aka "the Stocking Strangler" was a "granny-killer"
from Atlanta, Georgia with a genius IQ who raped and killed 7 elderly
women, strangling them with their own stockings. It is unknown if the
motive was sex, lust for murder, or robbery. He was eventually picked up
on narcotics dealing.
The most interesting thing about this case is that
he had a rival who operated at about the same time in Atlanta, a copycat
killer named William
Hance, who sent
letters to the newspapers saying "No more grannies" and threatened to
kill a black women every 30 days unless the granny killer stopped or was
arrested. Hance also demanded $10,000 for his services. His notes were
signed "the Forces of Evil". Hance was caught by police tracking the
last activities of the 3rd black woman he killed, and he turned out to
be a Fort Benning soldier.
The case proved to be
a good example of cooperation between Army CID and Atlanta police, both
assisted by FBI agent Robert Ressler. Both Gary and Hance received
sentences of death by electrocution.
Defense will get review
of bite mold
28 december 2005
The Stocking Strangler's
teeth are now in play, you could say -- pending further examination,
Attorneys for the man
convicted of three of the seven infamous serial murders of 1977-78 have
won their bid to have a long-missing bite-mark mold from the last victim
examined by an expert to see if it casts doubt on Carlton Gary's guilt.
Gary was not convicted of the April 20, 1978, strangling of 61-year-old
Columbus school teacher Janet Cofer, from whose left breast the
impression of an apparent bite mark was made. Prosecutors used evidence
from that case during Gary's 1986 trial "solely to show similar mode,
method and motivation," U.S. District Court Judge Clay Land writes in
his order Thursday granting funds for further examination of the bite
operated on the theory that one person committed all seven murders, so
any evidence that might clear Gary in one case could undermine the claim
that he is the "Stocking Strangler" -- so named for the killer's
strangling women with their own hosiery.
Gary's attorneys first
sought the bite-mark mold in 2003, but no one knew where it was. It was
not used as evidence in the 1986 trial, and Gary's defense attorneys
weren't told about it then. Prosecutors later claimed it was irrelevant
because Gary had dental work after the murders.
Last month, Muscogee
County Coroner James Dunnavant discovered the mold stashed back in an
old file cabinet in an office storage room. For years Dunnavant's
predecessor, Coroner Don Kilgore, had kept the mold in his desk drawer,
occasionally showing it off. But no one knew what happened to it after
Kilgore died in 2000.
Now that the cast is
available, the only way to determine whether it's relevant is to have it
examined, Land writes: "The bite mark was withheld, and it may, or may
not, be exculpatory, material evidence. It appears the only way to find
out is through discovery."
Prosecutors are required
to turn over to defense attorneys in such prosecutions any evidence that
might be "exculpatory," or of possible use in showing a jury that the
defendant is not guilty of the crime.
Land is granting the
defense $1,500 to have Atlanta forensic dentist Dr. Thomas David examine
the mold to see if it's still useful, and if it is, to make an
"exemplar" or "negative impression" of the cast. Land's order also
authorizes the defense to take a deposition from Dunnavant.
Whether the defense
seeks anything beyond this initial examination depends on David's
Gary was convicted of
killing Florence Scheible, Martha Thurmond and Kathleen Woodruff. He was
sentenced to death.
The case is before Land
as part of Gary's federal "habeas corpus" appeal process, in which the
District Court judge considers legal points not previously excluded in
appeals. The decisions of the District Court are subject to appeal to
the 11th U.S. Circuit Court of Appeals and, ultimately, to the U.S.
Seeds of Doubt: Fame & Scandal
Seeds of Doubt
For six months in the late 1970s, the quaint Southern
town of Columbus, Georgia, was terrorized by a serial killer known as
the "Stocking Strangler," who raped and garroted seven elderly white
women in their homes. In an exclusive book excerpt, David Rose uncovers
shocking new evidence that suggests the African-American man sentenced
to death for the crimes may be the latest victim of Columbus's Old Boy
brand of justice.
by David Rose - VF.com - May 7, 2007
Excerpted from The Big Eddy Club: The Stocking Stranglings and Southern
Justice, published by The New Press.
On March 22, 2001, I visited Georgia's most notorious
convicted serial killer on death row and made a request so intimate that
I would never dream of asking it of my best friend.
Nevertheless, I had every reason to believe he would
comply. You see, I was trying to save his life.
On May 30, 2007, Judge Clay Land of the Federal
District Court in Columbus, Georgia refused to grant Carlton Gary a new
trial over the bite cast that the state failed to disclose for 20 years.
His petition now goes to the 11th Circuit Court of
Appeals in Atlanta.For Land, the expert testimony he heard in February,
which stated that the cast from the wound on Janet Cofer's body does not
match Gary's teeth, was not enough to necessitate a new trial even when
considered along with all the other exculpatory evidence that was not
shared with the defense from the original 1986 trial.
It was "possible," but not "probable," that the jury
would have reached a different verdict, and thus, Land asserted, it
doesn't clear the bar set in precedents by the Supreme Court.As Land's
opinion put it: "Even assuming that the bite mark cast is favorable to
Petitioner, the Court finds that, when considering it collectively with
the other undisclosed evidence, it does not reasonably put the whole
case in such a different light to undermine confidence in the verdict."
He did not address the rather obvious question: if Gary did not bite
Janet Cofer, then who on earth did? There will now be a hearing in the
11th Circuit regarding the bite and all the other material that Land
originally rejected in 2004, such as the semen (serology) type—though it
will likely not take place for several months.
After that, there will be only the Supreme Court. If
Gary doesn't win in the 11th Circuit, his case could finally come to a
close.I arrived that morning at the Georgia Diagnostic and
Classification Prison, 40 miles south of Atlanta, for a meeting with
Carlton Gary, known throughout the Deep South as the "Stocking
Strangler"—the African-American man who had raped and strangled seven
middle-aged and elderly white women in Columbus, Georgia, between
September 1977 and April 1978. From a large plastic folder I produced an
open envelope, a sheet of plain paper, and some plastic cling wrap. I
gave them to Gary and explained what I wanted him to do: produce a
sample of his semen in the privacy of his cell, and then give it to me.
I had already met with Gary on five previous
occasions, starting in 1998. Like all the prisoners, he was wearing
regulation white slacks and a polo shirt, with white, rubber-soled
shoes. A big man, well over six feet tall, he was freshly groomed, his
hair, beard, and mustache trimmed. His eyes, which he said had been
damaged by a lack of light in jail before his trial, were protected by
dark glasses, but I could see them well enough. They met mine and held
my gaze. He was 50 but looked younger, a little thickset but muscular,
in visibly good shape.
When I returned after the lunchtime visitation break
that afternoon, Gary and I were locked in a little square room just off
an echoing hallway. Gary slid the envelope, now sealed, across the
table, and I slipped it into the folder. Then I gave him a second
envelope. In front of me, he plucked about 30 hairs from his head,
complete with roots.
He placed them inside the second envelope and gave it
back. By matching the DNA in the roots and the semen, I would be able to
prove that the semen was his. Thanks to funding from Vanity Fair, I
planned to conduct vital tests—tests that might prove his innocence.
For the first time in my life, I was acting not
simply as a reporter but as a participant in a legal drama. If that
meant relinquishing a measure of journalistic independence, I considered
it justified by what was at stake.
Columbus, population 185,000, is Georgia's second
city, 110 miles south of the state's capital, Atlanta. Founded in 1828,
it has long been prosperous and influential. The chemist John Stith
Pemberton first mixed the recipe for the soda drink that would become
Coca-Cola in his Columbus drugstore in the 1870s.
"Mother of the Blues" Ma Rainey was born and raised
in Columbus, as was the writer Carson McCullers. Most of her novels,
including her masterpiece, The Heart is a Lonely Hunter, were set there.
Running across the city is a racial fissure, its line
marked by the east-west thoroughfare known as Macon Road. With
exceptions unusual enough to be noticeable, white people—about 65
percent of the total—live to the north, and black to the south. No
longer legally segregated, they mingle at work and use the same stores
and restaurants, but in general they do not mix in their social lives,
or at home.
Most of the remains of antebellum Columbus are to be
found in Wynnton. Once known as the "millionaires' colony," its placid
exclusivity dates back to the period before the Civil War, when its
mansions were at the heart of a social whirl that is said to have
rivaled more famous centers of the old, slaveholding South, such as
Charleston and New Orleans. There were picnics, barbecues, marching
bands and orchestras, full-dress hunts and balls.
As late as the 1970s, Wynnton remained a kind of
Arcadia, writes the Columbus journalist and author William Winn: "Nearly
every house, however modest, has a lawn, and every spring Wynnton is
ablaze with pink and white azaleas, the neighbourhood's particular
glory." Later, the district was to become synonymous not with colorful
flowers but with rape and murder.
The horror began to surface on Monday, September 12,
1977, when an elderly woman named Gertrude Miller was raped and
strangled at her home, in Wynnton. She survived, but at 10 a.m. on
September 16 the police received a call that Mary "Ferne" Jackson, 59,
Columbus's director of public health education for 26 years, had not
shown up for work. When Jesse Thornton, the first police officer to
respond, arrived at Mrs. Jackson's house, he saw no sign that anyone had
forced an entry.
Many years later, Thornton would tell a murder-trial
jury what happened next. The first space he came to was the hallway, and
right away, "I could see something that wasn't right. There was stuff
laying on the floor, papers, articles, just scattered all over the
floor. There was a pillow on the floor, there was a suitcase that was
opened, the drawers had been opened on the dresser, and stuff was pulled
out and hanging out of it." He continued down the passage towards
Jackson's bedroom, his hand poised over his weapon. "Once I got to the
bedroom, I looked inside," Thornton said. "That's when I saw the body on
The Columbus medical examiner, Dr. Joe Webber,
conducted a postmortem. The killer, he wrote in his subsequent report,
had tied a nylon stocking and a dressing-gown cord together to make a
single ligature, which was wrapped around Jackson's neck three times,
leaving three "very deep crevasses."
The white of her left eye was "almost obliterated" by
bleeding—the result, he believed, of a massive blow to her head. The
white of her right eye was a mass of tiny, pinpoint hemorrhages, a
common sign of strangulation. Her sternum, or breastbone, had been
fractured, an act which would have required the application of enormous
force. Finally, her vagina was bloodied, torn, and bruised. Although
Webber could not find spermatozoa, he thought that Jackson had been
raped. Later, traces of seminal fluid would be found on the sheet.
There was no obvious motive for Jackson's murder.
Despite ransacking her house, the killer had left her jewelry and other
Ferne Jackson was murdered barely a month after the
capture of David Berkowitz, the sexually driven "Son of Sam," who killed
or seriously injured a dozen women in New York. The F.B.I. had recently
coined the term "serial killer." It took just eight days after Jackson's
murder for it to become apparent that one was at large in Columbus.
Jean Dimenstein, the owner of a small department
store, was 71. To enter her home on the night of September 24, the
strangler removed the hinges of the door leading from the garage to the
kitchen. Her murder was essentially similar in its method to Jackson's.
Florence Scheible, 89, who was almost blind and used a walker to get
around, was raped and murdered on October 21, and Martha Thurmond, a
retired teacher of 69, on October 25.
All the victims lived within a few blocks of each
other in Wynnton, as did Kathleen Woodruff, 74, the widow of former
University of Georgia football coach George C. Woodruff. A cultured
woman who had been a close friend of Carson McCullers, Kathleen was
strangled on December 28 with a varsity-football scarf.
The murders, and the evident inability of the cops to
do anything to prevent them, plunged the city into terror. Sales of
locks and firearms soared.
Augmenting the efforts of the local police was a
"stocking stranglings task force" assigned by the governor, with dozens
of agents from the Georgia Bureau of Investigation and extra patrols by
state troopers and soldiers from Fort Benning. None seemed effective. On
February 12, 1978, still known in Columbus as the "night of the
terrors," the strangler attacked Ruth Schwob, 74, the widow of a textile
tycoon. Trained in judo, she fought back and survived by pressing a
panic-alarm button, but even as the police rushed to her residence the
strangler made his way through an alley to the home of Mildred Borom,
78, two blocks away, and killed her with the cord from a blind.
Perhaps the night of the terrors scared even the
strangler. For his last murder, he moved out of Wynnton to Steam Mill
Road, a mile and a half away. There, on April 20, eight months after his
first attack, he killed Janet Cofer, 61, an elementary-school teacher.
And then, without apparent explanation, the stocking stranglings
stopped. Months, then years passed, with no new murders but no solid
leads either. Six years after Mrs. Cofer's murder, no one had been
Meanwhile, having examined the crime scenes and the
bodies, Columbus coroner Donald Kilgore made a public announcement. All
the victims had been white, but, after examining the pubic hair found at
the crime scenes, Kilgore said he was sure the killer was black. In the
Deep South, this was not an incidental matter.
I had first found my way to Columbus in 1996. With
the Summer Olympics in Atlanta fast approaching, my editors at the
London Observer had sent me to assess the state's handling of
death-penalty cases. I soon discovered that Columbus had sent more
prisoners to death row than anywhere else in Georgia, despite a
relatively low crime rate. It also had a long history of black men being
killed in other, uglier ways for alleged crimes against whites. From the
start of Reconstruction, in 1865, Columbus had been beset by lynchings—two
of them involving a family that would come to play a key role in Carlton
In the small hours of June 9, 1900, Simon Adams, a
black farm laborer aged 19, was caught trying to burgle the home of his
employer, E. H. Almond, who lived a little to the north of Columbus, in
Muscogee County. Adams made the mistake of trying to enter Almond's home
by the window of the bedroom used by his daughters, who were 10 and 17.
When one of them raised the alarm, he tried to hide in their wardrobe.
The men of the house found him and tied him up at gunpoint.
A few hours later, still bound and chained, Adams was
entrusted to the care of a bailiff, who was to take him to the county
jail in Columbus. The bailiff, from a powerful landowning family, was a
big, jug-eared man named A. Brewster Land.
There was a direct road from Almond's house to
downtown Columbus, but Land did not use it. Instead, he took a detour
through the woods, close to the Chattahoochee River. A vigilante posse
was waiting. They seized Adams, dragged him to a 30-foot, rocky overhang
known as Lovers' Leap, and pushed him into the water. Adams survived the
plunge, but the men on the shore began firing their rifles as the
current carried him downstream. The last time he came up for air a round
from a Winchester entered his skull behind his ear.
Land was asked later if he knew any of the men he met
in the woods. "I didn't recognize anybody," he said. "I was more
concerned with the guns than the men behind them." Like most lynchings,
Simon Adams's murder was officially put down by the coroner as the work
of "persons unknown."
In the late 19th and early 20th centuries, white
racists in Georgia public life defended lynching as necessary and
justified, especially in cases where black men were alleged to have had
sex with white women. In 1897, the Atlanta Journal columnist Rebecca
Latimer Felton wrote, "If it takes lynching to protect woman's dearest
possession from drunken, ravening, human beasts, then I say lynch a
thousand a week if it becomes necessary."
At a Georgia State Bar Association meeting in 1899,
J. F. DeLacy from Eastman declared, "The chief cause of the discontent
and disorder that occasionally breaks out is the uncertainty of
punishment for the commission of crime." If citizens could only be sure
that people they know to be guilty would be found guilty, DeLacy argued,
they wouldn't be forced to take matters into their own hands.
In June 1912, Brewster Land's nephew, 12-year-old
Cedron "Cleo" Land, was killed by a shotgun blast to the eye. The
supposed assailant was a black boy who was probably 14, T. Z. "Teasy"
McElhaney. Although they were of different races, Cleo and Teasy were
known to be frequent playmates, and Teasy claimed that they had been
playing with the gun when it accidentally went off. He was charged with
murder, but his all-white jury found him guilty only of manslaughter.
That wasn't enough for Cleo's father, Will Land, and his uncles,
Brewster and Ed. With the help of 15 other men, they seized Teasy from
the courthouse, bundled him onto a streetcar, and, when they reached
Wynnton, turned him out and, in front of a large crowd, opened fire. The
coroner was unable to determine how many bullets killed Teasy, but the
Columbus Ledger put the number at between 25 and 50.
Until then, no white person had ever stood trial for
killing an African-American in Columbus, but the brazenness of this
attack demanded a response. The three Land brothers were tried for
Teasy's murder. The all-white jury reached its decision immediately,
without discussion: "Not guilty."
Brewster Land went back to his farm. His son, John,
would become the first judge to handle the stocking-strangler murder
By May 1984, John Land had been a judge for 20 years
and was widely regarded as one of the most powerful men in Georgia—the
controller of the "Fish House Gang," a secretive, statewide network of
businessmen, lawyers, and rising politicians that held regular catfish
fries in Columbus. In the late 1940s and early 1950s, he had served in
the state assembly as a vehement segregationist.
Judge John Land, who handled Gary's case pre-trial.
From the Columbus Ledger-Enquirer Group.Carlton Gary appeared in Land's
court for the first time on May 9, 1984, indicted as the stocking
strangler. The police had been on his trail since the middle of March,
when Detective Michael Sellers had made a link between Gary and a gun
stolen in Wynnton at the time of the murders. It was possible, the
police thought, that this weapon had been taken by the strangler. Their
manhunt ended on May 3, when a swat team stormed a motel room in Albany,
Georgia, surprising Gary in bed. Later that day, Sellers and two
colleagues drove him back to Columbus.
In some ways, Gary was a perfect suspect. His
childhood, spent in Columbus until he was 13, was unsettled, and he had
been in trouble with the law since his teens. The most damning incident
from his past was his connection to the 1970 rape and strangulation of
Nellie Farmer, a woman of 84 who lived alone in a long-stay hotel in
Albany, New York. Gary told the Albany cops that he had merely acted as
a lookout for the killer, a habitual criminal named John Lee Mitchell,
and he testified against Mitchell at his trial. The jury, however, found
Mitchell not guilty. After pleading guilty to robbery charges, Gary was
sentenced to 10 years and released on parole after five.
In August 1977, less than a month before the first
stocking strangling, Gary escaped from another jail, in upstate New
York, where he had been doing time for handling stolen property, and
made his way back to Columbus. Just two months before his 1984 arrest,
he had again escaped from prison—this time in South Carolina, where he'd
been jailed in connection with a bungled 1979 restaurant robbery.
Other aspects of Gary's character were harder to
square with the profile of a serial killer, however. A talented
musician, he had played bass and keyboards in clubs across the
northeast, and at the time of the stranglings he had been modeling
clothes for a downtown Columbus fashion store, the Movin' Man, appearing
in advertisements on local television up to five times a night. He had
continued to live in Columbus for almost a year after the last murder.
Above all, he was a magnet for women. Several of his girlfriends would
later describe him to me as an attentive—and entirely normal—lover.
"He was always a very sharp dresser, real neat," says
Floyd Washington, the proprietor of the F&W Control Tower, a big
Columbus nightclub that Gary frequented. "He used to wear suits, sports
coats, top-line dress shirts, and sometimes a little bow tie. Let me
tell you, he never had a problem with no ladies. He was a very good
dancer. Over the years I knew him, he had many girlfriends, very
attractive ones, with good jobs. The idea that he would leave this club
with a beautiful girl and then leave her to go and rape an old woman—it
just don't add up."
Less than a week after Gary's arrest, Judge Land
assigned him a local attorney who had been raised in and around Wynnton:
William Kirby. "From an early stage, I formed a view about Carlton
Gary," Kirby told me in 2001. "I thought there was a pretty good chance
he was telling the truth—that he really was innocent."
But Gary wanted to choose his own lawyer, and he
settled on August F. "Bud" Siemon III. Siemon, a tall, well-built man
with startling blue eyes, was an Atlanta defense specialist who had
fought a dozen capital trials. His client had avoided a death sentence
each time. Gary decided to retain Kirby as co-counsel, along with his
experienced investigator, Don Snow.
August "Bud" Siemon, Gary's trial attorney. Deprived
of funds, he almost went bankrupt as a result of the case. From the
Columbus Ledger-Enquirer Group.
On August 28, 1984, after a meeting with Gary in
jail, Siemon appeared in Judge Land's court, saying he had been asked to
take over the defense. Land was happy to agree. But the conditions he
set were severe. First, Siemon would have to work for nothing. Georgia
law said the courts must give men on trial for their lives a lawyer,
Land said, but there was no stipulation that public funds had to be
provided for a defendant's counsel of choice. Second, he refused Gary's
request to keep Kirby and Snow as co-counsel and investigator.
Land recused himself the following May, when it
emerged that he might have to be called as a witness to Gary's treatment
by the police. But his replacement, Judge Kenneth Followill, took Land's
original ruling and applied it even more severely, deciding that it
meant that the state could not give the defense a single cent for any
purpose in any conceivable circumstance. Above all, there would be no
money to pay for experts to challenge what looked like the state's most
vulnerable point—its scientific testimony. "We couldn't even make
long-distance phone calls," says Siemon, whose efforts to cover the
case's costs would eventually drive him to the brink of bankruptcy.
In January 2001, John Land, by then long retired,
granted me an interview at his bungalow in north Columbus. There was
much in his life that he now regretted, he told me, including his
support for racial segregation. Land said he also had misgivings about
the trial of Carlton Gary. And while he stood by his decision not to pay
Siemon, he thought Followill had been wrong to deny Gary money for
experts. "If they had asked for expense money, I would have given it to
them, as being necessary to the defense, in order to have a fair trial,"
Land told me.
Gary's trial did not start until August 1986, and
when it did the case presented by District Attorney Bill Smith, a former
F.B.I. agent, looked overwhelming.
Police witnesses testified that they had found
fingerprints that matched Gary's from four of the crime scenes. Gertrude
Miller, the elderly woman who survived the strangler's first attack,
identified Gary in court, saying that he was the man who had raped her.
Detective Sellers said that, on the night of Gary's arrest, he and two
colleagues interrogated Gary for eight hours. They had neither taped nor
taken notes of these conversations, but Sellers testified that Gary had
admitted being at the crime scenes and had blamed someone else for the
murders—a friend named Malvin Crittenden. That, the prosecution said,
fitted Gary's standard pattern: his false accusation against Crittenden
called to mind his behavior after his arrest in Albany, New York, in
1970, when he accused John Lee Mitchell of murdering Nellie Farmer.
District Attorney Smith told the jury that Gary had surely killed her
Smith's case did have weaknesses. Gary's pubic hairs
had been compared to those found on the victims, and they did not match.
Similarly, it appeared that there were certain biochemical differences
between his bodily fluids and those left by the killer. But these
difficulties did not exclude him, Smith insisted, and they did not
trouble the jurors (of whom nine were white and three African-American).
They began to deliberate at 5 p.m. on August 26,
1986. Less than an hour later, they had their verdict: guilty on all
counts. The next day, after a short sentencing hearing, they condemned
Gary to death.
In the autumn of 1991, having lost his direct appeal
to the Georgia Supreme Court, Carlton Gary submitted a petition of
habeas corpus—a further appeal of a type that occurs in almost every
capital case, when a defendant seeks a new trial on the grounds that his
constitutional rights to due process, equal protection under the law,
and effective legal representation were violated in the original
hearing. The case was assigned to Judge Daniel Coursey, a lay preacher
in the United Methodist Church.
Bud Siemon had spent thousands of dollars handling
Gary's trial and first appeal; now he was off the case. Representing
Gary for his state habeas appeal was the Georgia Death Penalty Resource
Center, a not-for-profit organization. In the autumn of 1992 the
center's director, Michael Mears, asked his newest lawyer, Jeff Ertel,
to take on Gary's case. A rugged, down-to-earth figure with sandy hair
and an open, easy manner, Ertel had spent 11 years as a deputy sheriff
in upstate New York and was fresh out of law school. "I went back to
Buffalo to see my mom and she asked me what kind of work I was doing,
and I said, 'I'm fighting an appeal for a black guy on death row who was
convicted for raping and murdering a bunch of white women who were about
your age.' After that she didn't ask me about my work too much. But when
I met Carlton I began to think about it differently. I liked him
immediately, and when I got to know him better, I liked him a lot."
Throughout the United States, the 1963 Supreme Court
case of Brady v. Maryland means that prosecutors are supposed to turn
over evidence that might exculpate a defendant or weaken the state's
case. But in Georgia, the prosecution has wide powers to decide what is
in fact exculpatory before the trial, and if it decides that documents
or other testimony would not be materially helpful, it can keep them
secret. It is only later, during post-conviction appeals, that it can be
compelled to hand over almost everything—by which time, owing to the
strict rules of procedure, it may be too late to make any difference.
Working through the paper mountain, Ertel came across
evidence in the files of former district attorney Bill Smith (by this
time he had become a judge) that Ertel believed was tailor-made for a
habeas appeal. Some of it concerned the testimony of Gertrude Miller—the
rape survivor whom Smith had described as his "star witness." Smith had
laid heavy emphasis on Miller in his closing speech to the jury,
claiming that there was no way she could have gotten her identification
wrong: "Don't you know that every night Mrs. Miller has laid her head on
her pillow and closed her eyes to go to sleep, that she has seen the
face of this man? It's burned into her memory, it's forged into her
memory, she can't put it out of her mind if she wants to." Despite being
shown thousands of police photos, Smith said, she had never accused
anyone before she recognized Gary.
In fact, Ertel discovered evidence indicating that
Gary was by no means the first man she had identified. In Smith's files
were no fewer than nine separate documents that cast grave doubt on
Miller's testimony in court—prior identifications that she had made of
other suspects, and of police sketches of men who looked nothing like
Gary. Smith testified at the state habeas hearings that he had not
turned the statements over to Gary's defense because he had not
considered them to be exculpatory.
Ertel also discovered documents that cast doubt on
other parts of the prosecution's case. For example, the police had
always claimed that their best fingerprint match with Gary came from a
window screen at the home of Kathleen Woodruff, but the officers who
dealt with the crime scene had recorded there were "no legible latents
[print marks made visible when treated with chemical powder] developed."
Other papers raised questions about Gary's supposed
confession. He had always claimed that he made no incriminating
admissions, and that his interrogation was taped. Detective Sellers
still denies this, saying that he made the only record sitting at his
kitchen table at 4:30 a.m. on the night of the interrogation, when he
wrote Gary's confession out from memory. But in the Columbus files,
Ertel found what looked like early versions of Gary's statement—with
tape-recorder meter marks written in the margins.
These versions suggested that Gary had not admitted
being at the scenes of the crimes or having stood by while Malvin
Crittenden committed the murders. Sellers testified that although the
meter marks did refer to a tape, this had not been made by the police
when they questioned Gary. Instead, he said, the marks related to a
recording he had made of a meeting in his office shortly before the
trial, in which he had gone over the case with the detectives. He had,
he added, destroyed these tapes soon afterwards.
Ertel knew from his study of the trial that, on the autopsy report and
photographs of Janet Cofer's body, Bud Siemon had noticed a deep bite
mark left by the killer on Cofer's left breast. Siemon could not afford
to pay an expert to compare these images with Carlton Gary's teeth, so
he asked Judge Followill for funds. Followill, as usual, refused. Ertel
decided that it might be worth trying to open this issue again.
"I went through a list of accredited experts looking for a forensic
dentist," Ertel said. "I saw there was a guy in Atlanta called Thomas
David. I basically picked him at random and called his office. His
assistant took a message, and he called me right back. I told him what
it was about, and he said he wanted to talk to me right away, but not on
the phone. So I went over there to see him."
Dr. David had served for more than a decade as a consultant odontologist
with Georgia's State Crime Lab. More used to working with prosecutors
than defense lawyers, he was also accredited with the Georgia Bureau of
Investigation (G.B.I.), the state medical examiners, and the F.B.I. When
Ertel arrived in his consulting room, he came straight to the point. "I
know about this case," David said. "I'm the guy they showed the bite
Ertel stared at him in amazement. "Bite cast? What bite cast?"
David told him that two months after Gary's arrest, on July 6, 1984,
Bill Smith and his assistant, Doug Pullen, had been to see him. In
addition to the photos from Cofer's body, they had, he said, shown him a
solid dental mold, an impression made from the wound on her breast when
it was still freshly inflicted. The killer had left behind deep traces
of both his upper and lower teeth. "There appeared to be a 'crowding' of
the lower teeth that made them particularly distinctive," David told
David said he told Bill Smith that he should obtain a dental impression
from his suspect: "It was my belief that this comparison could either
positively prove or disprove that Mr. Gary was the attacker." Nine years
later, Smith's response continued to trouble him. David told Ertel that
Smith had then asked him whether, even if Gary's teeth did not match the
cast, "I would testify … that I could not necessarily eliminate that
individual as a suspect. I said I could not." In the state habeas
hearings, Smith testified that he had not contacted Dr. David again
because he insisted the bite only showed an impression of the upper
David's answer seemed not to have been what Smith and Pullen wanted to
hear. His affidavit went on: "As Mr. Gary's trial progressed, it became
apparent that the District Attorney was not going to use the bite-mark
evidence. I had strong misgivings as to why he chose not to use what I
believed to be highly relevant evidence."
It took only a short visit with his client Carlton Gary for Ertel to
confirm that, far from exhibiting overcrowding, Gary's lower and upper
teeth were well spaced and even. Gary told him that the only dental work
he had had done since the last strangling, in 1978, had been performed
while he was in prison in South Carolina. However, this had been only to
his upper teeth. The dentist who had treated him, Dr. Paul Hahn,
confirmed Gary's assertion.
The cast from the bite on Cofer's breast had been made by a dentist in
Columbus, Dr. Carlos "Sonny" Galbreath. Months before the trial, David
had spoken to him on the phone, and Galbreath told him that, instead of
keeping the mold in the Columbus Police Department's guarded-exhibits
store, Smith and Pullen had given it back to him. Sometime in June 1993,
Ertel phoned Galbreath and asked whether he still had the cast. "I think
I have it here at home," he told him. A few days later Galbreath phoned
Ertel. He sounded apologetic, and said he had been mistaken. The cast
had been lost or destroyed.
The bite cast was particularly important because Gary had no recourse to
DNA evidence. Soon after the first of (to date) 180 inmates was freed
from death row by DNA testing, in 1993, Ertel tried to obtain the semen
swabs from the stranglings. State officials told him they had all been
destroyed. By 1993, Doug Pullen had become Columbus's district attorney.
"It is considered, if I follow correctly, by the Crime Lab to be a bio-hazard
at the time it was taken," Pullen testified at Gary's state habeas
hearings, in December 1994. "There was nothing that could have been done
with it, and it was not maintained."
By 2006, Ertel had handled more capital cases than he could easily count,
and he told me that Gary's state habeas appeal was the strongest he had
seen: "I think with any case other than Carlton's, we would have got a
new trial at state habeas." Even as he filed the brief, he warned Gary
to expect the worst.
On November 13, 1995, Judge Coursey issued a 10-page decision denying
every count in the appeal. He gave no reasons for rejecting the fresh
Gary had one shot left: a second habeas petition, this time in federal
court. In the fall of 1998, I met with the attorneys Gary had been
assigned by the court in Atlanta, Jack Martin and Michael McIntyre.
Martin, a genial man with a rippling laugh, had successfully defended
Richard Jewell, the security guard wrongly accused of planting a bomb at
the Atlanta Olympics. I made him a promise. It seemed to me that the
case against Gary had one unexplained and potentially significant
weakness. I told him I would find an expert in England prepared to
examine the issue pro bono.
Even though DNA-testing techniques had not been invented at the time of
the stranglings, investigators did possess an older method for testing
semen that could be very effective: secretor typing. Most people, about
four-fifths of the population, are "secretors," meaning that in their
saliva, semen, and other fluids they secrete chemical markers that give
away their blood group. A "group-O secretor" would be someone from the
common O blood group whose semen contained a relatively large amount of
the relevant marker.
The tests carried out on semen collected from the murder scenes of Ferne
Jackson, Florence Scheible, and Martha Thurmond indicated that the
stocking strangler was a "non-secretor," whose body fluids contained
only tiny traces of the group-O marker. As for Carlton Gary, the police
took a saliva sample from him after his arrest. It revealed that,
together with some 40 percent of the population, he was an O secretor.
The state's main expert was a forensic serologist named John C. Wegel,
who worked at the G.B.I. crime lab in Atlanta. At the trial, he insisted
that, despite the apparent discrepancy, Gary could not be eliminated.
Perhaps, Wegel argued, the killer had not really been a non-secretor but
what he called a "weak secretor," because the tests had revealed tiny
traces of the blood-group-O marker. In that case, Wegel went on, it was
possible that Gary had produced the semen found at the murders after all.
Giving testimony in 2000, he said that in his view, it was appropriate
to use both absorption-elution and absorption-inhibition in order to
determine secretor status.
Even then, he had to explain what looked like a second big discrepancy—the
fact that Gary, according to tests on his saliva, was in fact a very
strong secretor, producing much higher concentrations of the chemical
marker in his bodily fluids than the killer had. Wegel testified that it
was possible that Gary secreted at higher levels into his saliva than
into his semen—which had never been tested.
It was also possible, he said, that Gary's levels had changed between
the murders, in 1977 and '78, and his arrest, in 1984: "Periodically, if
you would test someone over a given amount of time, you will see that
the titre [concentration] can in fact fluctuate." Deprived of funding,
Gary's trial attorney was in no position to challenge Wegel's testimony.
Near my home in England, I found Dr. David Roberts, a consultant
hematologist at the John Radcliffe Hospital, in Oxford. He agreed to
read the documents and trial transcripts relating to serology.
A few days after receiving the papers, Roberts called to say that he had
found the record not only baffling but apparently incomplete. First,
there was no evidence to support the theory that Gary secreted much less
in his semen than in his saliva—not only because Gary's semen had never
been tested but because semen invariably contained higher concentrations
of the marker chemicals than saliva.
Second, there was a gap in the record: Wegel's original laboratory
notes. Wegel had told the court that he had performed two separate tests
on the crime-scene semen—"absorption-inhibition" and "absorption-elution."
According to Roberts, the two tests were very different, and the correct
one to use when trying to identify a rapist was absorption-inhibition.
Because it was relatively insensitive, it would not detect the very low
levels of marker chemicals left behind by a non-secretor. Absorption-elution,
on the other hand, was about 100 times more sensitive. Its danger was
that its very sensitivity might give rise to "false positives"—findings
that someone was a "weak" secretor, when in reality they were not. (Wegel
rejected Dr. Roberts's analysis of his working procedures.)
Roberts wondered whether the reason Wegel had used both tests was that
he had drawn a blank with absorption-inhibition. If so, he should have
concluded that the strangler was a non-secretor—and so excluded Carlton
Gary as the source of the semen. Perhaps, Roberts speculated, Wegel had
then gone on to perform absorption-elution—and only then detected the
very low chemical levels that made him assert that the killer might be a
weak secretor. There was only one way to find out, said Roberts: he
needed to examine Wegel's lab notes from 1977 to '78. But the state had
never turned them over, and they appeared to be missing.
Martin and McIntyre filed a request under the Georgia Open Records Act,
asking to look at Wegel's laboratory worksheets. At the end of February
2000, they were told that the worksheets from the Florence Scheible and
Martha Thurmond killings could not be found anywhere, but that those
from the Ferne Jackson case were available.
The worksheets quickly confirmed Dr. Roberts's hunches. On April 22, he
swore an affidavit. The worksheets, it said, showed that Wegel had
indeed found no trace of chemical markers when he tried absorption-inhibition.
It was only when he used the hypersensitive absorption-elution method
that very low levels showed up. "Based on this new information," Roberts
stated, "it is my firm opinion that Mr. Gary is excluded as a possible
donor of the stains believed to be semen in the Jackson case." Repeating
his insistence that absorption-elution was not hypersensitive, Wegel
stood by the evidence he gave at the trial, stating that the worksheets
did not suggest that Gary was excluded at all.
Martin and McIntyre used Roberts's declaration to apply to the Federal
District Court for an evidentiary hearing. The hearing was set for
November 21, 2000, in the court of Judge Hugh Lawson, in Macon, a city
of similar size to Columbus, 60 miles southeast of Atlanta. The defense
attorneys also asked the judge to allow them to conduct a controlled
test of Gary's semen, to ascertain whether the state's assertion—that he
might have secreted at lower levels in his semen than in his saliva—was
On November 17, Jack Martin was at his desk in Atlanta when the phone
rang: it was Susan Boleyn, Georgia's assistant attorney general, the
lawyer responsible for resisting appeals on behalf of the state.
According to Martin, Boleyn sounded a little embarrassed as she told him
that Wegel's worksheets for his tests on the semen from the bodies of
Florence Scheible and Martha Thurmond had also just been discovered at
the G.B.I. laboratory.
David Roberts was unable to travel to Georgia, so in his place Martin
called another expert serologist, Rodger Morrison. He was the director
of Alabama's state crime lab, in Huntsville, and most of his 500-plus
courtroom appearances had been as a witness for the prosecution, not the
defense. Morrison said he knew of no case where anyone had been found to
be secreting less in their semen than in their saliva. But the way to be
sure, Morrison said, was to test Gary's semen now, because "what
literature there is available indicates that there does not appear to be
any significant change in secretor status over a period of time." Martin
asked Morrison what he would conclude if a new test on Gary's semen were
to confirm that he was a strong secretor. "It would be my opinion that
he would be excluded as a source of that sample," Morrison said.
By the beginning of 2001, it was starting to become apparent that my
relationship with Gary's lawyers ought to be put on a more formal basis,
since I was coming up with evidence that might prove decisive in Gary's
federal habeas appeal. Accordingly, Jack Martin appointed me as a
paralegal investigator on behalf of his law firm, Martin Brothers P.C.
My new role had one great advantage. It meant that when I visited Gary
on death row, I could carry documents and writing materials, and take
Almost five months after the hearing in Macon, there was still no word
from Judge Lawson as to whether he would grant the semen test that might
definitively prove Gary's innocence. Meanwhile, on behalf of the state,
Susan Boleyn had submitted a written brief opposing it: "Even if the
semen evidence in the Jackson, Scheible and Thurmond cases were … to
exclude petitioner as the rapist of these women, the semen evidence does
not in any way undermine the testimony of the surviving victim who
testified that petitioner Gary raped her and attempted to strangle her
with her stockings."
Over lunch one day in Atlanta, Martin and I agreed that Boleyn's
argument might seem less acceptable to Judge Lawson if we could show him
that the defense experts' claims about Gary were true. Afterward, I
decided to try to get the job done myself.
So it was that I made my death-row visit in March 2001 to collect a
semen sample from Gary. Afterward, I prepared the sample according to
instructions from Brian Wraxall, the head of the Serological Research
Institute, in Richmond, California, and sent it to him, together with
the strand of hair from Gary's head, and copies of all the documents
about the serology of the stranglings case.
Wraxall first DNA-matched the hair roots to the semen. Then he tested
the semen. "I determined that the donor is an O secretor and that his H
antigen was detected to at least a dilution of 1 in 3000," he told me
via e-mail on April 24, 2001. "In my opinion he is not a low level
secretor." Later, he elaborated, saying it was inconceivable that
someone secreting at such high concentrations could have been classed as
a "weak secretor" earlier in his life. The concentration of the blood-group
marker in Gary's semen was more than 3,000 times higher than the
Martin filed a new motion, enclosing affidavits from Wraxall and me and
asking the judge to take account of our tests. Predictably, Boleyn
opposed it: "It appears that the testing … was done, not at the
instigation of his attorneys, but at the instigation of a non-party
journalist," she wrote.
Another two months went by. Finally, on June 29, 2001, Judge Lawson
issued two orders. The first denied the defense's request for a test of
Gary's semen; the second denied to consider the test we had arranged
ourselves. "The court refuses to receive information regarding Mr.
Rose's and Mr. Wraxall's testing," he concluded. "The fact that
petitioner may be a normal secretor today does not prove he is innocent
of the rapes that occurred in 1977.
Apparently there is a disagreement among experts as to whether Mr. Gary
could be excluded as the perpetrator of the rape in the Thurmond case.
However, the existence of this disagreement does not prove he is
innocent." Moreover, he added, "Mr. Wraxall does not state that
petitioner could be excluded as the source of the semen in the Jackson,
Scheible, Borom and Woodruff cases."
That marked the end of Lawson's involvement in the stranglings case. He
was about to hand over the rest of the appeal to Columbus's new federal
judge, a man whose surname was very familiar. Once again, the fate of
Carlton Gary would be determined by a member of the Land clan.
A short, anxious-looking man with dark, bushy eyebrows and deep,
triangular furrows that ran to his mouth from the sides of his nose,
Columbus's new federal judge, Clay Land, was part of a new generation of
Southern conservatives, a group for whom the white-supremacist rhetoric
of the 1950s seemed as distant as his family's rural origins. Unlike his
predecessors, Land, appointed to the bench by George W. Bush in 2001,
was a Republican. Before he contemplated becoming a judge, he had served
three full terms for the G.O.P. as a senator in the Georgia State
Assembly. Judge John Land—the son of the lynch-mob leader Brewster Land
and the man who had made the rulings that had denied Gary funding—is
Part of the cast made from the bite mark left by the strangler in the
breast of his final victim, Janet Cofer, showing the wide gap that
existed between his two upper front teeth. One of the teeth is also out
of normal alignment.
Carlton Gary has never had such a gap, say those who knew him well at
the time of the murders.A few months before Land's investiture, I had
tracked down a man I had long been wanting to meet: Sonny Galbreath, the
dentist who made the mold of the bite wound left by the strangler in the
breast of his final victim, Janet Cofer.
Whether it had been lost or destroyed, the cast seemed irretrievable.
Yet there was one unanswered question that made me want to interview
Galbreath. Bill Smith and Doug Pullen had claimed in the state habeas
hearings that the cast was of the killer's upper teeth only, and that
because Gary had had dental work on his upper jaw while he was in prison
in South Carolina after the last of the stranglings, a comparison would
have no relevance. But Dr. David had said in his affidavit that the cast
included the lower teeth as well, and these, he said, had displayed a
distinct pattern of overcrowding.
I met Galbreath for coffee at the Columbus Hilton and soon discovered
why he had been forced to close down his practice in 1993. A devoted
weekend hunter, he had accidentally blown the ends off several of his
fingers with a rifle. He seemed friendly and easygoing, and was happy to
recall how he had made the cast by pouring a liquid alginate into
Later, he said, he had also made a replica of the bite itself, a
negative copy of the tooth cast known as a "cast stone" or "exemplar."
He also remembered being contacted by Jeff Ertel. At the time, he said,
Doug Pullen was still Columbus's district attorney."Doug and I go way
back," Galbreath said. "He and I go bird hunting.… So when the attorney
[Ertel] called me, I called Doug. I asked him if I could show the tooth
cast to the attorney." He slapped his thigh, unable to restrain a
chuckle. "Doug said, 'On no account can he look at [that] model. You
tell him his reputation precedes him.… You don't show him anything.'"
This was not what Pullen had said in his testimony at the state habeas
hearing in 1994. Although he confirmed that he and Galbreath were
friends, he said they had not discussed the cast at all.
Then Galbreath made an unexpected disclosure. "I do know this," the
dentist said. "The model is still in existence." Galbreath said that
when he shut his practice, he had given the cast to the Columbus coroner,
Donald Kilgore. Kilgore died in 2000, and Galbreath thought it must have
passed into the care of his successor, James Dunnavant. He promised he
would speak to Dunnavant the following week.
Over the next two months I called Galbreath at least nine times. On each
occasion he was adamant that the cast could be found, and was probably
with Dunnavant. But eventually, he admitted defeat. The cast, it seemed,
Later, when I visited Columbus in November 2002, Galbreath told me
something else that sounded significant: the cast had revealed that one
of the strangler's upper front teeth was "rotated" 20 degrees or more
out of its proper alignment.
I asked Gene Hewell, the proprietor of the Movin' Man fashion store, if
he had ever noticed that Gary had a rotated front tooth. "Hey, he worked
for me as a model," Hewell said. "Believe me, there ain't too many
models with twisted front teeth." Later, I asked Gary himself. "The
business I was in, if I'd ever had a twisted tooth, it would have stuck
out a mile," he said. He meant stealing, not modeling clothes.
After more than seven years, Gary's federal habeas appeal in Georgia's
Middle District Court was rapidly approaching a conclusion. Judge Clay
Land had already ruled that many of the issues on which Gary had fought
his state appeals were procedurally barred. Of those that Land did agree
to consider, only two looked as if they stood any chance of success: the
serology and, just possibly, the bite cast.
Land issued his final, 95-page order on September 28, 2004. With regard
to the cast, he said that he was bound to defer to the judgment made
seven years earlier by the state habeas court: that, even had the bite
cast been found, it would have been irrelevant, because Gary had had
work done on his upper teeth.
However, in the section that dealt with serology and Gary's secretor
status, Land's order contained two surprises. First, he ruled that the
state's failure to give John Wegel's laboratory worksheets to Gary's
defense at the trial constituted a violation of his right to due process
under the 14th Amendment of the U.S. Constitution. Second, although he
endorsed almost all the decisions made by his great-uncle Judge John
Land and later by Judge Followill to refuse Gary funding, there was one
exception: "Given petitioner's repeated requests for a serological
expert, the court finds that the trial court was unreasonable in denying
Having raised Gary's hopes, however, Judge Land set about dashing them,
more conclusively than any court had before. To vacate Gary's conviction,
Land said he would have to be convinced that had these mistakes not been
made the jury would have come to a different verdict. Instead, Land
ruled that there was enough evidence to convict without resorting to the
"inconclusive" serology results.
The concluding paragraph of his long order was particularly bleak: "It
has been eighteen years since twelve jurors unanimously found Petitioner
guilty of crimes that warrant the imposition of the death penalty under
Georgia law. Petitioner has avoided the execution of this sentence by
taking advantage of every legal avenue available to him. This Court
finds no legal obstacle standing in the way of Petitioner's journey to
his final destination."
Gary filed an appeal with the 11th Circuit U.S. Court of Appeals, one
tier below the U.S. Supreme Court, against Land's rejection of his
federal habeas petition, and an oral hearing was set for the first week
of December 2005.
On the afternoon of November 9, four weeks before the hearing, Michael
McIntyre was poring over some paperwork in his office when the phone
rang. Almost exactly four years earlier, Susan Boleyn had called Jack
Martin to say that the state had located the missing laboratory
worksheets. This time, she was calling with news that was even more
"Michael," she said, "I have something to tell you. The folks in
Columbus have found the missing bite cast." Two years after testifying
that the cast was nowhere to be found, she said, James Dunnavant had
discovered it in the back of an old metal cabinet. Martin and McIntyre
filed a motion with the 11th Circuit asking for Gary's December hearing
to be indefinitely postponed: the whole factual basis on which this part
of the appeal had been decided was, it was clear, mistaken. The three
judges agreed at once, and returned the case to Columbus, in the hands,
once again, of Clay Land.
The belated discovery of the cast produced no signs of contrition from
Boleyn and the state. When Martin and McIntyre asked Land to allow an
expert to examine the cast, and for funds to pay him, Boleyn resisted,
arguing that the cast was not enough to get Gary a new trial, whether it
matched or not. Land did not agree: "Undermining the state's evidence of
the bite mark on Janet Cofer would have resulted in undermining the
State's whole case," he declared. There was only one way to find out
whether it really did—to allow the bite to be compared with Gary's teeth.
The defense had already chosen its expert: the same Dr. Thomas David who
had been shown the cast by Bill Smith, Doug Pullen, and the cops more
than 20 years earlier. David, said Land, would be paid with public
In February 2006, Dr. David visited the G.B.I. crime lab and inspected
the cast by naked eye. Several things were immediately apparent. First,
it was still attached to a signed and dated label, establishing that it
had, as claimed, been made by Dr. Galbreath in April 1978 from a bite
around Cofer's nipple. Second, the cast found by Dunnavant was not the
positive, ridged model that David was shown in 1984, but the cast stone
negative or exemplar.
Making an exemplar is the usual first step in comparing a cast with a
suspect. Back in March 2001, when everyone assumed that the cast had
been lost, Gary told me that an exemplar had been made from his own
teeth in 1984, when he was waiting to be tried at the Columbus County
Examination of the cast by microscope, David wrote in a report in May
2006, showed that the strangler's lower teeth were, as he had always
stated, crowded and uneven. The exemplar he was finally able to make
from Gary confirmed that his are straight and well-spaced. In December
2006, he delivered his final report. The cast, he said, revealed "several
inconsistencies" with Gary's lower teeth, a conclusion he stated to "a
reasonable degree of scientific certainty."
He also confirmed what Galbreath had told me in 2002—that the killer had
a wide gap between his upper front teeth, one of which was rotated.
Meanwhile, there was still more new evidence. In October 2006, I
traveled to the lakeside home in Americus, Georgia, of Jim Covington, a
retired special agent with the G.B.I. In the 1970s, he spent almost a
year as co-leader of the stocking-stranglings investigation. Covington
told me that, 20 years after Gary's trial, his concern that an innocent
might be executed had prompted him to come forward, and he gave me yet
more documents that had not been shared with the defense.
On the night of the terrors, Covington had taken charge of the crime
scene at the home of Ruth Schwob, who had survived the strangler's
attack. On the air-conditioning unit that the killer had climbed on to
enter her kitchen window, Covington found a footprint from a sneaker
sized 9.5 or 10—four sizes smaller than Gary's feet. Covington had
pictures, as well as a map he had made of the trail of footprints from
Mrs. Schwob's house to the home of Mildred Borom, the strangler's sixth
victim. He also had a memo containing Schwob's description of her
attacker—he was, she told the cops, "small and muscular." Gary is six
feet three inches tall, large, and powerfully built.
Newly discovered footprint evidence also undermines the state's claim
that Gary had killed Nellie Farmer in upstate New York, back in 1970. In
the Albany police files, I found yet more documents that were not shown
to Gary's Columbus jury: reports from forensic scientists saying that,
after raping Farmer, the murderer had stood at her sink and washed
himself—leaving two shoe prints on the bathroom mat. They were size
Finally, new doubts were cast on the fingerprints at the strangling
crime scenes that allegedly matched Gary's. I had always found it
strange that none of the prints had been photographed in situ when they
were found. Eddie Florence, a Columbus print-ID officer for 18 years,
told me that such photos were taken as a matter of routine procedure,
even in relatively minor cases such as burglary.
There were dozens of photos of the strangler crime scenes, but none that
included fingerprints. The crime-scene prints exist only as pieces of
tape, supposedly lifted from the victims' homes, stuck on pieces of
cardboard: other than the testimony of the police, there is no
independent evidence that proves they are what they are supposed to be.
At Gary's trial, District Attorney Smith told the jury that the Columbus
detective Richard Smith (no relation) had visited Gary in South Carolina
after his arrest there for robbery in 1979. Smith, he said, was "obviously
a good detective … and it's just unfortunate, members of the jury, that
he didn't come back and have his [Gary's] fingerprints compared with the
stocking strangler's latent prints on file in the Columbus Police
Department." Had he done so, Smith suggested, the case would have been
solved five years earlier.
Smith left the police department in 1979 and is now the president of
Realogy Corporation, the largest U.S. residential-real-estate brokerage
firm. In 2003, I interviewed him in his huge office overlooking Central
Park in Manhattan and asked him why he had not brought Gary's prints
back to Columbus. He said that in fact he had, together with samples of
Gary's saliva and pubic hair. He asked Columbus's print-ID section to
compare them with the prints from the strangling crime scenes. No match
was found. "I don't know how they missed that one," Smith told me calmly.
"They should have caught it if there were prints which could have been
identified. To this day, no one knows why they didn't."
On December 4, 2006, Jack Martin filed a brief with Judge Clay Land,
asking for another evidentiary hearing—to hear testimony about the bite
cast, the footprints from Columbus and Albany, and Richard Smith's trip
to South Carolina in 1979. Land agreed to hear testimony about the cast,
and convened another hearing on February 14, 2007. Martin called a
series of witnesses who had known Gary well in the 1970s, including Gene
Hewell of the Movin' Man: all of them said that Gary had never had gaps
or twisted teeth.
On that basis, Dr. David was able to go further than he had in his
affidavit. Taking the physical evidence of the bite cast and the
descriptions of Gary in happier times, he stated with scientific
certainty that Carlton Gary was excluded beyond a reasonable doubt as
the man who bit Janet Cofer. The state had no evidence with which to
rebut his opinion, and could not attack his credentials—it had used Dr.
David as a prosecution expert witness many times in the past.
During the years I have worked on this book, I have been asked the same
question innumerable times: do I believe that Carlton Gary is innocent?
It is the wrong question to ask. One must assume that the prosecutors
who tried the 180 men who have been freed from death row because of new
DNA tests believed they were guilty, but belief is no substitute for
properly tested evidence.
Instead, we need to put two different questions. The first is, did he
get a fair trial? My answer to that is a resounding negative. He was
starved of resources and was facing a prosecutor who not only hid
exculpatory evidence from the defense but lied. The second question is
whether we should entertain real doubt as to Gary's guilt. The answer to
that seems equally clear.
For now, the final act of this drama cannot yet be written. Gary's
defense has submitted a brief arguing that he should now get a new trial:
in making up his mind, it states, Judge Land must consider not only the
bite cast but all the other areas of the original prosecution case which
have now been thrown into doubt. At the end of March 2007, Susan Boleyn
submitted her reply. It said that the court should not have looked at
the bite cast at all, and that the defense is at fault for failing to
find it years earlier. Land's ruling is expected in May or June. Even
then, the matter will not end: the losing side, whichever it is, will
undoubtedly submit a further appeal, first to the 11th Circuit, and then,
perhaps, to the Supreme Court.
For the time being, however, history has come full circle. The scion of
the Lands who led the mobs that murdered Simon Adams and Teasy McElhaney
must decide whether to halt Gary's legal lynching.
(260 Ga. 38)
(389 SE2d 218)
SMITH, Presiding Justice.
Murder, etc. Muscogee Superior Court. Before Judge Followill.
Carlton Gary was convicted by a jury in Muscogee
County on three counts each of murder, rape and burglary. He was
sentenced to death on each of the murder counts.
1. Between September 11, 1977 and April 19, 1978,
eight elderly women were raped in their homes. One woman survived;
the other seven were strangled to death. Seven of the victims lived
in the Wynton area of Columbus. One victim lived two miles away, but
had attended choir practice in Wynton the evening she was murdered.
Police had no viable suspects in the case until
1984, when a gun stolen from the Wynton area in 1977 was discovered
in Michigan -- a consequence of that state's gun registration laws
-- in the possession of Carlton Gary's cousin. After further
investigation, Gary was arrested for burglary on May 3, 1984. His
fingerprints matched those taken from the scenes of four of the
Gary admitted to law enforcement officers that he
was present at seven of the crime scenes (the eighth he could not
remember), but claimed he was only a burglar. He blamed the murders
on another. Further investigation revealed that in other instances
in New York and in South Carolina, Gary had committed violent crimes
and blamed others. For example, he raped and murdered an 89-year-old
woman in her home in Albany, New York in 1970. His fingerprints were
found at the crime scene. Gary claimed one John Mitchell committed
the murder. Mitchell, however, was acquitted by a jury. In another
New York crime involving rape and burglary, Gary admitted only to
being a "lookout" and blamed the rape on another. In all these
cases, no evidence other than Gary's own statements and testimony
supported his claim that another person was involved in the crime
The defendant does not question the sufficiency
of the evidence, and we find that it supports the conviction.
Jackson v. Virginia, 443 U. S. 307 (99 SC 2781, 61 LE2d 560) (1979).
2. Trial counsel, August F. Siemon, represented
the defendant pro bono. His first move upon entering the case was to
secure the dismissal of the defendant's court-appointed attorneys.
Later, he sought additional court-appointed counsel. The trial court
denied the motion after Siemon stated: "There is no question in my
mind that I have the personal resources to provide representation to
[the defendant]." We note that the defendant was represented by two
additional attorneys (acting pro bono) during much of the pre-trial
The defendant also sought funds for forensic and
investigative assistance. These requests for funds were denied.
During the pre-trial proceedings and at the
outset of the trial, attorney Siemon stated to the trial court that,
because of the lack of funds and time, he was not prepared to try
the case. The case proceeded to trial over his objection. When the
case came to this court on appeal, we exercised our discretion under
the Unified Appeal Procedure to remand the case to the trial court
for a hearing to determine whether, for any reason, including lack
of funds, the defendant was denied effective assistance of counsel.
See Rule IV (B) (1) of the UAP, Ga. Court & Bar Rules at 9-15.
On remand, the trial court appointed two
attorneys to represent the defendant on the question of
effectiveness of trial counsel. However, the defendant obtained the
services of attorney Frank L. Derrickson, who, as had trial counsel,
represented the defendant pro bono. The issue came on for hearing.
The defendant refused to waive his attorney-client privilege as to
any of his three trial attorneys, and none of them testified on
behalf of the defendant. Lead trial attorney Siemon was called as a
witness by the state; however, the defendant invoked his attorney-client
privilege to prevent his attorney from answering any questions about
tactical decisions he made while representing the defendant.
(a) The defendant contends the court erred by
refusing to dismiss his two court-appointed attorneys after attorney
Derrickson entered the case on remand. However, although these two
attorneys were not dismissed, they did not actively participate in,
or interfere with, the proceedings on remand. Hence, there was no
denial of the defendant's right to choose his own counsel. Cf.
McKaskle v. Wiggins, 465 U. S. 168 (104 SC 944, 79 LE2d 122) (1984)
(court may appoint standby counsel for pro se litigant over his
objection, so long as standby counsel's role is suitably limited).
(b) The defendant contends he was denied
effective assistance of counsel at trial. The trial court found as
In this case ample opportunity was given this
defendant through a hearing before this court with counsel of his
own choosing (and additional counsel, if he chose so to avail
himself) to urge any and all possible errors affecting the trial or
prejudicing the defendant. . . . This defendant refused to accept
the opportunities provided him. For reasons probably known only to
itself the State elected to go further than even the defendant in
the course of the hearing and attempted to elicit from defense
counsel, August F. Siemon, reasons for various avenues of conduct.
To each and every material question the defendant asserted his
privilege not to have his counsel testify. This is true even though
the defendant was made inescapably aware by this court that this was
his opportunity to have these possible issues litigated. It is
therefore the finding of this court that as to those issues
addressed during the hearings on remand that the defendant knowingly,
intelligently and voluntarily waived those issues after having been
repeatedly advised by the court that his conduct would amount to
such a waiver. Therefore these issues are resolved adversely to
A defendant cannot be forced to litigate an issue.
Cf. Morrison v. State, 258 Ga. 683 (3) (373
SE2d 506) (1988). The defendant was given an opportunity to
prove he was denied effective assistance of counsel. He declined to
do so. We agree with the trial court that he knowingly, voluntarily
and intelligently has waived any issue of effectiveness of trial
(c) In his post-remand brief, the defendant
continues to claim, as he did in his original appellate brief, that
the trial court's refusal (prior to the original trial) to appoint
additional counsel or to provide funds for forensic and
investigative assistance was an abuse of discretion, see Isaacs v.
State, 259 Ga. 717 (13) (a) (386
SE2d 316) (1989), and that he was penalized improperly for
exercising his right to retain his own attorney pro bono.
3. The defendant contends the conditions of his
solitary confinement while awaiting trial had an adverse impact on
his ability to stand trial. However, not only has this issue been
litigated -- and relief denied -- in a pre-trial habeas corpus
petition filed by the defendant on this issue, but also in a special
trial to determine his competence to stand trial. The special jury
found him competent to stand trial. See OCGA
While awaiting his trial, the defendant was held
in "administrative segregation" for his own protection. He was in a
private cell with a commode, a lavatory, a bunk with mattress,
blanket and pillow, a skylight, a television, a radio, and reading
materials. He was given extensive visitation privileges. The record
does not support the defendant's claim of unconstitutionally harsh
conditions of confinement. The special jury's finding that he was
competent to stand trial is supported by the evidence.
4. The defendant was caught in an escape attempt
in December of 1985. Afterward, he allegedly began behaving in a "bizarre"
manner. On March 10, 1986, the day his trial was scheduled to begin,
he filed a special plea of incompetence. He was examined by Doctor
Jerald Lower, who concluded the defendant was deliberately feigning
mental disorder. The defendant claimed he was suffering from sensory
deprivation disorder; however, expert testimony was presented at the
competence hearing that such condition is associated only with
complete sensory deprivation not at all consistent with the
condition of the defendant's incarceration.
The defendant contends he was denied due process
by the court's refusal to grant him funds for examination by an
independent mental health expert. See Ake v. Oklahoma, 470 U. S. 68
(105 SC 1087, 84 LE2d 53) (1985). We disagree. Nothing before the
court reasonably indicated that the defendant's sanity would be a
significant factor at trial. The denial of funds was not error.
5. The percentage of black citizens aged 18 and
over in Muscogee County is 31 percent. The percentage of black
citizens aged 18 and over in Spalding County, to which venue was
changed (see fn. 1, supra), is 24.7 percent. The defendant contends
the selection of new venue in Spalding County was racially motivated.
The circumstances of this case do not establish prima facie that
racial discrimination motivated the trial court's decision to change
venue to Spalding County.
6. There was no abuse of discretion in the
conduct of voir dire. Curry v. State, 255 Ga.
215 (336 SE2d 762) (1985).
United States v. Meester, 762 F2d 867, 884-85
(11th Cir. 1985). The defendant now contends Judge Followill should
have been recused because he was a party litigant represented by
counsel at the hearing on the motion to recuse. This contention was
not raised until more than a year after the hearing and is not
timely. Romine v. State, 251 Ga. 208 (2) (305
SE2d 93) (1983). Compare Isaacs v. State,
257 Ga. 126 (355 SE2d 644) (1987).
8. The defendant was given ample notice of the
date of his competency trial.
9. The only surviving victim of the defendant's
strangulation attacks testified at trial and identified him as her
assailant. The defendant was not entitled to a hearing outside the
presence of the jury to determine the admissibility of the testimony
of this witness, Watkins v. Sowders, 449 U. S. 341 (101 SC 654, 66
LE2d 549) (1981), and her testimony was admitted in evidence
properly. Pruitt v. State, 258 Ga. 583 (4)
(373 SE2d 192) (1988).
10. There was no error in the conduct of the
Jackson-Denno hearing or in the admission in evidence of the
defendant's pre-trial statements. See, e.g., Parks v. State,
254 Ga. 403 (1) (330 SE2d 686) (1985).
11. As noted in Division 1, supra, extrinsic
transactions demonstrating the defendant's modus operandi were
admitted in evidence. In view of the relevant similarities between
the extrinsic transactions and the crimes on trial, the court did
not err by allowing their admission in evidence. Hamilton v. State,
255 Ga. 468, 471 (339
SE2d 707) (1986).
12. The defendant was not, as he claims, denied
the right to present mitigating evidence by the trial court's
rulings on funds and travel expenses. Many potential witnesses who
had known the defendant during his life were either present at trial
or were readily available. The defendant can name no one whose
testimony he was unable to secure as a consequence of any ruling by
the trial court.
13. The jury found that the three offenses of
murder were committed while the offender was engaged in the
commission of the offenses of rape and burglary. See OCGA
17-10-30 (b) (2). The evidence
supports these findings. OCGA 17-10-35
14. After reviewing the record, including the
proceedings on remand, we conclude the sentences of death were not
imposed under the influence of passion, prejudice, or other
arbitrary factor. OCGA 17-10-35 (c)
(1). The sentences of death are not excessive or disproportionate to
sentences imposed in similar cases, considering both the crime and
the defendant. The similar cases listed in the appendix support the
imposition of death sentences in this case.
Douglas C. Pullen, District Attorney, Michael J.
Bowers, Attorney General, Susan V. Boleyn, Senior Assistant Attorney
General, Joseph L. Chambers, for appellee.
1 The defendant was arrested on May 3,
1984. He was indicted the next day. After numerous pre-trial hearings,
and a change of venue, the case was tried in Muscogee County by a jury
that had been selected in Spalding County. The trial began on August 11,
1986 and concluded on August 27, 1986. A motion for new trial was filed
on September 25, 1986 and was denied on October 18, 1986. The case came
to this court for review. We remanded the case to the trial court for a
hearing on the question of ineffectiveness of counsel. The case was
redocketed in this court on June 22, 1989, and the case was argued
orally on September 12, 1989.
August F. Siemon III, Frank L. Derrickson, for appellant.
DECIDED MARCH 6, 1990 -- RECONSIDERATION DENIED MARCH 28, 1990.