The presence of unwanted visitors on or near the Compound on February 28 might thus have triggered a hostile response that would interfere with the ATF raid whether or not ATF had the advantage of surprise. Increased activity around the Compound also could have impeded ATF assault by forcing agents to take care that reporters did not become hostages or casualties.

ATF did set up roadblocks around the Compound shortly before the raid. Press vehicles, however, already had begun patrolling the area. Setting up roadblocks too early would have its own risks, because roadblocks could have tipped off the Branch Davidians that some sort of enforcement action was imminent. More importantly, even if ATF had opted not to use roadblocks, the agency could have been far more sensitive to press activity in the area by identifying possible press vehicles and keeping them under surveillance the morning of the raid.

Had ATF attempted to monitor media movements in the area, it might have prevented KWTX cameraman Peeler from ever speaking to cult member David Jones. At the very least, ATF would have recognized the significance of Jones racing to the Compound after his conversation with Peeler, thus reinforcing Rodriguez's report that Koresh had been tipped off.

Media activity in the vicinity of the Compound was not the immediate cause of the casualties suffered by ATF agents on February 28. These were inflicted by Koresh and his followers, and could have been avoided had ATF's raid commanders called off the operation once they recognized that they had lost the advantage of surprise. But the media's conduct posed a substantial danger not only to the security of ATF's operation but also to the lives of agents and civilians alike. While it is not the purpose of this report to suggest what the media might do to minimize such dangers in the future, the media should further examine its conduct near Waco on February 28.

Section Four: The Flawed Decision to Go Forward With the Raid
On February 28, Koresh and his followers knew ATF agents were coming and decided to kill them. That the Branch Davidians, if forewarned, would try to lay such an ambush, however, should not have come as a surprise to those who planned the ATF operation. Indeed, the extraordinary danger posed by Koresh's arsenal and his propensity for violence were the reasons enforcement action was necessary. The issue addressed here is why ATF's raid-day decisionmakers proceeded with the raid, even though they should have realized--and indeed did realize--that they had lost the element of surprise, which was so critical to the raid plan.

The decision to proceed was tragically wrong, not just in retrospect, but because of what the decisionmakers knew at the time. Surveillance certainly indicated that something was amiss. There was none of the usual activity outside the Compound, and agents had seen David Jones, a known cult member, speeding toward the Compound after his conversation with one of the many media personnel who had begun to congregate in the vicinity. And there was no need to speculate about what Jones might have told Koresh. Once Rodriguez was able to report back to the command post, the key decisionmakers had to know that Koresh had been tipped off that ATF was coming. Why, then, did no one at ATF call off the raid?

The answer to this question lies in a complex set of factors that include the failure of the raid-day decisionmakers to adequately assess available information at the time of decision, the failure of those decisionmakers to appreciate the tactical significance of losing surprise close to an hour before the raid was to begin, serious deficiencies in the raid-day intelligence gathering and processing structure, and the placing of decisionmaking authority in the hands of individuals who lacked the requisite training and experience. In the end, this is less a story of wrong choices made than one of choices not made at all as the momentum of the massive operation--left unchecked by the raid commanders and ATF management--carried it inexorably forward, with speed substituted for reflection and inquiry.

ATF Decisionmakers Understood in Advance that the Raid Had Likely Been Compromised
Despite contrary public statements made by ATF officials in the days and weeks following the raid, it is now clear that the critical decisionmakers on February 28--Chojnacki, Sarabyn, and Cavanaugh--had sufficient information from Rodriguez to conclude that the raid had been compromised. They knew that Koresh had become upset and agitated after leaving to take a purported telephone call, proclaiming that neither the ATF nor the National Guard would ever get him, and commenting: "They're coming, Robert, the time has come. They're coming." Koresh's reference to the National Guard was particularly significant. Koresh had previously expressed hostility to ATF in Rodriguez's presence, and talked of ATF's coming to get him, but never before had he referred in this way to the National Guard. His reference to the Guard, which was indeed participating in the raid, was strong evidence that Koresh had specific information about the impending operation. In addition, Rodriguez told Cavanaugh and others in the undercover house that "Koresh knows we're coming," and, according to Sarabyn, the first thing Rodriguez told him on the phone was "Chuck, they know we are coming."

The actions and statements of Sarabyn, Chojnacki, Royster and Cavanaugh, after hearing Rodriguez's report, strongly suggest that they not only had reason to believe, but in fact did believe, that the raid had been compromised. Their solution was to hurry up. After his telephone conversation with Rodriguez, Sarabyn related its substance to an agent in the command post. When asked what he planned to do, Sarabyn drew comfort from Rodriguez's having left Koresh reading the Bible, with no firearms in sight, and he opined that the agents could still execute the plan if they went quickly. Raid preparations immediately moved into high gear. Sarabyn, Chojnacki and Royster had a brief discussion on the tarmac, where Sarabyn related his conversation with Rodriguez and offered his thought that if they hurried they could still do the raid. After that conversation ended in agreement to go ahead with the operation, Chojnacki and Royster hurried into the command post. Chojnacki called the National Command Center in Washington to say the raid was going forward, and they both rushed back to the helicopters. Royster told various raid personnel "They know we're coming," and expressed the need to hurry. Sarabyn rushed to the staging area, several miles away, and, on arriving, repeatedly exhorted the agents there to hurry up and "get ready to go, they know we're coming." Cavanaugh, though he had no place to rush to, commented to others in the undercover house, "We better do this ASAP."

Sarabyn and Cavanaugh concede fearing the raid had been compromised before it began. Royster likewise acknowledges that he understood that ATF had lost the element of surprise. In contrast, Chojnacki maintains that Rodriguez's report did not lead him to this conclusion, since he felt that Koresh's statements, as relayed to him, were not materially different from what Koresh had been saying to Rodriguez all along. Chojnacki, like Sarabyn, however, appears to have interpreted Koresh's statements as significant enough to accelerate the raid's timetable and get agents to the Compound ahead of schedule.

The Lack of a Control Agent
If any of the raid commanders, for whatever reason, found ambiguity in Rodriguez's report from the Compound, the fault lies not in Rodriguez's report--which was quite clear--but in ATF's failure to assign Rodriguez a control agent who could have obtained more details and, even more importantly, ensured that the undercover's information was understood.

The failure to give Rodriguez a control agent, not only on the day of the raid but also during the weeks before the raid, was an unusual departure from standard law enforcement practices. When an undercover agent makes repeated contacts with a target, particularly a target with a propensity toward violence and a powerful and influential personality, the agent must have substantial support. A control agent helps keep the undercover agent comfortable in that role, and attends to the undercover's needs, both physical and psychological. The control agent can also serve as a conduit for the undercover's information. Sarabyn, who was otherwise preoccupied and at a remote location, was certainly not the person to debrief Rodriguez, who was shaken by his experience in the Compound. This was a job that should have been handled in a face-to- face session with an otherwise unburdened control agent who knew Rodriguez well. Similarly, unlike Sarabyn, who focused his prepared questions on whether the cult members had openly exhibited weapons to Rodriguez or taken visible steps to resist law enforcement, a control agent could have patiently and objectively questioned Rodriguez about both the content and his impressions of his exchange with Koresh. More details on this score might have made the raid commanders recognize that ATF agents might face an ambush at the Compound.

Other Intelligence that Could Have Confirmed Rodriguez's Report that Koresh Knew ATF Was Coming
Even if Rodriguez's report did not convince the decisionmakers that the operation had been compromised, it should at least have led them to make further inquiries to determine whether Koresh's heated references to ATF, the National Guard, and their coming to get him were something more than an extraordinary coincidence. Such inquiries might well have caused them to abort the raid, since Koresh's behavior takes on a special significance when seen against the background of cult member David Jones' encounter with KWTX cameraman Jim Peeler. Although the agents in the undercover house did not know what Peeler told Jones, they did tell Cavanaugh that a known cult member had sped back to the Compound (while Rodriguez was still inside) after a conversation with someone they thought might be one of the many media personnel in the area. Cavanaugh believes that he, in turn, relayed this information to the command post. Had anyone stopped to consider it, this information might well have shed critical light on the specificity of Koresh's assertions and made it even clearer that the raid had been compromised.

The failure to evaluate properly the intelligence from the undercover house should not be seen simply as a product of undue haste, however. It is not enough for raw surveillance information by forward observers to be relayed to operational commanders. There must be some system for gathering such pieces, putting them together, and ensuring that a meaningful evaluation gets presented to those who need it. As discussed above in the analysis of tactical planning, ATF's National Response Plan called for an intelligence coordinator, and an agent was assigned to that role. Unfortunately, through no fault of his own, that agent was given virtually no intelligence coordinating responsibilities, and had none at all on the day of the raid. Instead, surveillance was coordinated by Cavanaugh, who, unfamiliar with the day-to-day routines at the Compound, viewed his job as consisting largely of relaying information to the command post, rather than independently assessing it. Cavanaugh had no single contact at the command post, and no one had responsibility for gathering, integrating and assessing all the various intelligence inputs. While, at some level, Sarabyn undertook the intelligence evaluation role, there was no structure to ensure that he received all the raw intelligence data. Moreover, given the other responsibilities he had assumed, he could not possibly have performed that role adequately.

As a result of the flawed intelligence structure, while Cavanaugh, the other agents in the undercover house the morning of the raid, and possibly someone at the command post knew about the Peeler-Jones encounter, no one put it together with Rodriguez's report. Similarly, no one in the raid's command structure saw anything suspicious in the media activity all around the Compound that morning. As Cavanaugh later explained, everyone had become desensitized to the media's presence, assuming that reporters were just following up on or reacting to the publication of the first two "Sinful Messiah" articles. Raid commanders took comfort in the fact that substantial traffic had been reported in front of the Compound on Saturday as well. As it happens, however, at least some of the media personnel in the vicinity of the Compound on Saturday had come because they had been tipped off about the raid and wanted to look the premises over in anticipation of coverage the next day.

More attention to the acquisition and flow of raid-day intelligence, coupled with better technical support, might also have led agents to obtain what could have been the most concrete evidence that Koresh was planning an ambush. Because ATF suspected that cult members were using amateur radio equipment, the undercover house had been outfitted with a scanner for monitoring radio traffic. Given the range of frequencies on which the cult members might have been operating and the limitations of radio scanners, it is not particularly surprising that the agents in the undercover house were unable to pick up any traffic from the Compound, and that no efforts were made to use the scanner on the morning of the raid. More sophisticated, but widely available, monitoring equipment, however, would have greatly increased the chances of overhearing Compound radio traffic. And there appears to have been radio traffic inside the Compound that morning. Two area residents overheard radio communications among people they later believed to be Compound residents. The Compound residents described approaching ATF agents as looking like "a covey of quail," and one said "If I had a shotgun I could flush them out and kill every one of them." Shortly thereafter, the scanner picked up the sound of gunfire. Had ATF been monitoring that same conversation, even the decisionmakers undeterred by Rodriguez's report might have recognized the need to abort the raid.

Decisionmakers Failed to Realize Unacceptable Risk of Proceeding Without Surprise
The chief reason why Rodriguez's report did not lead ATF's decisionmakers to call off the operation, or even to make further inquiries into whether Koresh had indeed been tipped off, appears to be that they did not appreciate that surprise itself was absolutely critical to the operation's success. Sarabyn and Chojnacki recall that, for them, the determining issue was not whether Koresh would be surprised, but whether the Branch Davidians were arming themselves in anticipation of ATF's arrival. That this was indeed their concern is suggested by the questions that they asked Rodriguez upon his return to the undercover house on February 28. Although Rodriguez had been sent into the Compound to see if the Saturday and Sunday Tribune-Herald articles had led the Branch Davidians to take up arms or otherwise vary their routine, he emerged with information of far more direct importance to the ATF operation. But the decisionmakers stuck to the questions that had been prepared earlier, asking what Koresh was wearing, whether the Compound's gates were open, and whether anyone in the Compound was armed. On hearing that Rodriguez had seen no weapons in the Compound, the decisionmakers decided that they could still succeed so long as they hurried up the raid and got agents to the Compound before conditions changed. Should Koresh mobilize his followers while the agents were en route, Chojnacki and Sarabyn assumed that they would learn of the danger from the forward observers positioned in the undercover house with sights on the Compound, and could abort the raid if necessary.

Chojnacki's and Sarabyn's calculations apparently rested on two false premises: first, that Koresh would not mobilize his followers as soon as he learned that agents were coming; and, second, that if an ambush were prepared, signs of it would be visible to the forward observers more than 250 yards away.

This underestimation of Koresh's resolve was inconsistent with the intelligence that had been amassed during ATF's investigation. Those familiar with Koresh's stockpile of weapons, ammunition, and explosives, his increasing propensity toward intimidation and violent rhetoric, and his prior statements expressing extreme hostility to the ATF, could have predicted how Koresh might react to a tip that the ATF and the National Guard were coming. Former cult member David Block had told Aguilera that he had left the Branch Davidian because Koresh would always remind his followers that if they were to have a confrontation with the local or federal authorities, the group should be ready to fight and resist. In light of the information provided by Block, Koresh's statement to Rodriguez that "the time has come" was also a strong indication that something Koresh had planned for was about to happen.

It is true that Chojnacki and Sarabyn lacked the firsthand or secondhand familiarity with Koresh that Rodriguez and Aguilera had, and therefore were less able to predict how Koresh would react to a tip about the raid. But they never turned to anyone for help. Instead, they asked Rodriguez only about whether he had seen defensive preparations, and they never made any inquiries of Aguilera or of the raid plan's other architects. Had they done so, they would have better understood how these new facts jeopardized a plan that depended entirely on the advantage of surprise. The Compound's structure, the firepower that Koresh had amassed inside, the loyalty and discipline of cult members, and the absence of cover in the surrounding terrain made a direct assault against forewarned assailants unacceptably risky.

Instead of seeking such counsel, the raid commanders thought they should hurry up. This, too, made no sense. If Koresh was not going to arm and deploy his followers, there was no need for haste. The raid commanders could follow the original plan and wait for the Branch Davidian men to begin their work in the pit, away from their weapons. If the men did appear, the forward observers or ATF's fixed wing aircraft would be able to tell the raid commanders. If, on the other hand, Koresh was going to resist the agents, any acceleration of the raid would again not help. It would still take at least 30 minutes from the time Sarabyn left the command post for the cattle trailers to get from the staging area to the Compound. This delay would give Koresh more than enough time to hand out weapons and deploy his followers in a Compound that appeared to be designed for just such defensive measures. And it was scarcely likely that anyone stationed outside the Compound would be able to tell that an ambush was being prepared. Cult members with access to machineguns and semiautomatic assault weapons should not have been expected to display their weapons out the window while they lie in wait.

Perhaps one explanation for why the raid commanders underestimated the ability and resolve of Koresh and his followers might be that they overestimated ATF's ability to intimidate their target simply by arriving at the Compound in force. No decisionmaker has said that he acted in the belief that Koresh would back down in the face of ATF's show of force. Several ATF raid participants, on the other hand, have said they never thought the Branch Davidians would fire on scores of uniformed agents. Such statements betray an insensitivity to the volatility of the situation that ATF should have known it was entering. Given that a small segment of the population harbors extreme hostility both to ATF and the federal laws it is charged with enforcing, the agency must always be wary of violent responses from the targets of its investigations. And here, Koresh's pronouncements left no need to speculate about his hatred of the agency and the apocalyptic violence with which he would greet its agents.

The narrow answer to why the raid was not called off Sunday morning is that ATF decisionmakers failed to realize that surprise was critical to the operation's success and why it was so critical. Looking only for indications of defensive measures that were unlikely to be seen by the forward observers, the commanders never paused to reflect on the consequences of Koresh's having been tipped off. They hurried up when they should have slowed down. This narrow explanation, however, is incomplete, for it must be understood against the backdrop of the momentum inevitably generated in an action of the type contemplated by ATF, and the upper-level management decisions within ATF that exacerbated the pressure imposed by that momentum.

Handling the Momentum of the Raid
Most major law enforcement actions develop momentum as the moment of execution approaches, particularly raids that are viewed as high risk. Anxiety, fear, bravado, and the desire to accomplish the raid's objectives combine to put pressure on the raid participants to go forward. As the point of no return approaches, the pressure to go forward increases.

A raid of the scope, expense, and logistical complexity contemplated by the planners of the Waco operation can generate a momentum that, if unchecked, can be inexorable. By the time Rodriguez left the Compound on Sunday morning and reported to Sarabyn, all was poised to go forward. The eight-month investigation had generated probable cause to believe that Koresh and his followers had amassed an enormous stockpile of weapons, ammunition and explosives. Raid planning had been in the works since early December. Over 130 ATF agents were in or near Waco for the raid. Three ATF special response teams and three arrest support teams, comprising 76 highly motivated agents with a common mission, had been training and living together for three days at Fort Hood. Dozens of other ATF agents had been brought in to participate in and support the raid. The agents had been drawn from seven different ATF field divisions and 18 different cities, and could not be kept in the Waco area indefinitely. National Guard and emergency medical personnel and equipment were in place, as were members of other federal, state and local law enforcement agencies. ATF headquarters personnel were assembled at the National Command Center to receive up-to-the minute information about the raid. No one associated with the venture could have doubted the fantastic cost and effort it would take if the operation were aborted and put off for another time.

It is difficult to measure what effect the operation's built-in momentum may have had on the raid-day decision to go forward. Several ATF agents who participated in the raid have strongly suggested that the raid probably would have been aborted but for this pressure. At the very least, the pressure to go forward might well have played a critical role in the failure of the raid commanders to seek more information from Rodriguez, Aguilera, and raid planners about the significance of Rodriguez's report from the Compound. Decisions that now appear flawed may well not have been decisions at all, but simply steps taken along what seemed at the time to be a preordained road.

The point is not that momentum must be avoided in large-scale raids. Surely it is inevitable, and may even be beneficial to the success of an operation. Such pressures, however, should not be permitted to infect the decisionmaking of those charged with giving the go-ahead for what, at best, is a high-risk endeavor. Many law enforcement agencies provide training in crisis management to those supervisory personnel likely to face high-risk situations where alternatives must be weighed under extreme pressure. ATF gave its supervisory agents no such specialized training.

The absence of such training was particularly unfortunate for the decisionmakers here. Neither Chojnacki nor Sarabyn had any experience remotely comparable to the raid attempted on February 28. The bulk of their experience was with typical street enforcement actions. Nor had they had any meaningful training in operations of this magnitude, or any relevant military tactical experience that might have compensated for that lack of training. As a result, they were ill-prepared for the command of a large-scale, high-risk assault on a large, heavily armed structure.

The pressures on Sarabyn were particularly great. Indeed, it is questionable whether any training could have prepared him for the many responsibilities that he took on. He served as the manager of the investigation, the manager of the tactical planning, and the de facto intelligence coordinator. He was also a principal liaison between the field agents involved in the raid and ATF headquarters, obliged to respond to high-level inquiries, prepare numerous reports on all aspects of the investigation and operation, and give briefings. Sarabyn's responsibilities continued to build as the investigation and tactical planning progressed. February 28 was to be the culmination of all of these efforts, and it would be understandable if Sarabyn was reluctant to postpone the long-awaited event and lacked the dispassion so critical to crisis management.

The pressures on the raid's commanders could only have been increased by the absence of any meaningful contingency planning for the raid. When presented with Rodriguez's report, they considered but two choices: proceed with the raid as planned or call it off. A third alternative that is always available, delay, apparently was not considered. Had the raid planners prepared a more productive tactical alternative, such as deploying for a siege in the event surprise were lost, the pressure to proceed with the raid would have been substantially eased. A range of meaningful options, considered and practiced in advance and accounting for the real possibility that surprise might be lost, almost certainly would have improved the crisis decisionmaking process.

The final check on the pressures of momentum faced by those in the field can come through monitoring by headquarters personnel, far removed from the scene. But no meaningful monitoring occurred here. On February 28, pursuant to the National Response Plan, ATF headquarters activated the National Command Center in Washington to follow the progress of the raid. Staffed as it was with high-level managers who had extensive experience in field operations, the Command Center could have served as a valuable check on the decision to go forward with the raid. Had the Command Center been briefed on what Rodriguez had learned inside the Compound, the raid might well have been aborted. At the very least, Command Center personnel might have recognized that caution and careful thought, not speed, was the appropriate response to Rodriguez's information. Instead, when Chojnacki called the Command Center after being briefed by Sarabyn, he said only that the raid was going forward, and made no mention of Rodriguez's report. Moreover, the person he spoke to was neither a superior nor someone with particular knowledge of the operation. No questions were asked, no further information requested. As a result, the Command Center involved command in name only. In reality, it served as a front-row seat for what everyone anticipated would be a major, successful operation. In that capacity, the Command Center simply reminded the decisionmakers in the field that headquarters was watching, and it could only have added to the pressure to go forward with the operation for which all had waited so long.

ATF Notifies Treasury of Impending Operation
As the planning for the raid on the Branch Davidian Compound entered its final stages, Associate Director (Law Enforcement) Daniel Hartnett, at the direction of ATF Director Stephen Higgins, asked Special Agent Christopher Cuyler, ATF's liaison to the Treasury Department, to brief the Assistant Secretary of the Treasury for Law Enforcement about the impending operation. Although ATF falls under the jurisdiction of the Assistant Secretary--as do the U.S. Secret Service, U.S. Customs Service, Financial Crimes Enforcement Network, and Federal Law Enforcement Training Center--no regulation then in force required ATF to seek approval from the Office of Enforcement for the execution of search or arrest warrants, even for an operation of this magnitude. Indeed, no regulation required ATF even to notify the Assistant Secretary that such an operation was about to be launched. Nonetheless, it had been ATF's practice to apprise the Assistant Secretary of significant events, especially those expected to generate substantial media attention.

Accordingly, on the afternoon of February 26, Cuyler prepared a one-page memorandum for Michael D. Langan, then Acting Deputy Assistant Secretary for Law Enforcement, alerting him that the Branch Davidian Compound would be raided on February 28 by SRTs from Houston, Dallas, and New Orleans, assisted by state, local, and military authorities. The memorandum noted that "approximately 75 people (men, women, and children)" were thought to be in the Compound, but it provided no details about the planned operation. However, the memorandum did assure: "A well-reasoned, comprehensive plan has been approved [that] allows for all contingencies." (See Appendix D.)

Shortly after preparing his memorandum, Cuyler gave a quick briefing to Ronald K. Noble, who had been designated by President Clinton, through the Secretary of the Treasury, to be Assistant Secretary of the Treasury for Law Enforcement. Pending his nomination and confirmation, Noble worked in the Office of Enforcement as a part-time consultant. He was authorized to give advice, but had no authority, operational or otherwise, over Treasury Department personnel. Because of his advisory status and because he was occupied with the bombing at the World Trade Center in New York City, which had occurred hours before, Noble suggested that Cuyler brief Langan and John P. Simpson, Deputy Assistant Secretary (Regulatory, Tariff and Trade Enforcement), who was acting as Assistant Secretary for Enforcement.

At the briefing, Cuyler added little to the one-page memorandum, except to say that the operation had been moved up from Monday to Sunday in response to the anticipated publishing of the Waco Tribune-Herald series. Questions were raised among officials at the Office of Enforcement as to whether there were alternatives to an operation of such magnitude and they decided to discuss the matter further. Meanwhile, Langan alerted Philip Diehl, then counselor to Secretary of the Treasury Lloyd Bentsen, and Joshua Steiner, then special assistant to Deputy Secretary of the Treasury Roger Altman, to the fact that a raid was scheduled to take place the following Sunday near Waco. Steiner in turn informed Altman of the existence of the planned raid. Secretary Bentsen, who was in England, was not notified. When Simpson sought clarification about the purpose of the ATF briefing, Director Higgins explained that ATF merely wanted to keep the Office of Enforcement informed and was not seeking Simpson's authorization.

Friday afternoon, Noble had a discussion with Simpson, Langan, and Stanley Morris, the former Director of the U.S. Marshals Service, now working in the Office of Enforcement. Langan and Morris had serious reservations about the operation. Noble, who had been a Deputy Assistant Attorney General and Assistant U.S. Attorney at the Justice Department, agreed. He observed that Cuyler's memorandum had not addressed critical questions as to why so much force was needed to execute the warrants, what precise precautions were being taken to ensure the safety of ATF agents and those inside the Branch Davidian Compound, and why ATF believed it could achieve its mission without a shoot-out. Based on the information he had available at that time, Noble noted that if he were in Simpson's position, he would not let the raid go forward.

Simpson, Langan, Noble, and Morris also discussed the extent to which the Office of Enforcement should defer to the decision of a bureau within its jurisdiction to proceed with an operation about which the Office had reservations. Simpson decided that the Office was obliged to intervene and prevent the operation from going forward. Simpson called Higgins and directed that the raid be called off.

While Higgins offered no objection at the time, a half hour later he called Simpson and asked him to reconsider his order to call off the raid. First to Simpson alone, and then in a conference call with Noble and Simpson, Higgins explained that the warrants had to be executed forcefully because Koresh was not likely to surrender voluntarily and because ATF feared that Koresh and his followers might destroy evidence or commit mass suicide if given the opportunity. Higgins assured them that ATF was familiar with the routine in the Compound and that the raid had been scheduled for a time when the men would be separated from the women and children, who would be inside, and from the weapons, which were stored in an "arms room" under Koresh's control. Higgins stressed that the raid needed to go forward that Sunday because the Tribune-Herald series might alert Koresh that he was the subject of law enforcement scrutiny and lead him to alter his routine.

Higgins asserted that those directing the raid were instructed to cancel the operation if they learned that its secrecy had been compromised or if those in the Compound had departed from their established routine in any significant way. Higgins explained that an undercover agent would be sent into the Compound shortly before the raid to determine whether there had been any such changes in routine. At the conclusion of this three-way telephone call, Noble and Simpson said that they were satisfied that their concerns about the raid had been addressed. Simpson revoked his earlier direction that the raid not go forward.

The next day, February 27, Higgins informed Simpson that the first Tribune-Herald article had appeared, but that it did not indicate that any law enforcement action was imminent. Higgins reiterated that an ATF undercover agent would be able to confirm the next day whether the investigation had been compromised, and he felt confident that the raid could proceed as scheduled. Simpson advised Noble of Higgins's call. The Office of Enforcement heard nothing more about the raid until late in the morning of February 28, when Higgins informed Simpson that the raid had been repulsed and agents had been killed and wounded.

Discussion
Although the Office of Enforcement is formally charged with overseeing ATF, ATF gave Enforcement fewer than 48 hours' notice that it was about to embark on the biggest raid in its history. Moreover, the notice the agency did send was minimal, a one page memorandum giving little sense of the nature of the operation and its risks. The presentation seems to have been made more to enable the Treasury Department to field media inquiries after a successful raid was concluded than to allow it to rigorously review the raid plan; perhaps, it was simply intended to keep Treasury from hearing about the raid from the media first.

The procedure that ATF followed, however, was consistent with prior practice established by previous Assistant Secretaries and still in force in February 1993. Indeed, ATF was not required to give the Office of Enforcement any notice at all of the impending raid. Nor was there any system in place for the Office to make regular inquiries about significant ATF operations. ATF's Director and the Assistant Secretary for Enforcement met once a month. The Office of Enforcement relied on the discretion and good judgment of the Bureau's senior management for making day-in, day-out decisions, and gave the Bureau no reason to believe that any enforcement issue was to be identified for special scrutiny. Thus, the Office of Enforcement must itself bear some responsibility for ATF's failure to treat the operation differently and to give anything but minimal information about the impending raid on the Branch Davidian Compound.

When information about the raid was finally presented to it, the Office of Enforcement did seek to exercise some oversight authority. The raid would not have been permitted to proceed if Director Higgins had been unable to answer critical questions raised by the Office about the operation's necessity and its risks. Had the Waco raid commanders adhered to Director Higgins's assurance to Simpson and Noble that the raid would not go forward unless ATF had the advantage of surprise, the operation might have ended differently. The manner in which the Office of Enforcement was brought into this case, however, demands that some thought be given to the role the Office can and should play in ATF operations in the future.

Theoretically, the Office of Enforcement could choose to review the plans for every ATF operation and, indeed, for every Secret Service and Customs operation as well. To do so, however, would either turn the Office into a rubber-stamping operation or make enforcement activities by these agencies come to a grinding halt. In 1992 alone, ATF's more than 2,000 agents executed 10,134 federal warrants. In addition, they participated with state and local agencies in the service of 12,884 search warrants throughout the nation. Given the speed with which most enforcement activities occur and the degree of familiarity that is needed before an operation can be assessed, involvement by the Office of Enforcement in most ATF raids is impossible. For routine operations, the Office must rely on ATF leadership. Indeed, any micro-management in this regard would be inappropriate as well as inefficient, because those who plan these operations should feel and be responsible for them.

For certain significant enforcement operations, however, Treasury Department oversight is both realistic and appropriate. Where an extraordinarily large raid is being planned, the Office of Enforcement can provide a critical check on a process that tends to develop a momentum of its own. Even while leaving law enforcement agency planners considerable discretion over operational details, the Office can assess the risks being taken and bring its independent judgment to bear on sensitive issues, such as criminal activities by religious cults, that the agencies are not used to dealing with. Charged with enforcing criminal laws, law enforcement personnel will understandably have a tendency to look to enforcement solutions that may not always be appropriate. Civilian oversight can check this tendency, and ensure that agencies consider other approaches as well, if suitable. And where enforcement action is required, the Office of Enforcement can ensure that an agency consults and coordinates with other law enforcement agencies with special expertise.

It is difficult to craft precise rules and guidelines regarding when ATF should seek approval from the Office of Enforcement for an operation. The raid that later becomes the subject of congressional or media attention will not always seem worthy of special scrutiny before it happens. Moreover, each assistant secretary, as with any other manager, will doubtless have his or her own particular concerns. The most effective way to communicate those interests and clarify what "significant" means is not simply through formal rules, but also through a close working relationship between agency heads and the Office of Enforcement.

When presented with a law enforcement agency's plans for a significant operation, the Office of Enforcement must give due recognition to the expertise and experience of the agents who put the plans together. While the Office can bring a critical outside perspective to bear on sensitive issues, such as whether the public believes the level of force the agency plans to employ is justified by the violations targeted or whether adequate measures have been taken to protect agents and civilians, the Office must allow agents considerable leeway in deciding how those concerns should be addressed. At bottom, however, the Office must ensure that tactical objectives will be accomplished in a manner consistent with public expectations of fairness and proportionality and safety.

For the Office of Enforcement to play a constructive role in any operation, it must be brought into the picture while the operation is being planned. Otherwise, agents will see its review as just another last-minute bureaucratic hurdle to overcome. When Simpson and Noble expressed their concerns about the Waco plan on February 26 and put the operation on hold less than 48 hours before the raid was set to go, their intervention was unlikely to spark any meaningful ATF reassessment of the plan. By this time, there were pat answers about how the element of surprise could be preserved and how contingency plans had been prepared. With such reassurances given quickly, the raid could go forward. A broader and more precise definition of the Office's interests in this area is needed.

The tragedy near Waco exposed deficiencies in the way the Office of Enforcement oversees its bureaus' activities. The process of addressing these deficiencies is underway. First, the Office of Enforcement has established a Treasury Law Enforcement Council comprising the directors of the U.S. Secret Service, ATF, and Federal Law Enforcement Training Center, the Financial Crimes Enforcement Center, the Commissioner of Customs, and the Assistant Commissioner for the Criminal Investigative Division of the Internal Revenue Service. Its purpose is to provide Treasury law enforcement leaders a forum to discuss significant policy or operational matters with one another and with the Assistant Secretary for Enforcement. Working closely with the council, the Assistant Secretary for Enforcement has developed formal reporting requirements and crisis management procedures for the bureaus. Second, the Office of Enforcement has instituted weekly meetings with bureau heads to ensure that policy-level officials are provided with timely information to allow them to conduct meaningful oversight.

In the end, however, the Assistant Secretary for Enforcement must have confidence in the judgment of ATF leaders and be ready to allow them considerable discretion in selecting the means to accomplish a worthy objective. These leaders must in turn clarify the degree of discretion that field supervisors will be allowed and be able to respect the judgment of these field supervisors within their allotted spheres of expertise. With discretion, at every level of the agency must come accountability, because no level of oversight will prevent tragedy where--as here--a plan presented for Office of Enforcement review is based on assumptions and contains preconditions that are ignored by the agents charged with implementing them. Any individual whose judgment or integrity cannot be trusted by those who must rely on those qualities must be removed from a position of discretionary authority. The Assistant Secretary will work with ATF to ensure that these goals are met.

Section Six: Operations Security
Among the many questions posed publicly after the raid failed were those concerning how the Branch Davidians could have known that ATF was coming. Many theories were circulated by the media and within the Waco community. Some theories suggested that a reporter had telephoned Koresh shortly before the raid. Others believed that the visibility of numerous agents in the Waco area before the raid had alerted local residents, particularly hotel employees, waitresses, and patrons of bars and restaurants.

The Review investigated the various theories that attempted to explain how Koresh was warned of the impending raid. The precise answer to how Koresh was alerted is now clear: KWTX cameraman Jim Peeler told David Jones that a raid was imminent, and Jones quickly passed that information to Koresh. Koresh's reaction to the news, as described by Rodriguez, strongly suggests that Jones' warning about the operation was the first to reach him. Contrary to some initial reports, Koresh was not consciously tipped off by a reporter. Similarly, reports that Koresh got wind of the raid due to careless ATF operations security practices are unfounded. Witnesses who thought they had seen ATF agents in Waco before the raid were actually recalling post-raid events during which ATF and other law enforcement were highly visible.

Although Koresh learned of the raid through the chance encounter between Peeler and Jones, not by ATF action, there were many other actions taken in the course of the investigation that could have alerted Koresh to ATF's investigation or the timing of its entry plan. Although many of these actions were needed to advance the investigation, some actions needlessly risked raising Koresh's suspicions. This was particularly dangerous because maintaining the element of surprise was vital to the plan's success. Although some criticisms of ATF's security practices were unfounded, a few of the salient actions are examined here so that ATF can improve the way it manages these inevitable risks in the future.

The Investigation
Although, in the course of their investigation, ATF agents pursued several avenues of inquiry that risked compromising the secrecy of the investigation, steps were taken to reduce the risk when possible and most of these risks were appropriately taken. During the compliance inspection of Henry McMahon, who was thought to be supplying Koresh with weapons and components, Special Agent Aguilera deliberately led McMahon to believe that the inspection was a routine administrative inquiry. In his interview of the Andrade family, Aguilera initially posed as a Texas businessman concerned about his relatives in the Compound. Aguilera did not reveal his true identity until he was satisfied that the family was trustworthy. When Aguilera and Special Agent Buford contacted several former cult members and the families of active cult members the agents identified themselves, but avoided releasing any details about the planned operation.

The Undercover Operation
The danger that Koresh would be alerted to possible enforcement actions against him vastly increased once ATF began its undercover operation. ATF appears to have given little attention to how cult members might have viewed the undercover house, even though the planners knew, or should have known, that the Branch Davidians were extremely suspicious of changes in their environment. The agents' cover was that they were Texas State Technical College (TSTC) students living in one of the houses that Gayle Peery, the owner, usually rented to his ranch hands.

Some efforts were taken to lend credibility to this cover. The agents obtained TSTC student identification cards and TSTC parking decals for their vehicles. Occasionally, some of the agents actually spent time at TSTC. A telephone was installed under one agent's undercover name, and the agents received mail at the house.

The routine at the house, however, could easily have undermined the agents' cover. Chosen for its view of the Compound, the house was small and only had two bedrooms. But it was occupied by eight agents, with four agents staying in it at one time, working two-man shifts. If the Branch Davidians had been keeping an eye on their new neighbors, this rotation would have been an odd, and suspicious, sight. Even if the constant changing of the house's occupants went unnoticed, the traffic created as agents came and went as well as the intermittent visits of technical operations officers and supervisors would have been difficult to miss.

Moreover, the agents selected to play the role of students did not fit the profile of TSTC students. Shortly after the agents moved in, Koresh visited the people in the house next to the undercover house and questioned them about the agents. In the course of the discussion, Koresh expressed doubt that the men were students because their cars were too new for most college students to afford and, according to Koresh, three of the four cars had no credit liens. (Koresh claimed to have found this out through an informant in the motor vehicles department.) In addition, Branch Davidians interviewed after the raid stated that Koresh had been suspicious of the men living across the road because they were too old to be students, their cars were too new, the men carried briefcases and the owner had previously refused to rent the house but then summarily rented it to these individuals. Cult members believed the men were law enforcement, but were not certain what agency they represented. And suspicions within the Compound could only have been heightened when the agents refused to allow cult member David Jones to enter the undercover house, despite his repeated efforts to do so.

The aspect of ATF's undercover operation that was carried out with the least regard for secrecy occurred on January 27, when a special agent posing as a UPS trainee accompanied the regular delivery person to the Compound. The UPS employee cautioned ATF that UPS requires its employees to keep their hair short and the agent's shoulder length hair might raise suspicions. The UPS employee expressed concern that this irregularity presented a safety risk to both men. His warning was ignored.

The agent's conduct during the course of the delivery was even more suspicious than his appearance. The ATF agent instructed the UPS delivery person to drive his truck into the Mag Bag's driveway, go to the door, and ask to use the telephone and the bathroom. The delivery person told the agent that he always parked on the street and had never driven his truck into the driveway. More importantly, he also told the agent that in the course of making numerous deliveries during an 18-month period he had never entered the Mag Bag. The delivery person expressed his concern that the appearance of a second person coupled with such unprecedented actions was certain to arouse suspicions.

Upon arriving at the Mag Bag, they were greeted by Woodrow Kendrick and Michael Schroeder. The delivery person complied with the agents' instructions. He was permitted to make his phone call and the agent was allowed to use the bathroom, however Kendrick asked many questions about the new person.

Upon leaving the Mag Bag, while on the way to the Compound, the agent told the delivery person to follow the same procedure as before. He should drive the truck to the front of the Compound and once they were inside, ask to use the telephone and the agent would ask to use the bathroom. The delivery person responded that he never drove to the Compound but always left the delivery at the gate; he never entered the Compound. The agent also instructed the delivery person that while he was using the bathroom, the delivery person was to drive away, as if he had forgotten his trainee. The delivery person refused to comply with this part of the plan.

Upon arrival at the Compound, the two men were met by David Jones and Koresh, which was unusual because Koresh rarely, if ever, accepted deliveries. Before the delivery person could ask to use the telephone, the undercover agent asked to use the bathroom. Jones already had a roll of toilet paper in hand. He gave it to the agent and told him to use the outhouse. (There were no toilets inside the Compound.) When the agent left, Jones questioned the delivery person about the trainee. Attempting to change the subject, the delivery person tried to engage Koresh in small talk. Koresh told the delivery person, "I know we're being watched." He then returned to the Compound. This undercover effort was so transparent that Koresh complained to the local sheriff's department. He accused the department of trying to infiltrate the Compound.

ATF's failure to exercise more discretion in conducting its undercover probes of Koresh was not responsible for alerting the Branch Davidians that a raid was being planned for a particular day. It did, however, confirm Koresh's conviction that some law enforcement action against him was being contemplated and lent urgency to preparations in the Compound to resist such an action.

Pre-raid Logistics
ATF appears to have given sufficient attention to concerns about operations security as it made final preparations for the raid on the Compound. Beginning February 25, large numbers of agents began to converge at Fort Hood for training. Some agents billeted in barracks at Fort Hood; a few opted to stay in local motels. Support team personnel arrived in Waco on Saturday, February 27, and lodged at local motels. Agents at all locations demonstrated adequate concern for operations security, despite early news accounts to the contrary. Agents did not wear ATF clothing, discuss the operation in public areas, or conduct themselves in a manner that would draw undue attention. Although agents lodging at local motels used government credit cards, if asked, they explained that they were participating in a training course.

Wherever possible, ATF obtained support services from military and other law enforcement sources. When the agency had to deal with private contractors, it was appropriately circumspect. ATF told representatives of a private portable toilet company that their services were needed for a construction project. Company employees were instructed to have a truck at TSTC on Sunday morning where they would receive further directions. Arrangements for the ambulance service were made with similar care. ATF dealt exclusively with the manager and assistant manager, who were told only that the company's services would be needed on February 28 and warned not to discuss this information with anyone else. The ambulance company, which had worked previously with the local sheriff's office, was considered trustworthy.

In hindsight, ATF's use of a private ambulance service did have tragic consequences. A dispatcher for the local ambulance service told Dan Mallony, a KWTX cameraman, about the timing of the raid. This tip started a chain of events that resulted in Koresh being warned about the impending raid. Peeler was directed to go to the Compound area, and ended up discussing the raid with cult member David Jones, who approached Peeler because he appeared lost. Naturally, Jones took this information to Koresh. ATF cannot be faulted for the decision to give the ambulance service information about the pending operation, however. Even though the risk of spreading knowledge about the raid might have been reduced if ATF had used an ambulance service from another city, the benefit gained would have been offset by using a company whose drivers might not know the local roads and hospitals.

The Raid
Three operations security problems stand out in reports about the day of the raid: the convoy was too conspicuous, the forward observers might easily have been seen during their deployment, and the raid commanders' communications were not secure.

Many agents expressed concern about the convoy in which they traveled from Fort Hood to the Bellmead staging area early on February 28. Once the cattle trailers and vehicles assembled for the 100-mile trip, the result was an 80-vehicle caravan, headlights on, with a cattle trailer at each end. This spectacle did not necessarily announce that law enforcement action was imminent, but it did suggest that something highly unusual was happening. Certainly, had thought been given to the convoy's visibility, steps could have been taken to avoid to the problem.

In addition, more thought should have been given to when to deploy six forward observers to the hay barn behind the Compound. Because the distinctly dressed observers deployed at 8:00 a.m., without the cover of darkness, they might have been seen by Branch Davidians and reporters who were traveling the roads.

Finally, Sarabyn and Cavanaugh's use of nonsecure cellular telephones to speak to each other as they traveled to the Compound in cattle trailers from the staging area violated basic operations security practices and was contrary to the communication plan. An antenna on the Compound indicated that the cult members might have had scanners that could have received the cellular telephone transmissions if tuned to the proper frequencies.

Conclusion
Operations security works best in tandem with intelligence. While an effective intelligence operation ensures that decision makers get reliable information about an adversary, effective operations security denies the adversary access to the same sort of information. It is thus not surprising that the operations security problems in the ATF operation mirrored the intelligence failures. Particularly when a raid is so dependent on surprise, operations security cannot be simply a matter of individuals thinking about the consequences of their own actions. Constant attention must be devoted to how the agency's activities might look to a target or his or her allies.

ATF never entrusted any particular individual with this responsibility, and, as was the case with the intelligence gathering effort, the tactical planners and raid commanders were too overwhelmed with other matters to pay sufficient attention to operations security. Consequently, on the morning of February 28, raid commanders underestimated the risk that the raid could be compromised other than by the Saturday and Sunday newspaper articles.

Even though Rodriguez returned from the Compound with information indicating Koresh had been alerted to the raid, those evaluating the information continued to focus on their earlier expectations, that after reading the articles, Koresh would distribute weapons and post sentries. If operations security concerns had been assessed properly, the agents would have treated the newspaper articles as just one of many possible threats. Having done so, they might have had a broader perspective from which to assess the significance of Rodriguez's information. This experience shows that operations security is crucial to the success of large-scale tactical operations.

Section Seven: ATF Post-raid Dissemination of Misleading Information About the Raid and the Raid Plan
Following a tragedy of this magnitude, it was inevitable that the law enforcement community, the Executive Branch, Congress and concerned private citizens would demand an accounting of these events.

In the wake of the tragedy on February 28, the raid commanders, who made the decision to proceed with the raid despite the clear evidence that Koresh had been forewarned, and their superiors in the ATF hierarchy endeavored to answer the call for explanations. But critical aspects of the information that they provided--to superiors, to investigators, and to the public--were misleading or plain wrong. It was not that they lacked access to the relevant facts. Rather, raid commanders Chojnacki and Sarabyn appear to have engaged in a concerted effort to conceal their errors in judgment. And ATF's management, perhaps out of a misplaced desire to protect the agency from criticism, offered accounts based on Chojnacki and Sarabyn's statements, disregarding clear evidence that those statements were false.

ATF Management's Misleading Post-raid Statements
In the aftermath of the Waco raid, perhaps the most frequently asked questions were: Had Koresh been tipped off that ATF was coming? And, if Koresh indeed was forewarned, did ATF commanders know this before they launched the raid? Certainly the news media representatives pouring into Waco sought answers for these questions from official and unofficial ATF spokespeople. The answers would also be significant for those looking toward a criminal prosecution of Koresh and his followers, since evidence that the Compound's residents had deliberately planned an ambush after getting tipped off would blunt any claims that they had merely acted in self-defense against unknown assailants. And ATF's leadership sought answers, that they might respond to media and official inquiries, and that they could work to prevent future tragedies.

ATF's top management appropriately set about to determine whether surprise had been lost, and how. They established a "shooting review" team, and that team systematically looked for answers. Even before a complete picture of the Waco tragedy had emerged, however, Associate Director for Law Enforcement Daniel Hartnett and Deputy Associate Director for Law Enforcement Edward Daniel Conroy, together with Intelligence Division Chief David Troy--who became ATF's principal spokesman about the incident--soon began to make false or misleading public statements about the raid. Moreover, Director Stephen Higgins, relying on their reports from Waco, unknowingly made similar misstatements. To some extent, these misstatements were the product of inaccurate, untruthful or misleading information from Sarabyn and Chojnacki about what they had learned from Rodriguez before deciding to go forward with the raid. In making his initial public statements, Hartnett appears to have consciously avoided confronting the truth and, at the very least, displayed a serious lack of judgment.

As top ATF officials began to receive additional information from line agents and other sources indicating that the raid commanders had proceeded with full knowledge that they had lost the element of surprise, those officials must have realized, had they not already known, that their earlier public statements were either misleading or flatly false. Yet they stuck to their original story, thereby misleading the public and undermining the integrity of their agency.

What follows is a brief summary of the relevant events as they unfolded after February 28.

The Shooting Review Team
On March 1, 1993, consistent with ATF policy, Hartnett and Conroy established a shooting review team to probe the circumstances of the firefight at the Branch Davidian Compound. The team consisted of ATF's Intelligence Division Chief David Troy, Bill Wood, Special Agent in Charge of ATF's Cleveland Field Office, and Dave Benton, the agency's Chief of Planning and Analysis. Troy was placed in command of the review; Benton was not able to participate in aspects of the inquiry, due to other duties. Between March 1 and 3, Troy and Wood interviewed the key participants in the decision to go forward with the raid. During this process, Troy took notes, and he and Wood kept ADLE Hartnett and DADLE Conroy apprised of what the review was being told. At the end of each day, Troy turned over his interview notes to Hartnett.

Shooting Review Team's Interview of Rodriguez

The team's first interview was with Robert Rodriguez, the undercover agent, who related what had happened in the Compound during his visit the morning of the raid.

In his interview with the shooting review team, Rodriguez said that he had been in the foyer with Koresh and others and that the Compound had appeared to be "normal." Koresh was preaching and reading from the Bible. Then Koresh was called from the room to take what was said to be an emergency telephone call. When he returned, he was visibly shaking and very nervous, and he repeatedly looked out the window and dropped the Bible he was carrying. He looked at Rodriguez and said, among other things, "He who kills me kills the Kingdom of God and that includes ATF and the National Guard." Rodriguez also recalled that Koresh said he "could only die once," and, upon looking out the window said, "They are coming for me but they can't kill me."

Rodriguez told the team that, upon hearing Koresh's proclamations, he said he had to leave. In response, Koresh walked up to Rodriguez, shook his hand and said, "Good luck, see you later," and told him to be careful. Rodriguez reported that Koresh had never done or said anything like that before. As a result, Rodriguez felt that he had been "burned," that is, he believed that his undercover identity had been revealed.

Rodriguez then told the team about how he had reported back to his superiors. Upon entering the undercover house, he told Cavanaugh what had happened at the Compound, then called Sarabyn and repeated his account. He specifically recalled informing both Cavanaugh and Sarabyn that Koresh had said ATF and the National Guard were coming. In response to Sarabyn's specific questions, Rodriguez had reported having seen no weapons or signs of preparations to resist a raid while he was at the Compound.

Shooting Review Team's interview of Mastin

After interviewing Rodriguez, Troy and Wood interviewed SAC Mastin, whose account of Sarabyn's actions and statements corroborated Rodriguez's claim to have informed Sarabyn that Koresh knew that "ATF and the National Guard" were coming. According to Mastin--and as the over sixty ATF agents who heard Sarabyn on the day of the raid have since recounted--when Sarabyn arrived at the staging area, he had "a sense of urgency about him." He told the agents, "Let's load up and go." Mastin candidly told the team that although Sarabyn had said "Koresh knows we are coming," he followed Sarabyn's lead and moved to get the trailers loaded and ready.

Shooting Review Team's Report to Hartnett and Conroy

Upon hearing Rodriguez's and Mastin's accounts of events, Troy was at a loss to explain why ATF proceeded with the raid and he doubted the wisdom of the decision to go forward. Troy and the shooting review team promptly let Hartnett and Conroy know what Rodriguez and Mastin had related. After being briefed, Hartnett was upset and expressed chagrin that Mastin had not tried to stop the raid or even questioned the decision to go forward after hearing what Sarabyn had said. Thus, as early as the day after the raid, Troy, Conroy and Hartnett were on notice that ATF's raid commanders might well have proceeded with the raid despite knowing that they had lost the element of surprise.

Shooting Review Team's Interview of Sarabyn

When the shooting review team interviewed ASAC Sarabyn the next day, March 2, he was unable to provide a detailed account of most of his critical conversation with Rodriguez, claiming that Rodriguez "was not real descriptive as to the ATF-National Guard statement," and that Rodriguez had said words to the effect that Howell (Koresh) must know something was going on but nothing explicit that Sarabyn could recall. However, Sarabyn clearly remembered that Rodriguez had said that he had not seen any guns or armed guards. Sarabyn also recalled very little about his conversation with Chojnacki on the tarmac at the command post, when the decision to go forward with the raid had been made. Furthermore, in contrast to Mastin's clear recollection, Sarabyn did not recall making any statements at the staging area to the effect that Koresh knew that ATF was coming.

Shooting Review Team's Interview of Cavanaugh

On March 3, the team briefly interviewed ASAC Cavanaugh, who reported that Rodriguez had returned to the undercover house extremely upset and reported that Koresh had said that ATF and the National Guard were coming to get him and that Koresh had said "our time has come." Cavanaugh had instructed Rodriguez to advise Sarabyn of what had occurred at the Compound.

Shooting Review Team's Interview of Porter

The team next interviewed one of the forward observers in the undercover house, Herman Porter, because they had heard that Porter was upset that the raid had gone forward even though the commanders knew that Koresh had been tipped. When interviewed, Porter was candid and distressed. He said that he had heard Rodriguez's report to Cavanaugh--which he confirmed had been accurately recounted by Rodriguez to the team- -and had been shocked that the raid had not been canceled. Indeed, Porter recalled that after hearing Rodriguez's account of what had happened in the Compound, he had been so certain the raid would be canceled he began putting his gear away.

Shooting Review Team's Interview of Chojnacki

When interviewed by the team, SAC Chojnacki could not recall anything specific that Sarabyn had told him about Koresh's statement regarding ATF and the National Guard. However, he was sure that Sarabyn had said that there were no guns or sentries; this information, Chojnacki claimed, had formed the basis for his decision to go forward with the raid.

After the interviews, the shooting review team was concerned because Sarabyn's urgency and his statements at the staging area about Koresh's knowledge that ATF and the National Guard were coming were inconsistent with his lack of any recollection that Rodriguez had told him that Koresh had been tipped about the raid. As a result, the team was prepared to conduct additional interviews. However, after being told by Hartnett that the local U.S. Attorney's office had directed ATF to stop the shooting review because it was needlessly duplicating the pending leak and murder investigations, the team concluded its efforts.

Hartnett, Conroy and Troy knew surprise was lost

By the conclusion of these interviews, Hartnett, Conroy and Troy were thus confronted with two conflicting versions of the events immediately preceding the decision to go forward with the raid. On one hand was Rodriguez's vivid account of Koresh's extraordinary behavior at the Compound and of his own reports to Cavanaugh and Sarabyn, reports that left little doubt that Koresh had been tipped off. Rodriguez's account was internally consistent and completely corroborated by Mastin. On the other side were Sarabyn and Chojnacki's statements. Not only did these raid commanders--who, given the magnitude of the tragedy at the Compound, obviously had a strong motive to conceal their own misjudgments--display a selective memory about critical facts, but also, what they "remembered" made little sense. Sarabyn's claim that Rodriguez had not informed him that Koresh had been alerted to the raid contradicted reports from agents at the command center of Sarabyn's announcements that Koresh knew ATF was coming. And Rodriguez's account offered the only plausible explanation for the sense of extreme urgency that gripped Sarabyn after receiving Rodriguez's telephone call.

ATF's Media Statements After the Shooting Review

The story ATF top management told the American people bore little resemblance to what had been told to the shooting review team, and had been relayed to Conroy and Hartnett. Uncritically accepting Sarabyn and Chojnacki's account, and disregarding the far more persuasive, and rapidly growing, evidence that the their account was false, ATF's top managers uniformly said, in substance, that ATF's raid commanders had not known that the element of surprise had been lost before they made the decision to go forward.

On March 3, 1993, three days after the raid, and the day the shooting review was terminated, Hartnett was asked during one of the press conferences held near Waco: "When the undercover agent [Rodriguez] heard this phone call [in the Compound on the day of the raid], did he realize at the time that this was a tip?" He responded that "[h]e did not realize this was a tip at the time." (CNN, March 3, 1993). Expanding on this line, Hartnett explained:


 * [T]here was an ATF agent in the Compound shortly before the execution of the warrant. When he left the compound everything was normal--children were outside, people were going about their business. While he was there, a phone call was received by [Koresh], and he began reading scriptures. There was more to it than that, but that was about what occurred.

Similarly, according to an article in the Los Angeles Times the next day:


 * On the morning of the raid, Hartnett said, the undercover ATF agent reported that "everything was normal" at the compound....But the agent left the compound just "as a phone call received" at the facility tipped off the sect members. The agent did not realize at the time that the raid had been compromised. ("Agents Prepare for a Long Cult Siege," Los Angeles Times at 1.)

Three days later, relying on what he had been told by Hartnett and Conroy, Higgins appeared on "Face the Nation," and flatly denied a report that ATF had known about Koresh's receiving a telephone tip about the raid:


 * Q. There has been some suggestion that perhaps your agents knew beforehand that the security had been compromised, that they were aware that Mr. Koresh had received some sort of phone call. Can you just give us your side of that?


 * A. Without being too specific, let me say as I did earlier, this plan was based on the element of surprise. It had to be done quickly, and it had to be a surprise. We would not send our agents into a situation where we didn't think we had the element of surprise....


 * Q. But your bottom line is that you absolutely did not believe the security had been compromised when the agents went into the compound?


 * A. Absolutely not, because as you can see, we walked into an ambush, and there's no way that our people, from the team members to the leadership, would have allowed that to happen had they known it.

When he made these remarks, Higgins apparently had not been informed about Rodriguez's and Mastin's statements.

The Texas Rangers' Reports

Hartnett and Conroy's hierarchical management style, which discouraged rank and file agents from speaking to them directly, effectively insulated them from hearing from the agents with contrary accounts. While Chojnacki, Sarabyn, Royster, and Cavanaugh had access to Hartnett, we have been unable to find a single rank and file agent who spoke to Hartnett about whether the raid commanders had known the raid had been compromised. But management style cannot explain Hartnett and Conroy's failure to change their public statements in the face of yet more evidence that Rodriguez's account was correct.

Within days after the raid, as part of the State's homicide investigation of the February 28 ambush, two Texas Rangers, David M. Maxwell and Coy Smith, interviewed Rodriguez and other ATF agents, including several of the agents who had been positioned in the undercover house when Rodriguez returned from his encounter with Koresh. Rodriguez's account was strikingly consistent with the statement he had recently given ATF's shooting review team. In addition, several of the forward observers informed the Rangers that, while they had not heard Rodriguez's telephone conversation with Sarabyn, they had heard Rodriguez clearly tell Cavanaugh that Koresh had returned from a "telephone call" visibly shaking and agitated and that he had been "tipped" that both the ATF and the National Guard were coming.

During the evening of March 3, shortly after their first interview of Rodriguez, the Rangers briefed Hartnett and Conroy about their interviews, noting that they had found Rodriguez to be a credible witness. Although Hartnett had already been briefed by Troy as to the shooting review team interviews, the Rangers recall that he seemed surprised to learn that Rodriguez positively recalled informing Sarabyn that the raid had been compromised. The next day, after speaking with Rodriguez again, and hearing the same account supplemented by minor additional details, the Rangers reiterated their view of Rodriguez's credibility to Hartnett and Conroy. But Hartnett's and other ATF top managers' public statements supporting the raid commanders continued.

The Rangers interviewed Sarabyn and Chojnacki on March 25 and 26 respectively, and thereafter told both Hartnett and Conroy that Sarabyn's and Chojnacki's accounts made little sense and were inconsistent with the weight of the evidence; and that they found the two men lacked credibility. Sarabyn now had claimed to have specifically asked Rodriguez if Koresh knew that ATF and the National Guard "were coming" and was told "no." The Rangers noted that Sarabyn's story could not be squared with his later announcements at the staging area that the agents should hurry up because Koresh knew they were coming. The Rangers also told Hartnett and Conroy that Sarabyn was evasive during his interview and had unfairly accused Rodriguez of changing his story.

Chojnacki had also claimed to have been unaware that Koresh had been tipped, but the Rangers stressed to Hartnett and Conroy that Chojnacki's claim was contradicted by his decision to join Sarabyn in rushing forward with the raid. Indeed, when pressed by the Rangers, Chojnacki could not offer a coherent explanation for why speed had been necessary. Like Sarabyn, Chojnacki had tried to blame Rodriguez for the flawed decision to go forward, saying, somewhat incoherently to the Rangers:


 * It's very disturbing to me that if Robert [Rodriguez], and I'm not trying to cast blame on anybody, because that I, I thought we had built in enough safeguards to cover ah a cowboy, you know, who would go under any costs or of any of those kinds of terms. Anything or one person would, would be willing to, ah, ah, put us in a risky situation, riskier than our typical work, and ah, his role was so key in this thing and was the key to the whole thing. And I can't believe that at the most critical time if he felt absolutely sure that that was the case that that couldn't be communicated, ah...or that we couldn't recognize that he was attempting to communicate....

At the conclusion of their briefing, the Rangers told Hartnett and Conroy that the morale of the rank and file ATF agents was suffering because they did not believe Chojnacki and Sarabyn's stories, yet the two were still high in the chain of command near Waco. The Rangers suggested to Hartnett and Conroy that morale might improve were Sarabyn and Chojnacki removed from their positions in the chain of command. Their advice was not followed.

The Late March and Early April ATF Statements

Although Director Higgins had begun to hear bits and pieces of information belying ATF's public statements about not having knowingly lost the element of surprise, Hartnett and Conroy failed to keep Higgins informed about the mounting weight of evidence that Sarabyn and Chojnacki's account was false. Higgins's own public statements thus deepened ATF's commitment to a story which was fast losing its credibility.

The occasion for these statements to the media came when, in the face of the agency's misleading public stance, agents "leaked" a competing story to the press. A March 28, 1993 New York Times front page headline proclaimed: "U.S. Agents Say Fatal Flaws Doomed Raid on Waco Cult." The article stated:


 * Contradicting the official version of events, four of the agents involved in the raid and in a review of its aftermath said that supervisors had realized even before they began their assault that they had lost any element of surprise but went ahead anyway.

The article initiated a barrage of ATF denials.

On March 29, 1993, on NBC's Today Show, Higgins, still unaware of Rodriguez's account of what had happened, engaged in the following exchange:


 * Q. Let's talk about one of the other charges, and that is that you have, in fact, said that cult members were tipped off, and now there are reports that bureau supervisors knew that the element of surprise had been lost and yet decided to go ahead with the raid anyway. Is that correct?


 * A. This was a plan which depended on the element of surprise. We would not have executed the plan if our supervisors felt like we had lost that element. So my position has been and continues to be we did not believe that we had lost that element of surprise.

In the next few days, Higgins heard from many agents who challenged the agency's public stance on the element of surprise issue. These contacts prompted him to request a copy of Rodriguez's statement to the Rangers with respect to the raid day events. Hartnett gave the statement to Higgins during the first few days of April. According to one of his top assistants, Higgins, usually a reserved person, exclaimed on reading the account: "What would Koresh have to do, paint himself up with war paint and shoot up the undercover house before we would have known enough to call off the raid?"

Troy also made a number of similar public statements during late March and early April. But, in contrast to Higgins, he was fully aware of the conflicting accounts of what had happened on raid day, and should have realized Higgins, Hartnett and others had overstated the agency's position in early March. As a result, many of his statements in late March and early April appear to have been carefully tailored to confirm that ATF did not realize that it had lost "the element of surprise" while artfully recasting the concept to accommodate the eventual release of Rodriguez's account of raid day events. On CBS's "Street Stories" program, on April 1, Troy stated:


 * There is no way that we would have executed the raid if the people running the operation would have realized that the element of surprise was lost. That would have been obviously a suicidal mission. We were not aware at that point in time, and did not become aware that the element of surprise was lost until they opened fire on us.

In response to reports that unnamed agents at the staging area had claimed that one of the raid supervisors had run around yelling that "we need to go" and "they know we're coming," Troy protested:


 * If the supervisory staff...was aware and convinced that the element of surprise was lost, there's no way we were going to go driving in there and execute a warrant because the element of surprise was a key factor.

The next day, Troy began to recast the issue when he told the Washington Post:


 * We felt that there wasn't compelling evidence that Koresh knew that a raid was planned for that day. Had agents known that the element of surprise was lost, the raid would have been halted. "Koresh Described as 'Nervous' After Call Before ATF Assault," Washington Post, April 2, 1993 at A3.

His remarks signalled a shift toward portraying the raid plan as requiring "compelling evidence" that Koresh had been tipped before the raid could be halted, rather than confirmation that conditions were right before the commanders would go forward. & also Houston Chronicle, "ATF Knew Koresh Tipped Off, Sources Say," April 2, 1993 at 1 ("Troy...said at a news briefing Sunday that agents did not know Koresh had been warned when they ordered the raid to proceed."). Similarly, on April 2, Troy appeared on ABC's "Good Morning America" and stated:


 * At this point in time our position is this. We know exactly what statements were made by our undercover agent to our tactical commanders when he came out of the compound Sunday morning.... [W]e know exactly what statements were made by our commanders when they were at the staging area prior to departing for that raid.... [T]he important thing that was said, that we feel, is the undercover agent saw absolutely no preparation for any kind of battle plan, there were no firearms displayed by anyone...we did not feel that they were gearing up, getting ready for any kind of offensive activity. 

(Emphasis added). In so doing, Troy held to the position that ATF had not known it had lost the element of surprise, but again subtly tried to redefine the concept of losing "the element of surprise," this time, to require outward signs of an ambush being prepared. By implication, therefore, according to Troy, so long as no weapons were visible, Rodriguez's information was not sufficient grounds to stop the raid. In keeping with this theme, on April 3, 1993, Troy told the Dallas Morning News: "We were not looking at a situation where we had a shrinking window of opportunity. We didn't say, 'this thing is turning bad, so let's go in before it does."' Troy further reconstructed the concept of the element of surprise on May 3, 1993, when he told Time magazine that "[t]he element of surprise does not mean they don't know you're coming. Only that they can't take control." By then, Troy had diluted the concept of surprise into a functionally meaningless term.

The Significance Of ATF's Misleading Statements

There may be occasions when pressing operational considerations--or legal constraints--prevent law enforcement officials from being less than completely candid in their public utterances. This was not one of them. And a desire to shield one's agency from public criticism cannot justify false or misleading pronouncements on matters of clear public concern. Hartnett, Conroy, and Troy permitted public statements to be put forward that were either most likely false or definitely false. Troy admitted to the Review that he wrongfully misled the press and the public. Troy provided two explanations for his actions. First, that he was trying to provide information to the press corps. Second, that he was acting at the direction of Hartnett, whose management style discouraged subordinates from challenging his instructions. Neither explanation is acceptable. Hartnett and Conroy, in contrast, were not subject to instructions to make misleading statements, and never gave their superior, Higgins, an opportunity to learn the truth.

The extent of Director Higgins's knowledge places him in a different category, since he was not aware of the falsity of Sarabyn and Chojnacki's account when he adopted that account in his public statements. However, Higgins must accept responsibility for continuing to take public positions on the issue when repeated questions from the media and information readily available to him should have made it clear that he was on shaky ground. Higgins never adequately questioned his subordinates to determine the facts until early April.

An oft-stated justification offered by top ATF management officials for their misleading statements, and their failure to inform the public about Rodriguez's and the other agents' conflicting accounts, is that they were prohibited from doing so. Various ATF officials have claimed that at various times that the local United States Attorney's office, the Texas Rangers, and officials in Treasury prohibited them from speaking to the public or the media on the subject of the loss of the element of surprise. And the Review has found, in fact, that representatives of both the Texas Rangers and the local United States Attorney's Office asked Hartnett, Conroy and Troy to refrain from commenting in specific terms about the loss of the element of surprise because of concern about how such statements might affect ongoing investigations and likely future prosecutions. Similar requests came from the Treasury Department. Over time, as ATF kept mischaracterizing the raid commanders' knowledge, these requests were sharpened and put more forcefully--and indeed, by early April, particularly with respect to Treasury's concerns, ripened into an effort to convince ATF to make no further statements on the subject. Still, since ATF officials obviously ignored these requests, and spoke regularly about this subject to the media, the requests offer no justification for making statements known to be misleading or false.

In addition to misleading the public, the statements by Conroy, Hartnett and Troy also had the effect of wrongfully pointing the finger at Rodriguez as being responsible for the failed raid. If the raid commanders were not informed that Koresh had been tipped, then the necessary corollary was that Rodriguez likely had failed to tell them what they needed to know. He was to blame. Moreover, despite the consistency of Rodriguez's recollection of what happened immediately before the raid, persistent rumors circulated that he was changing his story. As Rodriguez appropriately protested:


 * They're saying that [I've changed my story about what I saw in the compound and what I told raid commanders.] That's not true. Every time I told my story, I said it the same way every time. The Rangers know that too. There's no reason for me to go and make up stories. "ATF Agent Says He Saw Disaster Loom," Dallas Morning News, May 13, 1993 at 8A.

ATF's top managers should have acted swiftly to quash those rumors; they did not.

Sarabyn and Chojnacki lied to their superiors and investigators about what Rodriguez had reported. Their consistent attempts to place blame on a junior agent were one of the most disturbing aspects of the conduct of senior ATF officials. The recollections of Sarabyn and Chojnacki have diverged considerably since the immediate aftermath of the raid. After being confronted with the collective contrary recollections of dozens of line agents, Sarabyn finally admitted the accuracy of Rodriguez's account. In contrast, despite the weight of contrary evidence, Chojnacki steadfastly has contended that Sarabyn neither said nor did anything that alerted Royster and him that Koresh had been tipped.

The Alteration of ATF's Written Raid Plan
In addition to making misleading statements to their superiors and investigators about the basis for their decision to proceed with the raid, Chojnacki and Sarabyn altered documentary evidence, misleading those probing their operational judgments.

The Drafting of the Raid Plan

ATF's National Response Plan required that a written plan "for managing the critical incident or major ATF operation" be produced prior to the initiation of the operation. But the plan did not have to be distributed. The point of the requirement appears to have been limited to ensuring that multiple SRT activations were predicated only upon a well considered, reasoned and thorough raid plan.

Although the raid on the Branch Davidian Compound had originally been set for March 1, 1993, no one had even started to draft the mandatory documentation of the raid plan by February 23, 1993, when ASAC Darrell Dyer (Kansas City) arrived in Waco and was assigned to be the Support Coordinator for the operation. Dyer's past military service led him to assume that there was a detailed written raid plan, but, when he asked the raid planners for a copy he was advised that none existed. Thereafter, Dyer and agent William Krone took it upon themselves to produce one, even though they started with little knowledge about the work of the tactical planners. In a flurry of activity, the two conducted interviews, gathered information and eventually were able to generate a written raid plan. Due to the tight timetable, the plan did not meet Dyer's standards in terms of quality, and from his perspective was still a "work in progress." Nevertheless, the two of them had essentially finished a written raid plan the day before the raid. The plan, however, remained on Krone's desk; it was never distributed to any agents, or relied upon by any of the planners. (See generally Appendix C - Original Raid Plan, dated February 25, 1993.)

The Alteration of the Raid Plan

After the failed raid, authorities began to ask ATF officials for the raid plan. The Texas Rangers were the first to ask ATF's Houston Office for the raid plan. When Dyer was told of the request, he realized that the written plan had never been put in a satisfactory form. He advised Chojnacki and Sarabyn, and the three decided to revise the plan to make it more thorough and complete. Nowhere on the new version of the plan they crafted was there any indication that this was not the original document, or any identification of what had been added. The only hint that the plan had been modified was a handwritten notation in the margin of one page that did not indicate when and how the notation was made. Moreover, Chojnacki--the agent responsible for producing it for the Texas Rangers--never advised them that there was an original raid plan that differed substantially from the plan produced. Indeed, when the Review asked ATF for all documents relating to ATF's investigation of the Branch Davidians, initially only the altered raid plan was received, without any indication that it was anything other than a document prepared prior to the raid. In fact, the document received reflects yet another revision, since the handwritten margin note in the Ranger's copy was now incorporated into the typewritten text. At no time did any ATF official inform the Review that the plan submitted was not the original raid plan.

The alterations indicate not an attempt to create a plan that existed in the minds of the tactical planners and raid commanders on February 28. Rather, they suggest a self serving effort to clarify the assumptions on which the planners had relied and enhance the reader's sense of their professionalism. For example, to rebut criticism that ATF should have arrested Koresh when he ventured away from the Compound, the following language was added to the altered raid plan: "The subject has not left the Compound in months and has made statements that he does not plan to leave." (Appendix C - Altered Raid Plan, dated March 22,1993.) A second alteration sought to buttress ATF's initiation of the raid at 10:00 a.m. instead of the standard pre-dawn timing which law enforcement organizations customarily use to gain surprise:


 * The women, men and firearms are kept in different areas in the structure. Usually at approximately 10:00 am in the morning, the majority of the males and Howell should be in the underground area. SRT teams have been divided to handle the areas listed above. (Appendix C - Altered Raid Plan, dated March 22, 1993.)

Inquiries into the Alteration of the Raid Plan

The readiness of Chojnacki, Sarabyn, and Dyer to revise an official document that would likely be of great significance in any official inquiry into the raid without making clear what they had done is extremely troubling and itself reflects a lack of judgment. This conduct, however, does not necessarily reveal an intent to deceive. And, in the case of Dyer, there does not appear to have been any such intent. The behavior of Chojnacki and Sarabyn when the alteration was investigated does not lead to the same conclusion.

After the Review had obtained a copy of the original raid plan from a different source, and compared it to the revised document that ATF had produced, Dyer, Sarabyn and Chojnacki, the only three people who could have been involved in changing the document, were questioned. When asked about the alterations, Chojnacki denied knowing that the raid plan had been altered in any fashion except the handwritten comment in the margin of one page of one of the altered versions. (Appendix C - Altered Raid Plan, dated March 11, 1993.) Similarly, Sarabyn claimed that he had directed only that the date of the raid be changed from March 1 to February 28. Chojnacki and Sarabyn also denied knowing that other changes had been made, how they had been made, and who directed that they be made. Neither Chojnacki's nor Sarabyn's denial is credible.

When questioned about the alteration of the raid plan, Dyer recalled that it had been changed following the Texas Rangers' request because the original document had been incomplete, inaccurate in certain respects and had not fully articulated the reasoning behind the plan. He had advised Chojnacki and Sarabyn of these shortcomings, and the three decided to change the original plan in a manner that would "upgrade" it. Dyer candidly admitted to the Review that he had made certain changes to the plan. He said that, at the time, he had not thought he was doing anything wrong, but was simply "correcting" the original document. When questioned about the importance of identifying the altered plan as amended, Dyer agreed that it was a serious error in judgment not to properly label the altered document. In fact, he candidly stated it was a "stupid" mistake.

When advised by the Review that Chojnacki and Sarabyn had denied making any changes except the handwritten marginal comments Chojnacki had affixed to one of the already altered versions of the plan and Sarabyn's change of the raid date, Dyer seemed shocked. Obviously, as Dyer realized, when taken together, Chojnacki's and Sarabyn's denials amounted to a joint accusation that Dyer had directed or made all of the other changes.

The Review credits Dyer's account of events and believes that both Sarabyn and Chojnacki falsely denied participating in the alteration of the original raid plan. The assessments are reinforced by Dyer's relative lack of knowledge about the facts that were changed in the raid plan. Certain changes that were made went beyond Dyer's knowledge of the raid plan and the factual assumptions upon which it was built. Everything he knew came from someone else; he created nothing; he decided nothing. And, of course, as the only one of the three who was not intimately involved in planning the failed raid, he lacked motivation to lie about making changes to the plan. Sarabyn and Chojnacki's false statements with regard to altering the raid plan document is consistent with their failure to tell the truth about raid day events. And their readiness to blame Dyer indirectly is equally consistent with their efforts to do the same to Robert Rodriguez.

Introduction
During the investigation of the Branch Davidians and the subsequent raid on the Compound, ATF obtained assistance from the military, including the Texas National Guard. This support included the provision of training facilities and equipment, aerial reconnaissance missions, the use of helicopters during the raid, and advice concerning ATF's medical and communications plans. In the wake of the raid's outcome, specific questions were raised about the representations made by ATF in its effort to obtain the use of the helicopters which had been provided by the National Guard. This section responds to those questions.

ATF's Initial Contact with the Military
While investigating Koresh for violations of federal firearms laws in November 1992, ATF believed it required military assistance. ATF, therefore, approached the U.S. military and Texas National Guard for support. In early December, at ATF's request, the Department of Defense liaison to ATF briefed ATF officials about military support available for the Branch Davidian operation. During this briefing, the Department of Defense representative told ATF officials that ATF could obtain military assistance without having to reimburse the Department of Defense if the investigation was related to narcotics enforcement, i.e. had a "drug nexus." An ATF agent then met with officials of the Texas National Guard Counterdrug Support Program to determine what assistance the Texas National Guard could provide. During the meeting, the Guard and individuals representing the state of Texas reiterated the fact that nonreimbursable military support could be made available to ATF if the case had a drug nexus.

After these meetings, ATF officials investigated whether there was any drug activity at the Compound. The ATF case agent learned from an informant that parts of an illegal methamphetamine laboratory had been at the Compound when Koresh took control of the premises, and that the McLennan County Sheriff's Department had planned to collect this equipment. The informant, however, did not know whether such parts were ever collected. Upon inquiring at the sheriff's department, the agent found no records indicating that these parts had been collected by or turned over to the sheriff, raising the possibility that the illegal equipment might still have been at the Compound.

ATF acquired additional information that suggested there was drug activity at the Compound. An ATF agent who was acting in an undercover capacity during the investigation reported that Koresh had told him that the Compound would be a great place for a methamphetamine laboratory because of its location. Furthermore, information obtained from informants and a search of the criminal records of the Branch Davidians revealed that one cult member living at the Compound had a prior conviction for possession of amphetamines and a controlled substance, and that 10 other individuals associated with the Compound had previously been identified as having some involvement in illegal narcotics activity. The drug involvement of the 10 individuals varied; some had been arrested for alleged drug violations while others had been investigated for suspected drug activity.

After ATF had gathered this information, ATF officials informed representatives of the U.S. military and the Texas National Guard on numerous occasions about possible drug activity at the Compound. On February 4, 1993, ATF officials met with representatives of both groups to discuss the Branch Davidian operation. At this meeting, the military representatives were accurately informed of the results of ATF's investigation into the existence of a drug nexus. This briefing satisfied the representatives that a sufficient drug nexus existed to justify military assistance on a nonreimbursable basis.

ATF's Specific Requests for National Guard Support
On December 14 and 18, 1992, an ATF official wrote to the Texas National Guard Counterdrug Support Program requesting that the Guard take and interpret aerial reconnaissance photographs of the Compound. The National Guard subsequently conducted a total of six flights over the Compound and Mag Bag from January 6 through February 25, 1993. During the flights, the Guard used infrared scanning devices, which identified "hot spots"--heat sources--inside and outside the Compound. A Texas National Guard airman then provided ATF with an unofficial interpretation of the reconnaissance videotapes that suggested a hot spot inside the Compound was consistent with characteristics of a methamphetamine lab. ATF, however, never obtained an official interpretation of the videotapes.

In addition to the reconnaissance flights, the Texas National Guard supplied three helicopters and pilots for training exercises on February 27, and for the raid the following day. Prior to February 27, ATF officials told representatives of the Guard that the helicopters would be used as an airborne command platform and to transport ATF personnel and evidence on the day of the raid. During the training exercises, however, ATF officials informed the National Guard pilots that on the day of the raid, the helicopters were to arrive at the rear of the Compound shortly before the raid teams to draw the attention of the Branch Davidians away from the agents arriving in the cattle trailers. On raid day, however, the helicopter pilots encountered unexpected gunfire from the Compound as soon as their aircraft came within range, and they were forced to abort their mission.

Analysis
ATF did not mislead U.S. military or Texas National Guard officials in obtaining their assistance on a nonreimbursable basis. ATF conducted a legitimate inquiry into whether a drug nexus existed in the investigation after military representatives told ATF officials about the possibility of nonreimbursable assistance. ATF officials were aware that they could have obtained military support for the operation even if no drugs were involved in their investigation. However, in the absence of a drug nexus, ATF was told by both the U.S. military and the National Guard that the assistance would be reimbursable.

Once ATF gathered information about a possible drug nexus at the Compound, it presented this information to the U.S. military and the Texas National Guard. Representatives of these groups evaluated the information and found that it was sufficient to warrant assistance on a nonreimbursable basis. Because there is no formal standard by which the military defines a drug nexus in a law enforcement investigation, a substantive review of this decision cannot be conducted.

Nonetheless, the Review finds that the standards for nonreimbursable military support raise questions about the appropriate scrutiny that should be given when considering the issue of a drug nexus. The lack of a formal standard by which the military defines a drug nexus in a law enforcement investigation raised questions regarding the nonreimbursable assistance provided to ATF. It would be appropriate therefore that federal law enforcement, the U.S. military and National Guard develop more precisely defined criteria for determining when a drug nexus is sufficient to justify nonreimbursable military assistance.