After 28 years in prison for crimes that DNA evidence says he didn’t commit, Leroy Harris is a free man. His freedom is, however, not because of that evidence but because he took a plea deal in order to be with his family.
Back in 1983, the owner of a New Haven, Connecticut, nightclub was robbed at gunpoint. His car was stolen and then later used in the robbery and sexual assault of two women. According to the Innocence Project, a national group that works to exonerate the wrongfully convicted and which took on Harris’ case, Harris was identified as a suspect for the robberies by two men —Charles Myers and Jerome Downing. During the police investigation, the female assault victims twice failed to identify Harris when shown his picture.
Despite a lack of any physical evidence against Harris, or any reliable eyewitness testimony of his involvement, he was arrested and brought to trial in 1989. Only at the trial did the two female victims, plus the nightclub owner and his girlfriend, then positively identify Harris. Charles Myers and Jerome Downing would later admit their involvement and plead guilty to the crime. Downing recanted his earlier statements, saying Harris had not been involved and he had only implicated him in order to get a better plea deal.
Nevertheless, in 1989, Harris was convicted on three counts of first-degree robbery and one count of first-degree sexual assault. He was sentenced to 80 years in prison.
When the Innocence Project got involved with Harris’ case in 2012, they discovered new DNA evidence exonerating Harris and showed that prosecutor James Clark had not only failed to turn over exculpatory evidence during Harris’ trial but had also solicited false and misleading testimony.
In 2016, the Connecticut Supreme Court ruled that identifications such as the ones in Harris’ trial violated due process.
“Thus, we conclude for the first time today that any first time in-court identification by a witness who would have been unable to reliably identify the defendant in a nonsuggestive out-of-court procedure constitutes a procedural due process violation.” the court wrote in State v. Dickson.
Despite the exculpatory evidence and the state supreme court’s ruling, Harris’ conviction was not overturned.
On November 21, 2017, Harris accepted an Alford plea, a kind of plea bargain where the defendant concedes the state has enough evidence to convict them but does not admit guilt. New Haven State’s Attorney Patrick Griffin agreed to vacate Harris’ previous convictions if Harris pleaded guilty to charges of accessory to first-degree kidnapping and three counts of first-degree robbery.
According to the Innocence Project, neither Harris nor his lawyers were satisfied with the outcome, but accepted it because “it enabled him to return home to his family and maintain his innocence despite his “guilty” plea.”
“Although I agonized over accepting the plea, I couldn’t spend another day away from the people I loved.” Harris told the Innocence Project.
Harris’ story is an example of the adverse effects critics argue plea bargains have on criminal defendants. His story is unusual—his plea bargain came after 29 years in prison, rather than in lieu of an initial trial—but it nevertheless features a key complaint critics have of the practice generally. Namely, that defendants can be compelled to accept guilty pleas for crimes they’re ultimately innocent of. Downing’s identification of Harris in hopes that cooperating with police would get him a better plea bargain could also be interpreted as further evidence of plea bargaining’s potential adverse effects.
Harris’ example is also somewhat unique in that his initial case went to trial. The Constitution’s Sixth Amendment guarantees the right to a speedy and public trial to all individuals accused in a criminal prosecution. But that right is infrequently exercised in the modern American justice system. In most jurisdictions, at both the federal and state levels, few criminal defendants go to trial. The vast majority of cases are settled without trial largely thanks to the widespread use of plea bargaining.
When a defendant agrees to a plea bargain, they enter a guilty plea in exchange for prosecutorial concessions, such as lesser charges or a reduced sentence.
There are several different types of plea bargain. One of the most commonly used is charge bargaining, where a defendant agrees to plead guilty to a lesser charge, often carrying either a lower maximum sentence or lacking a mandatory minimum sentence, than the one with which they were originally charged or presented. Charge bargaining can also involve a defendant agreeing to plead guilty to one or more of the original charges, with the prosecutor agreeing to drop the rest.
Another type of plea bargain is a sentence bargain, where the defendant pleads guilty with the agreement that either the prosecutor will recommend or the judge will impose a lower sentence or sentencing range. Plea bargains can also involve fact bargaining, where the prosecutor and defendant negotiate over the facts that will serve as the basis for a charge.
It’s a practice that has both supporters—who argue that reduced sentences benefit defendants and help streamline the justice system—and detractors—who argue the practice is coercive, enables corrupt prosecutorial behavior, and erodes the right to trial.
Eligibility is another variable. According to Maureen Platt, State’s Attorney for the Judicial District of Waterbury, there are three primary factors that are almost always used to determine whether a defendant should receive a plea bargain.
First, Platt says prosecutors look at what occurred in a factual scenario, as not all crimes of the same type occur in the same way, so prosecutors take into account the factual pattern and its egregiousness.
Second, prosecutors look at the prior record of the person who’s being charged, including their age and the number of times they’ve been charged in the past.
Third, prosecutors look at the strength of the state’s case. “Many times when we make a decision to make a lower offer, it’s because the case is not as compelling. Take a robbery, where there’s video that is clearly of the defendant, where their cell phone data puts them at the scene, where there’s a confession, where the victim can make an ID, where DNA is found. Then it would be a stronger case.” Platt said.
An additional factor prosecutors take into account is the charges involved.
Determining the precise rate at which plea bargaining occurs is complicated by a variety of factors. In Connecticut, for example, data on the use of plea bargains in criminal cases isn’t stored in one centralized location. It’s split between the Judicial Branch and the Office of the Chief State’s Attorney.
While trying to obtain information on the rate at which plea bargaining is used in Connecticut, this reporter first reached out to the Judicial Branch, who suggested reaching out to the Division of Criminal Justice within the Office of the Chief State’s Attorney as they maintain the most complete records on the use of plea bargaining from the prosecutor’s side of a criminal case. They also cautioned that the chief state’s attorney’s office would likely refer that inquiry back to the Judicial Branch, which they did, stating the Judicial Branch had the most complete and accurate information. Efforts to follow up were not answered.
The Judicial Branch was able to fulfill a data request for trial outcomes in criminal and motor vehicle cases, categorizing the data by the manner in which cases were disposed and by year.
This data does not provide a comprehensive look at the rate at which plea bargaining is used. The data set shows that convictions and cases which were disposed of without going to trial make up the vast majority of case dispositions for criminal and motor vehicle conditions in the data provided. But they do not specify the reason cases were disposed without a trial.
According to a Judicial Branch employee, the data turned over for requests such as the one Connecticut Inside Investigator submitted was originally designed so that court clerks can manage case flows. It was not originally designed for data analysis, though additional information has been added over the years.
The data shows that approximately 99.6 percent of criminal and motor vehicle cases were disposed of without a trial between 2015 and 2020. That number rose to 99.8 percent in 2021 and then fell slightly to 99.7 percent in 2022. It’s worth noting that both 2020 and 2021 were abnormal years, with fewer cases being prosecuted because of the COVID-19 pandemic.
In 2023, the rate of criminal and motor vehicle convictions secured without trial is 99.6 percent through May 9, the date Inside Investigator received the data request.
But according to Platt it’s safe to assume that cases disposed of without a trial are the result of a plea bargain.
“If cases do not go to trial, I think it is extremely likely that they are the result of a plea bargain, because how else would they have happened?” Platt stated, noting that prosecutors handle 100 percent of cases that come before them and can oftentimes prevent someone from using a diversionary program.
The Judicial Branch’s numbers put the rate at which criminal convictions are secured by trial in Connecticut among the lowest in the nation. According to a 2023 report published by the American Bar Association’s Criminal Justice Section’s Plea Bargain Task Force Pennsylvania, New York, and Texas have trial rates of less than 3 percent. The report further notes that some jurisdictions in the country “report not having had a criminal trial in years.”
At the federal level, the trial rate is similarly low. According to U.S. District Court data for the 12-months between December 31, 2021, and December 31, 2022, just under 2 percent of all criminal defendants were convicted by either a bench or jury trial. Of the 71,896 individuals tried, 65,462 were convicted. Of the convictions secured during that time period, 97.8 were secured by a guilty plea rather than a trial. That means just 2.2 percent of convictions were secured by a trial. By contrast, of the 6,434 defendants accused during that time period who were not convicted, approximately 4.5 percent were acquitted by a trial rather than by a dismissal of their case.
While state-wide data suggests trials occur at a lower rate in Connecticut than at a federal level—which suggests plea bargains may also be used at a higher rate—data from Connecticut’s judicial districts suggests there is some variety in the rate at which some cases go to trial around the state.
Data collected annually by the Division of Criminal Justice (DCJ) contains demographic information about reported crimes, arrests, and trial outcomes in each of the state’s 13 judicial districts. Most cases within those jurisdictions remain in geographic courts in their respective area, but more serious cases, where the charges are greater than a Class D felony, are transferred to the Part A court within the same county.
The data collected by the DCJ includes information on how cases that are transferred to Part A court are disposed—with or without a trial. For all of the state’s judicial districts, the data continues to show that a high rate of transferred cases were disposed of without trial. Data for each district recorded transferred cases between 2016 and 2020 (though not all districts contained data for all years in this ranged) and showed that, of the state’s judicial districts, the number of Part A cases disposed of without trial varied between a high of 99.4 percent (Litchfield and Stamford/Norwalk in 2017) and a low of 80.9 percent (New Haven in 2017).
According to Platt, in 2022 44 Part A cases were tried to a verdict in the Waterbury judicial district. 1,989 cases were disposed of by plea, making the rate of cases that went to verdict 2.3 percent. The rate of Part B cases that went to verdict was even lower. In 2022, 31 cases were tried to a verdict and 99,441 were disposed of by plea, making the percent of cases settled by trial. Roughly 99.2 percent of part B cases were disposed of through plea.
In total, the judicial district settled 101, 204 Part A and Part B cases in 2022. 77 were disposed of by a verdict and 101, 130 were settled by a plea, meaning less than 0.1 percent of cases were settled by a verdict.
Available data shows that most criminal defendants in the state will not go to trial. For some, this may be because the charges against them were dropped, or because of pretrial motions. But most criminal defendants won’t see a trial because they choose to accept a plea bargain.
According to Hartford-based defense attorney Mario Cerame’s experience, plea bargaining in Connecticut is “almost exclusively used” in criminal trials. “As a private attorney, not someone who’s in the public defender’s office, we’re almost always not going to trial. It’s very rare that something is going to be tried. There’s a very strong presumption when someone goes in that you’re not going to have to go to trial.”
To some of plea bargaining’s critics, this presumption that someone isn’t going to trial is one of the practice’s biggest flaws. There are a number of reasons to presume a trial will not occur, and one of them involves sentence bargaining. Critics of plea bargaining argue the difference between the longer sentence awaiting someone who gambles on going to trial, where there’s no guarantee that any exculpatory evidence will be brought forwards, creates a trial penalty, thus coercing a defendant into accepting a plea bargain and sacrificing their Sixth Amendment rights. Another argument is that the trial penalty can lead to prosecutors overcharging defendants in an attempt to influence their decision to take a plea deal.
There is evidence to back up both these claims: A 2018 report from the National Association of Criminal Defense Lawyers (NACDL) on the trial penalty referenced the U.S. Sentencing Commission’s data on federal sentencing, which showed that the average post-trial sentence was more than three times longer than the average post-plea sentence for most primary offense categories in 2015. The organization reported that for convictions on burglary/breaking and entering and embezzlement that sentencing gap increased to be eight times as long.
According to the NACDL, the trial penalty is the result of shortcomings that exist across the criminal justice system: prosecutors having wide latitude at the early stages of criminal proceedings, federal guidelines that impose mandatory minimum sentences for a variety of crimes and limit a judge’s sentencing discretion, among other factors.
But the use of plea bargaining—and the presumption that a defendant will not go to a trial—is not inherently bad in every case.
At least for criminal cases in Connecticut, Cerame says that whether plea bargaining is abused is about the “particulars or circumstances” of a defendant’s case and the charges they are facing, but overall, the use of plea bargaining “systemically creates problems, creates problems in incentives, and problems for justice systems as whole.”
For individual clients, Cerame says, the presumption that they’re not going to trial can be a positive. “If the client has actually made a mistake, it’s usually in their interest to come to the fairest outcome, which is a plea bargain.”
It’s the cases that aren’t typical where plea bargaining can be more of a problem. “For a case that doesn’t fit the mold, or is different, then [the presumption a client will not go to trial] very well probably is not to their advantage,” said Cerame.
Cerame also notes that, as a private defense lawyer, he doesn’t have a broad view of the process. “Prosecutors do, and certainly public defenders offices can too.” Cerame noted.
Cerame also raised the issue of motion practice. “Because plea bargains don’t normally involve motion practice, that can sometimes be to your client’s disadvantage. And that’s a problem. Most of them might not be trying to do a motion to dismiss or suppress evidence.”
He also noted that even when a client goes to trial, motions still involve engaging in a sort of plea bargaining. Whether pretrial motions come up, and when they are heard, is all a form of bargaining between the parties involved in a case.
Another possible way the presumptive use of plea bargaining can be to the disadvantage of a client, Cerame says, is that the job of the defense in a criminal trial is to incentivize a quick resolution. And while prosecutors can’t legally punish someone for going to trial, that line can get blurred.
“On the one hand, are they punishing the person for trying the case or are they incentivizing a quick resolution?” Cerame said. “Sometimes they are and sometimes they aren’t.”
According to Platt, the perception of the trial penalty is incorrect. Platt pointed to mandatory minimum sentences, which require defendants charged with certain crimes to receive a penalty with a set minimum number of years. And that takes control of sentencing out of the hands of prosecutors.
“Let’s say someone is charged with the rape of a seven-year-old child. The legislature has determined that person should do ten years at a minimum. That’s not a prosecutorial or judicial decision. That is a legislative decision.” said Platt, who said this is typical of what happens with many Part A cases in the state.
Platt also noted that, in cases such as the one described, where the victim may not want to testify out of fear of the emotional trauma involved, prosecutors can offer the defendant a lower sentence to try to avoid it, but if the defendant refuses the offer, the case goes to trial.
Even when mandatory minimums are not in play, sentencing is also a judicial decision, not a prosecutorial one. While prosecutors can make a pretrial offer, it’s the courts that determine a sentence in the end, Platt notes. “Sometimes they mitigate the offer and sometimes they don’t.”
Placing the blame for longer sentences on those who go to trials at the feet of prosecutors isn’t fair, according to Platt, since sentencing decisions are more complicated than that and are often outside the control of prosecutors.
Another criticism of plea bargaining is that it undermines the structure of the criminal justice system and by extension the Bill of Rights. Along with the right to a speedy and public trial, the Sixth Amendment guarantees the accused the right to be confronted with witnesses against them, to have a compulsory process for obtaining witnesses in their favor, and to have the assistance of counsel for their defense. These rights are not only predicated on trials taking place, but also on being run by knowledgeable legal professionals whose actions help secure those rights.
But without trials, Cerame says, legal professionals don’t have that knowledge.
“It’s a systemic problem that we don’t do as many trials,” said Cerame. “The judges who are coming in now aren’t used to doing trials.”
That lack of experience extends to attorneys, too. Cerame says that he’s lucky enough to be at a trial firm and so doesn’t have to fear going to trial, but that’s not the case for a lot of other attorneys. Additionally, because trials occur at such a low rate, when they do occur, they are often handled by senior partners, leaving more junior partners without experience.
According to Cerame, this can also mean that when trials do occur, there’s a greater risk of a mistrial of justice.
“Cases that we do try are prone to accidents because it’s this rare exception where we go to trial.” said Cerame.
Because fewer cases are heard, Cerame says there isn’t a robust body of case law—laws based on the precedents set by previous judicial cases—being developed. And that means that bad laws stay on the books for longer periods of time, creating a risk of injustice for those who accept a plea deal and are convicted under them.
Cerame pointed to Cerame v. Lamont, a case he successfully argued violated his free speech rights,as an example. The case involved Connecticut’s “racial ridicule statute,” passed in 1917, which criminalizes advertisements that ridicule or show contempt for people on the basis of demographics such as race and religion. According to the Foundation for Individual Rights and Expression, which filed an amicus brief in the case, the law was being used by police, ignoring the law’s application to advertising, to arrest people who directed racial slurs against them.
In February 2020, two white UConn students were arrested and charged under the law for using racial slurs while walking through the parking lot of a campus apartment complex.
A 2008 memo produced by the state’s Office of Legislative Research (OLR) reporting on Connecticut’s hate crime laws even stripped the context of commercial speech from the statute, stating that a person “commits this crime if he ridicules any person or class of people on account of creed, religion, color, denomination, nationality, or race.” Violations of the statute were punished by either 30 days in prison or a fine of $50.
Cerame filed a federal lawsuit challenging the constitutionality of the law under the First Amendment. A federal district court referred the case to the Connecticut Supreme Court, asking them to clarify whether the law touched on noncommercial speech.
Cerame argued that he used speech within the statute’s scope to express camaraderie and in conversations with close friends. He also alleged that he used the same types of speech to ridicule people he considers harmful to society.
The court sided with Cerame, holding that the legislature’s intent in passing the law demonstrated it was not intended to include personal, noncommercial speech. The case is an example of how laws can be misapplied, or how unconstitutional laws can be enforced, without challenge when they’re not examined during criminal trials.
According to Platt, there are also drawbacks from a prosecutor’s perspective to the low rate of trials. Platt described Part B cases, which are less serious criminal and motor vehicle offenses tried in geographical area courts. Platt says that, statistically, the state decides to nolle, effectively a dismissal of charges, approximately a third of Part B cases. This can occur for a variety of reasons, including because the case against the defendant isn’t strong, the charges are minor and the defendant has shown remorse, and because the victim doesn’t want to prosecute. In another third of cases, defendants are granted entry into a diversionary program where charges are dropped upon completion.
Many of these, Platt says, didn’t exist ten to fifteen years ago. “Many, many times, defendants take advantage of those. And those don’t result in a conviction.” The use of such programs partially explains decreases in conviction rates.
According to Platt “substantial reasons cases don’t go to trial,” include cases being nolled or dismissed through diversionary programs. Indeed, 66 percent of cases don’t make it to a trial.
Another reason many cases don’t go to trial is limited resources. Platt notes that Part B cases are frequently less serious, which means prosecutors have to prioritize which cases they want to invest resources in to prosecute via trial.
“Connecticut is one of the only states where jurors are individually interviewed, which is incredibly time consuming.” said Platt.
On the whole, between the 66 percent of cases that will not be prosecuted and the prioritizing prosecutors have to do, very few Part B cases are tried. “Exceptions could be driving while intoxicated, which have mandatory minimums. Some gun cases have low mandatory minimums.” Platt added. As with sentencing, the decision to go to trial is not always in the hand of prosecutors.
Not all defendants are able to afford private legal counsel, which can put criminal defendants who are not as financially well off and who want to exercise their Constitutional right to trial, rather than accept a plea bargain, at a disadvantage. It can also put defendants who need to navigate accepting a favorable plea deal at a disadvantage. Cerame described a “donut hole effect,” where a certain percentage of defendants would be able to find affordable, experienced legal counsel, while the remainder struggled to obtain adequate representation.
“Public defenders are skilled but overwhelmed, but there’s a cutoff. And if you’re above the cutoff, that doesn’t mean you have the resources to afford private counsel. That’s a huge number of people.” said Cerame. He estimated that about 75 percent of defendants are able to find experienced legal counsel to handle their defenses, but that the remaining 25 percent, especially if they’re looking for someone to help them take a case to trial, were likely to have more of a challenge.
As with statistics on the use of plea bargaining, comprehensive statistics on the number of criminal defendants unable to find adequate legal representation are difficult to find. Connecticut does report annually on the cases referred to and handled by indigent legal services. According to the Connecticut Division of Public Defender Services’ annual report for fiscal year 2021-2022, the total number of assigned counsel attorneys for criminal, child protection, and family attorney for a minor child/guardians ad litem cases was 20,942 in fiscal year 2022. For the same year, the division had 222 full time attorneys on staff. But data on the caseloads for private defense attorneys, or on the number of defendants unable to find adequate legal defense, is not publicly available.
The lack of hard data makes it difficult to make any kind of comprehensive assessment about not just the rate at which it’s used in the state, but the ways in which prosecutors offer plea bargains. That’s something over which the state has been criticized before.
In 2019, the Connecticut Advisory Committee of the U.S. Commission on Civil Rights issued an advisory memorandum to Connecticut’s legislature recommending they pass legislation to increase prosecutorial transparency and data collection, provide additional funding for prosecutorial training, increase opportunities for public input in the prosecutorial appointment process, and support the expansion of the Chief State’s Attorney’s Early Screening and Intervention Program.
Connecticut is just one of three states, plus the District of Columbia, where prosecutors are appointed rather than elected. The DCJ, which houses the Office of the Chief State’s Attorney, State Attorneys, and the Criminal Justice Commission, is an autonomous body created by the state’s constitution.
According to testimony heard by the civil rights commission’s Connecticut advisory committee, Connecticut’s Criminal Justice Commission did not have “the outcome data to independently evaluate applicants” for prosecutors and instead relied on reviews submitted by the Chief State’s Attorney. At a briefing, Connecticut’s deputy chief state’s attorney stated that he could only recall one prosecutor not reappointed within the last 25 years.
Additionally, the committee heard testimony that the DCJ didn’t have the ability to produce meaningful data on prosecutorial policy and conduct because the state’s attorney’s offices didn’t have an electronic management system. The committee also noted that the DCJ is exempt from the state’s Freedom of Information Act in most instances, making it difficult for the public, including journalists, to learn about prosecutorial decisions.
The commission recommended any law the state pass regarding prosecutorial transparency include a requirement for the DCJ to collect and publicly report disaggregated case level by docket number, including information on the total number of defendants prosecuted for felonies and misdemeanors, demographic data, indigency, and plea deals—including the first and last plea deal offer by charging offense.
A 2021 Senate bill would have incorporated many of these recommendations, including requiring that performance reports for state’s attorneys include data on plea agreements, trial dispositions, and other data points. However, the law was never voted out of committee.
A much narrower prosecutorial accountability bill, which requires the Criminal Justice Commission to call each state’s attorney before it annually to testify and comment on annually collected data, passed during the legislature’s most recent session and has since been signed into law.
But without broader reforms, the public is largely in the dark on how state prosecutors use plea bargaining. And without that data the public has only stories, like that of Leroy Harris, to judge whether Connecticut has a plea bargaining crisis.
Republish our articles for free, online or in print, under a Creative Commons license.
