Just two seconds

by Gavin De Becker; Thomas A. Taylor; Jeff Marquart

  Who would have thought that in a book written for protectors, we'd be praising photographers? Given how much interference and hassle they've caused to protectors, there is perhaps some fairness in our extracting all we can learn from them -- and maybe also some fairness in our respecting their commitment.

  Protectors and photographers are often together at the scene of attacks, and though protectors routinely operate in the line of fire (or at least hope to), photographers are also sometimes at risk. However, only the protector has the safety of another person on his shoulders -- and sometimes the safety of many people. So we won't end this book with praise of photographers. Rather, we praise those protectors who have accepted the high stakes of their craft, understand the risks, know they might be blamed for the actions of others and for events outside their control, accept that they cannot control everything, commit to control everything they can, and know that Time is on their side only if they act outside of time -- in the Now. Above all, we praise those protectors who continue to learn, willing to receive valuable lessons from any source -- from other protectors, from photographers, even from assassins. The willingness to keep learning is not an attribute shared by everyone who claims to be a protector, of course, but as a reader of this book, you've shown your interest in new ideas and your commitment to improve your effectiveness.

  We hope that any dialogue opened in these pages will be continued by others in our field. Our highest wish is that protectors can be more effective as a result of our work, a wish we put our hearts into because we're so proud to be part of what you do.

  THE COMPENDIUM

  OF ATTACKS AGAINST AT-RISK PERSONS

  "It is even better to act quickly and err than to hesitate until the time of action is past."

  Carl von Clausewitz

  The Compendium Described

  This Compendium is a collection of attacks, near attacks, and incidents involving at-risk people all over the world, spanning nearly fifty-years (1960 to mid-2007). Targets were either under the protection of bodyguards at the time of the incident, or had reason to believe they were at risk by virtue of their position or situation.

  Methodology

  Some of the information in the Compendium was gleaned from journalistic reports; as is the norm when relying upon the news media, we encountered conflicting and inaccurate information. When no other sources were available, we drew upon later reports, since initial or so-called "breaking news" stories are often filled with speculation and inaccuracy.

  In each case we sought to learn the setting and situation of the attack, the weapons used, the proximity of attacker(s) to the target, the number of attackers, the duration of the attack, and of course, the outcome.

  By design, our analysis is sufficiently liberal that minor inaccuracies in individual news reports do not have a significant impact on the integrity of the overall conclusions. (See page 154 for a more detailed discussion of the data and its analysis.)

  This book does not concern itself much with motives for attacks, and we do not separate attackers into categories such as sane versus insane, nor ideological versus political. Determining motive and category of attacker is not always easy, in any event. For example, Mehmet Agca shot Pope John Paul II in 1981. He was described as a terrorist affiliated with the Grey Wolf terrorist group. He also had ties to the Bulgarian Secret Police, and, reportedly, to the Soviet Union. Further, he claimed at times to be Jesus Christ. All that aside, what matters for our purposes in the Compendium is that a heavily-protected person was wounded when one attacker armed with a pistol fired from close range while the target was riding in an open car, and the attack was over in seconds.

  We are protectors, not statisticians. Accordingly, we've analyzed and presented information about attacks in the way we feel will best inform and enlighten protectors.

  A Sample, Not a Complete Collection

  While the number of cases described and analyzed in these pages is impressive, the Compendium is not held out as a complete listing of every single attack that's occurred on Earth since 1960. (Trying to present that much data would fail in any event, given that in 2006 alone, according to the National Counter-Terrorism Center, there were more than 1300 government leaders, officials, and bodyguards killed or wounded.) Rather, we sought to assemble a large enough sample from which to draw insightful and meaningful conclusions.

  There were two fairly obvious criteria to be met before an incident could be considered for inclusion in this Compendium:

  The incident became known beyond the protective detail (often by becoming a news story); or,

  The incident became known to the authors through direct experience or trusted sources (law enforcement agencies, involved parties, credible publications, research, etc).

  Though there were just two criteria for inclusion of incidents, there were many criteria for exclusion. First, visits to the homes or work environments of prominent people by unwanted pursuers who did not attempt attacks are not included herein because there are literally thousands each year. Most of these visitors are mentally ill or emotionally invested fans who are fixated on a famous person. As just one example (already reported publicly), our firm provided protective coverage for the actress, model, and author Brooke Shields while she attended Princeton University. As a very famous and often topical cultural figure, her attendance was widely reported around the world. During the four years, ninety unwanted pursuers who visited the campus were interviewed, detained, or arrested by protectors or police. There was no precedent for a media figure as famous as Shields being so accessible in the exact same place every day for years, with the public frequently informed and reminded. (Later, President Clinton's daughter attended university under similar conditions, though she attracted fewer mentally-ill pursuers.)

  We also did not include many of the so-called palace coups. For example, South Vietnamese President Ngo Dinh Diem surrendered to opposing forces in Saigon in 1963. Being transported in an enemy tank, Diem was shot in the head by the platoon commander. While the incident goes down in history as an assassination, it was actually more like a common murder. It does not teach us anything about protective work, for at the moment of the shooting, there was no protection.

  We have also left out some incidents (and even attempted attacks) that have never been reported publicly and which the targets wish to keep confidential. For example, in the 1980s, a team of protectors from our firm was bringing a protectee back from a concert she had just performed. In an excited crowd waiting for our client outside the hotel, a man raised his arm with a knife. A member of our security team grabbed the knife and our people rushed the protectee into the hotel. The man retreated into the crowd and we were not able to learn his identity or intent. The incident was never reported publicly. (Even if included, that incident would not impact the statistics enough to change the conclusions arising from the analysis.)

  We did not include incidents involving angry or unruly crowds unless there was clear danger or sinister intent, such as protestors who became violent. For example (as publicly reported), our firm provided the protective team for singer Tina Turner when she performed before the world's largest-ever paying audience: more than 240,000 people at Maracana Stadium in Brazil. Order at Maracana was just barely maintained by police as hundreds of people scaled walls rather than purchase tickets. Outside the stadium, people had parked in the middle of streets we needed to use to depart. Accordingly, many small vehicles had to be literally picked up and moved by police and protectors. The departure, while dramatic, didn't rise to the level for inclusion herein. There have also been many incidents in which audience members swarmed stages at performances (thankfully not with our clients); unless danger was present or intended, they've not been included in the Compendium.

  An example of extreme danger from a crowd occurred when then Vice-president Richard Nixon's motorcade was attacked by an angry mob in Caracas. However, since that incident occurred two years earlier than our cut-off date of 1960, it is not include
d herein.

  We also did not include interpersonal attacks or homicides of public figures, such as the murder of comedian Phil Hartman by his wife.

  Another kind of case excluded from the Compendium involves unwanted pursuers who were, at the time of publication, still focused on the prominent person they have stalked. In other words, the cases are still evolving.

  Finally, we did not include incidents about which there was insufficient information to analyze.

  Even after all the exclusions, the Compendium nonetheless considers more than 1400 cases, an excellent number for useful analysis, and the largest collection of such information we've seen anywhere.

  Note: This Compendium is kept up-to-date. The Gavin de Becker Center for the Study and Reduction of Violence, a not-for-profit foundation, collects and organizes information on assassinations, attacks, and incidents involving at-risk persons. Qualified requesters can obtain information on the Compendium at www.gavindebecker.org

  STARS: Successful Tactics, Action, or Response on the Scene

  When attackers and plotters were unsuccessful, their failures occurred for a variety of reasons, however to draw particular attention to those cases in which attacks were foiled or harm was reduced as the result of on-the-scene protector action, some Compendium entries include the word STARS (Successful Tactics, Action, or Response on the Scene). The key words here are on the scene because that's the core topic of this book.

  4 STARS = Protector action during the attack entirely prevented injury

  3 STARS = Protector action during the attack favorably influenced safety/survival

  2 STARS = Protective Strategies or Resources at the scene favorably influenced safety/survival (e.g., effective positioning, an armored car, advance arrangements, weapons detection, deterrence)

  1 STARS = Management of the incident immediately after the attack favorably influenced safety/survival

  In cases that might fit into more than one category, we assigned the highest applicable number. For example, the attack on Pinochet (Case #120) includes protective strategies that favorably influenced outcome (order of vehicles in the motorcade had been shuffled), important protective resources (armored car), and most notably, protector action during the attack that entirely prevented injury (driver removed Pinochet from the line of fire). A clear 4-STARS case. The number of STARS assigned to a given case is a subjective matter of assessment, deduction, and opinion which others could see differently. It is not science and doesn't pretend to be. Setting aside the different ways people might analyze specific cases, here's what can be learned taking the statistics at face value for the moment:

  Among cases in which protector actions or strategies clearly had an important and favorable influence on safety, that benefit arose approximately 2% of the time from protector action after the incident. Approximately 37% of the time, benefit resulted from direct protector action at the scene of the incident, and 61% of the time, the benefit was the result of protective strategies or resources at the scene. What we learn here is that protectors at the scene are most likely to have a favorable influence on safety, not through the dramatic action of jumping in front of a bullet or resuscitating an injured protectee, but rather through protective strategies and resources (e.g., effective positioning, an armored car, advance arrangements, logistics, weapons detection, deterrence). This means that logistical decisions some might think of as small or routine, such as how close the car is placed to the building, which entrance or exit is selected, how close to the protectee crowds are able to be, how close to the protectee protectors are able to be, which footroute is selected -- are not, alas, small decisions at all.

  Special mention must go to a protective decision made far from the site of the attack, far from the moment of the attack. It's a decision usually made in a comfortable office, frequently after much discussion and resistance, a decision made by people who won't themselves be in the line of fire: It is the decision to have an armored car, or just as significant, the decision to not have an armored car. The Compendium cases speak for themselves on this subject, and speak loudly: There is no single precaution at-risk people can take that will mean more to their safety than traveling inside a fully armored car. Note the word fully, included here because of those cases in which the occupants of partially armored cars were killed during attacks. Also note the need for the protectee to stay fully inside the armored car, and to keep the windows up. These last observations might seem too obvious to include, however they haven't been obvious to everyone: They were not, apparently, obvious to Benazir Bhutto, who had the exceptional advantage of an armored car, yet elected to place half her body outside the vehicle through the sunroof -- and died for the decision. Similarly, Afgan President Karzai had his window down in the back of the car when an assassin shot at him. Karzai was fortunate enough to survive that day, though two other people died during the attack.

  The extent to which protectors reduced risk through deterrence or precautions cannot ever be fully measured. We can't know about all the attacks that weren't launched or were aborted. What we can know is that protectors made a favorable impact on target safety in 22% of all cases. However, since many targets who were attacked chose not have protectors with them, the more revealing truth is this: When protectors were present, their action was the reason attacks were unsuccessful about 57% of the time. History is giving a strong endorsement for at-risk people to have protectors!

  Though people typically think of 4-STARS as better than 2-STARS, any number of STARS reflects enhanced safety, and thus they are all good. Those 2-STARS cases in which protector strategies or resources prevent an attack from occurring altogether, arguably depict the most desirable situation and outcome. However, in a book concerned with events during the attack at the scene of the attack, 4-STARS go to those cases in which protector action during the attack prevented injury. The use of higher and lower numbers is not meant to communicate that effective response is somehow better than outright prevention -- it isn't.

  The Compendium also contains descriptions of many attacks in which the target was injured, and survived. Though the protectors in those instances did not prevent all injury, some made a profoundly favorable difference, and we draw attention to those cases in which harm was reduced as the result of protector action. Accordingly, you'll see the occasional STARS designation alongside a Compendium entry that describes a seemingly successful attack. In protective work, as in many aspects of life, it isn't over till it's over -- and the Compendium cases show that while the assassin's success is defined by just one outcome, protector success can occur in many ways.

  * TOTAL STARS = number of STARS cases out of the number of total cases (1A has 3 out of 65 cases; 1B has 24 out of 371 cases, etc.)

  Notes on the Compendium Categories

  Status of Target: In many cases, the target qualified for more than one category. Robert Kennedy was a Member of Congress, a national figure (presidential candidate), and a media celebrity. His assassination likely arose from being a presidential candidate; however, we counted him as a Member of Congress. Governor John Connolly of Texas was hit by gunfire as the result of riding in the car with President Kennedy, not because he was targeted as a governor, so his shooting in the motorcade in Dallas is not listed as a separate incident.

  Injury: There have been incidents in which the targets sustained injury and chose not to report it, or minimized the seriousness of injuries. Likewise, injuries have in some cases been exaggerated by targets and their advisors for political or PR benefit. When we could not be certain about the extent of injuries, we followed the version of events most frequently reported. If there was insufficient information to draw any reliable conclusions, we pointed out inconsistencies and controversies, or deleted the case entirely. Hence, there are some entire incidents intentionally left out.

  Assailants: There are incidents in which the number of assailants is not known. If an al-Qaeda suicide bomber walked up and exploded a bomb, we counted one assailant, though it'
s clear that more than one person was involved in such things as making of the bombs, transportation, and even sometimes assisting the bomber at the scene of the attack. If a terrorist shot a target and then escaped on the back of a motorcycle, we counted two assailants. What we sought to answer was, in effect: "How many bad guys did the protectors on the scene face?" If a bomb goes off when a target drives by, we count that as one attacker, unless reports specifically identify others (e.g., "three men were seen fleeing"). In some cases, we had to deduce the number of assailants on the basis of injuries to the target. For example, an incident in which a target is shot once in the head might be presumed to involve one assailant. Many of the attacks outside the US involved attackers who got away from the scene entirely, and in those instances, it is, of course, difficult to estimate numbers.

  Weapon: In a few instances, we had to guess whether the weapon was a handgun or shoulder weapon, however we feel confident about the total number of cases placed into the categories of firearms, bombs, knives, etc. With incidents in which assailants threw grenades and also fired weapons, we tried to determine which weapon caused the injury or death. In some cases it was necessary to select from among more than one weapon when counting cases in the analysis. For example, in the 1998 attack on Eduard Shevardnadze, his motorcade came under attack by gunmen armed with a grenade launcher and machine guns. The front of his car was badly damaged when it was hit by a rocket-propelled grenade. Though the car was also struck by gunfire, in our analysis we counted the primary weapon (the one doing most damage) as "anti-tank weapon." Similarly, though Sadat was attacked by firearm and grenade, the mortal injuries were caused by bullets.