Proportioning verdicts and punishments to evidence

[Music] Hi, I’m Barry Lam, Associate Professor of Philosophy at Vassar College, and producer of Hi-Phi Nation, a show about philosophy that turns stories into ideas. In this video, we’ll ask whether jurors in criminal cases should be able to choose from a range of verdicts, not just between Guilty and Not Guilty. George was walking home one night when someone attacked him and stole his wallet. George didn’t get a good look at the person. But police found footage of the incident, and facial recognition software determined a 60% match between the assailant’s grainy image and James Borden, a man with a record of robberies. At Borden’s home, the police found a black hoodie and an empty brown wallet. Borden had no alibi. He claimed to be home alone all night and that the wallet was his. Based on the evidence, the prosecutor figured there was an 85% chance that Borden was the perpetrator. 85% might sound pretty high, but if this case went to trial, Borden could just as easily be convicted as acquitted. That’s because the burden of proof in an American criminal trial is very high: it’s proof beyond a reasonable doubt. And a reasonable person can doubt Borden’s guilt. True, the wallet looked like George’s, but brown wallets are very common. A 60% facial recognition match is good, but that means there’s a 40% chance the image isn’t Borden’s. And black hoodies are extremely popular. The 85% chance Borden is guilty still leaves a 15% chance he’s innocent. In the U.S. criminal justice system, trial verdicts are all or nothing. The defendant is either unanimously declared Guilty beyond a reasonable doubt and punished, or declared Not Guilty and goes free. But why should the legal system force a Guilty or Not Guilty verdict when there’s significant uncertainty about whether the defendant committed the crime? Why shouldn’t judges and juries be able to issue verdicts that better reflect their true level of confidence in the defendant’s guilt? Scotland does something close to this. Scottish jurors choose from three verdicts: Guilty, Not Guilty, and Not Proven -- where Not Proven means that, given the evidence presented, you have reasonable doubt about the defendant’s guilt, but also reasonable doubt about their innocence. And why stop at three options? Some philosophers have proposed four: Definitely Guilty, Probably Guilty, Probably Innocent, and Definitely Innocent. You could be even more precise and issue verdicts of guilt by percentage. There are good arguments for giving jurors a wider range of verdict options. It allows justice to be carried out more honestly. It also better reflects how we make decisions under uncertainty in our everyday lives, like when we choose a career, buy insurance, or navigate relationships. We treat friends we know did something harmful to us differently from friends who probably did though we can’t prove it. And we treat both of these kinds of friends very differently from those we know to be innocent. Treating someone who probably robbed you the same as someone who definitely didn’t seems unwise and morally wrong. Yet it’s what our current legal system asks jurors to do. A defendant whose guilt is likely but not beyond a reasonable doubt is ruled Not Guilty -- the same verdict given to a defendant who’s conclusively innocent. Both defendants are set free, but both are also stuck with the stigma of an arrest record. This might be appropriate treatment for someone who’s probably though not provably guilty, but it seems unfair to forever stigmatize someone who’s provably innocent. Instead of treating every defendant as belonging to one of two categories, proportionate verdicts could lead to incarceration only for the definitely guilty, mere supervision for the probably guilty, a full expunging of the records of the probably innocent, and official apologies and restitution for the definitely innocent. Or the system could get even more precise: if punishment for assault and robbery is 100 days in prison, then someone like Borden, who’s 85% likely to be the perpetrator, might receive 85 days in prison, rather than either none or the full hundred. But while such a justice system has advantages, the objections to it are many. Imagine Lyla and Maria are the only people present when Paul is murdered. Evidence proves there was only one murderer but not who did it, so there’s a 50% chance each woman is the killer. Our current system would set both women free, since there’s reasonable doubt of guilt for each. This is unjust: the true murderer wouldn’t receive any of the punishment she deserves. But what would happen in a proportionate system? Here's one way it could go: If the full sentence for murder is 40 years, Lyla and Maria would be punished proportionately, each incarcerated for 20 years. Although the actual killer still escapes full punishment, at least she’s sure to get some of the punishment she deserves. Meanwhile, though, an innocent person is sure to spend 20 years behind bars! To most people, incarcerating an innocent person is significantly more unjust than letting a guilty person go free. The injustices of the alternative system seem far worse than the injustices of the current one. Here’s another objection. In our current system, the high standard of proof means that if you don’t commit a crime, it’s generally unlikely you’ll end up in prison. So the easiest way to avoid criminal conviction is to follow the law -- exactly what we want people to do. In the alternative system, though, following the law wouldn’t be enough. You’d also need to curb all activities that can count as circumstantial evidence of guilt -- like wearing clothes associated with criminal offenders or spending time alone with no alibi. There’s no societal benefit to preventing these activities, but in a system where any evidence can lead to some time in prison, people have strong incentives to avoid them. Ultimately, there seems to be a strong argument against systems that proportion verdicts and punishments to evidence. But here’s the catch: In America, this actually IS our system. Only 5% of criminal cases go to trial. Instead, most cases are settled through a plea bargain. For the Borden case, that means that the likely outcome would be a plea bargain negotiated by the parties, where Borden gets about 85% of the full sentence, reflecting the 85% certainty of his guilt. And this would happen regardless of whether Borden is actually guilty. So if proportionate punishment is unjust, it seems we must reject, not just some hypothetical system, but the one we live in today. What do you think? [Music]