A number of years ago, I served on a criminal jury. The defendant, who was African American, was charged with assaulting a police officer, who was white. As I recall, all of the prosecution witnesses were police officers, and all were white. All of the defense witnesses were African American, and they corroborated the defendant's story, which was that no assault had occurred.
The jury consisted mostly of African Americans. If you asked us whether we believed the defendant had assaulted the officer, I am fairly confident that a strong majority (and perhaps all of us) would have said yes. Although all of the witnesses seemed quite credible, the officers were a bit more so. Moreover, the defense witnesses included the defendant's friends and family, who could be expected to have a bias in his favor.
Nonetheless, it was clear that we couldn't convict him. Throughout the trial and during our deliberations, race was a significant (but unmentioned) issue - the elephant in the room. But that particular elephant had nothing to do with our refusal to return a guilty verdict.
What drove our decision was the reasonable-doubt standard. We might well have been able to agree that the prosecution had proved its case by a preponderance of the evidence. But we couldn't say that it did so beyond a reasonable doubt.
The verdict in the George Zimmerman-Trayvon Martin case can be understood in many ways, but it reflects, at least in part, the immense power of the reasonable-doubt standard. That standard ensures that juries will have to return a verdict of not guilty even if they really believe that the defendant is guilty.
Reasonable doubt is far more difficult to meet than other legal standards, including "preponderance of the evidence" (used for most civil trials), "clear and convincing evidence" (used for deportation proceedings) and "substantial evidence" (used for administrative agency decisions).
To be sure, any doubt must be "reasonable"; the law doesn't require absolute certainty. But a good defense lawyer is often able to obtain an acquittal even if most jurors essentially agree with the prosecution's account of the facts.
Lord Blackstone famously said, "It is better that 10 guilty persons escape than that one innocent suffer," and most people think the reasonable-doubt standard reflects a judgment to that effect.
But that judgment isn't self-evidently correct. If 10 guilty people escape punishment, then the deterrent effect of the criminal law will be significantly weakened, and wrongdoers will be set free to do more wrong, potentially putting innocent lives in jeopardy. The reaction to the Zimmerman verdict also raises a question: If the reasonable-doubt standard were put to a vote, would a majority support it?
As a matter of history, this standard wasn't centrally focused on protection of the innocent. As Yale Law School professor James Whitman has shown, the standard was rooted directly in Christian theology. The goal was to protect the souls of jurors who might otherwise be at risk of damnation.
Premodern Christian theology regarded conviction of the innocent as a potentially mortal sin. In these circumstances, the reasonable- doubt standard would serve, in Whitman's words, "to ease the fears of those jurors who might otherwise refuse to pronounce the defendant guilty."
The standard began to appear in England in the late 18th century, with no evident single author, as a way "to leave the soul of the Christian judge safe." In Whitman's telling, jurors felt "a general dread lest the charge of innocent blood should lie at their doors." A goal was to ensure not merely factual proof but also moral comfort. Paradoxically, reasonable doubt was meant in part to increase the likelihood of convictions by giving moral reassurance to juries that were willing to deliver a guilty verdict.
Whitman emphasizes the dramatic differences between the modern era and the historical period that gave rise to the reasonable-doubt standard. He writes, "The reasonable doubt rule arose in the face of religious fears. It is still with us today, a living fossil from an older moral world."
But there is a lot of continuity, too; it is no mere fossil. Whether or not they worry about eternal damnation, many jurors have strong religious or moral qualms about convicting a potentially innocent person (especially when that person is sitting in front of them). And whether or not they would have chosen the reasonable-doubt standard before they were empaneled as jurors, it reflects what a lot of them start to think once they are placed in that role.
To juries, moral comfort matters. Whether the verdict in the Zimmerman case was right or wrong, that is one of its central lessons.
Cass R. Sunstein, a professor at Harvard Law School, is a Bloomberg View columnist.