by Edward Humes
Ornelas had been searching for fallen branches to heap on a bonfire. He and ten other high school students had spent the evening busily getting stoned near a weedy drainage ditch outside Bakersfield, a godforsaken chunk of grit and rock surrounded by farms midway between the small Kern County towns of Shafter and Wasco. The spot had no name, but it had long been a popular place for teenage partying, handed down over generations of high school students. It was isolated yet convenient, tucked behind a convergence of highways and bathed in the dim amber afterglow of freeway light standards, littered with bottles, cans and condoms, the mummified remains of a thousand similar beer busts and trysts. The white noise of high-speed traffic on Interstate 5 provided a constant background, a dull hiss almost like rushing water. There was a clear area suitable for parking and drinking, then a winding path down an embankment to a dark drainage slough, where there was plenty of old growth and loose wood for fueling fires. It was there that Sandy Ornelas made his find.
He and another boy had walked the path gingerly, lighting their way with a makeshift torch fashioned from a beer carton. As the boys walked, they kept noticing dark spots of what looked like motor oil dotting the path. Ornelas’ buddy tried several times to ignite these spots with the tip of his torch, but they wouldn’t catch fire.
“You hold the light and I’ll go down and grab some wood,” Ornelas said when they reached the scrubby area at the bottom. He picked his way down through the darkness, grabbing a branch here, a stick there, moving fast, trying to get out of the claustrophobic space as quickly as possible. Then, as he stooped to pick up one last nicely sized piece of wood, the flickering orange light of the torch revealed that the gnarled-looking branch he was about to grasp had fingers on it. And the fingers were attached to an outstretched arm and, beyond that, a body, motionless and in shadow, reclining with its other arm flung across its forehead, as if in repose. The bloodstained figure didn’t move.
Ornelas started yelling a heartbeat later, dashing toward his friend with the guttering torch, then running right over the boy, bowling him over as he careened up the path to the parking area, to the comfort of a crowd, to the older boys who hooted at his terror for a moment, then saw his face and realized something must be very wrong. It was left to them to trudge through the darkness stepping over the spots of oil that were really blood, so they could see for themselves what had frightened their friend.
They found a teenage girl, shot through the back and the head, an inexplicable act of violence, a senseless crime, a—there’s really no other word for it—big-city crime. For all the wrong reasons, this community had finally earned the slogan it plastered on its street signs and letterheads and welcome-to-town placards. In a nation where violent crime had become the great common ground in small towns and large, Bakersfield was, at last, the “All-America City” it had so long claimed to be.
• • •
On the surface, the execution-style murder of Maria Madera Rodriguez couldn’t have been more different from Sandy Dunn’s disappearance and fatal stabbing, beyond the fact that both bodies were found in the desert. Maria was a petite and pretty Hispanic high school girl with an active, sometimes-risky sex life well hidden from her mother. She had been found within hours of her murder. There was a known crime scene, physical evidence, blood, witnesses, footprints. Everything missing in the Dunn case could be found in the investigation of Maria’s death. Yet Laura saw many parallels—too many, for her comfort—as Susan Penninger told her about the case.
By the time homicide investigators had cordoned off the area, much of the evidence—footprints, tire tracks, blood, plants and fibers—had been trampled over and thoroughly compromised by the boys, by paramedics, by coroner’s investigators and patrol deputies who wanted to see and examine the body. Before racing off to get help at a nearby fire station, all the kids had traipsed down to gawk at the body, including one boy whose wheelchair slid and skittered along the path, obliterating untold amounts of evidence in the dirt while leaving behind narrow tracks of its own.
Not all the evidence had been trampled, however. One of the deputies trained his flashlight on the parking area and found that the kids’ cars, and some of the official vehicles as well, were parked next to and on top of old tire tracks, footprints and, most important, puddles of blood that the partyers hadn’t even noticed when they pulled up in the darkness. The parked cars had preserved the evidence like lids on a jar, preventing anyone from walking over these areas in the rush to see the body. Some of the sheriff’s investigators and other witnesses would initially report that the blood under the cars appeared fresh and still damp when first discovered,24 which meant, in that hot desert climate, that the murder almost certainly had to have occurred after sundown. Otherwise, the heat of day would have left the blood dried and caked long before the body had been discovered. A print from a running shoe with a distinctive tread was found in one large, still-damp puddle of mud and blood—evidence investigators hoped to match up with a killer’s shoe.
Later, though, despite what was initially observed and reported, the authorities would claim all of the blood puddles were hard and dry when first found—they only looked wet because they had appeared shiny in the beams of deputies’ flashlights. This change was crucial, because it meant the murder could have occurred earlier in the day, a convenient revision, since the prime suspect had an airtight alibi for the late afternoon and evening. It was here, Laura decided, that the Rollins case began to resemble Pat’s, and Carl Hogan’s, for that matter. Each exhibited the same tendency of Kern County authorities to interpret and reinterpret evidence until it fit a desired theory. To the prosecution’s way of thinking, declaring the blood dry instead of wet was just correcting a misstatement, nothing more. But Laura had to wonder: If wet blood had been consistent with the prosecution theories in the case, would the blood still be wet in the police reports? If the secretary Ann Kidder had not contradicted Jerry Coble’s story, would she still be an officially credible person and a witness against Pat Dunn?
Susan Penninger had found another interesting fact that kept cropping up in the police reports in the Rollins case: The girl’s body, and everyone who went near it, came away covered with sticky brambles from a dry desert weed, the tamarisk pentandra, whose distinctive clinging burrs under a microscope resemble nothing more than a dark-winged angel with halo. The lowly tamarisk weed would become crucial in the case much later, because the killer should have been covered with the burrs. This would mark another point where evidence and theory in the case did not match up.
No identification was found on the girl, and the sheriff’s department cleared the crime scene with the body classified as a Jane Doe—identity unknown. Tracks and shell casings from a variety of firearms found on the ground suggested she had fought with one or two assailants, then had been felled by a shot, dragged, and shot again. Rape appeared unlikely, as her black stretch pants were still on. But her blood-soaked white blouse and bra had been pulled up as her body was dragged, and her bared chest was deeply abraded by the rough terrain. When the coroner’s technicians turned her body over, investigators spotted a smear of lipstick on her back, apparently from a dropped or discarded lipstick container found on the trail a few feet away. It remained unclear whether the body had simply been dragged over the lipstick inadvertently, leaving an unintended smear, or if the marks represented some kind of a message from the killer. The lead detective on the case seemed to choose the latter, writing in his initial report that the markings appeared to be evidence of a ritual killing. This belief would soon be discarded, but not because ritual murder had been disproved by actual evidence. Rather, it did not fit the emerging theory of the case.
Within forty-eight hours of the murder, two breaks turned the case firmly in one direction. First, a Bakersfield businessman named Dale Knox read in the newspaper about the unidentified body found in the desert and called the sheriff’s department. Knox said he had been driving near that very spot early on the afternoon before the bod
y was found and had seen an older-model maroon Buick or Oldsmobile speeding from the area, kicking up dust and driving in an erratic manner, even cutting him off after tailgating him. “It was some dumb Mexican,” Knox told Deputy Paul Hussey, an old friend.25 This information was passed on to Detective Randy Raymond, the homicide investigator in charge of the case.
A few hours later, Detective Raymond received a missing-persons report that had been filed with the police in Shafter, a small town a few miles from Bakersfield. A seventeen-year-old girl had disappeared from her home on Friday. She was supposed to have gone to a park to meet her boyfriend that morning, but she never returned home. The family had called all her friends and looked all over town, waiting overnight in hopes of hearing from her, then filed a missing-persons report Saturday morning. Raymond took one look at the accompanying photograph of a smiling Maria Madera Rodriguez and knew he had identified his Jane Doe.
Still, he had to make absolutely sure—and he had to know more. So he drove to Maria’s home and spoke to her distraught mother, Miriam, who knew nothing of the body in the desert. She greeted Raymond warily at her door and explained that she spoke little English. Raymond hadn’t thought to bring an interpreter, but a wiry man with a beard and an ample collection of tattoos on his arm approached and introduced himself. Don’t worry, Victor Perez offered, he could translate. Miriam gestured to him and said to Raymond, “My son-in-law,” and the detective nodded.
Raymond wanted the whole story before breaking the bad news, so he did not mention the dead girl in the desert at first. He only explained that he needed more information about Maria’s disappearance, and they talked for nearly half an hour, Victor acting as principal translator. Once he got what he needed, the detective informed Miriam, as gently as possible, that Maria was dead. The news of her child’s murder caused Miriam Rodriguez to collapse. The last thing Raymond did before leaving was summon an ambulance for her.
But he had his information. And he had a prime suspect. The story he extracted from Miriam via Victor Perez boiled down to this: Maria and her teenage boyfriend, Offord Rollins IV, had been going steady for the past six months. They had been arguing recently because Maria was seeing someone else. On Friday morning, the day she disappeared and died, Offord had called the house and spoke to one of Maria’s sisters, then to Maria, and arranged to meet her in a park across the street at 11 A.M. Finally, Victor translated this startling piece of information: Maria’s Aunt Lucy had seen Maria and Offord together Friday night in a maroon car—just like the one Dale Knox had seen speeding from the murder scene. That would mean Offord almost certainly had been the last person to see Maria alive.
When Raymond learned that Offord had been driving his father’s Oldsmobile that weekend—and that it was an older car, maroon in color—the detective knew he had just solved a murder. Within hours, he had a seventeen-year-old high school track star and class president named Offord Rollins behind bars, accused of first-degree murder and facing life in prison. Only the fact that he was underage saved him from a potential death sentence.26
• • •
So far as Laura was concerned, as soon as the sheriff’s department seized on Offord Rollins as the one and only suspect in the murder of Maria Rodriguez, the investigation’s resemblance to Pat’s became absolutely eerie. As with Pat, official pronouncements and press reports suggested a mountain of proof against the young athlete, but, in truth, there was little hard evidence linking him to the crime. There had been multiple searches of his home, with nothing to show for them, just like Pat’s case. No murder weapon had been recovered. There was no evidence Offord ever possessed a gun at all. (After initially reporting that Maria had been shot by two different weapons—a finding that would have contradicted the official theory of Offord Rollins as lone gunman—the authorities revised their opinion and asserted that only one twenty-two-caliber rifle had been used.) Nothing in Offord’s house or room matched the soil or plant material around Maria’s body, and no one who ran into Offord that day saw any signs of tamarisk burrs clinging to his clothing or car. None of Offord’s many athletic shoes—and as a football player and competitive long jumper, he had a closetful—matched any of the prints at the murder scene, including the one in the puddle of blood. The authorities could not even find anyone who actually saw their prime suspect and the victim together on the day of the murder, notwithstanding what Victor Perez had claimed.
There were, however, three pieces of evidence that supported the prosecution’s case: Authorities found Maria’s palm print on the outside (but not inside) of the maroon Oldsmobile that Offord drove that weekend, as if she had leaned against the car while waiting or talking. There were seventeen nylon fibers on Maria’s clothes that seemed to match fibers from the Oldsmobile’s seats—though the most that could be said of these is that they came from a similar car seat, as opposed to that specific car. And, finally, there was a single, tiny bit of tamarisk plant found clinging to the backseat of the Oldsmobile.
These were disturbing bits of evidence, Susan Penninger told Laura, but far from conclusive. Clearly, Maria had been around and possibly inside the car. But Offord said he met Maria at a store and she leaned against his car the day before she disappeared, which could provide an innocuous explanation for the prints and the fibers. Unfortunately, he didn’t tell the police this right away, and even led them to believe he hadn’t seen Maria at all, which not only increased official suspicions, but added a troubling building block to the case against him—an apparent lie. Offord’s account also required believing Maria had worn the same clothes two days in a row, something her family insisted had not happened.27
As for the bit of plant found in the car, tamarisk grew in many areas outside Bakersfield, and could have come from any one of numerous locations other than the murder scene. Furthermore, the significance of the burr was in question because of problems at the Kern County crime lab itself, a former dog kennel renowned for its contamination problems, which included ventilators that spewed soot into the examining rooms and a shortage of sterile, clean work areas.28 Controls in the lab were so loose that the technician who examined Maria’s tamarisk-covered body later conceded she may well have worn the same lab coat, uncleaned, the next day to examine Offord’s car, raising the possibility that she inadvertently deposited that quarter-inch bit of weed on the backseat where it was found. Another technician, who checked the car for fingerprints inside and out, had probed the interior first and saw no plant material on the seat. Besides, if Offord had really been the killer, the car—which had a dusty, cluttered interior that clearly had not been cleaned or vacuumed before police seized it—should have contained many more pieces of the clingy plant, not to mention blood and soil from the murder scene. But it did not, not even the microscopic amounts that resist cleaning and can persist for weeks. Nor were any strands of Maria’s long hair found anywhere inside the car.
Laura knew strange things sometimes happened in criminal cases, subtle changes in testimony and evidence that occurred, usually early on and almost always to the detriment of whoever the police suspected. Neither Pat Dunn nor Offord Rollins were unique in this regard. Witnesses’ memories often grew more specific with time, although common sense suggests the opposite should occur. Police officers would come to court to testify and recall crucial observations they had made months earlier, though they had somehow forgotten to note those telling details in their reports at the time—reports that were supposed to be all-inclusive. Detectives would tape-record lengthy interviews with suspects, then report key, incriminating statements were made when the recorder was switched off—through sheer happenstance, of course. These sorts of things happened everywhere, but Susan Penninger told Laura they seemed to come up in every case she worked in Kern County, leading her to name the phenomenon “The Bakersfield Effect.”
The testimony of Dale Knox, the man who had seen a maroon car like Offord’s speeding from the vicinity of the murder scene, provided a perfect example, she told Laura. Knox h
ad originally given his report on the day after Maria’s murder, speaking at length about the maroon car and the “dumb Mexican” he had seen driving it with his friend Deputy Hussey, who also had been one of the first patrol deputies called to the crime scene. Hussey had taken notes on the conversation, but after relaying the information to Detective Raymond, Hussey was told not to bother writing a report. Instead, he destroyed his notes, contrary to common police practice.29
When Raymond conducted his own interview of Knox the next day, he reported Knox had seen a black or Hispanic male driving the maroon car.30 This change was crucial to the case, because Offord Rollins was black. His very dark complexion and facial features made mistaking his ethnicity an unlikely proposition. Here was a case in which a witness’s uncertainty was more useful than a definite (and incorrect) opinion. And so, like magic, Penninger saw, the report of a Mexican driving a maroon car changed—right after Offord was identified as the prime suspect. Many months later, when the case came to trial and Knox took the witness stand, he couldn’t say it had been Offord in the car, but he wouldn’t rule him out, either.
This ambiguity could have been fatal to the prosecution, except that Detective Raymond remembered Knox describing the car as having some damage on its left side. The car Offord had driven that day also had damage on its left fender. Again, this was crucial, since it seemed to buttress the notion that it was Offord whom Knox saw fleeing the scene, not someone else in a similar vehicle. Yet Detective Raymond had not put this key piece of information in his police report at the time, nor did he ever mention it while testifying at pretrial hearings when asked to summarize what Knox had told him. It only came to light at Rollins’ trial. Prosecutors saw nothing nefarious here, just the innocent omission of one small detail in an enormous and complex case. To Penninger, it was another example of the Bakersfield Effect.