Free Novel Read

The Bulldog and the Helix Page 13


  “It’s first thing in the morning, and I’m sitting there having a coffee. I look outside the glass door, and there is Gurmit Dhillon having a cigarette with his sister. He puts out the cigarette in the ashtray, right in front of me, then walks in through the door past me, sneering at me.” Smith happened to be carrying his briefcase, which contained, along with notebooks, all manner of evidence-gathering materials.

  “I open my briefcase, take out a pair of gloves and an exhibit bag, walk outside, and take the one and only cigarette butt that’s in this ashtray—it’s a sand-type ashtray. I put it in the exhibit bag, seal the exhibit bag, mark it with date, time, and initials, and put it in my briefcase. I take off my rubber gloves.

  “The deputy sheriff witnessed all of this, and he was smiling because he had clearly clued in to what I was doing. I threw my gloves in the wastepaper basket and got my notebook out. I started to write notes about gathering the cigarette butt and felt pretty proud of myself because, if this case went to appeal, we wouldn’t need to rely on [the warranted sample] because now I had a cast-off DNA sample as well. Then I looked up, and Mr. Chamberlain was hovering over me. He said to me, ‘I see what you’re doing,’ thinking I was backdating my notes or something, not realizing I was making current notes about a seizure I’d just made from his client.”

  Once again, the effort proved to be unnecessary, but working on a prosecution so fraught with precedents, no one wanted to lose this case on a point of procedure. Well after the consent sample got thrown out, but after the Richmond testing brought the odds back up, Smith said it was merely a contingency. “I now know I can do this. And if there is an appeal, and the appeal is successful, and a new trial is ordered, I now have DNA from Dhillon that will be admissible.”

  LYLE PRICE WOULD also face Russ Chamberlain on cross-examination. But the defence attorney would discover that this was one confident, articulate (former) potato farmer. Despite his place in the chain of evidence, however, Price says he did not figure significantly in the Crown’s case—at least, not based on the amount of time on the stand. “I was [in Victoria] for quite some time, but I didn’t testify for very long,” he said. “They went over the details of finding [Carolyn]. Then I got cross-examined on, ‘How did I know it was a four-wheel drive vehicle that was in there?’”

  Eighteen years later, Price still reflects the exasperation of having had to spell out the obvious to a questioner who did not appear to have any knowledge on the subject. “He said, ‘So how did you know it was a four-wheel drive? It could have been any vehicle.’ I said, ‘Well, it was pretty muddy; it had traction tires all around.’ The four tires were traction-lug, because you could tell where they turned in the mud. I said, ‘Obviously, it would have spun if it hadn’t been an all-wheel or four-wheel drive type vehicle, because it was that slippery. It had been mixed rain and snow that night, and it was wet down there.’”

  Price said that, at that point, unable to establish much in the way of doubt as to the identity of the suspect vehicle, Chamberlain shut down the cross-examination. “He said, ‘Oh!’ and basically dismissed me,” Price said. “He didn’t want to get into it any further.”

  The photographic and plaster cast evidence of the tire tracks were especially clear because of the mud and the fact that there was a bit of an incline that caused the heavy tires to dig deeply and leave a clear impression, Price explained. The chances of a two-wheel-drive vehicle being mounted with heavy-lug tires all around were pretty much zero. It was a low-percentage line of questioning, and Chamberlain obviously didn’t want to waste a lot of time, he said.

  ON FRIDAY, NOVEMBER 13, the forensic pathologist, Dr. Rex Ferris, took to the stand. In 1988, after learning about the new science of forensic DNA, Dan Smith had contacted Ferris, who was considered Canada’s leading expert. Ferris advised the court that the autopsy of Carolyn Lee on April 16, 1977, had been performed by Dr. Robert Clark, who’d died the previous February. The pathologist recounted in excruciating detail the massive damage inflicted on Carolyn’s young body during the savage attack. The November 16 headline in the Alberni Valley Times was no less graphic: “Carolyn Lee Likely Face Down When Fatal Blows Dealt.” Ferris testified that two and “probably more” blows were delivered to the back of the girl’s skull while she lay face down in the mud. “The blows shattered a twelve-centimetre hole in her skull, he said, and caused a fracture down to the top of her spine.”

  Ferris testified that, while it was possible the damage was caused by kicks from a steel-toed work boot, he believed it was more likely that the assailant used an object such as a tire iron or a rock. “Her assailant would almost certainly have been splattered with blood and brain matter, he said.”

  A scrape on Carolyn’s right thigh indicated she had been dragged over a sharp edge—likely a rock—while she was still alive. Ferris testified that Carolyn was raped before she died, and likely survived for as long as thirty minutes after she was beaten. If that were indeed the case, according to the timeline established by Dhillon, the accused was already back in town, drinking a beer, while the child drew her final agonized breaths in the muddy potato field.

  MONDAY, NOVEMBER 16, was Gurmit Dhillon’s forty-eighth birthday. On that day, his ex-wife, Sharon McLeod, took the stand. McLeod testified that, although the police conducted a search of her home immediately after Dhillon’s Blazer and its tire tracks were linked to the crime scene, police did not interview her. When police arrived with a warrant, McLeod was alone in the house but waited outside during the search that included seizing her vacuum cleaner bags.

  McLeod testified that she did not give a statement to police until November 26, 1983, after she had separated from the accused. McLeod then recounted the sequence in which, after Dhillon did not arrive home by 6:30 PM, she drove to several bars to try to locate him before returning home and seeing him drive past on Third Avenue at about 6:45. The jury now heard what she had first told police fifteen years previously.

  In a cross-examination that extended into the next day, Chamberlain vigorously attacked McLeod’s credibility, accusing her of “putting together this story out of vengeance on your husband.” Calling it “gilded evidence” and “fake evidence,” the defence counsel attempted to break McLeod’s composure. “I say you’ve made the whole story up,” he suggested.

  “That’s not correct,” McLeod answered calmly. Noting discrepancies between her trial testimony and her 1983 statement to police, Chamberlain pointed out that she didn’t mention phoning the Beaufort Hotel or driving there, that she didn’t tell police she saw her husband drive past, heading toward Ship Creek Road, and that she didn’t mention finding an earring in the Blazer. “That’s correct,” she replied, equally calmly. When Chamberlain pointed out that, at the time of her statement, she said she did not know whether the mud she washed off the Blazer came from the foundry, she replied simply, “That’s true.”

  Chamberlain kept the pressure up on the second day of his cross-examination. Pointing out that McLeod testified that she knew Dhillon was a suspect in the Lee killing, he suggested that she must have been “a brutal monster” to have brought her daughter and two younger sons to live in his house. “That’s not correct,” she replied. “I simply didn’t believe he was the killer.” McLeod said she soon discovered that Dhillon was a hard drinker and a womanizer and that she would regularly make calls to local bars looking for him. And she also began to collect jewellery off the floor of his truck.

  McLeod testified that on the day she first spoke to police, she actually went to the detachment to speak to speak to officers about one of her sons who was involved with a youth gang. But while she was there, the conversation turned to her estranged husband and his possible involvement in the Lee murder. That visit turned into a marathon. McLeod first arrived at the detachment at two in the afternoon, and subsequently spoke to then-GIS head Sergeant Bob Martin. She did not leave the detachment until 8:40 that evening.

  Once again, Chamberlain accused her of fabricating t
he entire story. “You were mad at Gurmit,” he said. “You made up this story to cause trouble for him.”

  “That’s not correct,” McLeod replied. Chamberlain then suggested that Sergeant Martin and the other officers colluded with her at length to put together an elaborate sequence of events for her statement.

  That was a charge that Martin refuted when he took the stand following McLeod. He did concede that her statement, which ran to thirty-eight pages, would not have consumed the five hours that ensued after he took over the interview in mid-afternoon. Martin said there was likely a break in between, but it was not part of the written record.

  ON NOVEMBER 19, Anne-Elizabeth Charland took the stand to testify that the material taken from the body of Carolyn Lee matched blood samples collected from Gurmit Singh Dhillon. Charland, like Stefano Mazzega, was a reporting officer, but she was based in the biology section of the RCMP’s forensic laboratory in Ottawa. She began her testimony with a recounting of the long and at times frustrating sequence in which the crime scene evidence samples were successively tested for DNA.

  Charland testified that an “analyst” [Hiron Poon] also took DNA samples from Carolyn’s parents, John and Sau Lee, to recreate Carolyn’s DNA pattern and “build a family tree in reverse order.” (Eight months later, Dan Smith asked Poon to use the same strategy to establish a genetic family tree in the Jessica States investigation. But in this instance, it was the suspect’s parents, rather than the victim’s, that were sampled covertly.) Charland further noted that, despite the poor results yielded by the early DNA tests, an Esquimalt man had been eliminated as the mysterious “second suspect” because his DNA profile differed markedly from the genetic material collected at the scene. As soon as one comparison point does not match, it is considered “an exclusion,” she explained. But “there were no exclusions” in the 1996 test of Gurmit Singh Dhillon. The match was so good that the analysts considered it “a rare event.”

  Charland testified to the effectiveness of the new PCR technology used to analyze the five large swabs collected at the crime scene in 1977 and preserved for years by the investigators. According to Charland, the RFLP process required 500 billionths of a gram of genetic material to produce a DNA profile; the latest PCR technology required just one billionth of a gram. To process the material on the vaginal swabs, scientists first had to remove the bacterial DNA from the human DNA, then the female DNA of the victim (hence the reconstructed DNA profile of Carolyn) from the DNA belonging to the suspect—or suspects. The resulting male DNA profile was then compared to the DNA from the warranted sample from Dhillon.

  In his cross-examination the next day, Chamberlain attacked both the credibility of forensic science and the integrity of the scientists who performed it. The defence counsel told the jury that the Crown was asking them to convict a man of first-degree murder based on “something we can’t see.” He said to Charland, “We’ve got to trust you to convict this man. You knew Dhillon was a suspect. ‘Get Dhillon’ was what your mission was.”

  “I have no mission of that sort, no,” Charland replied. Chamberlain then cited a DNA request written by investigators that appeared to advise her to find a link to Gurmit Dhillon while at the same time excluding the second suspect identified in 1989. Charland conceded that the wording was inappropriate, but, “It doesn’t matter anyway; I depend on the results.”

  Chamberlain attacked the results of any test that analyzed “degraded” or “rotted” material. On this point, Charland was adamant: degraded samples might not produce a DNA profile, but there was no science to indicate that they could produce inaccurate profiles. “If you get a profile, it’s a true profile,” she said.

  Chamberlain then ran the witness through an extended review of the lab and the equipment used or worn by scientists. The focus on scientific minutiae led Justice Stewart to interrupt Charland’s testimony several times. “I don’t know about the jury, but using myself as a common denominator—I’m lost,” he observed. Chamberlain continued to demand detailed explanations of the procedures, and Charland kept returning to the central theme of her testimony throughout: four pairs of coded peaks on the crime scene samples that exactly matched four coded peaks on Dhillon’s warranted blood sample. And no exclusions.

  Chamberlain found a little more currency in his contention of procedural error in his cross-examination of Richard Guerrieri, the scientist who worked at Roche Biomedical Labs in North Carolina at the time the lab received the Carolyn Lee samples. By the time of the trial, Guerrieri was a forensic biological examiner for the FBI and considered one of the world’s top experts in forensic DNA. In 1993, he opened the package of samples from Genelex Corporation in Seattle containing swabs, hair, and blood samples. At that time, Roche Biomedical had PCR technology that was superior to that used by both Genelex and the RCMP’s central forensic lab in Ottawa. Guerrieri testified that he identified five DNA sites, wrote his report, and returned the sample package to the RCMP in September 1994. In Ottawa, one of the analysts subsequently discovered that some of the hair samples labelled as belonging to Dhillon actually belonged to the “second suspect.” Guerrieri testified that the labelling mistake had been made at the Genelex lab and that, once the labelling mistake was noted, his company notified the RCMP lab. That was in early 1995.

  Chamberlain pointed out that, in two instances, the Roche report contained two incorrect notations, indicating that DNA samples had been tested when they hadn’t. Guerrieri explained that the lab did mix up the labelling of two sample extracts. “Yes sir. That was an error—a transcription error,” Guerrieri said. Chamberlain maintained that this error had caused confusion for his DNA experts, who were unable to find the corresponding test results. In response, Guerrieri noted that the mislabelled samples were not used to produce results in the Dhillon case.

  On Tuesday, November 23, a population geneticist, Dr. Richard Carmody of Carleton University, testified that the chances that any man other than Gurmit Dhillon had sex with Carolyn Lee were one in 165 million. “I think most of us would agree a number like that is pretty rare,” Carmody observed, because it meant that “99.9999994 percent of the population would not have that profile.” Carmody testified that by compiling the results of the testing by Ottawa RCMP, Genelex, and Roche Biomedical, along with those from Helix Biotech, the result of Dan Smith’s rush trip to Vancouver on November 8, scientists had put together a “ten-area match.” Further, Dhillon was not excluded by any of the match points, the scientist explained. The odds were calculated using a database of Caucasians because, as Carmody explained, Indo-Canadian people are often considered Caucasian in genetic terms.

  The next day, Dr. Ranajit Chakraborty from the University of Texas expanded on the racial sub-groups within the nominally Caucasian population. Chakraborty, another population geneticist, was the scientist Prevett had contacted after Dan Smith speculated that combining the early meagre DNA results from separate laboratories might provide a better exclusionary result. It was Chakraborty, using the “product rule,” who had calculated odds of one in 165 million, using the Caucasian population. However, the scientist explained, using a population base composed entirely of Punjabi Sikh men, those odds fell to one in 13.3 million. Given a population base of Indians as a non-homogenous, composite population with sub-groups, that number could fall to as low as one in 7.6 million, he further explained. If Punjabi Sikhs were considered a homogenous population, the odds rose to fourteen million to one, but as a composite population with sub-groups, it was one in 13.3 million, he said. That excluded 99.9999925 percent of Punjabi Sikh men as donors.

  In his cross-examination, Chamberlain pointed out that, in 1997, the Crown based its case on an eight-loci match, which produced odds of two million to one, then subsequently claimed that amazing ten-loci match and the one in 165 million odds. Chamberlain told Chakraborty that his DNA expert would testify that the jump from eight loci to ten was invalid because those additional two DNA loci were not verified as independent from other, testab
le DNA locations on the human genome. At this suggestion, Chakraborty countered that Chamberlain’s DNA expert had fallen behind the times. Not only had those two specific additional loci since been identified as independent (which would allow analysts to compound the odds), but scientists had now identified another two dozen loci on the human genome that could be tested for DNA matches.

  In previous testimony, Richard Carmody had testified that in 1996, Genelex had made an error in a DNA suit involving a paternity case. Chamberlain had told Carmody that his expert would testify that the error rate in DNA match is one in 1,000. To that contention, Carmody said that tossing around a facile “error rate” was improper procedure. “The error rate in this case is zero,” he had concluded.

  Chakraborty concurred. “From examining the material, I don’t see any possibility of error,” he said. At this point, Justice Stewart cautioned the jury not to attach too much significance to the numbers, suggesting that statistics often amount to “pious fraud.” Stewart instead suggested that jurors pay more attention to the experts’ interpretation of those numbers, such as “very uncommon” or “rare.” On that point Chakraborty agreed, terming the results of the Dhillon profile as “very uncommon.”

  THE DEFENCE MAKES ITS CASE

  On Thursday, November 26, forensic engineer Jason Gough, of Vancouver-based Baker Engineering, took to the stand on behalf of the defence. Gough testified that he’d been hired by the defence to reconstruct the timeline of the crime as alleged by the Crown using the accused as a live prop.

  On September 17, 1998, Dhillon was asked to work a full day at Alberni Foundry, wearing clothes similar to those he had worn in the spring of 1977. At about 4:30 PM, Gough testified, a “very dirty” Dhillon washed his hands and left the foundry, as he told police he had done on the day of the crime. Gough and Dhillon then drove to the site of the old foundry at the foot of Second Avenue, where he gave Dhillon a mannequin dressed in clothes similar to those Carolyn Lee wore on the last day of her life—a nylon ski jacket, blue jeans, black body suit, blue panties, and white running shoes. The mannequin was dressed in a sweatshirt instead of a green sweater and grey leotards instead of white. With a video camera running, Dhillon and Gough drove to Third Avenue and Athol, where it was believed Carolyn was abducted. From there, they proceeded along Third Avenue to Ship Creek Road and continued onto Franklin River Road to the wooded area near the abandoned railway siding where Lee’s body was found.