How Duke Law Researchers Pried Open the ‘Black Box’ of Plea Bargaining

The vast majority of criminal cases in the United States are resolved without a trial―a fact summed up starkly by the U.S. Supreme Court during a 2012 ruling.

“Criminal justice today is for the most part a system of pleas, not a system of trials,” the Court said in Missouri v Frye.

Earlier commentators used even more pungent language. “Horse trading determines who goes to jail and for how long,” wrote the authors of a textbook on plea bargaining published in 1909.

“It is not some adjunct to the criminal justice system; it is the criminal justice system.”

A paper published by the Duke University School of Law, which cited both commentaries, said the process was a distortion of justice because most—if not all—of the deals were concluded in secret, hidden behind what many have called the “black box” of prosecutorial discretion.

Advocates and analysts don’t know if there are “infections” among plea negotiations, such as discrimination based on wealth, sex, age, race, pre-trial detention, education, intelligence or ability of counsel, the paper asserted.

There is also anecdotal evidence linking wrongful convictions to the use of plea bargaining, where an innocent defendant is persuaded to admit guilt in order to avoid the risk of a long prison sentence if he goes to trail and loses the case.

In today’s clogged court dockets, the emphasis is on speed and efficiency, critics of the system point out. That has generated what some call a “trial penalty,” in which the threat of a longer sentence is held over a defendant who wants to plead not guilty.

“Prosecutors may resolve cases for reasons that are benign, thoughtful, and well-calibrated; or discriminatory, self-interested, and arbitrary, (but) with very little oversight or sunlight,” “writes lead author Brandon L. Garrett, the L. Neil Williams, Jr. Professor of Law at Duke University, and Director of Duke’s Wilson Center for Science and Justice.

The authors of the study said they had found a way to pry open the “black box” by using data.

Working with prosecutors in Durham, NC and Berkshire, MA, they created what they called a “plea tracker dashboard” that has added transparency to court proceedings.

The first-of its-kind system is now successfully operating in both states, and may soon spread to a third, the authors said.

The North Carolina and Massachusetts Experiments

For years, academics and policymakers have called for meaningful plea-bargaining data to fill the information void. The authors of the study say their data dashboard can make a big difference.

“This is not a hypothetical or conceptual exercise, but rather the product of theory, design, and implementation work by an interdisciplinary team,” the authors write.

The authors detail how they designed and implemented a data collection system that was piloted in prosecution offices in Durham, NC, and Berkshire, MA.

Durham was selected because the city had no rigorous system for collecting data on plea outcomes, so everyone was flying blind. The first change the researchers implemented was requiring that all plea offers be drafted and discussions transcribed to ensure a record of events.

The documentation of transcripts and data collection proved helpful. In 442 of the 559 felony cases reviewed in the study, there was just one plea before the offer was accepted. There were 96 cases that had two pleas, 15 had three pleas, and just one with four, the researchers found.

In Berkshire, the elected district attorney announced that the data tracking systems would serve as a metric for their successes in developing more open court procedures.

 Six Months of Tracking

They tracked data for six months, beginning on April 1, 2021, and found that overall, most communication with someone’s defense counsel was done through email, with “phone, Zoom, in-person” being the most common. Most people also shared that they never had a formal mode of communication with defense counsel.

“Our Plea Tracker asks prosecutors to record how much they communicated with the defense and the mode of communication, and what charging and sentencing options they prioritized or felt were most important when they were negotiating a case,” the authors wrote.

“They are also asked to disclose the mitigating, aggravating, and other case factors they considered when creating the plea agreement or negotiating the plea.”

Within those post-meeting assessments with defense attorneys, the most important aspect of plea agreements that came up in Berkshire were rehabilitation and treatment (23.5 percent) and sentence length (21.1 percent).

The researchers also found that the least important aspect of plea bargains were changing the charges, (61.2 percent said “not important) and addressing fines and fees (56.7 percent).

Prosecutors using the Plea Tracker were then asked to provide their opinion regarding the importance of various aspects and topics of the plea negotiation process, on a Likert scale from one to six. This sliding scale then correlates to jail and prison time, or probation options.

At the end of their study, prosecutors told researchers that they began to see value in collecting data as a means to inform themselves and improve their work, proving that “plea tracking is not complex” and the system can widely benefit from this approach.

“An open prosecution approach is feasible, and, for the first time in the United States, it is in operation,” the authors wrote.

In addition to Prof. Garrett, the other authors of the study were:

William Crozier, the Research Director for the Wilson Center for Science and Justice; Elizabeth J. Gifford of the Sanford School of Public Policy at Duke University; Catherine Grodensky, a Ph.D. Candidate in the Sanford School of Public Policy at Duke University; and Adele Quigley-McBride and Jennifer Teitcher, both post-doctoral fellows at the Wilson Center for Science and Justice.

The full paper can be accessed here. 

Additional Reading: Does the Justice System’s ‘Black Box Secrecy’ Violate the Constitution?

Andrea Cipriano is Associate Editor of The Crime Report