After examining the case history, Dr. Steinschneider concluded that S.I.D.S. could be genetic, which would explain the deaths of multiple infants in the same family. Waneta gave birth to two more children, Molly and Noah, both of whom died while under Steinschneider’s care. In 1972, Steinschneider published a landmark paper in the journal Pediatrics, in which he argued that S.I.D.S. could be hereditary, and was related to sleep apnea. The paper was a great success, and sales of sleep monitors—which Steinschneider recommended—took off. For twenty years, the Hoyts, who were described in the paper simply as “the ‘H’ family,” served as a sad exemplar in medical literature.

However, in the early nineties, a district attorney in upstate New York, who had been tipped off by a suspicious forensic pathologist, took a closer look at Steinschneider’s paper and opened an investigation. As Richard Firstman and Jamie Talan recount in their book, “The Death of Innocents,” under questioning by police in 1994, Waneta Hoyt confessed that her children had died not from S.I.D.S. but, rather, because she had smothered them. She later recanted this confession. But, in 1995, she was convicted of killing all five children and sentenced to seventy-five years in prison.

On the surface, this case might seem unrelated to the story of Amy Bishop, which I recount in a piece in the current issue of the magazine (“A Loaded Gun”). One theme of the Bishop saga is the dangerously formidable power of parental love, which seems a far cry from Waneta Hoyt, who dispatched her own children, one by one. But the two cases share a common, troubling thread: neighbors, police officers, and even the medical establishment may be more likely to overlook glaringly suspicious behavior when the perpetrator of that behavior is a woman.

In the aftermath of the massacre at Sandy Hook Elementary School, last December, there has been much public discussion about the necessity of greater vigilance regarding mental-health issues—about our ability to recognize red flags early and get potentially dangerous individuals into treatment. It’s a reassuring notion, and less divisive, certainly, than calls for greater gun control or for censoring video games. But, as the Bishop story makes clear, this kind of early-warning system is often difficult to institute in practice. Amy Bishop shot her own brother, after all. She punched a woman at a pancake restaurant. She stood accused of mailing a bomb to one of her supervisors at Harvard. Red flags don’t get much brighter than that. Yet, nobody stepped in. Why not?

One answer, which I explore in the piece, is the role that collective denial can play, not just in families but in communities and local institutions, like police departments. But another answer may lie in Bishop’s gender. In a 1998 article about gender bias in the criminal-justice system, Larissa MacFarquhar, reviewing two books about female killers, observes, “The message that emerges from this collection of tales is clear enough: be a woman, act like a woman, and you may blithely strew your neighborhood with bloody axes and severed heads, since the chances are that you will get away with murder.”

In her article, MacFarquhar relates gruesome tales of so-called Black Widows (women who murder their husbands or lovers) and Angels of Death (women who kill those placed in their professional care). These women often skirt suspicion and kill prolifically because it doesn’t occur to the cops until much, much too late that a female could be capable of such a thing. In retrospect, the crimes seem gallingly obvious: Genene Jones, an Angel of Death who worked as a pediatric nurse in Texas, is believed to have murdered as many as forty-six children before authorities caught up with her, in 1983. But as the murders are actually being committed, investigators prove far too ready to attribute the mounting body count to accidents, medical error, or a male perpetrator.

The detection of crime, as MacFarquhar notes, is “one of the most stubborn redoubts of male chauvinism.” Women have fought to undo the patriarchal notions of gentle femininity that in the past have excluded them from suffrage, employment, and combat roles in the military. But in the criminal-justice system, they may still be construed as lacking in the moral and physical agency that is necessary to carry out a violent crime.

Consider the case of Karla Homolka, a young Ontario woman. In 1991, she married an accountant, Paul Bernardo. They were an attractive, apparently happy couple, but, in 1993, Canadian authorities began to suspect Bernardo in a string of unsolved rapes. Investigators also linked Bernardo to the recent rape and murder of two local schoolgirls. When Homolka was questioned by police, she informed them that her husband was indeed responsible for the crimes, and that he had beaten her savagely and forced her to become an accomplice. Homolka made a sympathetic witness—a middle-class white woman who worked at a veterinary clinic and had been battered by her domineering spouse. She had also suffered a recent tragedy: in 1991, her fifteen-year old sister, Tammy, had died after she passed out from drinking too much champagne at a Christmas party and choked on her own vomit.

Because Homolka had known about her husband’s crimes and played some role in them, prosecutors could not offer her full immunity. But they struck a plea bargain in which she agreed to testify against Bernardo in exchange for a light sentence. In assessing her culpability, Canadian investigators were reportedly influenced by a paper, written by Roy Hazelwood, an F.B.I. profiler, called “Compliant Victims of the Sexual Sadist.” It was only after Homolka’s deal was in place that she provided a fuller confession, explaining that she had not merely been a witness to Bernardo’s crimes but an active participant as well. Home videos that showed Homolka’s role in the sexual assaults to be anything but forced surfaced not long afterward. And it ultimately emerged that Tammy’s death was not an accident: Homolka had drugged her sister with an animal tranquilizer that she had stolen from work, then she and Bernardo had raped her.

To be sure, women are less apt to commit crimes—especially violent crimes—than men are; that has long been an accepted criminological fact. In recent decades, the “gender gap” in crime has narrowed somewhat, however, and there are a number of theories that might account for that change. Some believe that women are committing more crimes than they have in the past, but others argue that women are simply being investigated and arrested more often, as cops gradually come to terms with the concept of a female criminal.

Still, when you look at penalty statistics, it seems undeniable that our justice system remains warped by some measure of gender bias. In 1987, the year the Massachusetts State Police ruled that the death of Seth Bishop was an accident, twenty-two per cent of people arrested for “serious crimes” in the United States were women. But women were only ten per cent of those convicted of such crimes—and only five per cent of those imprisoned. A recent law-review article described the persistence of certain “chivalric norms,” which might explain the disproportionately low number of women who receive the death penalty in this country. (Only twelve women have been executed in the United States since 1976, as opposed to thirteen hundred and eight men.)

If a bias in favor of women is statistically obvious in the prosecution of crimes, there is likely a similar bias in the investigation of wrongdoing. This claim is harder to substantiate, of course—there is no way to accurately quantify the number of potential crimes and suspects that law enforcement opts not to pursue. But one useful counterfactual exercise would be to consider the events surrounding the death of Seth Bishop, on December 6, 1986, and wonder how things might have played out had the perpetrator been his brother rather than his sister. Would the police have accepted the notion that the shooting was an accident due to incompetence with a gun? If a young man had fled the scene of a shooting, brandished a shotgun in an effort to commandeer a vehicle, then ended up in an armed standoff with two police officers, would the cops have been so quick to release him without charges? Would officials have believed the explanation, which the Bishop parents still maintain, that this behavior was simply the product of “shock”? MacFarquhar points out that one feminist critique of gender bias in the treatment of violence is that it can “infantilize” women: “In supposing that a woman can’t be truly responsible for her violent acts, we are treating her as if she was a child.”

Of course, Amy Bishop was not just a woman—she was also an academic, and a related question is whether warning signs were overlooked because some degree of eccentricity is accepted, even nurtured, in a university setting. “Academe is often home to oddballs,” an article about the case in the Chronicle of Higher Education conceded. “Choosing to spend your life in a library or a laboratory is, by definition, out of the ordinary.” There is a fine line between eccentricity and instability—a line that is sometimes indiscernible. And in our rush to pre?mpt future mass shootings, we don’t want to start involuntarily committing any university professor who is given to erratic behavior.

But according to a lawsuit filed by the families of two of Bishop’s victims, Maria Ragland Davis and Adriel Johnson, Vistasp Karbhari, the provost of the University of Alabama at Huntsville, was well aware of Bishop’s “dangerous instability,” yet did nothing to intervene. “Bishop had a history of severe, observable mental instability and violence dating back more than 20 years,” the suit alleges. According to court filings, the university had procedures in place to report potentially dangerous people on campus to the police and to get them into therapy—procedures of a sort that were instituted at universities across the country in the aftermath of shooting incidents at other educational institutions, like Virginia Tech.

In something of a bombshell development, the Huntsville Times reported earlier this month that Amy Bishop herself wrote an affidavit from prison in which she described watching Karbhari flee the administration building along with David Williams, who was president of the university at the time, and an armed escort—because they believed that she was coming to see them. This is a damning image: if Karbhari was so concerned about his own safety, how could he not warn Bishop’s students or her colleagues? But a university spokesperson has dismissed the claim, pointing out that in the same affidavit Bishop mentions that she “sees dead people.” (Bishop told me that she has heard “voices” and suffered from “hallucinations” since shortly after her brother’s death.)

Bishop is also not exactly an objective source on the subject of Karbhari: her parents expressed confusion to me about her decision to shoot her fellow-professors, some of whom were friends of hers, when it was Karbhari whom she perceived as her greatest antagonist. I’m not suggesting that Bishop would lie in her affidavit simply to assist in a lawsuit against the provost, but there do seem to be grounds for assessing her story with a skeptical eye.

Another person named in the lawsuit is Amy Bishop’s husband, Jim Anderson. If anyone was in a position to spot red flags, it would presumably be her spouse, with whom she often collaborated in her research, and who drove her to campus on the day that she shot her colleagues. According to the lawsuit, Anderson “had direct knowledge of his wife’s impairment, her history of violence, her anger and threats toward colleagues over the denial of tenure, and her acquisition and use of a firearm shortly before the massacre.”

“He, too, knew of her dangerous propensities,” Allen Brinkley, an attorney representing the Davis and Johnson families, said in 2011. “He more than anybody.”

Anderson’s role is one of the most unsettling features of the whole affair, because either he grasped that his wife was dangerous and did nothing or he, too, was oblivious to the warning signs. While it might be tempting, even comforting, to conclude that the family members of a killer would be the first to register tremors of danger, it may actually be the case that those closest to an unstable person have too much invested, emotionally, to take a dispassionate view. Perhaps the closer you are to someone the more blinded you become to what they are capable of. Even after Waneta Hoyt confessed to smothering all five of his children, her husband, Tim, continued to assert that she was innocent. He insisted that Waneta’s confession was coerced even as she was tried, convicted, and sent to prison, and close friends and neighbors of the family agreed—as did Waneta and Tim’s one surviving child, who was adopted.

This raises a chilling question: If the killer’s own family and neighbors miss the red flags, what hope do the rest of us have?