While Baton Rouge was running a multi-agency manhunt with FBI support and a suspect pool of more than a thousand men, a second man was killing women in the same city, and had been doing so for years before that search began. He was not found by the task force, the behavioural profile or the DNA dragnet. He was found by a tyre.
Sean Vincent Gillis was born in Baton Rouge on June 24, 1962, to Yvonne and Norman Gillis. His father left soon after his birth and he was raised by his mother and grandparents in southern Louisiana. His record begins in 1980, when he was seventeen, and consists of traffic citations and minor infractions. He worked in a print shop and as a delivery driver, lived on Burgin Avenue with his long-term girlfriend Terri Lemoine, and was described by the people who knew him as amiable and unremarkable. He was forty-one when he was arrested. Nothing on his record before the first killing involved violence of any kind.
The first killing was the least like the others. In March 1994 he killed Ann Bryan, eighty-two, in her room at the St James Place retirement home. She had been stabbed more than fifty times and her throat was cut. Then five years passed. Her case went cold and stayed cold, and nothing connected it to anything else until he told investigators about it ten years later.
On May 30, 1999 he took Hardee Schmidt, fifty-two, who was out running in southern Baton Rouge. He had watched her for about three weeks after first seeing her jogging. He struck her with his car and put a cable tie around her neck, and her body was left in a bayou off Highway 61 in St James Parish.
Those two did not fit what followed. Bryan was an elderly woman in a care home and Schmidt lived in an affluent part of the city and was chosen because he had seen her running. The other six were women he picked up on the street at night, most of them working as prostitutes, most with some connection to the drug trade, aged from their late twenties to their fifties. Joyce Williams was killed on November 12, 1999 and her body was not found until January 2000. The others named in the record are Katherine Hall, twenty-nine; Lillian Robinson; Marilyn Nevils; Johnnie Mae Williams, forty-five; and Donna Bennett Johnston, forty-three, who was the last, in February 2004.
What he did after the killings is why the case has the name it has. He strangled them, mutilated the bodies, kept body parts and personal effects in his house, and photographed his work on a digital camera and kept the images. He posed at least one body after death. In at least one case he ate part of a victim. The photographs were later used in evidence at his trials.
The question the case actually poses is how a man doing this went unnoticed for a decade in a city that was, for the second half of that decade, looking as hard as it knew how for a man doing something very like it.
Part of the answer is that there was more than one. Gillis's killings overlapped with Derrick Todd Lee's in the same city, and for part of that period with a third man, Jeffrey Lee Guillory. An investigation already struggling to connect its own cases had three concurrent series in one metropolitan area, and some of Gillis's victims were briefly attributed to Lee. The two men's victims were in fact quite different. Lee took women who were settled and working. Gillis took women whose disappearance drew little inquiry. That distinction was a good deal clearer afterwards than it was at the time.
The rest of the answer is in that second sentence. The dragnet that swabbed more than a thousand men was assembled because women were disappearing whose absence was noticed at once. That was not how most of Gillis's victims were reported. Joyce Williams was killed on November 12, 1999 and was not found until the January.
Set the two lists beside each other and the difference is not subtle. The women Lee killed were a nurse, an LSU student, a graduate student a week past her MBA, an interior decorator who ran an antique shop, an Army veteran intending to join the Marines, and a doctoral student. The women Gillis killed were picked up on the street at night, and the record of several of them is thin enough that this archive cannot give their ages.
That is not a remark about the victims. It is a remark about what an investigation is built out of. A missing woman with a job, an address and a family produces a report within a day, a search, and a case file that stays open. A woman who is not reported produces nothing for a machine to work on, and a killer who takes only such women is not evading detection so much as operating outside the range of it. Gillis went on for ten years, and for the last of them he was doing it in a city that had a task force, a laboratory and a thousand swabs pointed at a man who was not him.
There is a figure from the time that gives the scale of it. When the task force confirmed in 2002 that it was hunting one man, police told reporters they were also going back through nearly three dozen unsolved murders of women in the area from the previous decade, to see which of them belonged to the same series. Three dozen open cases is not a backlog produced by one careful killer. It is what a city looks like when a particular kind of death has not been treated as a series at all.
The third man in this is barely known outside Louisiana. Jeffrey Lee Guillory was killing in the same area across part of the same period, which meant that at the height of the search there were three separate series running at once in one metropolitan area, and no reliable way for anyone working a single body to know which of the three, if any, it belonged to.
What ended it was physical evidence of the dullest possible kind. When Donna Bennett Johnston's body was left in February 2004, the car that brought it there left a tyre track. The state crime lab identified the brand, model and tread pattern, and established that the tyre had been manufactured over a period of about three years which had ended in 2003. Roughly ninety of them had been sold in the Baton Rouge area. Investigators worked through the purchasers, and on April 28, 2004 they visited the owners of vehicles fitted with that tyre and asked for DNA samples. One of the men they called on was Sean Gillis, who gave a sample voluntarily.
The sample matched evidence taken from the bodies of Katherine Hall, Johnnie Mae Williams and Donna Bennett Johnston. He was arrested at his house by a SWAT team early on April 29, 2004, and booked on three counts of first degree murder and three of ritualistic acts. It was the second time in under a year that a man in that city had been arrested as a serial killer. A senior officer of the East Baton Rouge Parish Sheriff’s Office told reporters that day that the goal now was to convict him and, he hoped, to execute him. In the event no jury sentenced him to death.
He confessed with very little pressure, first to the three he had been charged with, and then to five more that nobody had connected to him. The archive's own record of this is that he confessed at length and in detail, which is the part that separates his case from the confessions elsewhere in this collection that turned out to be worth nothing: what he gave investigators included killings they did not know about, and the physical evidence to go with them was in his house.
The prosecutions ran over three years and never covered everything he admitted. In August 2007 he pleaded guilty to the second degree murder of Joyce Williams and was given life. On July 31, 2008 a jury convicted him of the first degree murder of Donna Bennett Johnston; the defence argued that severe mental illness impaired his responsibility and the jury rejected it, then deadlocked at the penalty phase, so the judge imposed life without parole rather than death. On February 17, 2009 he pleaded guilty to the murder of Marilyn Nevils in Lafayette Parish and received another life sentence. Prosecutors could not bring charges for all eight, and put the uncharged killings before the court as evidence at a sentencing hearing instead.
He is held in maximum custody at the Louisiana State Penitentiary at Angola. He has written letters from there, including to a friend of Donna Johnston, in which he expresses regret in the same flat register he uses for describing how quickly she died. Eight deaths are attributed to him. He stands convicted of four.
The Louisiana press called him the Other Baton Rouge Killer, because the first name was taken. It is a poor description of what separated the two cases. Lee was found when a two-detective department went back to two old files and worked out which dates he had been out of prison. Gillis was found when a laboratory identified a tyre and somebody drove round to the ninety houses. Neither man was caught by the apparatus built to catch him, and in both cases the thing that worked was a small piece of physical evidence followed patiently to the end of a short list.