“Are you in favor of the death penalty for a person convicted of murder?”

For the past decade, about two out of three people in the United States have said yes, they are. But at the same time, the death penalty has been on the decline. In 2000, 224 people were sentenced to death in the United States. In 2010, half as many were given death sentences.

Click to read photo caption. Jason Smith

On a basic level, the moral conversation about the death penalty hasn’t changed. A majority of people think that death is the only just punishment for the most horrific of crimes. Others think exacting payment for a murder with yet another killing is moral nonsense.

It could be that neither of these two arguments will ever win out over the other. But the ethical stalemate may not matter as much as we think. While the debate goes on, researchers are finding out why the death penalty—right or wrong—is fading away. Capital punishment in the United States could disappear, they say, regardless of what anyone believes about an eye for an eye.

All through the 1990s, the number of annual executions in the United States was on the rise, climbing from twenty-three in 1990 to a peak of ninety-eight in 1999. Four of the prisoners executed in 1999 were from North Carolina, making the state sixth (tied with South Carolina) in number of executions carried out that year. Two of the men executed in North Carolina were white and two were black. But all five of the murder victims in those cases were white.

That same year, UNC law professor Jack Boger decided to find out if race was playing a role in death penalty cases in North Carolina. Before coming to UNC, Boger had litigated capital punishment cases—one of them, McCleskey v. Kemp, was heard by the U.S. Supreme Court. Warren McCleskey, a black man, had been convicted of killing a white police officer in Fulton County, Georgia, in 1978. In McCleskey’s final appeal to the Supreme Court, Boger used a 1970s study showing that a Georgia defendant was much more likely to receive the death penalty if the victim in the case was white than if the victim was black, and that black defendants were more likely in general to receive the death penalty. The Supreme Court, split 5-4, upheld McCleskey’s sentence because the defense hadn’t proved that there were racial factors at work in McCleskey’s particular case. In 1991, McCleskey was executed, thirteen years after the Georgia officer’s murder.

Click to read photo caption. Donn Young

But Boger didn’t give up. He wanted another study like the one in Georgia, but this time in North Carolina and covering more recent cases. So in 1999 he brought the idea to Isaac Unah. Unah, a UNC political scientist, had never been involved with death penalty issues, but he had the training in data collection and analysis to head up a study of the almost four thousand homicide cases that took place in North Carolina between 1993 and 1997.

Eight new graduates from UNC’s School of Law studied newspaper articles, visited courthouses, and interviewed defense attorneys and prosecutors all over the state about murder trials they had participated in. The group gathered data on everything from the evidence police had collected about crimes to the political leanings of the counties where the cases were tried to the dates when prosecutors were coming up for election.

The goal, Unah says, was to find out whether and how race played a part at each stop on the road to a death sentence: the prosecution’s choice to seek the death penalty, the plea bargaining stage, the trial itself, the verdict, and the sentencing. This was unusual; most studies of race and the death penalty had looked only at the final outcome.

Click to read photo caption. Jason Smith

They found that a defendant’s race sometimes does matter. The cases most likely to end in a death sentence were the ones where a black defendant was on trial for murdering a white victim. But the main thing that mattered was the victim’s race. A defendant accused of killing a white person faced odds of getting a death sentence that were 3.5 times higher than those for a defendant accused of killing a nonwhite person. And it wasn’t the prosecutors being racially biased—it was juries who came down harder when the victim of a crime was white.

After the study came out in 2001, some activists immediately called for a death penalty moratorium in North Carolina. Boger did as well. In 2002, the N.C. General Assembly considered the first version of a Racial Justice Act bill, backed by Unah and Boger’s study. It failed, and it would fail whenever it was introduced for several years to come.

While all this was happening, Frank Baumgartner, also a political scientist, was studying how public policies change. Sometimes, he says, laws and government policies change slowly and steadily, but sometimes they can move so fast it seems like the change happens overnight. What he’s most interested in is figuring out how the national conversation about a topic can suddenly take off in a new direction—not necessarily because the facts have changed, but because the issue is being talked about in a different way. This is what happened to the death penalty.

It’s sort of like what happened with nuclear power, Baumgartner says. “In the 1950s and 60s, nuclear power was seen as having the potential to save humankind. It would produce so much energy so cheaply, it wouldn’t be worth the cost to meter it.” And then, just like that, around 1968 the public discussion flipped. “Suddenly,” Baumgartner says, “it was all about mushroom clouds, danger, radioactivity, health risks to workers at the power plants and to people who lived near the plants.”

In the 1990s, he says, capital punishment seemed unstoppable. More and more people were getting death sentences. Politicians were spreading the “tough on crime” message. Newspapers were full of uncritical accounts of death penalty cases, focusing on the lives of the victims and the horrific details of the crimes. When exonerations of convicted murderers happened, they weren’t talked about much, Baumgartner says. “Back then, people who were working against the death penalty didn’t want press coverage,” he says. “They knew that what the papers said would be horrible for their clients. It was like opposing a freight engine—you might as well just get out of the way.”

Then, one day in 2002, someone asked Baumgartner how to go about reframing public perception of the death penalty. It doesn’t really work that way, Baumgartner told him. Framing can’t be controlled by one person or one activist group—it happens when a lot of factors come together at once.

But the idea of studying the death penalty intrigued him. Unlike a lot of public policy issues, the death penalty has straightforward outcomes you can count: numbers of death sentences and executions. That appeals to someone who would rather point to hard data than sit around discussing theory.

Baumgartner got on the train at the right time: in 2002, public discussion about the death penalty had just flipped, the same way that nuclear power had. What once seemed a sure thing was now fraught with questions. Were innocent people being executed? Was race a factor in convictions? How could a defendant be found guilty beyond a reasonable doubt, only to be exonerated through DNA evidence years later?

“A snowball effect happened,” Baumgartner says. “When bad news started, it accelerated.” There wasn’t some meeting where activists against the death penalty got together and decided to change how the United States was talking about the death penalty, he says. But it happened anyway.

Baumgartner and an undergraduate student, Cheryl Feeley, started analyzing every New York Times article between 1960 and 2000 that had to do with the death penalty. This eventually turned into a book-length project, The Decline of the Death Penalty and the Discovery of Innocence, which traces the downward trend not to the activism of politicians, lawyers, or prosecutors, but to the way the issue is talked about in society at large.

Exonerated from Death Row

Each of these 140 people was convicted and sentenced to death. In each case, one of the following two conditions applied: 1) the defendant’s conviction was overturned and a) the defendant was acquitted at retrial, or b) all charges were dropped; or 2) the defendant was given an absolute pardon by the state’s governor based on new evidence of innocence.

1973 David Keaton 1974 Samuel A. Poole 1975 Wilbert Lee, Freddie Pitts, James Creamer, Christopher Spicer 1976 Thomas Gladish, Richard Greer, Ronald Keine, Clarence Smith 1977 Delbert Tibbs 1978 Earl Charles, Jonathan Treadway 1979 Gary Beeman 1980 Jerry Banks, Larry Hicks 1981 Charles Ray Giddens, Michael Linder, Johnny Ross, Ernest Shuhaa Graham 1982 Annibal Jaramillo, Lawyer Johnson 1985 Larry Fisher 1986 Anthony Brown, Neil Ferber, Clifford Henry Bowen 1987 Joseph Green Brown, Perry Cobb, Darby Williams Tillis, Vernon McManus, Anthony Ray Peek, Juan Ramos, Robert Wallace, Richard Neil Jones 1988 Willie Brown, Larry Troy 1989 Randall Dale Adams, Robert Cox, Timothy Hennis, James Richardson 1990 Clarence Brandley, John C. Skelton, Dale Johnston, Jimmy Lee Mathers 1991 Gary Nelson, Bradley P. Scott, Charles Smith 1992 Jay C. Smith 1993 Kirk Bloodsworth, Federico M. Macias, Walter McMillian, Gregory R. Wilhoit, James Robison, Muneer Deeb 1994 Andrew Golden 1995 Adolph Munson, Robert Charles Cruz, Rolando Cruz, Alejandro Hernandez, Sabrina Butler 1996 Joseph Burrows, Verneal Jimerson, Dennis Williams, Roberto Miranda, Gary Gauger, Troy Lee Jones, Carl Lawson, David Wayne Grannis 1997 Ricardo Aldape Guerra, Benjamin Harris, Robert Hayes, Christopher McCrimmon, Randall Padgett, James Bo Cochran 1998 Robert Lee Miller, Jr., Curtis Kyles 1999 Shareef Cousin, Anthony Porter, Steven Smith, Ronald Williamson, Ronald Jones, Clarence Dexter, Jr., Warren Douglas Manning, Alfred Rivera 2000 Steve Manning, Eric Clemmons, Joseph Nahume Green, Earl Washington, William Nieves, Frank Lee Smith, Michael Graham, Albert Burrell, Oscar Lee Morris 2001 Peter Limone, Gary Drinkard, Joaquin Jose Martinez, Jeremy Sheets, Charles Fain 2002 Juan Roberto Melendez, Ray Krone, Thomas Kimbell, Jr., Larry Osborne 2003 Aaron Patterson, Madison Hobley, Leroy Orange, Stanley Howard, Rudolph Holton, Lemuel Prion, Wesley Quick, John Thompson, Timothy Howard, Gary Lamar James, Joseph Amrine, Nicholas Yarris 2004 Alan Gell, Gordon Steidl, Laurence Adams, Dan L. Bright, Ryan Matthews, Ernest Ray Willis 2005 Derrick Jamison, Harold Wilson 2006 John Ballard 2007 Curtis McCarty, Michael McCormick, Jonathon Hoffman 2008 Kennedy Brewer, Glen Edward Chapman, Levon Bo Jones, Michael Blair 2009 Nathson Fields, Paul House, Daniel Wade Moore, Ronald Kitchen, Herman Lindsey, Michael Toney, Yancy Douglas, Paris Powell, Robert Springsteen 2010 Anthony Graves

—Jason Smith

Arguments about the morality of the death penalty have always dominated the discussion, Baumgartner says. But at the end of the twentieth century, other arguments started getting more air time in the media. (His team compared their New York Times data with ten other national publications, he says, and all of them had similar death penalty coverage.) New DNA evidence coming out of old cases made it crystal-clear that some people who end up on death row are innocent. Articles about the possibility of innocent inmates and the constitutionality of the death penalty showed up more and more often starting late in the 1990s. When a wrongful conviction was discovered, the media paid a lot more attention than before. An exoneration didn’t just mean it was one convict’s lucky day—it was a sign that something major wasn’t working in the death penalty system.

These articles didn’t have immediate effects on public opinion. But they did seem to have a huge effect on juries. Positive articles about the death penalty meant more death sentences over the next several years, and negative articles meant fewer death sentences. These framing effects were much better at predicting death sentences than other factors social scientists have used to try to explain why people get sentenced to death. A rise in the national homicide rate, for example, is known to make juries more likely to give the death penalty. But the researchers found that the tone of media coverage was much more significant than how homicide numbers were moving.

“It’s not that we think juries are sitting down and reading the New York Times or any one of these articles,” Baumgartner says. “But we think those arguments find their way into the things that the juries are thinking about, the lawyers’ closing arguments, how the prosecutors make their case. It gets out there.”

Baumgartner looked at the effects of framing on public opinion of the death penalty, too. The influence was there and it went in the direction they expected—media coverage predicted public opinion much more strongly than public opinion predicted media coverage—but the effect was modest and slow. Changing public opinion is tough, Baumgartner says, because people hold on tightly to their moral beliefs.

North Carolina politicians might not really have changed their minds about the morality of the death penalty, Unah says. But after his and Boger’s 2001 study of race and death sentences, more research started piling up.

“The one really consistent finding in all those studies,” Unah says, “whether they’re done in the South, in the North, in the East, using a large sample or a small sample, is that defendants who kill white people come out significantly worse.” They are more likely to receive death sentences and more likely to have those sentences carried out. Eventually, he says, the weight of evidence for that conclusion convinced N.C. legislators to pass the Racial Justice Act in 2009.

“I have always been a supporter of the death penalty,” Governor Beverly Perdue said when she signed the act into law. “But I have always believed it must be carried out fairly.”

Under the Racial Justice Act, inmates on North Carolina’s death row can have their sentences reduced to life in prison if they can show that in North Carolina racial bias is involved in the death penalty. Almost all the inmates have appealed, and many of their lawyers have asked Unah for statistics.

Republican Party legislators elected in 2010 have said they’ll try to repeal the Racial Justice Act. That might not affect the cases already under appeal. “Unless,” Unah says, “the legislation is written in such a way that it puts an immediate stop to all appeals. There is precedent for that. A legislature can remove jurisdiction of a judge to make a decision on a certain case.”

He hopes that won’t happen. He thinks the Racial Justice Act is a first step in addressing the systematic problems underlying the death penalty in North Carolina. It starts long before a jury decides the sentence, Unah says. Studies show that when police are investigating a murder, they gather more evidence if the victim is white. Killings of black victims aren’t investigated as thoroughly. In turn, how much evidence there is and how compelling it seems influences whether a prosecutor will try for the death penalty.

But it isn’t just about race. Prosecutors may not be influenced by race in itself, but they are influenced by politics. Unah and Boger found that when prosecutors are coming up for reelection, they are more likely to seek the death penalty, because being seen as tough on crime is popular.

“For the process to be fair,” Unah says, “you’d have to remove the decision to seek the death penalty from the hands of prosecutors. You could set up an independent group made up of informed citizens who make that decision, completely and solely removed from politics.”

As for crime investigation, Baumgartner says, the problem is that investigators work for the accusing side. The dangers of this setup became clear in North Carolina in 2010, when a review of the State Bureau of Investigation turned up more than two hundred cases in which the bureau had withheld or distorted evidence. “If we want a truly neutral system of justice,” Unah says, “they would have to be working neutrally, unrelated to either the police or the defense.” But that kind of shakeup of the criminal justice system in North Carolina would take a huge amount of money and time, he says, and the political will to do it just isn’t there yet.

So what will happen to the death penalty on the national scale? Right now, the idea of innocent people getting executed still dominates the discussion. As long as that’s the case, Baumgartner says, death sentences and executions are almost sure to keep declining. In order for the death penalty to make a recovery, the conversation would have to shift away from arguments about innocence and fairness. And there’s nothing to say that that won’t happen.

It’s troubling to think that innocent people might be getting executed. But if the death penalty keeps declining, couldn’t there eventually be a backlash, a point where people think: “It’s great to know that we’re not executing innocent people, but it’s coming at the price of executing hardly any guilty people, either”?

Click to read photo caption. Jason Smith

“That could happen,” Baumgartner says. “But there’s another possible scenario. Say the number of death sentences declines and declines and declines, and there get to be more states like North Carolina that have long periods of history with no executions. At that point, a lawyer might be able to argue successfully that it’s arbitrary to execute their client because no one’s been executed in the last ten years. That would be the definition of unusual punishment.”

The death penalty has been ruled unconstitutional once before, in 1972, only to come back four years later. “Capital punishment is allowed under the U.S. Constitution,” Unah says. “Will there ever be an amendment to the Constitution outlawing it? Probably not.” But the death penalty could be outlawed by the Supreme Court on the grounds that it’s just too hard to apply fairly and correctly. “And if state legislatures are unwilling to spend the money to make it better,” Unah says, “then what would be the point of having the death penalty?

Frank Baumgartner is the Richard J. Richardson Distinguished Professor of Political Science in the College of Arts and Sciences at UNC. He conducted his research with Suzanna De Boef, a professor of political science at Pennsylvania State University, and Amber Boydstun, now an assistant professor of political science at the University of California, Davis. The three wrote The Decline of the Death Penalty and the Discovery of Innocence, published in 2008 by Cambridge University Press. Isaac Unah is an associate professor of political science in the College of Arts and Sciences at UNC and John Charles Boger is dean of the School of Law. Their study was funded by the Common Sense Foundation.