The Murders And The Journalists

by Evan Hughes

Produced in partnership with Tumblr Storyboard. Illustrations by Tully Mills.

In February 1970, at Fort Bragg in Fayetteville, North Carolina, a pregnant woman named Colette MacDonald and her two children, Kimberley, 5, and Kristen, 2, were slaughtered in their home. Colette’s husband, Jeffrey MacDonald, a 26-year-old doctor and Green Beret at the time of the crime, was convicted of the murders in 1979. MacDonald faces the next of countless court dates on September 17, still seeking exoneration. The MacDonald case has been an object of obsession and controversy for more than four decades and the subject of high-visibility journalistic debate. But respectable opinion has always vastly favored the jury verdict of guilt. Errol Morris is trying to change that.

In one highlight of his career as a documentary filmmaker, Morris’ 1988 investigative documentary The Thin Blue Line led to the release of Randall Dale Adams, who had been serving a life sentence in the killing of a Dallas policeman. After many years of reporting, Morris has written a new book, A Wilderness of Error, that argues that Jeffrey MacDonald, too, was wrongly convicted. Morris directly challenges prior accounts by Joe McGinniss and the much-revered Janet Malcolm. Morris’ book, published yesterday, infected me with the virus of his fascination with the case and sent me to consult other sources. Can Morris be right? Is a man who tried and failed to prevent his family from being killed now serving three consecutive life terms for the crime?

The MacDonald case gained widespread fame with the publication of Joe McGinniss’ massive bestseller Fatal Vision in 1983. MacDonald had granted McGinniss extraordinary access — for legal reasons he was actually formally added to the defense team — in exchange for a share of the proceeds of the resulting book. McGinniss then concluded that MacDonald was in fact guilty of the murders, though he kept that belief from MacDonald while consistently assuring him of his friendship. MacDonald discovered the truth on camera before a national audience when Mike Wallace read him passages from the manuscript on “60 Minutes.” More than thirty million people watched each episode of the TV miniseries adapted from Fatal Vision, engrossed by the story of a doctor with good looks and a sterling résumé — Princeton, Northwestern, Columbia-Presbyterian, the Green Berets — who butchered his whole family.

Janet Malcolm’s seminal book The Journalist and the Murderer, which originated as a bombshell 1989 two-part article for The New Yorker, covered the lawsuit that MacDonald brought against McGinniss for fraud and breach of contract. Malcolm badly damaged McGinniss’ reputation but posed his duplicity toward MacDonald as merely “a grotesquely magnified version of the normal journalistic encounter.” Her famous and infamous opening sentence: “Every journalist who is not too stupid or too full of himself to notice what is going on knows that what he does is morally indefensible.” Despite assailing McGinniss on several fronts, however, Malcolm did not question his verdict that MacDonald was the killer.

MacDonald’s story always seemed unlikely. The night of the murders, he telephoned for help at 3:33 a.m. Military police found the bodies of Colette MacDonald and her daughters in their respective bedrooms. An ice pick and two knives had been used to stab them between 8 and 48 times, and Colette and Kimberley were disfigured by heavy blows from a lumber-wood club. The word “pig” was written in blood on the headboard near Colette’s corpse. The policemen found Jeffrey MacDonald injured, but alive.

MacDonald’s account, offered from the first, was this: a group of three men attacked him on the living room couch, where he had fallen asleep after reading. A woman with a wide-brimmed, floppy hat and long blonde hair and boots stood by during the assault. She was holding what appeared to be a candle and reciting phrases such as “acid is groovy” and “kill the pigs.” One of the assailants was black and wore an Army jacket with service stripes. The two others were white. MacDonald was clubbed. His pajama top was pulled over his head, and when it became caught up between his arms, he used it to fend off attackers. MacDonald heard Colette and Kimberley call out to him but his struggle was fruitless and eventually he lost consciousness. When he came to, he went to his family’s aid, but his wife and daughters were dead and the intruders were gone.

The prosecution’s key contentions were these:

1. There was no evidence of intruders.

2. Jeffrey MacDonald’s pajama top, found draped on Colette’s chest, had 48 holes in it, front and back. The evidence suggested that MacDonald used the ice pick found outside the house to stab Colette through his pajama shirt — that her 21 ice-pick wounds and the 48 holes were made by the same blows, with the blade passing through multiple layers of fabric.

3. MacDonald intentionally injured himself; his wounds were too minor to match his story and to correspond with the number of holes in the pajama top. He staged the living room scene where he claimed to have been attacked; signs of struggle were too few, with items placed in improbable positions.

4. The lumber-wood club used in the killings came from MacDonald’s house. And three people contradicted his claim that his family did not own an ice pick.

Morris’ case centers on the following:

1. Helena Stoeckley, a drug-addicted young Fayetteville woman who mixed with a crowd with a volatile history, was known to wear a blond wig and a floppy hat and boots. A military policeman on his way to the crime scene saw a woman in a wide-brimmed, floppy hat standing in the rain at 3:45 a.m. near the MacDonald house, before he heard MacDonald’s description. Several witnesses saw Stoeckley or a woman matching her description nearby in the hours surrounding the murders, accompanied by several men, one of them black, the others white.

2. Helena Stoeckley, who died in 1983, said numerous times — with varying degrees of certainty but sometimes in significant detail — that she recalled being present in the MacDonald home during the murders and knew the identity of the true killers, her companions.

3. There is evidence to suggest that Stoeckley was in the MacDonald house the night of the killings.


After a major event, mainstream opinion tends to fall heavily on one side of any element of controversy — often, not coincidentally, the “official” side. The best-supported theory tends to become official, and the stamp of officialdom confers additional credibility on that narrative. The problem with most alternative theories, which usually involve conspiracies, is that too many implausibilities need to be explained away, such that Occam’s razor cuts the other way. For the buildings at Ground Zero to have been destroyed by controlled demolition, we need to accept, apart from extraordinary evil “on the inside,” the existence of a widespread, perfectly executed conspiracy without a leak. An elaborate network of improbabilities is required, whereas the official story has relatively few anomalies to account for — not zero, mind you, because inevitable anomalies like weird puffs of debris are what give rise to conspiracy theories to begin with, but relatively few. We balance the troublesome facts on each side to find the ones we can live with.

In the MacDonald case, true to form, the officially sanctioned judgment has carried the day: guilty. The title of Janet Malcolm’s book, after all, is The Journalist and the Murderer. Although the syndicated columnist Jack Anderson, for one, backed the defense, MacDonald’s supporters have gained little traction. Too improbable, they’re told. Given the conviction, the failed appeals, and the popularity of McGinniss’ book, mentions of potential innocence tend to meet with immediate resistance, as Morris laments. Gene Weingarten of the Washington Post wrote in a 2011 online chat:

It is impossible to read “Fatal Vision” and not KNOW that he killed his wife and child. Impossible.

My wife is good friends with the prosecutor from that case.

Jeffrey MacDonald did it. Whatever happens to him, including re-trial and exoneration: Jeffrey MacDonald did it.

I approached Morris’ book with a radical skepticism I tried my best to maintain. And I read the three other major books on the case. I am not so certain that MacDonald is guilty. On the contrary, the MacDonald case appears to present an instance where the official account faces the more substantial array of implausibilities in need of explanation.


Let’s balance the anomalous facts that trouble each position, beginning with the case for innocence. The fact that the murder club found outside the house clearly came from MacDonald’s supply of spare lumber, though MacDonald said he did not recognize it in a photograph, does not look good for him at all. (“Maybe it was around,” MacDonald said when questioned further.) The wood could have been a weapon of opportunity picked up and used by the intruders, but the more likely explanation would be that MacDonald wielded it, and his reaction to the photo seemed untruthful.

As the prosecution pointed out, MacDonald denied owning an ice pick, and three witnesses contradicted him. But the testimony of two of those witnesses can be heavily discounted; both had said the opposite years earlier, and one was Colette’s mother, who in the interim had turned passionately against MacDonald. The third witness, however, was a close friend of MacDonald’s who testified that at a party three months before the crime, MacDonald had been looking around the house for his ice pick, though he did not find it. In early questioning, MacDonald said, “No, I didn’t have an ice pick,” then added, “Not that I know of.” Under oath, MacDonald said he was certain that the ice pick entered into evidence was not his. Unlike with the club, nothing is clear, but if we suppose both those weapons did come from the house, an argument for innocence requires us to accept that MacDonald appears to have foolishly tried to mislead investigators anyway and that the intruders apparently came to the house with only a knife or two. On the other hand, Stoeckley claimed the plan was only to intimidate. Still, another mark against MacDonald.

Morris succeeds in showing that the argument that MacDonald staged the living room is remarkably weak, but what about the business of the pajama top, which Morris calls “the Shroud of Turin of the MacDonald case”? MacDonald said he took off the shirt when he went to Colette’s aid and laid it on her chest in a futile gesture to keep her warm and prevent shock, though he suspected she was dead. Prosecution witnesses, brought in to rebut that story, demonstrated that the top could be folded such that the 21 stabs of the ice pick into Colette’s body could make the 48 holes in the shirt. Following the defense’s line, Morris portrays this exercise as entirely meaningless because no serious attempt was made to match the way the top was folded to the crime-scene photos (which in any event differed from one another, with the original position of the top in dispute). Morris writes, “Could you also find a way to match up the holes if the pajama top was bunched up, as MacDonald said it was, around his hands? Yes. Call it the origami problem.”

Morris’ argument does not strike me as a knockout. Without having seen the demonstration, it’s hard to know what to make of it. It does seem problematic for MacDonald that if his alleged attacker was stabbing with the ice pick enough to produce 48 holes, one would expect a number of blows to have landed. The extent of his injuries is also disputed, and he exaggerated them at various times, a stupid mistake if he was innocent. He did have a head wound consistent with being clubbed, small puncture wounds, and a stab wound between his bottom two ribs that was shallow but partially collapsed his lung. Nevertheless, his self-defense seems to have been improbably effective. One might counter that the prosecution never offered a convincing reason that MacDonald would have stabbed Colette through his own shirt. As Morris writes, “Wouldn’t he understand that he was leaving incriminating evidence?”

Fibers from MacDonald’s torn pajamas were found beneath Colette’s body, a fact he could not account for. “[O]bviously the implication is real bad for me,” he said under intense questioning upon first realizing he was a suspect, “but I can’t — how can I explain that? I don’t know.” He later said, not entirely credibly, that he had moved Colette’s body slightly. Things may be looking dire for MacDonald here, though we should keep in mind that a government brief called the pajama top “the single most inculpatory piece of evidence.” It was the best the prosecutors could do to put any of the weapons in his hand, rather than the hand of an unknown assailant.


Now what if we turn things around and point out the problems facing any case for guilt? Suppose that MacDonald killed his family and contrived the home invasion story. Now consider his magnificent luck. How lucky he is that Helena Stoeckley’s appearance and movements that night corroborate his bizarrely specific and unpromising attempt to deflect blame, when he could have invented a hazier story that cast a much wider net. How lucky that MacDonald’s completely fictional tale of wild hippie intruders — borrowed from the recent Manson case, as the prosecution would have it — happened to pin a tail on nearby people without an alibi. And most of all, how lucky that Stoeckley then repeatedly confessed, falsely each time — to investigators and journalists, twice during polygraph tests that supported her truthfulness, and twice to her mother, by the mother’s account.

Stoeckley’s first admission, a tentative one, came the night after the homicides, when a policeman whom Stoeckley had worked with closely as a drug informant confronted her. He had heard MacDonald’s description and thought immediately of Stoeckley and her crowd. He later testified, “She backed off and hung her head. She says, ‘In my mind, it seems that I saw this thing happen’; but she says, “I was heavy on mescaline.’”

Stoeckley later said that the hopped-up perpetrators she accompanied were trying to teach MacDonald a lesson for how he treated addicts as a doctor — that he refused them methadone and, they suspected, reported them to authorities. Intimidation escalated chaotically to murder. According to Stoeckley’s brother (who surprised Morris by agreeing to an interview), Stoeckley told her mother, “It got out of hand, and it just went crazy.”

In 2007, Stoeckley’s brother summoned one of MacDonald’s attorneys to Fayetteville so that Stoeckley’s mother (also named Helena) could sign an affidavit about what her daughter had told her before she died. The mother had once felt that her daughter couldn’t be relied on, but she changed her mind. Of the second confession, her affidavit stated:

My daughter knew she was dying. She wanted to set the record straight with her mother about the MacDonald murders, and that she wished she had not been present in the house and knew that Dr. MacDonald was innocent.

Let’s try to explain all this away. Stoeckley seems to have been intelligent, but she was on some level disturbed. Accounts vary widely as to her mental fitness at different times — guess who says what? — but I think “disturbed” is a fair word. Evidence of intelligence and disturbance can both be found in this interview filmed for “60 Minutes” in 1982. (Morris notes that this confession should be viewed with more skepticism than others. The interviewer, a former FBI agent named Ted Gunderson, had hounded Stoeckley and possibly planted ideas. He himself was an odd character in the MacDonald saga — unhinged, “lost in a funhouse of his own contrivance,” Morris writes.) In any event, Stoeckley was at the very least using drugs heavily in 1970. She often contradicted her admissions, claiming instead that she had no memory of that night between taking mescaline with her then-boyfriend, Greg Mitchell, around midnight and returning to the house around 4:30 a.m.

Suppose Stoeckley’s confessions were all untrue, their details cribbed from news coverage. The lies were motivated by a desire for publicity, and she doubled down on her fabrications with her mother. Or perhaps she was so unstable that she mistakenly believed she was there. Two men she implicated, including Mitchell, underwent polygraphs that supported their denials, and in one statement she implicated a man who could not have been present because he was in jail. False confessions are shockingly common, and some of her questioners did push.

Here we encounter another problem because Mitchell, the boyfriend, later also confessed repeatedly, though not in detail and not to officials or journalists. After he said at a prayer meeting that he had murdered in the past, the next day he was seen running from a farmhouse where someone had recently painted on a wall, “I killed MacDonald’s wife and children.” (The former L.A. County coroner Dr. Thomas Noguchi, hired by the defense, examined the MacDonald autopsy results and gave the opinion that there were multiple killers and that one was left-handed; Mitchell was left-handed, and MacDonald is not.) Cathy Perry, an acquaintance of Stoeckley’s, also told the FBI that she was involved, but she got many details wrong and she was psychologically unstable (with a history of stabbings actually). Dismiss these two suspects as well, for the sake of argument.

Supposing they are all innocent, a possibility presents itself: could MacDonald have known Stoeckley and her friends and made a decision to pin his crime on them? This theory actually has a degree of surface plausibility. The prosecution suggested that MacDonald was making a poor attempt to cast blame on the city’s hippies — a grand jury witness noted that the phrase “acid is groovy” registered as dated and false, and it does sound phony — but not that he was framing specific people; the government hewed to the line of an invented web of lies.

MacDonald treated addicts both on the Army base and at his second job in a civilian emergency room. With troubled soldiers returning from Vietnam and mixing with a late-hippie element, Fayetteville, population 54,000 at the time, had a notable drug problem. Stoeckley said MacDonald had treated people in her circle, while denying that she herself had ever met him. But Stoeckley’s brother told Morris that she had encountered MacDonald through his work, that “somewhere their paths crossed.” In the 1995 book Fatal Justice, coauthors Jerry Allen Potter and Fred Bost investigate potential doctor-patient connections between MacDonald and the Stoeckley crowd, but do not discuss the idea that he therefore could have framed them, and neither does Morris. (Potter and Bost explore the territory to argue the other way — to posit potential motives on the part of the alleged intruders.)

For our frame-up theory, we still need MacDonald to be lucky indeed that Stoeckley appears to have been spotted with her companions several times nearby but crucially not during the crime. (No one contradicts MacDonald’s story of being home all evening, and if he was, how could he know?) And we still need to accept his great fortune that Stoeckley was so confused or in need of self-incriminating attention that she corroborated a made-up story. But can Morris rule it out?


Morris still has cards to play to establish that Stoeckley was there. (Some of them were already played during the appeals process and in Fatal Justice, where Potter and Bost also argued from MacDonald’s side, but A Wilderness of Error is a far better book. The reporting in Fatal Justice is credulous and tendentious.) On more than one occasion, Helena Stoeckley spoke of remembering bumping into, or trying to ride, a rocking horse with a broken spring, or a broken bolt or wheels, in the MacDonald house. There was a rocking horse in Kristen’s bedroom, though it had runners, not wheels. Stoeckley could have seen a newspaper photo that showed the rocking horse, though she denied it, but no one had publicly indicated it was broken, according to Morris’s research, and it’s impossible to tell from the available pictures, he says. Was it broken? Morris talked to a close friend of MacDonald’s mother, Helen Fell, merely to ask about the MacDonalds’ relationship with Colette’s parents. Out of nowhere Fell said:

And the story of the rocking horse. Nobody but nobody knew that rocking horse was broken. And I ended up knowing about it because, inadvertently, I was talking to Jeff the day that it happened. And he said, “Oh, God, I’m gonna have to listen to [Colette’s mother]. She got the kids a rocking horse. And she got it at probably a yard sale.” … And that was the day that the thing was broken…. And there’s only one way [Stoeckley] knew it — she was there.

Morris shows laudable restraint here: “I wish I could say that this is proof that Stoeckley was there. It is not.” Fell’s reliability might be questioned. It’s the accumulation of evidence that makes a case, Morris writes, and then he continues to pile it up.

The prosecution’s contention that there was no evidence of intruders not only rested on a thoroughly botched crime scene — more than 40 fingerprints were destroyed and gawkers were allowed to wander through the house — but also was, in fact, less than true. Wax drippings were found in the house, including on Colette’s bedding and on the coffee table next to the couch where MacDonald claimed he struggled with assailants and saw the unidentified woman with them. MacDonald had said, “And I just remember that my instinctive thought was that, ‘She’s holding a candle. What the hell is she holding a candle for?’” The wax did not match any of the candles in the MacDonald house. A burnt match was found in Kristen’s bedroom; no items in the room required flame and the MacDonalds did not smoke. Dark wool fibers were found on the club and around Colette’s mouth, and they could not be sourced to anything in the home. Blond synthetic hairs, 22 to 24 inches long and made of saran, were found on a hairbrush in the house, a fact never disclosed to the defense in the 1979 trial. The government explanation was that the fibers could have come from a doll’s hair and that saran was not used in the making of wigs at the time. Morris convincingly defeats both assertions, with the help of a doll expert — no doll of the time had hair that long — and the research of MacDonald’s attorneys. One could throw out every Helena Stoeckley confession and still point to these pieces of evidence to suggest — not prove, but suggest — that she was in the house, that MacDonald’s outlandish story is true.


Have I simply fallen under the spell of Errol Morris’s book? I don’t know the answer. It’s a question that goes to the heart of what makes A Wilderness of Error so fascinating. Like criminal investigation more generally, the book serves as a high-stakes testing ground for the ability to keep absorbing facts and holding off the onrush of opinion, the ability to fend off the psychological weaknesses and creeping biases that beset us all.

Morris does slip up on occasion, for instance bringing more skepticism to bear on evidence that threatens his position and more readily accepting the other kind. When Ken Mica, the military policeman who saw a woman in a floppy hat in the rain, tells Morris that he doesn’t think Stoeckley was the woman, Morris, taken aback, goes to some lengths to show that Mica appears not to have believed that until he decided MacDonald was guilty. Would Morris have been as skeptical if Mica had told him he was now sure it was Stoeckley he saw? Morris’ portrayal of Stoeckley’s reliability also seems generous in light of alternative accounts. At moments, Morris’ flashes of anger or rhetorical flourish serve to undermine him (though it can be enjoyable to watch him shift into High Dudgeon Mode — “This is absurd,” he says of a Janet Malcolm claim). It’s conceivable that Morris is nudging a few things out of sight. He does not discuss fragments of a rubber glove in the master bedroom that some say implicate MacDonald, perhaps having decided, with some reason, that the glove fragments do no such thing. Or perhaps he would rather direct attention elsewhere. But on the whole, Morris inspires trust. He prints many excerpts from transcripts of his interviews, and his interviewing approach — enthralling to observe, as it is in his films — is genial, curious. He does not lead the witness. He seems to want the truth.

WADE SMITH [a defense lawyer]: … it’s been so long. I don’t know whether [Stoeckley] said those things right then. I just don’t know.
ERROL MORRIS: Memory is so weird. Here is something else that bothers me. And if you think it’s completely off base, I’d love being corrected….

If Morris missteps, he does not err nearly so badly as investigators and prosecutors and indeed judges that he scrutinizes. Officials on the government side appear to have suffered a virulent contagion of confirmation bias. Once the central players had decided, too early, that MacDonald was guilty, giving an airing to any evidence in his favor became a counterproductive annoyance. The judge in the 1979 conviction, Franklin Dupree, was openly hostile to the defense, as even McGinniss acknowledges.

On the proving ground of careful reasoning and impartiality, Morris bests Joe McGinniss by a comfortable margin. Just as it is possible that prosecutors railroaded a man who happens to be guilty anyway, it is possible that McGinniss arrived at the correct verdict. But forgive me if I lose a little of my own judicious restraint for a moment. Fatal Vision is a dishonest and unserious book. Put aside McGinniss’ treatment of MacDonald. What about his victimization of the reader? The shoddiness of his book is of a dangerous kind, because McGinniss to some extent succeeds in giving the appearance of fair dealing; the malpractice is not so total as to exclude exculpatory passages. But it’s enough to poison the well.

Here is how McGinniss handles the central and troubling matter of Helena Stoeckley: for one, while he does recount her known admissions at the time of writing, he makes them sound more tentative and hazier than they were. He prefers to emphasize her testimony on the stand, when she devastated the defense by denying any memory of the house and saying she could not recall her prior confessions. In that testimony, she also said, oddly, that she burned her blonde wig two to three days after the crime “because it connected me with the murder,” but you wouldn’t know that from reading Fatal Vision. The judge then excluded any testimony implicating Stoeckley from the six people prepared to give it — a ruling that became the focus of appeals. (McGinniss could not have known that a retired Deputy U.S. Marshal named Jimmy Britt would come forward in 2005, citing a need to lift a “moral burden,” and state in a sworn affidavit that he was present when Stoeckley told a prosecutor immediately before testifying that she had been inside the MacDonald house during the crime. The prosecutor, James Blackburn, responded that if she testified to that effect, he would indict her for murder, the affidavit said. Blackburn denies it. Morris can’t resist going for the jugular and pointing out that Blackburn was later disbarred and imprisoned for forgery, fraud, and embezzlement.)

On various occasions, Stoeckley spoke about trying to quiet a German Shepherd near the MacDonald house, about lighting a candle and carrying it around inside, about an upside-down book on MacDonald’s chest while he slept with his glasses on the floor next to him, and about the rocking horse. When Stoeckley told police that “it did not begin to rain hard until after the homicide,” they asked how she knew. She replied, “I have already said too much.” Perhaps this was all a lot of nothing, scraps she had cobbled together over time, and on some occasions she added new bits that didn’t jibe, particularly under Gunderson’s questioning. But the inconsequential details like the upside-down book and the glasses — I find them eerily plausible.

In Fatal Vision, McGinniss writes Stoeckley off with flimsy justifications. He tells us that according to “her acquaintances,” “Helena Stoeckley had always liked to make up stories.” Seriously, that’s on page 233. Morris demonstrates that McGinniss mischaracterizes one of Stoeckley’s polygraphed confessions by citing the examiner’s judgment that “a conclusion could not be reached as to whether or not she knew who committed the homicides or whether she was present at the scene.” Here McGinniss opts not to mention that the examiner also wrote, “it is concluded that Miss STOECKLEY is convinced in her mind that she knows the identity of those person(s) who killed” the MacDonalds and “convinced in her mind that she was physically present when the three members of the MacDonald family were killed.” Speaking to Morris, the polygraph examiner clarified that owing to Stoeckley’s drug use and questions about her mental state, he felt it was possible that she was mistaken in her belief that she was involved. This was compatible with his other conclusion — that Stoeckley was convinced she was there.

A maestro of speculative insinuation, McGinniss manages to make incriminating molehills look mountainous, particularly by undermining the prevailing view of MacDonald’s character. McGinniss spends an awful lot of pages attempting to establish that MacDonald is not such a spotless guy. McGinniss tells us gravely that MacDonald had been wrong about what year he had sex with Colette back in college in her parents’ apartment, as if he has caught MacDonald in a terrible lie. McGinniss does show that MacDonald can be callous and petulant, a bit of a fancy-pants and a womanizer. Lulled by the book, we begin to dislike MacDonald. Then we remember that we have met a number of callous womanizers, and not one of them has murdered his entire family.

Building to the climax of Fatal Vision, McGinniss cites Christopher Lasch’s The Culture of Narcissism, a work of social theory and criticism, as if it were some sort of scientific text on psychopathology that damns MacDonald with its diagnostic power. McGinniss finally lays out his theory that diet pills sent MacDonald into a homicidal rage that unleashed his psychopathic fantasy of misogynistic violence. Morris memorably remarks that “even though it is a bluff, [McGinniss] plunks it down with ultimate bravura — as if he is the proud owner of a royal flush.”

When the jury found MacDonald guilty, he and McGinniss were in frequent contact. McGinniss wrote MacDonald about the verdict, “What the fuck were those people thinking of? How could 12 people not only agree to believe such a horrendous proposition, but agree, with a man’s life at stake, that they believed it beyond a reasonable doubt?”

Was McGinniss deeply dishonest or deeply confused when he wrote that letter? Whether MacDonald is innocent or not, you have to be troubled by McGinniss’ vast influence on the public perception of the case. Gene Weingarten again: “It is impossible to read ‘Fatal Vision’ and not KNOW that he killed his wife and child. Impossible.”


Morris does not spend much time on Janet Malcolm, but when he finally discusses her book, a dramatic head-to-head ensues. The Journalist and the Murderer centers on the lawsuit MacDonald brought against McGinniss, not the crime itself; Malcolm does not explicitly take a position on culpability, though it is implicit in the title and Malcolm does repeatedly refer to MacDonald as a murderer. Malcolm later blurbed the pro-innocence Fatal Justice (1995) by calling it “a quietly convincing book,” indicating a possible reversal on her part. When Morris called her, she declined to state her view. (And she declined to speak to David Carr for an article published in The Times last week, saying she had not read A Wilderness of Error.) In her book, she explained her reasons for allowing McGinniss’ conclusions to stand despite MacDonald’s pleas to look into the matter further:

MacDonald does nothing by halves, and, just as McGinniss had felt oppressed by the quantity of extraneous details in MacDonald’s tapes, so was I oppressed by the mountain of documents that formed in my office. I have read little of the material he has sent — trial transcripts, motions, declarations, affidavits, reports.… I know I cannot learn anything about MacDonald’s guilt or innocence from this material. It is like looking for proof or disproof of the existence of God in a flower — it all depends on how you read the evidence. If you start out with a presumption of his guilt, you read the documents one way, and another way if you presume his innocence.

Morris responds: “I find this passage to be disturbing and problematic. Also wrong.” Malcolm does seems to adopt a bizarre stance here — that it is pointless to learn the facts to try to get to the bottom of a crime. Given the context of Malcolm’s wider argument in the book, the passage could be read more generously as being in a sense confessional: You see, we journalists can’t be trusted. We can choose a storyline that suits our predilections. True as that may be, aren’t some storylines more well-founded than others? MacDonald offers evidence that is not yet public, but, Malcolm writes, “I do not want it…. The story of the murders has been told — by Joe McGinniss — and it has acquired the aura of a definitive narrative.” Alas, that is just so.

Morris’ book isn’t the definitive narrative either, because no such thing exists, but it shows just how many troublesome facts must be swept under the rug to accept McGinniss’ account — just how many improbabilities we need to accept to be confident that Jeffrey MacDonald should be living his life in prison. In an interview nine years after a North Carolina jury convicted MacDonald of triple homicide, one of the jurors spoke of his lingering concern about the matter of Helena Stoeckley: “We should have been told more about that woman.”

Evan Hughes is the author of Literary Brooklyn. He’s on Twitter.