Mad City: The True Story of the Campus Murders That America Forgot

by Michael Arntfield

  The second theory is more frightening. It suggests that, while the number of domestic and intimate partner murders has been decreasing in recent generations through improved social services and intervention techniques—in fact, the US murder rate as a whole has been decreasing steadily over the last thirty years—the number of stranger-on-stranger murders has actually been increasing. Historically speaking, domestic murders pointed to clear and obvious suspects and motives, suspects frequently being collared by uniformed first responders at the scene. On the other hand, murders with no obvious suspects and no apparent motive—whether random, personal, or sexual in nature—are essentially impossible to solve absent some lucky break such as a tip from the public or a deathbed confession. A single random-stranger crime is essentially a needle in a haystack. In the case of a serial killer, new individuating details are often revealed with each new case until linkages can be made after a comfortable and emboldened killer eventually cuts corners and makes a sloppy mistake that puts the police on to him or her.

  But there is also a third and more controversial theory that tries to explain why America’s unsolved murder rate is climbing while violent crime and the total number of murders are actually going down—the X-Factor. It describes the unforeseen human variable that is less about a lack of investigative smarts as it is the deactivation or otherwise purposeful derailment of investigations to suit the objectives of those occupying middle and senior management positions. These are, for the most part, the decision makers and pencil pushers who often have little involvement in day-to-day investigative work and who essentially occupy bureaucratic roles—the puppet masters. In the business world, this myopic and often toxic management style has little regard for what effect these decisions have on the morale and lives of subordinates and other stakeholders. Such decision makers have even been called corporate psychopaths—people whose insatiable need for admiration and penchant for grandiosity is a veiled form of the same sadistic behavior exhibited by criminal offenders in the outside world. In fact, some variations of the corporate psychopath have been found to score on par with convicted sexual murderers in the standardized forty-point Psychopath Checklist Revised (PCL-R) test that has for the last several decades, despite some controversy, been the gold standard for evaluating offenders in both clinical and forensic environments.

  Recognizing the damage done by psychopaths who manage to infiltrate the workplace (positions such as CEO, lawyer, surgeon, television personality, and police officer being among the most overrepresented of jobs), many progressive organizations have implemented their own test to screen for psychopathic indicators during the hiring process. Because no such testing yet exists in law enforcement beyond some rudimentary—and easily defeated—psychometric true-false evaluations completed at the time of initial hire, there is by comparison no way to measure or account for such psychopathic tendencies in the police workplace that might eventually be exhibited by senior managers. As the proverbial elephant in the room, the role of psychopathic behavior in police case-management practices—an investigative X-Factor that no one seems prepared to confront or address—remains the most plausible explanation for how and why such egregiously outrageous decisions are often made regarding unsolved homicides. It might also explain the reluctance or outright refusal to revisit what appear to be wrongful convictions. This sometimes means willfully ignoring new developments and exculpatory evidence, the infamous cases of the “Central Park Five” and “West Memphis Three” being but two examples that still resonate widely with the American public. Leaders in law enforcement in such cases are effectively prepared, it seems time and again, to ignore or betray their oaths in name of empire building—tin-pot tyrants.

  In the case of the Julie murders, apart from the mishandling of the other Mad City sex slayings that came before and after and the influx of tips provided by Linda as she pursued Jorgensen, the X-Factor looms large as the most plausible explanation for how the dual investigation went so wrong. It remains, in fact, the only explanation for linkages between the Julie cases being so summarily denied, the pathologist’s ruling of suspected homicide in the Speerschneider case being questioned, and the maligning of the victim in the local and state press by police. It’s also the only explanation for the apparent refusal or failure to recognize the identical surface burials, transportation distance and vector, the specificity and limited access to disposal sites, the stripping of the bodies and collection of souvenirs, and the indeterminate cause of death—manual strangulation or suffocation implied in both cases.

  After factoring in the similarities in victimology—age, appearance, last seen at or leaving a bar near UW—there is simply nothing in law, current criminal investigative analysis methodologies, or old-fashioned detective work to justify the way the Julie murders were handled, particularly the Speerschneider case. Someone, for reasons we may never know, perhaps pride, perhaps a “my way or the highway” management style, stood in the way of these cases being properly linked or investigated. Although the discovery of Speerschneider’s body should have been an appropriate time to raise the alarm, the opportunity was missed given the way the Julie murders were characterized as being unrelated. Even if the Capital City Killer legend of years past was fiction, these last two cases at least had some striking similarities. It was not inappropriate to at least consider that a serial killer may have been at work. The Madison powers that be pulled the oxygen out of any such talk by taking the stance they did, effectively failing the Julies—and the whole of Madison.

  Just over ten years after the recovery of Julie Speerschneider’s body, all documents relating to her investigation were also destroyed. On whose orders, no one ever could determine, and we will also likely never know. What is clear is that, by then, whoever was behind the curtain, pulling the strings on the Julie murders, certainly would have had reason to worry, to cover his tracks. As early as that summer—July of ’81—the cavalier manner in which the Julie murders had been dismissed by area police would reveal itself as a mistake with another grisly find in the offing.

  Ghost Camp

  Out in the hinterland of Dane County, in a densely wooded area near the town of Westport, four archeology students from UW made their way into the woods on the muggy afternoon of July 16, 1981, a Thursday, in order to map new territory for a proposed nature preserve in the tradition of the university arboretum. Lake-effect humidity moving in from the majestic nearby Mendota made the trip through the region, plunked squarely within the Madison census metropolitan area, more arduous than expected as the foursome lugging cumbersome surveying equipment and paper maps made their way to Camp Indianola just south of what locals knew to be the old Reynolds farm. The farm, owned by local Westport resident Henry Reynolds, had become the de facto landmark in the area since the summer of ’67 when Camp Indianola was leveled by a late-season tornado that killed two women living nearby. The camp had just let out for the summer, saving hundreds of potential child victims who had vacated the grounds the day prior, but the owners never recovered. The camp, first run by a UW psychology professor named Frederick Mueller, never reopened. By the summer of ’71 it was a veritable ghost town peppered with the remains of summer memories—the detritus of the wooden lodgings and outbuildings that had once stood there. It was a lost summer camp in much the same way that Lake Forest had become Madison’s forsaken city.

  The private boys’ summer camp, first opened in 1910, had seen some notable Wisconsin and other Midwestern residents pass through the gates over its fifty-seven-year history, particularly after the operators started placing an increased focus on theatrical productions and dramatics ahead of athletics. These included famed Hollywood actor, director, and venerable luminary Orson Welles, later followed by National Public Radio fixture Scott Simon, and others. That’s not to say that all time at Indianola was devoted strictly to the arts. In fact, another one of the camp’s defining features was its “Gypsy Trip,” a five-mile expedition that required campers to obtain detailed cartographic kno
wledge of every hidden nook and cranny, every isolated county road and waterway in the area—a combination of bird’s-eye reconnaissance and Boy Scout grit. Campers were also required to swim five miles across Lake Mendota and understand its current, and learn how to follow and respect the force of the Yahara River. Whoever abducted and killed Julie Ann Hall, Susan LeMahieu, and Julie Speerschneider also had an encyclopedic knowledge of the greater Madison countryside and most of its wooded areas. He knew and respected the Yahara—he was following its path. He also knew the Lost City, maybe as a lovers’ lane voyeur, maybe as an adventurer who’d scouted it more than once. As the hapless survey team dispatched by UW would soon discover, he also knew Camp Indianola.

  The decaying old boathouse at the abandoned Camp Indianola borders the spot on the map where, in July of 1981, the body of Shirley Stewart was discovered by UW archeology students in the nearby woods. Courtesy: Wisconsin Historical Society.

  Just prior to 4:00 p.m. on the sixteenth, the archeology majors happened upon what appeared to be an addendum to the Julie murders—the decomposed remains of missing seventeen-year-old Shirley Stewart. The native of Middleton, a Madison suburb, had been placed by her killer in an identical fashion to that of other recent victims. Covered in deadfall only a little over fifty feet into the woods and off the trail, Stewart’s skeletonized remains were still relatively well intact considering the wildlife and scavengers that roamed freely in the area. The maid uniform she was last seen wearing at the Dean Clinic the night she vanished was, however, found nowhere near the body. As with the others before her, all of Shirley’s clothing, jewelry, and personal items had been plucked from the corpse at some point prior to the killer leaving her naked and supine on the forest floor near a set of old summer camp storage buildings. As with the other deaths, Phil Little, the coroner replacing “Bud” Chamberlain, revealed that no cause of death could be firmly established. Again, as with the last three deaths, the state of the skeleton seemed to rule out severe trauma, such as a gunshot or bludgeoning. The inference then, as it is now, was that the murder amounted to another choke job on a random stranger—a sexual homicide.

  Given that all estimates pointed to Shirley being at the site since the night she vanished or shortly thereafter, the arithmetic was simple. Her death brought the total number of victims to four, found in similar ways, in Madison between May 1, 1978, and January 2, 1980—less than two years. Four women ranging in age from seventeen to twenty-four had been snatched off the street, murdered, and dumped naked in remote locations no further than thirty minutes from the city center and in locations strategically selected for their isolation—not to mention their apparent expressive value. All victims had been stripped of all clothing and all jewelry save the lone ring left on LeMahieu, the elastic hair band seemingly overlooked in the Speerschneider murder, and a Timex watch—later ordered destroyed—discovered near Stewart’s body. All the women had been covered in deadfall in carefully executed surface burials, and all had been strangled or suffocated in some capacity based on the available evidence. Whether sexually assaulted or not, the motive in every case had been explicitly sexual, and the offender’s signature remained constant with all four victims. Although the Capital City Killer might have started as a concoction of the lunatic fringe, this latest volley of murders—every murder after Debbie Bennett—seems to have made it a reality.

  As Linda followed the trail of Jorgensen—the city’s first serial killer of the era—on one side of the country, it appeared that another one had made his way to the Mad City just as Jorgensen had once done. Maybe it was the legend itself that had drawn him. Or maybe, unlike Jorgensen, he was an existing local who had been created by the city itself—the harvest of what the previous two decades had sown. Maybe he was once a small boy with incipient perversions and violent predications who had been sent to summer camp at Indianola to get straightened out. Maybe that’s exactly where he ended up cutting his teeth.

  Repudiated

  The reality is that the center block of murders to plague the Mad City during its thirty-year reign in sexual violence—the Julie murders plus the LeMahieu and Stewart slayings—overwhelmingly point to the work of a single offender. It’s the legacy of a serial killer, one born from a legend the police sought to hide from rather than repudiate, one who disappeared as quickly as he materialized. By the time he’d come and was gone again, like Jorgensen before him, other lone wolves would follow in his wake and target Madison’s young, most vulnerable, and most easily forgotten. It’s a story of a serial killer whose crimes filled the middle innings of one city’s sordid but suppressed criminal past and, as a consequence, it’s a story without an ending. The killer—the heir apparent to a local legend that started soon after Camp Indianola was shuttered—will in turn most likely never be caught. Assuming he’s still alive, absent a corroborative confession, there is simply nothing left from these cases to test using modern scientific techniques save one—what’s known as geographic profiling.

  Owing its origin largely to the pioneering work of senior Canadian police detective D. Kim Rossmo’s PhD dissertation, geographic profiling employs a series of calculations based on what’s known as the “buffer distance” between known crime scenes, including body disposal sites, attack locations, areas where evidence is recovered, or simply an estimated location where offender-victim contact was first initiated. Based on the calculated buffer distance, a pattern in an offender’s movements can be inferred by relying on the assumption that, especially in the case of murder, the killer comes and goes from a regular home base, usually his primary residence. The more of these confirmed locations the police have to work with, the more accurate the ensuing calculations and the more precise the estimated “hot zone”—the most probable area where the offender’s home base is located—with the operant theory being that serial offenders in particular will tend to follow repetitive patterns in their behavior. While the MO and disposal sites may change between crimes, the primary and secondary crime scenes will, when examined in aggregate, eventually begin to paint a picture of how the killer navigates his space and exploits his physical environment in selecting attack and disposal locations. The proprietary program designed by Rossmo for this purpose, known as RIGEL, has been employed successfully in a number of high-profile serial crimes, as have variations of RIGEL relying on the same mathematical principles. These cases include, most notably, when geographic profiling was used to zero in on the home base of a serial rapist named Randy Comeaux in Lafayette, Louisiana.

  By 1995, Comeaux had snuck into the homes of at least seven women in a middle-class Lafayette neighborhood after beginning his spate of home invasions in 1982. Using a flashlight to blind his victims, he ambushed them while they were asleep. Comeaux would then sexually assault and torment the women before forcing them to shower and later giving them a bizarre fatherly lecture on the security oversights of their homes that had been their Achilles’ heel—how they had essentially allowed him to break in and brutalize them through their own negligence. Once the attacks were definitively confirmed as connected to a common serial rapist, the task force investigating the crimes was nonetheless disbanded in 1998 after coming up dry.

  Then, an article on Rossmo’s innovative and—unlike conventional behavioral profiling—scientifically rigorous geoprofiling method was seen by a senior detective in the pages of The Police Chief magazine. In a last-ditch effort, the Louisiana cops contacted Rossmo and a profile using his formula was developed and indicated that the hot zone was in fact immediately next to the neighborhood where the attacks had clustered over the thirteen-year period in question. Later, Randy Comeaux—actually a police officer with the Lafayette PD who previously had inside information on the task force’s activities—was revealed to be the elusive serial rapist. It was also revealed that he lived in the mean center of the hot zone. It was the first of several big wins and validations for geoprofiling and the RIGEL program, one that has been used in cases of serial murder, serial rape, and serial a
rson—all compulsively sexual crimes that will bear out common denominators in attack-retreat patterns.

  The method has, since Comeaux’s arrest in January of ’99, seen generous fictionalized treatments in some prime-time cop shows, most accurately in the FX series The Shield, but is otherwise strangely still very much in a beta stage. It’s a system that has been woefully underutilized by law enforcement, and for no justifiable reason. Somewhere between being confused by the applied mathematics and computer schematics, and being irrationally suspicious of outsourcing serial investigations to the handful of experts around the world who know how to apply the formula—Dr. Rossmo having rightfully since left policing for academia—geoprofiling remains curiously arcane in spite of its successes. Many police managers also, in fairness, simply don’t know it exists and that it’s a logical next step in cold cases when looking to narrow the size of the haystack. They should.

  A geographic profile, applying Dr. Rossmo’s formula and using a variation on RIGEL, indicates that the four connected Madison sex slayings were all committed by a killer living within this darkest region at the city’s center and circumscribing both Lake Mendota and the UW campus. It’s a narrow circular band that simultaneously points most strongly to the nearby community of Five Points and also to the village of Windsor, both exurbs of Madison. One or both areas reflect the most probable “home base” of a single killer over the course of the murders, factoring in both the abduction sites and disposal sites as the data points. The combined male populations between both locations would, in 1981, have numbered fewer than three thousand. Courtesy: Western University Cold Case Society & Peter Leimbigler at the University of Toronto.