Every number in this archive comes from somewhere. Most come from a court. A few come from the man himself, and those are the ones worth looking at, because a confession is the only kind of evidence that grows when nobody checks it.
Henry Lee Lucas is the case everyone knows. He confessed to about six hundred murders, cleared 213 cases for the Texas Rangers, and was convicted of eleven. Task forces came from several states to sit with him and he obliged all of them. The number attached to his name for years was a measure of how many detectives had visited, not how many people had died.
Albert DeSalvo confessed in 1965 to the thirteen Boston Strangler murders and was never charged with a single one of them. He was in custody on other matters, the confession came through another inmate's lawyer, and he died in prison in 1973 with the murders formally unsolved. In 2013 DNA linked him to Mary Sullivan, the last of the thirteen. The other twelve remain unattributed, which means that half a century on, the only one of his thirteen confessions anyone has tested is the one that turned out to be true.
Roger Andermatt, a Swiss care assistant, confessed to twenty-seven killings and was convicted of twenty-two. Kiyotaka Katsuta, a fireman in Nagoya, confessed to twenty-two murders and hundreds of robberies over a decade and was convicted of eight; he was hanged in 2000 with the other fourteen never tested in court.
Sailson José das Graças told police in Rio that he had killed forty-one people, and in some accounts forty-three. His entry records that the number formally established against him in the Rio courts is considerably smaller. Ali Asghar Borujerdi was convicted of thirty-three killings in Iran in 1934 and gave a figure of twenty-five for a separate series in Baghdad that was never tried.
Stephen Nash's entry is the shortest version of the problem. The figure of eleven is his own.
Julian Kennedy is the clearest illustration of what these confessions were for. He was thirty, an escaped prisoner connected with outlaw motorcycle clubs, confined at Lake City, Florida in 1973, and he began describing killings in five states going back to 1958. Detectives came from several of those states to interview him. None of it produced a prosecution. He was convicted in Georgia of one murder and sentenced to life, and the figure of nine or more that sits in this roster is drawn entirely from what he said in a cell that year.
The pattern in all of these is the same, and it is not that the men were lying, though most of them were. It is that a confession is worth something to everybody in the room. The man gets attention, status, transfer out of a cell, sometimes a better prison. The detective gets a closed case. Nobody in that room has an interest in testing it.
What breaks the pattern is the same thing every time: physical evidence, arriving late. DeSalvo's one confirmed killing was confirmed by DNA forty-eight years afterwards. Nothing has confirmed the other twelve, and nothing has confirmed the fourteen Katsuta was not tried for, or the greater part of the 213 cases Lucas closed for the Rangers.
This archive carries convicted counts and claimed counts as separate figures for exactly this reason. Where a case rests on a man's own account and nothing else, the entry says so. It is not scepticism about whether he was a killer. Most of them were. It is that the number is a thing he said, and saying it was in his interest.