During Kevin Steele’s successful election campaign for District Attorney
of Montgomery County, Pennsylvania, in 2015, he attacked the longtime
incumbent, Bruce Castor, for having “refused to prosecute Bill Cosby”
and promised “tough sentences for sexual predators.” After taking
office, District Attorney Steele immediately moved on his promise to
vindicate Bill Cosby’s victims, arresting and charging Cosby for the
sexual assault of Andrea Constand, one of nearly sixty women to have
accused Cosby of sexual assault over several decades. But Cosby’s
criminal trial, on three counts of indecent assault for the 2004 incident,
ended in a mistrial due to a hung jury, after six days of deliberations
produced neither conviction nor acquittal.

The sheer number of Cosby accusers who have come forward, and the
consistency of their descriptions of his modus operandi, are so
overwhelming that they produce little doubt that Cosby used his fame and
power to lure women, give them incapacitating drugs, and have sex with
them without their consent. When one views Cosby and Constand as
stand-ins in a narrative of rapists acting with impunity against
powerless victims, it is tempting to consider the failure to convict
Cosby as one of the highest-profile examples yet of assaulted women
being disbelieved and devalued.

But the legal meaning of the Cosby mistrial is both less and more than
that. Standards that enable criminal conviction and punishment for a
specific act differ markedly from the ones that lead to personal beliefs
that someone must be guilty of wrongdoing. In the midst of fifty-two
hours of deliberation, the jury requested clarification of what proof
“beyond a reasonable doubt” meant. The judge explained that a reasonable
doubt is a real doubt that causes a juror to hesitate. In other words,
if a juror were mostly or nearly convinced that a defendant was guilty,
but still had some doubts that seem reasonable, the appropriate vote
would be for acquittal. Apparently, some of the jurors did have such
doubts, while others did not.

The extraordinarily high prosecutorial burden of proof in any criminal
trial is intentionally designed to heavily favor defendants, because we
long ago embraced as a society Blackstone’s principle. Formulated in the
seventeen-sixties by the English jurist William Blackstone, the
presumption is that it is better to have ten guilty people go free than
that one innocent person suffer. Hard as it is to stomach today,
embracing that calculus means that we should even want ten rapists (not
to mention terrorists and murderers) to go free in order to protect the
one falsely accused. Unfortunately, Cosby is one of those to escape
criminal punishment. And, to put a fine point on the over-all gendered
impact of requiring proof “beyond a reasonable doubt,” the inevitable
effect of the heavy tilt toward defendants is that in sexual-assault
trials, which involve mostly male defendants and mostly female accusers,
men are favored over women.

This structural bias in favor of criminal defendants does not create
anything like an even playing field between accused and accuser—and it
is not supposed to do so. But when the testimony of a female complaining
witness is the centerpiece of a sexual-assault trial, as was true of the
calm testimony of Constand at Cosby’s trial, we fear a repeat of our
sexist legal history of putting the victim on trial, of pointing to her
sexual past or reputation, to insinuate her dishonesty. Rape-shield laws
adopted from the nineteen-seventies through the nineteen-nineties now
preclude defense lawyers from doing exactly that. But, because it is
essential to a fair trial that criminal defendants have the chance to
try to show that the stories of witnesses against them are untruthful or
inaccurate, the credibility of the accuser is inevitably still on trial.

It was the job of Cosby’s defense to attack Constand’s credibility, to
try to sow reasonable doubt in the minds of the jurors about whether her
account should be accepted as true. That is an uneasy fact for those who
think sexual-assault complainants should always be believed. But raising
doubts about a witness’s credibility is basic to any legal trial
involving testimony under our legal system. As expected, Cosby’s defense
lawyers pointed to inconsistencies and inaccuracies in Constand’s past
statements about the incident. The case most crucially turned on whether
there was consent for the sexual acts in question. Constand said Cosby
gave her pills that immobilized her and made her semi-conscious, and
then touched and penetrated her with his fingers without consent. The
defense said the two had a romantic relationship, that the pills were
merely Benadryl, and that the sexual acts were consensual.

But the jurors who voted to acquit did not actually need to conclude
that Constand was a scheming liar. They could have thought that Constand
was being truthful as to what happened that night, but that some of the
other evidence produced reasonable doubt regarding lack of consent.
Evidence that she called him at least fifty-three times after the night
in question, and that Cosby offered to pay for Constand’s schooling
afterward (intended by the prosecution to show that Cosby felt remorse
for assaulting her), may have suggested that each party wanted something
from the other. Perhaps he wanted sex and she wanted his help with her
career. In that context, Constand agreeing to take the pills he gave her
at his home may have created a cloud of ambiguity for some jurors
regarding her consent.

In a deposition in Constand’s 2005 civil suit against Cosby, which was
read into evidence at his criminal trial, Cosby explicitly acknowledged
that he kept Quaaludes to give to women to have sex with them. (Cosby
paid her an undisclosed settlement amount, in 2006.) While this appears
akin to a “smoking gun,” Cosby’s admission is not literally inconsistent
with his giving the pills as party drugs to women to ingest voluntarily
for the purpose of sexual arousal. Perhaps some jurors thought that
there was a small but real chance that Constand accepted the pills as
part of a consensual sexual encounter on this particular night. That is
all it takes for reasonable doubt—and just one juror’s reasonable doubt
can hang a jury.

There is also a natural reluctance on the part of many people to
confidently rely on accounts of an incident that occurred thirteen years
ago. Personal experience with how the mind over time tends to distort or
alter memories could have led some jurors to believe Constand but still
hesitate to incarcerate Cosby. Jurors need not have thought Constand was
a liar to have been swayed by the passage of time toward reasonable
doubt.

Many Cosby accusers will never have their day in criminal court, as
Constand did, because the statutes of limitations for the crimes have
expired. So the denial of vindication here can feel like defeat. But
criminal charges are not the only legal path, and many of Cosby’s
accusers have pending civil cases against him. Some have also filed
civil slander cases against Cosby for saying that they were lying about
their assaults. The burden of “beyond a reasonable doubt” does not apply
in a civil trial, where a jury need only conclude that it is more likely
than not that Cosby assaulted the plaintiff.

Steele was elected as District Attorney on a platform that focussed on
Bill Cosby and on empowering victims. The jury is out on whether his
ambitious pursuit of a criminal trial here helped sexual-assault victims
as much as his own profile and career. But it would be wrong to take
away from the hung jury the message that sexual-assault victims are
widely and simply disbelieved. The jury result instead reflects exactly
the conflict, and ambivalence, we have between honorable legal
principles that for good reason structurally bias criminal trials
against conviction, and our cherished ideal that sexual-assault victims
should be believed and vindicated.