FINE, sharp grains of ice blow in sheets across the highway and brush against the windshield. A gray-white sky covers an unrelievedly flat expanse of Canadian prairie. It's late March and well below freezing on a road north of Winnipeg. Ahead lies a beige fortress called Stony Mountain Institution. A medium-security prison, it holds about 465 convicts. One of them is David Milgaard, 38 years old and Canada's longest-serving prisoner. He's been in jail since he was 16, convicted of murdering a young woman.
In the entryway, the bars and walls are painted a pale yellow. Mr. Milgaard comes through the door silently, wearing worn moccasins. He has on a tired velour pullover and faded olive-green pants. His handshake is soft, weak; his fingernails are too long. An index finger is tawny from nicotine.
Milgaard is at the center of a controversy about whether he is guilty. He has consistently maintained his innocence. He has also escaped twice and attempted suicide.
Milgaard's lawyers and advocates have raised doubts about his conviction - at least in the court of public opinion. Newspapers in Western Canada have covered the story intensively for the last two years, and Milgaard's case has been examined on the ``Fifth Estate,'' one of this country's premier television newsmagazines. ``Free David Milgaard'' bumper stickers appear on Winnipeg cars.
Several disclosures have increased the attention on this case: A crucial witness at Milgaard's trial says he lied on the stand, forensic pathologists have discredited the physical evidence, and Milgaard's supporters have identified a possible perpetrator who is in jail on a rape and attempted murder conviction.
``I will not say [whether or not] the man is innocent,'' says John Harvard, a Liberal Party member of Parliament from Winnipeg, ``but I think that the case against him lies in tatters.''
Many people, however, are certain that Milgaard is guilty, including A. Kim Campbell, Canada's minister of justice. There ``is no basis to conclude that a miscarriage of justice may have occurred here,'' she wrote last February in a letter denying a formal request to reopen the matter.
There are also indications that the process used in Canada to review convictions like Milgaard's - those upheld by appeals courts but still in dispute - is flawed. Four Canadian lawyers and law professors interviewed for this article say the process needs reform. A top-level government commission reached the same conclusion in 1989.
Some critics are concerned about the impartiality of the Canadian Justice Department officials who investigate cases like Milgaard's. Four people involved in the Milgaard case, three who were interviewed by Justice Department counsel Eugene Williams and one who witnessed a fourth interview, say Mr. Williams did not appear open-minded or impartial. One of the four, forensic pathologist James Ferris, says: ``They were trying to confirm in their own minds that the new facts and new evidence introduced [on Milgaard's behalf] were in fact not new facts or new evidence, or [were] irrelevant.'' Williams disputes these accounts.
Reviewing the facts
Milgaard is serving a life sentence for the Jan. 31, 1969, murder of Gail Miller, a 20-year-old nursing assistant who was stabbed to death in Saskatoon, a small city in Canada's Saskatchewan province. On the day of the murder, Milgaard had arrived in town with two traveling companions and picked up a third.
The victim was found partially unclothed and there were indications of sexual assault. Four days after the crime police found two small clumps of frozen, yellowish liquid in the snow where the body had lain.
A month later the police investigation was ``at a standstill,'' recalls Eddie Karst, a retired Saskatoon detective who was involved in the case at the time. The break came when one of Milgaard's companions, Albert Cadrain, told the police he'd seen blood on Milgaard's pants on the morning of Jan. 31. He later collected a C$2,000 ($1,850) reward for the information.
The prosecution built its case on the testimony of Milgaard's other companions, Ronald Wilson and Nichol John. They said they had seen him with a knife, and Mr. Wilson said Milgaard had told him he'd ``hit a girl in Saskatoon.'' He said he also saw blood on Milgaard's pants. A police specialist said one of the samples of liquid found at the scene contained human semen and raised the possibility that it could be linked to Milgaard.
Two last-minute prosecution witnesses told the jury that they had spent time with Milgaard in May 1969, just before he was arrested. After they had seen a television news report on the case, one of the people with Milgaard sarcastically accused him of the crime; everyone knew he had been questioned by police. Milgaard then reenacted the crime by stabbing a pillow and saying ``I killed her,'' among other things.
After 11 hours of deliberation, the jury returned a guilty verdict. Milgaard's lawyer appealed the conviction to the provincial appeals court. He attacked the reading aloud of inconclusive ``eyewitness'' testimony, arguing that it was unfair to Milgaard. The appellate court disagreed; the trial judge had handled the situation properly, it ruled. The Supreme Court of Canada refused to review the case. By the end of 1971, Milgaard's appeals were exhausted.
`Not a choirboy'
The most vocal advocate for Milgaard's innocence has been his mother, Joyce, although the other members of the family are involved in the attempt to exonerate him.
``David was certainly not an innocent, choirboy type of boy,'' she said in an interview. ``He was living a wild lifestyle, but he was never vicious.'' At age 16 Milgaard was a frequent drug user and was sexually active, she says.
After the conviction and the unsuccessful appeals, recalls Mrs. Milgaard, the family was derailed. ``I was so numb at the time,'' she adds.
Milgaard was an unruly, uncooperative prisoner during his early years in jail - behavior brought on by unjust imprisonment, he says. Parole seemed like a Catch-22: He had to show remorse to convince the board of his rehabilitation, and remorse requires an admission of guilt. He wasn't prepared do that, he says. Frustration led to suicide attempts and two short-lived escapes during his first decade in prison.
``I was dying a little bit,... something inside me, every day.'' So, he explains, ``I stole my freedom.''
In 1980, Milgaard was on the run for several months, making his way to Toronto, where he found a job and got an apartment. He even started a relationship with a woman. But in November 1980, the Royal Canadian Mounted Police found him and chased him on foot. He was shot in the back and captured.
The shooting spurred Joyce Milgaard to renew the attempt to exonerate her son. She offered a reward in Saskatoon for information, and started to interview witnesses and gather evidence for a new legal effort. In 1986, she contacted Winnipeg defense lawyer Hersh Wolch, who asked a young associate, David Asper, to look into the case.
In December 1988, Mr. Asper made a formal application to Canada's minister of justice to reopen Milgaard's case. Canadian law gives the minister the right to exercise ``the mercy of the Crown'' by ordering a new trial, or asking an appellate court to render an opinion or hold a hearing. It is an extraordinary procedure, and the applications are rarely successful. But it is regarded as a safety valve designed to prevent miscarriages of justice.
Doubts about evidence
The Milgaard application included a report from Dr. James Ferris, a forensic pathologist who works in Vancouver. Ferris wrote that the forensic evidence at the trial ``could reasonably be considered to exclude [David Milgaard] from being the perpetrator of the murder'' because there was no evidence that the frozen yellowish substance could be tied to Milgaard.
Another report from Peter Markesteyn, the chief medical examiner for Manitoba province, concluded that the substance was probably ``of canine origin.''
``You can shoot a cannon through the whole forensic case,'' the Manitoba medical examiner says now. He adds that his review of the record suggests that the forensic evidence may have bolstered the prosecution's case in the eyes of the jury.
Mrs. Milgaard found another person who was with Milgaard when he ``reenacted'' the murder. Deborah Hall, now a hairstylist in the Saskatchewan capital of Regina, said Milgaard wasn't stabbing a pillow, he was fluffing it up. His acknowledgement of the crime was merely a sarcastic response to a sarcastic question: No one in the room thought twice about it, she said.
Milgaard's lawyer and his mother encouraged reporters to write about Milgaard after concluding in early 1989 that the Justice Department wasn't moving on the case quickly enough. Stories in the print press began to break in late summer of 1989. Soon Canadian television news shows were featuring David Milgaard.
In January 1990 Asper's senior partner, Mr. Wolch, got a phone call from a man named only ``Bud.'' He said he knew who killed Gail Miller, and it wasn't David Milgaard.
James McCloskey, who runs a Princeton, N.J., organization that investigates the cases of convicts he believes are innocent, says the anonymous phone call and the identification of a possible perpetrator convinced him to get involved in the case. His investigations have freed eight people from US jails over the past eight years; in each case, prosecutors or judges have concluded that Mr. McCloskey's clients were innocent of the serious crimes for which they were convicted.
McCloskey sent Paul Henderson, a Seattle-based private sleuth, to work on the case.
The investigator and Mrs. Milgaard looked into the caller's allegation that Larry Fisher - a former Saskatoon resident now in prison on a rape and attempted murder conviction - was responsible for Miller's murder.
Mr. Fisher's ex-wife Linda told them he stayed out late the night before the murder. In a heated argument the next morning she angrily accused him of killing Gail Miller - more out of a desire to antagonize him than out of a belief that he was responsible, she says.
What stunned her, in retrospect, was her husband's reaction. ``I picture Larry's shock to this day,'' she says now - the color drained from his face.
Henderson and Mrs. Milgaard weren't the first people to hear Linda's story about the morning of Jan. 31: In 1980 she told the Saskatoon police. Saskatoon Deputy Chief Ken Wagner remembers taking Linda Fisher's statement, although not the outcome of any further investigation. ``We're through with that file as far as I'm concerned,'' he says, adding that he is convinced of Milgaard's guilt.
But recent research into Larry Fisher's past has led to some startling discoveries, at least to Milgaard's supporters. According to police records and newspaper clippings, Fisher raped two women and assaulted a third in Saskatoon in the three months before Miller's murder. In 1971 he pleaded guilty to these crimes, as well as other attacks against women in Saskatoon and Winnipeg.
In December 1968, six weeks before the Miller attack, Saskatoon police urged women to be careful, in light of two recent rapes and an assault, according to a newspaper article uncovered by Winnipeg journalist Carl Karp. Police ``said the alleged assailant first talks to women and then takes them into alleys,'' the article says. Fisher lived in Gail Miller's neighborhood, and Milgaard's lawyer David Asper says Fisher rode the same city bus she did. She was found in an alley near the bus stop they apparen tly shared.
He remains in jail, serving time for attacking a woman while he was on parole in 1980, but is due out in 1994. In an interview aired on Canadian television he denied killing Gail Miller.
The Milgaard team sent information about Fisher to the Justice Department, to be included in the review of Milgaard's case.
A witness recants
In June 1990 Paul Henderson interviewed Ronald Wilson and Albert Cadrain at their homes in British Columbia. Both gave him statements that cast doubt on their trial testimony.
Wilson, calling his interrogation by police a ``sweat session,'' said he had no recollection of seeing blood on Milgaard's clothes. He also denied seeing a knife in Milgaard's possession before their arrival in Saskatoon or that Milgaard had said he'd ``hit a girl.''
``I was manipulated into lying against him [by police],'' Wilson told Henderson, ``manipulated into believing my own lies.''
Cadrain did not deny seeing blood on Milgaard's pants, but called his police questioning ``hell and mental torture'' in a statement to Henderson. The investigator says he was not able to interview Nichol John, the third person with Milgaard the day of the crime.
By the time they finished their submissions to the Justice Department, the Milgaard team felt they had demonstrated that the forensic evidence put before Milgaard's jury was completely meaningless and possibly misleading, that the account of the reenactment was suspect, and that the testimony of the car companions was now undermined. And they had pointed the way to a possible perpetrator.
Officials from the Justice Department were not very cooperative, says lawyer David Asper, but they were traveling around Canada interviewing many of the people in the case, including Larry Fisher, the forensic experts, and people who testified at the trial.
Asper had filed the application for mercy in December 1988 and forwarded new information as it developed until June of last year. Last February 27 the Justice Department faxed him and Wolch a 12-page letter explaining Minister of Justice Campbell's decision. The answer was no.
She agreed that the forensic testimony was meaningless; but she said that this was clear to the jury. Deborah Hall's contradiction of the reenactment scene was a matter of ``interpretation.'' And Ronald Wilson's ``retraction ... of much of his trial evidence is unconvincing.'' As for Larry Fisher, the minister concluded: ``This application reveals no evidence to connect him with the killing of Gail Miller.'' She denied the allegations of police coercion. (Retired Saskatoon detective Karst al so disputes the accounts of police pressure; he says Milgaard is guilty.)
Canadian lawyers and law professors who have been involved in these kinds of cases say the process used to review possible wrongful convictions is flawed.
Anne Derrick is a lawyer in Halifax, Nova Scotia, who represents Donald Marshall Jr., a Micmac Indian who was exonerated of a murder conviction in 1983 after serving 11 years in prison. Ms. Derrick says one problem with Section 690 of Canada's criminal code - which discusses the application for mercy - is that it is vague.
Because ``no clear standard is articulated,'' she says, applicants do not know what they have to prove in order to convince the minister that further judicial process is warranted. David Asper, in representing Milgaard, grappled with this problem. He and partner Wolch met in Ottawa with Justice Department officials during the review. ``They said to us,'' Asper recalls, ```What is the standard that we have to apply here?'
``And our position was [that the applicant has] to show that there may have been a miscarriage of justice. Period. Because if you show that there was a miscarriage of justice you don't need the minister - then the guy's entitled to a free pardon. So what they did was they took something in between and they said that you have to establish that there was likely a miscarriage of justice,'' Asper says.
He adds: ``Now, I don't know the difference, frankly, between `likely' and `was.' `Likely' may be a little bit more equivocal, but it's still more than `may,' and so they define it as they go along. So that the applicant never knows what test you have to meet.''
Applying for mercy ``is a hopelessly flawed process,'' says Michael Jackson, a University of British Columbia law professor. He has handled one unsuccessful Section 690 case and has just started working on a new one.
He notes that the standard for conviction in a criminal case is ``beyond a reasonable doubt.'' So raising reasonable doubt to the minister, he says, ``is as far as you should have to go.'' But ``the reality is you have to ... demonstrate the conviction was wrong.''
The statute also does not specify how the minister should carry out her discretion. It does not say that she is to assume the role of a judge, although Asper says Campbell has done so in the Milgaard case. Of the forensic material, she writes: ``With the benefit of hindsight, it may have been preferable had the evidence simply not been tendered. Nevertheless, the case against Milgaard was a strong one.''
Campbell also judges Wilson's retraction ``unconvincing'' and says Deborah Hall's conflicting account of the reenactment ``would not have detracted'' from other evidence about it. Asper calls these conclusions ``factual determinations,'' and says that function is normally performed by a jury or judge, not a government official.
Douglas Rutherford, associate deputy minister in Campbell's Department of Justice, acknowledges in a telephone interview that the minister's role in scrutinizing applications for mercy is ``quasi-judicial.''
``In this day and age,'' says Mr. Rutherford, ``when the courts are overcrowded, you don't just throw a complex matter to the court without some responsible basis for doing so.'' The minister ``has to look at the evidence and find some reasonable basis,'' he stresses, ``for concluding that a miscarriage has occurred with some degree of probability.''
There is, he adds, ``some onus on an accused ... to do more than say something may have gone wrong here.'' The standard for conviction in a criminal case should not be confused with what is required in a 690 application, he says. ``This remedy of mercy begins where legal rights end.''
`By their own rules'
Section 690 cases also are not handled in accordance with time-honored principles of how officers of legal systems should carry out their work. These ideas are known in the US as due process and in Canada as natural justice.
In criminal cases, these lawyers say due process demands that witnesses and evidence be presented in an open hearing where both sides are present and may speak. A public record should be made of the events. And the facts should be decided by an impartial judge or jury.
But Stephen Aronson, who as a private lawyer won Donald Marshall's exoneration and who now works for the Justice Department, says that reviews of 690 applications lack these characteristics.
Mr. Aronson notes that Justice Department investigators interview witnesses privately - nowhere near a lawyer for the applicant. The applicant also does not know what evidence the department is gathering and has no way to challenge its appropriateness or validity, he says.
``It's not an open process,'' agrees Professor Jackson. ``These investigations are conducted by the department by their own rules.''
Also, says Jackson, ``there's enormous resistance on the part of government officials and [senior Justice Department managers] to acknowledge that their paradigm of justice can in fact be flawed.'' He says their impartiality in these cases is questionable.
Justice Department investigator Eugene Williams did much of the legwork during the review of the Milgaard case. According to the impressions of three people he interviewed and a lawyer who was present for a fourth interview, Mr. Williams seemed primarily interested in fitting new information to the conclusion that Milgaard was guilty:
* Linda Fisher says Williams repeatedly encouraged her to admit that it was ``possible'' that Larry Fisher's shock at her accusation could have resulted from his concern that she might have learned about his earlier rapes. She had said she was surprised to find him home and not at work; Williams asked her to acknowledge that it was ``possible'' that he had returned home mid-morning.
* Deborah Hall says that during her interview with Williams ``everything was twisted to their advantage and interpreted to their advantage.'' She says Williams ``made things fit where they wanted them to.''
* Ronald Wilson's lawyer, Ken Watson, was present for Williams's questioning of his client. ``What [Williams] personally thought,'' Mr. Watson says, ``I think came through to me at the time - that he probably thought there was no case.
``There's some justification for saying that the investigation was approached on the basis that there would have to be something pretty substantial surfaced before there would be any changes,'' Watson adds. He said Wilson wouldn't speak to this reporter.
* Dr. Ferris, the Vancouver forensic pathologist, says the minister's decision was a foregone conclusion when he was interviewed by the investigator. ``Mr. Williams was here to defend the decision that was going to be made'' confirming Milgaard's guilt, he says.
The Monitor questioned five people Williams interviewed, but only one found him objective and openminded. ``My impression,'' says Manitoba chief medical examiner Peter Markesteyn, ``is that without putting any pressure on me, he wanted to hear what I had to say and why I said it.''
Williams disputes the four critical accounts, saying that transcripts of his encounters with Deborah Hall, Linda Fisher, and Ronald Wilson would disprove the charges. (Neither the transcripts, nor any other documents uncovered by the department's investigation, are public records.)
``I have a duty to examine these matters thoroughly and impartially. I have no stake in the outcome and I approach them on that basis,'' Williams says. ``Whether a witness's perception is this or is that is entirely their business.'' He says he conducted three dozen interviews during the Milgaard investigation, but will not say how many were done in person. This reporter spoke to Williams over the telephone.
Both Williams and Rutherford say that 690 investigations are secret because the department must promise anonymity to witnesses who do not want to speak publicly. Investigators do not have the power to compel testimony. Rutherford also says making the 690 process conform to due-process standards would eventually create a new court of appeal in the Canadian justice system.
``There's nothing fundamentally unfair about [the 690 process],'' Rutherford insists, ``but it's a different kind of proceeding than the adversarial proceeding that occurs in the courts.''
In Nova Scotia, after Donald Marshall's conviction was overturned, a Royal Commission was formed to analyze the problems that the Marshall case exposed in the justice system. The commission's first recommendation was that Canada consider ``constituting an independent review mechanism - an individual or a body - to facilitate the reinvestigation of alleged cases of wrongful conviction.''
A vestige of royalty?
Prof. Ed Ratushny of Ottawa Law School calls the existing mechanism a ``vestigial bit of [royal] prerogative. I don't think it's appropriate in a modern society.'' Like Professor Jackson and lawyers Derrick and Aronson, he says reform is in order.
Back at Stony Mountain Institution, where David Milgaard is talking to yet another reporter, the calls for reform aren't heard. Here the concerns are more particular.
Of his case - the material accumulated by his mother, his family, his lawyers, his supporters - he says: ``All these things are real. They're tangible, you can touch them, you know, there's no reason they should be ignored. But they seem to be a house of cards by the time it gets to a person or a group of people in a bureaucracy ... because they suddenly disappear. And that's very hard on me.''