GEOGRAPHIC PROFILING

Dr. Grover M. Godwin (angmg1@uaa.alaska.edu) works in the Justice Center at the University of Alaska where he teaches courses in criminal profiling and serial murder. He has published Hunting Serial Predators, which offers a unique multivariate analysis approach to profiling, and his own computer system for geographic profiling is called Predator. Holding a doctorate in investigative psychology, he lectures widely and has served as a law enforcement consultant in developing geographical profiles. Currently, he is one of the leading authorities on this method. To his mind, geographic profiling is more scientific and more accurate than the inferential methods used by the FBI.

According to Godwin, the steps in developing a geographical profile first involve deciding on which of two different ideological positions you will take:

1. The sole use of body dump sites
2. The use of both the body dump sites and victims' abduction locations (or where a victim was last seen).

His research (with David Canter) on fifty-four American serial killers suggests strongly that the abduction sites significantly affect the predictive power of a program when determining the offender's home base area, so he prefers option #2. This involves getting more data, but it also makes the analysis more solid.

"The site where the victim was last seen," he says, "can be developed from any number of sources, such as eyewitness accounts, visual sightings, telephone conversations, and official documents like traffic citations, police field reports, jail booking logs, long distance calls, toll records, and credit card receipts."

Using this information, the next step is to determine the geographical coordinates of both the location where a body was found and the physical address of where the victim was last seen or abducted from. Godwin feels that the best measures are made by visiting all crime locations and obtaining geographical coordinates with a global position unit (GBS).

He then enters the data into Predator, although he points out that besides CGT, there are two other programs available: Dragnet, developed in the Investigative Psychology Department at the University of Liverpool, England, and Crime Stat, developed at the National Institute of Justice.

The program produces the same type of 3-D map as Rossmo's, displaying various colors that depict different results. "For example," Godwin explains, "in the Predator program, a light-colored area suggests a high probability of the offender living in that area, while darker color areas suggests a low probability."

A 3-d profile

His main interest is to provide a psychological theory about the spatial behavior of a certain offender, i.e., what caused the offender to travel the way he did, based on data about how killers interact with their victims. Godwin's ideas were affirmed in 1996 in a serial murder case in North Carolina.

He had read about the crimes in a Raleigh newspaper and fed whatever facts he could find into his database, which at the time consisted of information from 728 victims. The police resisted the idea that the murders were connected, yet Godwin was able to show clear links. The victims were all black women who were choked and beaten, and the geographical pattern indicated a relatively small area of operation, mostly around railroad tracks. When asked by the newspaper to provide a profile, he said that the offender was an explosive person between 28 and 35 who would commit his acts in an unplanned but stylized burst of violence. He also indicated where he believed the killer probably lived.

The day Godwin presented his profile, John Williams, Jr. was arrested for one of the crimes. He soon became a suspect in the others, as well as in five rapes. It turned out that Godwin was correct in his overall assessment, but most specifically in the geographic pattern as a clue to the offender's home. "In the John Williams, Jr. case," Godwin states, "I predicted within one block where he livedwhich is more accurate than any other geographical profiling system."

To explain how his program works, he says, "Killers have a certain kind of place in mind, where experience has taught them that suitable victims can be found. Each subsequent trip to these locations forms something of an analogy with previous successes, modified by experience and perhaps intelligence gained from previous murders. The killer's perception will be shaped both by actual characteristics and those inferred from factors such as where a victim hangs out and with whom."

Then the killer tends to go about his routine activities until the opportunity arises to snatch someone. He may have a passing conversation with the potential victim, see her from afar, or even work with her.

"The situational context within which network interactions occur is critical to understanding the hunting patterns of a predator," Godwin explains. "For example, if the killer targets victims in a location at which contact is likely to be witnessed, the chance of detection will increase."

That means that analyzing potential settings where a victim may have had contact with a killer can help to narrow the focus of an investigation to promising areas for locating witnesses and for people who may have survived an attempt at grabbing them.

Furthermore, according to Godwin, the areas of greatest risk from serial killers include:

• Urban subcultures (bars, night clubs, and red light areas)
• Isolated landscapes (parking lots, jogging paths, and rest areas)
• Areas with a high concentration of elderly and poor individuals
• Derelict areas of a city
• University campuses

"These landscape layouts," he points out, "provide the serial killer with ease of access and escape routes to avoid detection. For example, of the five high-risk victim-targeting areas, the university campus appears to be a safe place. However, university campuses have certain landscape features, such as large isolated parking lots, which make them ideal for hunting and abducting victims. Ted Bundy is a classic example of an offender who targeted victims on college campuses."

For criminals, the home environment is familiar and predictable, with certain people becoming a focal point of the "experiential space." Godwin adds that "the tangible home area becomes an enduring symbol of self, of the continuity of their
experiences. If their crimes develop as an elaboration of their daily activities, then it would be predicted that the home would be geographically and symbolically central to their criminal activities."

In other words, the places where criminals shop, eat, and get involved in recreation play a significant role in defining their crime awareness space.

Even if the emphasis on targeting specific areas at high risk for crime provides a more accurate profile assessment than trait analysis, all profiling approaches need to be used as one tool among many in any criminal investigation. Profiling is not magic; it has its limits.