The Riverman: Ted Bundy and I Hunt for the Green River Killer

by Robert Keppel; William J. Birnes; Ann Rule

  Ted tried to keep me in his corner by complimenting our HITS program. He wanted to continue our correspondence about serial murder. One last time, he inquired about the crime scene photo I had shown him back in 1984. He asked if I had any more to show him. It was almost as if he needed to be energized. I told him maybe next time.

  The stage had been set. Bundy had carefully prepared his student. In an indirect way, by talking about what would and wouldn’t work when interviewing the Riverman, he told me what it would take for him to talk about the murders he committed. I “would have to give him something.” Ted was talking about his life; he didn’t want to die in the electric chair.

  Bundy emphasized that he had kept his secret for so long that any statement of admission would be difficult and would have to be made in degrees from the less painful to the most shameful. Ted had nominated me to be his confessor by grooming me from day one under the camouflage of solving the Green River murders. He had warned me about saying “I understand” before he was able to tell me what “understanding” him was all about. My approach to him was that I was always looking for information that contributed to my understanding of the Riverman and, through the Riverman, him. It was an ongoing process that didn’t stop even with his execution.

  He force-fed me with advice on maintaining a low-key approach and remaining patient at all times. My patience in developing the facts necessary to understand Ted on his terms was constantly being tested by him against the whetstone of his disturbing admissions and pretentious lectures on the Green River Killer. Maybe he had already convinced himself that I would be ready when the time came. Maybe this had always been part of his own exit strategy from the very beginning of our extended dialogue.

  I did not know that Ted had made the decision to confess to me in February 1988, about one year prior to his death. If it hadn’t been for his attorney’s talking him out of it, Bundy would have begun telling all. I was always curious why Bundy kept changing the day of my visit in February at the last minute. Now I know. He was negotiating with his appellate attorneys about the propriety of giving incriminating statements. In 1988 he ultimately chose not to confess.

  What controlled Ted’s ability to communicate with me was the stage of his appeals. By February 1988, he had received another execution delay. The next stage was less than a year away. His advisors apparently concluded that confessing to new murders might undermine his chances to win an appeal of his conviction for the Florida murders. At the close of our 1988 conversations, I informed him that I would return to visit him anytime he desired.

  Unfortunately, Ted waited too long. My next visit was scheduled four days prior to his execution in January 1989. The conditions for his interview were horrible, and it was conducted in a circuslike atmosphere. Representatives from every major television and radio station and newspaper from Washington to Florida were congregating outside the gate of the Florida State Penitentiary to cover the story of his execution. My role was to participate in his “debriefing,” which was not exactly the forum that I expected for his voluntary confession. One more time, Ted was going to manipulate me and everyone else around him; this time it was his disorganized and unsuccessful effort to save his own life.

  12

  The Signature of Murder

  Ted Bundy, while he never did catch the Green River Killer as he wanted to do, still played a very, very indirect role in the capture of one of Washington’s most outrageously violent serial killers. And he did it from beyond the grave. I wouldn’t have believed it possible had I not seen it with my own eyes. It all began with Bundy’s advice that police agencies develop a violent-crime tracking system on computer. That eventually became reality with the FBI’s VICAP program that was housed in the basement of the organization’s headquarters at Quantico. Bundy was indirectly responsible for the development of the program because his cases, spread out as they were across hundreds of miles, were the types of crimes VICAP trackers and profilers were trying to investigate.

  However, in the state of Washington, we had other plans. We went on to develop our own computer tracking system that ultimately helped us establish a pattern for and then solve a serial-murder spree that would have slipped right through our fingers if we had used normal investigative methods. The case we solved was that of George Russell, who was caught, in no small measure, because of Ted Bundy’s comments years earlier and our subsequent homicide tracking system based partly on our experience with Bundy. And, like most of my cases, it started with a simple request over the phone.

  Inception of HITS

  Two months after I left the Green River Murders Task Force, I was surprised by a telephone call from Terry Green at VICAP, who invited me to participate in a meeting at Quantico to discuss changes that would reduce the size of the existing VICAP crime report form. The call amazed me because I thought I was pretty much persona non grata after my friend Pierce Brooks left the unit. My problems with VICAP had more to do with where the system should have been housed—it belonged anywhere but inside the FBI, as far as I was concerned—than the size or content of the form. In spite of my criticisms of the program, I was still one of VICAP’s most vocal supporters. For the first six months of the system’s existence, one third of VICAP’s database contained murder cases from the state of Washington that I had diligently researched for them. Differences aside, I was willing to advise, and my input about what and how information should be collected seemed to be highly regarded by those who ran VICAP.

  The initial VICAP form was long—three volumes holding 68 pages of questions. It contained everything you always wanted to know about murder and more. Pierce Brooks had done a stellar job of consulting with experienced homicide detectives to produce the most comprehensive homicide investigation data collection form ever conceived. Unfortunately, its length frightened some officers. Typically, old-timers balked at filling out additional paperwork, especially a 68-page form for only one murder. The fact that the form was created without securing local law enforcement agencies’ full commitment created some doubt about their ultimate participation. Concern from police over the size and complexity of the form itself was an obstacle that threatened VICAP’s survival. In addition, I believed that certain information that was not collected on the form needed to be added and entered into VICAP’s database. Without that information, comparisons could not be made effectively among the different homicides the database system was created to relate. Therefore, the form had to be completely revised—information had to be added and deleted.

  In order to encourage local police agencies to use the form, VICAP decided to greatly reduce its size. There was a small group that was to decide what stayed on the form and what went—Terry Green, VICAP homicide case specialist; Ken Hanfland and Jim Howlett, VICAP crime analysts; Captain Gary Terry, Hillsborough County Florida Sheriff’s Department; Lieutenant Dan Scribner from the New York State Police; and me. Pierce Brooks was invited, but didn’t come. He disagreed vehemently with any form reduction.

  Pierce called beforehand and warned me that the abbreviated contents of the new form had already been decided. He felt the FBI was only using us for support and would say someday, if the form was a failure, it was our fault. Pierce and I had talked long hours about how to make VICAP a success, and it wasn’t necessarily by reducing the content or size of his form. VICAP needed to regionalize its activities.

  In the few hours that we discussed the reduction in size of the VICAP form, an idea hit me. The VICAP program was exclusively a serial-murder tracking program. Its only function was to compare an incoming murder case against the database of serial homicides to determine if another murder was committed by the same person. As we discussed which questions on the long form to eliminate, I saw the importance of collecting information about all types of murders, not just those in a series. I asked Terry Green, who would later assume the duties of program manager of VICAP, “If a detective called you and asked if ‘so and so’ was a murder victim, would you give him the information?
Or if he asked, ‘Do you have a victim who was shot with a .22-caliber firearm and found in a Dumpster?’” Initially, Green said, “Sure.” Then, suddenly, he saw my point. There were specific questions related to nonstranger murders that could be answered by querying the database. If VICAP provided that type of service to local law enforcement officers on a daily basis, their personnel would be overwhelmed with inquiries from thousands of investigators. They could not concern themselves with routine murders; there were too many and, of course, too many questions about them.

  The situation was quite different for a statewide system that could monitor fewer murder cases. Therefore, I decided to use the VICAP form as the basis to collect information, and develop a state system in Washington State that could forward serial murder cases to VICAP and at the same time provide information to local investigators about routine murders.

  Back at the FBI, the VICAP analysts were eager to deal with a form that was more suited to crime analysis and less of a format for the greater understanding of all murder investigations. They wanted to make some questions on the form more general, such as including broader categories for locations of events rather than specific locations. Also, they wanted to remove the redundancies from one volume of the original form to another. We were obliging and went along with their proposals. Still, I had a lingering bad feeling about the form’s contents. I wondered if they would be effective at connecting murder cases committed by the same person with that trimmed-down form.

  When I returned to Seattle, I consulted my closest fellow detectives. We discussed the possibilities of a state system, but came up with the same scenario over and over: no funding and nary an agency willing to bear the cost. One day, I was looking through the National Institute of Justice’s program announcement. It had a specific section that dealt with the apprehension of violent offenders. I reasoned that it was possible to research the ways murder cases were solved in order to better understand how murderers were apprehended.

  For years, the viewing public had watched homicide investigation as portrayed in Dragnet, Naked City, Kojak, and Columbo. In addition, entire communities have felt the terror of a real-life brutal killer such as Ted Bundy, William Heirens, Ed Gein, Edmund Kemper, or the Boston Strangler in their midst and disrupting their lives. Much to my amazement, with all that fascination with the gruesome details of murder and its investigation, no one had ever studied, empirically and objectively, the methods by which police investigators catch killers. Researching solvability factors in murder cases would entail the establishment of an information system to store and analyze the data on murder. So in September 1987, based on a grant proposal that I and others in the state law enforcement hierarchy put together, the Criminal Division of the Washington State Attorney General’s Office was awarded a $228,000 grant to study murder cases in Washington State and set up HITS.

  HITS has become a murder and sexual assault investigation system that collects, collates, and analyzes the salient characteristics of all murders and predatory sexual offenses in Washington. HITS relies on the voluntary submission of information by law enforcement agencies in the state of Washington. To date, these agencies have submitted data on murders, attempted murders, missing persons where foul play was suspected, unidentified dead persons believed to be murder victims, and predatory stranger rapists.

  We developed the HITS form, in place of the VICAP crime report, to collect information about murders in Washington State. The HITS form was a combination of the shortened version of the VICAP form and Pierce Brooks’s long form. In all, there were 250 questions.

  HITS Starts Working

  HITS, similar to the national FBI-run serial murder tracking system, VICAP, had a different mission. HITS was at first confined to Washington state and was designed to help homicide detectives throughout the state by providing them with a relational database to which they could report homicides and from which they could retrieve the salient data that would help them compare and contrast their cases.

  From the start, HITS was an immediate success, helping us bring data together to solve murders committed in different parts of the state by the same person. Our first significant connection involved two murder cases whose victims were found on opposite sides of the state. A Spokane detective filled out the HITS form for a murder of a male hobo found stabbed to death in a railroad yard. The investigation had long since been inactivated. Since HITS contained all murders and not just the potentially sexually sadistic serial murders the VICAP database contained, the Spokane murder was compared to all murders in the state. I found a similar case investigated by Cowlitz County authorities about 150 miles south of Seattle. This time a male transient had been found stabbed to death in a railroad car two years after the Spokane murder. The HITS comparison revealed that a person previously considered as a possible witness in the Spokane case was listed as the suspect in the Cowlitz County case. Spokane detectives were alerted that their “witness” was a possible suspect in the Cowlitz County case and their own. The same person was then identified as a suspect in a similar murder in a Midwestern state. When this same information was disseminated to other law enforcement agencies throughout Washington State another similar case was reported to HITS from Thurston County authorities, 60 miles south of Seattle. Without HITS, the three police agencies would not have known they were investigating similar murders.

  In another example, members of the Green River Murders Task Force were trying to determine the full name and additional information about Diane Merks, who was believed to have been raped and murdered somewhere east of the Cascade Mountains about five years earlier. An anonymous caller had reported that she was killed and found in much the same manner as some of the Green River victims. Task force members were unfamiliar with that death and wanted to look into the circumstances of that case to determine if there was a connection to their series of murders. Prior to HITS, a search for the victim’s case would have consumed many hours of an investigator’s time. But the HITS computer, with a Soundex program for name searches, identified in a matter of seconds a female homicide victim from Franklin County, Washington. The name of Diane Merckx was provided to Green River investigators, a spelling that would not have been previously found with a name search in any existing criminal justice computer system. Her skeletal remains were discovered in 1978, four years prior to the publicized beginning of the Green River cases.

  Since HITS was enhanced to include sexual assaults in July 1990, the program has been successful in providing information on over 300 murder and rape investigations. After an extremely brutal rape and attempted murder, a request was made to the HITS unit for information about rapists who have certain physical characteristics and a previously recorded modus operandi. HITS staff were able to provide the investigating detective with a list of known sexual offenders who had been released from prison in the past five years and the areas to which they had been released. The detective assembled a montage of photographs from those rapists, and the victim immediately identified one of the offenders as her assailant.

  The major financial benefit to local law enforcement agencies in Washington is that they incurred no cost for the implementation of HITS; its development and implementation costs were borne by the U.S. Department of Justice and the Washington State Attorney General’s Office. The ongoing operational costs are also paid for by the Washington State Attorney General’s Office.

  The only cost to local agencies is the investigator’s time for filling out the HITS form, which is 20 to 30 minutes. This time is negligible when compared to the time previously spent by an investigator trying to obtain by other methods the information that HITS can provide in a matter of minutes.

  Local agency participation in the FBI’s VICAP program has become systematic in the state of Washington, resulting in added conservation of investigator’s time, since the investigator is not required to complete two forms. The VICAP information is collected on the HITS form as well. Therefore, investigators and departments receive the
benefits of participating simultaneously in both programs.

  HITS Goes Regional

  On October 8, 1991, Washington State Attorney General Kenneth O. Eikenberry and Superintendent of Oregon State Police Reginald Madsen signed an agreement to link the two law enforcement agencies in their efforts to fight violent crime. The agreement allows Oregon access to Washington’s HITS program. The Oregon State Police (OSP) are able to electronically transfer their violent crime information into the HITS computer. Through remote access, OSP has the capability of coordinating their violent crime information. Data from over 700 murders committed in Oregon during the previous six years have been entered into the HITS computer. As a result, police and sheriff’s investigators from Washington and Oregon have ready access to violent-crime information for their investigations.

  HITS underscores the importance of open lines of communication and coordination among police agencies, prosecutor’s offices, coroner/medical examiner’s offices, and crime laboratories in murder investigations of common interest. The HITS unit functions as the central location and repository where Washington’s various murder and sexual assault investigators can readily find information about murder and sexual assault cases. The lack of such a process was exploited by serial killers such as Ted Bundy and the Riverman. Its implementation demonstrates that the timely and coordinated sharing of comprehensive information is the key to successful violent crime investigations.

  HITS is a model that other states and local agencies have replicated or adapted to their own needs. The creation of the methods and procedures for data collection, the collection instruments, routine analyses, and computer programs are already benefitting other jurisdictions that have discovered the importance of coordinating and sharing violent-crime investigation information.